Introduction^{Footnote 1}

In recent decades the studies on African women have expanded dramatically. Yet, most of these publications have focused on the twentieth century, during the colonial period, or after independence.^{Footnote 2} Scholars have also started exploring the arrival of Christian missionaries in the end of the nineteenth century and how they transformed the lives of African women with their own notions of morality, femininity, and motherhood, but also regarding labor and property control.^{Footnote 3} Fewer publications focus on women before the end of the nineteenth century, although an important number of these studies centered their analysis on West Central Africa.^{Footnote 4}

Facing the challenges of writing a more inclusive gendered history of the nineteenth- century West Central Africa, where not only wealthy women are seen as historical actors, but also those in a marginal position, such as farmers, petty traders, and enslaved women, I searched for notarial records following the footsteps of historians who have indicated their existence of such sources in Angola.^{Footnote 5} Inspired by Latin Americanists who have made extensive use of parish records to reconstruct the experiences of women, slaves, and other social actors marginalized in more traditional primary sources, I made several attempts to visit the Archdioceses of Luanda, in Angola, in order to locate more information about common free and the non-free women in the nineteenth century Benguela.^{Footnote 6} Parish records, alongside wills, nominal lists, and a variety of colonial documents registered the lives of common women in Benguela, yet not many specialists on the African past have explored their potential.^{Footnote 7} Africanists have much to gain from the methodological contributions advanced by specialists on Latin American history, who have managed to write the history of the impoverished and enslaved population, where women also played important roles.^{Footnote 8} Among other things, baptism, marriage, and burial records allow a reconstruction of family history and a better understanding of social mobility, complementing other sets of historical evidence, such as travelers’ accounts and official correspondence. Ordinary women who lived in the center of Benguela or its suburbs did not leave diaries or memoirs, and very few of them were able to pay an attorney or priest to write down their wishes in wills or other official documents. Further, no one in twenty-first century Benguela, claims to be their descendant, which might have allowed for the existence of an oral family history, as can be found in other parts of the continent.^{Footnote 9} Many of the colonial documents, particularly official correspondence exchanged between authorities in Lisbon and in Benguela, underrepresented the general African population, particularly the poor, peasants, women, slaves, and the illiterate. In sum, the existence of the petty traders and the enslaved women of Benguela could be easily forgotten in Angolan historiography.

Alongside colonial documents, such as residential lists and censuses, notarial records offered me the possibility to get a glimpse into the lives of poor and enslaved women in Benguela during the nineteenth century. As historians working on Europe and Latin American have shown, parish records are full of short life stories, which can assist us in reconstructing how African women maneuvered around a colonial society that otherwise ignored them.^{Footnote 10} These sources also permit us to see the power of the archives and the state in creating classifications and social hierarchies. The notarial records might give the impression of being unbiased sources, appearing as straightforward sets of statistical information collected by colonial officers, who were well informed and able to classify the subject population “objectively,” recording self-evident differences. But as other scholars remind us, these sources are “coded texts, in which social categories were manipulated in function with gender and gender relations.”^{Footnote 11} It is clear that what colonial officers and priests reported as “observations” were informed by social expectations regarding place of residence, affiliation, language skills, and economic activities. We should not forget how gender also affected and affects perceptions of skin color and racial classification. These documents, although produced by colonial officers invested in maintaining patriarchal values and colonial subjects’ subordination, reveal the importance of women in the social fabric of Benguela. And, despite constant efforts to control and regulate women’s activities, colonial officers and priests recorded how women occupied public and private spaces, and shaped the operation of the colonial society in nineteenth-century Benguela.

There is a substantial literature about merchant women on the coast of Africa. Scholars have researched the role of African women as cultural intermediaries and economic agents, such as the case of the signares of St. Louis or Gorée, the nharas of Cacheu, and the donas of Luanda and Mozambique.^{Footnote 12} Studies have also emphasized the role of women in Christian missions, women as farmers, as subjects who suffered economic and social exclusion at the eve of European colonialism, and women as rulers and matriarchs.^{Footnote 13} These studies recuperate women’s role, retrieving them from previous analysis that had insisted in on invisibility. Women’s historical agency is evident in case studies that indicate the variety of roles occupied by women and the significance of their contributions.^{Footnote 14}

In Angola, as elsewhere, African women were largely excluded from the colonial administration and other political processes that affected their lives, from the interpretation of metropolitan legal codes to the administration of the bureaucracy, as well as the enforcement of ecclesiastic, civil, and criminal laws. If we limited our study to the official correspondence, the legislation, or even the writings of colonial officers stationed in Benguela, we might think that African women were not active or were not important. However, these same documents, when read against the grain, reveal how the Portuguese administration failed in implementing and imposing these legal, economic and social systems and restrictions over the population they were supposed to rule and control. The local population negotiated, challenged and even ignored many of the colonial legislation, as scholars have demonstrated.^{Footnote 15}

The traces of how enslaved and poor women showed up in colonial documents allow scholars to write a history of those excluded from the highest levels of political and economic decisions. Rather than writing a history from the point of view of the colonizers, parish records allow us to read between the lines and reconstruct how enslaved and poor women lived in Benguela for most of the nineteenth century. These were not powerful women in control of their options, but they were aware of their position in a colonial society and they made efforts to build social networks and to guarantee protections, as the parish records reveal.^{Footnote 16} Thus, it is possible to write about the underrepresented sectors of society, mainly the unfree and poor women, before the twentieth century. It is inspired by the work of historians who have shown that even when a person is in a single document, that one reference is evidence that contributes to examining marginalized women and uncovering their agency, despite the silence of other records.^{Footnote 17}

Baptism Records and Women in Benguela

In the case of Angola, scholars such as José Curto, Lucilene Reginaldo, and Roquinaldo Ferreira have begun to explore the ecclesiastical documents and have consulted the parish archives held at the Arquivo do Bispado de Luanda (ABL).^{Footnote 18} However, their focus was not necessarily on poor and enslaved women, although their research revealed a great deal about the social life of slaves in Luanda and Benguela. Through the ecclesiastical evidence they reconstructed social links and Atlantic connections of some of the men and women who lived in Luanda and Benguela before the twentieth century. And it is in their footsteps that I started using the documents of the ABL. Yet, very few scholars have made used of this archive, in part due to the difficulty in accessing the collection. The ABL is not a public archive, nor is it under the control of the national archive of Angola (Arquivo Histórico de Angola) or the Ministry of Culture. The ABL is a private archive, located in the offices of archbishop of Angola and its access depends on personal connections, favors, and the approval of the Bishop himself.

The first time I was allowed to consult the collection, in 2009, I had to schedule a personal audience with Bishop Damião António Franklin (1950–2014 ) to explain my research and my interest in the ecclesiastical documents. After the first audience, I had to come back twice for more chats and personal negotiation. After some weeks of long chats, attending masses, and insistence, I was granted access to the archives. The second time, in 2011, was relatively easier, since it only involved submitting a written request and waiting ten days for its approval.^{Footnote 19} In the end, I was only able to consult the documents for two days, since I was leaving Angola shortly. But the access was less personal and less arbitrary.

In both circumstances, I encountered parish records without any obvious organization, including finding some books on the floor, and several books that were damaged or at risk from insects, climate, and lack of conservation. Rats wandered through the rooms where books and documents were stored, further compromising the historical documents. The lack of air conditioning or any kind of preservation has harmed books and one can only wonder how many other baptism, marriage, and burial records might have been lost in the past three hundred years.

Very few resources are devoted to conserving these documents, and very few priests or laypeople recognize the value of these documents for the study of the Angolan past.^{Footnote 20} These are seen and labeled “arquivo morto,” or useless archives by the ABL’s staff. In a country where many people have been displaced during four decades of colonial and civil wars, with some having lost their government issued identity documents, Angolan citizens search the ABL for more information about deceased relatives or in search of proof of their birth in order to get a new identity cards. Parish records are accepted as official evidence for Angolans in search of showing a family link with Portugal, for example, in order to emigrate. Thus, the staff is there to help Angolans locate baptism, marriage or burial records of parents, grandparents, or great-grandparents. The “arquivo morto” is considered old paper, with no value or importance in contemporary Angola. Those documents, however, reveal an important aspect of Angolan cultural heritage and should be available for anyone trying to reconstruct the social history of West Central Africa.

The lack of an inventory or a registry of books makes it extremely difficult to identify what has been misplaced, lost or never produced.

For the parish of Nossa Senhora do Pópulo in Benguela I have been able to locate nine baptism books, for the following years.

Table 1. Baptism Records of the parish of Nossa Senhora do Pópulo in Benguela

In these books, 10,020 baptismal entries have been identified (see Table 1). It is not clear what happened to other baptism books. In the case of churches located in the interior of Benguela, we know that a fire struck the Nossa Senhora de Santa Ana located in the Vila Viçosa village, a short-lived Portuguese center in the the sobado of Kitata, sometime between 1769 and 1774, and the baptism books were lost, according to an annotation made in 1774.^{Footnote 21} Similar episodes likely happened in Benguela, leading to the loss of important historical documents.

The Concílio de Trento (11 November 1563) and, later on, the Constituições Primeiras da Bahia regulated the collection of baptismal, marriage, and burial records, and recognized their central importance to the expansion of Portuguese colonialism and conversion of the so-called “heathens.”^{Footnote 22} Despite a Vatican decision regarding what to record and what to ignore, the ecclesiastical agents had a certain degree of autonomy on the format and the details entered in each entry, called an assento. As a result, each priest left his imprint on the sacraments he performed and collected, which is very clear in the records available for Benguela. While some priests provided detailed information on those participating in the sacraments, others were more succinct, limiting the records to names and dates.

In general, the registers before 1814 recorded the name of each child or adult baptized, his or her skin color and legal status, and place of birth. Registers classified infants and adults according to their legitimacy in the eyes of the Catholic Church, that is, if the baptized was born outside of wedlock, she or he was registered as natural. Natural was also the label used for children only registered by their mothers, which opened the space for a future marriage and a possible recognition of paternity, which would result in the “legitimation” of the offspring. Baptized children or adults could also be considered legítimo if the parents had married in the Catholic Church. In the case of slaves, the name of the owner was registered. Following the initial information, the priest recorded the name of the mother, her skin color, place of origin, and status (free, freed, or enslaved). In the case of enslaved mothers, once again, the name of the slaveholder was recorded. The same information was recorded about the father, godfather, and godmother. In the case of representatives, the priests also noted their names. After dating the register, the priest signed the form.

Information about skin color, status, age, and economic activities were not necessarily recorded for every single individual present at the baptism or otherwise involved. Or even worse, in some cases, priests insisted in recording skin color, while in others this information is neglected. For example, on 5 December 1807, the priest João dos Santos baptized Maria, who was freed upon her baptism. No information about Maria’s skin color was recorded, but her mother was identified as black Ana, or Preta Ana, slave of Maurício Nunes Soares. The father was incognito or unknown. The fact that the young Maria was freed by Nunes Soares suggests that he might have been her father.^{Footnote 23} In the Portuguese empire, and Benguela seems not to be an exception, children freed at the baptismal basin tended to maintain family links with the slave owners. Although there is no recognition of paternity, similar cases in Luanda suggest that the father of the child was the owner of her mother.^{Footnote 24} This case also shows how reference to skin color was subject to the interest of the priest. While Ana was registered as black Ana, there is no mention to the color classification of her daughter and her slave master in the baptism register. While the skin color of mothers was carefully scrutinized and recorded by priests, the same cannot be said about the children baptized. Moreover, this case highlights the widespread experience of illegitimacy among the children baptized in the Benguela church. Although the expansion of Catholicism was considered an important factor in the Portuguese colonial enterprise, very few people married in the Catholic Church. Illegitimacy was evident in baptismal and marital records, and it was also mentioned when people petitioned for administrative positions or even disputed the content of wills.^{Footnote 25}

By 1846 some changes were introduced in the baptismal register. The individual names, status, and condition of legitimacy were recorded, but “nação” was introduced to capture the place of birth of the adult or child baptized.^{Footnote 26} As previously in the case of slaves, the priest recorded the name of the slave owner, making sure that bondage was established in the ecclesiastical document. Priests also recorded the names of the parents, their skin color, status, and their master if they were not free. References to skin color were made when the priest found that information to be relevant. For example, on 18 October 1846, when Amália, who was considered an adult woman of approximately eleven years old, was baptized, the priest did not mention her skin color. She was identified as from the nação Hera and her owner, Josefa Mendes, was registered as a black woman. There was no information about the parents of Amália, since she was probably brought from the interior in a slave coffle to Benguela and sold there to Josefa Mendes. Although the priest Bernando José Pinheiro recorded the skin color of the slave-owner, Josefa Mendes, he failed to mention the color of the godparents, José Gonçalves and Clementina Mendes. Or perhaps the lack of reference was intentional, allowing José Gonçalves and Clementina Mendes to enjoy some degree of color freedom in a society strongly influenced by ideas of captivity, blood purity, and legality.^{Footnote 27} The fact that this lapse might have been intentional becomes clear when compared to other cases. The same priest, Bernardo José Pinheiro, carefully recorded the skin color of other godparents, such as the case of Luzia Pedro, registered as a black woman. Luzia was the godmother of Teresa, an adult woman of Caluquembe baptized on 24 October 1846.^{Footnote 28} Curiously, no reference was made to the skin color of Luzia’s owner, Dona Joana Mendes, one of the famous merchant women of Benguela. Again, it might have been a failure of priest to record this information. But in a different set of document, Dona Joana Mendes was identified as a black woman. In her 1861 will, Dona Joana Mendes was described as a single black woman, born in the interior, in the region of Libolo.^{Footnote 29} As she was a wealthy resident of Benguela, the priest might have avoided racially classifying her. It is difficult to understand why and how skin color was ignored in some records, such as parish records, but emphasized in others, such as wills. It was probably related to the social position of the individual being observed and the relationship established between that person and the official who was recording attributes that referred more to social and economic position than to phenotype classifications.

After 1877, priests started recording the date of birth of the baptized, as well as the place of residency and occupation of parents. The name of the father was, for the first time, recorded before the name of the mother of the person baptized. For example, in July 1877, the two month-old Carolina was baptized. She had been born on 14 May 1877 in Benguela.^{Footnote 30} Her father, Francisco Quimuca, was from the fortress of Caconda in the interior, but resided in Benguela. Quimuca was single and worked as a stonemason. It is not clear how long he was in a relationship with Catarina Gueva, the mother of young Carolina. Catarina was a single woman from Bailundu, who lived in Benguela working as a laundress. The godparents, both resident in Benguela, were Augusto Rodrigues Varela, who worked in public works as an officer in the colonial administration, and Alexandrina Salvador. Although the record does not reveal the skin color of the godparents, we know that the father, Quimuca, as well as the godfather, were able to sign their own names. Their occupations and ability to read and write suggest colonial insertion and cultural adaptation.^{Footnote 31}

In most records slaves, freed slaves, and poor people were mentioned briefly, sometimes just by their Catholic names and place of origin, but the in-depth analysis of these records reveals the social networks that slave families established. Curiously, the baptisms of free and enslaved individuals were registered in the same book, unlike in Brazil, where even the ecclesiastical records were segregated. Baptismal records also indicate social mobility and the role of manumission in keeping the colonial slave society in check and the slave population under control. Like in Luanda and other parts of the Portuguese empire, manumission was relatively common and allowed the incorporation of former enslaved people into colonial society, sometimes as slaveholders.^{Footnote 32} This was the case of the slaves of José Antonio de Carvalho, who died in Benguela in 1798. In his will he acknowledged that he was the father of three children, Guiomar, Isabel, and Antonio, from three different enslaved women. Guiomar was the daughter he had with one of his slaves, named Bibiana. In his will, Carvalho freed Bibiana and offered her 50,000 réis in textiles and a molecona, a young slave woman “of the many she has in her company besides a slave girl I offered her.” Another enslaved woman, called Josefa, who was pregnant at the time of his death, was also freed, and received 30,000 réis in textiles and a young slave girl. Besides his former lovers, his daughters were also rewarded with slaves.^{Footnote 33}

In the case of parish records, many baptisms were accompanied by priests’ observations regarding the manumission of the baptized infant. On some occasions, priests recorded the amount paid for the manumission at the baptismal font, the reasons for freeing the enslaved, and the witnesses. The baptismal record, in these cases, acted as a letter of liberty for children or adults. For example, in 1809, Maria, a young girl was freed at the baptismal font. Her mother, Marcelina, was a slave of António do Campos Trovão. The father was declared incognito, suggesting that the slave master was the father.^{Footnote 34} In the 1794–1814 baptismal book, 72 children were freed upon their baptism (an average of 3.6 per year), while a few decades later, in the 1846–1849 book, only one case was registered, a decrease to 0.25 per year. In 1848, an adult man named Gabriel, from Ganguela, was freed by his owner, Francisco Higino Craveiro Lopes, at his baptism without any stipulation or justification.^{Footnote 35} Manumissions at the baptismal basin diminished dramatically between 1814 and 1849.

More analysis of the parish records and manumissions at the font will have to be carried out to confirm what seems to be a consensus in the literature related to manumission in the Portuguese empire: that enslaved women were more likely to be manumitted than male slaves.^{Footnote 36} Another aspect that deserves attention concerns the selection of godparents, particularly for enslaved mothers baptizing their children from a non-identified father to see if, as in Brazil, “god parentage was not generally used to reinforce the paternalist aspects of the master-slave relationship.”^{Footnote 37} It will be important to explore the skin color and legal status of godparents to learn whether the choice of godparents, particularly of children of the enslaved population, replicated the color and status hierarchies of Benguela.

There are a series of problems with the baptism records available for Benguela and its interior. First, only a minority of the population is represented, since few individuals turned to the Catholic Church to baptize their children or their slaves, as can be seen by the small number of registers per year. As a result, the records are more detailed and better representative of the Portuguese and Brazilian residents in Benguela than of the African population. Second, an unknown number of books have been lost or misplaced and those that survived sometimes contain illegible records due to pest or water damage. Third, demographic analyses of the population based on the parish records can be a challenge since there are several individuals registered with a single name, such as the dozens of Marias, Josés, or Franciscas recorded without any last names; or even two or three João Carvalhos or Maria Josés, as such common names could belong to more than one person. Another challenging task, however, is dealing with pages that are apparently missing from some books, suggesting that they have been removed for reasons which are not clear. These issues are not exclusive to Benguela, as scholars working with notarial records elsewhere have noticed the same difficulties and challenges.^{Footnote 38}

Regarding the missing records, it is worth mentioning that in some cases people asked for the original record to be inserted in the baptismal book, suggesting that the original baptism was not located. From the records, it appears that the priests did not register baptisms immediately after the ceremony, and the delay between the baptism and the annotation in the parish books may explain the reason behind some records being misplaced. For example, in 1800 Joana Pereira da Silva requested the inclusion of her original baptism in the Nossa Senhora do Pópulo records. She claimed that her original baptism records were missing, despite the fact that her mother, Dona Aguida Gonçalves da Silva, and several witnesses insisted that the baptism had taken place in 1785. In this case, the identity of Joana’s father was crucial in a legal dispute related to his will. According to the witnesses in this process, Nuno Joaquim Pereira da Silva had registered Joana as his daughter, and he had accompanied Dona Aguida in the baptism ceremony.^{Footnote 39} If the displacement of the baptismal records of wealthy residents happened, one can only imagine the vulnerability of poor and enslaved women, who sometimes depended on the parish records as a proof of paternity or even for their freedom.

Despite their limitations, these records provide an abundance of data on people who remain invisible or unaccounted for in other historical documents. Moreover, these records provide information about social networks. Baptism and marriages were not only religious sacraments that emphasized belonging to a religious community, they were important events, where free and enslaved women expressed and reinforced solidarity links, or even forged new ones with the godparents of their children.^{Footnote 40} Thus, despite the problems related to the collection of the data, the limited scope of the population, the colonial gaze that was revealed, plus the difficulties related to scholars having access, parish records provide information about family life, solidarity networks, and dependency that are not evident in other historical records. However, the fact that people sought to participate in the Catholic sacraments for themselves or their children should not be read as acculturation or even assimilation into the Portuguese Catholic world. Belonging to the Catholic community was almost a requirement for those who lived in the colonial urban centers. Baptism protected people against enslavement, for example, and created a series of rights that differentiated assimilated people from the rest of the population. People, however, continued to worship their ancestors and to follow habits that were unwelcome in the Catholic world.^{Footnote 41}

In Angola, ecclesiastical documents recorded the presence of people not seen as important in oral traditions, including rulers who were forgotten. One example was Joanes José Gaspar, who was omnipresent in colonial records and ecclesiastical documents in the early nineteenth century and who is not remembered in official oral traditions.^{Footnote 42} Gaspar was the soba, or the local ruler, of Dombe Grande, a chiefdom southeast of Benguela, for most of the first half of the nineteenth century. Yet he is not remembered even by the mais velhos, the elders, of Dombe Grande nowadays. Records, however, reveal the centrality of his family in the interactions between the colonial administration and the local political elites. Gaspar and his children dominated the parish records of Benguela and Dombe Grande for most of the nineteenth century, yet their existence has been erased from oral tradition.^{Footnote 43} Thus, ecclesiastical documents can provide a look into the past that other sets of sources, including oral traditions, neglect.

Marriage Records and African Women

While baptismal records include information about the enslaved population, the marriage records refer mainly to the free population. As can be seen in Table 2, of the 144 records from 1806–1853, only on four occasions were there references to the legal status of women, and those referred to freed women (two were registered as freed black women and two as mixed race). Although there is no mention to the legal status of the other women and men, the statement that four women were freed suggests that all the other records refer to free people. More revealing is the fact that all the freed women married freed men. Missing from the marriage book is any record of marriage between enslaved people, suggesting that in Benguela marriage in the Catholic Church was not as important as the baptism for the general population.^{Footnote 44}

Table 2. Marriage Records of the parish of Nossa Senhora do Pópulo in Benguela

The records in the marriage book shed some light on the lives of freed women and even poor women, but do not offer clues on the relationships of enslaved women. As mentioned earlier, no case of a slave getting married was found. For example, in 1806, Josefa Pereira, a freed black woman “natural of this continent” married Joaquim dos Santos Pires, a freed black man “also natural of this continent.”^{Footnote 45} Not much information is available about this couple, except their names, their status as free people, and their classification as black people born in the African continent. More revealing is that freed black women in Benguela married men in the same condition, freed men like themselves. There is no record of a free woman marrying a free man, for example, indicating how racial and status classifications limited the number of partners available. The other cases of freed women who were married in Benguela are Marta Maria da Conceição, a freed black woman who married Francisco Xaviers da Costa, a black freed man from São Tomé; Dona Dionizia Teodora da Silva, a freed parda from Rio de Janeiro who contracted marriage with Alvaro Ferreira da Souza, a freed pardo from Bahia; and Inês Xavier de Almeida, a freed parda woman, with no reference to her place of origin, who married João Modersto, a freed pardo man from Rio de Janeiro, who acted as surgeon assistant in the army.^{Footnote 46}Pardo is particularly a difficult term to translate into English, since it meant different things in different contexts. In late eighteenth-century Angola, pardo was a synonym for fusco or “civilized mixed race” people.^{Footnote 47}Pardo designated those who, regardless of being born in Africa of African parents, were seen as integrated into the Portuguese colonial society, acting and behaving in a way that the Portuguese administration considered acceptable. More pertinent for the purpose of this study is that fact that all the other marriages registered in the Livro de Casamentos were between free people. In Benguela, marriage in the Catholic Church was limited to free people, and only rarely were those who were freed from an enslaved condition married in the Church. Unlike other regions in the Portuguese empire, slaves faced difficulties in making their marriages public and “legitimate” in the eyes of the Portuguese colonial society.^{Footnote 48}

Curiously, when comparing the marital records with baptisms, a different picture emerges. The baptismal records from 1797–1805 revealed that some enslaved women married. It is curious to note that the legitimate children of slave couples were registered during the 1797–1805 period when no record of enslaved people getting married was located in the parish of Nossa Senhora do Pópulo. Between 1797 and 1804, eleven children born to enslaved mothers were baptized as legitimate, suggesting that the parents had been married in the Catholic Church. Unfortunately, I have not been able to locate marriage books before 1806, thus I cannot cross reference the baptism and marriage information for the period before 1806. After 1806, however, the records suggest that only free people married in the Catholic Church in Benguela. The few books available indicate that marriage at the Catholic Church was not easy and accessible for everyone. Only the economic and social elite of Benguela looked for Church recognition of their relationships, while the majority of the poor and enslaved population carried on with their family lives. They baptized their children, but avoided legalizing their relationship in the eyes of the Portuguese colonial state.

The data from the Livro de Casamento of Benguela also revealed that women classified as filhas naturais, that is from a couple not married, did not have trouble finding suitable partners in Benguela. Thirty-two women, out of 152 cases, were classified as naturais. One of them, Dona Felizarda de Nazareth, originally from Luanda, was already widowed by the time she married Bernardo Jose de Oliveira, from Braga in Portugal.^{Footnote 49}

Marriage records also raise interesting questions about color classification in Benguela in the nineteenth century. It becomes clear that those classified as pardos faced different opportunities, including marriage options, which could offer greater social mobility. Sheila Farias argues that in Brazil, pardos baptized their children and married in the Catholic Church.^{Footnote 50} While three women were classified as blacks (pretas) in Benguela (two of them were freed women), two were parda (and both of them were freed), and 147 were married without any notation of their skin color. It appears that priests preferred to ignore the racial classification of women, particularly those who married to foreign traders and colonial officers. As Elizabeth Kuznesof stressed for seventeenth and eighteenth century Mexico, the skin color or ethnicity of women was not recorded, perhaps allowing them and their descendants to assume the skin color of their spouses, who were likely to be better positioned socially and economically.^{Footnote 51} Thus, a free African woman who married a colonial officer or a trader in Benguela was usually listed without any reference to her skin color. References to color were restricted to poor women or women who married men of lower social standards, such as the few pardas and pretas explored earlier.

Yet, it is not clear in the records who were the pardos in Benguela. And, once again, I look for inspiration in the historiography about slavery in Brazil to understand events in West Central Africa. In the Brazilian context, Hebe Mattos has argued that pardo acquired a new meaning in nineteenth century with the increase of the number of free blacks in Rio de Janeiro. Pardo came to refer to all non-white people who were born free, thus differentiating them from the freed population, who carried the stigma of slavery. Bert Barickman and Douglas Libby, however, have shown that there were enslaved pardos in Bahia and in Minas Gerais, respectively.^{Footnote 52} A new generation of scholars, such as Roberto Guedes and Larissa Viana, are showing how color classification was flexible in the nineteenth-century Brazil, and the same person could be recorded as white, pardo, or black in different set of documents.^{Footnote 53} Skin color classification revealed more about social perceptions, wealth, and family connections, than an individual’s actual skin color. In the case of the parish records of Benguela, pardo was a skin color restricted to free people and only rarely used to designate the enslaved population. Between 1794–1814, 171 children were classified as pardo, most of them free or freed Benguela residents. On one or two occasions pardo was used as the color classification of an enslaved person. In 1800, the priest baptized Jacinta and registered her as a parda enslaved girl. Her mother, Teresa, was black.^{Footnote 54} The other parda enslaved girl, was Maria, who was freed on her baptism. Both girls labeled as pardas belonged to the same slave owner, Antonio do Campos Trovão, and in both cases, the names of their fathers was unknown, suggesting that Trovão himself had fathered children with his slaves.^{Footnote 55} So, before 1814, parda was a term associated with former slaves, never with free people. Three decades later, parda continued to be a common term in Benguela. For example, of the 1,998 people baptized in Benguela between 1846 and 1849, only one was listed as parda – Eugenia, an adult woman from Bihe in the interior of Benguela, was classified as a free parda.^{Footnote 56} Perhaps she spoke some Portuguese and behaved in a way that the priest considered “civilized.” Unlike her, most of those baptized between 1846 and 1848 were listed without any reference to color. By 1880, the term pardo was not used in the 119 baptism recorded. Thirty-three mothers were classified as blacks and three as gentia, or heathen. And in 82 cases, there was no reference to skin color.^{Footnote 57}

In the end, parish records, censuses, and other documents that registered color classification revealed more about the people who collected the information than those who were labeled with a particular racial description. More than designating skin color, pardo in Angola indicated a cultural mixing, where people were inserted into the colonial world as traders and colonial administrators, such as capitão-mores and the donas, giving shape to a culturally mixed society that blended Portuguese and local values from the Ndombe and others.^{Footnote 58} The meaning of pardo and racial classification in Benguela deserves attention, and it is important to stress the role of these documents in creating and assigning identifiers to people, in the past as well as in the present, when scholars engage with this historical evidence.

Death Records and the Study of African Lives

Another set of ecclesiastical data were information about women is available is the burial records. As with baptisms and marriages, deaths were also counted and recorded in Benguela. So far, four burial record books have been identified for the town of Benguela:

Table 3. Burial Records of the parish of Nossa Senhora do Pópulo in Benguela

Priests usually, recorded a full set of personal information in the burial records, including the full name of the dead person, the date of death, place of origin, status (free, enslaved, or freed), the sacraments the dead person had received, their place of residence and death, burial location, and the signature of the priest. In some records, the recorded information might provide further detail, for instance the parents’ names, skin color, marital status, spouse’s name, slaveholder’s name in the case of enslaved or freed people, and cause of death. The economic standing was implicit in such expressions as “with a will,” with a solemn will, or poor. In some cases, which also related to property control and social status, there were references to brotherhoods and occupations, particularly if the deceased was a colonial officer or military personnel.

Detailed burial record indicated the deceased’s social and economic position. For example, when the Portuguese born José António Marques died, the priest registered:

As this case shows, deceased men received a lot of attention. The priest in charge of the record detailed the family life and the wishes of the person buried. The same attention was not paid to women. Even in the case of important women, such as the wife of the governor of Benguela, the priest provided much less information. Ana Guilhermina Brissae Ferreira, who died in 1861, was not fully described despite being the wife of the governor. The lack of attention in this case might be linked to the fact that Ana Guilhermina was not Catholic. She was probably Protestant, since she never received any sacrament, indicating she was not baptized.

In the case of an enslaved woman, the registry entry was even shorter:

These records reveal how status social affected the burial records. Despite the brevity, such records provide much needed information on the enslaved and poor population, who were usually invisible in other types of documents.

Burial records reveal the possibility of studying age, gender, and cause of mortality among the free and unfree populations, as well as exploring the cultural practices related to death.^{Footnote 62} Similarly to the baptism and marriage records, there are a series of problems with the burial records. First, they are uneven on the number of people registered (as Table 3 indicates). Second, there are important gaps in the records available, with no burial record book located for the 1831–1858 period, a moment of profound shifts in Benguela associated with changes in the transatlantic slave trade.^{Footnote 63} Third, as the passages above demonstrate, the data is better detailed for the male free population than for the enslaved people. Fourth, only those buried on the colonial cemetery, initially located around the Nossa Senhora do Pópulo, received attention from the local priests. It is not clear what happened to those who died when there were no priests available or to those who did not receive a Catholic burial.

Ecclesiastical data is also available for places in the interior of Benguela, such as the Portuguese fortresses of Caconda and Quilengue, as well as Dombe Grande, and detailed study of these sets of documents will provide much needed information to reconstruct the lives of ordinary people in the urban center of West Central Africa.