SciELO - Scientific Electronic Library Online

vol.2 special edition author indexsubject indexarticles search
Home Pagealphabetic serial listing  

Novos Estudos - CEBRAP

Print version ISSN 0101-3300

Novos estud. - CEBRAP vol.2 São Paulo  2006


The dynamics of slavery in Brazil: Resistance, the slave trade and manumission in the 17th to 19th centuries1


A dinâmica da escravidão no Brasil: Resistência, tráfico negreiro e alforrias, séculos XVII a XIX



Rafael de Bivar Marquese

Translated by Anthony Doyle
Translation from Novos Estudos - CEBRAP, São Paulo, n.74, p.107-123, Mar. 2006.




O artigo examina as relações entre o tráfico negreiro transatlântico para o Brasil, os padrões de alforria e a criação de oportunidades para a resistência escrava coletiva (formação de quilombos e revoltas em larga escala), do final do século XVII à primeira metade do século XIX. Valendo-se das proposições teóricas de Patterson e Kopytoff, sugere uma interpretação para o sentido sistêmico do escravismo brasileiro na longa duração, sem dissociar a condição escrava da condição liberta, nem o tráfico das manumissões.

Palavras-chave: escravidão; história do Brasil; tráfico negreiro; alforrias; resistência escrava.


The article examines the relationships between the Brazil-bound transatlantic slave trade, manumission patterns and the creation of opportunities for collective slave resistance (formation of Maroon slave communities and large revolts), from the end of the 17th Century to the first half of the 19th Century. Based on the theoretical propositions of Patterson and Kopytoff, it suggests an interpretation for the Brazilian slave system over the long term without dissociating the condition of the slave from that of the freedman, nor the slave trade from the manumissions.

Keywords: slavery; Brazilian history; transatlantic slave trade; manumissions; slave resistance.



The Palmares enigma

The Palmares War was one of the most notable episodes of slave resistance in the history of the New World. Though population estimates by coeval sources and historians vary wildly – from as few as 6 thousand to some 30 thousand people -, there is no doubt that the Palmares communities, given their sheer territorial extension and the number of fugitive slaves they absorbed, constituted the largest 'Quilombo', or Maroon slave community, in the history of Portuguese America. While Palmares dates back to the beginning of the 17th Century, its formation as a large runaway slave centre only took hold with the Dutch invasion of Pernambuco, when droves of slaves took advantage of the military chaos to flee to the south of the Captaincy.  The rebel communities that formed at this time held out against various incursions by the West India Company and, after the expulsion of the Dutch, also staved off various raids by Luso-Brazilian troops.   

In virtue of the threat they represented to Portuguese colonial order in the Americas, during the 1670s and 80s the metropolitan authorities viewed the Africans, Creoles and their descendents based in Palmares as “coloured Dutch”. Their military defeat would only be secured halfway through the next decade, after a century of struggle against two of the greatest European colonial powers of the modern world. Prior to the Saint-Dominigue slave rebellion (1791-1804) and the major abolitionist revolts of the British Caribbean in the first third of the 19th Century, the Palmares episode was only equalled by the First Maroon War in Jamaica (1655-1739) and the Saramaca War in Suriname (1685-1762). In these two cases, however, the runaways managed to defeat their armed oppressors, thus forcing the authorities and slave owners alike to recognise their liberty2.     

The history of the defeat of the Maroon Kingdom of Palmares gave rise to an enigma that has long held the attention of specialists in Brazilian slavery: why weren't there other Palmares in the history of Brazil? The point is an important one, as Maroon activity intensified throughout the 18th Century with the expansion of the transatlantic slave trade and the formation of mining centres in the hinterlands, and assumed different modalities from the north to the south of Portuguese America. Beyond the countless runaway slave communities, of varying dimensions and lifespans, early 19th-century Brazil saw the emergence of another form of collective slave resistance, one long installed in the British Caribbean: the spate of African revolts that flared up the Recôncavo Baiano from 1807 to 18353.   

The explanation historians generally give for the enigma hinges upon a change in Portuguese slave law. After Palmares, they say, the functions of the slavehunter (the capitão-do-mato, or 'bush captain', responsible for tracking down runaway slaves in various regions of Portuguese America) underwent progressive specification, as did the definition, in legal terms, of what constituted a Maroon. The argument goes that the institutionalisation of the figure of the slavehunter and the defining of a Maroon colony as any settlement comprising as few as a handful of runaway slaves snipped the formation of any future Palmares in the bud4. However, I believe there may be another explanation, one which – without refuting the arguments of those who study the subject – lies in the configuration Brazilian slavery acquired as early as the end of the 17th Century.     

The aim of this essay is precisely to understand why there were no other Palmares in the history of Brazil? In order to pursue this goal I have chosen to focus on the relations between the transatlantic slave trade, manumission and the creation of opportunities for collective slave resistance (such as the formation of Maroon communities and large-scale revolts) from the end of the 17th Century to the first half of the 19th. The idea is that events like Palmares, the Jamaican Maroon War and the Saramaca campaign were directly related to the configuration of a particular type of slave system, which I will call “plantation slavery”. In this system, economic production was concentrated in a singe product and the social fabric was marked by a demographic unbalance between free whites and black slaves, mostly of African origin, with few opportunities for manumission and high levels of master absenteeism.    

This kind of slave system, so typical of the British and French Caribbean colonies of the 18th Century, and whose basic characteristics achieved only partial development in Portuguese America in the first half of the 17th Century, was closed out in the subsequent two centuries of Brazilian history. The commencement of the mining activities served to accentuate this shift in the base nature of Brazilian slavery. The institution became socially and spatially diffuse, as slave ownership spread throughout the social fabric and highly complex ethnic and cultural hierarchies took hold. The old plantation lands, like the Zona da Mata in Pernambuco and the Recôncavo Baiano, witnessed these transformations in equal measure, though slave labour continued to propel the sugar production5.    

From the end of the 17th Century, the Brazilian slave system came to depend upon a fine balance between a voluminous transatlantic slave trade and a constant number of manumissions. This equation meant that slave traffic could be intensified, injecting larger numbers of enslaved Africans, without endangering the social order. In the wake of the Palmares defeat, a slave revolt or Maroon community's chances of success fell substantially in Brazil. It was not by chance that the Portuguese colonial authorities and Brazilian Imperial representatives, with the exception of one brief episode in the 1670s, still during the Palmares War, refused to negotiate with rebel slaves or large Maroon communities in Brazil. This policy, which reflected the relationships of power between master and slave, contrasts with the approach adopted by the British and Dutch, who were forced to recognise Maroon and Saramaca victories on the battlefield in their peace treaties. 

It is important to point out here that historians have been aware of the close relationship in Brazilian history between the heavy transatlantic slave trade and the high rates of manumission6 for at least the last thirty years. What I believe is missing, however, is a more substantial theoretical framing of this connection that relates it with the limited prospects for successful collective slave resistance in Brazil.  

Drawing upon the available studies, I tried to re-read the results in the light of the theoretical propositions of Orlando Patterson and Igor Kopytoff, who do not separate the experience of the slave from that of the freedman; as both face slavery, the slave society and manumission as parts of the same institutional process. As Kopytoff suggests:

slavery ought not be defined as a status, but as a process of transformation of a status that could take an entire lifetime and even extend into subsequent generations. The slave begins as a social outsider and passes through a process whereby he becomes an insider. An individual stripped of his or her prior social identity is placed on the margins of the new social group, which confers a new social identity. The outsidedness is therefore sociological rather than ethnic7.

With this proposition as a base, I will attempt to suggest an interpretive schema for the systemic sense of Brazilian slavery over the long term, without dissociating the condition of slave from that of the freed slave or the slave trade from manumission. As in all essays, the high level of generalization invites a certain risk, exacerbated by the fact that this systemic sense was by no means clear at the time. Awareness of the institutional process of Brazilian slavery only dawned at the beginning of the 19th Century, more specifically within the context of independence, both for the foreign travellers then roaming Brazilian territory and, more importantly, for the architects of the Empire of Brazil. This is my final destination. Put in other terms, what I hope to show is that the perception of the historical colonial experience, which combined the slave trade and manumission, played an important role in defining the future of slavery within the framework of the Brazilian nation state.      


Plantation slavery

In the centuries that followed the fall of the Roman Empire, slavery did not entirely disappear from Western and Mediterranean Europe. However, throughout the Early Middle Ages, slavery as a labour system did cease to exist in the European West, with the exception of the Mediterranean countries of the Iberian and Italic peninsulas. That said, even there, slavery was an exclusively urban institution throughout the 14th and 15th Centuries and of only limited importance to the economic set-up, as the large scale use of captives in agricultural production had all but disappeared in these regions. The reinvention of slave society, with the massive employment of slaves in agricultural graft, would only come in the second half of the 15th Century, as the Portuguese and Spanish introduced slave manpower to their sugar plantations on the western Atlantic islands (The Canaries, Madeira, San Tomé), and, later, with the colonisation of America in the 16th Century8.  

Encouraged by the experience accumulated in sugar production on the islands of Madeira and San Tomé, from the 1530s onwards the Portuguese Crown endeavoured to stimulate the construction of sugar facilities in Brazil. However, up to the 1570s the colonists faced serious difficulties in laying solid foundations for a sugar mill infrastructure along the coast, including problems in hiring manpower and a lack of capital to finance the construction of the mills. With these problems finally overcome, by tying Brazilian production to the northern European mercantile centres and forming a slave trade supply line between Africa and Brazil, the slave-based sugar industry of Portuguese America could finally and definitively take off, which it did between 1580 and 1620, when Brazil's accelerated production growth outstripped all other regions supplying to the European market.    

A few words are needed here on role the transatlantic African slave trade played in this boom in Brazilian sugar production. The manpower engaged in constructing the Brazilian sugar mills had been predominantly indigenous. Some of this indigenous workforce (recruited at Jesuit settlements on the coast) received paid wages, though the majority was submitted to slavery. The first African slaves were imported in the mid-16th Century, though these were mostly employed in specialist activities at the Brazilian mills, which meant they were more expensive than their Amerindian counterparts. In the latter half of the 16th Century an African slave cost around three times the price of an Indian slave. After 1560, with the various epidemics that hit the Brazilian coast (such as measles and smallpox), the Indian slaves began to die in such alarmingly high numbers that the workforce at the mills required constant replenishment. The following decade, under pressure from the Jesuits, the Crown partially proscribed the enslavement of Indians. At the same time, the Portuguese were fine-tuning the functioning of their transatlantic slave trade, especially with the conquest of Angola towards the end of the 17th Century. The figures for the slave trade bear this out: between 1576 and 1600 some 40 thousand enslaved Africans came ashore at Brazilian ports, during the first quarter of the following century (1601-1625), this volume more than tripled, jumping to 150 thousand Africans disembarking in Portuguese America as slaves, with most of these destined for the sugar cane plantations and sugar mills9.   

The success of slave-based sugar production in Portuguese America soon drew the attention of the other European colonial powers and by the end of the 16th Century there was growing involvement of British and Dutch businessmen in the sugar trade between Brazil and Europe. The Dutch invasions of Bahia (1624) and Pernambuco (1630) were largely motivated by the dynamism of the sugar economies in these captaincies. However, the members and shareholders of the Dutch West India Company (WIC), which commanded the invasions of the Brazilian sugar producing regions at the time, did not possess a thorough knowledge of the secrets behind the commodity's production, which boiled down to three basic necessities: techniques for processing sugar cane; techniques for administrating slaves and the ability to organize a transatlantic slave trade. The invaders soon realized the extent of the geo-economic relationship between Africa and the American slave-based plantations. Their Brazilian territories would be worth nothing unless they could also conquer the outposts that supplied the slaves on the other side of the Atlantic. Thus, in 1638, under the command of Maurício de Nassau, the WIC set about conquering the Portuguese-controlled port of São Jorge da Mina, followed in 1641 by the invasion of Angola10.

Dutch control of Pernambuco did not last long. In 1645, an uprising by Luso-Brazilian colonists led to Holland's definitive expulsion from Portuguese America in 1654. Prior to this, in 1648, Luso-Brazilian colonists from Rio de Janeiro claimed direct responsibility for the Dutch expulsion from Angola. Given the failure of its Brazilian and Angolan ventures, the WIC set sugar production to one side and began to focus on buying the product from regions not directly under its command. To this end, the Dutch traders sought to encourage sugar production in the British and French Caribbean. During Dutch occupation in Brazil in the second half of the 1640s, traders from Holland exported the know-how of the Brazilian sugar mills to the British colonists in Barbados and the French in Martinique and Guadalupe, along with slaves from the WIC outposts in the Gulf of Guinea. Sugar production with slave manpower on the British and French islands began to show notable growth in the 1660s, while traders from the homelands became directly involved in the transatlantic slave trade. By the beginning of the 18th Century, the physical and human landscape of the Caribbean had been completely altered, with the islands transformed into vast sugar plantations and the population now overwhelmingly black, and almost entirely enslaved11

In the course of the wars against the Dutch in the South Atlantic, the supply of slaves to the Brazilian mills suffered perceptible reductions. If, during the period 1601 to 1625, 150 thousand enslaved Africans had reached Portuguese America, the following quarter century saw this volume cut to a mere 50 thousand. One way or another, the Dutch invasion of Pernambuco and the subsequent wars against the Luso-Brazilian colonists opened up ample opportunities for resistance among the slaves who had disembarked in such large numbers during the first quarter of the 17th Century. Indeed, it was no accident that the cultural support that was so decisive in configuring the policy of the “Neo-African” kingdom of Palmares came from groups originally from central and southern Africa, precisely the zones in which the Portuguese slave traders had been most active from the end of the 16th Century12

The size and power of the Palmares Maroon can be explained not only by the context of Imperial conflict between the Portuguese and the Dutch, but also by the very demographics of the region of Pernambuco in which the sugar plantations were located. Any categorical assertions as to the composition of the colonial population prior to the 18th Century would be risky, but I believe I can safely affirm that when the Dutch invaded Pernambuco the black slaves outnumbered the white population – and even the “domesticated” Amerindians. Based on the few available sources, we can also wager that the free black population was relatively tiny. What we see here is therefore a demographic profile that would have been extremely propitious to the ignition of collective movements of slave resistance, as earlier events in the British Caribbean would clearly demonstrate.     

After the expulsion of the Dutch, the Luso-Brazilian troops turned their relentless combat to the Palmares settlers. However, the colonists biggest problem was on the economic front. The rapid establishment of a slave-operated sugar complex in the Antilles starting in 1650 soon took a heavy toll on the sugar industry of Portuguese America. Growth in British and French Caribbean production dumped prices on the European markets, while the demand for black labourers for the plantations in the Antilles pushed up the prices of slaves along the African coast. As if this wasn't enough, the Luso-Brazilian sugar barons had two further problems to contend with. First of all, the British and French market policy of stimulating sugar production in the Antilles in the second half of the 17th Century by offering monopolistic protection meant that these two European markets were all but closed to Brazilian sugar. Secondly, between 1640 and 1668, Portugal was locked in a sapping war of independence with Spain, just as the “Pepper Empire” began to crumble in the East. In the second half of the 17th Century, the New World territories became an economic prop for Portugal. Heavy taxation was placed on Brazilian sugar to help cover the Realm's expenses on diplomacy and defence.  

However, this heavy tax burden did not signal the demise of the sugar industry in Portuguese America. Despite the havoc wrought by the South Atlantic wars from the 1620s to the 1650s, the high taxation post-1650, stiff competition from the Antilles and the restricted access to certain European markets, the Luso-Brazilian sugar barons managed to keep sugar production stable. In order to achieve this, it was essential to consolidate the bipolar Atlantic system that connected Africa with the Brazilian ports, which came with the retaking of Angola in 1648. In the second half of the 17th Century some 360 thousand African slaves were shipped to Brazil. By ensuring the Brazilian sugar industry a constant supply of slaves at low cost, this system enabled sugar production in the Colony to weather such an adverse international market environment.    

There is evidence to suggest that, during this turbulent period for the sugar industry, manumission acquired fresh impetus. Of course, manumission had been occurring in the Colonies since the very beginning, but the fact that the first sequential documentation related to the practice dates only to the second half of the 17th Century could indicate that it only became widespread after this time. One of the first studies on the theme in the Brazilian historiography focused precisely on Bahia – a neighbour to Pernambuco, the heartland of the colonial sugar industry – and covered the years 1684 to 1745. The researcher Stuart Schwartz recorded and analysed a series of practices related to manumission that were later repeated throughout Portuguese America and the Brazilian Empire at different times and in different places. Among the more than one thousand manumission letters examined by the author, there was a constant ratio of two freed women to every manumitted male. Given the numeric predominance of men on the transatlantic slave ships and very composition of the slave-stock, Schwartz writes that “women received manumission in a far greater proportion than the statistical estimate”. Slaves born in Brazil – meaning the Creoles and, principally, the Mulattoes - were likewise statistically privileged, accounting for 69% of manumissions against 31% for African slaves. Finally, many of these manumissions were granted to children aged 14 or under. The tendency to manumit slave-women of childbearing age, concluded Schwartz, compromised the possibilities of self-sustainable reproduction among the Brazilian slave population, which accentuated still further the structural role of the transatlantic slave trade in restocking the workforce13.    



This demographic pattern was consolidated and geographically expanded with the discovery of gold reserves at the turn of the 17th into the 18th Century. The allure the chance of rapid enrichment exercised upon the metropolitan and colonial population was immense and triggered a gold rush to the mining regions. For one specialist, this influx constituted “the first mass migration in Brazilian demographic history”14. Besides the internal displacements, the mines also lured an even greater number of Portuguese immigrants, estimated at 400 thousand individuals over the course of the 18th Century. However, much of this migratory wave that flooded the region was compulsory in nature. The volume of transatlantic slave traffic into Portuguese America, which was already the largest in the New World, doubled in the first half of the Seventeen Hundreds. From 1701 to 1720 some 292 thousand African slaves came ashore at Brazilian ports, and most of these were destined for the gold mines. From 1720 to 1741 this rose to 312.4 thousand. The slave trade reached its peak in the two decades that followed, with 354 thousand enslaved Africans shipped to Portuguese America between 1741 and 1760.    

The enormous territorial and demographic expansion of Portuguese colonization in America in the 18th Century saw a corresponding rise in economic, social and political tensions. In the specific case of Minas Gerais, a captaincy established in 1720, the tumultuous nature of its occupation caused an escalation in conflicts: food shortages sparked terrible famines during the first years, followed swiftly by price speculation on the staple foods brought into the region; clashes between the first discoverers/settlers (from São Paulo) and the various blow-ins from the rest of the Colony and from Portugal led to the War of the “Emboabas”; The Crown's normative efforts to impose its power in the region by establishing villas and installing a bureaucratic apparatus were met by stiff resistance from the settlers15. For the purposes of this essay, however, our main interest is in another kind of social conflict, that expressed in slave escapes, the formation of Maroon colonies and broader plans for a slave uprising.       

Many authors have argued that the specific conditions of the mining operations in the region gave the slaves more room to exercise autonomy and resist the control of their masters. That the gold mines were scattered throughout the captaincy, that the workers could receive a stake in the yield and that they enjoyed ample control over their own work (as in the case of the slaves from the Coast of Mina, who were famed for their mining skills) conferred a considerable degree of slave autonomy. This meant that their masters rarely needed to employ coercive measures to ensure the regularity of extraction, which, in turn, enabled slaves to put money aside with which to buy manumission16.      

While the existence of channels for slave autonomy may well have made them more likely to stick to the status quo, it also provided greater scope for resistance. In relation to the latter, historians note that many Maroon colonies in Minas Gerais very often maintained close commercial relations with the surrounding communities. João José Reis suggests that this boom in Maroon activity may have stemmed from the authorities' own repressive drive to clear them out, as the “miserly definition” of a Maroon colony…

as any grouping of five or more fugitive blacks encamped on vacant lands [...], conceived with the purpose of better controlling escapes, actually ended up aggravating the phenomenon in the eyes of contemporary and later scholars17

Whether this assessment is correct or not, one thing that is certain is that there were at least two large, thousand-strong Maroon colonies among the myriad small runaway encampments: the 'Quilombo' of Ambrósio, defeated in 1746, and Quilombo Grande, which fell in 1759. In addition to these, researchers have also identified three planned slave uprisings (1711, 1719 and 1756), all of which were foiled in advance.   

Here we return to the question posed at the beginning of this essay: faced with this explosive social context, with a white population clearly outnumbered by the black, why did nothing similar to Palmares occur in Minas Gerais? The question becomes all the more intriguing when we recall that the precedent set by the Palmares haunted the public authorities of Minas throughout the entire first half of the 18th Century. The warnings issued to King João V by the Count of Assumar in 1718 are famous: according to the governor of what was then the captaincy of São Paulo and Minas de Ouro, fighting the Maroon colonies was a matter of supreme relevance, as the “conservation or ruin of this nation could depend upon it […] as I see in the Negroes of this government the inclination toward something similar to the Palmares of Pernambuco”18.  

As we have already seen, the prevailing response to the question is that hard-line legislation institutionalizing the slavehunter prevented the flowering of any more Palmares in Portuguese America. However, some historians offer an alternative explanation. Donald Ramos, for example, suggests that the very proliferation of small runaway slave settlements in Minas Gerais served to water down the region's power of resistance against slavery. The trade that many of these small communities struck up with the surrounding society would indicate that they functioned more as “escape valves” than as fronts of opposition to the slave system19. Most interesting of all in Ramos' argument is the observation that manumissions fulfilled a similar role in shoring up the slave society.      

Indeed, the practice of manumission became widespread in Portuguese America from the 18th Century. It is no accident that a substantial number of the studies on this issue look to Minas Gerais during this period. Given the impossibility of reviewing all or even the most relevant of the works available on the theme, John Russell-Wood's recently presented summary proves extremely useful. Two points set the mining experience apart within the context of Portuguese America: firstly, the fact that more slaves received their liberty during the height of the gold mining activities (second half of the 18th Century) than during the decline; secondly, the more extensive presence of the practice of 'coartação', that is, the slave's purchase of his or her own freedom via installments. Of all that can be said about the practice of manumission, for Russell-Wood, one thing all studies on the 18th-century mines agree on is that

[…] slave women were preferred to men, Mulattos to Negroes, Brazil-born to African, urban slaves to rural slaves and that many slave owners preferred to manumit babies rather than adults20.  

In its general outline, manumission in Minas Gerais basically repeated the model Stuart Schwartz encountered in Bahia at the close of the 17th Century. This standard obeyed a basic rule: the further removed from the transatlantic slave trade, the greater were a slave's chances of receiving manumission. African men, the majority on the slave ships, hardly ever received manumission, though after one or two generations their descendents often did. 



In the late 18th, early 19th Centuries, Portuguese America presented a demographic configuration without parallel among the New World colonies. In order to properly understand this difference, let us take a bird's-eye view of the other European colonies up to that time.   No fim do século XVIII e início do XIX, a América portuguesa contava com uma configuração demográfica ímpar no quadro das sociedades coloniais do Novo Mundo. Para compreendê-la devidamente, vale dar uma olhada a vôo de pássaro nas demais colônias européias de então.

Throughout the entire 18th Century, the various sugar islands of the British and French Caribbean, their own slight differences apart, presented enormous unbalance between the number of free whites and black slaves. The latter wildly outnumbered the former, even in colonies with a relatively higher number of settlers of European origin. Barbados is a case in point, where there were always about four black slaves to each free white throughout the Seventeen hundreds. In Saint-Domingue (now Haiti) on the eve of the uprising, however, there were as many as fifteen slaves to every white, while the number of freed blacks and mulattos never rose to match that of the slaves. In Saint-Domingue, these groups (freed blacks and mulattos) – which were fundamental to the start of the revolution that would lead to the downfall of slavery and emancipation from French rule – numbered fewer than 30 thousand individuals, a number equivalent to that of the white population. In Jamaica, the proportion was even smaller21.        

The South of Continental British America and, later, the southern states of the North-American Republic, constituted another New World slave society of a bi-racial character. If here the number of freed blacks and mulattos was far smaller than in the British and French Caribbean, there was demographic parity between the free white and black slave populations. 

By far the greatest demographic variety between the European colonists (albeit with decisive support from the indigenous populations in continental colonies) and the African slaves and their descendants was to be found in Spanish America. The concentration of black slaves in the cities or enclaves (such as the regions of Caracas, Chocó and the Lima coast) were so low that we cannot really characterize Spanish America as a genuine slave society22

Portuguese America, on the other hand, undoubtedly was, although of a different type to the British and French Caribbean colonies and those in the southern United States. The difference was Brazil's large freed black and Afro-descendant populations living side-by-side with a significant white society and vast slave majority, largely comprising Africans, though also with a lower percentage of American-born Creoles and Pardos. Variations from captaincy to captaincy (the extreme north and south were both still predominantly indigenous) and the imprecision of the available demographic statistics aside, the colonial Brazilian population at the beginning of the 19th Century was roughly as follows: 28% white, 27.8% freed black and mulatto, 38.5% black and mulatto slave and 5.7% Amerindian23.

The amassment of this large population of freed blacks and mulattos fundamentally stemmed from the dynamic between the transatlantic slave trade and manumission. The enslavement of Africans, their transportation to Brazil, the slave labour they did here (mostly unskilled rural and urban labour), the recomposition of familial and cultural bonds, the production of descendants, who, within one or two generations, would certainly earn manumission – these and other aspects can be seen as parts of a large-scale institutional process of status transformation, as proposed by Patterson and Kopytoff.

Luiz Felipe de Alencastro captured this movement with rare perspicacity in the conclusion to his book O trato dos viventes, in which he examines what he called the “invention of the mulatto”. According to Alencastro, the favouritism towards mulattos in Portuguese America can be seen in the facts that they were hired more frequently for skilled work, were admitted into the auxiliaries and, above all, were privileged when it came to manumission. Alencastro contrasts this reality with that of Portuguese Africa, where from the very beginning mulattos were placed on a par with the blacks. In Alencastro's own words:       

Brazil reveals a specific process whereby miscegenation – the simple demographic result of a relationship of domination and exploitation – became multiracialism, a complex social process that generates a racially plural society. The fact that this process was layered and eventually ideologised, even sensualised, does not erase its intrinsic violence, a consubstantial part of Brazilian society: there are mulattos in Brazil, but none in Angola, because here we had the systemic oppression of colonial slavery, while there did not24.  

In summary: in order to establish an abiding Brazilian slave society, one founded upon the incessant influx of foreigners, it was necessary to create safety mechanisms that could defuse the kind of tense social environment as simmered in the British and French Caribbean or even in Pernambuco in the 17th Century. The gradual liberation of the descendants of African slaves – no longer foreigners, but Brazilians – was the cornerstone of this mechanism. The definitive proof of the validity of this equation is how free and freed blacks and mulattos related with the slave system: their main economic and social ambition was precisely to acquire slaves of their own, that is, to become masters themselves.    

Various recent works have documented the common practice of free, freed and even enslaved blacks and mulattos owning slaves. Given the dynamic of the slave flow into Brazil, the heaviest in the history of the Atlantic slave trade, the African slave was a socially inexpensive commodity25. This was what enabled slavery to spread throughout the fabric of Brazilian society and become such a singular system. This mechanism, in turn, proved a heavily decisive factor in the equally unparalleled economic configuration of Portuguese America. 

As has long been consensus in Brazilian historiography, starting in the 18th Century, with the impact of the mining operations, the colonial economy began to diversify greatly, principally with the appearance of active production for the internal market, such as the cattle ranching in Rio Grande do Sul and in the São Francisco valley, or in the production of provisions in the captaincy of Minas itself, as well as in São Paulo and Rio de Janeiro. The emergence of various urban centres throughout Minas Gerais and the growth of the old cities of Rio de Janeiro and Salvador also fuelled the domestic economy. Tobacco production in the Recôncavo Baiano was another boom activity, as tobacco was the core merchandise used to acquire slaves on the Coast of Mina, as it was highly valued in the mining zones. Nor can we forget that the enclaves of sugar plantations in the Recôncavo Baiano, Zona da Mata in Pernambuco and Campos dos Goitacazes retained their vitality throughout the century, despite the competition from the Antilles that was barring their produce from the French and British markets.      

What matters from the perspective of this analysis is that all of these segments – whether rural or urban – were based upon slave labour and relied upon a slave ownership structure that spanned various levels of wealth, such that slave stocks were not monopolised by the wealthiest barons or even by the white landowners. Portuguese America therefore combined its diverse economic activities with the range of modalities of slave exploitation present throughout the New World: the mining and urban slavery of Spanish America, the slave plantations of the Caribbean and the production of provisions of the Chesapeake region.   

One could even argue that this was also the configuration present in Spanish America, which had plantation slavery in the region of Caracas. However, there were three basic differences between the two that must not be forgotten. Firstly, we must compare the economic weight of the indigenous population in central Spanish America against the generalized use of slaves in Portuguese America. Secondly, we can contrast the lack of economic integration among the colonies of Spanish America with the very respectable integration (for the means of transportation of the period) that mining brought to the Portuguese colony, from Rio Grande do Sul in the south to Pernambuco up north – despite the deep rift between the Amazonian valley and the rest of Brazil. Thirdly, and most importantly, the transatlantic slave trade played a crucial role in fuelling economic growth in Portuguese America. And here is a point of substantial divergence with the French and British colonies, where the slave trade was always controlled from the respective fatherlands. In Portuguese America, from the beginning of the 17th Century, the slave trade was organized directly from the Brazilian ports, that is, the large slave dealers that ensured the supply were actually based in Recife, Salvador and Rio de Janeiro, rather than in Lisbon. 

The slump in mining and the expansion in slave-based agro-exportation that occurred at the turn of the 18th/19th Century - with the emergence of new productive areas such as Maranhão (cotton) and western São Paulo (sugar), coupled with the recovery of the traditionally productive areas like Pernambuco, Bahia and Rio de Janeiro -, did not alter the systemic nature Brazilian slavery had acquired in the previous century. Quite the contrary, in fact, as it was precisely that social and economic configuration that had prepared the ground for such a quick response from the slave-owning producers of Portuguese America to the new and favourable conditions of the global market.   

For the purposes of this essay it would be interesting to examine the response from Bahia, as this will prove to be central to its argument. The slave revolt on Saint –Dominigue in the 1790s signalled deep changes for the sugar industry of the Americas. Prior to this date, the French colony accounted for 30% of global sugar production and was the world's largest coffee producer. With the uprising of 1791, sugar and coffee production went into collapse on Saint-Dominigue, opening enormous space for the production of these commodities in other colonies in the Americas, coupled with the increase in demand for tropical products in countries at the threshold of the industrial revolution. Faced with this new context, the transatlantic slave trade into Bahia accelerated in order to meet the sugar industry's demands for new workers. The reactivation of agro-exportation in the Recôncavo Baiano stimulated the cultivation of provisions in parishes unsuited to growing cane, but which likewise employed slave labour on a large scale. The resulting rise in the number of captives caused even the population of Salvador to swell26

From the late 17th Century the zone of preference for slave traders supplying Bahia was the Coast of Mina, though some also operated out of Angola. At the turn of the 18th/19th Century the dealers from Bahia benefited from a sharp upturn in the supply of slaves on the Coast of Mina. This increase occurred for two reasons: firstly, the French and British were out of the equation because both had ceased to ship slaves to their colonies; secondly, regional warfare, sparked by the Jihad of Usman dan Fodio, was generating a steady supply of captives, many of whom were sent to Bahia.

These groups plucked from the Coast of Mina, from various ethnicities (Nagô, Hauçá, Jeje, Tapa), were responsible for the gravest spate of African slave revolts in the history of Brazil. Anything like this systematic resistance to slavery had only ever been seen in the Palmares War and would only be equalled again in the abolitionist movement of the 1880s. From 1807 to 1835 Bahia endured a period of continuous rebellions by African slaves, the zenith being the Malês Revolt, “the most serious urban slave uprising ever staged in the Americas”27

What came of this resistance movement? Unlike similar slave uprisings in the British Caribbean, the spate of revolts in Bahia between 1807 and 1835 had no cumulative effect capable of putting the order of Brazilian slavery in check. The broader Atlantic context helps us to understand the real dimensions of the Bahia insurrections. While the revolts of 1816 (Barbados), 1823 (Demerara) and 1831 (Jamaica) were decisive in stirring the anti-slavery campaign in the British Empire, the slave resistance of the 1880s that eventually led to abolition in the Empire of Brazil had little to do with the historical experience of Bahia in the revolts of 1807-1835. In a nutshell: though serious and violent, they did not undermine the slave system in Brazil.    

The key to understanding this failure resides exactly in the rift that radically separated the African slaves and their Brazilian-born black and mulatto descendants. The latter did not participate in the African-led slave revolts in Bahia. Much to the contrary, as João José Reis explains:    

Mulattos, Cabras and Creoles were the majority of those employed in controlling and repressing the Africans. It was they who did the dirty work for the whites in maintaining order at the fountains and in the squares and streets of Salvador, invading and plundering the religious shrines on the outskirts, hunting down runaway slaves throughout the province and subduing slave rebellions wherever they arose28.  

The commitment of the Creoles and Mulattos – especially when either free or freed – to preserving the institution of slavery, and not just the commitment of the white masters, was the decisive factor in ensuring the survival of the Brazilian slave system. 



The armoured plating provided by this systemic configuration in Brazil not only prevented the occurrence of any further Palmares, but, above all, any chance of a slave revolt like that of Saint-Dominigue. In the 19th century, with Brazil now a nation state, this utterly stable internal slave-based social system spurred unheard of expansion in the transatlantic slave trade – though strictly speaking illegal since 1831 – and of Brazilian slavery itself. During the forty-year period between the arrival of the Royal family in Brazil (1808) and the definitive end of the slave trade in 1850, more than 1.4 million slaves were shipped to the Empire, that's 40% of the total number of Africans who made port in Brazil over the course of three centuries of slavery. In this sense, the changes brought upon the slave system in the Eighteen hundreds, especially by the incredible boom in coffee production in the Paraíba valley, which rapidly made Brazil the world's premier supplier of the commodity, rested upon longstanding practices that made it possible to introduce droves of foreign slaves without any risk to the internal security of the society. 

In the 19th Century, the greatest threat to Brazilian slavery came from abroad, in the form of pressure from the British anti-slavery lobby29. Not surprisingly, the ideological response the British diplomatic and military action provoked from Brazilian slave owners and politicians appealed, among other things, to the very logic of the systemic functioning of Brazilian slavery. In so doing, they inverted the predominant ideological vision in the Colony. Effectively, with one or two exceptions, the Metropolitan authorities stationed in Portuguese America always harboured the view that the free Black and Mulatto men presented more of a threat than a boon to the colonial order30. In other words, most of the Metropolitan administrators had never really understood the institutional process of Brazilian slavery.        .

This perception began to change at the beginning of the 19th Century, largely by agency of the quills of European travellers passing through or living in Brazil. The Englishman Henry Koster, for example, a slave owner in Pernambuco in the second decade of the Eighteen hundreds, did not fail to observe the ease with which Creole and Mulatto slaves obtained manumission in Brazil, which he contrasted with the difficulties endured by slaves in the British Caribbean31. It is accounts like this by European travellers that gave rise to the image of Brazilian – and even Iberian – slavery as having been more “benign” that the Anglo-Saxon counterpart.

The theme was swiftly instrumentalized by the architects of the Brazilian Nation State. The view that the freed slaves and their descendants were the allies rather than the enemies of the Brazilian slave masters emerged in 1822 at the Court of Lisbon debates convened to resolve the issue of Brazilian independence. On that occasion, while arguing the criteria for citizenship and political participation to be adopted in the future Constitution with Portuguese parliamentarians, the deputy for Rio de Janeiro Custódio Gonçalves Ledo declared:  

There are no grounds whatsoever for depriving freedmen of this right [to vote]. There are many freedmen in Brazil, of great use to society, and who operate in important segments of industry; many have families; which is why it would be the greatest possible injustice to deprive these citizens of the power to vote, and one could even go so far as to say it would aggravate the evil of slavery32.  

The definition of citizenship defended by Custódio Gonçalves Ledo in Portugal was crystallized in the Political Constitution of the Empire of Brazil. According to Article 6, paragraph 1 of the Constitution of 1824, freedmen born in Brazil were to be considered Brazilian citizens. However, freed Africans were to be excluded from the social body of the nation. This constitutional norm, in turn, franchised Brazilian freedmen with the right to vote: according to Articles 90 to 95, former slaves native to Brazil possessing a net income of a hundred thousand reis per year were entitled to vote in the primary elections, which chose the members of the provincial electoral colleges, but could not run for office. Ingênuos, on the other hand, i.e. the children of freedmen or women (whether African or Brazilian), could vote and run in the provincial elections, so long as they met with the census requirements.  

It was therefore a highly inclusive definition of citizenship. This constitutional paragraph ended up being used as pro-slave trade propaganda in the face of mounting British pressure. In 1838, José Carneiro da Silva, the future viscount of Araruama and a notable conservative politician, defended the annulment of the law of 1831 and the legalization of the slave trade precisely on the basis of the historical experience of Brazilian slavery:    

I have seen slaves the masters of slaves, the owners of plantations, herds of cattle and horses, endowed with vast and profitable estates. I have seen slaves earn their freedom, become landowners, soldiers, officers of rank, the holders of public posts of great utility to the State.  
How many officials and members of even higher orders were once, in other times, slaves, but now live with their families, cooperating toward the good of the State in the trades and positions in which they are employed, increasing the population and splendour of the nation that has naturalized them!33

This became a pet theme in historiography in the 20th Century. One need only recall the theses of Gilberto Freyre and Frank Tannenbaum on the supposedly benign character of Brazilian slavery, which soon found their way into the ideology of racial democracy. It is not my purpose here to throw yet more earth on this coffin. Nonetheless, what can never be forgotten is that this entire equation was rooted in the largest compulsory human migration of the modern world – a veritable crime against humanity, however reluctant countries like Portugal, England and Holland may be to recognise it as such.   



Received for publication on January 17, 2006.



Rafael de Bivar Marquese is a lecturer at the Department of History at the University of São Paulo.
1 Text originally presented at the I Encontro entre Historiadores Colombianos e Brasileiros (First Meeting of Colombian and Brazilian Historians) hosted by Ibraco in Bogotá, Colombia, in August 2005.
2 For more on Palmares, see Décio Freitas' Palmares, a guerra dos escravos. Rio de Janeiro: Graal, 1990 (1st ed. 1971) and República de Palmares. Pesquisa e comentários em documentos históricos do século XVII. Maceió: Editora da Ufal, 2004. On slave resistance in the English and French Caribbean and in Suriname, see Patterson, Orlando. “Slavery and slave revolts: a socio-historical analysis of the First Maroon War, 1655-1740”. Social and Economic Studies, vol. 19, nº 3, set. 1970; Craton, Michael. Testing the chains. Resistance to slavery in the British West Indies. Ithaca: Cornell University Press, 1982; Price, Richard. First-Time. The historical vision of an Afro-American people. Baltimore: The Johns Hopkins University Press, 1983; Dubois, Laurent. Avengers of the New World. The story of the Haitian revolution. Cambridge, MA: Belknap Press, 2004.
3 On Maroon activity in Minas Gerais, see Guimarães, Carlos Magno. Uma negação da ordem escravista. Quilombos em Minas Gerais no século XVIII. São Paulo: Ícone, 1988; on the spate of slave revolts in Bahia, see Reis, João José. Rebelião escrava no brasil. A história do levante dos malês em 1835. Ed. revista. São Paulo: Companhia das Letras, 2003.
4 This is the explanation proposed by Stuart Schwartz and broadly developed in the work of Silvia Lara. See, respectively, the essays “Repensando Palmares: resistência escrava na Colônia”. In: Escravos, roceiros e rebeldes. Bauru: Edusc, 2001, and “Do singular ao plural: Palmares, capitães-do-mato e o governo dos escravos”. In: Reis, João José & Gomes, Flávio dos Santos (orgs.). Liberdade por um fio. História dos quilombos no Brasil. São Paulo: Companhia das Letras, 1996.
5 The idea underlying this differentiation derives in part from Robin Blackburn's suggested distinction between “Baroque slavery” and “modern slavery”. See, The making of New World slavery. From the Baroque to the Modern, 1492-1800. London: Verso, 1997. However, Blackburn did not pay due heed to the presence of pockets of Baroque slavery within modernity, within the logic of the global market. On this, see the criticisms lodged by Stuart Schwartz in “Review of the Making of New World Slavery: From the Baroque to the Modern, 1492-1800, by Robin Blackburn”. In: William and Mary Quarterly, series 3, vol. LV, nº 3, Jul. 1998.
6 For more on this theme, see the following works: Schwartz, Stuart. “Alforria na Bahia, 1684-1745”. In: Escravos, roceiros e rebeldes, pp. 165-212; Slenes, Robert. The demography and economics of Brazilian slavery: 1850-1888. Doctoral thesis in History. Stanford: Stanford University, 1976; Alencastro, Luiz Felipe de. “La traite négrière et l'unité nationale brésilienne”. Revue Française d'Histoire d'Outre-Mer, nos 244-245, 1979; Eisenberg, Peter. “Ficando livre: as alforrias em Campinas no século XIX”. In: Homens esquecidos. Escravos e trabalhadores livres no Brasil, séculos XVIII e XIX. Campinas: Editora da Unicamp, 1989; Karash, Mary. A vida dos escravos no Rio de Janeiro, 1808-1850. São Paulo: Companhia das Letras, 2000; Mattos, Hebe Maria. “A escravidão moderna nos quadros do Império português: o Antigo Regime em perspectiva atlântica”. In: Bicalho, M. F.; Gouvêa, M. de F. & Fragoso, João (orgs.) Antigo Regime nos Trópicos. A dinâmica imperial portuguesa (séculos XVI-XVIII). Rio de Janeiro: Civilização Brasileira, 2001; Florentino, Manolo. “De escravos, forros e fujões no Rio de Janeiro Imperial”. Revista USP. Dossiê Brasil Imperial, nº 58, Jul.-Aug. 2003.
7 Kopytoff, Igor. “Slavery”. Annual Review of Anthropology, vol.11, 1982, pp. 221-22. Also see Patterson, Orlando. Slavery and social death. A comparative study. Cambridge, Mass.: Harvard University Press, 1982.
8 Cf. Miller, Joseph C. “O Atlântico escravista: açúcar, escravos e engenhos”. Afro-Ásia, nos 19-20, 1997.
9 Cf. Schwartz, Stuart. Segredos internos. Engenhos e escravos na sociedade colonial, 1550-1835. São Paulo: Companhia das Letras, 1988, pp. 22-73; Alencastro, Luiz Felipe de. O trato dos viventes. Formação do Brasil no Atlântico Sul, séculos XVI e XVII. São Paulo: Companhia das Letras, 2000, p. 69. All subsequent figures on the transatlantic trading of African slaves to Brazil have been taken from this source.
10 Cf. Alencastro, O trato dos viventes, pp.188-246; Marquese, Rafael de Bivar. Administração & escravidão. Idéias sobre a gestão da agricultura escravista brasileira. São Paulo: Hucitec, 1999, pp. 42-49; Puntoni, Pedro. A mísera sorte. A escravidão africana no Brasil holandês e as guerras do tráfico no Atlântico Sul, 1621-1648. São Paulo: Hucitec, 1999.
11 Cf. Emmer, P. C. “The Dutch and the making of the second Atlantic system”. In: Solow, B. (org.). Slavery and the rise of the Atlantic System. Cambridge: Cambridge University Press, 1991.
12 Cf. Schwartz, “Repensando Palmares”, pp. 244-55.
13 Cf. Schwartz, “Alforria na Bahia, 1684-1745”, pp. 165-212.
14 Marcílio, Maria Luiza. “A população do Brasil colonial”. In: Bethell, Leslie (org.). História da América Latina. Vol. 2: América Latina Colonial. São Paulo: Edusp/Funag, 1999, p. 321.
15 For an overview of this context, see the synthesis by Souza, Laura de Mello & Bicalho, Maria Fernanda. 1680-1720. O império deste mundo. São Paulo: Companhia das Letras, 2000.
16 Among these studies, see Vallejos, Julio Pinto. “Slave control and slave resistance in colonial Minas Gerais, 1700-1750”. Journal of Latin American Studies, vol.17, nº 1, May 1985.
17 Reis, João José. “Quilombos e revoltas escravas no Brasil”. Revista USP. Dossiê Povo Negro – 300 anos. nº 28, Dec. 1995-Feb. 1996, p.18.
18 Apud Lara, Silvia. “Do singular ao plural: Palmares, capitães-do-mato e o governo dos escravos”, p. 90.
19 Cf. Ramos, Donald. “O quilombo e o sistema escravista em Minas Gerais do século XVIII”. In: Reis, João José & Gomes, Flávio dos Santos (orgs.). Liberdade por um fio. História dos quilombos no Brasil. São Paulo: Companhia das Letras, 1996.
20 Russell-Wood, A. J. R. Escravos e libertos no Brasil colonial. Rio de Janeiro: Civilização Brasileira, 2005, p. 315.
21 Cf. Watts, David. Las Indias Occidentales. Modalidades de desarrollo, cultura y cambio medioambiental desde 1492. Madrid: Alianza Editoral, 1992, pp. 355-70.
22 On slavery in continental British America and in Spanish America, see Blackburn, The making of New World slavery, pp. 457-508.
23 Cf. Marcílio, “A população do Brasil colonial”.
24 Alencastro, O trato dos viventes, p. 353.
25 Cf. Florentino, Manolo. Em costas negras. Uma história do tráfico atlântico de escravos entre a África e o Rio de Janeiro (séculos XVII e XIX). Rio de Janeiro: Arquivo Nacional, 1995.
26 Cf. Barickman, B. J. Um contraponto baiano. Açúcar, fumo, mandioca e escravidão no Recôncavo, 1780-1860. Rio de Janeiro: Civilização Brasileira, 2003.
27 Reis, Rebelião escrava no Brasil, p. 9.
28 Rebelião escrava no brasil, p. 322.
29 Cf. Needell, Jeffrey. “The abolition of the Brazilian slave trade in 1850: historiography, slave agency and statesmanship”. Journal of Latin American Studies, vol. 33, nº 4, Nov. 2001.
30 On this ideological view see the following works by Sousa, Laura de Mello. Desclassificados do ouro. A pobreza mineira no século XVIII. Rio de Janeiro: Graal, 1983, and Lara, Silvia H. Fragmentos setecentistas. Escravidão, cultura e poder na América portuguesa. Thesis presented  in application for a permanent teaching post. Campinas: IFCH/Unicamp, 2004.
31 Cf. Koster, Henry. Viagens ao Nordeste do Brasil. Recife: Fundação Joaquim Nabuco/Editora Massangana, 2002, chapters 18 and 19, 2 vols. (1st ed. 1816).
32 Apud Berbel, Márcia Regina & Marquese, Rafael de Bivar. “A escravidão nas experiências constitucionais ibéricas, 1810-1824”. Text presented at the International Seminar: Brazil – from One Empire to Another (1750-1850) (History Department, USP, Sept. 2005). Available at
33 Cf. Marquese, Rafael de Bivar & Parron, Tâmis Peixoto. “Azeredo Coutinho, Visconde de Araruama and Memória sobre o comércio dos escravos de 1838”. Revista de História, vol.152, 1st semester of 2005, p. 122.