Population and Slavery in Vila Rica de Ouro Preto (1712-1770)

Analysis of the slave import market

Up until the mid-17th century, the Spanish colonies in America were the main destination of the Atlantic slave trade.⁹On this issue, see the interesting monograph edited by Alex Borucki, David Eltis, and David Wheat (2020). However, the Portuguese political crisis, which broke out in 1640, accelerated the reconfiguration of the organization and logistics of the slave business and put an end to the Portuguese monopoly in Africa. Likewise, the expansion of sugar plantations in north-eastern Brazil led to a gradual change in the destination of the trade, as a result of which the Portuguese territory in America surpassed the Spanish colonies in the volume of imports.

This way, Belém, São Luis, Recife, Salvador, and, to a lesser extent, Rio de Janeiro and Santos, became important ports for the landing of African slaves (Alencastro, 2000, p. 250Alencastro, L. F. de (2000) O trato dos viventes: formação do Brasil no Atlântico Sul. São Paulo: Companhia das Letras.). Well into the 18^th century, however, Minas Gerais interrupted its slave activity and caused a new dislocation in the transatlantic and regional slave flows (Florentino, 2014, pp. 233-235Florentino, M. (2014) “Aspectos do tráfico negreiro na África Ocidental (c. 1500-1800).” In: J. Fragoso, and M.ª de F. Gouvêa, orgs., O Brasil colonial (1443-1580). Rio de Janeiro: Civilização Brasileira, 1, pp. 229-270.). We have already hinted at the redistribution routes departing from the major slave ports, especially Salvador¹⁰On Salvador de Bahia as a catalyst city and its role in the transatlantic slave trade, see Verger (1987), Ribeiro (2008), and Souza (2018). For a study on the commercial circuits connecting Minas Gerais with Bahia, see Santos (2013). and Rio de Janeiro¹¹On the case of Rio de Janeiro as a slave port, see Florentino (1997), Rodrigues (2005), and the monograph coordinated by Florentino, especially Cavalcanti (2005)., towards Minas Gerais. Thus, for instance, Antonio Alves Rozerio received a power of attorney from Gregorio Gonçalves Ramos, a resident of Rio de Janeiro, to sell in Minas Gerais “hum moleque por nome Miguel, ladino de nação Angola” (a young boy called Miguel, ladino of the Angola nation). He was sold in May 1722 to Antonio Coelho Barboza, a resident of Vila Rica who, in addition to the slave’s price, had to pay for “os gastos que fez o dito moleque em este rio da Parahiba [do Sul], do Caminho Novo” (the expenses that the above-mentioned young man made in the river of Parahiba [do Sul], along the Caminho Novo).¹²AHMI-CP: C. 66, A. 737, 2º Of., 1722. This route, which was shorter than the one connecting Minas Gerais with Bahia, would end up becoming the main artery of communication and exchange. The urgency for labor and the wealth generated by the mining activity caused about 2,300 slaves to travel from Rio de Janeiro to Minas Gerais every year between 1715 and 1727, i.e., almost 70% of the slaves disembarked at the Rio de Janeiro port (Florentino, 1997, p. 37Florentino, M. (1997) Em costas negras: uma história do tráfico de escravos entre a África e o Rio de Janeiro: séculos XVIII e XIX. São Paulo: Companhia das Letras.).

As the literature has already pointed out, little is known yet about the slave import market of Minas Gerais during the 18^th century (Libby, 2007, p. 414Libby, D. C. (2007) “As populações escravas das Minas Gerais do século XVIII.” In: M.ª L. de Resende, and L. C. Villata, orgs., História de Minas Gerais. As Minas setecentistas. Belo Horizonte: Autêntica & Companhia do Tempo, pp. 407-438.), which is indeed a complex issue. In Bergard’s opinion (2004, p. 222)Bergard, L. W. (2004) Escravidão e história econômica: demografía de Minas Gerais, 1720-1880. Bauru (São Paulo): EDUSC., this lack of knowledge is explained by the absence of reliable data on slave imports. The issue has been approached mainly through the analysis of notarial records, namely last will and testaments and post-mortem inventories. However, it is important to take into account that this type of document refer to slave property that may have been formed long before but registered only at the end of their owner’s life. Therefore, these sources reflect a complex temporality of practices and phenomena that may involve various situations at the same time. On the one hand, the purchase of new slaves trafficked across the Atlantic or within the region, as a result of business operations carried out relatively close to the date of the documentary record or, on the contrary, years before. On the other hand, the logic of natural reproduction (Libby, 2007, p. 421Libby, D. C. (2007) “As populações escravas das Minas Gerais do século XVIII.” In: M.ª L. de Resende, and L. C. Villata, orgs., História de Minas Gerais. As Minas setecentistas. Belo Horizonte: Autêntica & Companhia do Tempo, pp. 407-438.). Likewise, it is also necessary to consider the possible repetition of slaves that become part of other slaves properties and that are again inventoried, so that the quantification from this type of documentary could be overestimated. On the contrary, there are two documentary series that make it possible to quantify the slave import markets and the calibration of their evolution over time. On the one hand, the fiscal documentation belonging to the Casa dos Contos, which has enabled Carrara to reconstruct the imports made in the Minas Gerais captaincy at two moments, between the years 1716 and 1717, and since 1757. However, “se as fontes são fragmentadas e escassas até 1717, a partir desse ano até 1757 guardam um silêncio tumular,” so that this series would serve to cover, above all, the second half of the 18th century (Carrara, 2007, p. 114Carrara, A. A. (2007) Minas e Currais: produção rural e mercado interno em Minas Gerais, 1674-1807. Juiz de Fora: Universidade Federal de Juiz de Fora. ). On the other hand, it is also possible to count on the records of baptisms of adult slaves, the series used here to quantify and calibrate over time the evolution of the Vila Rica slave import market. In effect, the short time elapsed between the arrival of the slaves at the Brazilian ports and the time of their baptism allows us to validate the relationship between the administration of this sacrament and the activity of the slave import market. There were doubts about whether the slaves had been baptized in their regions of origin, and the royal and ecclesiastical authorities recommended the administration of the sacrament to them, if necessary sub conditione, to avoid the risk of their dying unbaptized and therefore deprived of salvation.¹³As pointed out by Iraci del Nero Costa about the parish of Nossa Senhora da Conceição de Antônio Dias (Costa, 1979), or more recently by Patrícia Porto de Oliveira about the other parish in Vila Rica, Nossa Senhora do Pilar (Oliveira, 2022, p. 10). For a recent analysis of the administration of baptism among adult slaves in Seville during the 16^th and 17^th centuries, its connection to the problem of evangelization in the Iberian peninsula and Africa, and its relationship with the slave import market of the city, see Corona Pérez (2021).

Figure 2 shows the evolution of the slave import market in the parish of Nossa Senhora do Pilar in Vila Rica, which is used as a reference. As seen in the figure above, there is a documentary gap between 1720 and 1735. However, the results allow for estimating the size and access of the slave market of Vila Rica to the circuits and dynamics of the transatlantic slave trade. The increase in imports is also put concerning the production of gold at the captaincy of Minas Gerais. This comparative analysis is interesting because, despite the lack of data for the above-mentioned years, the curve of the slave import market basically reproduces that of the production of gold in the region, with a sharp increase at the end of the 1730s and a gradual decline after that date until the end of the period studied. The discovery of gold originated a strong migratory movement within the region, motivated by the prospect of enrichment. This movement of population and its economic power, together with the constant flow of supply from the slave trade circuits enabled the development of a dynamic market of African slaves until, at least, the middle of the 18^th century.

This is confirmed by Figure 3, where the evolution of the landing of African slaves at the Bahia and Rio de Janeiro ports is compared with the slave imports in Vila Rica. It is interesting to observe the relationship between these three locations because they share a dynamic of growth until, approximately, the beginning of the 1740s. Despite the recovery experienced in the period 1746-1750, São Salvador was gradually surpassed in a number of landings by the Rio de Janeiro port from the 1750s onwards. On the other hand, even if it is true that the slave import market of Vila Rica expanded between 1751 and 1755, it seems that, to the end of the period under study, it distanced itself from the transatlantic slave trade dynamics, especially during the 1760s, when it evolved independently of the disembarkation of African slaves in Bahia and Rio de Janeiro. Therefore, there seems to be a complete correlation between the production of gold and the import of slaves in Vila Rica: the discovery and exploitation of gold in the region catapulted the demand and access to a well-stocked supply of slaves, while the decline of the gold mining activity, which began at the end of the 1730s and accelerated after 1760, brought the contraction of both phenomena.¹⁴A relation also observed by Carrara for the Minas Gerais of the second half of the 18th century (2007, p. 145). The fall in the production of gold in the “comarca” of Vila Rica, which had the highest number of mining operations in the captaincy (45.2%), caused the decay of the local slave market, which suffered more than those of other “comarcas.” In fact, the gold mining crisis did not, by any means, reduce the slave imports from other areas in the Minas Gerais region. The “comarca” of Rio das Mortes, which had developed an export-oriented agricultural and livestock economy closely linked to the slave supply from Rio de Janeiro, soon replaced Vila Rica as the main population centre within the captaincy, in terms of both total population and slave population (Almeida, 2010, pp. 50-63Almeida, C. M.ª C. de (2010) Ricos e pobres em Minas Gerais: produção e hierarquização social no mundo colonial, 1750-1822. Belo Horizonte: Argvmentvm.).

Between the 16^th and 18^th centuries, Iberian societies were thought of and represented as an orderly whole composed of autonomous and unequal though hierarchized parts. The “quality” and, above all, the “legal status” of their members-in other words, whether they had been born free, or were “forros” or slaves-was absolutely essential. The political inequality between those statuses defined their specific position and role within that world (Hespanha, 2010, p. 55Hespanha, A. M. (2010) Imbecillitas. As bem-aventuranças da inferioridade nas sociedades de Antigo Regimen. São Paulo: Annablume.). We should therefore not be surprised at the variety of descriptive categories revealed by the sample of 213 slaves here presented, obtained from the analysis of 123 post-mortem inventories from the years 1720-1750 found at the Arquivo Histórico do Museu da Inconfidência - Casa do Pilar in Ouro Preto.¹⁵Many of the documentary funds of the first half of the 18^th century preserved at the Casa do Pilar were lost in a fire in the mid-20^th century. This sample is representative of approximately one-half of the notarial records preserved. The documents reflect that the slave workforces of Vila Rica were mostly made up of African slaves (67%) with similar profiles to the ones identified in the neighbouring “comarcas” of Rio das Velhas and Rio das Mortes during the 18^th century (Paiva, 2022, p. 126Paiva, E. F. (2022) “Por meu trabalho, serviço e indústria.” Histórias de africanos, crioulos e mestiçados nas Minas Gerais, 1716-1789. Belo Horizonte: Autêntica.). The classification is based on the one proposed by Mary Karash (2000, pp. 5066)Karash, M. C. (2000) A vida dos escravos no Rio de Janeiro (1808- 1850). São Paulo: Companhia das Letras., with a first division by region of origin, and a second one by qualities. As seen in Table 1, there is a prevalence of individuals arriving from Western Africa, a large region comprising Upper Guinea, the Gulf of Guinea, and the archipelagos of Cape Verde and São Tomé.¹⁶For an approach to the African slave trade in Brazil during the 18^th and 19^th centuries, see Florentino, Ribeiro & Silva (2004) or Acioli & Menz (2008). Among those coming from that region, the majority were “Mina,” a broad category that ultimately designated not only the slaves originally from the so-called “Coast of Mina”- named after the castle of São Jorge da Mina, built at the end of the 15^th century-but those from the whole western coast of the Gulf of Guinea, including several ethnolinguistic groups. Such were the cases of Domingos, “negro de nação mina cobû;” Francisco, “negro de nação mina courano;” or Antonio, “negro de nação mina sabarû.” The term “negro” refers to the individual’s legal status, given its semantic equivalence with “slave” (Paiva, 2015, pp. 200-202Paiva, E. F. (2015) Dar nome ao novo. Uma história lexical da Ibero-América entre os séculos XVI e XVIII (as dinâmicas de mestiçagens e o mundo do trabalho). Belo Horizonte: Autêntica.); “nação mina” refers to the individual’s geographical origin, while “Cobû,” “Courano” and “Sabarû” denote their ethnolinguistic group.¹⁷All the cases in AHMI-CP: C. 33, A. 374, 2º Of., year 1731. In addition, broad categories like “Mina” were also used to organize and distribute the slaves to different destinations and to meet certain demands, whether for technical skills or labor capacity, as in mining (Paiva, 2015, p. 146Paiva, E. F. (2015) Dar nome ao novo. Uma história lexical da Ibero-América entre os séculos XVI e XVIII (as dinâmicas de mestiçagens e o mundo do trabalho). Belo Horizonte: Autêntica.). Another African region often represented among the slave population of Vila Rica was the central-western area.¹⁸On the organization of the slave trade from that region, see Curto (2002) or Cândido (2013). This large area extends from Cape Lopez in the north to the Cunene River basin. Among the slave groups coming from this region, we can mention the “Benguela,” originally from southern Angola; the “Angola,” who were native of the area around Luanda; and the “Congo,” a term used to describe a multitude of ethnical groups (Karash, 2000, pp. 50-58Karash, M. C. (2000) A vida dos escravos no Rio de Janeiro (1808- 1850). São Paulo: Companhia das Letras.). The last of the African regions recorded, although much less often than the two previous ones, is Eastern Africa, where the “Mozambique” came from.

The rest of the descriptive groups comprise those born in America. Thus, the second group in number was that of the “crioulos” (14.1%), a category used to designate the children of African mothers born in Brazil and, often, their descendants (Paiva, 2015, pp. 202-204Paiva, E. F. (2015) Dar nome ao novo. Uma história lexical da Ibero-América entre os séculos XVI e XVIII (as dinâmicas de mestiçagens e o mundo do trabalho). Belo Horizonte: Autêntica.). The “crioulos” were followed by other categories qualifying American indigenous groups, most particularly, the “Carijó.” According to Paiva, possibly during the 17^th century the term “carijó” became a synonym for “Indian” or “Indian slave,” thus becoming a general term of designation (Paiva, 2015, p. 196Paiva, E. F. (2015) Dar nome ao novo. Uma história lexical da Ibero-América entre os séculos XVI e XVIII (as dinâmicas de mestiçagens e o mundo do trabalho). Belo Horizonte: Autêntica.). The eleven references to “carijó” correspond to the slave property owned by Francisco Leme da Silva, a resident of the “freguesia” (parish) of São Bartolomeu, located 20 kilometres to the north of Vila Rica. Except for Ana and Gregorio, who were single, the rest of the “carijó indians” of this slave property belonged to five different families. Three of those families were formed, respectively, around the “Carijó” couples of Matias and Maria, Gregorio “e sua mulher” (and his wife), and Tomé and Rufina, aged between 50 and 60. The three remaining “Carijó” belonged to mixed families. Mariana was married to João, who was a “Mina.” Maria, who already had a child from a previous partner (Simão), married João, who was a “Mina,” and from that relationship Domingos and Maria were born. Biological and cultural dynamics of miscegenation were indeed very intense in Iberian societies between the 16^th and 18^th centuries. Thus, the “mestizados” were also represented among the slave population of Vila Rica. In this sense, the product of the relationships between Indians and African Blacks, such as Maria and João, was often called “cabra,” a term used to classify these people in 18^th-century Brazil (Paiva, 2015, p. 205Paiva, E. F. (2015) Dar nome ao novo. Uma história lexical da Ibero-América entre os séculos XVI e XVIII (as dinâmicas de mestiçagens e o mundo do trabalho). Belo Horizonte: Autêntica.). The other designation found in the records is “mulato,” which, like the term “pardo” described “mestizado” people.¹⁹Although it was also used to refer to a skin tone lighter than black and darker than white. In the case of the Iberian peninsula, see Fernández Chaves (2016).

As for the sex of the imported slaves, in a sample of 2,315 people gathered from the adult baptismal records of the Pilar and Conceição parishes, 1,693 were men (73.1%) and 622 were women (26.9%), with one case of sex unknown due to the poor condition of the document.²⁰BD-AENSP: B (1712-1770); AENSC: LB (1710-1740). This way, the sex distribution pattern of the slave import market of Vila Rica was similar to that of other population centres in Minas Gerais (Libby, 2007, pp. 420-423Libby, D. C. (2007) “As populações escravas das Minas Gerais do século XVIII.” In: M.ª L. de Resende, and L. C. Villata, orgs., História de Minas Gerais. As Minas setecentistas. Belo Horizonte: Autêntica & Companhia do Tempo, pp. 407-438.; Paiva, 2022, pp. 161-163Paiva, E. F. (2022) “Por meu trabalho, serviço e indústria.” Histórias de africanos, crioulos e mestiçados nas Minas Gerais, 1716-1789. Belo Horizonte: Autêntica.), as well as to that of Rio de Janeiro (Florentino, 1997, pp. 55-56Florentino, M. (1997) Em costas negras: uma história do tráfico de escravos entre a África e o Rio de Janeiro: séculos XVIII e XIX. São Paulo: Companhia das Letras.), and fully reflects that of transatlantic trade, with an almost 3 to 1 predominance of men over women. However, if each descriptive group is analysed separately, the results may differ, as seen in Table 2.

The African slave population also reflects the pattern of the import market, and presents a structurally unbalanced demography (Florentino, 1997, p. 59Florentino, M. (1997) Em costas negras: uma história do tráfico de escravos entre a África e o Rio de Janeiro: séculos XVIII e XIX. São Paulo: Companhia das Letras.), with over 76% of men versus 22% of women. In the remaining groups, it is obvious that the distribution does not correspond to the commercial dynamics of the slave trade but to natural reproduction, and shows some balance between men and women among “crioulos,” indigenous, and “mestizados” slaves.

Age is anecdotal information in baptismal records, and the analysis of this source does not allow us to infer the age at which imported slaves arrived in Minas Gerais. In contrast, it is frequently mentioned in post-mortem inventories, where age is “pouco mais ou menos” (approximately) taken into account to calculate the price of the slaves. As a result, we know the age of around 80% of the slaves inventoried in our sample (171 of 213 cases). The data show that the average age was high, around 29.9 years globally. If we look at the distribution by sex, the average age of men slaves was higher (30.7 years) than that of women slaves (26.6 years) but there are no significant differences except in the “Mestizado” group, possibly because of the small size of the sample. By descriptive groups, the average age of African slaves was 33.3 years, while it was 9.9 years for the “Crioulo,” 48.3 years for indigenous slaves, and 13.4 years for the “Mestizado” group.²¹Of. 1: C. 1, A. 5; C. 8, A. 81; C. 19, A. 180; C. 21, A. 214; C. 22, 233; C. 23, A. 239; C. 23, A. 253; A. 24, 255; C. 26, A. 291; C. 34, A. 401; C. 38, A. 457; C. 38, A. 459; C. 41, A. 484; C. 42, A. 486; A. 42, A. 492; C. 43, A. 505; C. 43; A. 507; C. 43, A. 572; C. 47, A. 572; C. 47, A. 576; C. 48, A. 587; C. 50, A. 609; C. 51, A. 619; C. 57, A. 608; C. 58, A. 695; C. 65, A. 784; C. 66, A. 800; C. 69, A. 828; C. 71, A. 846; C. 72, A. 859; C. 73, A. 865, C. 73, A. 866; C. 74, A. 883; C. 77, A. 913; C. 77, A. 915; C. 80, A. 983; C. 91, A. 1107; C. 92, A. 1115; C. 92, A. 1121; C. 93, 1198; C. 98, A. 1194; C. 98, A. 1198; C. 98, A. 1199; C. 104, A. 1306; C. 104, A. 1308; C. 104, A. 1310; C. 104, A. 1313; C. 104, A. 1756; C. 104; A. 1313; C. 105, A. 1323; C. 105, A. 1324; C. 111, A. 1414; A. 113, A. 1440; C. 116, A. 1481; C. 119, A. 1507; C. 122, A. 1530; C. 123, A. 1545; C. 134, A. 1687; C. 140, A. 1755; C. 141, A. 1785; C. 142, A. 1789; C. 142, A. 1792; C. 218, A. 3493; C. 298, A. 6438; C. 333, A. 7017; C. 336, A. 7054; C. 435, A. 8986; C. 438, A. 8986; C. 449, A. 9458; C. 460, A. 9769. Of. 2: C. 2, A. 26; C. 11, A. 114; C. 12, A. 123; C. 12, A. 126; C. 13, A. 127; C. 16, A. 165; C. 16, A. 174; C. 17, A., 181; C. 17, A. 175; C. 21, A. 222; C. 21, A. 228; C. 22, A. 236; C. 29, A. 324; C. 29, A. 328; C. 30, A. 342; C. 33, A. 373; C. 33, A. 374; C. 38, C. 421; C. 39, A. 440; C. 40, A. 452; C. 40, A. 455; C. 41, A. 463; C. 41, A. 463; C. 42, A. 469; C. 43, A. 485; C. 46, A. 504; C. 46, A. 506; C. 47, A. 519; C. 48, A. 432; C. 50, A. 550; C. 50, A. 560; C. 51, A. 570; C. 52, A. 584; C. 52, A. 585; C. 54, A. 613; C. 55, A. 621; C. 56, C. 636; C. 58, A. 650; C. 59, A. 665; C. 60, C. 682; C. 62, A. 700; C. 63, A. 705; C. 66, A. 737; C. 68, A. 758; C. 101, A. 1257; C. 135, A. 1975; C. 135, A. 1976; C. 173, A. 3098; C. 180, A. 3319. It is clear that “crioulo” and “mestizado” slaves were the product of the natural reproduction of women slaves. However, it may be asked whether the average age of the African slaves inventoried is not distorted by the very nature of the source. As underlined when we described the documents used to measure the import market, notarial inventories were made after the death of the slaveowners, presumably at an old age. Therefore, the recorded age of African slaves was not representative of their “commercial age.” In effect, nine out of ten slaves trafficked by the slavers to the port of Rio de Janeiro were between 10 and 34 years old (Florentino, 1997, p. 59Florentino, M. (1997) Em costas negras: uma história do tráfico de escravos entre a África e o Rio de Janeiro: séculos XVIII e XIX. São Paulo: Companhia das Letras.), while there are studies on the slave market of the Iberian peninsula and the Spanish territories in America based on the analysis of letters of purchase and sale-a type of document that was not preserved among the notarial records of Minas Gerais-which estimate a lower average age of between 20 and 24 years.²²As it happened in Ayamonte (González Díaz, 1996, pp. 56-58), Huelvas, Palos de la Frontera, Moguer (Izquierdo Labrado, 2004, vol. 1, pp. 134-137) or Seville (Corona Pérez, 2022, pp. 168-180). For a recent example in Spanish America, see the doctoral thesis of Luis Molinari Mundim on the slave market of the city of La Plata, present-day Sucre, in Bolivia (Mundim, 2022, p. 526-528). In this sense, and even if age is specified only in 21 of the more than 2,000 cases, the adult slave baptismal records of the Pilar parish indicate an average age of 17.3 years.²³BD-AENSP: B (1712-1770).

The analysis of the average price of slaves is a complex issue because inventories provide a limited amount of data. Table 4 shows the average price of slaves, first by descriptive group and sex, and, secondly, by age range and sex. To make a more precise estimation, all slaves in the sample who had a flaw, whether physical (disease, mutilation) or behavioural (escapes), were disregarded, as were those who had been valued jointly. The information in the table allows us to draw some conclusions. The first is that the highest-rated profile in Vila Rica during the first half of the 18^th century was that of African men. This is confirmed when analysing the age ranges, in all of which the average price of men slaves is higher than that of women slaves. These results suggest that, in a mining-based economy, the capacity to work in the mines is more highly valued than the reproductive capacity of women slaves (Bergard, 2004, p. 264Bergard, L. W. (2004) Escravidão e história econômica: demografía de Minas Gerais, 1720-1880. Bauru (São Paulo): EDUSC.). Indeed, mining was the activity to which most slaves in the “comarca” of Vila Rica were assigned (Almeida, 2010, p. 56Almeida, C. M.ª C. de (2010) Ricos e pobres em Minas Gerais: produção e hierarquização social no mundo colonial, 1750-1822. Belo Horizonte: Argvmentvm.). As we will see later on, the average fertility rate of women slaves was low. The issue of the higher-rated capacity by sex lends itself to debate. Women, despite being a minority among the slave population of Minas Gerais, were freed more often than men during the 18^th century.²⁴This is not the place to critically examine the state of the art concerning this issue in the captaincy of Minas Gerais. We will only refer to the estimations made by Lima (2021, pp. 39-47), Silva (2017, pp. 32-41), and Monti (2020, pp. 15-28), which are probably the most up-to-date and complete ones on the Minas Gerais region during the 18^th century. In the case of Vila Rica, see the monumental work by Marcos Magalhães de Aguiar, the first chapter of which focuses on the analysis of the “alforrias” (1999, pp. 6-74), as well as the more recent monograph by Precioso (2011). In addition, many of them were freed after paying a certain amount, accumulated from their earnings (Paiva, 2022, pp. 2016-225Paiva, E. F. (2022) “Por meu trabalho, serviço e indústria.” Histórias de africanos, crioulos e mestiçados nas Minas Gerais, 1716-1789. Belo Horizonte: Autêntica.). Is it possible that the greater opportunities that women slaves had to buy out their freedom reflected their lower value in the market? In addition, the data for women slaves shown in the table may be distorted because access to freedom may have prevented the full consideration of these women’s capacity for work, which was possibly more highly rated than the figures suggest.

As for age as an element of valuation, the higher prices are obviously found in the intermediate age ranges, when slaves were most productive and could also be revalued and resold in the market. The lower prices, even when compared with older ages, were found in the group aged 0 to 9 years. As we will see later on, the low chances of survival during infancy, i.e., the high infant mortality rates, together with the children’s null or little production capacity until they are old enough to perform certain tasks, explain the very low value attributed to slaves in their youngest years.

Sociodemographic analysis

What was the size of the slave contingent of Vila Rica about the whole population? What was its evolution in time like? These questions may be answered by analysing baptismal records and estimating the percentage of non-adult slaves baptized over the total number of individuals who were administered the sacrament (Cortés, 1987, p. 94Cortés, E. (1987) Esclavos en la Extremadura meridional. Siglo XVII. Badajoz: Diputación Provincial de Badajoz.).

As shown, the curve of the slave population dominated the local birth rate until the 1750s, when it was surpassed by that of the free population. It seems that the progression of the slave population was very much linked to the production of gold, with an extraordinary increase in the number of births in absolute terms from 1721 to 1740. From that year onwards, with the decline of the gold mining activity, the curve flattened, showing a negative trend until the end of the period studied, as was the case with the evolution of the slave market (see Figures 2 and 3). On the other hand, the number of births of the free population increased throughout the period, with a less aggressive but more sustained progression over time. It is important to take into account that the presence of European women in the region was minimal during the 18^th century (Figueiredo, 1997, pp. 20-28Figueiredo, L. R. de A. (1997) Barrocas famílias. Vida familiar em Minas Gerais no século XVIII. São Paulo: Editora HUCITEC.). Therefore, the natural growth of the free population and the contribution made by the descendants of freed women slaves explain the evolution of the group. Finally, the curve of the freed slaves reflects the small size of this group, which was the outcome of the process of manumission at the baptismal font. Up to 8% of all the newborns of slave mothers benefitted from that process during the period studied, which ate up another “chunk” of the slave contingent. As a result of this situation, there was a 30% drop in the representation of the number of births of the slave population concerning the overall number of births in Vila Rica, falling from 76% in 1712-1715 to 40% in 1766-1770. Why was it so?

The demographic evolution of the slave population was the result of interrelated, non-exclusive factors of growth and degrowth. Thus, the contingent grew naturally when its population growth rate was positive, that is, when its birth rate reached or exceeded the replacement rate, as well as the mortality rate. This also depended on whether the nutritional, socio-affective, physical, material, and environmental conditions of this population group were good enough to be able to cope with daily life and disruptive situations, such as epidemics or famines, with confidence. Of course, slave populations could also be increased from the outside. This happened especially when the slave market had easy access to the import flows-but not only. There were other possibilities to replace slave workforces, although the flows were quantitatively smaller than those of the slave trade. It happened in the case of slaves who accompanied their owners when they temporarily or definitely migrated to other places. Or when the so-called “escravos de ganho” (slaves of gain) were temporarily displaced to another region to work, thus increasing the captive population at their point of destination. Exactly the opposite can be said of the decreases, which are explained by birth rates that are low or insufficient to surpass the mortality rates and reach the natural replacement rate due to interruption or withdrawal of access to the slave import flows in local or regional markets, or to the processes of manumission. In Vila Rica, during the period ranging from 1712 to 1770, the slave contingent showed, in contrast with the free population, a decreasing trend, a tendency to lose demographic forces. And this was due to sociodemographic factors and also to the specific social, economic, and cultural dynamics of that society related to the process to access freedom.

As proved in a previous work, the natural population growth of the slave population of Vila Rica was negative across this period, with a drop of more than 160%. These results contrast with those for the free population, which experienced an increase of more than 120%. They can be attributed to the slave population’s low birth rate and high infant mortality rate, which hindered the natural replacement of the group. On the one hand, the average number of children per slave mother was 1.3, compared with 2.1 children among free mothers. On the other hand, the infant mortality rate among slaves was over 350‰, which practically doubled the rate of the free population, which was almost 180‰. In our view, this is due to the different obstetrical problems that affected the reproduction of a large group of slave mothers, especially those who had been trafficked across the Atlantic. Many African women slaves probably became sterile as a result of the physical and psychological hardships associated with captivity, transportation, and living in a foreign land. In fact, only 13% of the baptized adult women slaves were recorded as having children. Moreover, 50% of the women slaves inventoried in Vila Rica seem to have conceived during their adolescent years, and 8% at advanced ages, which carries risks for the development of the foetus and the survival of the newborn, on the one hand, and for the mother’s health, on the other. The information on interpregnancy intervals also helps explain the birth and mortality rates of the contingent because more than half of the slave mothers registered in the baptismal records conceived in short and long intervals, i.e., before the first 24 months and after the first 48 months following the previous birth. This means that at least one of every two women slaves must have conceived in periods when foetal complications are likely to occur, thus augmenting the risk of maternal or perinatal mortality, and the incidence of miscarriages. To these problems we must add the psychological disorders generated by a life of slavery, which may have also affected these women’s pregnancies and the early life of their children.²⁵For a first methodological approach to the issue in the context of Seville, see Corona Pérez (2022, pp. 321-342). For the implementation of the proposition based on the results obtained for Vila Rica, see Corona Pérez (in press). In light of these findings, it seems evident that slavery was synonymous with worse conditions for the reproduction of most women slaves and the survival of their newborns. However, this situation must have been definitely more difficult for African women than for “crioulas” or “mestizadas” women born in Brazil.²⁶It has also been observed during the second half of the 18^th century and the first years of the 19^th century (Campos, 2011, pp. 217-220).

Baptismal records do not usually specify the colour or the origin of the slaves. As seen in Table 5, the information of most slave mothers does not include any references to their African origin or ascendence, or to their being “crioulas” or “mestizadas” (whether “mulatas,” “pardas” or “cabras”). However, when that information is provided-in 30% of the cases-, it allows us to build a sample that is representative enough to be able to draw some conclusions about the fertility rate of these women. Thus, African mothers, who were more numerous, fit in the above-mentioned pattern, with almost 70% of them having only 1 child, and an average of 1.5 children. In the “crioulas” and “mestizadas” groups, on the contrary, only a little over half of the women had only 1 child (53.5%), while 46.5% of them had 2 or more. These mothers had an average of 2.7% children, thus exceeding the natural replacement rate. They were familiar with the local social and cultural codes, and probably had better living conditions. As Botelho rightly pointed out in the case of Catas Altas do Mato Dentro, a process of “Brazilification” of the slaves’ profiles took place in Minas Gerais as the 18^th century progressed (Botelho, 2007, p. 468Botelho, T. R. (2007) “A família escrava em Minas Gerais no século XVIII.” In: M.ª L. de Resende, and L. C. Villata, orgs. História de Minas Gerais. As Minas setecentistas. Belo Horizonte: Autêntica & Companhia do Tempo, pp. 455-476.). This was due to the gradual increase of legitimate relationships, which, together with the fall of slave imports from the mid-1730 onwards, the deconcentration of slave ownership, and the formation and living together of stable family groups, favoured natural reproduction, especially after the 1760s (Botelho, 2007, pp. 469-479Botelho, T. R. (2007) “A família escrava em Minas Gerais no século XVIII.” In: M.ª L. de Resende, and L. C. Villata, orgs. História de Minas Gerais. As Minas setecentistas. Belo Horizonte: Autêntica & Companhia do Tempo, pp. 455-476.). This process can also be observed in the case of Vila Rica.

Figure 5 shows the evolution in absolute terms of the population of African, “crioulas” (descendants of African women born in Brazil), “pardas” and “cabras” (two categories that are the result of biological intermingling) mothers (Paiva, 2015Paiva, E. F. (2015) Dar nome ao novo. Uma história lexical da Ibero-América entre os séculos XVI e XVIII (as dinâmicas de mestiçagens e o mundo do trabalho). Belo Horizonte: Autêntica.). The three groups had a positive progression until the 1760s. Up until then, the profile of Vila Rica mothers had been dominated by African women. But after the 1760s, when the decline of the production of gold and the fall of the slave import market became evident, the process described by Botelho for Catas Altas took place: the number of mothers from transatlantic trade decreased, while that of mothers born in Brazil, whether “crioulas,” “pardas” and “cabras,” increased. This process, however, implies the parallel development of another one: the process of manumission.

As shown in Figure 6, the number of children of slave mothers gradually decreased as the century progressed due to the obstetric and perinatal problems mentioned before, which weighed down on natural reproduction, and to the drop in slave imports from transatlantic trade which had fuelled the market of Minas Gerais after the gold mining crisis but also due to the “alforrias” (processes of manumission) taking place in Minas Gerais. The loss of a large number of notarial records from the first half of the 18^th century at the archives of Casa do Pilar hinders the possibility of connecting the number of slaves inventoried with the number of manumissions, which would have enabled us to calculate the level of access to freedom in Vila Rica and its district. This information is indeed available for other “comarcas.” Around 20% of all the women slaves documented during the 18^th century in will and testaments and in post-mortem inventories in the mining “comarcas” of Rio das Velhas and Rio das Mortes, were freed or “coartadas” (Paiva, 2022, pp. 219-220Paiva, E. F. (2022) “Por meu trabalho, serviço e indústria.” Histórias de africanos, crioulos e mestiçados nas Minas Gerais, 1716-1789. Belo Horizonte: Autêntica.). These “forra” women would free their children from the legal principle of partus sequitur ventrem: once the women were emancipated, their descendants were born free. On the other hand, we can know the scope of these dynamics of manumission among the newborns of slave mothers because in the two Vila Rica parishes around 8% of those children were freed at the baptismal font between 1712 and 1770. As a result of it, the slave contingent of Vila Rica required a contribution from outside to renovate itself, mostly from transatlantic trade flows but also from immigration of slaveowners who moved with their slaves from other regions. Since the birth rate was not high enough to naturally replace the losses generated by mortality, a deficit to which the processes of manumission also contributed, the maintenance of the slave population depended fully on the provision of slaves from the transatlantic trade.