3 Slavery and the Economy in the Eighteenth Century

The growth of the West Indies plantation system in the seventeenth and the early eighteenth centuries did not put an end either to the Brazilian sugar industry or to the thriving slave system upon which the Brazilian economy rested. The Dutch occupation and the subsequent growth of the West Indies sugar industry did, however, seriously affect the colonial economy. Not only was a large part of the Pernambuco sugar industry destroyed, taking a long time to recover, but Brazilian export markets were also reduced and production stagnated for most of the late seventeenth and the early eighteenth centuries. Bahia did continue to grow, but the golden age of profitability had passed. Competition from the West Indies sent sugar prices into a decline relative to the first half of the century, and West Indian demand for slaves meant rising African slave prices, thus squeezing planter profits. By the last two decades of the century, the Brazilian economy was in a relatively depressed state, and an anxious Crown began seeking new markets and products to revive the colonial economy.

Among the many attempts to develop new resources, the Crown began to explore the interior with hopes of finding mineral wealth. The success of the paulista bandeirantes in supplying Indian slaves at mid-century had led to government subsidization of systematic surveys of the interior.¹ After numerous discoveries of minor deposits of gold and precious stones throughout the second half of the seventeenth century, a major expedition in 1689–90 discovered substantial alluvial deposits of gold in the region of what is today Minas Gerais, some 200 miles inland from the port of Rio de Janeiro. Thus, at the end of the century, an entirely new type of slave economy would emerge on Brazilian soil, that of slave mining. Gold and then diamonds would be the basis for this eighteenth-century phenomenon, and Brazil would again be the initiator of a system of production that would soon be replicated in Spanish America. Although silver mining would still be based on Indian free wage and forced contract labor in the Andes and Mexico, gold mining in the isolated regions of the interior of Spanish America would often be worked by African slave labor, much as it was in Brazil.²

The rush to these gold deposits by the coastal whites with their slaves was immediate. Before the 1690s, the interior region of Minas Gerais in the heartland of the gold region had been populated by only unconquered Indians. As early as 1710 there were probably some 20,000 whites and an equal number of blacks there; by 1717, the slaves had increased to 35,000, and by the early 1720s, passed 50,000. The 100,000-slave population figure was probably reached in the 1730s, and by the 1760s, there were 249,000 free and slave colored and only 71,000 whites in the province. At the time of the census of 1776, there were 266,000 colored, of

GRAPH 3.1. Average annual Brazilian gold production, 1700–1799.

Source: Pinto, O ouro brasileiro e o comércio anglo-português, p. 114.

whom 157,000 were slaves and an extraordinary 109,000 were free.³ By 1808, the free colored had passed the slave population in total numbers, and although the slave population continued to grow – eventually reaching some 383,000 by 1872 – the free colored remained the dominant population in the province until the end of slavery.⁴ The rapidity of the growth of the slave population, its size, and its makeup marked Minas Gerais as an unusual zone of slave labor in Brazil. In turn, this gold-mining boom lasted into the second half of the eighteenth century (see Graph 3.1) and was also sustained by a boom in diamond exports in the later eighteenth century.⁵ Although the first six decades were a true gold rush, by the last quarter of the eighteenth century there was an increasing diversification in the regional economy, which would also be reflected in changes in the structure of the slave labor force. Finally, all of this growth led to the emergence of major urban centers in this interior province by midcentury, with Ouro Preto alone reaching twenty thousand persons, only about ten thousand fewer than in the key port cities of Salvador de Bahia and Rio de Janeiro in this period.⁶

In the first half of the eighteenth century, during the great gold rush period, the rapidly expanding slave population of Minas Gerais primarily came directly from Africa. In a sample of early eighteenth-century censuses from the principal districts of the Minas Gerais, Africans consistently made up more than 95 percent of the total Afro-Brazilian slave population. There were also a significant number of Indian slaves, who made up 2 percent of the total slave labor force, compared to Creole colored, who were 5 percent, and Africans, who in this total made up 93 percent of all the slaves (see Table 3.1). Even as late as 1738, in a census of some eight thousand slaves in the diamond district of Serro do Frio, the ratio of African born among the slaves was 95 percent, and the Indian slaves had disappeared from the census.⁷ But as time went on, the ratio of Africans declined even as the total provincial slave population continued to grow. In Congonhas do Sabará in 1771, the percentage of Africans declined to 68 percent and by the census of 1804, among the slaves whose origin was known, only 41 percent were Africans, and the rest were born in Brazil (see Table 3.2).⁸ This rise in importance of a native-born slave population is a theme we will return to in discussing the developments of Minas in the nineteenth century. However, it is clear

TABLE 3.1. Origin of the Slave Population in Selected Minas Gerais Districts, 1718–1738

Source: Luna, “Estrutura da Posse de Escravos em Minas Gerais (1718),” tabela 7, and Minas Gerais: Escravos e senhores, tabelas 2, 7, 9, and 13.

that already by the end of the eighteenth and beginning of the nineteenth century, the growth of the slave population was no longer being sustained by just the incorporation of new African arrivals.

The African origins among these first two or three generations of Mineiro slaves shifted over time and place. Of the African-born slaves

TABLE 3.2. Division of Slaves by Origin in Selected Minas Districts in 1804

Source: Luna, “Estrutura da Posse de Escravos em Minas (1804),” tabela 7.

listed in the 3 district sample of 1718, some 42 percent were from the Cape Verde, Senegal, or West African regions (the majority being Nagos and Minas from the Bight of Benin), and 58 percent were from the southern African regions of Congo-Angola, who were mostly Bantu speakers.⁹ This ratio between Western and Central Africans fluctuated among different districts of the province over time, with the available data for the whole eighteenth and early nineteenth centuries showing a progressive domination in all towns and districts of Southern Africans.

The dominance of these Africans had a profound impact on the demographic characteristics of the local slave population, and initially the miners and other slave owners made no attempt to alter these developments. The gold fever initially did not encourage masters to think about long-term population concerns or family structural arrangements for their slaves, and they primarily purchased adult African males. The fact that the Crown granted out mining concessions on the size of slaveholdings and the immediate need for adult slave workers also encouraged this trend. The granting of tracks of land along gold-producing riverbanks – except for the discoverer – was based on the number of slaves owned by the miner. To obtain a grant of 66 meters (called a data) required a minimum of twelve slaves, the assumption being that they were all adult workers.¹⁰ Moreover, the Crown finally resolved on a tax on mining based not on gold production, which was difficult to calculate and register, but on the basis of a head tax on slaves.¹¹ Thus, the early sexual balance in the gold fields and towns of Minas Gerais was heavily biased toward adult males,

TABLE 3.3. Distribution of Slaves by Sex in Selected Districts of Minas Gerais in 1718

Source: Luna, “Estrutura da Posse de Escravos em Minas Gerais (1718),” tabela 7, and Minas Gerais: Escravos e senhores, tabelas 2, 7, 9, and 13.

often reaching an extraordinary ten or more males for every female slave (see Table 3.3). In fact, males represented 93 percent of all slaves in this year. Such a ratio meant that the only way to maintain the slave population at the more than 100,000 population range was through heavy and constant migration of slaves from the coastal ports.¹²

In a census of four districts of Minas in 1718, the ratio of African born among the 12,842 slaves whose origin was known was 88 percent, and an extraordinary 92 percent of the male slaves were Africans.¹³ Another study found that among the 402 adult slaves appearing in inventories in Sabará between 1725 and 1759, 88 percent were Africans and their sex ratio was 532 males per 100 females. These ratios remained the same for such inventories for Sabará written in the period of 1760 to 1808. Although the importance of adult Africans declined to 76 percent of all adult slaves listed, their sex ratio remained extraordinary at 844 males per 100 females. This compared to a rate of 166 Creole slave males per Creole female adult slaves in the first period and a sex ratio for adult Creoles of 115 in this second period.¹⁴ But this pattern began to change quickly in the second half of the eighteenth century. By the census of 1804, the majority of slaves were already Creoles (or native born) and thus the

TABLE 3.4. Division of Slaves by Sex in Selected Minas Districts in 1804

Source: Luna, “Estrutura da Posse de Escravos em Minas Gerais (1804),” tabela 6.

sex ratio had become far more balanced, although the 40 percent or so who were Africans guaranteed that there were still significantly more men than women among the slaves (see Table 3.4). The overall ratio for the entire slave population whose sex was known now stood at 198 males per 100 females, a major change from the figures for the early eighteenth century.¹⁵

Not only were the Africans predominantly male, but they were also overwhelmingly adults. Thus, all studies of age among the African slaves in the eighteenth century show a ratio of fewer than 10 children (0–15 years of age) per 100 Africans, a ratio unlike the 40 to 50 children per 100 Creoles, or native-born slaves.¹⁶ Given the adult nature of the African migration, this extremely low incidence of children is not surprising, but along with the sex ratio it is another indicator showing that growth of the slave population could not come from natural growth of the African population itself. As long as Africans dominated the local labor force, importation of more Africans was necessary to keep the servile population growing despite the probably positive natural growth rates of the Creole slave population.

Slowly this distortion in the age structure of the resident slave population also began to change in the second half of the eighteenth century. In a census of slaves in Sabará in 1776, children ages 0 to 15 years made up approximately 28 percent of the total 21,268 slaves who were registered. Moreover, the overall sex ratio of the children was 117 males per 100 females compared to double that rate for adults.¹⁷ Clearly then, the first half of the eighteenth century was one dominated by Africans who greatly influenced the age and sex ratios of the resident slave population. This would change slowly over the course of the century in Minas Gerais as native-born slaves began to become ever more important in the labor force. Given this rise in importance of Creole slaves, the ratio of children in the total slave population increased, along with the ratio of women. In the census of Vila Rica of 1804, for example, children now represented 22 percent of the total slave population of the 2,763 slaves whose age was known.¹⁸

The distribution of slave ownership became increasingly concentrated over time. The Crown reserved the well-defined mine sites for their discoverers and for miners who held twelve or more slaves; those with fewer slaves were allowed to get smaller claims proportional to the number of slaves they possessed if any areas were left unclaimed, but the majority of these small slaveowning miners tended to concentrate on itinerant prospecting, which meant extraordinary freedom for their few slave miners who in fact were paid in gold for any minerals discovered.¹⁹ Slaves were worked in gangs and were carefully supervised by white or free colored overseers, but this was only on the large scale and fixed works called lavras. Thus, in certain clearly delineated gold fields, such as those in the environs of the cities of Vila Rica and Vila do Carmo, heavy concentration of slaves guaranteed a certain stability on a par with the discipline found in a controlled plantation environment.²⁰ Here and at other well-defined alluvial gold fields, heavy investment was carried out in the construction of the lavras, which had elaborate sluice constructions or dredging operations, and that required major hydraulic works that in their more elaborate development led to channeling of rivers, excavation of riverbanks, or alternatively, the construction of hillside terraces and the setting up of sluices and other water-diverting projects. All these activities required a high ratio of more skilled slaves who were carpenters, masons, and smiths, but also a great deal of physical labor.²¹ These tightly controlled and well-developed mining camps probably absorbed the majority of the mining slaves in the province. In a year of mining decadence, that of 1814, it was estimated that there were still 6,662 workers (of whom 6,493 were slaves) in some 555 lavras or hydraulic mine works, and 5,747 individual prospectors of whom 1,871 were slaves in Minas.²² These 8,364 slaves made up only about 6 percent of the estimated 149,000 slaves in the province at this time.²³

Probably half the miners and a quarter of the slaves were the itinerant prospectors, even in the earlier period. Many of these itinerant miners were in fact unsupervised slaves in scattered river sites throughout the province of Minas Gerais, and then further west into the provinces of Goiás and Mato Grosso. Already by 1735, Goiás had 10,263 mine workers, the majority of whom were slaves and forros working in both gold and diamond mining in both fixed mines and in prospecting²⁴; whereas Mato Grosso had 11,910 slaves by 1797.²⁵ In these cases, slave owners late on the scene and initially with little capital to develop elaborate works or with enough slaves to obtain a royal grant relied exclusively on itinerant slave miners and prospectors known as faiscadores. These faiscadores usually spent considerable time away from their masters prospecting for gold, eventually returning a fixed amount of gold dust to their owners, otherwise paying for all their own expenses and even receiving a wage in gold for their efforts or sometimes their freedom.²⁶ Although local governments attacked this itinerant style of mining as dangerous for social control, it was simply too widespread to destroy. In contrast, the formal mine works, the lavras, employed large groups of slaves under very close supervision.²⁷

Even though slave ownership was restricted to a minority of the mineiro population, the ratio of slave-owning households was relatively high, accounting for a third or more of the households in the towns of Minas Gerais in the eighteenth century.²⁸ As can be seen from the census of 1718, the majority of 2,120 slave owners (60 percent) held five or fewer slaves (of the total of 14,665 slaves) and controlled just under a quarter of all slaves (see Table 3.5). This pattern was repeated in a larger sample of owners and slaves in three zones of Minas in the early 1720s, that of Vila Rica, Sabará, and São José del Rei, which encompassed some 3,163 owners and 19,820 slaves (see Table 3.6). In the 1717 and 1718

TABLE 3.5. Distribution of Slaves and Owners by Size of Holdings, in Fourteen Districts in 1715–1717 and Seventeen Districts in 1718

Source: Seção Provincial, Arquivo Público Mineiro; for 1718, Luna, “Estrutura da Posse de Escravos em Minas Gerais (1718),” tabela 7.

samples, more than 60 percent of the owners held four or fewer slaves, and the figures for the three zones of 1721–22 were quite similar with close to 60 percent of the owners holding five or fewer slaves, and in both cases these small slave owners controlled about a fifth of the slave labor force. In both cases, the next largest group owners in terms of slaveholdings – those who held 6 to 10 slaves in the 1710s or 5 to 9 slaves in the 1720s – were one-quarter of the owners and roughly owned a quarter of the slaves. For those who owned more than ten slaves, there

TABLE 3.6. Distribution of Owners and Slaves by Size of Holding; Selected Towns, Minas Gerais in the 1720s

Source: Tarcísio Rodrigues Botelho, “População e escravidão nas Minas Gerais, c. 1720.”

was more variation among the communities studied, and there were very few owners with more than fifty slaves. In a 1718 sample of comarcas, the average slaveholding consisted of seven slaves and the average in the three communities in 1721–22 was between six and seven slaves as well.²⁹

These distributions show a surprising lower level of inequality in terms of slaveholdings among the slave-owning class compared to later slave distributions. The GINI index of inequality in the distribution of slaves among owners shows relatively low inequality levels from the mid-.40s to the mid-.50s, but also a very high mid-.60s for two mineiro districts. These high variations in the GINI suggest that there were some significant variations in the holding of slaves that were related to local economic conditions (see Table 3.7). Nevertheless, most of the GINI indexes generated for most of these eighteenth-century districts suggest that in general, these distributions of slaves among owners were probably less unequal than would be the case in nineteenth-century Brazil. As we shall see in the next chapter, an almost complete census carried out in Minas between 1831 and 1832 found a GINI of .57 among slave owners, a figure on the higher end compared to most of the GINIs so far calculated for eighteenth-century Minas Gerais. Finally, it is worth noting that these mineiro slave owners were surprisingly literate. A study of 263 male and female slave owners in Vila Rica in 1718 found that an extraordinary 87 percent of them were literate, and for Vila do Carmo in the same year, the figure was 85 percent for the 176 slave owners of both sexes whose literacy was known.³⁰

A study of postmortem inventories for the Minas zones of Rio das Mortes and Rio das Velhas during the eighteenth century also showed a pronounced unequal distribution of slaves (see Table 3.8), which probably reflected the undercounting of poorer slave owners who may not have made out wills. What is interesting is that when the owners are broken down by sex, women tended to be found among the smaller slaveholders, with 70 percent holding ten or fewer slaves, compared to

TABLE 3.7. Slave Owners and Their Slaves with Indices of Ownership and Distribution

Sources: Luna, “Estrutura da Posse de Escravos em Minas Gerais (1718),” tabela 7, and Minas Gerais: Escravos e senhores, tabelas 5, 30, and 34.

65 percent of the males in this category, a finding probably due to the overrepresentation of forras (free colored women) among the female slave owners in Minas.³¹

The sexual divisions are more pronounced for the forros who owned slaves and made out their wills in these two districts of Minas Gerais in

TABLE 3.8. Structure of Slave Ownership in Postmortem Inventories in the Comarcas of Rio de Velhas (1720–1784) and of Rio das Mortes (1716–1789)

Source: Eduardo França Paiva, “Por meu trabalho, serviço e indústria: Histórias de africanos, crioulos e mestiços na Colônia – Minas Gerais, 1716–1789” (São Paulo: Tese de doutorado, FFLCH-USP, 1999).

this period. Clearly, their average size of slaveholdings were smaller – just half of the ten-slave average among the non-forro owners. Nevertheless, it is interesting to note that among the forros, women held on average more slaves than the males and were more likely to be slave owners than were the forro men, playing a much more prominent role than females in the non-forro population (see Table 3.9). This same pattern could be found among some forty-nine forro slave owners in Sabará in 1720, where the thirty-seven women slave owners among the forro population controlled more than two-thirds of the ninety-nine slaves that this group owned, but again, the average holding of the freed women slave owners was less than the male owners.³²

In the breakdown of ownership by occupation, which exists only for the district of Carmo in 1718, the largest slave owners were officials, who most probably were miners. Clearly defined miners held on average nine slaves, whereas faiscadores held on average half that number.³³ It is evident from this distribution that only a few held large numbers of slaves, and these appear to have been concentrated in the mining sector. Of course, the very largest slave owners in the province in the first half of the eighteenth century were the miners. In São Caetano in 1804, for

TABLE 3.9. Structure of Slave Ownership among Forros in Postmortem Inventories in the Comarcas of Rio de Velhas (1720–1784) and of Rio das Mortes (1716–1789)

Source: Paiva, “Por meu trabalho, serviço e indústria,” loc. cit.

example, miners averaged twenty-three slaves, compared to just three on average for the faiscadores.

The availability of all the data on slaves in eighteenth-century Minas has a lot to do with the Crown's attempt to control gold smuggling and to tax this resource. Unable to control the illegal extraction of gold, the desperate Crown in 1735 gave up attempting to tax smelted gold (the usual Iberian manner of determining output and extracting taxes) and resorted to charging a slave head-tax for all masters in the mining zones and collecting tolls on all goods moving in and out of the province. Although there were variations of this taxing system over time, it was also initially adopted – although at a higher rate – for the slaves in the diamond districts as well, although here a full royal monopoly was established with strict control on exports.³⁴ Even this was insufficient, and in 1719, all goldsmiths were ordered to leave the province and all working goldsmiths in Brazil were banned from working in 1766.³⁵

GRAPH 3.2. Slaves freed, self-purchased, or remaining slaves in postmortem inventories in the Comarcas of Rio das Velhas (1720–1784) and Rio das Mortes (1716–1789) by size of slaveholding Note: (n = 10,600 slaves).

Source: Eduardo França Paiva, Escravido e universo cultural na colônia, Minas Gerais, 1716–1789 (Belo Horizonte: UFMG, 2001).

The Crown's attempt to control mining was more successful than its halfhearted attempt to control the rise of a mixed population of free colored. Although the Crown opposed the emergence of this class of person, the normal patterns of settlement, rewards for labor, and general miscegenation that became the norm for social relations in the province all led to the extraordinarily rapid rise, through manumission, self-purchase, and reproduction, of a large free colored population in the mining camps. Probably in no other slave region of America did the population of free colored grow as rapidly or become as important an element so early in the settlement process as in Minas Gerais, with this growth coming from very high rates of natural reproduction as well as a relatively steady movement of slaves into free status. Although it is difficult to estimate the annual flow of manumitted slaves (known as forros, or free persons who were born slaves) into the total free colored class, one study of inventories of two comarcas (districts) in Minas Gerais in the eighteenth century showed that 14 percent of all slaves were either gratuitously manumitted or purchased their freedom over the course of the century. Of these 932 ex-slaves (out of 6,656 listed), some 36 percent purchased their own freedom. Interestingly, as can be seen in Graph 3.2, the smaller the holding, the more likely that manumission and self-purchase were to occur.

We also have some census data from individual towns and districts that list forros as a separate group. In these estimates for some six regions in Minas Gerais in the period 1735–1749, the average ratio of forros was only 1 to 1.5 percent of the total population in these districts. But these numbers are based on tax records, which essentially taxed both slaves and forros, with the class of forro slave owners excluded because they already paid taxes on the slaves they owned and thus were excluded from the forro censuses. As early as 1714 in the district of Sabará, there were ninety-one black and mulatto slave owners who represented 7 percent of the total group of slave owners listed in that year. In Serro do Frio in 1738, there were 387 forros (244 of them women) who owned slaves, and they represented 22 percent of the slave owners and held 10 percent of all slaves.³⁶ That forros made up a significant part of the population is evident in a household census of the Minas parish of Freguesia de Congonhas do Sabará in 1771. Of the 393 households in the parish, 112 were headed by a forro and 72 percent of these forro household heads were women. This compared to the much lower ratio of only 16 percent of the households that were headed by a woman in the non-forros households.³⁷

By 1786, when there were some 174,000 slaves in the province, the number of free colored, both those freed in their lifetime and those born free, had already passed the 123,000 level. Their growth now continued even more dramatically than that of the slave population. By the first decade of the nineteenth century, freedmen outnumbered slaves and had become the largest single group in this fast-growing provincial population. That growth would continue into the nineteenth century despite the continued expansion of the slave population. Although the Portuguese government protested the growth of this class and charged that it was based on theft of gold and other minerals, there was little that it could do to stop its expansion. One perceptive Royal judge, Teixeira Coelho, noted in a 1780 “Instruction for Governing Minas Gerais” that the rules against vagabonds (vadios) should be suspended with regard to Minas:

…although these vagabonds are prejudicial in other places, here they are useful. Negro forros and mestiços – in the majority made up of mulattoes, caboclos, and carijós [Indians] perform a service by populating distant locals such as Cuiete, Abre Campo, and Pecanha, where they establish frontier forts, they make up the bands who enter the forest and destroy the quilombos and fugitive slaves, they cultivate subsistence farms, in short, they carry out a series of tasks which could not be carried out by the slave labor force.³⁸

Free blacks and mulattoes played an important part of colonial mineira society and by the end of the century became the dominant group within the region. The elite always despised them and attempted to prevent their social mobility in every way possible, but to little avail.³⁹

The Brazilian gold-mining economy also gave rise to an important regional urban culture. By the second half of the century, Minas Gerais had numerous urban centers in the ten- to twenty-thousand range that supported a highly developed urban lifestyle based heavily on both skilled and unskilled slave labor. The restriction of clergymen in the province also led to the massive organization of a complex group of confraternities or lay religious brotherhoods that were responsible for most of the church constructions and the elaborate celebration of religious holidays. These organizations served both religious functions and as mutual aid and burial societies and were vital centers of social activity for its participants. Most of these Irmandades, or brotherhoods, were for the poorer elements of the society and most were organized along color lines, with the most famous one for blacks being dedicated to the Nossa Senhora do Rosário (Our Lady of the Rosary), which was organized by both slaves and free persons of color. The elite tended instead to congregate in tertiary order confraternities sanctioned by the provincials of the various missionary orders.⁴⁰ The Nossa Senhora do Rosário black brotherhoods were established in all the towns in Minas by the 1720s and the records of some 62 of them have survived (out of 322 irmandades scattered throughout colonial Minas), making them the largest single such organization among the 52 types of brotherhoods established in colonial Minas Gerais.⁴¹ Much of colonial town life revolved around these confraternities and they were fundamental in patronizing the arts and music as well as being responsible for major church construction.⁴² So vibrant was this religious activity – much of which involved African and Afro-Brazilian artists, architects, and musicians – that the whole movement has been defined as a distinct barroco mineiro culture. In towns like Vila Rica, which reached a population of twenty thousand by the 1740s, the mining elite as well as artisans and even the poor supported a surprisingly rich local cultural development, which was expressed in a rather sumptuous display of architecture, the plastic arts, and music along with public ceremonials.⁴³ There is little question that the numerous black and brown religious brotherhoods were extremely important in the creation of this mineiro culture. In Minas Gerais, the most famous sculptors and architects were free colored. Antônio Francisco Lisboa, known as Aleijadinho, was the son of a slave woman and a white Portuguese-born architect father. His sculptures and decorations of eighteenth-century churches of Minas earned him the reputation as Brazil's leading artist of the Rococo period, and he, like his father, also worked as an architect.⁴⁴ Another was the slave-born Manuel da Cunha, who was the leading portraitist of the age and also painted many walls and altars of Brazil's leading churches. He was trained in both Brazil and Portugal and had already achieved an outstanding career before his manumission. In music, the composers of Minas were almost all mulattoes. The most outstanding was Emerico Lobo de Mesquita, who was organist to a major white brotherhood, a member of the mulatto brotherhood, and a composer totally current with the latest in European Baroque composition. A more prominent if less skilled composer was the Jesuit Padre José Mauricio Nunes Garcia, whose mother was a forra (slave woman freed in her lifetime) and whose father was a mestiço and who himself was appointed court composer when the imperial family moved to Brazil in 1808.⁴⁵

Although gold was the initial metal exported first from Minas Gerais, then from Goiás in the 1720s and Mato Grosso in the 1730s, it was not the only mineral produced. In 1729, in the northern end of Minas Gerais, it was announced that diamonds had been discovered in Brazil.⁴⁶ Like gold, diamonds were found in alluvial deposits, on the beds or banks of rivers, or in wadis left by seasonally active rivers. Slave labor was used to obtain these precious stones in the same manner as for gold, through panning, hydraulic works, and active washing of soils. The impact on the European market of the diamond finds in Minas Gerais and Goiás was immediate, and international prices dropped by two-thirds as a result of the discoveries. The Crown tried to create a royal monopoly on the extraction of these stones, but it was only partially successful. In fact, diamonds would prove harder to control than gold, because the latter required smelting. The eighteenth-century diamond boom, which started and peaked later than gold, tended to use fewer slaves in far more scattered holdings than in the gold-washing operations.⁴⁷

The rise of mining centers in the central interior zone of Minas Gerais would also have a profound impact on the subsequent growth of slavery and black populations in other parts of Brazil. The gold-mining boom of Minas Gerais powerfully shifted the center of gravity of the Brazilian economy and population from the north to the center and south. The logistics of interior transport guaranteed that the balance of trade to and from the interior provinces would be directed both to Bahia and to the southern cities.⁴⁸ Thus, the mines of Minas Gerais, Goiás, and Mato Grosso became the crucial hinterland of the southern port of Rio de Janeiro as well as the entire southeastern region. Rio de Janeiro soon outpaced Bahia in international shipping and trade and quickly approached the 50,000-population size of the imperial capital. The Crown recognized this new geographic reality by shifting the capital of the colony from Salvador, in Bahia, to Rio de Janeiro in 1763. This only furthered the city's dynamic expansion, and by the end of the century, Rio de Janeiro was not only Brazil's leading slave-trading port and the major port for Minas trade, but it was also Brazil's leading urban center with more than one hundred thousand persons. That made Rio de Janeiro, along with Mexico City, one of the two largest cities in America.⁴⁹

Other southeastern regions also benefited tremendously from the growth of this new interior market. Although the gold rush fever initially disrupted coastal production by attracting large numbers of speculators and coastal planters with their slaves, it soon created dynamic new markets that only the coastal zones were equipped to supply. In the first few years, the gold fever absorbed the workforce in Minas to such an extent that few interior workers, free or slave, were able to systematically engage in agriculture or stock breeding. Although this quickly changed and farming and ranching became established, especially along the major trade routes (the so-called caminhos), demand still outstripped local supply.⁵⁰ Thus, many of the food and animal needs for this growing interior province were supplied by the coastal provinces. The central and southern highlands around São Paulo began producing animals and foodstuffs for the mineiro market, but these quickly proved incapable of satisfying demand. To supply beef, hides, and the crucial mules for the great inland shipping caravans, a whole grazing industry was fostered in the open plains of Rio Grande do Sul and as far south as the eastern bank of the Plata River (in modern Uruguay). A major series of interior trails were now opened between these southern zones and São Paulo in the 1730s.

São Paulo would ultimately be the region most affected by the opening up of the interior mines. Until the late eighteenth century, the province of São Paulo at the southern limit of the Portuguese colony of Brazil was, like much of North America at the time, a lightly settled, forested frontier. However, it was home to a rather unusual combination of Indians, mestiços, and whites. By the standards of the rest of Brazil, it was a backward region, peopled by the marginal elements of Portuguese society. Probably in no other region of the Americas had Indian slaves and Indian and mestiço free workers been so fully integrated into a white-dominated colonial regime. As allies, dependents, and slaves, the local Indians were tightly woven into the fabric of paulista society and formed the base of its armies, farm laborers, and even its urban workers. This gave local society an unusual mestiço aspect and created a frontier population famous for its military prowess, its exploring ability, and its extraordinary mobility, attributes that carried paulista bands over all of Eastern South America from the Amazon to the Rio de la Plata.⁵¹

Although some sugar and its derivative aguardente (brandy made from sugar) were produced in the province of São Paulo from the beginning, these industries only satisfied local needs. Because of the difficulties of transport from the interior highlands, they did not become important exporters of these products until the second half of the eighteenth century. Far from the seat of colonial authority, this region was characterized by slash-and-burn agriculture, small coastal settlements, modest highland villages, and scattered peasant farms producing subsistence food crops in a densely forested interior. São Paulo was the province slowest to develop and the shift to African slave labor here took the longest time, only getting fully underway in the second half of the eighteenth century. Although a few Africans were to be found in the province from its earliest settlement, it was the local Indian population, either as slaves or as settled villagers under the control of the white and mestiço colonists, who were the fundamental labor force until well into the eighteenth century.⁵² Large numbers of Indians had been reduced to subservient, pacified villages (called aldeamentos) that were exploited for labor,⁵³ even as the local colonists continued to enslave Indians. Given the comparative poverty of the region, free and enslaved Indians remained the core of the labor force available to the small white and mestiço population. Although the extent of this Indian slave trade is debated,⁵⁴ there is no question that it was crucial in the evolution of the paulista economy because it provided one of the few sources of export income for a province that was still largely oriented toward subsistence agriculture.⁵⁵ But increasing government and church opposition to Indian enslavement – together with the growth of economic activities that permitted the paulistas to buy expensive African slaves – gradually led to the substitution of African slaves for Indian slaves after 1700, resulting in the end of Amerindian enslavement by the middle of the eighteenth century. In turn, the Indians settled in the aldeias did not survive as an autonomous culture after the middle of the nineteenth century.⁵⁶ In the early development of Minas Gerais, the original paulista explorers brought their Indian slaves with them, and such slaves remained important for the first twenty years of the colonies' development, after which they were totally replaced by African slaves.

By the time of the first available census of São Paulo in the 1760s and 1770s, the African and Afro-Brazilian slaves were well distributed in the province, with a very high ratio of slaves to free in the coastal communities, in the capital region, and in the west paulista zone (see Table 3.10). Moreover, there was clearly a progressive increase of the importance of

TABLE 3.10. Census of Free and Slave Population of São Paulo in 1765/68 and 1777

Source: Luna and Klein, Slavery and the Economy of São Paulo, table 1.1, p. 25.

the native-born Afro-Brazilian slaves, as shown by the relative low and declining sex ratios of the slave population in the two census periods.

Also, when the paulistas began to import Africans in significant numbers, their distribution among slave owners was more weighted to the smaller slave owners than those of Minas Gerais in this same period. In contrast to Minas districts in this period, the São Paulo slave owners tended to be more evenly distributed and the more than two-thirds of the owners who owned fewer than six slaves accounted for a very significant third of all slaves, as can be seen in a census of 1777–1778 (see Table 3.11). This difference was due to the greater poverty of the paulistas slave owners and to their concentration in agriculture, with little significant mining.

The Paulista's discovery of the Minas gold fields initially created a boom in local foodstuff production and finally provided the capital to get local sugar production up to international standards and encouraged the beginnings of an important local sugar industry. But the opening of more direct routes from Rio de Janeiro to the Minas gold fields (the so-called caminho novo that was opened in stages from 1707 to 1720), and the fact that only eighteen thousand persons occupied the paulista plains

TABLE 3.11. Distribution of Slaves and Owners by Size of Slaveholdings in Seven Districts in São Paulo, 1777–1778

Source: Arquivo Público do Estado de São Paulo, Mapas de População.

meant that São Paulo could not respond fast enough to the demands of the mining markets, so it was replaced by the provincial producers in Rio de Janeiro. This involved Rio de Janeiro producers in everything from supplying foodstuffs and locally produced sugar to Rio becoming the chief port for all of the interior mining provinces’ imports (slaves included) and exports. In turn, São Paulo's economy also became more dependent on the far larger Rio de Janeiro market for its capital and markets.

Given the slower growth of the paulista African slave population with relatively lower levels of African immigration compared to that of Minas, it is not surprising to find a more balanced sex ratio among the provincial slaves in the eighteenth century. Although knowledge about the specific origin and/or color is limited for the slaves in the census of 1777–78, the overall sex ratio of the 22,607 whose age was known was 143 male slaves per 100 female slaves, and the overall age of the 21,602 slaves whose age was known was a relatively high average of 26.5 years. Although these age and sex ratios were less biased than the comparable age and sex ratios of the slave population of Minas in the first half of the eighteenth century, these numbers still show that slaves arriving directly from Africa were having an important influence on the local population. The few data we have from this census on the origin and color of the slaves (for fewer than two thousand of these slaves) suggest a sex ratio for the

GRAPH 3.3. Age pyramid of slaves in São Paulo, 1777–1778.

African-born slaves of 155 males per 100 females, compared to 99 males per 100 females among the Creole or native-born slaves, with their average age being 36 years for Africans compared to 21 years for the native-born ones. The total slave population showed this African influence in terms not only of the sex distribution, but also the bunching of the population in the most important economically active ages – and the relatively low percentage of children among these Paulista slaves – only 29 percent were children of either sex under 15 years of age (see Graph 3.3).

Although Rio de Janeiro profited most from the opening of the interior gold mines, the backward linkages of the mining sector resulted in a more even distribution of population within Brazil and the spread of slavery to all sectors of the colonial economy. Slaves now reached the frontier working in food producing farms, and they also joined the burgeoning cattle industry in both the central northern coastal region as well as in the new cattle zones of the southern pasturelands.

The case of the southern province of Rio Grande do Sul was typical of these developments. The early part of the century brought an active opening up of the southern grasslands of the region, for both political reasons to prevent Spanish expansion northward and as a response to the demands from Minas Gerais. By the end of the century, there were some 21,000 slaves and 5,000 free persons of color, both forros and those born free in a population of 71,000. The slaves were linked into the export sector of the economy. Whereas the cowboys on the cattle ranches were mostly Indians or free peon gauchos, the salting and beef-drying establishments were run with slave labor. Jerked or dried beef (called charque) was produced in special factories (charqueadas), which usually used from sixty to ninety slaves. By the early nineteenth century, these charqueadas of the Rio Grande do Sul region were in full production and were employing some five thousand slaves. The market for the dried beef was domestic, because Brazilian products did poorly in competition with Spanish output from the Rio de la Plata region. The consumers of Brazilian jerked beef were almost always slaves, the dried beef of Rio Grande do Sul being a major source of protein in the diets of plantation and mining slaves in central and northern Brazil.⁵⁷

Whereas the gauchos of the cattle ranches of the Rio Grande do Sul region were mostly free and Indian laborers, those further to the north in the so-called Campos Gerais area – the area around the city of Curitiba – employed both free and slave labor in the ranches where both mules and horses as well as cattle were kept. Slaves would also be employed in various crafts needed on the ranch and transporting products to and from the fazendas.⁵⁸ In this entire region of southern São Paulo known as Paraná, the overall ratio of slaves to total population was 20 percent by the first available census of 1798, whereas in the local ranching districts of Castro and Palmeira, the rates of slave ownership were a high 52 percent and 39 percent, respectively, of all households.⁵⁹ On these and on the ranches further south, there were also a large number of free colored servants, employees, and dependents who worked on the estates, usually in less export-oriented capacities. Finally, in all the southern towns, some

TABLE 3.12. Distribution of Slaves and Slave Owners in Paraná, 1804

Source: Arquivo Público do Estado de São Paulo, Mapas de População.

of which were reaching the ten-thousand level by late in the century, slaves formed the largest single element in the workforce and were the majority of skilled craftsmen. The three southern provinces of Santa Catarina, Rio Grande do Sul, and Paraná had a combined slave population of some 27,000 slaves and some 13,000 free persons of color by 1811.⁶⁰

A census of Paraná in 1804 provides a more complete view of this frontier ranching economy. With a relatively small population of slaves, it was evident that the cattle breeders and fatteners – the ranchers – were the elite of the society and concentrated a large group of local slaves. They were mostly middle-sized slaveholders with an average of eight slaves per unit, and although they made up just 17 percent of all slave owners in the province, they controlled a quarter of the provincial slave population in this year (see Table 3.12).

By this period, the ratio of Africans in the labor force of this region was relatively small, accounting for only 13 percent of the local slave force. Moreover, this was an aging population whose average age was nineteen years older than the average Creole slave, who was just twenty-one years of age. By this time, the Creole slaves had a sex ratio of 97 males to 100 females, compared to the still highly imbalanced ratio of the Africans (who had 147 males to 100 females). Without more massive imports of Africans, it appears that the expanding local slave population was progressively being supplied by children born to the local Creole population.

The southern grasslands provided a new area for slave labor to supply mules and other animals to the mining economy and the rest of the society. But once settlement got underway to the south, local industries could be developed, which created a new labor market for slaves. With strong settlements now established along the coast of Rio Grande do Sul and Santa Catarina to support the grazing industry, Brazilians began to engage in commercial fishing activities with important slave participation. Although offshore coastal whaling had been practiced in Brazil from the beginnings of colonization, the industry became a major factor only when the southern provinces were successfully opened to colonization in the eighteenth century. From Cabo Frio in the province of Rio de Janeiro south to Laguna in Santa Catarina, whaling became a major industry from the second half of the eighteenth century until the first decades of the next century. The center of the industry was the island of Santa Catarina in the province of the same name, which had a commercial whale oil-producing factory (or armação) as early as 1746. By the 1770s, the region of Santa Catarina alone was taking over one thousand whales per annum. During the June to September whaling season, free colored, poor white, and slave fishermen in open boats did the harpooning and bringing of the whales inshore. Once beached, they were then cut down and boiled for their oil, which was sold both nationally and internationally for use in illumination. These very costly and elaborate cutting and boiling factories were run mostly with slave labor. A typical armação was a major operation, on average employing between fifty and one hundred slave workers. One of the biggest in the early nineteenth century was the Armação de Nossa Senhora da Piedade on the island of Santa Catarina, which owned 125 slaves, of whom 107 were working adults.⁶¹ Along with unskilled laborers, the slave workforce included free colored and white carpenters, blacksmiths, and coopers, as well as the specialized skills relating to the cutting of the whale and the production of the spermaceti. The seven leading whaling companies on the island in 1817 held a total of 329 slaves.⁶² Although a highly seasonal occupation, the factories could employ as many as two to three thousand slaves in a good season.

The opening up of the Brazilian interior stretched Brazilian settlement both southward and westward and also encouraged the creation of major transportation networks to tie these vast markets together. Slaves were vital in the large canoe fleets and mule trains made up by the coastal and southern merchants to supply the enormous import needs of the interior mining provinces. Given the poor records of these activities, it is difficult to estimate the number of slaves involved. But another major area of transport fostered by the interior and southern markets was coastwise shipping. In this case, there is a basis of estimating the relative role of the African- and American-born slaves. Contemporary reports list high rates of participation of slaves as sailors in all types of coastal shipping. A rough estimate of interregional coastwise shipping at the end of the eighteenth century would suggest the number of vessels employed at approximately two thousand ships. Assuming a minimum of five slaves per crew on these ships (or one-third of the average coastal trader's complement of sailors), then something like ten thousand slaves were sailors involved in cabotagem trade in the late eighteenth century.⁶³

Brazil was also rather unusual in its use of slave sailors in international shipping as well, especially so in its Atlantic slave-trade routes. Because of its direct trading relations with Africa, in which no triangular linkages existed with Portugal, Brazil developed a very powerful merchant marine early on. Hundreds of Brazilian-owned ships plied the South Atlantic, taking Brazilian aguardente, gunpowder, tobacco, and European and American manufactured goods to Angolan and Mozambican ports and exchanging them for slaves, who were then brought to Brazil. Brazilian-owned vessels also controlled most of the carrying trade to Europe, in sharp contrast to the Spanish American areas. Given the crucial role slaves played in all aspects of the Brazilian economy, it was no accident that even on slavers there were typically slaves listed as members of the crew. In 147 of the 350 slave ships that arrived in the port of Rio de Janeiro between 1795 and 1811, Brazilian-owned slaves were listed as crew members. These slaves numbered 2,058 out of the 12,250 sailors engaged in the trade. On average, there were fourteen slave sailors per ship, or just under half the total crew on a typical slaver. Because the registers always justified the need to use slaves due to the lack of free sailors, this would suggest that slaves were even more important in the other international routes of the period.⁶⁴ Nor was Rio de Janeiro unique in this respect. The governor of Bahia in 1775 reported to the Crown that the port of Salvador only had 678 sailors who were free persons, and although the majority of these were white, they were insufficient to man the slave ships leaving for Africa, so that “the navios and corvetas which carry out our commerce with Africa, usually equip themselves with a small crew of four to seven white sailors and make up the rest of the crew with black slaves.”⁶⁵

The growth of mining was matched by the development of new agricultural products coming from newer production zones as well as the revival of the northeastern sugar industry and the growth of a major new center of sugar production in the south. All this growth of traditional and new exports led to a major expansion of the colonial economy of Brazil in the second half of the eighteenth century. The emergence of a dynamic administration in Portugal under the Marques de Pombal from 1750 to 1777 also brought about the further development of the Brazilian economy and a new slave-based industry in the north of the country. A typical Enlightenment regime, the Pombal administration used classic mercantilist procedures to encourage the growth of previously neglected regions of Brazil. With the interior and the south booming, it turned its attention to the major northeastern regions of Pará and Maranhão, which until the second half of the eighteenth century were backward and sparsely settled areas. In 1755 and 1759, respectively, he created two major monopoly trading companies: the Grão-Pará e Maranhão Company and the Companhia Geral de Pernambuco e Paraíba. Both were given economic support by being allowed monopoly rights to slave importation into these two regions – the only break in the usual free-trade policy that Portugal allowed. In turn, these companies were required to invest in the commercial development of the northeastern regions. All told, these companies imported some eighty-five thousand Africans to the four northeastern regions of Pará, Maranhão, Pernambuco, and Paraíba in the period 1756–1787.⁶⁶

After much experimentation, a major new export crop was developed under Pombaline company initiatives in both Maranhão and Pernambuco. This was cotton, which was produced on plantations using slave labor. At approximately the same time cotton was developing in the British colonies with the aid of slave labor, it was also becoming a major staple export of Brazil. Beginning in the 1760s, Maranhão cotton plantations began to export to Europe. Production rose steadily in the next decades and quickly spread to the neighboring province of Pernambuco. The typical cotton plantation in these two states contained fifty slaves per unit, not too different from what the average size of a cotton plantation in the southern states of the United States would be in the nineteenth century. With the steady increase in European prices came a continuous increase in production. So aggressive was the Brazilian response that by the early 1790s it accounted for 30 percent of British raw cotton imports.⁶⁷ By the first decade of the nineteenth century, more than thirty thousand slaves were involved in cotton production in the northeastern states. The cotton plantation system continued to expand for two decades more until ginned U.S. cotton production wiped out its comparative advantage and brought a long-term decline to the industry.

The efforts of the Pombaline companies were also important in finally reviving the sugar plantation economy in Pernambuco in the 1770s and 1780s. Although Pernambuco never regained its dominant position in the industry, it became the second-largest northeastern producer after Bahia. In turn, Bahia had grown not only from the expansion of its sugar industry but also from its close ties with Minas Gerais. Its location near the São Francisco River, the only major inland river route to the mines, guaranteed steady contact with the mines. At first, the Crown tried to prevent trade with the mines and feared for the loss of crucial slave labor from the plantations. However, the rise of sugar prices after 1711 eased the pressure on the Bahian sugar industry, so the Crown lifted its ban on the sale of Bahian slaves to the interior. Trade with the mines also encouraged the expansion of the interior northeastern manioc and foodstuffs frontier and promoted the growth of an important livestock industry, which now supplied both the coastal plantations and the interior mines.

But the major change in sugar in the late eighteenth century was not so much the revival of the older northeastern region as the growth of new sugar products and new sugar production regions. Rio de Janeiro and São Paulo became the centers of production of both mascavo (brown sugar) and of aguardente (brandy made from sugar). Although sugar had been cultivated in the Campos dos Goitacases region of Rio de Janeiro for well over a century, there began a major expansion of the sugar estates in the second half of the eighteenth century. By 1779, Campos already had five sugar estates that contained more than 100 slaves each, with an average of 15 slaves per unit in 159 large and small sugar-producing engenhos. Moreover, slaves represented more than half of the estimated thirty thousand local population.⁶⁸ At the end of the colonial period, Rio de Janeiro would rank third in Brazilian production and account for two-thirds of mascavo sugar output. It was also Brazil's major producer of aguardente, which was exported to Africa as well as supplying the internal market. By this time, Rio de Janeiro had some eighty-four thousand slaves,⁶⁹ of whom possibly a quarter were employed in all aspects of its sugar industry. Neighboring São Paulo, although only a moderate sugar producer and slave zone at the end of the century, finally began exporting from both its coastal enclaves and highlands to the west of the city of São Paulo into the international market in this late eighteenth-century period, and this evolution marked the beginning of what would prove to be the most important slave and plantation region of Brazil in the nineteenth century.

Despite the growth of new sugar production areas and the fact that sugar still accounted for one-third of the value of all Brazilian exports, the industry was relatively depressed through most of the eighteenth century. Whereas colonial production was still averaging some 36,000 tons per annum in the 1730s, by the 1770s it was down to 20,000 tons and probably accounted for less than 10 percent of total American sugar output. Nevertheless, volume and importance fluctuated during the eighteenth and early nineteenth centuries. In the middle of the eighteenth century, Brazil's 27,000 tons per annum output placed it in third place behind Saint Domingue (at 61,000 tons) and Jamaica (at 36,000 tons). It also became an important alternative source for northern European markets in the frequent imperial wars that France and England fought in the eighteenth century, which temporarily would halt the West Indian trade to Europe. Thus, in the 1760s, Brazilian sugar captured about 8 percent of Europe's market for sugar and in the warfare of the 1790s, took a 15 percent share of the market. This continued vitality of the Bahian and Rio de Janeiro sugar plantations guaranteed that even with the massive growth of mineral exports in the eighteenth century, when Brazil became the world's greatest single source for gold, sugar still represented the single most valuable Brazilian export and alone accounted for half the value of its total exports. At this time, the number of slaves involved in all forms of sugar production, which involved both the exporting of finished white sugar and the semi-processed brown sugar, as well as the production of aguardente for both national consumption and export to Africa, was probably fewer than 100,000 persons. By the early 1780s, European tensions and the disruptions of trade were beginning to affect prices and to encourage national production, and in the 1790s, the profound impact of the French Revolution and the subsequent Haitian revolution would create a new era of expansion for Brazilian sugar.⁷⁰

The final major development in the colonial economy of Brazil related to slave labor was the surprising diversification that was taking place in the province of Minas Gerais by the end of the eighteenth century. As first gold output and then diamond production declined after the middle decades of the century, the mineiro economy was faced with a serious economic crisis. By the first decade of the nineteenth century, mining for both gold and diamond production was in full decline, yet the slave population of the province at this time stood at more than 150,000 persons and would continue to grow for the rest of the century. Urban decay had set in with the decline in mineral extraction and diminished even further the opportunities for slave use. The free colored population, moreover, was now employed everywhere and was greater in number than the slaves. Yet the slave population continued to grow at a steady pace through the nineteenth century, and by the time of abolition at the end of the century, it had more than doubled, which meant that at both the beginning and the end of the nineteenth century, Minas Gerais had the largest slave population of any province in Brazil.⁷¹

The major developments that accounted for Minas retaining and expanding its slave labor force seem to have been a combination of diversification in agricultural production, which supplied the internal market, and then, several decades later, an expansion into coffee for international export. In the southern and eastern regions of the province, a diversified agriculture developed in the late eighteenth and early nineteenth centuries based on slave production. Sugar, coffee, staples, and cattle were produced in Minas on farms using slave labor.⁷² The total number of slave owners in the free population was higher and the number of slaves held per owner was lower in Minas than in the coastal provinces, and under the impact of agricultural diversification, this pattern was accentuated even further. What this diversification meant for the development of slavery in Minas Gerais in the nineteenth century has created a very important debate within Brazilian historiography, a theme that will be treated in the next chapter.

Although much of mineiro economic history is still poorly understood, the vitality of slavery in its borders in the late eighteenth and early nineteenth centuries made for a nontraditional and highly unusual slave economy by American standards. Some have even argued that slavery was essentially dedicated to subsistence agriculture from the late eighteenth-century decline of mining to the mid-nineteenth century rise of commercial coffee production, but this seems too extreme a position.⁷³ More likely, it would appear that local output was being successfully exported into a national market and that Minas Gerais had reversed the direction of its relations with the coastal economy, for it now became a major supplier of the foodstuffs needed to run the coastal plantation regimes, but these are themes that will be explored in the next chapter.⁷⁴

In all of Brazil by 1800, there were now close to one million slaves. Brazil thus held the largest single concentration of African and Creole slaves in any one colony in America and also accounted for probably one of the most diverse economic usage of slaves to be found in the Western Hemisphere. Although a detailed breakdown of the slave population by economic activity is always difficult, it is evident that no more than one-quarter of all the slaves were to be found in plantations or mines. The rest were spread widely through the cities and rural areas of the nation, engaged in every possible type of economic activity. As many as 10 percent of the total slave population may have had an urban residence, but the rest were involved in rural activities, employed in farming, fishing, transportation, and every conceivable type of occupation. Brazil, with its half a million free colored, was also the largest center of the new class of black and mulatto freedmen in America. Although sugar, gold, diamonds, and other export products went through the classic colonial boom-bust cycles, the vitality of the Brazilian economy was such that new products were developed, new regions opened up, and a lively internal market created. All this guaranteed that the flow of slaves would not cease. In the last decade of the century, an estimated 28,000 African slaves were arriving annually in the ports of Brazil, above all Rio de Janeiro and Salvador da Bahia. By the first decade of the new century, that number would rise to 34,000 per annum and would keep increasing every decade until the 1830s.⁷⁵ Brazil was also home to a thriving free colored population, which at this time numbered almost 500,000 persons. Without question, then, Brazil in 1800 had the largest population of Africans and Afro-Americans among the European colonies and was the largest slave system in the Americas.⁷⁶

¹ During the last thirty years of the seventeenth century, the Crown directly involved itself in the question of gold production. There is abundant documentation with respect to this theme in the correspondence between the Court and the paulistas. The Crown furnished some resources, sending some mining specialists, offering prizes to the discoverers, and establishing the legal norms in mining, but it aimed to have the paulistas invest most of the capital and effort. See Pedro Taques de Almeida Paes Leme, Notícias das Minas de São Paulo e dos sertões da mesma capitania (São Paulo: Prefeitura do Município de São Paulo, 1954). For the contract written by Bartolomeu Bueno da Silva with the Crown to discover gold in Goiás in 1720, see Alida C. Metcalf, Families of Planters, Peasants, and Slaves: Strategies for Survival in Santana de Parnaíba, Brazil, 1720–1820 (Austin: University of Texas Press, 1983), p. 57.

² For a good introduction to the evolution of the mining economy in Minas in this period, see Charles R. Boxer, The Golden Age of Brazil, 1695–1750: Growing Pains of a Colonial Society (Berkeley: University of California Press, 1962); Virgílio Noya Pinto, O ouro brasileiro e o comércio anglo-português (São Paulo: Companhia Editora Nacional, 1979); and Sérgio Buarque de Holanda, “Metais e pedras preciosas,” in Sérgio Buarque de Holanda, ed., História geral da civilização brasileira (10 vols.; São Paulo: Difusão Européia do Livro, 1960), tomo I, vol. 2, chapter 6, pp. 259–310; and Francisco Vidal Luna, Minas Gerais: Escravos e senhores (São Paulo: FEA-USP, 1980); Francisco Vidal Luna and Iraci del Nero da Costa, Minas Colonial: Economia e Sociedade (São Paulo: FIPE/PIONEIRA, 1982); João Pandiá Calógeras, As Minas do Brasil e sua legislação (Rio de Janeiro: Imprensa Oficial, 1905); W. L. von Eschwege, Pluto Brasiliensis (São Paulo: Ed. Nacional, 1944, 2 vols.); Alice P. Canabrava, “João Antonio Andreoni e sua obra,” in André João Antonil, Cultura e Opulência do Brasil. Introdução e vocabulário por A. P. Canabrava (São Paulo: Ed. Nacional, s/d).

³ A good survey of the available census data for Minas Gerais in the eighteenth and early nineteenth centuries is found in Eduardo França Paiva, Escravos e libertos nas Minas Gerais do século XVIII (São Paulo: Annablume, 1995), pp. 67–78. For a rather detailed census of Minas Gerais in 1814 with birth and death listings, see ANRJ, cod. 808, 1, fols. 130–2.

⁴ Laird W. Bergad, Slavery and the Demographic and Economic History of Minas Gerais, Brazil, 1720–1888 (Cambridge: Cambridge University Press, 1999), table 3.2, p. 91.

⁵ According to Carlos Leonardo Kelmer Mathias, already by the period 1711–1717, Minas Gerais exhibited a higher value of slave sales than occurred in Rio de Janeiro in the same period (1711–1720), which shows the intense nature of economic activity in Minas in these first twenty years of its colonization. Mathias, “O perfil econômico da Capitania de Minas Gerais na segunda década do século XVIII, notas de pesquisa, 1711–1720,” in Anais do XII Seminário sobre a Economia Mineira (2006), pp. 6–8.

⁶ Dauril Alden, “Late Colonial Brazil, 1750–1808,” in Leslie Bethell, ed., The Cambridge History of Latin America (11 vols.; Cambridge: Cambridge University Press, 1984), II, table 3, p. 605.

⁷ Luna, Minas Gerais: Escravos e senhores, p. 82, tabela 9.

⁸ Francisco Vidal Luna, “Estrutura da Posse de Escravos em Minas Gerais (1804),” in Iraci del Nero da Costa, ed., Brasil: História Econômica e Demográfica (São Paulo: IPE/USP, 1986), tabela 7.

⁹ Francisco Vidal Luna and Iraci del Nero da Costa, “Algumas características do contingente de cativos em Minas Gerais,” in Anais do Museu Paulista (São Paulo: USP) (1979) tomo XXIX, tabelas 5, 7.

¹⁰ Luna, Minas Gerais: Escravos e senhores, p. 39 and footnote 9, p. 60.

¹¹ According to the “Regimento de 1702, artigo 5°,” which announces the discovery of an area of potential gold mines, it was required that the Superintendente das Minas and his representative the Guarda-Mor were to divide any new mining region by lottery among interested wealthy miners: “regulando-se pelos escravos que cada um tiver que em chegando a doze escravos e daí para cima, fará repartição de uma data de trinta braças [corresponde a 66 metros] conforme o estilo e àquelas pessoas que não chegarem a ter doze escravos lhes serão repartidas duas braças e meia por cada escravo para que igualmente fiquem todos lucrando da mercê que lhes faço (…),” which corresponds to 5.5 meters per slave. The original discoverer of the gold deposit was to get first choice of mine along with one mine reserved for the Crown before the lottery began and none could participate if they did not have the twelve slave miners. For the details of the law, see Leme, Notícias das Minas de São Paulo e dos sertões da mesma capitania, p. 190; also see Luna, Minas Gerais: Escravos e senhores, p. 39.

¹² For censuses up to 1759, the sex ratio among slaves was usually in the 500+ males to 100 females ratio; see Francisco Vidal Luna, “Estrutura da Posse de Escravos,” in Francisco Vidal Luna and Iraci del Nero da Costa, Minas Colonial: Economia e Sociedade, p. 51, tabela 11.

¹³ Francisco Vidal Luna, “Estrutura da Posse de Escravos em Minas Gerais (1718),” in A.E.M. Barreto et al., História Econômica: Ensaios (São Paulo: IPE/USP, 1983), tabelas 7 and 9.

¹⁴ Kathleen J. Higgins, “Licentious Liberty” in a Brazilian Gold-Mining Region: Slavery, Gender, and Social Control in Eighteenth-Century Sabará, Minas Gerais (University Park: Pennsylvania State University Press, 1999), pp. 74–5, tables 2.7 and 2.8.

¹⁵ Luna, “Estrutura da Posse de Escravos em Minas Gerais (1804),” tabela 6.

¹⁶ Francisco Vidal Luna and Iraci del Nero da Costa, “Estrutura da Massa Escrava de Algumas Localidades Mineiras (1804),” in Revista do Instituto de Estudos Brasileiros (1981), tabela 2.

¹⁷ Mariana L. R. Dantas, “Black Townsmen: A Comparative Study of Persons of African Origin and Descent in Slavery and Freedom: Baltimore, Maryland and Sabará, Minas Gerais, 1750–1810,” (Ph.D. diss.; Baltimore: Johns Hopkins University, 2003), p. 132, table 2.4.

¹⁸ Iraci del Nero da Costa, Vila Rica: População (1719–1826) (São Paulo: IPE/FEA-USP, 1979), p. 245, tabela 5.

¹⁹ On the organization of the mining industry and government policy, see Francisco Vidal Luna, “Mineração: Métodos extrativos e legislação,” Estudos Econômicos, vol. 13 (1983), pp. 845–59; and “Economia e Sociedade em Minas Gerais (Período Colonial),” in Revista do Instituto de Estudos Brasileiros, vol. 24 (1982), pp. 33–44. On the technology used to extract gold in Minas in this period, see the detailed study by Flávia Maria da Mata Reis, “Entre faisqueiras, catas e galerias: Explorações do ouro, leis e cotidiano das Minas do Século XVIII (1702–1762),” (Dissertação de mestrado, Belo Horizonte: FFCH-UFMG, 2007). The Crown even went so far as to declare in 1752 that all miners who owned thirty or more slaves – a distinct minority – were exempted from their slaves or mining equipment being seized for debt. Boxer, The Golden Age of Brazil, p. 184.

²⁰ There were already almost 21,000 slaves in Vila Rica by 1735. See Alda Maria Palhares Campolina, Cláudia Alves Melo, and Mariza Guerra de Andrade, Escravidão em Minas Gerais: Cadernos do Arquivo (Belo Horizonte: Arquivo Público Mineiro/COPASA MG, 1988), p. 31.

²¹ A.J.R. Russell-Wood, “Colonial Brazil: The Gold Cycle, c. 1690–1750,” in Leslie Bethell, ed., The Cambridge History of Latin America (11 vols.; Cambridge: Cambridge University Press, 1984), II, p. 573. Detailed distraction of the extractive methods can be found in Francisco Vidal Luna, Minas Gerais: Escravos e senhores; João Pandiá Calógeras, As Minas do Brasil e sua legislação; Eschwege, Pluto Brasiliensis; André João Antonil, Cultura e Opulência do Brasil, Introdução e vocabulário por A. P. Canabrava (São Paulo: Ed. Nacional, s/d).

²² Eschwege, Pluto Brasiliensis, vol. II, pp. 20–49.

²³ Bergad, Slavery and the Demographic and Economic History of Minas Gerais, table 3.2, p. 91.

²⁴ Gilka V. F. Salles, Economia e escravidão na Capitania de Goiás (Goiânia: CEGRAF/UFG, 1992), p. 231. On the gold mining developments in this province in the eighteenth century, see Eschwege, Pluto Brasiliensis, I, pp. 88–118. Interestingly, the overwhelming majority of the slaves in early eighteenth-century Goiás were Africans from the Bight of Benin, the so-called Mina slaves, almost all of whom were brought into the province from the port of Salvador. Maria Lemke Loiola, “Trajetórias Atlânticas, percursos para a Liberdade: Africanos e Descendentes na Capitania dos Guayazes” (Dissertação de mestrado, FCHF, Universidade Federal de Goiás, Goiânia, 2008), tabela 1, p. 35 and figura 4, p. 46.

²⁵ Lucia Helena Gaeta Aleixo, “Mato Grosso: Trabalho escravo e trabalho livre (1850–1888)” (Dissertação de mestrado, PUC São Paulo, 1984), p. 63, tabela 13. On the gold mining developments in this province in the eighteenth century, see Eschwege, Pluto Brasiliensis, I, pp. 119–36.

²⁶ On “wages” for slave miners and other positive incentives, see Boxer, The Golden Age of Brazil, pp. 162–203.

²⁷ Russell-Wood, “Colonial Brazil,” II, pp. 581–2; Eschwege, Pluto Brasiliensis, II, p. 126.

²⁸ Francisco Vidal Luna and Iraci del Nero da Costa, “Demografia Histórica de Minas Gerais no Período Colonial,” Revista Brasileira de Assuntos Políticos (Belo Horizonte, UFMG), vol. 58 (1984), tabela 5.

²⁹ These ratios were roughly the same as was found for a census of Sabará in 1720 for 894 masters and 5,908 slaves. Some 78 percent of the masters owned nine or fewer slaves, but their slaves made up only 42 percent of all slaves. Higgins, “Licentious Liberty,” p. 49, table 2.2.

³⁰ In both cases, the number of owners whose literacy was not known was quite small, only thirty-two in Vila Rica and sixteen in Vila do Carmo. Rodrigo Castro Rezende, Mariângela Porto Gonçalves, Regina Mendes Araújo, and Karina Paranhos da Mata, “Os proprietários de escravos nas Minas Gerais em 1718–1719: Um estudo comparativo dos distritos de Vila do Carmo e Vila Rica,” XIII Encontro da Associação Brasileira de Estudos Populacionais, ABEP (2002), tabelas 7–10, pp. 19–21.

³¹ This differed from the important sugar zone of Itu in São Paulo, where the value and size of estates for women and men were roughly equal; see Joseph Cesar Ferreira de Almeida, “Entre engenhos e canaviais: Senhoras do açúçar em Itu (1780–1830)” (Dissertação mestrado, FFLCH-USP, 2007), tabela 5, p. 55.

³² Higgins, “Licentious Liberty,” p. 81, table 2.11.

³³ Luna, “Estrutura da Posse de Escravos em Minas Gerais (1718),” tabelas 1–5.

³⁴ For a detailed analysis of the evolution of the taxing structure in gold, see Luna, “Mineração,” pp. 845–59; and for the gold and diamond taxing and monopoly arrangements, see Boxer, The Golden Age of Brazil, chapters 7 and 8.

³⁵ Boxer, The Golden Age of Brazil, pp. 317–18.

³⁶ Luna, Minas Gerais: Escravos e senhores, p. 133, tabela 38.

³⁷ Arquivo Nacional, Rio de Janeiro, “Rol das Pessoas que Confessam e Comungam na Freguesia de Congonhas do Sabará, 1771,” Caixa 202, pacote único.

³⁸ “[P]orque estes vadios, que em outra parte seriam prejudiciais, são ali úteis. Negros forros e mestiços na sua maior parte – mulatos, caboclos, carijós – , serviam para povoar locais distantes como Cuiete, Abre Campo e Peçanha, onde se iam estabelecendo presídios; engrossavam os contingentes que entravam mato adentro destruindo quilombos e prendendo foragidos; cultivam plantaçöes de subsistência, enfim, realizavam uma série de tarefas que não podiam ser cumpridas pela mão-de-obra escrava.” Reproduced in Laura de Mello e Souza, Desclassificados do Ouro: A Probreza Mineira No Século XVIII (4th ed.; Rio de Janeiro: Graal, 2004), p. 104.

³⁹ On royal officials and their expressed opposition and/or ambivalence to the free colored population, the standard work is that of Mello e Souza, Desclassificados do Ouro, see especially pp. 144ff. Also see Boxer, The Golden Age of Brazil, pp. 164ff. As the royal Governor Lourenço de Almeida declared, “even the so-called great ones [of the province of Minas Gerais] being bred in the milk of servitude” (p. 164). The Crown went so far as to prohibit free persons of color from being members of the municipal council, tried passing sumptuary laws against black and brown women, and even tried to prevent free colored from having arms, although permitting capitães do mato to organize armed bands and of course putting free persons of color into militias organized on color grounds.

⁴⁰ Caio César Boschi, Os Leigos e o Poder (Irmandade Leigas e Política Colonizadora em Minas Gerais) (São Paulo: Editora Ática, 1986), pp. 19–20. As he noted, “ser admitido numa ordem terceira significava pertencer à ‘elite social’ e ser de ‘origem racial branca e católica incontestável.’” Moreover, the tertiary order of Nossa Senhora do Carmo attracted the miners and elite merchants, whereas the order of S. Francisco de Assis was the preferred association of military men and bureaucrats. Boschi, Os Leigos e o Poder, pp. 162, 164. This social division between irmandades and ordem terceiras associations explains the domination of the former organization and the relative lack of the latter ones in colonial Minas.

⁴¹ Boschi, Os Leigos e o Poder, pp. 187–8. Also see Elizabeth W. Kiddy, Blacks of the Rosary: Memory, and History in Minas Gerais, Brazil (University Park: Pennsylvania State University Press, 2005); Célia Maria Borges, Escravos e libertos nas irmandades do Rosário: Devoção e solidaridade em Minas Gerais, séculos XVIII e XIX (Juiz de Fora: Editora da UFJF, 2005); and Julita Scarano, Devoção e escravidão – a Irmandade de Nossa Senhora do Rosário dos Pretos no Distrito Diamantino no século XVIII (São Paulo: Ed. Nacional, 1976).

⁴² Boschi notes that the precarious nature of the official church presence in Minas meant that the irmandades were in fact the ones solely responsible for the erection of churches in the local towns. Boschi, Os Leigos e o Poder, p. 23.

⁴³ For an introduction to the urban architecture of this culture, see Suzy de Mello, Barroco mineiro (São Paulo: Editora Brasiliense, 1985).

⁴⁴ For a recent review of his life and times, see Ana Helena Curti, ed., Aleijadinho e seu tempo: Fé, engenho e arte (Rio de Janeiro: Banco Central do Brasil, 2006), and for the latest catalog of his religious sculptures, Myriam Andrade Robeiro de Oliveira, et. al, O Aleijadinho e sua oficina: Catálogo das esculturas devocionais (São Paulo: Editora Capivara Ltda., 2002).

⁴⁵ On the musicians, see Daniela Miranda, “Músicos de Sabará: A prática musical religiosa a serviço da Câmara (1749–1822),” (Tese de doutorado; Belo Horizonte: FFCH/UFMG, 2002), and Francisco Kurt Lang, “A musica barroca,” in Sergío Buarque de Holanda, ed., História geral da civilização brasileira, II, pp. 121–44.

⁴⁶ It has been pointed out that diamonds had probably been known and exploited before this date, but that this was the official date of announcement of the discovery to the Crown. Buarque de Holanda, “Metais e pedras preciosas,” pp. 241–2. A good survey of the establishment of the exclusive diamond district and the far greater royal control exercised over this mining activity compared to that of gold is found in Boxer, The Golden Age of Brazil, chapter 8.

⁴⁷ The Crown initially limited those renting the annual royal diamond monopoly from using more than six hundred slave miners – a policy designed to keep production low and prices high. Eschwege, Pluto Brasiliensis, II, p. 120. In fact, in the key diamond mining district of Serro do Frio, the average slaveholding in 1734 was five slaves (Arquivo da Casa dos Contos, “Serro do Frio: Escravos, Livro de Matrícula.” Codice no. 1068). This compared to an average of seven slaves in four gold mining districts in 1718. Luna, “Estrutura da Posse de Escravos em Minas Gerais (1718),” p. 6, tabela 7.

⁴⁸ The standard source on the provisioning of the mines of Minas Gerais is Mafalda P. Zemella, O abastecimento da Capitania das Minas Gerais no século XVIII (São Paulo: Hucitec-Edusp, 1990) and recently, for a case study of individual merchants, Cláudia Maria das Graças Chaves, Perfeitos negociantes, mercadores das minas setecentistas (São Paulo: Annablume, 1999).

⁴⁹ On the population of Mexico City in this period, see Herbert S. Klein, “The Demographic Structure of Mexico City in 1811,” Journal of Urban History, 23, no. 1 (November 1996), pp. 66–93.

⁵⁰ Zemella, O abastecimento da Capitania das Minas Gerais, chapter 8.

⁵¹ On the settlement and social and economic evolution of the colony of São Paulo, see the fundamental works of Sérgio Buarque de Holanda, Caminhos e Fronteiras (2nd ed.; São Paulo: Cia. das Letras, 1995); Monções (2nd ed.; São Paulo: Alfa Omêga, 1976); Visão do Paraíso (São Paulo: Editora Brasiliense, 1994); “Movimentos de população em São Paulo no Século XVIII,” Revista do Instituto de Estudos Brasileiros, 1 (1966), pp. 55–111; and Raízes do Brasil (Rio de Janeiro: José Olympio Editora, 1956); those of Alfredo Ellis Jr., “Ouro e a Paulistania,” Boletim de História da Civilização Brasileira, 8, (1948); “O Ciclo do Muar,” Revista de História, vol. I (São Paulo, 1950) and with Miriam Ellis, “A economia paulista no século XVIII,” Boletim de História da Civilização Brasileira, 11, 1950; Leme, Notícias das Minas de São Paulo e dos sertões da mesma capitania; the studies of Caio Prado Junior, Evolução Política do Brasil e Outros Estudos (8th ed.; São Paulo: Brasiliense, 1972) and those of Alcântara Machado, Vida e Morte do Bandeirante (São Paulo: Livraria Martins, 1965) and Cassiano Ricardo, Marcha para Oeste (Rio de Janeiro: José Olympio, 1942). Along with these classic works, there are the studies of Alice P. Canabrava, “Uma economia de decadência: Os níveis de riqueza na Capitania de São Paulo, 1765–67,” Revista Brasileira de Economia (Rio de Janeiro), 26, no. 4 (Out./Dez. 1972), pp. 95–123; and “A repartição da terra na capitania de São Paulo, 1818,” Estudos Econômicos, 2, no. 6 (Dez. 1972), pp. 112–15.

⁵² Only at the beginning of the eighteenth century did Africans become numerically important in São Paulo. Alcântara Machado, who did an elaborate study of the wills and testaments in the first centuries of Paulista colonization, found approximately 100 Africans in these documents in the seventeenth century. Interestingly, their values were always superior to those for the Indian slaves. Machado, Vida e Morte, p. 181. Ellis Jr., “O Ouro e a Paulistania,” p. 4, analyzed two thousand inventories in the sixteenth and seventeenth centuries and found fewer than three hundred Africans. Queiroz, from a study of alternative sources, found Africans in the earliest period but concluded that they were of little importance in the first two centuries of Portuguese occupation. Suely Robles Reis de Queiroz, Escravidão Negra em São Paulo (Rio de Janeiro: Livraria José Olympio Editora, 1977), p. 12.

⁵³ Several aldeamentos became vilas or districts (bairros) of the city of São Paulo. On this theme, see Pasquale Petrone, Aldeamentos Paulistas (São Paulo: EDUSP, 1995), pp. 84–351.

⁵⁴ There exist disagreements in the literature with respect to the magnitude of these transactions, but not to their occurrence. Ellis Jr., “O Ouro e a Paulistania,” p. 53; Roberto C. Simonsen, História Econômica do Brasil (São Paulo: Companhia Nacional 1977), p. 218; and Buarque de Holanda, “Movimentos de população,” among others, emphasize the importance of the Indian enslavement. On the other side are scholars like Monteiro (Negros da Terra, pp. 76–81) who admit the existence of this traffic in Indians to other capitanias, but refute its importance.

⁵⁵ For a more detailed analysis of the paulista economy in the sixteenth and seventeenth centuries, see Francisco Vidal Luna and Herbert S. Klein, Slavery and the Economy of São Paulo, 1750–1850 (Stanford: Stanford University Press, 2003), chapter 1.

⁵⁶ Indian slavery was systematically attacked by the Crown in the second half of the eighteenth century, but the continued persistence of the aldeamentos allowed forced labor to continue in a disguised form. In 1802, the aldeamentos were supposedly abolished, but the use of Indian forced labor continued, however precariously, until the middle of the nineteenth century. In 1846, there was a final attempt to resurrect these villages, but they had disappeared by then. Among the abundant literature on this subject, see Agostinho Marques Perdigão Malheiro, A escravidão no Brasil: Ensaio Histórico-Juridico-Social (2 vols.; Rio de Janeiro: Typografhia Nacional, 1866); Georg Thomas, Política indigenista dos portugueses no Brasil, 1500–1640 (São Paulo: Edições Loyola, 1982); Rodrigo Otávio, Os selvagens americanos perante o direito (São Paulo: Companhia Editora Nacional , 1946); Warren Dean, “Indigenous populations of the São Paulo-Rio de Janeiro coast: Trade aldeamento, slavery and extinction,” Revista de História, 117 (1984), pp. 3–26; Alexander Marchant, Do escambo à Escravidão: As relações econômicas de portugueses e índios na colonização do Brasil, 1500–1580 (São Paulo: Companhia Editora Nacional, 1980); John Manuel Monteiro, Negros da Terra: Índios e Bandeirantes nas Origens de São Paulo (São Paulo: Companhia das Letras, 1994) and his essay “From Indian to Slave: Forced Native Labour and Colonial Society in São Paulo during the 17th Century,” Slavery & Abolition, 9, no. 2 (September 1988), pp. 105–27; Petrone, Aldeamentos Paulistas; and Manuela Carneiro da Cunha, ed., História dos Índios no Brasil (São Paulo: Companhia das Letras, 1992).

⁵⁷ By the early nineteenth century, there are good data on several charqueadas. The one owned by João Nunes Batista had sixty-six slaves on his estate, of which thirty-one worked as carneadores, charqueadores, and sailors in the production and transport of jerked beef. Ester J. B. Gutierrez, Negros, Charqueadas, e Olarias: Um estudo sobre o espaço pelotense (2nd ed.; Pelotas: Editora e Gráfica Universitária – UFPel, 2001), pp. 61–62. A survey of numerous charqueadas at mid-century found an average of eighty-four slaves per estate, with most of these slaves being skilled workers directly employed in the processing and transportation of meat products. Gutierrez, Negros, Charqueadas, e Olarias, p. 91. By the mid-nineteenth century, a typical charqueador employed 80 slaves and 10 free workers and slaughtered some 200 to 250 cows per day. Mário José Maestri Filho, O escravo no Rio Grande do Sul: A Chaqueada e a gênese do escravismo gaúcho (Porto Alegre: Escola Superior de Teologia São Lourenço de Brindes, 1984), p. 89.

⁵⁸ The largest ranching estates in this region, centered in the Castro district, on average owned seventeen slaves. Horácio Gutiérrez, “Escravidão e pequena propriedade no Paraná” (unpublished manuscript).

⁵⁹ Horácio Gutiérrez, “Donos de terras e escravos no Paraná: Padrões e hierarquias nas primeiras décadas do século XIX,” História, 25, no. 1 (2006), pp. 102–3.

⁶⁰ Herbert S. Klein, “Os homens livres de côr na sociedade escravista brasileria,” Dados (Rio de Janeiro), no. 17 (1978), tabela 1.

⁶¹ Myriam Ellis, A Baleia no Brasil Colonial (São Paulo: Edições Melhoramentos, EDUSP, 1969), p. 100. On the extensive use made of slaves in the oil-producing factories on shore, see Fernando Henrique Cardoso and Octavio Ianni, Côr e mobilidade social em Florianópolis (São Paulo: Companhia Editora Nacional, 1960), pp. 23–6. The Santa Catarina whaling factories were the most important ones in the mid- to late eighteenth century, but there were also major factories in the ports of Salvador de Bahia and Rio de Janeiro going back to the seventeenth century – all of them being shore-based whaling companies; see Dauril Alden, “Yankee Sperm Whalers in Brazilian Waters, and the Decline of the Portuguese Whale Fishery (1773–1801),” The Americas, 20, no. 3 (January 1964), pp. 270ff.

⁶² Ellis, A Baleia no Brasil Colonial, p. 190. For a full discussion of the different employments of slaves and free colored in the industry, see Ellis, A Baleia no Brasil colonial, chapter 3.

⁶³ This is a rough figure suggested by early-nineteenth-century shipping data. In the 1830s, for example, there was an average of 1,807 ships (averaging per annum 123,733 tons) listed as being cabotagem or coastal vessels arriving just into the port of Rio de Janeiro. Jornal do Comércio (Rio de Janeiro), 4 Janeiro 1840, for all the 1830s.

⁶⁴ Herbert S. Klein, The Middle Passage: Comparative Studies in the Atlantic Slave Trade (Princeton: Princeton University Press, 1978), pp. 58–9.

⁶⁵ “…esta falta faz que os navios e corvetas que trilham o comércio d’Africa, costumam equiparse com uma pequena equipagem de quatro ou seis marinheiros brancos, suprindo os pretos captivos para o resto da mareação de que carecem.” IHGB/CU, Arq. 1–1–19, “Correspondência do Governador da Bahia, 1751–1782” folios 228v-230, 3 Julho 1775, Bahia. The governor reported that slaves made up 64 percent of the 1,905 sailors registered in the province of Bahia and were 39 percent of its 2,069 fishermen.

⁶⁶ António Carreira, As companhias pombalinas de Grão-Pará e Maranhão e Pernambuco e Paraíba (2nd ed.; Lisbon: Editorial Presença, 1982), p. 249.

⁶⁷ Serious exports of cotton began from Maranhão in the 1760s, by the 1770s Pará became an important but minor producer, and finally in the 1790s, Pernambuco replaced Maranhão as the leading producer. Alden, “Late Colonial Brazil, 1750–1808,” II, table 8, pp. 636–7. By 1796–1800, cotton exports reached 4,443 tons and were worth almost the same as sugar exports in that quinquenium, and now represented some 30 percent of British cotton imports. Jorge M. Pedreira, “From Growth to Collapse: Portugal, Brazil, and the Breakdown of the Old Colonial System (1760–1830),” Hispanic American Historical Review, 80, no. 4 (2000), p. 843. Cotton even temporarily surpassed the value of sugar in the first decade of the nineteenth century and was then Brazil's leading export. José Jobson de Andrade Arruda, Brasil no comércio colonial (São Paulo: Editora Ática, 1980), pp. 353–4.

⁶⁸ Silvia Hunold Lara, Campos da Violência. Escravos e senhores na Capitania do Rio de Janeiro, 1750–1808 (Rio de Janeiro: Paz e Terra, 1988), pp. 138–9.

⁶⁹ Lara, Campos da Violência, p. 136.

⁷⁰ A good overview of the sugar trade in this period is found in Galloway, The Sugar Cane Industry, chapters 4 and 5; Schwartz, Sugar Plantations in the Formation of Brazilian Society, chapter 15; and Noel Deerr, The History of Sugar (2 vols.; London: Chapman and Hall, 1949–1950).

⁷¹ The best single source on the nineteenth-century slave population by province is Joaquim Norberto de Souza e Silva, Investigações sobre os recenseamentos da população geral do Império e de cada província de per si tentados desde os tempos coloniais até hoje ([1870]; reprint, São Paulo: Instituto de Pesquisas Econômicas, 1986).

⁷² For an analysis of the regional economies within Minas Gerais, see Clotilde Andrade Paiva, “População e economia nas Minas Gerais do século XIX” (Tese de doutorado, São Paulo: FFLCH/USP, 1996); Marcelo Magalhães Godoy, “Espaços canavieiros regionais e mercado interno subsídios para o estudo da distribuição espacial da produção e comércio de derivados da cana-de-açúcar da província de Minas Gerais,” X Seminário sobre a Economia Mineira (2002) and Marcelo Magalhães Godoy, Mario Marcos Sampaio Rodarte, and Clotilde Andrade Paiva, “Negociantes e tropeiros em um território de contrastes, o setor comercial de Minas Gerais no século XIX,” Anais V Congresso Brasileiro de História Econômica, ABPHE (2003).

⁷³ Roberto Borges Martins, “Growing in Silence: The Slave Economy of Nineteenth-century Minas Gerais, Brazil,” (Ph.D. diss., economics, Vanderbilt University, 1980); and Amílcar V. Martins Filho and Roberto B. Martins, “Slavery in a Non-Export Economy: Nineteenth-Century Minas Gerais Revisited.” Hispanic American Historical Review, 63, no. 3 (1983), pp. 537–68.

⁷⁴ Francisco Vidal Luna and Wilson Cano, “Economia escravista em Minas Gerais,” Cadernos IFCH/UNICAMP (Outubro 1983).

⁷⁵ The trade would fluctuate between fifteen and nineteen thousand African arrivals per annum until the 1780s, when it would begin a long, secular rise, reaching thirty thousand per annum by the 1790s. These figures are based on the latest estimates of volume of the trade in the Emory data set accessed on August 18, 2008.

⁷⁶ For comparative data on slave and free colored in the Americas, see Herbert S. Klein and Ben Vinson III, African Slavery in Latin America and the Caribbean (2nd rev. ed.; New York: Oxford University Press, 2007), appendix tables 1–3.