So you have dared to enter this forest, which was never yours? Do you not know that this is our homeland, passed down from our parents and ancestors? … Are you so poor that you must take your riches from our forest, robbing the leaves of our trees to make a drink?Footnote 1
These words were ascribed by the Jesuit missionary Martin Dobrizhoffer to an anonymous Guaraní warrior in the eighteenth-century Atlantic Rainforest of South America, allegedly delivered during an Indigenous attack on Spanish yerba mate collectors.Footnote 2 While it is impossible to know how much poetic licence Dobrizhoffer took with this speech, it does resonate with themes raised by contemporary Guaraní thinkers, such as the threat posed to the Atlantic Rainforest by colonisation and the centrality of this biome to the Guaraní way of life.Footnote 3 As Dobrizhoffer notes, this particular attack was successful in its aims, expelling Spanish colonists from this region of the forest for a number of years.Footnote 4 In this article, we argue that acts of resistance such as this one, carried out by Guaraní and other Indigenous forest-dwelling peoples, had a cumulative ecological impact of global significance, by obstructing tropical deforestation over the long term in South America.
Tropical deforestation is one of the most pressing environmental issues of the twenty-first century, with the survival of tropical forest biomes being an essential component of the global campaign to prevent runaway climate change. The escalation of land-use change in tropical forests has been identified as an indicator of the so-called ‘Great Acceleration’, part of a series of human-driven planetary changes that have arguably intensified over the last seven decades.Footnote 5 Recent scholarship has traced the historical origins of commodity-driven tropical deforestation back considerably further, to the sixteenth century, with the onset of European colonisation and the capitalist world-market.Footnote 6 However, the potential role of Indigenous resistance in obstructing and shaping deforestation patterns over this historical period, and the legacies this has left for current patterns of deforestation and forest cover, has received relatively little attention.Footnote 7 Recent research is increasingly recognising the importance of Indigenous tropical forest territories across the Americas, Africa, and Asia as key sites of biodiversity and carbon sequestration, and of Indigenous peoples in these regions as key players in the fight against climate change.Footnote 8
With this in mind, this article first maps the relationship between the capitalist world-economy and tropical deforestation in South America since 1500, then analyses how Indigenous resistance has affected these deforestation patterns in two biomes: the Atlantic and Amazon Rainforests of South America. These latter two sections focus on the long-term impacts of Indigenous resistance during the colonial period in these two biomes, while also assessing how these Indigenous struggles continue in both these forest regions today. Sven Beckert et al.’s recent, agenda-setting study of global commodity frontiers alluded to the importance of local resistance in imposing constraints on capitalist expansion.Footnote 9 This article takes the next step, providing a detailed analysis of this phenomenon in tropical South America, acknowledging Indigenous forest-dwelling groups as significant actors on both a regional and global scale.
Due to the five-century-long history of European colonisation on the continent, tropical South America offers an ideal region to analyse the relationship between local deforestation processes and the capitalist world-economy over the longue durée.Footnote 10 South America is also at the centre of global debates about the relationship between Indigenous peoples and tropical forests. Some researchers argue that Indigenous agriculture in the Amazon and Atlantic Rainforests caused significant deforestation before European colonisation, with the subsequent mass death of Indigenous populations being a key driver of forest regrowth, potentially even contributing to global climatic shifts in the seventeenth century.Footnote 11 However, other studies have contested this position, demonstrating that precolonial Indigenous swidden and other agroforestry practices in South America caused far less deforestation than colonial commodity frontiers, and in various cases appear to have increased forest cover and biodiversity.Footnote 12 We align with the latter position, and argue that an exclusive focus on factors such as Indigenous demographic collapse can obscure the role that Indigenous resistance has played in conserving South American forests. In other words, Indigenous peoples have conserved South America’s forests not by dying, but by living.
We define ‘resistance’ as the sum of the activities carried out by Indigenous peoples to maintain their political and territorial autonomy, along with their ways of life.Footnote 13 In the cases studied here, this includes violent conflict, labour withdrawal, mass flight into the forest, economic sabotage, and tactical temporary alliance-making. While colonial documentation only occasionally recorded Indigenous groups justifying these acts in explicitly conservationist terms, collectively these acts made an important contribution to long-term forest conservation, by obstructing Iberian occupation of various territories.Footnote 14 We focus on Spanish and Portuguese colonisation in the Amazon and Atlantic Rainforests, with Indigenous struggles against Iberian gold mining acting as one of the article’s major historical through-lines.
The investigation utilises unpublished texts and maps from a variety of archives in the Americas and Europe, including the Archivo y Biblioteca Nacionales de Bolivia, the Arquivo Público do Estado do Pará, the Biblioteca Nacional do Rio de Janeiro, the Archivo Nacional de Asunción, the Archivo General de Indias, the National Archives at College Park, Maryland, and the Archivum Romanum Societatis Iesu. With one of the co-writers of this article being Nhandeva Guaraní (from the Atlantic Rainforest) and another being Paiter Suruí (from the Amazon Rainforest), we then put this material into dialogue with Guaraní and Paiter Suruí cosmology and oral history.Footnote 15 In doing so, we hone in on the negative space in the colonial archives, cross-referencing archival documents with Indigenous perspectives to identify where and why European colonists and their financial backers were unable to achieve their goals. Following Ann McGrath and Lynette Russell’s lead, we seek to bridge the artificial disciplinary divide between Indigenous history and global history.Footnote 16 Key themes of global history, such as the expansion of capitalism and the retreat of the world’s forests, can be seen in a new light if we recognise Indigenous resistance not just as a survival strategy, but also as a fundamentally disruptive material force with a planetary political and ecological impact.
Capitalism and tropical deforestation in South America
Between 1950 and 2020, more than 8 million square kilometres of tropical forest were estimated to have been cleared, an area larger than Australia.Footnote 17 This period saw both the intensification and extensification of commercial agricultural, livestock-raising, mining, and tree plantation land use in tropical forests.Footnote 18 However, widening our historical frame of analysis can help us trace the political and economic pathways that have led to the current global ecological crisis. According to environmental historian Michael Williams’ calculations, more than 4 million square kilometres of tropical forest had already been cleared between 1700 and 1950.Footnote 19 While commodity-driven tropical deforestation has increased dramatically in the last seven decades, the emergence of this socioeconomic relationship to tropical forests can be traced back five centuries earlier. Karl Marx noted in the first volume of Capital that ‘the modern life-history of capital can be dated from the creation of modern world-trade and a world-market in the sixteenth century’.Footnote 20 The modern life-history of tropical deforestation is tightly linked to this process of colonial and capitalist expansion.
There have been extensive debates about the origins of global capitalism and its environmental impacts, with Andreas Malm, for example, identifying the advent of fossil fuel usage in Britain in the late eighteenth century as a crucial pivot point.Footnote 21 The Industrial Revolution and the Great Acceleration both escalated capitalist deforestation in the tropics. However, as scholars associated with world-systems analysis such as Giovanni Arrrighi and Jason W. Moore have identified, the Iberian powers, backed by Genoese, southern German, and Flemish capital, had already established several key components of this system in the Americas in the sixteenth century, such as large-scale ore extraction and tropical plantations supported by European credit systems.Footnote 22 We broadly align with this position, arguing that a capitalist world-economy had emerged since the sixteenth century, with evolving cycles of accumulation, spearheaded by a series of European capitalist agents (the Genoese and the Dutch, followed by the British and then the United States), driving intensifying waves of deforestation on commodity frontiers in the tropics.Footnote 23 Nevertheless, we also contend that world-systems analysis has tended to neglect the historical impact of Indigenous actions, particularly those of non-state Indigenous groups during the colonial period, on the periphery of the capitalist world-economy.Footnote 24 In this article we open a dialogue between South American Indigenous forest history and world-systems analysis, to situate these histories of Indigenous resistance in a global historical and environmental framework.
An analysis of historical deforestation in the tropical rainforests of South America illustrates some of the key phases in the development of the capitalist world-economy. While the Amazon Rainforest captures contemporary headlines, the Atlantic Rainforest of modern-day Brazil, Paraguay, and Argentina is also a major biodiversity hotspot, and was approximately one-fifth of the size of the Amazon when Europeans arrived in South America. Today, less than 17% of the Atlantic Rainforest remains.Footnote 25 Figure 1 provides a provisional, two-stage visualisation of deforestation in South American tropical rainforests, depicting clearing between 1500 and 1950, then clearing between 1950 and 2020. Satellite photography dating back to the 1970s, along with US and Brazilian government surveys from the 1950s, gives us a solid scientific baseline for identifying forest cover in 1950.Footnote 26 To trace forest cover changes prior to 1950, we have assessed a mixture of archival texts and maps, while also synthesising secondary historical studies. To account for the limitations and biases in colonial maps and texts, we have triangulated these sources with Indigenous perspectives and the aforementioned satellite data.
Figure 1. Deforestation in tropical and subtropical South American rainforests, 1500–2022.Footnote 29
As Figure 1 indicates, if we focus on the Amazon, we can see a series of smaller deforestation fronts in that biome before 1900, such as colonial sugar and coffee plantations in the Guianas; gold mining in eighteenth-century Mato Grosso; and cotton crops in nineteenth-century Maranhão.Footnote 27 A mosaic of isolated clearings for cattle raising and agriculture appeared in the Brazilian Amazon before 1950, in conjunction with the first rubber boom (1879–1912) and earlier colonial-era commodity cycles. Nearly all rubber production in the Amazon depended upon wild trees rather than plantations, and was therefore selectively extractive, with the majority of Amazonian cacao and Brazil nut production following the same pattern.Footnote 28 This meant that overall, although these industries commodified various forest plant species and had a devastating effect on Indigenous populations in many areas of the Amazon, they did not cause large-scale deforestation in themselves. Within this broader context, a 1937 land tax chequebook (talonário) reveals a minor cattle-ranching frontier emerging in the savanna area around Monte Alegre on the Amazon River in Pará (maximum property size, 1,058 hectares).Footnote 30 Another 1944 talonário lists smaller agricultural sites established to feed workers on major rubber, cacao, and Brazil nut extraction zones in Pará at Altamira (maximum agricultural property size, 150 hectares). In addition, infrastructure projects linked to the rubber boom, such as the Madeira-Mamoré railway, also caused patches of localised deforestation.Footnote 31 Assessed as a whole, this data nevertheless indicates that while the Mato Grosso, Maranhão, and Pará fronts in Brazil were significant for influencing the main direction of deforestation, the majority of full-scale clearing in the Amazon Rainforest has occurred since 1950.Footnote 32
The picture changes, however, if we zoom out to include the Atlantic Rainforest, revealing a massive earlier wave of deforestation. This article builds on Arrighi and Moore’s work by offering an overview of the evolving export and credit systems driving this deforestation in South America from the sixteenth century to the present day. This process started in the sixteenth century with sugar cultivation along the Brazilian coast (funded in many cases by Flemish and, later, Dutch capital); expanded in the seventeenth and eighteenth centuries with the opening of the Brazilian gold frontier in Minas Gerais (spurred on by British capital); and advanced rapidly westward in the nineteenth and early twentieth century via the coffee frontier of Rio de Janeiro and São Paulo (servicing the US consumer market).Footnote 33 As Figure 1 visualises, the strongest deforestation front in South America has been in Brazil, where the majority of the Atlantic Rainforest was cleared before 1950, and where the largest share of the Amazon Rainforest has been cleared since 1950. A smaller scale, but still noteworthy, pre-twentieth century deforestation front is also visible in the Pacific and Caribbean forests of Ecuador, Colombia, and Venezuela, where cacao and coffee plantations drove earlier waves of clearing.Footnote 34 Livestock farming, particularly cattle ranching, has also been a constant deforestation driver across all South American tropical rainforests in both the colonial and post-independence eras.Footnote 35
Debt chains linking colonial empires and independent states in South America to the leading agents of capitalism in Europe and North America have also influenced long-term deforestation rhythms in both the Atlantic and Amazon Rainforests. The debts of the Hapsburg monarchs to south German bankers (the Fuggers and Welsers), then to the Genoese from 1557, drove Spanish ore extraction efforts in the Americas, including the search for gold in the Amazon.Footnote 36 From its independence in 1822 onwards, Brazil was increasingly indebted to British banks, with subsequent British investments in Brazilian railways extending the coffee frontier into the western Atlantic Rainforest after 1870.Footnote 37 In line with this pattern, from the 1970s onwards, the US oversaw a period of global financial expansion (the neoliberal era) during which Bretton Woods organisations such as the International Monetary Fund (IMF) and the World Bank took on a new, augmented role in global monetary regulation.Footnote 38 These institutions provided loans for the Brazilian military government, which had taken power in a US-supported coup in 1964, to launch a colossal infrastructure scheme in the Amazon, combining road and dam construction with cattle-ranching and mining megaprojects such as Jari and Grande Carajas.Footnote 39
To pay off its subsequent international debts, Brazil became increasingly dependent on export-led growth via commodities such as soy, beef, sugar, and coffee, along with gold and other minerals. This economic boom has driven continued deforestation in the Atlantic Rainforest and has opened a gigantic new beef and soy frontier in the Amazon.Footnote 40 Today, Indigenous territories, such as those of the Tupi-Mondé corridor, which includes the Sete de Setembro Indigenous Land of the Paiter Suruí community, form some of the last remaining barriers preventing runaway deforestation on the southern edge of the Amazon (see Figure 1).Footnote 41 These territories are of critical importance to the global carbon cycle, with Wayne Walker et al.’s recent study calculating that Indigenous territories in the Amazon store 32.8% of the biome’s aboveground carbon.Footnote 42 As the following analysis demonstrates, however, Indigenous resistance had already played an important role in constraining and redirecting deforestation patterns during previous centuries.
Deforestation and Indigenous resistance in the Atlantic Rainforest
In the Atlantic Rainforest, our research indicates two general regional trends. In the Brazilian section of the forest, where Portuguese colonists appropriated Indigenous labour to assist in the opening of both the sugar frontier and the Minas Gerais gold-mining frontier, deforestation advanced rapidly. Nevertheless, even within this region, localised resistance by Macro-Jê groups impeded deforestation in the eastern forests of Minas Gerais. In the western section of the Atlantic Rainforest, known as the Paraná Forest ecoregion, Indigenous resistance occurred on an even larger scale, with Macro-Jê and Guaraní-speaking Indigenous groups successfully obstructing colonisation for several centuries, contributing to a far slower deforestation process.Footnote 43
In the sixteenth century, groups from the Tupi-Guaraní linguistic subfamily (a branch of the Tupian language family) predominated in most of the major river valleys and along large stretches of the coast within the Atlantic Rainforest, practising manioc and maize swidden farming that supported substantial populations. Within this subfamily, Guaraní-speakers predominated south and west of Cananeia in modern-day São Paulo state, on the Brazilian coast, while Tupinambá speakers predominated to the north and east.Footnote 44 Interspersed with these Guaraní- and Tupinambá-speaking populations were a range of other groups from the Macro-Jê linguistic family, including the ancestors of the contemporary Kaingang, Xokleng, Krenak, Maxakali, and Pataxó peoples.Footnote 45 While Tupinambá, Guaraní, and Macro-Jê groups engaged in a diverse range of farming, foraging, and hunting practices, all of these groups were politically decentralised, with no identifiable states existing in the Atlantic Rainforest in 1500.Footnote 46
After the European invasion, Indigenous labour played a critical role in establishing the Brazilian sugar and gold frontiers, with these frontiers, in turn, initiating the largest wave of deforestation in colonial South America. During the seventeenth century, Portuguese colonists fanned out from São Paulo in long-distance slave raids and mineral prospecting expeditions known as bandeiras. These raids accumulated a population of enslaved Indigenous workers on the São Paulo plateau who cultivated wheat to feed the administrators, soldiers, and workers on the nearby sugar plantations of Rio de Janeiro.Footnote 47 Environmental historian Warren Dean calculates that Brazil’s early sugar frontier in Rio, Bahia, and Pernambuco cleared 2200 km2 in the Atlantic Rainforest.Footnote 48 In the late seventeenth century, as the sugar frontier stagnated, Paulista colonists pushed inland, using predominantly Indigenous soldiers, porters, and guides to discover the first gold-mining sites in Minas Gerais.Footnote 49 Following the Methuen Treaty of 1703, the majority of Brazilian gold then flowed through Portugal to England, to cover the deficit in the Anglo-Portuguese balance of trade.Footnote 50
According to Dean’s estimates, gold mining itself cleared around 4000 km2 of forest in Minas Gerais.Footnote 51 In the eighteenth century, the Minas Gerais gold frontier became one of the principal motors of the transatlantic slave trade, with close to 2 million African workers transported to Brazil during this period.Footnote 52 This primary mining frontier thus generated a market of workers and administrators that were fed via the opening of secondary livestock and agricultural frontiers. These secondary frontiers, in turn, possibly cleared another 26,000 km2 of forest in the eighteenth century (see Figure 2).Footnote 53
Figure 2. Deforestation, commodity frontiers and Indigenous resistance in the southern Atlantic Rainforest, 1500–1800.
Note: The references for this figure can be found in footnotes 47 to 62.
While Portuguese control over Indigenous labour played a crucial role in opening the gold-mining frontier in the eastern Atlantic Rainforest, the successful resistance of Guaraní and Macro-Jê groups played an equally important role in slowing down deforestation in the Paraná Forest. Although Spanish colonists from Paraguay initially colonised the Guayra region within the forest in the sixteenth century, constant uprisings by their Indigenous subjects made it impossible for them to consolidate their settlements.Footnote 54 The Paraguayan governor Hernando Arias de Saavedra lamented in 1607 that the Indigenous populations of Guayra ‘only served how and when they want to because the Spanish don’t have the force to be able to conquer nor subject them’.Footnote 55 This instability left the Spanish vulnerable to the Portuguese bandeiras, obliging them to withdraw from the forest in the mid-seventeenth century.Footnote 56
Acting both autonomously and in alliance with Spanish Jesuit missionaries, Guaraní and Macro-Jê Indigenous groups then defeated numerous bandeiras themselves, creating an independent buffer zone in the forest outside of Iberian control. Autonomous Indigenous forces beat back the Portuguese slave raiders in the late 1620s, forcing the Bandeirantes to redirect their attacks against the Spanish Jesuit missions amongst the Guaraní.Footnote 57 After retreating to the forest edge, the Guaraní-Jesuit forces inflicted a decisive defeat on the Portuguese at the battle of Mbororé on the Uruguay River in 1641, imposing a limit on Portuguese raids.Footnote 58
In the eighteenth century, autonomous Indigenous groups employed violent ambushes and psychological warfare (such as the case mentioned by Dobrizhoffer in the introduction to this article) to obstruct Spanish attempts to recolonise the forest.Footnote 59 By 1735, attacks by independent Indigenous forces in the Paraná Forest ‘had begun to gravely endanger the government of Paraguay’, according to the Spanish colonist Pedro Cavallero Villasanti.Footnote 60 Occasional retaliations occurred, such as in 1825, when the newly independent Paraguayan government burnt a large stand of coconut trees to flush out hostile Mbayá (Kadiwéu) Indigenous forces on the far-western edge of the Paraná Forest.Footnote 61 However, the overall deforestation rate in nineteenth-century Paraguay was minimal compared to neighbouring Brazil, where the coffee frontier was advancing rapidly.Footnote 62 Much of the Paraná Forest was only cleared in the twentieth century, culminating with the expansion of the Brazilian soy frontier (and Brazilian migration) across the border into Paraguay from the 1970s onwards. Notably, this soy frontier detoured around the Argentinean province of Misiones, where a significant remnant of the Paraná Forest still survives, overlaying a key area of historical Guaraní and Macro-Jê resistance.Footnote 63
In areas where Indigenous resistance was defeated, coerced Indigenous labour aided colonisation. The enormous concentration of gold in the eastern portion of the Atlantic forest undoubtedly acted as a major pull factor for Portuguese colonists, and contributed to heavier colonial-era deforestation in this region. However, these colonists were only able to discover and extract this gold through their appropriation of Indigenous knowledge systems, following a pattern seen across the colonial world.Footnote 64 The colonisation of the South American interior was made possible by the enslavement of Guaraní and other Indigenous women, in a violent strategy that appropriated their bodies (both through manual labour and sexual assault) and their knowledge. Paulista colonists appropriated the agricultural knowledge of Tupinambá and Guaraní women to plant manioc and maize along their expedition routes, while also exploiting Indigenous canoeing skills to negotiate the biome’s extensive network of waterfall-strewn rivers.Footnote 65
In the western Atlantic Rainforest, where Guaraní groups were able to maintain their autonomy, the Guaraní conception of the forest, not just as a bundle of resources but as a ‘body-territory’, has contributed to the biome’s conservation. A key element of Guaraní cosmology is that human relationships with other species and the surrounding environment are overlapping and enmeshed, forming the concept of the body-territory. As Sandra Benites notes, the body of the earth itself is the body of a woman, Nhandecy eté, the First Mother. Nhandecy eté showed Guaraní women the routes through the forest, and these women continue to tread lightly on her body as they follow these ancestral pathways today.Footnote 66 The strength of such cosmological knowledge lies in its resilience despite colonial transformations, and it continues to provide a framework for Guaraní people to defend the areas of forest where they live today. The collective character of this knowledge also contributes to its resilience, as seen in the potirõ/nhopytyvõ (roughly translatable as ‘mutual help’) regimes among various Guaraní-speaking groups. Potirõ/nhopytyvõ systems come in a variety of forms, including collective assistance amongst community members to construct houses in Guaraní villages, coupled with regulations concerning the sharing of food during group celebrations. These collectivist practices continue to form a barrier against the development of more individualistic, capitalist conceptions of social and ecological relations in many Guaraní communities.Footnote 67
Explanations for deforestation patterns in the Atlantic Rainforest that focus exclusively on the geographical distribution of resources or the motivations of European colonists can thus lose sight of the historical impact of Indigenous labour and Indigenous resistance, along with the importance of Indigenous cosmologies in maintaining this resistance. Even in Minas Gerais, at the heart of the gold-mining frontier, Indigenous resistance shaped the course of deforestation on a local scale. To the east of the principal gold-mining region, a number of predominantly Macro-Jê-speaking Indigenous groups, known in the colonial period as the Botocudo/Aimoré, Coropó, Coroado, Goitacá, and Puri, vigorously resisted incursions by mining prospectors throughout the eighteenth century.Footnote 68 In 1700, construction of a road began between Minas Gerais and the Espirito Santo coast, cutting through the territory of these Indigenous peoples, the ancestors of the contemporary Krenak and Maxakali groups, amongst others. In 1702, the Portuguese crown cancelled the roadbuilding project, and instead prohibited colonists from circulating through this region, in order to direct gold exports through Rio de Janeiro, where they could be more effectively taxed.Footnote 69 This royal decision was influenced indirectly by Macro-Jê resistance, which prevented the development of any potential alternative gold routes and customs ports in Espirito Santo. Even after the Portuguese government switched tack and gave military support to the colonisation of the eastern forests after 1808, violent Macro-Jê resistance hindered settlement of the region into the late 1820s.Footnote 70 Geographical surveys from the 1880s indicate that this was still the least deforested part of Minas Gerais over half a century later.Footnote 71
One rebuttal to the argument that Macro-Jê resistance influenced the direction of the trade routes from Minas Gerais to the coast would be that Espirito Santo had no port that could have rivalled Rio de Janeiro, with Rio’s prominence built on its pre-existing sugar plantations. However, even here it is worth remembering that in the mid-sixteenth century, when attempts were made to establish a captaincy with sugar plantations in Espirito Santo, Indigenous forces sacked the colony, hampering its subsequent development.Footnote 72 Portuguese colonists subsequently shifted their attention to Guanabara Bay, where they inflicted a famous defeat on allied French-Tupinambá forces and founded Rio de Janeiro in 1567.Footnote 73 The capacity at a local level to overcome Indigenous resistance directly affected where Portuguese colonists established their initial settlements and transport routes, but the battles that the colonists lost, the cities they failed to found, and the roads they failed to build have all faded from historical memory.
Sustained waves of Indigenous resistance, in interaction with environmental constraints, economic imperatives, and the decisions of colonists, have therefore shaped the direction of even the most aggressive deforestation fronts in the Atlantic Rainforest. As a result, the majority of the Paraná Forest, as well as a large section of forest in the Krenak and Maxakali homelands of eastern Minas Gerais, survived well into the twentieth century. However, the majority of these remaining forests have been swallowed by expanding commodity frontiers over the last seventy years, during the Great Acceleration.Footnote 74 Overall, Indigenous resistance has impeded and redirected, rather than halted, deforestation in the Atlantic Rainforest. The global economic forces driving this process at times seem unstoppable, but Indigenous campaigns to protect the surviving remnants of the Atlantic Rainforest and communicate the cosmological importance of this biome to the wider world continue today.Footnote 75
Deforestation and Indigenous resistance in the Amazon
In contrast to the Portuguese gold frontier in the Atlantic Rainforest, Spanish attempts to extract gold from the Amazon Rainforest were a spectacular failure. These expeditions are often dismissed as quixotic pursuits, as epitomised by the legend of El Dorado, in which the gold of the Amazon was said to emanate from a single city deep in the jungle.Footnote 76 However, as the modern-day gold rush in the Amazon attests, there are significant gold deposits within this biome, even if they are not concentrated in a single mythical site. Brazil, for example, is exporting as much gold today as it was during the eighteenth-century gold rush, with much of this new gold coming from the Brazilian Amazon.Footnote 77 This raises two questions: why were the Spanish unable to take control of the decentralised Indigenous networks that distributed Amazonian gold in the sixteenth century; and why was this pattern so different to that seen in the Portuguese section of the Atlantic Rainforest? To address these questions, we focus on examples of Indigenous resistance to Spanish colonisation in the Amazon Rainforest, then compare them with Indigenous resistance against Portuguese and Brazilian colonisation in the Amazon.
It is crucial here to distinguish between Tahuantinsuyu, the Inca Empire, which the Spanish conquered in four decades, and the neighbouring non-state societies in the Amazon basin, which successfully resisted Spanish incursions for centuries. While there was also fierce opposition to Spanish invasion in the Andean highlands, Francisco Pizarro’s capture of Tahuantinsuyu’s ruler, Atahualpa, in 1532, gave the invaders a critical edge in the conflict. Pizarro took advantage of Atahaulpa’s position of authority to seize control of the Inca state, using the Indigenous polity’s hierarchical structure against it.Footnote 78 From the 1570s onwards, while there was extensive Indigenous resistance in the Andean highlands, it predominantly took the form of class conflict within the Spanish colonial state, and lies outside the scope of this study.Footnote 79 The Spanish exploration of Inca silver-mining areas on the Andean altiplano led to the discovery of Potosí in 1545, with the gargantuan flow of silver from this site allowing the Hapsburg crown to pay its mounting debts to first south German, then Genoese bankers.Footnote 80
Simultaneously, the Spanish ventured into the Amazon Rainforest to discover the sources of the gold that Indigenous intermediaries transported to the Andes.Footnote 81 Here, however, the Spanish encountered a mosaic of non-state Indigenous peoples who had successfully resisted Inca expansion in the fifteenth and early sixteenth centuries. In the sixteenth century, the Amazon Rainforest contained a plethora of linguistic groups, including the Tupian, Macro-Jê, Arawak, and Carib language families, amongst others.Footnote 82 There has been considerable debate about the possible presence of ‘chiefdoms’ with characteristics comparable to ‘small states’ along the Amazon basin’s central and lower riverways on the eve of the European invasion.Footnote 83 Even so, societies in the Amazon Rainforest that were potentially more stratified, such as the Omagua, were decimated during the first two centuries of contact with the Spanish and Portuguese, with decentralised Indigenous groups in the biome mounting the most successful resistance to the invaders.Footnote 84
Faced with Spanish aggression, non-state Indigenous groups in the western Amazon engaged in both violent resistance and flight from mining sites, preventing the Spanish from gaining full control of their labour supply or transport networks. The Spanish had to abandon their gold mines in the Quijos valley of modern-day eastern Ecuador in the 1570s, after local Indigenous groups razed the settlements of Archidona and Avila.Footnote 85 In the late sixteenth and early eighteenth centuries, Western Ava Guaraní and Yuracaré groups in the forested Yungas on the southern edge of the Amazon threatened the operation of the Potosí silver mines and blocked access to the lowlands.Footnote 86
At the Zamora gold mines to the south of the Quijos Valley, continuous attacks by Jivaro/Shuar groups between 1582 and 1616 expelled the Spanish.Footnote 87 After these military failures, the Spanish crown turned to Jesuit missionaries to regain control of the Zamora gold-mining frontier. An unpublished 1689 letter by the Jesuit missionary Francisco Viva laid out plans for a new incursion, noting that:
Close to our Missions there is the nation of the Jivaros, in revolt for the last 90 years, in their hills is so much gold, that they say without doubt here and in Spain, that in all the Indies they have not discovered a richer land … His Majesty has continuously rushed out decrees to pacify the aforementioned Jivaros, and even though many have attempted to conquer them, none have succeeded.Footnote 88
However, the Jesuits of the Maynas missions were also unable to suppress the Jivaros, with these groups continuing to defend their autonomy in subsequent centuries.Footnote 89 The gold supplies of this region were not a figment of Viva’s imagination, with contemporary Shuar/Jivaro groups in Ecuador engaged in an ongoing campaign against the modern Mirador and San Carlos Panantza gold and copper mining projects. The Mirador mine alone holds an estimated 110,000 kilograms in gold reserves, roughly equivalent to the 112,000 kilograms of documented gold extracted from Minas Gerais between 1741 and 1750, at the height of the Brazilian gold rush.Footnote 90 Colonial Spanish miners could have accessed a significant portion of the east Ecuadorian gold reserves via both alluvial and hard-rock mining, had they not been obstructed by Shuar/Jivaro resistance. This gold frontier, in turn, could have opened a significant new deforestation frontier in the western Amazon, just as the Brazilian gold frontier drove a wedge into the Atlantic Rainforest that was soon filled by food crops, livestock, and coffee plantations. However, with Indigenous resistance preventing the Spanish from consolidating political control over the western Amazon, various plans to establish new cacao, coca, and coffee plantations failed.Footnote 91 This pattern continued into the eighteenth century, with Ashaninka groups achieving a wholesale expulsion of Franciscan missionaries from the central forest zone of what is now Peru in the 1740s, as part of a general trend whereby Spanish frontiers on the edge of the Amazon actually retracted in many areas when compared with earlier Inca boundaries.Footnote 92
As Figure 3 shows, by 1800 a clear distinction had emerged between the deforestation rates within the Spanish and Portuguese territories of South America. In both the western Atlantic Rainforest and the western Amazon, the zones of successful Indigenous resistance against the Spanish were still heavily forested. In Brazil, by contrast, significant tracts of the Atlantic Rainforest had already been devastated, and Portuguese colonists had used the techniques they had acquired in this forest as a springboard to attack the Amazon. On the southern and eastern edges of the forest, Portuguese commodity frontiers were already causing significant deforestation, such as the gold mines of Mato Grosso and the cotton frontier of Maranhão.Footnote 93 This divergence has continued into the twenty-first century, with Mato Grosso and Maranhão forming the two spearheads of the Brazilian deforestation arc in the Amazon.Footnote 94 The rate of Amazonian deforestation in the former Spanish colonies, by comparison, while still worryingly high, has been consistently lower than that in Brazil. By 2013, Brazil had cleared 17.63% of its share of the Amazon Rainforest, more than double the accumulated Amazonian deforestation rate across the Spanish-speaking countries of 7.98%.Footnote 95
Figure 3. Examples of Indigenous resistance in the Amazon and Atlantic Rainforests before 1800.
Note: The references for this figure can be found in footnotes 47 to 115.
We argue that the contrasting Iberian encounters with non-state Indigenous groups made a significant contribution to this long-term divergence in deforestation levels. The Spanish Empire’s South American labour management systems, such as the encomiendas, the mita, and the early reducciones during the period of the Toledo reforms, were adapted to the organisation of labour within the former Inca state.Footnote 96 The Spanish struggled, in contrast, when confronted with mobile Indigenous forest-dwelling societies that continuously resisted being corralled into work sites. Groups that were more decentralised and prone to internal conflict, such as the Shuar/Jivaro, the Ashaninka, the Kalina/Caribs, and the Guaraní, were able to resist the Spanish more effectively.
The Portuguese better adapted to stateless Indigenous modes of movement, in particular Tupian agricultural and warfare strategies, to extend their sphere of influence in both the Atlantic and Amazon Rainforests. In both biomes, Portuguese colonists appropriated Tupian riverine navigation techniques, mobile styles of warfare, and linguistic networks, along with Indigenous women’s botanical knowledge and labour.Footnote 97 Crucially, these colonists learnt from Tupi women in the Atlantic Rainforest how to cultivate and process manioc, a root crop that facilitated mobility.Footnote 98 The subsequent transfer of manioc to Africa assisted the Portuguese in establishing long-range slave raiding networks on both sides of the Atlantic.Footnote 99 Fernand Braudel noted the parallels between the mobile Bandeirantes in Brazil and the French fur trappers in Canada, who also advanced inland rapidly along Indigenous transport routes.Footnote 100 However, we would note that whereas the French lost Canada to Great Britain in 1762, the Portuguese successfully interlinked their inland Indigenous networks with coastal plantation slavery and an alliance with the British, a distinctive combination that cemented their presence in Brazil.Footnote 101
Nevertheless, along the southern tributaries of the Amazon, as in eastern Minas Gerais, noteworthy cases of resistance by a variety of Indigenous groups impeded Portuguese expansion, such as the Juruna on the Xingu river, the Sateré-Mawé on the Tapajós river, and the Mura on the Madeira River.Footnote 102 As with the Macro-Jê resistance in Minas Gerais and Espirito Santo, this Indigenous resistance in the Amazon ensured that while the Portuguese advanced inland more rapidly than the Spanish, significant areas of forest still escaped their grasp.
In the seventeenth and eighteenth centuries, the Spanish finally found a way to break through the lines of resistance on the western edge of the Amazon and incorporate a number of lowland stateless Indigenous groups via the Jesuit mission system.Footnote 103 However, the constant threat of Indigenous rebellion and flight also forced the missionaries to concentrate the majority of their reducciones outside the forest, at sites such as the Orinoco llanos (plains), the llanos of Moxos, and the Pampas transition zone on the southern edge of the Atlantic Rainforest.Footnote 104 Spanish Jesuit missions within the forest, such as the Maynas missions in the western Amazon and the early Guaraní missions in Guayra, were notably less successful, with mobile Indigenous populations constantly abandoning the reducciones and returning to their previous ways of life.Footnote 105 The differences between Spanish and Portuguese forest colonisation strategies began to diminish in the eighteenth century, with the decline of the bandeiras, the expulsion of the Jesuits, the attempted rationalisation of colonial policy, and the centralisation of power in both states.Footnote 106 Nevertheless, before the mid-eighteenth century, contrasting Iberian interactions with non-state Indigenous societies contributed to the eastern Portuguese sections of both the Amazon and Atlantic Rainforests becoming more thoroughly integrated into the expanding capitalist world-economy.Footnote 107 Successful Indigenous resistance to Spanish advances delayed this process in the western Amazon and Atlantic rainforests, with a consequent divergence in deforestation rates.
One counterargument here would be that geographical features, such as the Andes mountain range, constituted the principal barrier to economic development and deforestation in the western Amazon. However, in areas where the Spanish were able to utilise the pre-existing Inca road system, the geographical impediment caused by the Andes was reduced significantly, as the rapid Spanish takeover of the Inca Empire attests.Footnote 108 It was only when moving into the Amazonian lowlands, beyond the reach of Incan infrastructure, that the advance of Spanish colonists slowed down. If either the Incas before 1530 or the Spanish after this date had been able to suppress lowland autonomous Indigenous groups and extend the Inca transport network, Spanish colonists would have been better placed to overcome the geographic obstacles of the western Amazon and exploit this region economically. In a significant boost for the survival prospects of the western Amazonian forests, both empires failed in the face of sustained, decentralised Indigenous resistance.
Geographic determinism alone also fails to explain the low deforestation levels in the Guianas, with this region being accessible from the Atlantic Ocean and criss-crossed by Indigenous gold distribution networks that attracted attention from various European states in the early colonial period.Footnote 109 Here, the violent conflicts and geopolitical manoeuvring between multiple colonial forces and local, decentralised Indigenous groups contributed to subsequent territorial fragmentation, with this section of the Amazon Rainforest today split between five nation-states (Venezuela, Guyana, Suriname, France, and Brazil).Footnote 110 Tellingly, the English navigator and privateer Walter Raleigh imagined a gold-producing offshoot of the Inca state in the Guianas, with the illusory ‘imperial city’ of Manoa as its capital. As with the Inca Empire before it, Raleigh claimed that this prosperous realm was ripe for European conquest.Footnote 111
The reality in the Guianas, however, was entirely different, with loosely organised Arawak and Carib (Kalina) groups controlling the distribution networks radiating out from gold-producing sites. Throughout the seventeenth and eighteenth centuries, these groups pursued shifting alliances with European colonists attracted by the promise of gold, playing Spanish, Dutch, English, French, and Portuguese forces off against each other. These colonial powers were able to establish footholds in the region and conduct slave raids into the interior, but none of them was able to assert complete territorial control or gain sustained access to local gold flows.Footnote 112 Indigenous resistance was directly acknowledged by colonists in the region as a major economic and military impediment: as a 1599 Dutch report noted, the ‘Spanish also said that upwards there was much gold but that they dared not come there on account of war with the aforesaid Charibus [Caribs]’.Footnote 113
In the seventeenth and eighteenth centuries, African workers from plantations on the Atlantic Amazonian coast escaped into the forest, forming maroon communities (in British, Dutch, and French Guiana) and quilombos (in Portuguese Brazil) that also impeded inland colonisation. Maroon groups such as the Saramaka in Suriname, who incorporated both African and Indigenous members, formed an ongoing obstacle to the expansion of the Dutch plantation system, and continue to campaign against logging and gold mining in their territories today.Footnote 114 While the environmental impact of quilombo and maroon groups requires more detailed research, as a preliminary observation, it is worthy of note that the Guiana Shield region still hosts a large number of Indigenous, maroon, and quilombo communities, while also having the lowest deforestation rate in the Amazon.Footnote 115
Although Brazil has been at the historical forefront of deforestation in South America, contemporary Indigenous organisations in the country have attempted to challenge this trend. Following the end of the military dictatorship (1964–85), Indigenous campaigns led to the introduction of a relatively progressive constitution in 1988, which assisted in the creation of new Indigenous territories.Footnote 116 The consolidation and expansion of these territories in the 1990s and 2000s contributed significantly to the 70% drop in the Amazonian deforestation rate in Brazil during President Lula Inácio da Silva’s first two terms in office (2003–10).Footnote 117 These campaigns in South America have occurred within the broader context of a global Indigenous rights movement that in recent decades has emphasised the role of Indigenous forest dwellers in combating climate change.Footnote 118 Increasingly, Indigenous groups from South America and other tropical forest regions, such as the Dayak people of Borneo in Indonesia, have used a common conceptual framework to defend their land rights and advocate for forest conservation.Footnote 119
The experience of the Paiter Suruí in the Brazilian Amazon during the last century offers another insight into this dramatic recent history. During the rubber boom in the early twentieth century, the Paiter remained hostile to outside interlopers, a position which they maintained into the 1960s, when they killed multiple agricultural colonists invading their territory. In 1969, they made official contact with Brazilian government representatives, inaugurating a traumatic period of transformation in which around half the community lost their lives to diseases, while the majority of those who survived converted to evangelical Protestantism.Footnote 120 Important traditions, such as the Mapimaí, a festival in which two halves of a clan, one living in the forest and the other in the village, would exchange gifts, are now only rarely performed. Despite these changes in their cosmology, the Paiter have maintained a strong connection with the rainforest, with clan leaders reorganising their way of life to protect the forest from the advancing cattle and soy frontiers. Since the 1970s, the Paiter have established their settlements along the borders of their designated territory, to prevent the entrance of outside colonists. They have also engaged in agroforestry projects, such as Brazil nut harvesting, to provide an alternative income source to more destructive practices such as logging, mining, or ranching.Footnote 121
The success of groups such as the Paiter in slowing down recent deforestation in Brazil is all the more notable given the previous political and geographical factors that have contributed to Brazil’s higher historical deforestation rate. Nevertheless, major challenges remain. Across the Amazon, contemporary miners are exploiting the gold reserves that eluded the Spanish in the sixteenth century. Both small-scale, illegal mining and large-scale, legal mining are opening new deforestation frontiers and exposing Indigenous communities to mercury and cyanide poisoning. From the lands of the Shuar in Ecuador to those of the Yanomami and Paiter in Brazil, the 500-year Indigenous struggle against mining, ranching, and other forms of exploitative commodity extraction in the Amazon continues.Footnote 122
Final considerations
Tropical deforestation is one of the ultimate longue durée historical processes, with incremental factors building up over centuries, before sudden political, economic, environmental, and technological shifts drive rapid bursts of destruction. This process has been marked by long-range path-dependencies in which the outcomes of past struggles between colonisers and Indigenous populations have acted like a ratchet, either blocking or releasing the forces of ecological destruction. In the current epoch, the runaway effects of climate change and the Great Acceleration threaten to wipe out the areas of rainforest that earlier Indigenous resistance helped preserve. The Amazon Rainforest is a case in point, with a network of natural white sand savanna clearings to the north of the Amazon River identified as a fire-vulnerable ‘Achilles heel’ that could engulf the surrounding moist evergreen forests as temperatures increase.Footnote 123 Combined with the advance of the Brazilian deforestation arc from the south and east, this could soon dry out the Amazon irreversibly, converting most of the biome into open savanna.Footnote 124 To avoid reaching these tipping points, ongoing campaigns to defend and consolidate Indigenous forest territories are essential, with these campaigns potentially having a similar ecological ratchet effect in the future as Indigenous resistance had in the past. The preservation and transmission of Indigenous cosmologies, offering perspectives on human relations with nature different from the views prevailing within the capitalist world-system, will continue to play an important role in these ongoing struggles.
Global histories of capitalism often focus on European powers or, less commonly, incorporate analysis of Indigenous states such as the Inca Empire. Addressing a notable blind spot in world-systems analysis, this article has shown that non-state Indigenous populations also played a major role in constricting and shaping commodity circulation and extractive frontiers in South America. While specific local circumstances must always be taken into account, this analytical framework could also be applied to forms of Indigenous political resistance in the tropical rainforests of Africa and Asia. To analyse this phenomenon, researchers must look at political actions beyond the bounds of the state, honing in on the silences, ellipses, and failed projects hidden in colonial archives.
We have cross-referenced these archival documents with Guaraní and Paiter oral histories and cosmological insights, alongside contemporary deforestation data, so that new knowledge can emerge from the dialogue between these contrasting ontologies. This methodological approach reveals that while commodity frontiers on the edge of the expanding capitalist world-economy in South America have been a powerful force for environmental transformation, they have also encountered frequent obstacles in the form of both direct and indirect Indigenous opposition. In the Atlantic Rainforest, resistance by Macro-Jê and Guaraní-speaking groups prevented Spanish and Portuguese colonists from taking political control of both the Paraná Forest and a significant swathe of the forest in eastern Minas Gerais until the nineteenth century, with major deforestation not occurring in these areas until the twentieth century. Direct violent opposition, labour withdrawal, and flight, underpinned by a cosmological framework opposing colonisation, all played a role in this process.
While the Portuguese accessed the gold of Minas Gerais by appropriating the knowledge and labour of Tupinambá, Guaraní, and Macro-Jê groups, the Spanish failed to gain control of Indigenous gold distribution networks in the Amazon. The Spanish state’s early structural focus on labour management within the former Inca Empire, and its corresponding inability to exert sustained control over neighbouring, mobile, non-state Indigenous groups in the western Amazon, was a significant factor in this failure. Spanish, English, Dutch, French, and Portuguese colonists were all unable to capture the various gold-producing sites in the Guianas, where Indigenous communities impeded the economic exploitation of the forest hinterlands. Even on the southeast edge of the Brazilian Amazon, where the deforestation front advanced most rapidly, the resistance of groups such as the Paiter directly resulted in the conservation of critical forest corridors during the last half century. In all these locations, as elsewhere in the American tropics, the long-term interplay between Indigenous actions and the protection offered by the forest itself has hobbled numerous European economic projects.
Episodes of resistance by stateless Indigenous peoples have often been analysed in isolation, rather than as an ongoing, structural component of world history. To address this issue, historians must take Indigenous cosmologies and actions seriously as forces with global ecological and political consequences. The fact that Indigenous populations have restricted and shaped commodity frontiers and capital flows in the past will make them key agents of change and land care as we seek to charter a more just course through the current planetary ecological crisis. As this article has argued, the modern-day capacity of Indigenous tropical forest territories to absorb carbon dioxide and harbour biodiversity, rather than being a product of external economic or geographic factors alone, is directly linked to local histories of political resistance.
Acknowledgements
The authors would like to acknowledge the assistance and participation of Mbyá Guaraní, Nhandeva Guaraní, and Paiter Suruí communities in the creation of this article. In particular, we would like to thank the interviewees Carlos Papá from Aldeia Rio Silveira, along with Marcos Tupã and Altino Wera Mirim from Aldeia Jaexaa Porã, as well as Pamatoa, Cristine Takuá, Anai Vera, and Timóteo Verá Tupã Popygua.
Financial support
Part of the archival data and research trips were conducted with the financial support of the Max Planck Society. Freg J. Stokes, Laura Furquim, Jürgen Renn, Ricarda Winkelmann, and Patrick Roberts are supported by the Max Planck Society.
Competing interests
The authors declare none.
Freg J. Stokes is a postdoctoral researcher at the Max Planck University of Geoanthropology, specialising in history and cartography.
Sandra Benites is a Nhandeva Guaraní anthropologist working in Brazil, with a focus on education and Guaraní women’s history. She is currently serving as Director of Visual Art at Funarte.
Anita Ekman is a researcher, curator, and artist from Brazil who investigates precolonial art and rainforest history. She is currently working as a project coordinator with the Goethe Institute.
Uraan Anderson Suruí is a Paiter Suruí linguist and president of COOPSUR, a Suruí cooperative in the Sete de Setembro Indigenous Territory, Brazil.
Laura Furquim is a visiting researcher at the Max Planck Institute of Geoanthropology, working in the Amazon with Historical Ecology and Archaeobotany.
Ricarda Winkelmann is a climate scientist and the Director of the Department of Integrative Earth System Science at the Max Planck Institute of Geanthropology.
Jürgen Renn is a historian of science and the Director of the Department of Structural Changes of the Technosphere at the Max Planck Institute of Geoanthropology.
Patrick Roberts is an archaeological scientist and the Director of the Department of Coevolution of Land Use and Urbanisation at the Max Planck Institute of Geoanthropology.
Open access
