People of Faith

People of Faith
A book in the series

Latin America In Translation / En Traducción / Em Tradução
Sponsored by the Duke–University of North Carolina
Program in Latin American Studies
PEOPLE OF FAITH
\
Slavery and African Catholics in
Eighteenth-Century Rio de Janeiro
mariza de carvalho soares

translated by jerry d. metz
duke university press

Durham & London 2011
∫ 2011 Duke University Press
All rights reserved
Printed in the United States of America on

acid-free paper $ Designed by Jennifer Hill.
Typeset in Quadraat by Keystone Typesetting, Inc.
Library of Congress Cataloging-in-Publication

Data appear on the last printed page of this book.
For Bruno and Juliana,
Sources of inspiration for my work and my life
contents
ix List of Tables
xi Acknowledgments
1 Introduction
part one
19 | 1 From Ethiopia to Guinea
40 | 2 Commerce with the Mina Coast
67 | 3 African ‘‘Nations’’ and Provenience Groups
101 Gallery of Illustrations
part two
113 | 4 Urban Life and Brotherhoods in the City
146 | 5 Constructing a Religious Norm
183 | 6 Conflict and Ethnic Identity among Mahi
223 Postscript
241 Appendix
249 Notes
293 Bibliography
309 Index
tables
9 | 1 Total Slave Baptisms Valid for Analysis, 1718–1760
54 | 2 Baptisms of Adult Africans in Sé Parish, 1718–1726
58 | 3 Baptisms of Adult Africans in Sé Parish, 1744–1750
71 | 4 Total Baptisms of Slaves in Rio de Janeiro, 1718–1860
77 | 5 Principal Proveniences of Slaves in Rio de Janeiro, 1718–1760

(with percentages for each period)
78 | 6 Principal Proveniences of Slaves in Rio de Janeiro, 1718–1760

(with percentages of the total sample)
94 | 7 Mina Adults Baptized in Rio de Janeiro, 1718–1760
97 | 8 Marriages of Female African Slaves in Rio de Janeiro, 1718–1760
99 | 9 Distribution of Child Baptisms according to Mother’s

Proveniences, 1751–1760
100 | 10 Distribution of Mina Baptisms by Parish, 1718–1760
124 | 11 Distribution of Houses by Parish in Rio de Janeiro, 1779–1789
130 | 12 Distribution of Slave Obituaries from Candelária Parish, by Locale

of Burial in the City of Rio de Janeiro, 1724–1736
134 | 13 Distribution of Slave Burials by Year and Sex in the City of Rio de
Janeiro, 1724–1736
x | List of Tables
242 | 14 Total Slave Baptisms in Rio de Janeiro, 1718–1760
242 | 15 Total Slave Baptisms Valid for Analysis, 1718–1760
243 | 16 Slave Baptisms in Sé Parish, Rio de Janeiro, 1718–1726
245 | 19 Slave Baptisms in Candelária Parish, Rio de Janeiro, 1751–1760
246 | 20 Slave Baptisms in São José Parish, Rio de Janeiro, 1751–1760
247 | 21 Slave Baptisms in Santa Rita Parish, 1751–1760

acknowledgments
The initial version of what would become this book was my
\ doctoral dissertation for the Department of History at the Fed-

eral Fluminense University, Rio de Janeiro. I thank the History
Department, where I already worked, for allowing me two years of unen-
cumbered time to organize and write the dissertation. In particular, I thank
my colleagues at the Laboratório de História Oral e Imagem, where my
relationships with Ana Maria Mauad, Ângela de Castro Gomes, Hebe Mat-
tos, Ismênia de Lima Martins, and Paulo Knauss always provided attentive
and fruitful collaborations. In compiling and organizing data from a range
of sources, I was ably assisted by Fluminense history students Luciana
Gandelman, Mônica Monteiro, and Juliana Barreto Farias.
I gained access to critical documents through the friendly generosity of
the members of the Brotherhood of Santo Elesbão and Santa Efigênia,
especially Américo Bispo da Silveira and Roberto Machado Passos. I also
wish to express thanks to the sta√s of the other archives and libraries
where I carried out research, of which the personnel at the Arquivo da Cúria
Metropolitana do Rio de Janeiro deserve particular recognition. Of course I
am also indebted to all the authors whose works I have cited in the follow-
ing pages, but I feel a fundamental obligation to those writers whose works
shaped my questions, research, and analysis in a more indirect or un-
noticed way; undoubtedly, my bibliography would be more complete if my
memory were a bit better.
Manolo Florentino and Luciana Villas-Boas were responsible for bring-
xii | Acknowledgments
ing attention to this work and fostering its publication in Brazil in 2000.
Since then, many people and institutions have engaged with it, and in the
process brought wider attention to it. I will highlight only some: Bernard
Vincent wrote a flattering review in Cahiers du Brésil Contemporain; Ambas-
sador Alberto da Costa e Silva read the text with exquisite care; with Hebe
Mattos and Silvia Lara I shared my work and permanent friendship. I owe
my utmost gratitude to my students, and to the great majority of my
readers, who are personally unknown to me. They have consistently pro-
vided the most meaningful recognition and incentive to do better.
The barriers of distance and language can appear formidable, but Devo-
tos, as my book came to be called in Brazil, was able to cross them in the
present edition thanks to the e√orts of several scholars and institutions.
Stuart Schwartz, A. J. R. Russell-Wood, and Mary Karasch were among the
first non-Brazilian historians to lend the book some prominent encourage-
ment. In 2001 I achieved my first ‘‘foreign’’ visit, at the History Department
at the University of Maryland, College Park, thanks in particular to Barbara
Weinstein. This position provided me a valuable opportunity to begin
establishing contacts with American universities, and since then I have
often traveled to the United States and Canada; I am very grateful to the
various institutions that have supported these academic exchanges. In
2003, thanks to a postdoctorate grant from Coordenação de Aperfeiçoa-
mento de Pessoal de Nível Superior—capes /Brazil, I spent a year at the
History Department of Vanderbilt University; Marshall Eakin, the chair at
that time, was tireless in his support. During that year I enjoyed frequent
visits to York University at the Harriet Tubman Resource Centre on the
African Diaspora (presently Harriet Tubman Institute), where I served as an
associate. During those years Jane Landers, Paul Lovejoy, and Elisée Sou-
monni became colleagues and friends. My first papers were published in
the United States thanks to Toyin Falola and Matt Childs (The Yoruba Dias-
pora in the Atlantic World, 2004), and to Gwyn Campbell, Suzanne Miers, and
Joseph C. Miller (Women and Slavery, 2008).
Finally, if the arrival of this English edition of Devotos da Cor (whose
original title could not be translated directly into English without losing
some force) is an expression of the book’s success in Brazil, its fortunes in
this new form will depend on a much wider circle of readers, critics, and
interrogators. I am grateful to all those participants at workshops and
Acknowledgments | xiii
conferences who heard my ideas and, even when debating them, encour-
aged me to continue. The final revisions for this edition were completed at
the Gilder Lehman Center of Yale University, thanks to a fellowship in
March of 2007. I thank the sta√ of the center, especially Dana Scha√er.
During this period, the friendship of Íris Kantor, Maria Jordan Arroyo, and
Stuart Schwartz was indispensable.
At last I want to thanks Jerry Metz for his translation and constant con-
versation; Leia Pereira and Monica Carneiro for helping to prepare the
folder with the images; and the team of editors at Duke University Press.
introduction
I paid my first visit to the Church of Santo Elesbão and Santa
\ Efigênia one Wednesday morning, back in 1989, and it was an

appointment with a friend that happened to take me there. The
church fronts onto a bustling street in central Rio de Janeiro, lined with
stores o√ering inexpensive clothing, shoes, and housewares. The din of
commerce is constant; enthusiastic inducements to buy blare from tinny
speakers mounted on shopfronts, and from sidewalk amplifiers. But in-
side, the church maintains an air of quiet, inconspicuous reserve, its se-
renity protected by a stately wooden panel standing in its open entrance that
discreetly bu√ers the noise while preventing passersby from glancing to-
ward the pews. Architecturally, it is small and unremarkable, especially
when one compares it to the other more sumptuous examples of eight-
eenth-century churches nearby; and perhaps because of this, it is not well
known. I had gone to meet Alberto Lobo, who suggested the location. ‘‘Seu
Lobo,’’ as he was also known, was born around 1900 and had come to Rio
de Janeiro from Bahia (in northeastern Brazil) around 1930. He was a
practitioner of candomblé (a Brazilian religion with an African background),∞
and had also joined the Church of Santo Elesbão and Santa Efigênia soon
after arriving in Rio. I was working on my master’s thesis, and I hoped to
share with him some of my discoveries and uncertainties about the history
of religious practices in Brazil—and to learn, in turn, from his stories and
perspective.
As it happened, we would have several delightful and (for me) helpful
2 | Introduction
conversations, speaking in hushed voices as we sat in the benches of the

old church. But Seu Lobo passed away the following year. I thought of him
regularly, especially when I began doctoral work in 1994. I often replayed
our discussions in my mind, looking for hints of insight or ideas that I had
not grasped at the time, but I never returned to the Church of Santo Elesbão
and Santa Efigênia.
However, a chance discovery, also in 1994, raised a series of questions
that would ultimately lead me back to an exploration of the humble little
church. I happened to pick up a copy of the Guia Brasileiro de Fontes para a
História da África, da Escravidão Negra e do Negro na Sociedade Atual (Brazilian
Guide to Sources on the History of Africa, Slavery, and the Negro in Con-
temporary Society), published by the National Archive. I was intrigued to
notice an unusual reference to the ‘‘Statutes of the Congregation of the
Minas Makii Blacks in Rio de Janeiro (1786), copy of document in the
National Library (1907).’’≤ The 1907 document I read at the National Ar-
chive was a handmade copy of part of a 1786 manuscript whose where-
abouts were uncertain. When I inquired at the National Library, the sta√
could determine only that the information pertaining to the document’s
accession (such as date and method of acquisition) had been lost. The
catalog had the description of the document,≥ with two call numbers for it,
an older—bn(MA)5,3,12—and a newer—bn(MA)9,3,11. There was also
another form of numeric descriptor, 11447, which turned out to be a
registration number from the important Exhibition of Brazilian History in
1882. The document had been displayed in a section related to Brazilian
ethnography (specifically, ‘‘Classe X Natural History/2nd Ethnography and
Linguistics/A. Brazilian Ethnography’’), rather than to slavery or religious
organizations. Indeed, it was classified as belonging to the study of so-
called heathen peoples, utterly disconnected from the histories of slavery
or the Catholic Church. Based on the exhibition catalog,∂ it seemed certain
that the document had been maintained at the library at least since the
exhibition. Through fruitless searches, my curiosity grew.
Upon the first reading, I was unaware that this document, substantial as
it was—nearly seventy pages—would transform the premises of my re-
search. It had been written by a member of a congregation of enslaved and
freed Africans referred to in the text as the ‘‘Kingdom of Maki’’ (or ‘‘Mahi,’’
according to modern spelling), which, I was astonished to learn on the
Introduction | 3
thirteenth page, was based in the very Church of Santo Elesbão and Santa
Efigênia where I had met with Seu Lobo two centuries later. Structurally,
the Mahi Manuscript contains two separate narratives (actually two di-
alogues).∑ The first recounts the election and administration of Francisco
Alves de Souza as the leader of the Mahi Congregation, and includes a
statute, dated 1786, that lays out the congregation’s devotional activities.
The second dialogue (undated but probably from the same period) de-
scribes the Mina Coast, with attention not only to its ports and castles but
to the various groups of people that lived there. This type of account was
common to the era and popularly known as a derrota.∏
For the better part of a year I pored over the Mahi Manuscript, intrigued
by the challenge to reconcile its components and situate the whole docu-
ment geographically and historically. At one point, the writer claimed that
the congregation numbered some 200 people in 1786, all of them originally
from the ‘‘kingdom’’ of Mahi on the Mina Coast. Drawing on what litera-
ture on the history of Africa I could compile, I determined that the Mahi
came from the hinterland of the Bight of Benin, north of the ancient
Kingdom of Dahomey. These Mahi arriving in Rio de Janeiro from the Mina
Coast were considered members of the Mina ‘‘nation,’’ whatever their
actual provenience. Since at least 1740, the Mina that composed the Mina
Congregation had elected a king to govern them, and that king had been
without exception a Mina himself. But one of the great surprises the Manu-
script o√ers is its portrayal of a group of Africans in Brazil, many of them
freed and all converted to Catholicism and deeply concerned for the salva-
tion of their kinfolk in Africa who had remained dedicated to their pagan
beliefs. I came to doubt much that I had studied previously about African
slavery in Rio de Janeiro, and gradually realized that—whatever my pre-
vious, tentative ideas for dissertation topics had been—the project emerg-
ing from the provocations of the Mahi Manuscript clearly focused on the
meanings and importance of Mahi identity within the African community
of eighteenth-century Rio.π
The first problem was to find additional sources that could help illumi-
nate the nature of the city in that era, and the composition of its slave and
freed populations. Most of the research on Rio’s urban slavery focused on a
later period, while works on the slave trade to colonial Brazil rarely ac-
knowledge the presence of West Africans in the captaincy of Rio de Janeiro
4 | Introduction
in the eighteenth century. Almost without exception, Brazilian studies of

slavery concentrated narrowly on the work relations and daily life of slaves
without considering the larger context of diaspora, and the transfer and
negotiation of identity among transplanted Africans. Anthropologists have
highlighted the fundamental contributions of peoples from the Bight of
Benin to Afro-Brazilian religious practice, but as historical societies these
peoples remain strangely anonymous. Thus it was by both interest and
necessity that I turned to the history of Africa, although this was still fairly
uncharted terrain for me and for most Brazilian historians at the time.
The Mahi are part of the Gbe language groups, and shared their territory
with di√erent Yoruba speakers.∫ Gbe speakers were and still are in what are
today Togo and Benin, and came to be called Jêje in Brazil, while Yoruba
peoples extend today across Nigeria into Benin and Togo, and were usually
termed Nagô in Brazil.Ω But these terms could be combined, as in the
expressions Jêje-Nagô and Jêje-Mahi found in the candomblé practiced in
Bahia.∞≠ Still, while one must never ignore the dangers inherent in imput-
ing ethnic veracity to classificatory terms from colonial sources, occasional
specific references to the Mahi in anthropological literature support the
notion that there were Mahi elements in the foundation of Bahian candom-
blé. In a collection of African myths related to the orixás (deities) of can-
domblé, Pierre Verger relates the story of Obaluaê-Xapanã (the orixá asso-
ciated with smallpox) who with his warriors left Empé, the region of Nupe
(called Tapa in Brazil), to lay waste to the Mahi. Before the attack, the Mahi
consulted an oracle, who suggested certain o√erings be made to Obaluaê-
Xapanã. The o√erings carried out, the deity was not only placated but
charmed into dwelling with the Mahi; and since that time, the Mahi king-
dom prospered.∞∞ Perhaps the Manuscript refers to such tales, as well as to
the somewhat more easily verifiable episodes of fierce resistance to inva-
sion by neighboring forces, when it asserts that the Mahi Kingdom was
one of the ‘‘most powerful of the Mina Coast.’’
Although there is also evidence of the Mahi in Bahia by the middle of the
eighteenth century, we have been unable to trace their arrival or existence
earlier than the 1740s, when the Manuscript clearly indicates that they were
not merely present but actively organizing. A preliminary survey of the
growing literature on African history shows that Mahi people probably
constituted less a kingdom than a series of dispersed small villages, with
Introduction | 5
no centralized authority or government. It also seems that during this

period, the Mahi were caught up in the expansion of both the kingdom of
Dahomey and the slave trade; the historian Akinjogbin notes that traders
were coming to consider the general region that included the Mahi lands as
one continuous ‘‘hunting ground’’ for human captives.∞≤ According to
Verger, there was a marked presence of Dahomeans in Bahia from the late
seventeenth century until 1735.∞≥ From around 1730 onward, the frequent
incursions on Mahi land by Dahomean aggressors led to a gradual but
consistent increase in the numbers of Mahi shipped to Bahia throughout
the remainder of the century; this demographic wave included one Fran-
cisco Alves de Souza, who arrived at the port of Salvador, Bahia. When he
went to Rio in 1748, he fell in with a group of so-called Mina blacks, all of
whom originally came from the Bight of Benin and spoke the ‘‘general
tongue of Mina.’’∞∂ Souza would later become head of the Mahi Congrega-
tion, and would author the Mahi Manuscript.
Nina Rodrigues, a Bahian medical doctor and scholar of African cul-
tures in Brazil, referred to the dwindling population of Mahi in the city of
Salvador at the end of the nineteenth century:
The number of Jêjes [Ewe or Eves] has markedly diminished in Bahia.∞∑

Those that remain lack a proper district or headquarters of their own in the
city, and wander dispersed here and there, some in Campo Grande, some in
the Rua dos Sapateiros. The population seems to be evenly split among
males and females. I have encountered a few from Dahomey, but almost all
are from the coast, having left behind the cities of Ajudá, Ouidah, Popo,
Agbomi, and Kotonu. Many of them say they are Efan, and display their
distinctiveness from the Dahomeans with a characteristic tattoo that in-
cludes a small burn mark. Others are Maí, a small group residing to the
north of Dahomey that has su√ered cruel persecution from its southern
neighbor. Two Maí families that lived near the fountain on São Pedro hill
furnished me with much valuable information on the history of Bahia’s jêje
colony; but today, those two families are reduced to one aged survivor.∞∏
My own research also benefited from the kindness of informants. At the

Church of Santo Elesbão and Santa Efigênia I conducted several helpful
interviews. I also came across a bound folio containing (in painstaking
transcription) the brotherhood’s compromisso, or foundational statutes∞π —
6 | Introduction
in Portuguese—as well as the complete ecclesiastical correspondence be-

tween the worshippers of Santo Elesbão, the bishop of Rio de Janeiro, and
the Tribunal of Conscience and Orders in Lisbon, from 1740 to 1767, per-
taining to their approval. Unfortunately, the volume lacks information
about its own origins—date of compilation, name of author, or any men-
tion of how or where the transcription of the statutes took place; despite
repeated e√orts, I could not locate the original documents. The compro-
misso states that the brotherhood had been organized by a diverse group of
Africans from the Mina Coast, Cape Verde, the island of São Tomé, and
Mozambique. It also specifies that ‘‘blacks from Angola’’ would never be
permitted membership in the brotherhood. This is notable because at least
since the sixteenth century, so-called blacks from Angola had held leader-
ship positions in the Irmandade de Nossa Senhora do Rosário dos Homens
Pretos (Brotherhood of Our Lady of the Rosário of the Black Men), estab-
lished since the 1720s in Rio’s ‘‘Church of Rosário.’’
While researching Portuguese archive catalogs, I came across entries
referring to a proposal to form a congregation dedicated to Our Lady of
Remédios, also in the Church of Santo Elesbão and Santa Efigênia, dating
from 1788. The first reference appeared in the records of M. A. Hedwig
Fitzler and Ernesto Enes, published in 1928; the other was included in the
inventories of the Arquivo Histórico Ultramarino in Lisbon.∞∫ In both ar-
chives, the document is catalogued the same way,∞Ω which helped me locate
it on microfilm at the National Library in Lisbon. Examining it, I began to
perceive a connection between the three Mahi Catholic organizations—the
Mahi Congregation, the Devotion to the Souls of Purgatory, and the Frater-
nity of Nossa Senhora dos Remédios—all within the Church of Santo
Elesbão and Santa Efigênia. Rarely have scholars been able to identify such
organizations, and pursue their histories through diverse primary sources,
at such an early period in the Americas. Of course, the ultimate goal is to go
beyond identifying and listing these organizations, and to use the range of
documentary evidence surrounding them to begin to reconstruct how they
were conceived, negotiated, and lived by their members. Indeed, this far-
flung set of documents not only contains information about the Mahi
associations but was for the most part actually written by them as well.
These documents, particularly the Mahi Manuscript, can be read as some-
thing of an ‘‘ethnic text’’ whose analytic possibilities extend far beyond the
parameters of this study.≤≠
Introduction | 7
But I had a more immediate question to address. What could I make of

this substantial presence of slaves from the Mina Coast in eighteenth-
century Rio de Janeiro, a city where Bantu peoples had long been assumed
to be the principal, perhaps near-exclusive African group? Where could I
get a more complete portrait of the Africans brought to Rio in the early and
middle parts of the century? I first turned to the baptismal records of slaves
at the Archive of the Cúria Metropolitana in Rio de Janeiro–acmrj.≤∞ It can
be di≈cult to explain to nonhistorians why baptismal records are some-
thing to be excited about; but particularly in Brazil, before the 1890s, they
not merely documented the performance of baptisms but served the key
function of identifying people—free, freed, or slave—in detailed ways. They
also provided a medium in which other important events or commemora-
tions touching colonial families could be inscribed,≤≤ as in the case of one
newborn slave whose master freed him in the baptismal basin because he
was ‘‘the first progeny of the household.’’
The Constituições Primeiras do Arcebispado da Bahia (1719), the first o≈cial
book of guidelines for ecclesiastical practice in Brazil, includes regulations
for baptismal records but does not substantively address how to record the
conversions or baptisms of African slaves; in this, it echoes an omission in
the earlier Ordenações Filipinas (1604).≤≥ Still, the general instructions for
record keeping are clear: there should be a bound book, with numbered
pages, each page signed at the top by the vicar or attendants; the front and
back cover of the book should indicate the number of pages. In the form of
a statement by the priest performing the baptism, a record should contain
the date of the ceremony, the parish in which it occurred, and the name of
the child as well as the names of his or her parents and godparents.≤∂ This
model was standard for recording baptisms in Brazil throughout the eigh-
teenth century, but in practice the existing notations often included infor-
mation about the juridical status of the enslaved or freed individuals under-
going the ceremony. Close reading of entries can also help us hear the
individual authorial voice of the functionaries who wrote the books, mak-
ing the basic exercise of comparing and contrasting their content quite
rewarding. It is at once the standardization of the information required by
the book, combined with baptismal records’ subtle peculiarities of detail,
which suggested to me that they are one of the conceptual and legal spaces
where colonial society constructed slave identity. For this reason I turned to
them in hopes of better understanding the composition of slave demo-
8 | Introduction
graphics in Rio de Janeiro, and, proceeding from there, in evaluating the

nature of Mahi identity within the city’s cultural tapestry.
But again, whatever the potential of such a research method, its success
depends on solid and abundant sources; and despite the heroic e√orts of
many librarians, there were predictable frustrations in this regard. I had
access to only three books containing slave baptism records from the two
parishes of Rio in the first half of the eighteenth century: Livro de Batismo de
Escravos, Freguesia da Sé (the seventh Baptism Book of Captive Slaves of the
Parish of Sé, 1718–26); and two others, lacking covers or titles—Livro de
Batismo de Escravos, Freguesia da Sé, 1744–61, and Livro de Batismo de Escravos,
Freguesia da Candelária, 1745–44 (from the Parish of Candelária). Other
books dated from 1751 or later, when the city added two more parishes: for
example, Livro de Batismo de Escravos, Freguesia de São José, 1751–90 (records
from the Parishes of São José), and Livro de Batismo de Escravos, Freguesia de
Santa Rita, 1751–99 (Santa Rita, 1751–99). In the last-mentioned book, the
e√ects of humidity had rendered illegible all the pages for 1751, and some
of those for 1752.
From this incomplete series, I framed my collection of data within two
periods—before 1751, and the decade from 1751 to 1760. The earlier period
must be characterized more tentatively, since its analysis is based on data
first only from the Parish of Sé (1718–26) and then on both Sé (1744–50)
and Candelária (1744–50). The second period is a shorter time frame but
includes the whole city, which is vital for comprehension of the spatial
distribution of Mina slaves across Rio’s urban landscape. In the end, there
are a total of 9,578 registered baptisms of supposed slaves whose masters
lived within the four parishes of the city. I say ‘‘supposed’’ because there are
occasionally children of freed slaves among the lists, or a few children of
indigenous mothers who should not be recorded as slaves. Where there
were significant doubts, those cases were excluded, reducing the total to
9,269. It should also be noted that the number of individual registers will
be larger than the number of records of baptism ceremonies, which often
included more than one individual participant.≤∑ Those 9,269 registers can
be broken down in the following manner as seen in table 1.
All the calculations used throughout the study are based on the 9,269
registers considered valid for analysis. Still, although a large and verifiable
number seems to o√er a certain reassurance to the researcher, far more
Introduction | 9
Table 1. Total Slave Baptisms Valid for Analysis, 1718–1760
parish and years baptisms total

Baptism with Mother Baptism of Adults
Sé (1), 1718–26 983 855 1,838
Sé (2), 1744–50 1,381 283 1,664
Sé (3) 1751–60 893 324 1,217
Candelária, 1751–60 1,648 586 2,234
São José, 1751–60 1,117 321 1,438
Santa Rita, 1751–60 587 291 878
Total 6,609 2,660 9,269
Source: acmrj, Livro de Batismo de Escravos, Rio de Janeiro: Sé, 1718–26, 1744–61;
Candelária, 1745–74; São José, 1751–90; Santa Rita, 1751–99. The parishes of São José
and Santa Rita were created in 1751.
profound questions hover over this data. For instance, how are we to
interpret such descriptive labels as ‘‘gentio (‘‘heathen’’ [literally ‘‘gentile’’])
from Mina’’ or ‘‘Mina nation,’’ which are common not only in the parochial
books but in all the ecclesiastical documentation? How did the Mahi arrive
in Rio de Janeiro? How did they begin to organize? What was the signifi-
cance of religious organizations in the life of urban slaves in the eighteenth
century? And what does the historiography say, or not say, about such
questions? These were the overarching concerns of my research project.
Following the trail of Africans from the Bight of Benin to Rio de Janeiro
involved making lists of countless names and miniscule details which
might, or might not, provide key connections later. This book is the prod-
uct of two years’ worth of coaxing tiny pieces around in an Atlantic-sized
jigsaw puzzle; the emerging image was finally clear enough, at least, to
draw some conclusions from this phase of the research.
One of the greatest challenges I faced was in grappling with the com-
posite names given to Africans, which combined Christian first names
(with or without a last name) with kingdoms, locales, or some other
geographic allusion: ‘‘Maria Antunes of the heathens of Guinea,’’ ‘‘Pedro,
black of the Mina nation,’’ ‘‘Elório Cabinda.’’ There is always some sort of
group designation to these names, indicating that beyond the utility of
10 | Introduction
individual identification the names were partly a strategy to classify Afri-

cans brought as slaves to Brazil into recognized categories of physical
aspect, behavior, ancestry, language, or beliefs. Of course, it was not only a
Brazilian phenomenon. The French artist Jean-Baptiste Debret, who lived
in Rio from 1816 to 1831, drew a street vendor of cornmeal who, he wrote,
‘‘looks to be Congo, judging by the shaved head and the attitude of the
turban.’’ When the famed naturalist Louis Agassiz and his wife, Elizabeth
Agassiz, visited Rio in 1865, she commented that ‘‘Mina women are very
pretty . . . with an almost noble bearing.’’≤∏ Such descriptions were part of
the general practice for constructing ethnic-functional ‘‘types’’ by which
Africans were comprehended and inserted into colonial society—as do-
mestic workers, field hands, miners, or, for the women (lest one forget), as
desirable sexual partners. A plethora of such terms composed a crude sys-
tem for conceiving, evaluating, and representing themes of such complex-
ity that we still lack the basic theoretical and methodological approaches to
grasp them.
This quandary is not alleviated by recourse to the Brazilian historiogra-
phy, which has been far more concerned with tracing the expansion of
Europeans in Brazil than in looking closely at the Africans and their par-
ticular situations. Neither have historians been immune to more recent and
broad-based generalizations about racial and cultural diversity, at least
when applied to African-born peoples. The three concepts of race, eth-
nicity, and origins have come to refer almost inextricably to each other,
although each subject has di√erent theoretical requirements and, while
their empirical terrains may overlap, they are not interchangeable. Accep-
tance of the study of ethnic identity has come to historians slowly, through
the gradual interdisciplinary transformation of social science as a whole.
Since the 1980s interest has grown in understanding the social and cultural
diversity of the African populations brought to Brazil. While new sources
are necessary, it is also critical to interrogate known sources in light of
these new queries and sensitivities. Three scholars in particular were pi-
oneers in the field: Stuart Schwartz, A. J. R. Russell-Wood, and Mary
Karasch. Perhaps it is not a coincidence that they work in institutions in the
United States, where the study of African history is far better established
than in Brazil. However, two other influential scholars, Pierre Verger and
João José Reis, have explored the history of slavery and Africans in Bahia,
where the presence of slaves from the Bight of Benin was quite marked.≤π
Introduction | 11
Rather than purely following these authors’ leads and creating a social
history of Mina slaves in Rio de Janeiro, I elected to base my study in
religious behavior—more specifically, in the participation of Africans
(slaves or freed) in Catholic organizations such as the black lay brother-
hoods. Lay brotherhoods have been a newly emergent theme in the Bra-
zilian historiography, with works by Julita Scarano and Caio C. Boschi
focusing on Minas Gerais.≤∫ Religiosity is an example of what Fernand
Braudel calls ‘‘mental frames’’ of long duration,≤Ω but here the conversion
of Africans under slavery in the Americas is treated as a source of infinite
possibilities that can only be perceived in the short term. To that point,
Braudel’s notion of the danger hidden in everyday occurrences has influ-
enced the tone of my work.≥≠ But the brief moment of disputes, tensions,
and conflicts—so viscerally present in the proceedings of the Mahi Congre-
gation—emerged from the long run of history. My work is placed squarely
at that temporal intersection, with the objective of viewing in their integrity
both the particular circumstances of the religious activity of the Mina in
Rio, and the long unfolding of slavery, the slave trade, and the African
Diaspora that surrounds it.
Anthropological theory, notably from such Latin American anthropolo-
gists as Roberto Cardoso de Oliveira, João Pacheco de Oliveira, and Miguel
Bartolomé, has informed my interdisciplinary approach to history. By em-
phasizing the fundamental relations between ethnic identity and culture,
these and other anthropologists help demonstrate that a historiography
excluding the long, diverse, and often violent episodes involving African-
born slaves in Brazil, or in any other place in the Americas, will be only
an impressionistic outline.≥∞ Norbert Elias’s classic sociological work on
France,≥≤ delicately transposed to colonial Rio de Janeiro, has helped me
imagine the textures of a society pervaded by the rules of sociability of the
ancien régime. In Rio, as throughout the Portuguese empire, black lay
brotherhoods engaged in expressions of imperial pomp through elaborate
festivals, processions and funerals, the election of kings and queens, and
the organization of royal courts, all within their ethnic and provenience
groups. This book, by focusing on the establishment of lay brotherhoods
among freed and enslaved Mina linked to the Church of Santo Elesbão and
Santa Efigênia from the early 1740s to the late 1780s, attempts to compre-
hend the importance of Catholicism in the new social configurations these
people adopted living under slavery.
12 | Introduction
The diversity of sources I encountered enabled me to follow a relatively

small group—around 200 people in the Mahi Congregation, and several
thousand Mina dispersed around the city—for a fifty-year period. Notwith-
standing my interdisciplinary approach, this work is grounded in the em-
pirical standards and methods of history. My overarching concern is to
understand how Mina people in general, and the Mahi in particular, assem-
bled and organized themselves through meaningful religious activity un-
der slavery.≥≥ Thus I read widely in the history of slavery and of the Catholic
Church in Brazil. The historiography of Brazil and Brazilian slavery pro-
vided the contours of the ‘‘historical situation’’ where my questions of
ethnic and religious identity would be directed.≥∂ Slavery, ethnicity, and
religiosity are essential factors in how a group of Africans in colonial Rio
de Janeiro constructed their identity and social organization, dialoguing
with the past and the new rules and determinants of a Brazilian slave
society.≥∑
Another central challenge I faced was to compose a su≈ciently rich and
meaningful portrait of the era, which would be the stage on which these
various actors lived their lives and engaged each other. I was reminded
many times of Boschi’s advice:
If one endeavors to study religion in the colonial setting, one cannot keep as
a parameter or yardstick the historical norms of doctrinal Catholicism,
uttered by theologians and canonical law. If indeed this parameter ever
existed, the Portuguese colony in the Americas would have been an unlikely
setting for their realization. Rather, what one sees here is a popular Catholi-
cism characterized by both the precarious conditions of evangelism and
expanded devotional and protective spheres. Thus, analysis of colonial reli-
gion has to somehow get inside this pretended exteriority. Only in that way
can one begin to comprehend, for example, the real meanings of sacred
festivals in that context; and to identify, under the diaphanous mantle of
superficiality, the original expression that is a hallmark of Brazilian colonial
religiosity.≥∏
Studies of slavery in Brazil were another framework to be considered. I

traversed the di√erent moments of the historiography, including classics
by João Lúcio de Azevedo, Maurício Goulart, Pierre Verger, and newer work
by Manolo Florentino (who is the first author to explore in depth the
Introduction | 13
commerce of slaves in the city of Rio de Janeiro). Still, none of these

scholars dwelled overlong on the commercial relations between Rio and
the Mina Coast in the first half of the eighteenth century.≥π Florentino does
look at the Atlantic slave trade, especially the dealings between Angola and
Rio in the years after 1790, when the documentation is more substantial.
Works treating the years before 1790, such as Luiz Felipe de Alencastro’s
on the seventeenth century and Herbert S. Klein’s on the mid- to late
eighteenth, have also tended to focus on Angola.≥∫
Inspired by Hebe Mattos’s analysis of the Das Cores do Silêncio (silence of
color) in the nineteenth century,≥Ω I could observe something di√erent in
the eighteenth century—a period in which social status was determined
overwhelmingly by color. Elites were supposedly ‘‘white’’ and of ‘‘clean
blood,’’ and the ‘‘blacks’’ were slaves or slaves who had been liberated.
Occasionally there were pardos, somewhere in between the two extremes
but di≈cult to grasp in a comprehensive way from the sources. In this era,
color bespoke social condition and one’s place in the social hierarchy. Take
the woman known as Páscoa (Easter), said to be ‘‘preta mina forra’’ (freed
Mina black): Before being identified as from Mina, she was called ‘‘black,’’
and even though she was a freed woman her freedom comes last in her
name, after the insistent ‘‘preta mina.’’
I relied on the period chroniclers of city life who showed an interest in
the religious practices of Rio’s African population. Among them, Augusto
Maurício recorded much insightful information about churches, and
quoted from eighteenth-century documents. Other authors—such as Mon-
senhor Pizarro, Moreira de Azevedo, Padre Perereca, Vieira Fazenda, and
Vivaldo Coaracy—helped fill in the gaps in the documentation about the
brotherhoods or the city itself.∂≠ When taken as a whole and read in se-
quence, these sources are also extraordinarily challenging because they not
only regrettably refrain from citing the additional material they draw from,
but they often address comments, allusions, and arguments to each other
in a style of partially closed or coded conversation. All these writers shared
the goal of creating a consensus about the past, whose doubts or mysteries
would be gradually resolved as knowledge was accumulated in a method-
ical (if dangerous) procedure. This may explain why one author often
seemed to address or refer to another with whom he agreed, while ignor-
ing or grumbling about an author who held a divergent position. Such
14 | Introduction
intertextuality has to be kept in mind to analyze or contextualize any given

single document from the group of chronicles.∂∞
Images and engravings are rare for this period, and the few that do exist
still await concentrated analysis. Lygia da Fonseca Fernandes Cunha should
be recognized for, among other things, publishing the plates of Carlos
Julião that are likely from the 1780s. Maps are rather more numerous and
have been studied by Gilberto Ferrez and Eduardo Canabrava Barreiros.
Occasionally I refer to images from the first decades of the nineteenth
century, but only when the images convey what were understood to be
places or practices that dated from the eighteenth century. Still, even these
images are rarely unaccompanied by text that ‘‘explains’’ them, whether the
image is meant to illustrate something in the text or (as in the case of
Debret’s work) the text is meant to provide context to the image.∂≤
One of the risks for historians in the recent trend to ‘‘read’’ images for
what written text does not or cannot articulate is that the content of the
image can be unmoored, forced to do interpretive work far from its own
historical circumstances. For instance, we have no record of how the Mina
people in eighteenth-century Rio dressed, or arranged their hair, or applied
physical markings; for the period, there are only Julião’s images of uniden-
tified Africans involved in special festive events. The Mina woman pointed
out for praise by Elizabeth Agassiz was seen on the streets of Rio in the
middle of the nineteenth century. Demographics suggest that that woman
was a Nagô, that is, Yoruba speaker, who had likely come to Rio from Bahia
after the Malê uprising.∂≥ In the era I address, Gbe speakers were predomi-
nant in the city compared to Yoruba-speaking people. These are distinct
languages, traditions, and cultures, which likely had an influence on local
social identities and strategies adopted to cope with slavery. But still, since I
lack images to create more complete portraits, the people in this book must
remain to an extent faceless. All historical research projects contain omis-
sions, but this is one I particularly regret.
I make one final comment about the appearance of period texts. Al-
though transliteration is a common practice today, I opted to transcribe
directly, because it seems that to change the words of authors from a
distant century adds an additional interpretive burden to the historian
already struggling to understand them in their fullest historical context.∂∂
Nonetheless, punctuation has been made current, and obvious errors in
Introduction | 15
grammar or orthography have been corrected. In this edition all the En-
glish translations are written in correct modern English.
The narrative exposition of a research project must di√er from the non-
linear, sometimes chaotic, and often highly obscure process of researching
itself. Therefore, although I began the investigation long ago with the
transcription of the Mahi Manuscript, that document is specifically ad-
dressed only in chapter 6. The chapters can be imagined as a series of six
photographs taken of the same subject, each with a tighter zoom and
sharper central focus: starting with European exploratory formulations of
Africa and the Mina Coast, and ending with the Mahi community in Rio de
Janeiro that formed a lay brotherhood in 1786.∂∑
The book is further organized into two parts. The first presents the Mina
Coast, the commerce of slaves to Brazil, and the formation of what I call
grupos de procedência (provenience groups). Chapter 1 analyzes the conquest
of Guinea in the fifteenth and sixteenth centuries; it also attempts to trace
how designations such as Guinea and Mina were selectively applied in
practice, and how they came to be lodged in Western thought. Chapter 2
looks at the slave trade from the Mina Coast to Rio de Janeiro in the
eighteenth century to begin to understand who were the so-called Mina
heathen in the city, and where they came from. Chapter 3 considers the
Mina from the perspective of the classifications used by the church, such as
baptism records, to try to tease out some averages for population size, their
diverse relationships, and the texture of their presence in Rio. Key to this
discussion is the period term gentio, (heathen), which appears throughout
the book because of frequent use in the documentation; it refers to people
ignorant of the ways and teachings of Christianity (unlike pagans, who
learn about but reject baptism).
Part two situates the Mina within the panorama of enslaved and freed Af-
ricans in the city, with particular attention to religious practice and internal
organization. Chapter 4 focuses on the development of black lay brother-
hoods in Rio during the seventeenth and eighteenth centuries. Chapter 5
articulates an analysis of the twenty-year process of constructing the com-
promisso of the Brotherhood of Santo Elesbão and Santa Efigênia and the
multiple negotiations involving members of the brotherhood themselves
and the representatives of the Portuguese Crown. Chapter 6 details the
organization of the Mahi Congregation, as well as issues that arose from
16 | Introduction
their ethnic identity, exploring the Mahi Manuscript for hints of how the
process of identity construction intersected with strategies for power.
My research goal was to find a consistent path through diverse docu-
ments that pointed to a new understanding of African slavery in the Por-
tuguese Empire. Twelve years after the dissertation was written, I have been
able to augment the African dimensions of the research. As I carried out
research in the 1990s, there was still an almost total lack of interest in
African history in the Brazilian academy, which also meant that essential
books on the subject written in other languages remained inaccessible to
readers of Portuguese. All of that is certainly changing, although not as
quickly or substantially as some of us would like. Still, to have included
references here to all the related, expanding themes and bibliographies—
from the history of slavery and of Africa, to diaspora studies and the
Atlantic world—would have doubled the size of this book, or demanded a
second one. For that reason, I decided to leave the text almost as it was
already published in Brazil—with the addition of a postscript that contains
details about the Mahi that I have subsequently learned. My last fifteen
years of e√ort have been dedicated to understanding the specific experi-
ences of Africans taken as slaves to the Americas, and how they collectively
reconstructed their lives within, and at times apart from, captivity. This
book emphasizes the importance of groups, and not isolated individuals,
in this historical context. I hope it stimulates scholarly debate but, also and
especially, more research into ethnic identity and provenience groups
among the Africans in the New World.
Finally, I’d like to note that in the Brazilian edition of this book, I used the
term ‘‘procedência’’ (provenance) to refer to the trajectories of captive Afri-
cans, from the starting points of their journeys until landing in Brazil. The
term was borrowed from Nina Rodrigues, one of the founders of this type
of study in Brazil; I extended its application from individuals to groups,
called grupos de procedência (that analysis appears here in chapter 3). For the
American publication, however, I opted for the term ‘‘provenience’’ utilized
by Melville Herskovits. Thus I use ‘‘provenience group’’ to designate the
form of organization created by Africans in the New World that is rooted in
the reference to a shared provenience.
1 | from ethiopia to guinea
As Renaissance Europe probed its southern frontier through
\ trade networks branching across the Mediterranean, its mer-

chants, scholars, royalty, and commoners alike gazed in de-
lighted wonder at the bags, chests, and bundles arriving from distant lands
by ship and caravan. Precious metals, ivory, ostrich feathers, strange furs,
and hides—these and other exotic stu√s were joined (and often physically
conveyed) by dark-skinned slaves, who were themselves both a highly
valued import commodity and a provocative object of study for Renais-
sance elites eager to situate themselves at the apex of a world market-
geography as yet incompletely formulated.
In this period, knowledge of faraway realms was mostly obtained
through the voyagers’ accounts that had been accumulating in the litera-
ture since antiquity. According to Herodotus (fifth century bce), beyond
the Saharan Desert sands one could reach a region of great forests and
bogs, with a vast river full of crocodiles that swam upstream, in a river
flowing from the sunset to the Levant (supposedly the Niger River). The
Greek historian had also alluded to entire cities of black people there; and
he gave an account of the formidable Garamantes, who spent their days
hunting the Troglodytes—a strange people who subsisted on snake meat
and communicated with batlike squeals. Four centuries later, another
Greek, the geographer Strabo (1 bce –ace 1), provided a harrowing de-
scription of the ‘‘numerous deserts’’ that formed a barrier to the explora-
tion of ‘‘the country of the Occidental Ethiopes.’’ By the beginning of the
20 | Chapter One
Christian era, Roman explorers and philosophers added their own ac-
counts to the Greek repertoire. With those writings in hand, Pliny the Elder
(first century ace) devised his own list of the peoples inhabiting the
remote area south of the Sahara.∞
Until the early fifteenth century, Europeans based their understanding
of far-o√ lands and peoples on materials such as these. They had traveled
only as far as the Mediterranean coast, parts of Egypt, the edge of the
Sahara, and the northern stretch of the western coast of Africa (up to the
Atlantic archipelagos). The Portuguese, however, were soon to open a new
era in Atlantic exploration in general, and African exploration in particular.
Their conquest in 1415 of Ceuta (today an enclave in Morocco), a strategic
port city in North Africa, eventually came to be a vital Portuguese commer-
cial base with established links to the Muslim world. Portugal thus had
new access to the caravan routes that fanned across the north, west, and
east of Africa, while it took advantage of Ceuta’s seaside location to launch
its own navigational forays up and down the African coast. It would be
through a combination of their own actual exploration and the assimila-
tion of Muslim familiarity with the region’s physical and astronomical
features that the Portuguese could, in the mid-fifteenth century, start to
devise a new African geography.≤
To the south, the Sahara separated Portugal and North Africa from the
city of Timbuktu (in modern Mali), already an important center of learning
in the early fifteenth century as well as a key trading post where caravans
would exchange cargo, haggle for supplies, and water their camels. The
city’s location—near the banks of the Niger River, and at the intersection of
trade arteries bearing salt, gold, and other goods—was favorable for both
culture and commerce. Salt, extracted from the mines at Taghaza (in today’s
Mali), was conveyed from Timbuktu on to West Africa, south of the Sahara.
In exchange for the salt, as well as for other merchandise imported into
West Africa, the black kingdoms sent back to Timbuktu gold, slaves,≥ and
ivory, along with particular commodities prized by the Mediterranean mar-
ket (such as black pepper, cola nuts, and amber). The west coast was ac-
cessed by three land routes: one leading to Arguin (o√ Mauritania); one to
the city of Safi; in Morocco, and one to Cantor, in the Lower Gambia.
Renaissance maps suggest that due east of the Guinea Coast, if one could
cross or circumnavigate the entire African landmass, lay Oriental Ethiopia,
From Ethiopia to Guinea | 21
a place of Christian kingdoms where Santo Elesbão (Ethiopia) and Santa

Efigênia (Nubia) originated. According to these cartographies, Guinea
denominated a narrow slice of the western coast, situated at around the fif-
teenth parallel.∂ Little was known of the surrounding territories or the peo-
ple that inhabited them, particularly south of the Saharan sands. That would
begin to change with the Portuguese arrival on the Mina Coast in 1470.
If one compares the depictions of Africa produced by European car-
tographers during the first half of the fifteenth century to those produced a
century before, the similarity implies that European conceptions of African
geography had scarcely changed in a hundred years’ time. Even by the
1450s and 1460s, exploratory expeditions did not penetrate far enough to
revise the boundaries established during antiquity. But if the early car-
tographic limits were not as yet being challenged in a fundamental way,
increasing contact between explorers and the so-called native forest peo-
ples, who had been known only from their fabulous descriptions in ancient
texts, would soon transform European thought and society.
To understand this process, it is appropriate to begin with the writings
of an influential contemporary, the crown’s royal chronicler Gomes Eanes
de Zurara (1410–74). Whatever his personal opinions or intentions, Zura-
ra’s work constitutes a detailed toponymy that gave narrative form and
lexical texture to the West Coast of Africa. In a sense, Zurara initiated the
entire scheme of categories and classifications that would underpin the
way slavery was understood in the Portuguese Empire.∑ The expression
terra de negros (land of the negroes) already appears in his writing in the
middle part of the century, designating the region to the south of what is
today called the Senegal River.
the land of the black moors

Gomes Eanes de Zurara was the author of many books, including the
Crônicas de Guiné (Chronicles of Guinea), which remained as a manuscript until
1841. The chronicle had been written at the request of King Afonso V. João
de Barros cited the Chronicles in 1552, but by 1556, Damião de Góis claimed
that it had been lost; the Manuscript was ultimately discovered by Ferdi-
nand Denis in the Paris Library in 1837, and published in 1841.∏ The book
contains ninety-seven chapters, of which the first seven are devoted to
22 | Chapter One
lauding the notable deeds of Infante Dom Henrique (1394–1460, also

known as Henry the Navigator), organizer of Portugal’s seaborne expedi-
tions of Africa. While listing Henry’s lofty motivations to ‘‘search for the
lands of Guinea’’ (gain true knowledge of the region, assess the power of
the moors, search for Christian princes, save the souls of heathen), Zurara
found a delicate way to suggest that Henry’s underlying personal inspira-
tion came from a di√erent dimension altogether:
It is because his ascendant sign was Aires, which is the house of Mars, and is
the exaltation of the Sun, and his Master is in the eleventh house, accom-
panied by the Sun. And the aforementioned Mars was in Aquarius, which is
the house of Saturn, and in a place of hope, which meant that he should
struggle to accomplish great feats, especially seeking things hidden to other
men, and secrets according to the quality of Saturn, in whose house he is.
And because he was accompanied by the Sun, and the Sun was in the house
of Jupiter, he knew all of his achievements would be faithfully and gracefully
done, to the pleasure of his king.π
From the eighth chapter on, Zurara recounted various expeditions along
the African coast. It should be emphasized that since around 1440 in
Portugal, the conquest of new lands and the commerce in both gold and
slaves had been tightly associated. This association had developed out of
earlier strategies and victories, such as the taking of Ceuta, an important
trading city, from the Spanish in 1415; and the successful crossing in 1434
of the ‘‘sea of darkness’’ that had swallowed many European ships around
Cape Bojador, the Bulging Coast of North Africa. The first African gold in
Portugal arrived in 1442 in the form of a ransom for two Moorish hos-
tages;∫ the gold arrived with ‘‘ten black slaves, and some trinkets produced
on the coast.’’Ω In 1443, when Portuguese navigators made it to the island of
Arguin, a base was established to trade with the caravans for a variety of
goods such as wheat, horses, pitchers, bowls, combs, bracelets, shawls,
linens, eyeglasses, and needles.∞≠ A year later the expeditions reached Can-
tor, near the borders of (as the chronicles told it) the ‘‘land of the negroes,’’
or Guinea; it is here that the Portuguese began assembling large shipments
of slaves. An African enterprise in the region lived handsomely for years
through the commerce of slaves to Algarve, in the south of Portugal.
The expedition reports that Zurara compiled allowed him to interpret
and demarcate the transition from Islamic Africa (including the Sahara) to
the ‘‘land of the negroes’’ (equatorial Africa). Zurara commented on the
passage of the navigator Dinis Dias across the border:
He arrived in the land of the negroes, who are called Guineans. And as we
have had occasion to say several times in this narration, Guinea was the
other place where the pioneers explored. We by necessity write of the land as
continuous from the north to the south, but there are in essence two distinct
places, marked by great di√erences, and lying far apart from each other.∞∞
Drawing from Dias, Zurara specified a physical marker of the border be-
tween the two regions—a pair of palm trees—which aided the explorers
who soon followed, such as Gonçalo de Sintra, who took this path to
pass from the ‘‘land of the moors’’ (Sahara) to the ‘‘land of the negroes’’
(Guinea). Zurara related de Sintra’s journey: ‘‘The caravels having left
behind the lands of the Sahara, they soon espied the two palm trees which
Dias identified as the beginning of the land of the negroes.’’∞≤ And he
emphasized that ‘‘the peoples of this green land are all black [negros], and
that is why it is called the land of the negroes, or Guinea, and the people are
called Guineans, which means the same thing as negro.’’∞≥ In these writ-
ings, Guinea was emerging as the uncharted land between the better-
known regions of coastal Northwest Africa, and Christian Ethiopia far to
the east.
Zurara terminated his account in 1448, when he replaced Fernão Lopes
as the principal royal chronicler of the House of Bragança. That move was
probably not unrelated to Lisbon’s creation of the House of Guinea (Casa de
Guiné) in the southern Portuguese city of Lagos in 1445 to better adminis-
trate its oversea activities, in particular on the African coast. Zurara soon
complained that there had been a regrettable change in Portuguese strategy
with respect to maritime expansion and African exploration, a deterioration
that the mostly commercial functions of the House of Guinea represented.∞∂
It is clear that Zurara had the ambitious sense of being the chronicler of
Portugal, not merely the biographer of Dom Henrique.∞∑ Nevertheless, he
reiterated the broader spiritual goals attributed to Henry the Navigator in
looking for Guinea in the first place, such as the possibilities of finding
Christian kingdoms and of saving souls, and misconstrued the new era and
its commercial goals.
24 | Chapter One
This helps us understand why he would come to insist in the Chronicles

on describing the inhabitants of the ‘‘land of the negroes’’ as Moors, a
curious fact which has already caught scholars’ attention.∞∏ However, I
argue that Zurara was neither careless nor ingenuous. The roots of his
argument lie in the papal bulls of the period. The Dum Diversas of 1452
conceded the right to Portugal to conquer the Moors; and the deeply
‘‘imperialist’’ 1454 Romanus Pontifex gave to the Portuguese royalty all the
lands discovered beyond the Capes Bojador and Num, while extending the
permission to conquer ‘‘Indians’’ as well as Moors. Finally, the Inter Coetera
of 1456 gave to the Order of Christ (with Henry the Navigator as their
Grand Master) the legal and spiritual authority over lands discovered to be
non-Christian.∞π Thus, Zurara gave the Guineans the unlikely designation
of Moors so that the 1452 papal bull could be used to justify and legitimize
their conquest by the Portuguese Crown.
Zurara’s preoccupation with nomenclature in turn helps us date the
completion of the original edition of his Chronicles of Guinea, a point of
contention in the Portuguese historiography. A. J. Costa has argued that
the texts were written in their entirety after the death of Dom Henrique,
between 1464 and 1468, while Duarte Leite suggested that Zurara began
writing in 1451 and finished between 1460 and 1466.∞∫ My research sug-
gests that Leite is more accurate. But Zurara would have written his chroni-
cles for the king after the 1452 Dum Diversas, and before the 1454 Romanus
Pontifex, which is why his discourse assumed the programmatic conversion
of Moors but not of Indians or heathens.∞Ω
If Zurara knew how to distinguish Moors from heathens, he also called
for a project to convert ‘‘Indians’’ (permission for which would be granted
in the 1454 bull). He suggested that Indians are more easily converted than
Moors because they ‘‘do not come from the lineage of Moors, but of
heathen [ gentios]’’; thus they are more readily brought to the path of salva-
tion.≤≠ But to demonstrate how those black Africans who might not be
Moors still could and should be converted, Zurara reminded his readers of
how e√ortlessly several young boys and girls born of black African ‘‘gen-
tio’’ families in the city of Lagos (in Portugal) had become ‘‘good and true
Christians.’’≤∞
To the extent that the Portuguese e√orts of the first half of the fifteenth
century were, as Zurara claimed in the Chronicles, to gain true knowledge of
the region of Guinea, the second half of the century would be marked by an
intensification of commercial relations and the insertion of the ‘‘discov-
ered’’ peoples into networks of imperial and religious power. These three
imperatives were perhaps less distinguishable than Zurara tried to main-
tain, and he holds a complex, ambivalent legacy at the center of them.
the mina of guinea

Zurara was later replaced as royal chronicler of the House of Bragança and
High Guard of the Tower of the Tombo by Rui de Pina (1440–1523), who
wrote accounts of the contemporary conquest of Kongo as well as continu-
ing to compile information about Guinea. Two other coeval writers added
to the corpus of knowledge about Guinea: Duarte Pacheco Pereira (dates of
birth and death unknown), and João de Barros (1496–1570).≤≤ Pereira—
soldier, navigator, knight of the house of Dom João II, and inventor of the
cosmograph—finished his Esmeraldo de Situ Orbis between 1506 and 1508.
Named captain of the Fort of São Jorge in Mina (also known as Mina
Castle) in 1519, he resided there from 1520 until 1522, when he was accused
of various malfeasances and taken back to Lisbon in chains. João de Bar-
ros, manager of the Portuguese administrations for the House of India and
for Mina, wrote extensively on history, although some of his work has been
lost. In Asia 1a. Década, he also recounted the conquest of the West African
coast and furnished detailed descriptions of Mina in particular.
By the mid-1450s, after the caravels had first reached the island of
Arguin, they were continuing their southward explorations past the estuary
of the Gambia River.≤≥ At the time, Portugal was receiving around 800
slaves a year. In 1460, as the expeditions reached what is today Sierra Leone,
Henry the Navigator died. Because he had no descendants, the crown
incorporated his patrimony, which took nine years to complete; that same
year, 1469, a substantial contract to develop both slave commerce and
Portuguese territory in Guinea was awarded to Fernão Gomes. The five-
year contract stipulated that annually, Gomes had to secure 100 more
leagues of land beyond the region of modern Sierra Leone.≤∂ Acting to
fulfill this contract, João de Santarém and Pero Escobar reached modern
Ghana by 1470, bartering for gold there, and by 1472 had arrived at the
islands of São Tomé and Príncipe.
26 | Chapter One
Of course, Portugal was not the only European power exploring the
region. A Flemish ship landed on the Mina Coast in 1475, but the Por-
tuguese accounts declare that ‘‘the negroes ate all thirty-five Flemings on
board.’’≤∑ In 1479, the Treaty de Alcáçovas, which brought to an end the
War of the Castilian Succession, gave to Portugal the dominion over
Guinea, Madeira, the Azores, and Cape Verde, reserving the Canary Islands
for Castela. Anxious to centralize its control over the growing commerce in
gold and slaves, the Portuguese Crown in 1481 created the House of Slaves
in Lisbon to regulate and tax the trade. A year later, in 1482, they ordered
the construction of a fort on the Mina Coast, in today’s Ghana, to protect
the trade from other nations (especially Spain).
The task of building the fort was assigned to Diogo de Azambuja, a royal
knight and commentator of the Order of São Bento (Benedictine Order).
He chose to name the fort for São Jorge (Saint George), of whom King João
II was a votarist. According to Duarte Pacheco Pereira, all the material used
to build the Mina Castle, every stone of it, was brought from Portugal. Final
touches to the construction of the first major European building in tropical
Africa were performed in 1486. Anchored by the fort, the Portuguese
conception of the Mina Coast began to coalesce around this extension of
Iberian identity and bureaucracy into what had been Guinea’s vast and
di√use complex of ports, ethnic groups, business deals, and maritime
routes. By late 1486 the growing population of Portuguese and converted
Africans in the environs of the fort led to the designation of a city there, also
named Mina.
From the first, the Portuguese allied themselves with particular groups
of amenable natives who might be seen less as local agents than as partners
in a globalizing commercial system based on gold and slaves. Individual
dealers carried the gold to the fort, and little was communicated or re-
corded about its origins, although most of it seems to have come from west
of the Portuguese outpost. We do know that as early as the beginning of the
sixteenth century, slaves were being taken from various kingdoms along
the coast, principally Benin but also those farther southeast, into coastal
Central Africa and beyond. But the central importance of the kingdom of
Benin as a reliable, nearby source of slaves quickened the interest of Por-
tuguese traders and explorers. Even before the Mina fort had been com-
pleted in 1486, João Afonso de Aveiro had been awarded a contract to
establish Portuguese-controlled trade in a region favored by slave hunters,

neighboring the kingdom of Benin, east of the fort. No less than the king
of Benin himself agreed to supervise the capturing of slaves for sale to João
Afonso, who agreed to build a warehouse ( feitoria) in a nearby place called
Guato. From there, the slaves would be marched to the fort and either
traded to other merchants for gold, or dispatched on a waiting ship for
Portugal.
Africans were the middlemen in a system that seemed to benefit every-
one, except of course the slaves. An African could capture slaves for politi-
cal reasons, but the presence of Europeans eager to barter o√ered addi-
tional motivations or conveniences. Even those African traders who dealt
in other forms of merchandise could use slaves to carry their goods in
caravans, only to trade those slaves at the end of the route for the gold also
prized by Europeans. In global partnership, these complicated local net-
works produced a voluminous commerce of human beings that by the
1600s led the whole region, from the falls of the Volta River to the Niger
estuary, to be designated the ‘‘slave coast.’’
But in the midst of all this, the Africans who negotiated with the Por-
tuguese were also, from the beginning, targets of missionary e√orts aimed
at their conversion to Christianity.≤∏ Somewhat more rarely, locals could
also request conversion, as occurred during the period João Afonso was
negotiating with the king of Benin. A representative of the king ap-
proached the missionaries, asking for a group indoctrination into the
Catholic faith. This was apparently unsuccessful because the king had to
be converted first in order to then convert his subjects, but according to
João de Barros, ‘‘The King was too attached to his idolatries.’’ When João
Afonso died in 1486, the contract was suspended and the missionaries
abandoned the area, but the slave trade continued.≤π
The coast of Kongo was first reached early in the 1470s. Regular contact
with the Kongo Kingdom and Angola was established later, by Diogo Cão,
in 1482. Historians have noted that the Portuguese presence in this part of
Africa was characterized by violence and an expansion of slave trading,
despite all e√orts to introduce the Catholic Church there.≤∫ Arriving in
Kongo, the Portuguese realized that what they called Guinea (the realm
between the Islamic Sahara and the Catholic Ethiopia) was in fact much
longer than Zurara’s writings had led them to anticipate. Whatever the
28 | Chapter One
case, the discovery of these territories unknown to previous explorers and

cartographers stretched the geographical concept of Guinea to encompass
both the Mina Coast (including the islands of the archipelago of Cape
Verde), and the south-central area of Kongo and Angola. Changing con-
ceptions of the scale and composition of the area are reflected in the words
of the royal chronicler Rui de Pina, who located the kingdom of Kongo ‘‘in
Guinea, far beyond Mina.’’≤Ω This toponymy also justified the addition of
another title to the King of Portugal: Lord of Guinea (attributed in the 1479
Treaty of Alcáçovas, but adopted by Dom João II in 1485): ‘‘King John, by
God’s Grace King of Portugal and the Algarves, whose dominion covers
both sides of the Sea, Lord of Guinea in Africa.’’ Rui de Pina argued that the
title was valid because of the growing importance of the region in the
Portuguese Empire: ‘‘Kings can legitimately lay claim to a dominion
through its contributions and Apostolic concessions; but in the past, until
very recently, Guinea was a very small thing, and not worthy of being
claimed by a King.’’≥≠ Once Guinea was no longer a ‘‘very small thing,’’ a
new royal title was in order.
The first shipments of slaves leaving the Kongo were routed through the
fort at Mina to be resold. Soon another commercial route was established,
with stops in São Tomé, to accommodate the growing numbers of slaves
from the area. The increasing importance of the Mina Castle was not only
financial but symbolic, coming to represent the robust Portuguese colonial
initiative. Other forts were soon constructed at strategic locations in the
empire, such as Luanda, Mozambique, Goa, Macau, and Itamaracá.
The Portuguese missionary project gained a nominal victory with the
conversion of the king of Kongo, baptized as Dom João I, around 1491. The
king soon abandoned Christianity, but his son, baptized Dom Afonso I,
stayed truer to the faith and worked to continue conversions in the king-
dom early in the 1500s.≥∞ Still, the apparent fickleness of Africans with
respect to Christianity was not the only limitation faced by the Catholic
Church. It was simply not easy to find missionaries willing to risk the
area’s inhospitable conditions and many diseases to work there. Afonso’s
son D. Henrique was educated in Portugal and made a bishop in 1520, but
he never returned to Kongo. Portugal’s suggestion to instruct converted
Africans to teach the catechism was deemed too dangerous to ever put into
practice.
In 1488, with the rounding of a notorious rocky headland o√ modern

South Africa, the so-called Cape of Torments or Cape of Storms was re-
named the Cape of Good Hope. With this achievement, the entire West
Coast of Africa had been e√ectively delineated by Portuguese navigators,
and Guinea could be accessed by a series of Portuguese ports. The new sea
routes to the Orient that this made possible were crowded and disputed by
Portuguese, Dutch, French, and English ships; meanwhile, forts, towns,
and factories of various European nations dotted the coastal territories.
The new maritime routes transformed not only coastal linkages but also
helped expand trade into the West African hinterland. Between 1450 and
1500, Portugal’s inventory of slaves traded across the Atlantic totaled ap-
proximately 150,000 people.≥≤ In a classic study of Atlantic history, the
French scholar Pierre Chaunu observed that ‘‘black Africa was definitively
sundered from the Maghrib and set adrift in the sea.’’≥≥ Of course, there
were no concrete new divisions between coastal and Saharan Africa, but
the connections between them had been remade.
Renaissance maps of the African landmass would have to be redrawn,
and period maps depicted an Africa growing in detailed complexity with
every generation of cartographers. The Insularium Illustratum, a planisphere
drawn by Henricus Martellus in 1489, shows the southern coast as traced
by Bartolomeu Dias. From the Cape of Good Hope onward, lack of infor-
mation left the drawing vague, and the eastern coast lacks the island of
Madagascar (that would first be called São Lourenço). For their part, the
contours of Asia still largely resemble what Ptolemy had imagined. Soon,
however, the increasing expeditions toward both Asia and the Americas
resulted in an innovative 1502 planisphere called Cantino’s Map, after its
creator, Alberto Cantino. This work shows the Southern Hemisphere,
including the American continent and the Caribbean, in remarkable detail,
and includes a recognizable depiction of North America.≥∂
Across the Portuguese empire by the sixteenth century, the word Guinea
becomes a generic label for the source of slaves. Requests for slaves in
Brazil, for instance, never indicated a specific region from which slaves
should be drawn. The letter to the king from Duarte Coelho Pereira (1485–
1554), donatary captain of the captaincy of Pernambuco, was typical in
asking for ‘‘some slaves from Guinea.’’ Padre Manoel da Nóbrega similarly
appealed for ‘‘Guinea negroes’’ to work at the Colégio dos Jesuítas in
30 | Chapter One
Bahia. But within the royal administration that regulated the slave trade
through such controls as licensing and taxation, the record keeping was
more precise. This body responded to the accumulating requests from
Brazilian colonists for ‘‘Guinea slaves’’ by authorizing the captain at São
Tomé, in 1559, to send up to 120 ‘‘slaves from Kongo’’ to each plantation
owner who received certification from the governor of Brazil.≥∑
The city of São Paulo de Luanda, founded in 1575, reflects how the
commerce in slaves had come to give both form and direction to the
Portuguese presence in Africa. Between 1575 and 1591, Angola exported
some 50,000 slaves to Portugal, Spain, and their respective colonies.≥∏ By
contrast, the trade in both slaves and gold from the Mina Coast that had
produced such notable riches for Portugal the previous century was in
evident decline. In 1585, the crown attempted to lease this part of its
commercial holdings to suitable investors for 24 contos per year; once the
first contract had expired, given the lack of interested parties, Mina re-
verted to the king’s control.≥π
During the period of the Union of the Iberian Crowns (1580–1640), the
Portuguese explorer and merchant Duarte Lopes proposed the construc-
tion of two new principal facilities to rationalize the warehousing and
marketing of slaves in each country: one in Lisbon, focusing on slaves from
Kongo, São Tomé, and Angola; and the other in Seville, handling slaves
from Cape Verde and the so-called Guinea Rivers.≥∫ His proposal was not
acted upon, but it indicates the growing demarcation between two general
geographical sources of slaves on the West Coast of Africa—and it also
reflects Portugal’s sharpening interest in the central part of the coast. This
is further revealed in Duarte’s exertions to create a bishopric in the Kongo
city of São Salvador, where he lived for several years; the bishopric was
approved in 1596. By the beginning of the seventeenth century, both com-
mercial and religious activity had divided the formerly shapeless Guinea
into two distinct realms: the Mina Coast, and Kongo/Angola, both of them
advanced posts in the Portuguese Empire.
A wide survey of maps, o≈cial sources of various kinds, and even period
literature about the Portuguese vision of Africa in the fifteenth and sixteenth
centuries has left me with two overall conclusions. First, it is unwise for a
historian to adopt phrases such as ‘‘Guinea Coast’’ or ‘‘Guinea slaves,’’
which are common in Brazilian documents until the eighteenth century,
without first carrying out a vigorous study of the meanings of the label
‘‘Guinea’’ in the particular contexts involved. Second, and whatever Guinea
can be understood to signify, by the early fifteenth century the Mina Coast
had already taken on a distinct identity in Portugal’s toponymy of the
western African coast. It is possible to conceptualize it as a territory corre-
sponding to a stage of conquest (the contract of Fernão Gomes); a deter-
mined constellation of African ‘‘nations’’ that together constitute the ‘‘hea-
thens of Mina’’; and the fort, often called the Mina Castle, which both
symbolized and inspired the Portuguese colonial project.
the map of africa in the historiography

of the slave trade
The changing Portuguese presence in Africa can be viewed in a series of
four phases: first, the expeditions advancing along the coast, as Zurara
recounted—1415–48; second, arrival in the ‘‘land of the negroes’’ and the
recognition of Guinea—1448–69; third, growing commercial bases in the
Mina Coast, São Tomé and Kongo, described by Rui de Pina, and admin-
istered by a burgeoning royal authority—1469–82; and fourth, the explora-
tion of Angola, after 1482. As the geographical concept of Guinea grew
during this period, the western boundaries of a Christian Ethiopia, antici-
pated by Zurara, were never reached. At the same time, older topographical
characterizations of the western coast were enriched by a new multiplicity
of villages, ports, ethnic groups, and commercial relations.
Based on this perspective of the history of Portugal in Africa, I decided to
examine some of the classic texts on Brazilian slavery and slave commerce
to both isolate their underlying visions of Portuguese expansion, and raise
points of contrast with the present work. I start, in chronological order,
with the positivist historian Francisco Adolfo de Varnhagen (1816–78). He
cited Guinea, the Mina Coast, and Kongo (here probably including An-
gola), but highlighted the events and importance of the Mina Coast. Incor-
rectly, he attributed to this region the ‘‘nations’’ of Nagô speakers,≥Ω who
were heavily represented in the shipments of slaves to Brazil in the first half
of the nineteenth century, during the years when Varnhagen was complet-
ing his education. Given his interests in politics and power, he gave greater
emphasis to the nation as the fundamental identifying unit of local (politi-
32 | Chapter One
cal) organization than to the categories of ‘‘heathens’’ or ‘‘pagans,’’ which

open into sociocultural and religious dimensions.∂≠
Nina Rodrigues (1862–1906) was a medical doctor and observer of the
so-called human races, a popular theme in the waning years of the nine-
teenth century. A pioneer in calling for the serious study of the provenience
of African slaves in Brazil, he cites Varnhagen as an example of the defi-
ciencies of extant historiography. At the same time, like the previous author
he linked provenience with nationality. Thus the ‘‘kingdom of Guinea,’’
which continued to mushroom through ‘‘Portuguese colonial grandiosity
. . . from the Senegal to the Orange Rivers’’∂∞ —that is, from modern Sene-
gal to South Africa—contained within its purportedly discrete boundaries
numerous places from which slaves were taken. But there was never a
kingdom of Guinea, either of African extraction or by Portuguese extension
(the only real Portuguese occupation was in Angola).
In a 1906 essay, Sílvio Romero (1851–1914)—philosopher, literary critic,
and scholar of African influences in Brazilian culture—indicted what he
viewed as bias in the historiography, as well as in Brazilian society:
No one has deigned to concentrate on the negroes, which is a most abomi-

nable expression of ingratitude. What was the ethnographic map of Africa
when Brazil began to import slaves from beyond the seas? And in the
eighteenth century, when this enterprise continued? And in the nineteenth,
until 1850, when this terrible trade overpassed the previous centuries? What
about the classification of races, and the political situation of African states?
What about the social organization of those people? From which tribes did
we take slaves, and in what number? What do we owe them, in economic,
social, and political terms?∂≤
Interestingly, despite the assertion of African cultural diversity in this

passage, Romero’s writings as a whole betray the assumption of the exis-
tence of a sort of blanket Bantu unity that contradicts his own occasional
claims (as well as those of Rodrigues).
The historian João Lúcio de Azevedo (1855–1933) drew heavily from the
chronicles of Zurara to describe a changing geography of color and belief
across Africa, emphasizing the rainbowlike diversity of the first generation
of captives: ‘‘bright whites to more or less tarnished; mixed-bloods to
negroes, the dark hue a measure of whence along the southward expedi-
tions they were taken. Islamic at the start, and by the end a gleaming black,
all barbarous idolaters.’’∂≥ This recalls Zurara’s reference to the progressive
darkening of the native peoples of Guinea.
The most important historian of the Atlantic commerce of slaves in
Brazil was Maurício Goulart (1908–83), a pioneer in attempting to estimate
the total numbers involved in the trade.∂∂ But he was less interested in
comprehending the diverse cultural and geographic proveniences of the
slaves in Brazil than he was in defending the notion of four historically
dominant trading routes across the ocean: from High Guinea, Mina Coast,
Angola, and Mozambique (or east coast). Of the sixteenth century he
stated, ‘‘In popular conception there was no Africa, there was Guinea. The
entire region from which negroes were ransomed, the West Coast of Af-
rica, from top to bottom was Guinea.’’∂∑ Still, because his ultimate goal was
an estimate of the numbers of slaves sent to Brazil, his distinctions were
few. He delineated the two main commercial origins, Mina Coast and
Angola, and suggested where the majority of slaves from each source was
sent—Mina slaves to Bahia and Angola slaves to Rio de Janeiro. Even more
recent authors have been satisfied to stay within this dualistic version of the
proveniences of Brazilian slaves, a perspective that was not born with
Goulart but that emerged in the sixteenth century with the way Portugal
established and conceptualized its African possessions and trade routes.
Period maps, such as that of Sebastião Lopes (1558), accentuate the princi-
pal features of Portuguese Africa: the Mina Castle and the Catholic Church
erected in Kongo.∂∏
Charles R. Boxer (1904–2000), an English historian with many substan-
tial contributions to the history of colonial Brazil, analyzed Portuguese
slave trading in Angola but largely ignored the Mina Coast.∂π To the degree
that he was interested in the Portuguese empire, his preference for Angola
is comprehensible, given the larger Portuguese presence there. But the
e√ect was to exclude the Mina Coast from his analysis of Portuguese
history. (His brief consideration of the Mina Coast focuses on the Dutch
presence.)
There are several reasons why the historiography has been built with
recurring gaps and shadows over the circumstances that created commer-
cial and cultural ties between Brazil and the Mina Coast in the sixteenth and
seventeenth centuries. When Frédéric Mauro wrote his highly influential
34 | Chapter One
book about the Portuguese Atlantic of the seventeenth century, little infor-
mation was accessible about either the Mina Coast or São Tomé. The
resulting imbalance, or di√erence in emphasis, perhaps contributed in its
own way to the historians of the period that followed in Mauro’s wake.∂∫
The historiography on the slave trade to Bahia is mostly dedicated to the
eighteenth century and later, an era for which documentation is compara-
tively abundant.∂Ω For Rio de Janeiro, the connection to Mina has been prac-
tically ignored. The ties between Rio and Angola in the seventeenth century
—embodied in the figure of Salvador de Sá, a theme expertly studied by
Boxer—have left the illusion of exclusivity and continuity that do not hold
for the eighteenth century, when (as shown in the tables of chapter 2), a
significant number of Africans were brought to the city from the Bight of
Benin.∑≠
It should be noted that the complex question of the cultural diversity of
African slaves has been of more interest to the social sciences generally
than to history proper. Following Rodrigues’s lead, Arthur Ramos was the
first to attempt an elaborate map of cultural origins as well as an evaluation
of academic production on the topic. The French sociologist Roger Bas-
tide, working within the cultural approach of Herskovits, produced impor-
tant work with a consistent focus on religion. Rodrigues, Ramos, and
Bastide are, each in his own manner and time, the principal names of what
we might call the Bahian school of Afro-Brazilian studies.
Arthur Ramos (1903–49) presented, in 1936, the first general classifica-
tion of the African peoples brought to Brazil. His work criticized the
partiality of Rodrigues (for the Sudanese Africans) and Romero (for the
Bantu); he chided Romero for not answering his own forceful questions on
the study of origins, and proposed responding to them himself. Deploying
a range of racial and cultural criteria, he reformulated the two basic groups
of Africans familiar to scholars—Sudanese and Bantu—into four new cul-
tural categories: Sudanese (Yoruba, Dahomean, Fanti-Ashanti, and oth-
ers); Guinean-Islamic Sudanese (Fula, Mandinga, Haussa, and others);
Angola-Kongo Bantu; and East African Bantu. In his map of the Brazilian
distribution of these groups, he situated the Bantu in Maranhão, Pernam-
buco, Alagoas, Minas Gerais, and Rio de Janeiro. Sudanese groups joined
the Bantu in Maranhão, but they were the dominant African population
in Bahia.∑∞
The anthropologist Melville Herskovits (1895–1963) led a group of

American researchers to Brazil, by way of Africa, to study the acculturation
of Africans and their descendants. Herskovits had studied at Columbia
University under Franz Boas, as had Gilberto Freyre; both these men were
deeply influenced by the teachings of the elder anthropologist. In Brazil,
Herskovits and his team observed among Afro-Brazilians what they called
survivals, or the surviving traits of early African cultures. Their conclusion
was that because the Fon and especially the Yoruba (from modern Benin
and Nigeria) had been among the last African peoples to be taken to Brazil
as slaves, between the late eighteenth century and mid-nineteenth, their
cultural memories were more intact and perceptible.
Herskovits’s project on the transfer of African peoples and cultures was
the largest and most widely discussed such study to date. It produced an
impressive documentation in the 1930s of the historical conditions sur-
rounding the transfer of African populations to Brazil, along with maps to
suggest the origins and destinations of these populations. Herskovits pub-
lished two books about Dahomey—Outline of Dahomean Religious Belief (1933)
and Dahomey (1938)∑≤ —whose influence Karl Polanyi explicitly acknowl-
edged in his own important historical study, Dahomey and the Slave Trade.∑≥
However, the fact remains that despite the interest Herskovits expressed in
Fon speakers in Africa, and their descendants in Brazil, his curiosity did
not inspire substantive studies (by him or his contemporaries) either of
slaves from that region, or of the ‘‘Jêje candomblés’’ of Bahia. Indeed,
anthropologists as well as historians in the first half of the twentieth
century concentrated on Yoruba speakers, neighbors of the Fon, and their
descendants in Brazil. If historians tended to focus on the history of slav-
ery, not of Africa, anthropologists often homed in on the religious and
cultural heritage of Yoruba derivation.
Such was the case with Roger Bastide (1898–1974), a French sociologist
who arrived in Bahia in 1940. He published extensively, but is perhaps best
known for the 1958 O Candomblé da Bahia (Rito Nagô) and the 1960 As Religiões
Africanas no Brasil, books in which he applied first an ethnographic and then
a historical-sociological perspective to the religiosity of slaves and their
descendants in Brazil. Bastide envisioned a continuum of religious culture
extending from Africa to Bahia. From there it went to Rio de Janeiro, where
it was broken down and assimilated into such syncretized Brazilian forms
36 | Chapter One
as ‘‘macumba.’’ Both books privilege the purity and authenticity of the

Nagô legacy—a predilection that is implicit in the first work, and explicit
(even insistent) in the second.∑∂
In Religiões Africanas, Bastide included a critical survey of the literature on
the Atlantic commerce of African slaves, and suggested that Arthur Ramos
had glimpsed the ‘‘definitive solution’’ to the problem of classifying Afri-
can cultures. He proposed one qualification to Ramos’s four cultural cate-
gories—subdividing the Angola-Kongo Bantu group into eastern and west-
ern—and then proceeded to a succinct century-by-century chronology of
how the Portuguese slave trade mapped Africa: sixteenth century, the ex-
ploration of Guinea (then including what would be known as Angola);
seventeenth century, the establishment of primary commercial relations
with Angola; eighteenth century, the reversion of the commercial base to
the Mina Coast, and the replacement of shipments of mostly Bantu by
‘‘Mina or Sudanese’’; nineteenth century, Angola’s distinction once again
as the center of the trade. Bastide made the following observation about the
designation of slave origins:
In the beginning all the African slaves sent to Brazil were called ‘‘Guinea
negroes,’’ but one should not read too much into this because in the six-
teenth century Guinea extended from Senegal to Orange. The so-called
‘‘Guineas’’ arriving in Brazil were most likely Bantus. After all, do not the
records of the Inquisition in Bahia often include references to the ‘‘Guinea
negro, son of the Angola race’’? However, it’s probable that during the first
years of colonization most slaves came from regions above the Equator.∑∑
Bastide did not su≈ciently explain why the slave trade of the sixteenth
century should be understood as mostly Bantu, and his recycling of stereo-
types and undefended assertions contribute little to the historical problems
at hand. If anything, the popularity and influence of his work have acted to
discourage research programs that might challenge such inherited views.
In sum, we can derive from the above authors three basic chronological
criteria that they applied in di√erent ways to the task of locating the sources
of Brazil’s slave trade across time. First, there is the perspective of stages of
Portuguese expansionism, starting with the approach to Guinea (as de-
scribed by Zurara) and the occupation of the Mina Coast, followed by the
drying up of Mina’s gold trade and Portugal’s growing relations with
Kongo and Angola. Although these authors all seemed to recognize that
the construction of the Mina Castle was contemporaneous with the con-
quest of the central-western coast, they wrote as though these were succes-
sive events and represent distinct historical epochs.
Second, there is the perspective of di√erentiating ethnic or provenience
groups by ‘‘nation.’’ Period documents use the terms gentios (heathens) and
nações (nations), although they are not interchangeable. Heathen from
Angola might also have been considered a nation, from the point of view of
Portuguese or Brazilian observers, but the reverse is not true, since a nation
was a broader social construct that could have contained both heathens
and converted individuals. And while there were ‘‘Guinea heathen’’ in
general, there never existed a Guinea nation. Guinea, a term of toponymy
from the fifteenth century, corresponded only to a heathen person of ge-
neric ethnicity from West Africa. Over a relatively short span of time the
term fell into disuse and was progressively substituted by nação (nation) in
the colonial documentation.
Finally, there is the periodization of the Portuguese historiography
that emerged with the Avis dynasty. This corresponds to the reigns of
Dom Afonso V (1438–81),∑∏ and of Dom João II (1477/81–1495), who were
principally responsible for impelling the explorations to the Mina Coast
and Kongo/Angola.∑π These two kingdoms were associated with di√erent
strategies and geographies of maritime expansion. If the mysteries of
Guinea were first penetrated under Dom Afonso V, the more targeted
e√orts of João II led to the construction of Mina Castle (1482–86), the first
forays up the Congo River (1482–85), and the occupation of Angola’s coast
(1486).
Whereas some of the authors discussed above used these criteria to
discuss the commerce of slaves, others applied them to the study of African
cultures and contact between cultures in the colonial setting. But it is
notable that none of them attempted to question what is a sort of received
consensus about the spatial distribution of African peoples. Certainly all
were concerned with the proveniences of the African slaves who were sent
to Brazil, but they all also tended to forget or ignore the histories of these
peoples and the contours of both short-term and long-term change—
before but especially during the eras of European contact. Slave tra≈cking,
whether for external sale or for the myriad internal political disputes be-
38 | Chapter One
tween African groups, was one force a√ecting the movements and distri-
bution of people; other primary factors included droughts and the spread
of Islam. But even beyond spatial distribution, these forces could have
impacted the social organization and cultural expression of these peoples.
We confront the possibility that while much of the west African coastal
region has been mapped by the historiography, the populations who lived
there may themselves remain quite unknown to us, as they changed and
regrouped and adapted throughout the nearly five centuries of the Atlantic
slave trade. Nina Rodrigues was keenly aware of the need to better compre-
hend Africa, and the lives of people there, throughout the period.
In the records of the trade in human flesh, customs receipts should have
indicated the provenience (procedência) of each tragic shipment, at least
with reference to the port in Africa where they were loaded aboard ship. But
this was rarely done, and when done, rarely with accuracy. The best one
could hope to see was an indication of the port, and this says nothing of the
hunting grounds from which the slaves were snatched, by war or by fraud;
nor of the central markets where the slaves were taken to be sold before
being herded to port. It is only with such information that one could discern,
in the cold statistics of black immigration, the provenience of the people
brought to Brazil in the slavers’ holds.∑∫
In the dialogue, or absence of a dialogue, between these authors we can

see the distance imposed by their di√erent theoretical starting points, but
also by the methods of analysis adopted by each one. For a long time,
historians of Brazil paid little attention to culture. The Brazilian historians
Francisco Adolfo Varnhagen and João Pandiá Calógeras reflected the pre-
occupations of their era in their focus on politics and diplomacy. Rodrigues
was in a certain sense the most modern and ambitious in his work, but he
was hindered by polarized period debates over the meaning and verifiabil-
ity of true knowledge. Avoiding the tendency of some historians to em-
brace oral histories of elderly Africans as a means of studying the past, he
went the opposite way, deferring to o≈cial documents (and he loudly
protested the destruction of the archives of the slave trade after abolition).
In distinct ways both he and Romero suggested provocative alternative
methods and sources to present-day historians—Rodrigues in his focus on
migrations, using oral reports of older Africans in Brazil, and Romero in
his deep immersion in the African influences in folklore and national

culture.
I cannot enter here into the various historiographical impasses over
race, ethnicity, and color in the general field of slavery in Brazil. My inten-
tion in this chapter has been to demonstrate that all of these important
authors grappled with one particular problematic—the relations between
Africa and Brazil—and the possibility of identifying continuities from one
place to another. Texts and maps of the fifteenth and sixteenth centuries
construct an idea of Guinea that was based on Portuguese conceptions of
African geography and African culture. They served as the basis for the
struggles of modern social scientists to comprehend the African presence
in Brazil. The use of African and Portuguese toponymies could be alter-
nated or combined, as with the formulaic social description reproduced by
Bastide, ‘‘Guinea negro, son of the Angola race.’’∑Ω
Several points should be emphasized before we begin to look more
closely at the slave trade in the Mina Coast. First, as should be clear by now,
the phrases in documents of the fifteenth through seventeenth centuries
that mention Guinea—such as ‘‘Guinea Coast’’ or gentio de Guiné (heathen of
Guinea)—must be understood within their particular historical contexts.
E√ective studies of the changing meanings and borders of ‘‘Guinea’’ have
yet to be undertaken. Second, the Mina Coast emerges by the end of the
fifteenth century as a distinct place within Guinea, characterized by the
construction of the Mina Castle, the development of a settlement there, and
the initiation of a cyclical and complex relation of global commerce and
power. Finally, I am not asserting that social scientists of the period who
have highlighted the importance for Brazil of Angola, while downplaying
that of the Mina Coast, are somehow simply wrong or benighted. Few
historians could argue with the general claims that Mina was less impor-
tant than Angola within Portuguese America; and that in the case of Rio de
Janeiro, Angola was the source of most slaves both to the captaincy and the
city. But by focusing on the connections between Mina and the city of Rio
de Janeiro, I hope to reformulate the Portuguese colonial world in a more
complex way than such general claims allow, while at the same time linking
that research to a previously untold story of how some individual Africans
remade their lives in the diaspora.
2 | commerce with the mina coast
The physical act of slavery—capture, dislocation to the markets,
\ commercialization, forced labor—directly impacted millions of

Americans, Africans, and Asians throughout the Portuguese
Empire from the fifteenth century through the nineteenth. In the higher
echelons of Portuguese society, the nature of those enslaved was a topic of
contentious debate not only among merchants, plantation owners, and tax
collectors, but also the Catholic Church. What were the slaves’ moral and
intellectual capacities, and what were owners’ responsibilities toward their
slaves? When and how should spiritual values be allowed to override finan-
cial concerns?
slave trading in the seventeenth century

On 30 July 1609, during the union of the Portuguese and Spanish Crowns
(1580–1640), Felipe III of Spain (Portugal’s Felipe II) stunned most Brazil-
ians—not least the slaveholding class—when he declared that all the indig-
enous slaves (called Indians or gentio da terra, ‘‘native heathen’’) there were
to be freed. According to the decree, they should be placed under the juris-
diction of the Jesuits, who would instruct them in the catechism and ‘‘civi-
lize’’ them (to prepare them to live as vassals of the Portuguese Crown).∞
Thus began a juridical dispute that would last centuries.
The law was greeted with loud protest in the captaincy of Rio de Janeiro,
where agricultural labor provided by converted indigenous people was in-
Commerce with the Mina Coast | 41
dispensable. Slave and free Indians toiled not only in mills and plantations;
they carried out a range of critical functions in the city of Rio de Janeiro,
from the household delivery of water to construction and public works.
Local clamor against the law was such that in 1611 another law was written
to permit enslavement of Indians, by ‘‘fair war’’ (sometimes called ‘‘just
war’’) for a period of ten years—a term that was rarely observed in practice.≤
The second law incited a rush across south-central Brazil: expeditions set
out from Rio de Janeiro to capture Indians in the Paraíba Valley, while
explorers and Indian hunters departing from São Paulo made it as far as the
Paraguay basin.
At this time, Rio de Janeiro had around 3,850 inhabitants, who were
then referred to as ‘‘souls’’: roughly 750 Portuguese, 100 Africans, and
3,000 Indians and mestiços (mixed-bloods).≥ The African demographic was
increasing, but historians have neither a clear sense of its patterns early in
the seventeenth century, nor of its relations to the progressive decline of the
indigenous population from the captaincy of Rio de Janeiro as a whole. The
smallpox epidemic of 1613 that decimated most of the indigenous in the
region might have been a factor encouraging the importation of Africans.
Three clues about the presence of Africans in the captaincy in the following
years suggest that the profile of the local slave population was undergoing
a transformation soon after 1613. First, in his massive account of the
history of the captaincy, Monsenhor Pizarro mentioned an early contract
for the importation of slaves from the Mina Coast. He did not provide a
date for the contract, but did state that the initial request had been ap-
proved by o≈cial permit in 1615.∂ Second, in 1618, Governor Rui Vaz Pinto
gave his brother Duarte Vaz the privilege of using African slaves to load and
unload ships at port.∑ Third, a local 1625 law, eloquent in its implications
for both the size and the restiveness of the African population, established
a reward for the capture of fugitive black slaves.∏
The present challenge is to interrogate available sources to assess the
proportions of slaves from di√erent African proveniences imported into
the city of Rio de Janeiro. We know that in this period there were three
principal maritime routes: from Kongo/Angola, often stopping in São
Tomé; from Mina; and from so-called High Guinea (where the trade was
centered along the Cacheu River in modern Guinea-Bissau), with a regular
stop in Cape Verde.π Unfortunately, data indicating the numbers of slaves
42 | Chapter Two
traded along these routes in the period are not reliable. And specific data
for how many slaves went through which routes to the captaincy of Rio de
Janeiro seem to be nonexistent. There were few local surveys or demo-
graphic studies, and no population censuses, all of which makes compre-
hending the general population of the city of Rio—much less its assem-
blage of African slaves—an unlikely feat. The only claim that can be made
with relative certainty is that by the second decade of the seventeenth
century, a population of African slaves—most of whom were probably
taken from Kongo/Angola∫ —were working alongside indigenous slaves
around the Guanabara Bay, both within Rio de Janeiro proper and through-
out the city’s outlying fields and hollows.
In 1632, one Father Dr. Lourenço de Mendonça arrived in Rio de Janeiro
to serve as the prelate of Sé. He quickly aligned himself with the Jesuits of
the local mission in their stance against the enslavement of Indians. In
swift response came the predictable protests and hectoring of the prelate,
but the force of local antipathy to Mendonça’s position was expressed
when his house, liberally dusted with gunpowder from a barrel carried
boldly up to his door, was set ablaze. Mendonça survived, making him
more fortunate than his predecessor Mateus da Costa Aborim, who was
killed by poisoning in 1629 for, it was said, defending the cause of indige-
nous emancipation.Ω Such events suggest the determination of Rio’s resi-
dents, particularly its landholding classes, to protect their access to Indian
labor. These conflicts were occurring at the same time that the Portuguese
were struggling to maintain control of their factories at di√erent slave
ports in Africa, defending them from attacks (especially by the Dutch). The
future of Brazil’s labor force seemed increasingly insecure. Religious com-
mitments were constricting the possibility of relying on Indian slaves,
while clashes between European powers threatened to disrupt Brazil’s
acquisition of Africans.
The Dutch provoked the Portuguese with a series of a√ronts. They had
created the West India Company in 1621, and within a few years were
dealing in slaves for the lucrative Caribbean market. In 1625 they attacked
Bahia. In 1637, they took possession of the Mina Castle on the Mina Coast.
In 1639 they occupied Pernambuco, and in 1641 Angola too went under
Dutch control. Simultaneously, piracy corroded Portuguese authority and
profits in the various Atlantic routes between Lisbon, the West African
coast, Rio de Janeiro, and Bahia. The French, for their part, created the
Compagnie des Îles d’Amérique in 1635 and began the occupation of
Guadalupe. In 1638, they founded the city of Saint Louis in modern Sene-
gal, where they engaged in the profitable export of gum arabic (a water-
soluble binder derived from certain species of the acacia tree). Stretches of
land in the French-held island of Martinique were converted to sugar cane
fields in 1654, after seedlings were spirited there by Dutch Jews angered at
their expulsion from Brazil; ten years later, using profits from its sugar
exports, France formed the Compagnie des Indes Occidentales. The entire
Portuguese imperial project seemed at risk.
In the midst of these tensions, Portugal confronted Spain to restore the
Portuguese Crown (1640) and attempted to reassert its relevance in the
Atlantic trade between Africa and the Americas. The Dutch occupation of
Luanda a year later was both a symbolic and strategic blow to the Portu-
guese Empire, since Luanda was the principal supplier of slaves to Brazil.∞≠
The captaincy of Rio de Janeiro was a√ected by the lack not only of slaves,
but of other merchandise that was carried along the human-trading routes.
The increasing disarray that gripped Portugal’s Atlantic commerce is re-
flected in the definitive insertion of the Dutch into the slave trade to the
Americas—and also in the establishment of a competing sugar market in
the Dutch Antilles, which served England and France as well as Holland,
with plants taken from Portuguese holdings. Brazilian planters were hurt
twice over, as Dutch sugar forced them to lower prices on their own exports
while the demand for slave labor to cultivate and process the sugar in the
Antilles raised the costs of importing slaves.∞∞ To centralize its e√orts to
regain control of its Atlantic possessions, Portugal created the Conselho
Ultramarino (Oversea Council) between 1642 and 1643. One of the coun-
cil’s first recommendations was that trade between Portugal and Brazil
must no longer be carried out in a piecemeal fashion, with small groups of
ships that were easily attacked. The mercantile ships should be organized
into fleets, protected by accompanying armed galleons.
One person closely following these developments was Salvador Correia
de Sá, a landowner with possessions in Brazil (farther south, near the Prata
River) and in Angola. A scion of Estácio de Sá, founder of the city of Rio de
Janeiro, Salvador de Sá and his family controlled great swathes of land in
the captaincy of Rio de Janeiro, and they dominated local politics in Rio. He
44 | Chapter Two
served as governor of the captaincy several times, starting in 1637. In 1643,

he wrote to the king in response to a royal inquiry about the best means to
open trade relations with the Prata region, especially Buenos Aires. He
observed that in the face of the Dutch occupation of Portuguese Angola, it
would be di≈cult to trade with the Spanish in Buenos Aires, since the
merchandise most sought after there was African slaves. It would therefore
be better to retake Luanda first. He urged action, reminding the king both
of the need for the meat and leather that the Spanish o√ered, and of the fact
that a Portuguese presence along the Prata River would open access for
Brazil to the silver mines at Potosí.∞≤
Although there was a clear intention by the crown to make all trade more
organized, e≈cient, and safe, in practice the powerful merchants linked to
trading monopolies had more sway than individual businessmen, who
either had to appeal their much smaller contracts and licenses to royal au-
thorities, or who sought other ways to get what they needed more cheaply.
An example of Salvador de Sá’s influence in this regard can be seen in his
response to local protest over the new law prohibiting trading ships to
operate outside of the o≈cial fleets. These naval convoys sailed only twice a
year between Lisbon and Brazil. To attempt an Atlantic crossing without the
protection they o√ered would leave ships isolated and vulnerable to pirates,
but Rio’s businessmen evidently preferred this risk to what they saw as too
egregious a constraint on their abilities to operate. In 1645 Salvador de Sá,
then the captain-major of the captaincy of Rio de Janeiro,∞≥ even though he
was residing in Lisbon, went to Rio to confront the protest. Soon after
gathering in assembly to debate the law with him, local merchants declared
their unified opposition and revolted against the law.∞∂ Salvador de Sá
negotiated a few small concessions but sided with the crown in maintain-
ing the overall structure of the law; by a curious coincidence, he was quickly
named the fleets’ commander, which included being given all the rights
and privileges of a military general.∞∑ Dumbstruck, residents of Rio de
Janeiro carried accusations of corruption against him to the Finance Coun-
cil in Lisbon, which was persuaded to call for the opening of a formal
inquiry.∞∏ Resisting the accusations, Salvador de Sá was ultimately nomi-
nated governor of the captaincy of Rio de Janeiro. All of this suggests that
the crown and the powerful shipping monopolies were at least as con-
cerned about clamping down on what may have been a thriving informal
system of Atlantic commerce as they were about protecting Portuguese

trade in general from pirates and international competitors.
Rather than staying in Rio de Janeiro, Salvador de Sá raised funds and
organized a squadron to battle the Dutch in Angola. He helped retake the
colony in 1648, and stayed on there as governor. In 1649, he, the renowned
Jesuit priest Antônio Vieira, and other notables raised capital from private
investors (especially from the Portuguese New Christian network) and
created the General Company of the State of Brazil. Their first fleet was sent
to Brazil in 1650, a vivid demonstration of the victory of large-business
interests, including their representatives in Rio de Janeiro, over less power-
ful commercial segments in the city. A private company, with few tangible
connections to the crown or royal commerce, the General Company’s fleets
traversed the Atlantic on their own schedules, circulating merchandise
(including slaves) throughout the Portuguese colonies.∞π
Mina Coast in the seventeenth century experienced a growing demand
for slaves for the expanding American market, and the solidification of
commercial networks increasingly linked it to the Atlantic world. Early in
the century the kingdom of Hueda, in modern Benin, encouraged European
traders to focus slave-hunting e√orts in the Bight of Benin. Soon all of the
di√erent European nations that were present on the Mina Coast had estab-
lished alliances with groups of local people to facilitate both access to
slaves and the occupation of coastal territory. In this era many ‘‘castles’’
resembling Portugal’s São Jorge da Mina were built on the littoral of mod-
ern Ghana to serve a similar dual function: defense (as a fort) and com-
merce (as a factory). In 1665 the English claimed Cape Corso, where a
century earlier the Portuguese had constructed a modest fort. They reno-
vated that old building into what came to be called Cape Coast Castle, and it
served as the base of operations for expanding English interests in the area.
Christiansborg Castle, constructed by the Danes in 1679, was purchased by
the Portuguese and renamed São Francisco Xavier. A contingent from Bran-
denburg-Prussia also built a fort, Friedrichsburg, in 1683. Some less im-
posing structures were also erected along the coast to facilitate and defend
slave trading, such as the French warehouse placed in Ouidah in 1671.
Three stretches of the Mina Coast became especially important in the
trade: the Gold Coast, where Mina Castle sits and where the scale of trade
began to mount after 1660; the kingdom of Allada, east of Mina Castle,
46 | Chapter Two
incorporated into trading patterns after 1670; and the kingdom of Hueda,
whose people were significant participants in augmenting the trade by the
end of the seventeenth century. By the early eighteenth century, they had
lured away many of the European trading partners of the kingdom of
Allada.∞∫
The historian Karl Polanyi has described the Gold Coast as a long, rain-
drenched strip of sand separating the mountains from the sea. The region
is intercut with lagoons, where in the sixteenth century the Portuguese
encountered communities of fishing villages with little apparent political
centralization. The origins of Portugal’s Gold Coast slave trade were here,
with most of the early negotiations involving only a few slaves (even one or
two) o√ered by the natives directly to the Portuguese; the gold trade began
under similar circumstances.∞Ω But it was precisely this loose form of social
organization, poorly observed among fishermen in the wetlands, that the
Europeans would assume were kingdoms, with discrete boundaries and
authoritative central governments. The political system that the Portuguese
projected onto local communities was one highly conducive to the expan-
sion of slave commerce. Thus began a complex relationship between the
emergence in the region of territorial states with centralized power, the
voracious demand for slaves in the Americas, and the infrastructure and
networks that developed around the business of slave commerce through
ports throughout West Africa.
This was the context surrounding the first shipments of slaves to Brazil
from the Bight of Benin around 1670. In the same period, in 1675, a group
of merchants created the Cacheu Company to coordinate Portuguese com-
mercial relations in High Guinea. But the Dutch intervention had been
decisive, and Holland, occupying Mina Castle, soon became the dominant
player in West Africa’s European slave trade.≤≠ An insightful portrait of the
era comes from William Bosman, who relocated to the area late in the
seventeenth century as subcommander of the Dutch West India Company.
In 1705, Bosman published a book called Voyage to Guinea, which includes
numerous stories and asides about the West India Company as well as the
Royal African Company (founded 1660 in England) and the Company of
Senegal (founded 1673 in France). He believed that these three European
companies, each of which from the beginning had scrambled to outflank
the others, finally succeeded in destroying each other. Still, the abundant
riches produced by trading in African gold and slaves meant that Holland,
England, and France all came out well ahead in the end. Bosman also
observed that the most desirable slaves from the Mina Coast region were
under the age of thirty-five; that the price of women was 20 percent less
than that of men; and that Dutch ships usually carried between 600 and 700
captives per trip.≤∞
In the eighteenth century, the most formidable kingdom of the Gold
Coast was that of the Ashanti. Located in the interior, they skirmished
incessantly with their southern coastal neighbors, the Akim, who sat on
the source of the gold considered by European merchants to be the finest in
the region.≤≤ As the Ashanti battled to enlarge their territory, and the Akim
to defend theirs, English merchants were fortuitously situated to benefit
from the additional flows of gold and slaves resulting from the strife. This
was contemporaneous with the bloody but profitable struggles raging west
of the Gold Coast, where the kingdom of Dahomey was fighting its way to
the littoral, killing or displacing people of the Allada and Hueda kingdoms
to control the maritime commerce of the Slave Coast. Directly or indirectly,
both the Ashanti and Dahomey kingdoms were fighting over not only ports
but the myriad land routes used to march slaves through the interior to
waiting European pens and ships.
The Portuguese presence on the Mina Coast, limited after the mid-
1600s, was even further diminished by these developments. A booming
Atlantic demand for African slaves, combined with the power consolida-
tion of several dominant African groups on the coast, led to a redefinition
of relations between Europeans and West Africans to which the Portuguese
were mostly a distant witness. They were losing their grip not only within
the European commercial circuit, but also on the trade to Brazil. There,
local merchants and investors, principally from Bahia, were underwriting
their own slave-buying trips to the Bight of Benin to help meet a demand
that the ports of Kongo and Angola seemed increasingly unable to satisfy.
Portugal, unable to regulate or even e√ectively participate in the commerce
of slaves across the Atlantic, while it was still sharply restricting access to
indigenous labor in Brazil, helped usher in a transformation of labor
relations in Portuguese America. Now, African slavery, obtained by local
means if necessary, was looking like the definitive solution to the labor
shortage a√licting some regions of Brazil—among them the captaincy of
Rio de Janeiro.
In the city of Rio, disputes between Jesuits and residents over whether
48 | Chapter Two
Brazil’s indigenous should be subject to forced labor had continued

throughout the seventeenth century. In 1680 the Prince Regent and future
Dom Pedro II of Portugal decreed once again the liberty of Indians, provok-
ing yet more violent local protests. That law was modified in 1684 to allow
for Indians to be maintained in relationships of private ‘‘administration,’’
as long as they were instructed in the Catholic faith and incorporated into
the church. Meanwhile, the increasingly urgent calls for importing more
African slaves to the captaincy were silenced by a virulent epidemic in
Angola that temporarily choked the trade, to Bahia as well as Rio de
Janeiro. It was in this context that, in 1685, approval was granted by the
governor of Bahia to a proposal, submitted by local merchants, to send a
ship to the Mina Coast on a slave-buying trip. The initial petition made
reference to the ‘‘news that comes from Angola that smallpox has so direly
a√licted it, one may fear that in many years the loss of negroes dead of the
disease will not be recouped.’’ Smallpox wrought devastation in Angola
during the 1680s, and it is not surprising that the epidemic should be
adduced in the arguments for turning to the Mina Coast, especially the
Bight of Benin. It is notable that the Bahian slave traders sought authorized
access to the ports of Mina. At any rate, their preferred mode of convey-
ance, the patacho—a nimble but small vessel with only two masts—could
carry but a miniscule proportion of the human cargo handled by the large
European shipping companies of the day.≤≥
rio de janeiro’s slave trade

in the eighteenth century
In his analysis of the evolution of Rio de Janeiro’s commercial activity from
the eighteenth century to the nineteenth, the historian Manolo Florentino
shows how the centrality of local traders to the city’s development, in
its myriad aspects, often led to conflicts of interest with the Portuguese
Crown.≤∂ This was also the case in the seventeenth century. Even if the
documents for the earlier period are less abundant, they are quite expres-
sive of such tensions and indicate how they could persist and multiply with
the passage of time.
Late in the seventeenth century, a call went out to send slaves to the new
mining region north of Rio de Janeiro proper, a place increasingly referred
to as ‘‘Minas’’ (referring to the gold mine lands) and later ‘‘Minas Gerais’’
(General Mines), where gold had been discovered. In 1699, the crown
legalized slave trading from the Mina Coast to Brazil, creating what quickly
became a busy and prosperous circuit of gold and slaves flowing between
Minas (the mines in Brazil’s central interior) and Mina (the Bight of Benin
in West Africa). Pernambuco, Bahia, and Rio also partook in the trade,
anxious for access to slave labor. In the immediate rush, however, proper
taxes—principally the quinto (‘‘fifth’’)—were not always paid to Portugal.
This allegation was the premise of a letter written by D. Rodrigo Costa,
governor of Bahia (1702–5), to the crown on 20 June 1703. In a rambling
diatribe, Costa accused the traders in Rio of having maximized their per-
sonal gain in recent years by not only avoiding or underpaying the taxes,
but by using the gold carted to its ports from Minas Gerais to buy slaves
from other European brokers instead of from Portugal’s appointed agents
in Benin.≤∑ Whether Costa’s motivations stemmed from personal loyalty to
the crown or indignation at what he saw as unfair competition remains
unclear.
However, the response was prompt. That same year, 1703, King Pedro II
prohibited the departure of slavers from Rio de Janeiro for the Mina Coast,
and instituted quotas for the allocation of Mina slaves between Rio, Bahia,
Pernambuco, and the gold mines of Minas, with the additional stipulation
that all slaves should be imported through Bahia. This measure was in-
tended not so much to limit the importation of slaves as to more closely
regulate the movement of Brazilian gold, with the objective of keeping it
out of the hands of foreign establishments plying the Mina Coast trade.
Bahia’s primacy in the region did not last too long. The 1703 law was never
fully complied with, and it was abolished in 1715 in the face of a rapid
expansion of mining in the area and a concomitant need for more slaves.≤∏
It is unclear how many slaves the traders in Rio de Janeiro helped bring to
Minas Gerais between 1703 and 1715, and targeted research is urgently
needed to illuminate this poorly known chapter of Brazil’s commerce with
the Mina Coast.
Meanwhile, Rio’s merchants were locking horns with the colonial ad-
ministration over what they argued was prejudicial treatment. In 1704,
o≈cials from the city council sent a representative to the Oversea Council
in Portugal to denounce alleged abuses of power by then governor, D.
50 | Chapter Two
Álvaro da Silveira e Albuquerque (1702–4). According to the letter drawn

up by the council, which included some of Rio’s most important local
businessmen,
The o≈cials of the City Council of Rio de Janeiro inform His Majesty that
when several vessels arrived into that port carrying shipments of slaves from
Mina Coast and São Tomé, Governor D. Álvaro da Silveira obliged the ships
to remove themselves to a deserted island to be examined for contagious
diseases. He did not permit the ships to disembark in the city until each one
had provided him with the best negro specimen on board, all of which slaves
being then informed they were property of the Governor, and commanded to
hoist him on his palanquim [sedan] and carry him to and fro.≤π Beyond this
o√ense he demanded the right for himself and his family to buy the best of
the remaining slaves for a trifle, and they then sold these slaves to the people
of the city at a profit, each person making at least 40,000 or 50,000 réis.
When they do this again, with the ships coming from Angola, we will deliver
another letter of complaint to His Majesty.≤∫
This curious manner of advertising a future complaint about the same

o√enses should not be taken to indicate that traders maintained exclusive
relationships with the di√erent sources of slaves—that is, that some mer-
chants in Rio de Janeiro dealt only with Mina slaves, and others only with
Angola slaves, so that a distinct set of merchants would be hurt by the
governor’s actions in each case. Rather, the letter indicates the formation
of complex political alliances and interest groups in the city, for whom a
campaign of letter-writing to the Oversea Council represented a strategic
move. Unfortunately, I could not discover what the council made of this
communication. But, coincidence or no, Álvaro da Silveira was replaced as
governor of the captaincy that very same year, with D. Fernando Martins
Mascarenhas taking his place.
In 1706, D. Rodrigo da Costa, the ex-governor of Bahia, was still preoc-
cupied with Rio de Janeiro and the slave trade even as he headed back to
Lisbon. In a letter he wrote to the king ostensibly about Brazil’s gold
mines, he quickly returned to the themes of commerce with the Mina Coast
and Rio’s evasion of the royal fifth. In this context he also alerted the king
to the apparent fact that slaves in the mining region were worth two to
three times more than they were in the plantations. He went so far as to
propose the outright prohibition of any shipment of slaves to Rio de

Janeiro, be they ‘‘from Angola or the Mina Coast, Bahia, Pernambuco,
Paraíba, the Islands, Lisbon, Mozambique, India, or any other place in or
out of the royal dominions of His Majesty.’’≤Ω The ex-governor suggested
mandating the trading of slaves through the Cacheu Company, a Portu-
guese concern boasting a substantial naval fleet that was already active in
the trade. (Slaves arriving at Bahia usually had departed from the Bight of
Benin, but at least some of those in Pernambuco and Paraíba came from the
region called High Guinea, and appeared in the documentation in Rio de
Janeiro as cacheos or cacheus.) Again, the idea was less to limit the absolute
numbers of imported slaves than to channel the trade into the hands of the
so-called Junta de Cacheu (Cacheu Council), an assortment of rich mer-
chants, politicians, and insiders who owned the company. In fact, as an
investor in the company, D. Rodrigo had a personal interest in seeing the
Cacheu Council gain exclusive slave-trading rights. He was exasperated at
all the business lost to local traders and smugglers in Brazil,≥≠ especially in
Rio, although in his missives to the king he tended to dwell on the argu-
ment of the unpaid taxes. It remains unclear the extent to which local
traders may have participated in larger commercial and financial networks,
including Portugal and even other European countries.≥∞ Apparently swayed
by D. Rodrigo’s insistence, the Oversea Council threatened, in a 1706
document, to denounce ‘‘in public or in private’’ all plantation owners in
Brazil who sold their slaves to the mines for an untaxed profit.≥≤ A reading
of various related documentation indicates that the ministers of the Over-
sea Council were increasingly amenable to the idea of reinstating the
‘‘exclusive system,’’ in which large Portuguese companies would dominate
the tra≈cking of slaves to Brazil and local autonomy to trade externally or
within the colony would be diminished.
By the first years of the eighteenth century, there existed two routes
linking Minas Gerais with Atlantic trade. One went south-southwest from
Salvador, crossing the interior scrublands, following when possible the
course of several rivers until arriving at the villages of Minas, especially Vila
Rica (later named Ouro Preto) and Vila do Carmo (later Mariana), the
location of the principal gold mines of the time—often called the Caminho
do Sertão, a hinterland route described by Antonio Andreoni. He was a Jesuit
who published under the name André João Antonil while living in Bahia in
52 | Chapter Two
the late seventeenth century and early eighteenth. The second route ex-
tended north from Rio de Janeiro by sea until Parati, then across the
Mantiqueira mountain range to the same villages in Minas. It has received
comparatively less study, although it is makes an appearance in the narra-
tive of the travels of the Count of Assumar, who arrived as the governor of
Minas in 1717.≥≥
The trading of slaves within Brazil was not as pervasive as its critics
intimated, although there were internal commercial routes based on the
rough trails linking Rio de Janeiro, Minas Gerais, and Bahia. These ter-
restrial routes were also the final stage of an Atlantic journey bringing
slaves from Africa to the mines at Minas. In the first half of the eighteenth
century, according to Maurício Goulart, some 2,000 captives a year were
marched from Salvador to Minas down the Interior Trail. Between 1728 and
1748, 40 percent of the slaves arriving in Salvador were destined for the
mines. The same author has demonstrated that starting in 1715, somewhat
more than 2,200 slaves a year passed through the port at Rio de Janeiro
from the Mina Coast on their way to Minas. Between 1725 and 1727, around
5,700 captives a year arrived in Rio from the Mina Coast and Cape Verde; of
these, 2,300 were dispatched to Minas. The rest were divided in various
ways between the captaincy of Rio de Janeiro and the cities of Rio and São
Paulo.≥∂
There are no systematic studies of the Parati route that linked Rio de
Janeiro and Minas Gerais. Historians seem to have dismissed its impor-
tance in the larger networks of trade connecting Brazil to the Atlantic
world. This neglect echoes the longstanding lack of interest in the pres-
ence of Mina slaves in Rio de Janeiro. That the two phenomena are related
in complex ways encouraged me to explore the available primary sources
more widely, and apply the insight or evidence gained about one of the
historical problems back to the other, thereby keeping them in dialogue.
This is where baptism records helped fill in some of the gaps left in the data
on Rio’s slave imports and the city’s retention of a small percentage of
those sent on to Minas. Where it exists, baptism data tend to be relatively
reliable because of the nature of the ritual and the singular importance it
held for contemporaries. Of course, when adult African slaves were bap-
tized, broader data appealing to subsequent researchers (such as the pro-
portion of baptized versus nonbaptized slaves in the parish or city, or the
identity of the company or merchant that brought a given slave to the city)
were hardly ever included in the record books. Nonetheless, these records
constituted a critical source of information for my inquiries.
In the first half of the eighteenth century, the city of Rio de Janeiro
contained two urban parishes: Sé and Candelária. The oldest record book
of slave baptisms available to me at the time of research was the 7\ Livro de
Batismo de Pretos Cativos da Freguesia da Sé (Seventh Book of Baptisms of Black
Captives in Sé Parish), covering the years from 1718 to 1726. Data from the
earlier period (1702–17), when slaves began arriving in Rio in substantial
numbers, would have been extremely helpful. Even still, suggestive trends
emerged from these records that could be correlated elsewhere. Over the
period covered by the records, significantly more baptisms were per-
formed for adult slaves brought from the Mina Coast than from any other
provenience (even though it is well established that more slaves were
arriving from Angola than from anywhere else),≥∑ and these numbers were
especially concentrated in the years 1722 and 1724.
Those were years of intense conflict along the Slave Coast. The kingdom
of Dahomey invaded the kingdom of Allada in 1724, and the kingdom of
Hueda in 1727, and took control of the commerce of slaves along a vast
expanse of prized coastline. According to Polanyi, at this time French and
English traders operated individually or in small associations up and down
the coast, while the Portuguese and the Dutch—accustomed to the ‘‘exclu-
sive system’’ of large organized companies—tended to negotiate their car-
goes in the Bight of Biafra (in modern Nigeria) or in Kongo.≥∏ This ap-
proach might explain why the importation of slaves from the Bight of
Benin to Rio de Janeiro went into sharp decline after 1725, although it had
been falling gradually for several years. The instability of full-out war in the
region, combined with Portugal’s current preference for other sources of
slaves, led to a slowdown in the local trade to Brazil. By late 1730 or early
1731, the kingdom of Dahomey had secured the coast and had reversed its
battle maneuvers to push northward, into the hinterlands of the Bight of
Benin and deeper in the interior, where it lay siege against the kingdom of
Oyo. This was also where the Mahi territory lay, and the region soon
became, as the historian Akinjogbin observed, a ‘‘hunting ground’’ for the
Dahomean incursions to gather prisoners, slaves, or both.≥π
Pierre Verger analyzed the inventories kept by Bahian slave owners be-
54 | Chapter Two
Table 2. Baptisms of Adult Africans in Sé Parish, 1718–1726
year provenience total

Mina Guinea Central Cabo Cacheo East
West Verde Coast
# % # % # % # % # % # % # %
1718 57 7.2 1 0.1 2 0.3 1 0.1 0 0.0 0 0.0 61 7.7
1719 64 8.1 5 0.6 7 0.9 1 0.1 0 0.0 0 0.0 77 9.7
1720 50 6.3 5 0.6 5 0.6 1 0.1 0 0.0 6 0.8 67 8.5
1721 95 12.0 13 1.6 2 0.3 0 0.0 1 0.1 11 1.4 122 15.4
1722 107 13.5 5 0.6 0 0.0 0 0.0 1 0.1 2 0.3 115 14.6
1723 73 9.2 11 1.4 0 0.0 0 0.0 0 0.0 0 0.0 84 10.6
1724 125 15.8 19 2.4 0 0.0 0 0.0 0 0.0 0 0.0 144 18.2
1725 79 10.0 9 1.1 0 0.0 0 0.0 0 0.0 0 0.0 88 11.1
1726 31 3.9 1 0.1 0 0.0 0 0.0 0 0.0 0 0.0 32 4.1
Total a 681 86.2 69 8.7 16 2.0 3 0.3 2 0.2 19 2.4 790 100.0
Source: acmrj, Livro de Batismo de Escravos, Freguesia da Sé, 1718–26. Calculations were made
based on the proveniences declared in the record; these designations are analyzed in chapter 3.
The column ‘‘Central West Coast’’ corresponds to slaves imported from that region, including
such designations as Kongo, Luanda, Angola, Benguela, and Quissamã. The column ‘‘East
Coast’’ includes slaves imported from Mozambique and the island of São Lourenço (modern
Madagascar). As noted in the text, the higher percentage of Mina baptisms is explained by the
Portuguese practice of baptizing slaves in Angola before taking them from Africa to Brazil.
a
Rounding of decimals accounts for discrepancies in some of the totals.
tween 1737 and 1841. Out of a total of 187 Africans clearly originating from
Dahomey, Verger identified 180 ‘‘Jêjes,’’ three ‘‘mondobi,’’ one ‘‘ladá,’’ two
‘‘maquim (maí),’’ and one ‘‘savanu.’’ He isolated thirty-one other slaves
identified only as ‘‘minas,’’ and one as ‘‘guiné,’’ denominations which he
considered vague enough to refer to either ‘‘Dahomeans’’ or ‘‘Yorubas.’’≥∫
Somewhat earlier, in Rio de Janeiro, the growing number of adult slave
baptisms in the 1720s might be a function of the demand for slaves in Minas
Gerais, or also of the local demand in Rio. The city was growing as its
trading functions expanded, which occurred in large part due to the discov-
eries at Minas. Table 2 presents the Africans baptized in Sé Parish between
1718 and 1726, which gives an idea of the distribution of proveniences of
slave imports to Rio de Janeiro in the period. The great variation from one
year to another prevents us from drawing associations between import

levels and distributions and the wars in the Bight of Benin.
While Portuguese companies were already struggling to maintain their
share of the slave trade in the 1720s—in an Atlantic Ocean crowded with
Dutch, English, and French ships—internal disputes between merchants
in Portugal and in Brazil were beginning to heat up. In the twenty years
between 1720 and 1740, traders based in Lisbon and in Bahia engaged in
raucous conflict over control of the tra≈c to Brazil. In 1721, the Portuguese
constructed a fort, São João Batista de Ajudá, in Ouidah (it was more of a
warehouse, although it was often referred to as a fort). The fort went under
the dominion of Dahomey in 1727, when the Dahomeans conquered the
kingdom of Hueda containing the fort, and its occupation marked a new
phase in the commercial relationships between African kingdoms and
Portuguese and Brazilian traders. The Portuguese had long allied them-
selves with the kingdom of Allada, now defunct, and su√ered stormy
relations with a newly powerful kingdom of Dahomey. Nonetheless, the
fort (called Ajudá, Judá, Uidá, Ouidah, or Whidah, depending on the docu-
ment) remained the principal Portuguese commercial structure on the
Mina Coast until around 1770.≥Ω
Most current analyses indicate that the wars waged by the Dahomeans to
control sources of gold and slaves, as well as territorial routes and coastal
access, resulted in mass imprisonments of men and women who were ulti-
mately sold as slaves on the international market. However, table 2 shows a
reduction in the appearance of slaves from the Mina Coast in Rio de Janeiro
after 1724. Turning once again to Polanyi, we get a bit more perspective on
this drop-o√. He suggested that the wars’ turbulence directly impacted the
capacity of local traders to successfully negotiate with nervous European
traders, even if the absolute number of slaves available for sale might have
been increasing. And one should not take the latter point as a given. Many
of the prisoners would have been disposed of by other means, either through
ritual sacrifice or through local slavery on the fields of the Dahomey King-
dom. Before the attacks bore down on the realm of Allada, that kingdom was
probably responsible for the sale of around 20,000 slaves a year through
Ouidah. Polanyi doubted that the subsequent commerce run by the Daho-
means reached similar levels.∂≠ Those are all reasonable assumptions. But
we are left with the puzzling coincidence of the baptism of no fewer than
56 | Chapter Two
144 Mina slaves in Sé Parish in Rio de Janeiro in 1724, one year after
Dahomey laid siege to Allada. Since the numbers decreased to 79 in 1725,
and to 31 in 1726, were these slaves the bitter fruit of that war, or were they
the last harvest before the wars began?
I could find no contracts, receipts, or other specific primary sources
relative to the buying, selling, or internal redistribution of slaves before
1725, but the documents from that year provide an outline of how the
transactions were undertaken. In 1725, Jerônimo Lobo Guimarães was
granted a new contract, which he managed to extend for three years, to
administer the apportionment of slaves entering Rio de Janeiro from Mina
Coast and Cape Verde.∂∞ In another three-year contract, dated a month
later, the Oversea Council instructed Guimarães to send most of these
slaves on to Minas Gerais.∂≤ Guimarães cropped up in an earlier document,
from 1723, petitioning for multiple contracts to distribute slaves from
ports at Bahia, Pernambuco, Paraíba, and Rio de Janeiro to the gold mines
at Minas.∂≥
The various correspondence pertaining to Guimarães reveals that his
achievement of those two contracts in 1725 was met with skepticism and
dissension among other merchants, and even among members of the Over-
sea Council itself. That very year, doubts were raised about the contractor in
the council of the Treasury in Lisbon,∂∂ and provoked the governor of Rio de
Janeiro to convene discussions with the local representatives of the Portu-
guese Treasury.∂∑ For his part, Guimarães sent five petitions defending his
honor and demanding the observance of his contract.∂∏ He was allowed to
appear before the commission in Rio, although the denunciations contin-
ued, many coming from disgruntled contenders for the choice contract
Guimarães had won.∂π In October of 1725, an irate Guimarães succeeded
in lodging a formal complaint against the Treasury for the damages he
claimed the public mudslinging had caused him; soon after, his contract
was guaranteed until 1727.∂∫ We can only imagine the reaction of the Over-
sea Council in 1726, when one André da Costa Faria petitioned the council
to be recognized as Guimarães’s personally nominated representative in
Rio de Janeiro, and thus as the e√ective administrator of the contract which
the embattled Guimarães had clung to so tenaciously.∂Ω
This episode is eloquent in its implications for how the social webs of
contracts, influences, negotiations, and reputations encircled and compli-
cated the o≈cial slave trade from the Mina Coast to the city of Rio de
Janeiro and beyond. The scope and nature of all the transactions (legal or
otherwise) directly and indirectly associated with the commerce of slaves
remain much harder to tease out of the documents. After 1730, voyages
from Brazil to the Bight of Benin required the prior permission of the
viceroy. Whether Portugal’s intention was to limit local autonomy or more
pointedly to clamp down on smuggling, the clandestine trades (which had
reached levels shocking to o≈cials in Lisbon) clearly continued. By 1734–
35, the illicit trading of gold to West Africa to buy slaves had grown almost
too obvious to deny. It was discovered that not only individual smugglers
but a highly organized secret company operated in the trade. One had
agents based in Bahia, Pernambuco, Rio de Janeiro, Sacramento, São
Paulo, and the island of São Tomé, where no less than the local judge was
arrested as one of the outfit’s masterminds.
The African slave trade to Brazil was carried out through the singular
deals of freewheeling merchants and adventurers; o≈cial contracts, ob-
scurely awarded but loudly disputed; and clandestine enterprises spread
across the Atlantic. There was constant tension between free commerce
and the exclusive system of the Portuguese monopolies, but both forms of
the trade were implicated in the shadowy transfer of Brazilian gold to other
European traders.∑≠ However the myriad actors went about the business of
trading in human captives, all involved sought privileges, personal ad-
vancement, and riches. In that sense, the slaves themselves were viewed
less as living beings than as merchandise, a means to an end.
The numbers in table 3 present baptism records from the Parish of Sé in
the city of Rio de Janeiro from 1744 through 1750. They can be cautiously
compared with the data in table 2 for a sense of the flux of imports of Mina
slaves in particular. Some hints about the general decrease in total num-
bers might be derived from an earlier document, from 1738, that mentions
a reduction of slaving voyages from Rio to the Mina Coast.∑∞
The numbers in tables 2 and 3 need further contextualization, of which
I can provide only a preliminary sketch. Maurício Goulart found that be-
tween 1735 and 1740, the slave population of Minas Gerais remains stable,
although it had been clearly growing in previous decades. There is even a
slight decline perceptible after 1740.∑≤ Over the same period in the city of
Rio de Janeiro, there was a continued increase in the levels of Mina slaves
58 | Chapter Two
Table 3. Baptisms of Adult Africans in Sé Parish, 1744–1750
year provenience total

Mina Guinea Central Cabo Cacheo East
West Verde Coast
# % # % # % # % # % # % # %
1744 12 4.5 1 0.4 3 1.1 3 1.1 0 0.0 0 0 19 7.2
1745 12 4.5 0 0.0 0 0.0 3 1.1 0 0.0 0 0 15 5.7
1746 33 12.5 0 0.0 1 0.4 6 2.3 0 0.0 0 0 40 15.1
1747 24 9.1 0 0.0 1 0.4 0 0.0 1 0.4 0 0 26 9.8
1748 37 14.0 0 0.0 0 0.0 4 1.5 0 0.0 0 0 41 15.5
1749 49 18.5 2 0.8 0 0.0 0 0.0 0 0.0 0 0 51 19.2
1750 68 25.7 2 0.8 0 0.0 3 1.1 0 0.0 0 0 73 27.5
Total a 235 88.7 5 1.9 5 1.9 19 7.2 1 0.4 0 0 265 100.0
Source: acmrj, Livro de Batismo de Escravos, Freguesia da Sé, 1744–61. See the additional explana-
tions in source note for table 2.
a
that roughly tracks with the growth of the city proper. The number of Mina
adult slaves baptized in Sé went from 12 in 1744 to 68 in 1750, while the
total baptisms for all African adults grew from 19 to 73 in those years. The
increase between 1744 and 1750 in the absolute number of slaves in the city
might be attributed to the slight lowering in the price of slaves, which
might itself be a reflection of a tapering o√ of demand in Minas Gerais. But
more likely the main factor was the rapid transformation of Rio de Janeiro,
in terms of new urban development, population growth, and businesses;
as well as the creation in 1751 of the second High Court in Brazil (the first
was in Salvador, the colonial capital until Rio earned that distinction in
1763).∑≥ Just as slaves imported from the Mina Coast had been introduced
to meet an urgent demand in the mines, which the trade with Angola
appeared insu≈cient to satisfy, so would Mina slaves in later years be
brought to serve the needs of the new capital of the colony.
After 1730, the decline of direct commerce between Rio and the Mina
Coast is probably based on the requirement for prior authorization from
the viceroy, associated with the machinations of the powerful Bahian trad-
ers whose interests the restriction served. According to a survey carried out
by Pierre Verger in the Hague, between 1724 and 1740 there were 212 pass-
ports awarded by the Dutch West India Company to ‘‘Portuguese vessels’’
docking at the Mina Castle: 129 of them were from Bahia, 73 from Pernam-
buco, 9 from Paraíba, and 1 from Rio de Janeiro.∑∂ Even granting the steep
decline in baptisms of Mina slaves from 1725–26 (110) to 1744–45 (down
to 24), the appearance of these new arrivals needs to be comprehended.
The likeliest explanations are that they were brought from unlicensed ships
between Rio and the Mina Coast, or that they had first arrived at port in
Brazil in a di√erent city, such as Salvador or Recife.
the trade in gold and tobacco

We enter very briefly here into the ways that historical documents describe
slaves who derive from di√erent regions (and I must refrain from pursuing
how such ideas ultimately fed into the wider colonial-era processes of
identifying Africans in both Brazil and Africa). Boxer linked the trading of
slaves from the Mina Coast at the end of the seventeenth century with the
impelling demand at the gold mines, but also with the physical force of
Mina slaves perceived by contemporary observers.∑∑ Referring to roughly
the same period, from 1700 to 1730, Russell-Wood suggested that the bad
reputation of Mina slaves in Bahia created downward pressure on their
prices, which made them appealing when Angola slaves could not be
produced in su≈cient number to satisfy local demand.∑∏ But a letter from
the governor of Rio de Janeiro to the crown in 1726 plays up their ostensibly
unique and valuable knowledge: ‘‘No miner can live without a Mina Black;
it is said that only with them is there good luck.’’∑π Along those very lines,
an eighteenth-century manuscript, cited by Scarano, highlighted the sup-
posed familiarity these slaves had with the methods of prospecting: ‘‘The
Mina Blacks from the Tombuo Bambui Kingdoms are generally the best
miners of gold in Brazil, and perhaps it was they who instructed the
Portuguese on how to remove the gold from the alluvium, since it is so like
a method they already know.’’∑∫ This line is highly valuable, since it indi-
cates that the Portuguese searched out specialists in gold mining from the
region of Timbuktu. Although it might be utterly anecdotal, it (and other
references to a magical ability of ‘‘Mina blacks’’ to sni√ out gold) reinforces
the idea that West Africans were known for gold mining, which suggests
60 | Chapter Two
that this might have been the motive for choosing slaves from the Mina
Coast region to work in the mines of Brazil. It was an idea with little
substance behind it, since most of these slaves had never mined in Africa.
But in a form of popular alchemy, their very presence came to connote good
luck, good fortune, and happy endings to those eagerly running the min-
ing and wildcatting operations in Brazil.
The question of the particular qualities accorded to slaves by prove-
nience must be treated with care because, as Russell-Wood pointed out, the
qualities attributed to such groups by those in power vary with place and
epoch. For instance, we know that most slaves sent from Bahia to Minas
Gerais early in the eighteenth century walked o√ ships that had come from
the Mina Coast; and we have a sense why, based on the previous para-
graphs. But I am also interested in the practical dimensions of the trade in
these people—the financial, mercantile, and shipping logistics involved in
their commerce, the di√erent merchants and agents they connected, who
distributes them where and why—because it is only when considered in
this totality can we begin to comprehend the meanings and values that
di√erent African peoples held for plantation owners and mine operators.
This level of research and analysis remains largely an ideal.
If the Portuguese perspective of the slave trade to Brazil in the eigh-
teenth century involved a tightening of the reins on colonial upstart trad-
ers, from the point of view of Rio de Janeiro the goal seems to have been to
continually try to thwart metropolitan controls. From the early decades of
the century, gold was packed into vessels waiting in Rio’s port to embark
for the slave markets of the Mina Coast without the slightest concern for
the royal fifth, or other taxes—much less for the need, articulated later, for
o≈cial permission to make the route. Brazil’s main ports all participated,
as Boxer argued: ‘‘Gold was smuggled out by unfrequented rivers and
paths through the sertão to Rio de Janeiro, Bahia, and Recife (whence some
of it was again diverted illegally to the slave trade in West Africa).’’∑Ω Unlike
Bahia, which could use its tobacco as a homegrown resource in the trade
with African kingdoms, the city of Rio de Janeiro relied on Minas gold—
and, increasingly, Bahian tobacco, in its commerce with the Mina Coast.∏≠
Even though it depended on carrying goods from other Brazilian regions
across the Atlantic, this trade, in concert with Rio’s other mercantile en-
deavors, created many rich men in the city. The Pereira de Abreu family had
long roots in the slave trade, but when Cristóvão Pereira de Abreu began
combining commerce in tobacco and slaves in the first half of the eigh-
teenth century, he increased the family’s wealth. His heir Joaquim José
Pereira de Abreu continued in Cristóvão’s gilded footsteps and traded in
slaves until well into the nineteenth century.∏∞
Tobacco from Brazil, principally Bahia, was prized all over the world,
from Europe and China to Dahomey.∏≤ In his book published in 1711,
Andreoni stated that Rio de Janeiro was receiving 3,000 arrobas of Bahian
tobacco a year at the time.∏≥ I encountered a permit dated 1 April 1712
authorizing the use for trade on the Mina Coast of tobacco ‘‘of the third
category,’’ which was prohibited in Europe, as long as the rolls did not
exceed two and a half arrobas in weight.
Rio de Janeiro’s trade in tobacco was plainly inferior to Bahia’s,∏∂ but it
is not without interest to the present study, because the African kingdoms
on the Bight of Benin were well-known customers of the commodity. The
city’s first contract for brokering tobacco was created in 1695, by way of a
royal decree that had been solicited by Rio’s legislative assembly (the
Câmara) in order to help pay for city sta√. On 28 April 1728, Manoel Corrêa
Bandeira won the tobacco contract for Rio de Janeiro from the Oversea
Council, which authorized him to take possession of 4,000 arrobas of
Bahian tobacco; the contract was valid for three years, based on the pay-
ment of 35,000 cruzados and 50,000 réis.∏∑ A law of 10 January 1757 man-
dated the opening of the city’s tobacco contract to competition, and also
authorized the local planting and selling of the crop.∏∏ The Duarte Nunes
Almanac of 1799 indicates that in 1794, there were twenty-four merchants
focused on reselling tobacco, a number that jumped to thirty-five by 1798.∏π
But if we go all the way back to the first waves of voyages from Rio de
Janeiro to the Mina Coast to exchange gold for slaves, it is possible that
tobacco was also on board those ships as a trade good. Its visibility to
o≈cial record keepers would have been lower in the seventeenth century
than later, and it had the additional virtue of being directly appealing to
African slave traders themselves, whereas European brokers in charge of
much of the slave trade would have demanded gold.
What remains to be done in this chapter is to suggest how the church
hierarchy, lay religious brotherhoods, tobacco merchants, and Mina slaves
all intersected in the city of Rio de Janeiro throughout the eighteenth
62 | Chapter Two
century. One perspective on this network comes from a letter written in

1709 from the bishop of Rio de Janeiro, D. Francisco de São Jerônimo, to D.
João V, the king of Portugal. D. Jerônimo served as temporary governor of
Rio de Janeiro three times—during the years 1704, 1708, and 1709. He
wrote to ask the king for financial support to construct a new church in Sé
Parish, and he o√ered a proposal that shows a canny awareness of both
Rio’s trading activity and the metropolitan bureaucracy. Rather than trans-
ferring funds from the treasury in Portugal, which D. Jerônimo noted faced
‘‘so many expenses,’’ the bishop-cum-governor suggested that permission
be granted for Rio’s merchants to take 700 arrobas of Bahian tobacco
beyond what the contracts allowed for. This way the crown would pay
nothing directly, and D. Jerônimo could discreetly mention that the con-
tract for tobacco as it stood was insu≈cient for local needs: ‘‘For many
months the warehouse sits empty, causing great harm to the whites but
even worse damage to the blacks, who sustain themselves with the pre-
cious smoke of the dried leaves.’’∏∫
The letter describes the decadence of the aging Church of São Sebas-
tião (Saint Sebastian), constructed in the sixteenth century when the city of
Rio de Janeiro was founded atop a hill overlooking the bay. During the
seventeenth century the city expanded to the lowlands, and many of the
important families of the area began to visit the newer churches, built by
lay brotherhoods along the plain. There was more autonomy in those
churches than in the older Church of São Sebastião, where the bishop kept
the faithful under a tight rein—especially since it was there he had estab-
lished the Cabido da Sé, or administration of the bishopric. Soon this
church, also referred to as Sé, came to be frequented almost exclusively by
the poor population that lived in the surrounding area, and especially by
the blacks and pardos who had installed their brotherhoods there. The first
years of the eighteenth century had been marked by intense conflict be-
tween the diocese and the black brotherhoods. During the period from
1700 to 1704, the bishop had expulsed one association outright—the Broth-
erhood of São Domingos, directed by men said to have come from ‘‘Guinea
heathen’’—and had struggled to get rid of the Brotherhoods of Our Lady of
the Rosário and of São Benedito (Saint Benedict from Palermo, also known
as ‘‘the Moor’’) led by ‘‘Angola blacks.’’ There is no information about
brotherhoods of Mina slaves at this time; the first notice is from 1716. But
the bishop’s reference to blacks’ benefiting from commerce in tobacco

suggests the possibility that those individuals might have been so-called
Mina blacks, brought to Rio on the same boats that had carried tobacco to
the Bight of Benin.
The letter thus hints provocatively at the relations between traders of
tobacco and slaves, slaves and free blacks, and the bishop who was trying
to gain something himself from the network.∏Ω The proposed increase of
700 arrobas of Bahian tobacco over the contracted 3,000 suggests that the
potential margin, as contributed by the traders, would have been substan-
tial. Indeed, the bishop intimated that this contribution from the increased
local trade would be worth more than the 30,000 cruzados the royal trea-
sury had committed to pay. The bishop further emphasized that the current
amount of tobacco stipulated in the contract was not su≈cient ‘‘for the
expenses that the merchants have in the city,’’ and alerted the crown of the
fiscal dangers facing not only white but black traders. This implies that
already in 1709 there was an assortment of whites and blacks, free and
slave, connected with each other through the tobacco trade and through
proximity or fealty to the Church of Sé. Who those people were, what they
did, how and for how long, were fascinating questions that I could only
begin to answer within the limits of this book.
Nowhere in the documents that address the identity or activity of Rio’s
merchants in the first half of the eighteenth century could I find allusions
to the existence of blacks from the Mina Coast, who were freed, working,
and residing in the city.π≠ However, the existence of such a group is sug-
gested by the participation of freed Africans in the lay religious brother-
hoods. By 1740, we can determine that there was a group of freed Mina
blacks living in the Parish of Candelária who organized their devotion to
Santo Elesbão and Santa Efigênia in a brotherhood.π∞
The ties that are just perceptible in Rio between the bishopric and the
tobacco merchants are even clearer in Bahia. In 1745, Teodósio Rodrigues
de Faria, a captain in the Portuguese navy who had arrived in Salvador five
years earlier, founded there the Brotherhood of Senhor do Bonfim. A
successful and notable man, president of the Tobacco Commission, Faria
quickly found the support necessary to construct a church.π≤ Between 1751
and 1759, the director of the Portuguese fort at Ouidah was a Portuguese
soldier named Teodósio Rodrigues—perhaps a kin of the same name, but
64 | Chapter Two
perhaps it was the benefactor of Senhor do Bonfim himself.π≥ The spheres

of commerce and religion overlap again in the case of Felix Simões de
Azevedo, administrator of the contract for Mina Coast slaves. In 1763, Felix
Simões donated to the Brotherhood of Senhor Bom Jesus dos Martírios in
Salvador—an association composed of ‘‘Jêje’’ Africans, recent arrivals from
the Bight of Benin—a chapel in the neighborhood of Barroquinha, which
came to be one of the more popular sanctuaries in the city.π∂
It is important to remember that it was not only traders linked to the
Mina Coast who had the habit of financing or participating in brother-
hoods; nor was it a recent practice. The relation between economic activity
and religious devotion is old and diverse. In Rio, one could mention the
Brotherhood of Santo Pedro Gonçalves, created in the seventeenth century
by a consortium of wealthy navigators. In the eighteenth century, a hum-
bler group of merchants constructed a public oratory containing the image
of Our Lady of Lapa (Our Lady of the Cave). Years later they united with
another coalition of merchants to form the grander Brotherhood under the
name Our Lady of Lapa dos Mercadores (Our Lady of the Cave of the
Merchants).
In Rio, by 1740 or before, there existed at least four established Catholic
devotions among the Mina blacks: two in the Church of Our Lady of the
Rosário (Santo Antônio da Mouraria, Saint Anthony of the Moors; and Our
Lady of Lampadosa, Our Lady of the Candle) and two in the Church of São
Domingos (Santo Elesbão and Santa Efigênia, and Menino Jesus). The
Brotherhoods of Our Lady of Lampadosa and Santo Elesbão and Santa
Efigênia were the wealthiest and most popular; by the late 1740s, both had
raised su≈cient funds to begin constructing their own chapels.π∑ The Mina
who associated in Santo Elesbão and Santa Efigênia seem to have been
especially well endowed, since they inaugurated their new church a mere
eight years after receiving authorization to begin breaking ground.π∏ It
hardly seems conceivable that a chapel could be built in such a short time
with only the contributions of slaves and poor freed blacks. There must
have been other sources of capital, whether from allied devout patrons or
from more heterogeneous methods of acquisition. This is where the bish-
op’s letter to the crown about the renovation of the church at Sé appears
especially relevant. The construction of a chapel meant something more
than any other building could have in this historical context. It demon-
strated a solid economic base, openly recognized connections with the
Catholic Church, and the strategic capacity to attract allies in both devotion
and other dimensions of public life. These would have been acknowledged
by anyone in Rio at the time, black or white, slave or free, as real advantages.
The slave trade involved more than the displacement of Africans to
Brazil; it constructed an extensive, complex network of commercial inter-
mediaries and interested parties, from the moment of capture until the
slave arrived to make his or her way on Brazilian soil. The gaps and
partialities of record keeping on the trade prevent us from being certain
about where, not just individual slaves, but entire groups of slaves came
from. We cannot know for sure where the slaves called ‘‘Mina’’ in eight-
eenth-century Rio came from, or to what ethnic groups they belonged.
Europeans were little preoccupied with these details, and perhaps the only
agent within the trade who knew anything about particular slaves were
their African captors, probably from neighboring groups, who had first
caught and imprisoned them. At the same time, it is the African groups
who engaged with the Europeans as trading partners and local agents
(such as the Dahomeans) that were the best known, because Europeans
dealt with them directly and had shared business concerns to protect. That
is, the most helpful information in period documents is about African
peoples who practiced the trade, and not about those who su√ered it.
Polanyi’s fine book, for example, focuses with admirable sensitivity and
insight on the Dahomeans as a slave-trading people. But about the peoples
whom they traded, Polanyi had less to say. He noted that the strip of land
corresponding to Dahomey was made up of disorganized, mutable territo-
ries (except Ouidah and Porto Novo) and diverse, rebellious tribes, many of
them habitually battling each other.ππ
Given the lack of African sources, New World documentation obviously
holds more promise for the historian interested in slave provenience as
well as in the ways that captive peoples confronted slavery in a new society.
That this is a complicated trajectory to reconstruct with imperfect sources
is a truism that the researcher learns and relearns every day. Sometimes, a
people could change from being slave traders to the slaves being traded by
others, or vice versa. This was the case of the Dahomeans, who were the
principal source of human merchandise when the kingdom of Allada dom-
inated the Portuguese commerce. Later, the Dahomeans conquered Allada
and became the largest exporter of slaves on the Bight of Benin.
It appears that the Brotherhood of Santo Elesbão and Santa Efigênia in
66 | Chapter Two
Rio de Janeiro was composed of Africans from Cape Verde, Mozambique,

and the Mina Coast. In 1740, those members from the Mina Coast reunited
in a separate congregation exclusively of Mina blacks. Later they entered
into conflict and split into two groups, one led by freed slaves from Daho-
mey, and the other led by freed slaves from the Mahi territory and its
environs (places such as Agonli and Savalu). All of them spoke the so-
called general tongue of Mina,π∫ and all of them are nearly invisible in most
of the period documentation. The challenge of this book is to advance a few
steps conceptually and methodologically in the identification of these
small groups—or rather, in the perspective of how they identified and
organized themselves. This required compiling a wide range of sources
that shine light on the slave trade and on the life of slaves and freed slaves in
Rio de Janeiro, where the di√erent identities within the slave population
were reconceived, elaborated, and articulated.
3 | african ‘‘nations’’ and provenience groups
In colonial Brazil, as throughout the Portuguese Empire, there
\ was not a clear distinction between civil and religious admin-

istration. This apparent ambiguity was part of the structure of
the Padroado, a pact between the Catholic Church and the Portuguese
Crown, which gave a religious dimension to the political bureaucracies of
the king and also allowed the church to be active in civil a√airs.∞ Baptism
records are an example of how the Catholic Church acted to order the social
lives and identities of people (in large part because the church assumed
that everyone in the population was or would be a Catholic). A baptism
record is a written register of the occurrence of the religious ritual on such-
and-such a date, at so-and-so a place. But more than that, in the absence of
any secular civil registers, the baptism document often provided the only
o≈cial written information attesting to a person’s identity. It listed such
basic information as the names of parents and godparents or, in the case of
slaves, the name of the owner.≤ These records can begin to reveal how
informal, individual choice—in areas of personal names, selection of god-
parents, family relations—bumped up against or interfered with the o≈-
cial sphere of an obligation to abstract norms.≥
Slave names are a rich source for comprehending the personal values or
social calculations of their masters, who gave them those names. There
were innumerable slaves, many facing especially tragic destinies, who were
baptized under the beatific designations of Felicidade (Happiness), Ven-
tura (Venture), and Esperança (Hope). Another trend, emerging from the
68 | Chapter Three
documents of the Parish of Sé, was the significant number of slaves bap-
tized under the unusual name Hyeronimo. This was probably in homage to
D. Hyeronimo Barbosa, an adjunct to the bishop, who was responsible for
most of the baptisms at the time. Records for slaves who had come from
Africa typically mention the provenince of the slave, what they usually refer
as his or her nação (nation). In the case of slaves born in the city and bap-
tized as infants, records regularly mention the provenience of the mother,
that means, her nation. Thus, the parochial books are helpful not only for
indicating how the size of the local slave population changed with the
arrivals and births of new slaves, but for showing the categories by which
each new individual is absorbed into the conceptual structure of Brazil’s
slave society. Those categories, in particular gentio (heathen) and nação
(nation) will be scrutinized in the following section ‘‘African Heathens and
Nations.’’ As I have shown earlier, the criteria for identifying slaves started
to be formed during the first years of the Portuguese presence in Guinea,
with the preliminary organization of the Atlantic slave commerce. Those
identities were o≈cialized in the parishes of Brazil, when baptisms were
performed and recorded.
The most fundamental point that distinguished slaves from each other
in the colonial context is whether they were born within Brazilian territory
or outside it, in Africa. The first group is categorized by color in the parish
records, with three predominant terms: preto, pardo, branco (black, pardo,
white). Two words that modern readers might expect, mulato (mixed-race)
and negro, did not appear in the documents; but the distinctions of preto,
pardo, and branco were widely recognized and socially accepted in the era
under study. If their conceptualization and ascription may sometimes seem
vague or inconsistent from our perspective, they were nonetheless carefully
assessed by contemporaries and had meanings and ramifications through-
out colonial society. For instance, the Brotherhood of Santo José (Saint
Joseph) only admitted brancos; pretos and pardos had their own associa-
tions, such as the Brotherhood of Our Lady of the Rosário and São Bene-
dito of Black Men, or the Brotherhood of Our Lady of the Conception and
Boa Morte (Good Death) of Pardo Men. It was common to refer to Africans
in Brazil as pretos, although the term was not only applied to them. An
intensive study of the internal social relations, forms of identification, and
broader social hierarchies that cohered around these divisions is beyond
‘‘Nations’’ and Provenience Groups | 69
the scope of this work. I focus on the other subgroup of slaves that colonial
society created: those born beyond the reach of Portuguese colonial society,
who arrived in Brazil by ship, and who were categorized there on the basis
of a presumed nation or provenience group.
Typically historians have worked with a slightly di√erent set of param-
eters, in which creole slaves were those born in the New World, versus
natural-born Africans, who in this case might also be referred to as pretos,
and were imported. But the documents reveal that some slaves born in
Africa were called creoles, and not gentios (heathens). At the same time,
indigenous slaves in Brazil—often called negros da terra (native negroes) or
‘‘native heathen’’—were obviously born in Brazil, but not within the world
framed by colonial references and mores; and they were sometimes called
gentios, the same as Africans were. The existence of these variations in the
documents suggests to me that the underlying criteria for choosing desig-
nations that include or exclude have to do with the perceived limits or
boundaries of colonial society and the Portuguese Empire. The simulta-
neous possibility of, as documents reveal, creole slaves born in Pernam-
buco, Braga (Portugal) and Angola indicates the fragmented identity of
Portuguese society in the eighteenth century—but it also shows that the
society was seen as genuine in certain realms beyond Portugal and Brazil.
We can only guess at how much critical mass it took to be implicitly
recognized as a branch of society rather than as a mere outpost in the wilds.
That I never found reference to a Mina creole might be a reflection of the
precarious Portuguese presence in that part of Africa, unlike in Angola.∂
The Constituições Primeiras∑ —rules that carefully regulated the formats for
recording baptisms, weddings, and deaths—did not require doing any-
thing dramatically di√erent for slaves; but they did incorporate the current
means of identifying African peoples: Guinea, Mina, Angola, and so on.
The frequent use of phrases such as ‘‘of the gentio of X’’ or ‘‘of the nation X’’
in parochial documents shows the preoccupation with including this in-
formation. At times the curate, vicar, or whoever was writing the record
would have to ask around to determine the proper label of derivation for
the slave being baptized. When an adult slave was being baptized, his or
her own participation in the ritual depended on his or her comprehension
of the Portuguese language and level of Christian indoctrination. On such
occasions the position of authority held by the priest (of which the power to
70 | Chapter Three
write was a critical part) permitted the religious functionary to impose,

even if involuntarily, the conditions by which the record was created.∏
Thus, through the lens of the baptism records, it is possible to envision
how the church participated in reinforcing, combining, highlighting, or
obscuring di√erences and variations in the composition of Rio de Janeiro’s
slave population.
The Livro de Batismo de Escravos, Freguesia da Sé (Book of Slave Baptisms
from Sé Parish) of 1744–61, contains records taken down by a number of
di√erent people, starting with the curate Manoel Rodrigues Cruz (1744–
46), who seems to have adopted the expression ‘‘Gentios from Guinea’’ from
its use in earlier books. By 1753, after several other functionaries had
contributed records, the priest Manoel Fazenda de Castro began a relatively
lengthy tenure behind the pen. In 1753 and 1754, he distinguished himself
from the other scribes by noting all of the newborn slaves as creoles but
without an indication of the mother’s provenience. From 1755 onward, he
methodically included the mother’s provenience in all the records of infant
baptisms. This change reflects the importance of the maternal derivation
in identifying the child, and also reinforces the definition of creole as the
first generation born of an African mother within the limits of colonial
society.π It was curious that, starting in 1750, the standard expression ‘‘João
of the Angola nation’’ or ‘‘João of the Mina nation’’ was substituted by
‘‘João Angola’’ or ‘‘João Mina,’’ with the provenience incorporated into the
personal name. All of this implies that there was a fairly wide margin
within which the ecclesiastics could take records, a fact which should be
kept in mind when it comes to comparing data across time or performing
statistical analyses. It also seems that there was a learning process that each
record taker went through, determining what was appropriate to include
and how precisely to do it—a process in which the regulations in the Con-
stituições Primeiras would have played a significant but not all-encompassing
role. When the nation or provenience group of a slave was not declared, for
instance, the ecclesiastic was usually required to recognize the external
marks and physical attributes of each common group, or at the least speak
with other slaves about where the individual might have come from, in
order to include that reference in the register. It should be emphasized that
listing the nation of a baptized slave was not a required part of the norms for
baptism records. It more likely therefore derived from some sort of local
Table 4. Total Baptisms of Slaves in Rio de Janeiro, 1718–1860
parish baptism total

Children Adults
(with Mother)
# % # % # %
Sé, 1718–26 983 10.6 855 9.2 1,838 19.8
Sé, 1744–50 1,381 14.9 283 3.1 1,664 18.0
Sé, 1751–60 893 9.6 324 3.5 1,217 13.1
Candelária, 1751–60 1,648 17.8 586 6.3 2,234 24.1
São José, 1751–60 1,117 12.1 321 3.5 1,438 15.5
Santa Rita, 1751–60 587 6.3 291 3.1 878 9.5
Total 6,609 71.3 2,660 28.7 9,269 100
Source: acmrj, Livros de Batismo de Escravos, Rio de Janeiro: Sé, 1718–26, 1744–61;
Candelária, 1745–74; São José, 1751–90; Santa Rita, 1751–99.
social imperative, especially given that the importance of these records as

the central mode of noting and implementing o≈cial identity.
Although there was a substantial proportion of freed slaves in the broth-
erhoods, the books of slave baptisms provide the best source of informa-
tion relative to the provenience of Africans in the city. They also maintain
the basic distinction mentioned earlier about slave groupings: gentios
versus creoles, or those born within the ambit of colonial society. They
show that the first group was typically composed of adults, and the second
of children or infants. Table 4 presents a total of 9,269 individual registers
of slave baptisms from 1718 to 1860, including 6,609 children born in Rio
de Janeiro and 2,660 adult African gentios.∫
The analysis presented here takes as its base these two basic groupings.
In the case of children born in the city, the focus of attention turns to the
mothers. One qualification needs to be made about the category of moth-
ers: women could be counted multiple times if, for example, a woman with
three children returned for three separate ceremonies. The category is thus
artificial, but accurate insofar as it suggests the distribution of prove-
niences encountered in the city—assuming that women from particular
provenience groups did not have more or less children than women from
other groups.Ω The category of ‘‘adults,’’ on the other hand, e√ectively
72 | Chapter Three
corresponds to the numbers of adults baptized. Another point to be raised

is that among mothers, there would have been both Africans ( gentios, or
‘‘heathens’’) and creoles, while the adults contained only Africans. This
chapter is dedicated to the analysis of African gentios by way of the slave
baptism records, with special attention to those called Mina blacks (or
simply Mina). Before that can be done, however, we need to more fully
consider the female slaves born within the ambit of colonial society.
female slaves born in the portuguese empire

These women were usually described as crioulas (female creoles), pardas,
cabras (variations of race mixture), pretas,∞≠ or simply as ‘‘a female slave
belonging to ——.’’ For the years between 1718 and 1726 I found a total of
267 registers divided as follows: 126 pardas, 47 crioulas, 5 cabras, 2 pretas,
85 labeled only ‘‘slave’’ and 2 freed women.∞∞ The term creole typically re-
ferred to a male of female slave, born to a gentia (female heathen) mother
within colonial society; a creole was often assumed to be the first genera-
tion born in the city, who nonetheless maintained connections with the
mother’s native language and culture. In the designation of creole identity
there was a connotation both of the past, and of current or racial and social
ascendance —details of which would be adduced in the register. We read,
for instance, of Magdalena Costa, a ‘‘freed creole black,’’ the natural daugh-
ter of Josepha da Costa, ‘‘Mina black.’’ Magdalena had her own small boy,
Custódio, who was baptized in 1745; he was first entered in the ‘‘Book of
Captives,’’ but that entry was annulled and placed in the ‘‘Book of Emanci-
pated.’’∞≤ Whatever the story of Custódio, other sources also suggest that
there might have been only one generation of creoles per lineage, with the
next o√spring reverting to the category of slave (but not African or gentio).
Perhaps that is one reason why there were only 47 explicitly creole mothers
in the data, compared to 267 born within the limits of colonial society. The
phrase ‘‘creole nation’’ occurred, but very infrequently. The creoles con-
stitute a coherent group from the point of view of statistics, but in terms of
social relations and shared interests they are much more di≈cult to recog-
nize and comprehend.∞≥ For the purposes of analysis, I counted mothers as
slaves only when they seemed not to fit any other available category.
The group of children born within the ambit of colonial society presents
a double gradation: from black to white, passing through pardo, and from
slave to free, passing through emancipated status. The baptism records

show how complicated these paths were. Most fundamentally, the book in
which each child’s baptism was registered—that of the captives or the
emancipated—was already a critical and durable indication of the place in
the social hierarchy in which each ‘‘innocent’’ had been born. The majority
of slave emancipations involved colonial Portuguese or Brazilian women
liberating the children of female slaves they owned, which might indicate a
certain proximity between mistress and slave. That remains a tentative
supposition.
Until 1871, parishes in the Bishopric of Rio de Janeiro maintained two
di√erent books in which to record baptisms.∞∂ One was called the Livro de
Batismo de Pretos Cativos (Book of Baptisms of Black Captives). The other
was called the Livro de Batismo de Brancos (Book of Baptisms of Whites),
which refers to people of Portuguese or European descent, born free. But
this book had a dual role that exemplifies the complexity of racial and social
designations in colonial Brazil. When a slave child was emancipated in the
baptismal basin, his or her baptism should have been registered in the
Book of Baptisms of Whites, but that very same book in these situations
was called by a di√erent name. It was referred to—in annotations within
those two books, or in other documents—as the Livro dos Forros (Book of
the Freed). However, some of these freed individuals were listed in the
Book of Black Captives instead, for no apparent reason. The complications
involved in freeing a slave, in social and normative terms, can in a sense be
measured by the frequent, occasionally testy annotations scribbled along
the margins of a page when such an act was performed during the baptis-
mal ritual and had to be recorded. In one situation, a woman declared to
the vicar Ignacio Manoel ‘‘with a letter, written in her hand and recognized,
that she had given liberty to a slave and wanted him baptized in his free-
dom.’’ To the extent that the required baptism represented an act of free-
dom or not, it was clear that the slave had been freed, but the register of the
event went straight into the Book of Captives. A subsequent annotation by
the vicar even included the amount of money paid to free the slave, young
Agostinho, as if to substantiate the circumspect and o≈cial liberation of
the boy. But the register stayed where it was, in the Book of Captives, and in
this case did not even include the facile gesture of a note that the record of
Agostinho’s baptism and the proof of his emancipation should be trans-
ferred to the Book of the Freed.∞∑
74 | Chapter Three
Sometimes, with no added commentary, a register was annulled from

the Book of Captives and placed in the Book of Whites.∞∏ One inevitable
conclusion that these incidents suggest is that there was reluctance on the
part of ecclesiastical authorities to bear witness to the bureaucratic and
symbolic transformation of a black to a white. Even when the documents
o√er abundant proof that church o≈cials knew of the changed legal status
of the child, the register could persist in the Book of Captives, which shows
that emancipation represented a substantial step in the social hierarchy. In
actually passing from slave to freed, a slave had to e√ect the similarly
unlikely passage from one church book to another; emancipation in the
baptismal basin was only considered complete when it was recorded in the
Book of Whites. But even when a register of such an event was placed there,
reference is made to the book as the Book of the Freed to maintain the
critical di√erentiation between freed blacks and free whites. The new jurid-
ical status of a freed slave was certainly valuable, but in and of itself it had
little capacity to challenge colonial social hierarchies.
The question of pardos also deserves attention here, not least because it
can be especially di≈cult to separate them from the black population.∞π We
know that in Bahia at around the same time, pardos never reached more
than 10 percent of the slave numbers.∞∫ There are no equivalent data from
Rio de Janeiro, but the figures for mothers with children registered in the
city between 1718 and 1726 can give a rough idea of the distribution of
pardos. The 983 registers from Sé in those years (table 4) correspond to 716
mothers with an African nation attributed (72.8 percent), 126 pardas (12.8
percent), 47 creoles (4.8 percent), and 94 with other designations, most of
them slaves (9.6 percent). However rightly or wrongly, these limited num-
bers are in accord with the findings for Bahia. Studies are clearly needed to
shine light on the identities and social circumstances of the pardos, but the
present work is based on the clear predominance of blacks whose designa-
tion included reference to African nations or gentios.
african heathens and nations

One thing the historiography on African slavery in Brazil lacks, and which I
cannot provide, is a detailed historical and etymological study of the words
gentio and nação. I have had to develop my own interpretations, based on
extant discussions in the historiography, as well as the nuances discovered

in my own research. In documents from the seventeenth century, these
words apply both to Africans and to indigenous Brazilians. A 1680 law
mentions ‘‘defensive or o√ensive war, against a nation of Indians.’’ In
1680, the crown resolved to ‘‘find a convenient way to reduzir [attract,
assemble and convert] the Gentio of the State of Maranhão,’’ which in-
volved bringing ‘‘these gentios to the embrace of the Church.’’∞Ω
As discussed in chapter 1, use of the term gentio (heathen) derives from
the earliest epochs of Portuguese expansion in Africa, and persists through
the baptismal records of slaves in Rio de Janeiro in the eighteenth century.
The word nação (nation) also appears in those records, and a casual glance
at the documents would suggest that the terms were used interchangeably.
But they represent two distinct classificatory systems for colonized peo-
ples: gentios is consistently used to describe those who were to be cate-
chized, whereas nations refers to di√erent groupings of people that the
Portuguese interact with in the process of colonial expansion. The overlap
is clear, but awareness of the distinction is vital to any close analysis of
period documents.
The word gentio (heathen) was usually used to describe peoples located
outside the scope of the Catholic Church—people who had never been
exposed to a Christian religion and who were considered easier to convert
(as noted in chapter 1). Nação (nation), on the other hand, dealt with any
people who occupied a given territory, and a shared language, laws, cus-
toms, and systems of governance. The term was often used in the period
context of race and caste.≤≠ In this sense nation refers to a people united in
common interests and understandings, which may broadly include Chris-
tians, Jews, Muslims, pagans, or gentios.≤∞ A nation could be any unit of
people—gentio, pagan, or Christian, politically centralized or not—with
whom the Portuguese related, at the time usually for diplomacy or com-
merce, in times of war or peace. Nation was also used to identify those who
were enslaved, whether or not they were black or African. I was able to
observe that the word nação appears in ecclesiastical records from the
fifteenth century to the nineteenth, while gentio, also present in the fif-
teenth, disappears from most documents along the course of the eigh-
teenth century.
I found no mention of a ‘‘Guinea nation.’’ The reference to Guinea was
76 | Chapter Three
always intended to describe or situate a gentio, although ambivalence as to

its particular geographical setting is apparent throughout the documents.
Consider two of the slaving contracts awarded to the merchant Bartolomeu
Marchione in the late fifteenth century. One permitted Marchione to oper-
ate in the Slave River (1486–93), which was a well-known waterway that
gave access to the kingdom of Benin. Another contract (1490–95) was for
slaving in the Guinea Rivers, with little additional information about where
they were.≤≤ Making matters more complex for the historian is the fact that,
as shown in the first chapter, the region called Guinea in the fifteenth
century did not match up with the Guinea of the eighteenth century. A map
published in 1781 bore the title Carte de Guinée contenant les isles du Cap Verd, le
Senegal, la Côte de Guinée poprement dite, les royaumes de Loango, Congo, Angola, et
Benguela avec les pays voisins autant qu’ils sont connus. That is, this map of
Guinea depicted an area from Cabo Verde to Angola, although somewhere
in the middle was a region actually referred to as Guinea.≤≥
The terms gentio and nação did not necessarily refer to an ethnic group,
since they were applied to assemblages of people of possibly numerous
ethnic groups arriving together at the same port (in the case of Africans) or
gathered within the same village (in the case of indigenous). For Africans,
these are often place names that should not be taken as ethnic or tribal
indicators, but as the place where the ship picked up the slaves in Africa
(hence such names as Angola, Loango, Benguela, Mina). Similarly, natives
were referred to by place names such as Carijó, Guarani, Botocudo, and
Caeté. It bears emphasizing again that gentio and nação were equally applied
to the indigenous and to Africans, whether or not they were slaves. And the
word negro could describe an African or a Brazilian native, while the con-
cept of índio (Indian, American native) included natives from anywhere, as
long as they were in their proper cultural and geographical environment.
By the time they reached Brazil, the Portuguese had developed a sort of
commercial anthropology for understanding and classifying peoples in the
Orient and in Africa, a perspective that they essentially transferred whole
onto Brazil’s indigenous peoples.≤∂ In that sense, any analysis of the disap-
pearance of the term gentio to describe African slaves in the eighteenth
century would also have to consider the term’s trajectory in relation to
indigenous peoples (who were called, among other things, gentio ‘‘da terra’’
that means ‘‘local’’ or ‘‘native’’ heathen,’’ in this case, American natives).≤∑
Table 5. Principal Proveniences of Slaves in Rio de Janeiro, 1718–1760

(with percentages for each period)
parish provenience total

Guinea Mina Angola
# % # % # % # %
Sé, 1718–26 516 75.2 107 15.6 63 9.2 686 100
Sé, 1744–50 442 54.8 189 23.4 175 21.7 806 100
Sé, 1751–60 45 22.6 62 31.2 92 46.2 199 100
Candelária, 1751–60 5 7.7 34 52.3 26 40.0 65 100
Santa Rita, 1751–60 4 1.3 88 28.2 220 70.5 312 100
São José, 1751–60 60 9.8 150 24.5 402 65.7 612 100
Total 1,072 40.0 630 23.5 978 36.5 2,680 100
In the baptism books from Rio de Janeiro in the early eighteenth cen-
tury, some personal information about slaves was often left out (such as
birthdates), but the slave’s provenience was always included. A slave was
identified by a composite appellation listing the baptism name, prove-
nience (in terms of nation or gentio), followed by the full name of his or her
owner: for example, ‘‘Josepha of the Mina Heathen, slave of . . .’’ The name
of the owner could change in the documents or even disappear over time,
but the slave’s provenience, once entered in the baptism records, became a
permanent attribute of the slave’s name and public identity even if the slave
managed to attain freedom.
Of the African mothers from principal gentio groups who had their
children baptized in Rio de Janeiro’s Sé Parish between 1718 and 1726, 75.2
percent were called Gentio of Guinea; 15.6 percent Gentio of Mina; and 9.2
percent Gentio of Angola (see table 5 for these and the following figures).
Between 1744 and 1750, also among African mothers in Sé, those propor-
tions had changed to 54.8 percent Guinea, 23.4 percent Mina, and 21.7
percent Angola.≤∏
In Rio de Janeiro the term gentio was common in baptismal records early
in the eighteenth century; by the end of the century, it had all but vanished.
In Sé from 1718 to 1723, nearly all the entries were written by priest Hyeron-
78 | Chapter Three
Table 6. Principal Proveniences of Slaves in Rio de Janeiro, 1718–1760

(with percentages of the total sample)
parish provenience total

Guinea Mina Angola
# % # % # % # %
Sé, 1718–26 516 19.3 107 4.0 63 2.4 686 25.6
Sé, 1744–50 442 16.5 189 7.1 175 6.5 806 30.1
Sé, 1751–60 45 1.7 62 2.3 92 3.4 199 7.4
Candelária, 1751–60 5 0.2 34 1.3 26 1.0 65 2.4
Santa Rita, 1751–60 4 0.1 88 3.3 220 8.2 312 11.6
São José, 1751–60 60 2.2 150 5.6 402 15.0 612 22.8
Total 1,072 40.0 630 23.5 978 36.5 2,680 100.0
imo Barbosa, who regularly used the expressions Guinea Gentio, Mina
Gentio, and Angola Gentio. The first use of nação appears to be from the
pen of the curate Manoel Rodrigues Cruz in mid-1725,≤π and as it grew in
prominence over the years, the specific expression ‘‘Guinea Gentio’’ nota-
bly receded. If we look at who was called a Guinea Gentio, most were
mothers present to register their children, while the slaves called Mina
(who had passed from gentio to nation) were almost all adults being
baptized. The category of Guinea slaves had been present in the city over a
longer period of time and was as a group more socialized to the ways of
colonial society. That might also explain why, for most of the legitimate
children of married slaves recorded in the period, both parents were
Guinea Gentios (forty-three out of sixty-two).≤∫
Although the period of 1744 to 1761 was relatively short, it saw the
greatest number of births (806, or 30.1 percent; see table 6), a fact that
correlates with the patterns of Rio de Janeiro’s growth overall. Baptisms of
the children of mothers who were Mina slaves represented 23.5 percent of
all baptisms. Looking more closely at the Parish of Candelária in the decade
of the 1750s, the presence of Mina slave mothers reached 52.5 percent of
the total for that period. This parish, which included the city’s port, had a
conspicuously large population of Mina slaves compared to the other par-
ishes. The number of male Mina slaves, not only female, was proportion-
ally higher here too, as the records of adult baptisms indicate.
Comparing data from the first half of the century with that of the 1750s
involves as much art as science on the part of the researcher. Two related
phenomena a√ected records for the later period: the expansion of the city,
and the creation of two new parishes to serve it. In simple terms, part of Sé
was separated and called Santo José, and Candelária was similarly divided
to create a new parish called Santa Rita. One can derive a rough estimate for
Sé’s demographics across the 1740s and the 1750s by comparing the origi-
nal Sé in the years 1744–50 with the combined figures for the new, smaller
Sé and its o√shoot, Santo José, for the 1750s. There remain the problems of
city growth that are complicated to factor in, not to mention that there exist
data for only seven years of the 1740s, compared to ten years in the 1750s.
Given all those qualifications, the numbers of baptisms of children whose
mothers were slaves of the three main African categories for the two time
periods in Sé are as follows: 105 Guinea (45 vs. 60); 212 Mina (62 vs. 150);
and 494 Angola (92 vs. 402), for a total of 811 (199 vs. 612).
Taken all together, the numbers in tables 5 and 6 show a general ten-
dency for the Guinea Gentio slave identity to disappear from the baptismal
records over the period 1718–60, while the Angola identity enjoyed a dra-
matic increase, and the Minas a slight one. Part of the e√ect here was
overall demographic change from urban growth, but part also must have
been the normative reallocation of slaves who would earlier have been
considered Guinea Gentios into the Angola nation. All the evidence sug-
gests that most of the Africans called Guinea Gentios had come from the
central-western coast, and probably more specifically from Luanda in An-
gola; commerce with the ports of Benguela and Cabinda would reach a sig-
nificant scale only decades later.≤Ω Some of the Mina had been called gen-
tios, but not many, and they were more easily absorbed into the category of
Mina nation.
These trends lend support to the argument advanced in the previous
chapter about the progressive delineation of western Africa in Portugal’s
worldview: the contours of Guinea were increasingly di≈cult to delimit
(conceptually and on the map), while the Mina Coast had relatively clear
boundaries and features. Slaves who would have been called Guinea Gen-
tios would need to be redistributed into the African ‘‘nations’’ that were
80 | Chapter Three
emerging in the universe of Portuguese colonial and commercial relations.

The term gentio—applied initially to heathen peoples or communities who
should be converted, enslaved, or both—was replaced by nação—still a
generic term, but one whose meanings better fit the demands of the rapidly
expanding slave trade. The discourse was secularized. The individual hea-
then or slave was no longer identified by his or her position within a project
of Catholic evangelism. The new parameters by which such individuals
would be o≈cially recognized took on a geographical refinement that, if
flawed in its accuracy (with the consolidation of purported ‘‘nations’’
around African ports of call serving the trade), was also strategic in its
awareness of a vast, diverse tapestry of colonial relations in the Atlantic
world. The gradual substitution of terms in Rio’s baptismal records thus
expresses much larger transformations in the social and commercial ex-
changes throughout the Portuguese Empire.
the ‘‘nations,’’ or
african proveniences
The words used in o≈cial documents to describe a slave’s provenience
point to a heterogeneous assortment of places and social entities—from
islands and ports, to kingdoms and small villages, to cultural or kinship
groupings. Not infrequently I encountered in the declared provenience of a
given slave a reference to a specific locale or group that was impossible to
find in other sources. The provenience terms should not necessarily be
interpreted as implying the status of an ethnic group, although wider stud-
ies indicate that a few African groups semantically preserved by colonial
society—the Cobu, Coura, and Mahi—might qualify. For instance, the liter-
ature on African history mentions Cabu, Caabu, Kabu, Ngaabu, Caaabu,
and similar variations of Cobu (although not precisely Cobu itself ) as a
subgroup of the Mande, in the region of Senegambia.≥≠ There was little
naval tra≈c between this part of West Africa and Rio de Janeiro, but it is
known that slaves from throughout the region were shipped out to Brazil
through the island hub of Cape Verde (although they were often described
as Mina, not Guinea). The Coura or Couranos, shown in the work of Luiz
Mott to have been present among the slave population of Minas Gerais,≥∞
were a subgroup of the Mina ‘‘nation,’’ as were the Mahi. The Mahi inhab-
ited a mountainous region north of the kingdom of Dahomey, as I stated
earlier, and the final chapter of this book will address them in more depth.
The slave baptism records from Rio de Janeiro present a lexicological rain-
bow of proveniences and derivations. There was of course the Mina from the
Bight of Benin, including its subgroups,≥≤ and the Cacheu, coming from the
port of Cacheu on the São Domingos River in the captaincy of Cacheu, in
modern Guinea-Bissau. The central-western coast of Africa was typically
imagined as divided into two regions, each a distinct source of slaves. From
Kongo, principal subgroups included Muxicongo, Loango, Cabinda, and
Monjolo. Angola’s subgroups included Loanda (later Luanda), the port city,
as well as Kasanje (or Cassange), Massangano, Rebolo, Cabundá, Quissamã,
and Ambaca. Benguela expanded the colonial topography of African social
diversity when it became a third distinct region later in the eighteenth cen-
tury.≥≥ The island of São Tomé, whose Portuguese presence dates to 1485, was
a vital node of Atlantic slave trading; for centuries it served as a warehouse
and factory for the commerce of Africans from the Bights of Benin and
Biafra, Kongo, and Angola. But perhaps because of São Tomé’s isolated
island geography and well-known clearinghouse functions, its name seems
to have been adopted only as a nation, not a gentio group, in the documents.
Between 1745 and 1761 in the Church of Candelária, nearly all the entries
for slave baptisms were written by the vicar Ignacio Manoel. In his detailed
inscriptions we see a pattern of recording the rituals for Mina slaves and
Kongo or Angola slaves in di√erent ways. For Mina slaves, who were
probably reasonably assumed to be undergoing the ritual for the first time,
the entry was typically concise: ‘‘I baptized and placed the Holy waters on
So-and-So, Mina adult.’’ But for slaves from the central-western coast,
where the Portuguese missionary project had long and deep roots, the
ritual was more elaborate. It had to be done sub conditione, or under the
condition of a verbal guarantee from the slave that he or she had not been
baptized before.≥∂ To the extent that this more complex interaction might
reflect a di√erent attitude toward the Central African slave, whose prove-
niences were in a region where the church was present and active, is a
question awaiting careful analysis.
In the baptisms of adult slaves throughout the various parishes of Rio de
Janeiro, slaves from the Mina Coast were predominant (2,063 out of a total
of 2,660); as a rule it was their masters who instigated the ritual. Because
we lack more complete sources to measure the population of Mina slaves
entering Rio de Janeiro in the early and mid-eighteenth century, turning to
82 | Chapter Three
ecclesiastical documents is a necessary recourse. One important conclu-

sion from these documents is that Mina slaves baptized in the city were
usually those who stayed there, while those who passed through Rio un-
baptized were sent on to Minas Gerais, where the ritual was performed. In
1718, the crown wrote to the archbishop of Bahia to take umbrage that
Mina slaves were being sent from the major port cities of Brazil on to Minas
without receiving the sacrament, which, he said, was ‘‘against my recom-
mendation.’’ ‘‘It is enough of an abuse to the service of God and to the souls
of these unfortunates that they were taken to Brazil unbaptized, but to then
allow them on to Minas and the interior is reprehensible.’’≥∑
Baptisms of adults from Angola, Kongo, or Benguela were rare in the
documents. Where they occurred ( just as with the Mina baptisms), the sub
conditione proceedings were only one layer of the complexity surrounding
them. In the first place it needed to be ascertained that the slave was ready
to receive the baptism; as two registers phrased it, the slave was ‘‘first
examined and approved’’ or ‘‘first examined and approved in the Doc-
trine.’’ Other religious authorities could be called in for their opinions both
on a slave’s readiness, and whether he or she had been baptized before.
The vicar Ignacio Manoel noted of one adult from Benguela that ‘‘she
declared to her owner that she had been sold unbaptized’’; of another
female from Benguela that ‘‘she a≈rmed that she was not baptized when
she was taken from her land.’’ He reported of a third, from Angola, who
was the property of a freed black, that his owner knew him to be unbap-
tized because of the testimony of an Angolan ship’s captain who ‘‘has
known the slave since he was very small.’’ It was the clearest case of an
adult slave arriving from the region who was unbaptized. Manoel also
noted in the register of an adult Gentio of Guinea that ‘‘he declared that he
had not been baptized in Luanda because he had been captured and smug-
gled on board.’’≥∏ In the records of adults who had arrived from Kongo,
Angola, or Benguela without having been baptized at some point, it is often
possible to intuit something obscure or illicit in the manner by which the
slave embarked. The majority of cases suggest that the trade from central-
western Africa was done in accord with the norms of resgate, that is, re-
demption—the idea that buying the slave ultimately involved the slave’s
spiritual transformation, not just his or her forced labor, so that the bap-
tism should be done before the slave set foot aboard ship. The method for
indoctrinating adults before baptism, whether or not they were slaves, is

laid out in the Constituições Primeiras, but it is di≈cult to know the degree to
which it was followed in the actual situations involving slaves.≥π In the
Bishopric of Rio de Janeiro, the bishop D. João da Cruz had a short-lived
but notorious episcopate (1741–45) in which he reportedly showed the
fiery conviction of a reformer. The historian Julita Scarano reproduced one
of Cruz’s polemics, transcribed in the Monumenta Missionária Africana, that
deauthorized the standard practice of baptizing Africans before leaving
port for Brazil. He alleged that slaves could only be baptized after they had
arrived and learned the Portuguese language su≈ciently well to compre-
hend the doctrine and be able to respond to questions during the baptism
ritual.≥∫
The port of Rio de Janeiro saw very few vessels that served East Africa
during the eighteenth century. According to Antonil, in Minas Gerais there
were some slaves ‘‘from Mozambique, who had come on ships from In-
dia.’’≥Ω But it was highly improbable that ships from India, returning to
Portugal, had stopped in Rio de Janeiro; Rio was out of the way, and it was
bu√ered by crosswinds that would have made such a voyage almost impos-
sible.∂≠ Still, between 1720 and 1722, seventeen adult slaves described as
being from São Lourenço—by which Madagascar was previously known—
were baptized in Sé Parish. These were the only references to that island
that I encountered in all the records. Also in those years, the entries ap-
peared of two adult slaves from Mozambique. That nineteen slaves from
the eastern coast were baptized within three years does imply the pos-
sibility that a ship serving that route had made it to Rio de Janeiro. Of the 17
from São Lourenço, 4 were baptized in 1720, 11 in 1721, and 2 in 1722; the 2
from Mozambique were baptized in 1720. The concentration of rituals
around 1720 and 1721 suggests that most of the slave owners were follow-
ing church guidelines, which stipulated baptizing slaves within six months
of arriving on Brazilian soil, or at most within two years.∂∞
The segment of baptism records that presents the greatest diversity with
respect to proveniences is that of slave mothers having their children
baptized. As discussed previously, the principal gentios were Guinea, An-
gola, and Kongo, but there were other gentios or nations adduced. I found
a few cases where the mother was described as an ‘‘infidel,’’ but in each of
them the mother in question had given birth aboard the slave ship, adrift in
84 | Chapter Three
the Atlantic.∂≤
Of the 6,609 (table 4) registered baptisms of innocents (newborns and
infants) during the period under study, 2,952 were brought to the church
by mothers described as gentias (female heathen). Of that number, 2,680
belonged to the principal three provenience categories—Guinea, Mina,
and Angola (table 6). The remaining 272, not mentioned in table 6, were
divided among various minority derivations: Benguela (120), Cape Verde
(70), Kongo (33), Ganguela (26), Massangano (9), Monjolo (4), Ambaca
(3), Coura (2), Rebolo (2), São Lourenço (1), Quissamã (1), Luanda (1). It
should be remembered that the data being examined in this study are from
the years 1718–60, which was before the slave trade intensified in the later
eighteenth century. Among the minority proveniences outside the main
Guinea-Mina-Angola circuit, Cape Verde (often written as one word, cabo-
verde) had a special prominence. Although slaves bearing this designation
have long been regarded as an unremarkable fringe in Rio de Janeiro’s
slave population and overall social life, the baptism records show that they
were present in a large enough number to be recognized as a subgroup.
That is especially true given that over the 1750s, female Cape Verde slaves
were concentrated in the parishes of Sé (10) and Santo José (19).
Among other minority groups—such as the Ambaca, Luanda, and Quis-
samã—the simple fact that they were named as such at all suggests that
there was something significant in the reference for comprehending the
slave’s identity in colonial Brazil. These examples also indicate that slave
identification could be based on place names, and not a purported ethnic
group. The ethnic group was one component in the construction of the
identity of the nation, from the point of view of colonial society, but neither
the only nor even the most important one. At the same time, and this is
fundamental, once a given reference to an African locale was established as
a social identifier in the context of Brazilian society, it took on a new
character. The individuals coming from a determined locale or region
ultimately were viewed as constituting not merely loose demographic as-
semblages but coherent social groups composed of members who could
recognize each other as such. They could interact at deeper levels, creating
new forms of sociability and organization—among them, the lay Catholic
brotherhoods.∂≥
the organization of provenience groups

As I said in the Introduction, I use the word ‘‘provenience’’ to speak about
the displacement of people from Africa to the Americas in order to give
attention not only to their cultural background but also to the social condi-
tions and geographical routes of the displacement. Sometimes it is possi-
ble to trace the routes from people’s homelands; in other instances we only
know markets and ports along their routes and will never know their point
of departure or the places and peoples to which they belonged.
Roger Bastide (1898–1974), walking in the footsteps of the first cultur-
alists, and in particular of Melville Herskovits, produced voluminous stud-
ies of the ‘‘African religions’’ in Brazil, among them the already-mentioned
Le Candomblé de Bahia (Rite Nagô) (Mouton, 1958) and Les Religions Africaines au
Brésil. Vers une sociologie des Interpénétrations de Cililization. (Pressses Univer-
sitaires de France, 1960). His attention was addressed to candomblé in
Bahia since it was, as he declared, the most African religious practice he
could find in Brazil. From my point of view, we should highlight its impor-
tance in the opposite direction and see the candomblé as a religious prac-
ticed performed by descendants of Africans under an entirely new social
context that cannot be close to African practices, since enslavement and
forced migration imposed an entirely di√erent status on them. The change
of place, the transfer of people to a new set of social circumstances, is not
somehow irrelevant or secondary to the study of religion. African religious
practices can be, in all apparent aspects, performed in Brazil, but even
when a group with a shared prior understanding of this practice unites to
pursue it, the surrounding social conditions, perceptions, and attitudes are
di√erent from the (itself presumed) integrity of the past. Even when the
shared goal is to recuperate or reconstruct the past, this remains an im-
possibility, although new social meanings are created that should not be
downplayed.
In distinction from these authors who have zeroed in on displacements,
and established a direct relation between places of displacement and na-
tions or gentios, as Rodrigues, Herskovits, and Bastide I am proposing the
idea of provenience groups. This concept acknowledges the potential pres-
ence, even the potential power, of a shared provenience underlying a group
of people in the New World. But without lingering on what the content of
86 | Chapter Three
that shared provenience should be in order to be legitimate, I focus more

on the group’s construction of shared experience in the state of captivity
and in the new social circumstance, and on the tension between individual
and collective strategies throughout the process of social organization.
Proveniences refer to names of the places or social groups from which
African people were taken as slaves by the various agencies involved in the
trade. It is an identification attributed by others that might or might not
have a relation with local realities, in the form of actual place names,
kingdoms, and internally recognized ethnic groups. The concept of prove-
nience groups, on the other hand, takes the proveniences as one attribute
of identification, and looks at how people organized themselves around
them. There is a process of self-attribution, at the level of the group in the
new context, of an identity attributed or imposed from outside. It is also in
this sense that there can exist proveniences with a thousand, or a few,
representatives, even though a slave can still be called by the name of that
‘‘nation.’’ Unlike places of displacement or provenience, then, provenience
groups only begin to cohere when people appropriate aspects of that at-
tributed identity and begin to act in society as some reflection of that
identity. The term nação remains slippery in period documents, because it
can easily convey either individual proveniences or provenience groups—
but in fact its power is that it conveys both at the same time. Also, the
barriers between individuals’ proveniences and provenience groups are
di≈cult to establish in a temporally or geographically consistent way,
because individuals who may not be in contact with each other but share
particular references to provenience in the New World context are able to
act to create personal and collective identities in dialogue with that set of
places of displacement that could be their homelands or not.
The concept of provenience group has a profoundly organizational char-
acter, and is related to what the Norwegian anthropologist Fredrik Barth
called ethnic groups—particularly because Barth was preoccupied with
how ethnic groups form and express their organization. He noted that
although ethnic configurations take into account cultural di√erences, there
is not an equivalence between ethnic groups and culture, because each
group selects the components of culture that it considers most meaningful
and important.∂∂ This openness to alternatives gives social actors space—
culturally, socially, historically—to create di√erent provenience groups at
di√erent times, in di√erent places, even if they share the same background.
And we do find that slaves from the Mina Coast who were taken to di√erent
places or in di√erent eras to the New World organized themselves in diverse
forms, depending on the place and circumstances. In Bahia, they adopted
and used di√erent labels (Mina, Nagô, Jêje), but in Rio de Janeiro they were
usually under the same provenience group called Mina.
The idea of provenience groups also di√ers from the analysis of Nina
Rodrigues. I emphasize the importance of social arrangements that de-
velop in the state of captivity (or freed from it, but still in its shadow), while
Rodrigues’s attempts to discern ‘‘racial origins’’ of Africans and their de-
scendants implied a racial ideal and a ‘‘pure’’ past. I also di√er from the
view of Herskovits and his followers because I privilege these actions as
social and collective arrangements that result from people’s agency, rather
than particular African cultural traits that one might find both in Africa and
in di√erent parts of the Americas. Here, culture and ethnicity are among
the components more or less deliberately used to construct meaningful
social arrangements in the new circumstances that the provenience groups
confront. The question of the relation between culture and ethnic identity
has been explored far more by historians and anthropologists in the con-
text of Hispanic America, with reference to pre-Columbian indigenous
peoples, than in the contexts of African peoples or African descendants in
the New World. For this relatively unexplored area, the theoretical ap-
proach of the anthropologist Miguel Alberto Bartolomé would seem to
o√er promising directions. He has stated that although a collective identity
tends to reflect cultural norms of a given society, the identity does not
depend on those norms to exist. Therefore, the configurations and textures
of identity are not joined at the hip with the dominant culture, and can
change at di√erent speeds and in di√erent directions from culture. That is,
culture is one of the aspects of identity, but not the totality of identity.
Culture can change without identity necessarily changing. In that sense—
and this is the key point—Bartolomé concludes that even if an ethnic
identity is expressed through a past culture, or images of a past culture, in
fact that identity is deeply entrenched in the actor’s present and always in
conversation with it. He observes that the Spanish conquest shows that
even societies who were submitted to a wrenching cultural transformation
could still maintain their ethnic identity.∂∑
88 | Chapter Three
In a practical sense, however, attempting to study provenience groups is

no simple matter. In the context under discussion, Africans in the city of
Rio de Janeiro, one immediate problem is how to define ethnicity. The first
matter at hand is to distinguish ethnic groups from nations, because while
the historiography often uses either or both words at will, they are not
interchangeable. Nation has dimensions of culture and ethnicity, to be sure,
but this social unit was defined within the framework of the Portuguese
Empire and the Atlantic world, and imposed on slaves by colonial agents.
The nation was in essence a box of attributes and references that were
presented, aspects of which the formative group could adopt, adapt, re-
elaborate, or do their best to ignore. As the di√erent individuals in the
group reinterpreted the contours of what nação (nation) meant in their
personal and social lives, a creative process occurred through which a
shared language and set of common cultural practices developed. Within
this complex process of building a new identity and organizing a new
group based on common provenience, it should be noted that culture and
identity are both present but to a degree dissociable and in dialogue, as
Bartolomé proposes.
Because nations and provenience groups are historical phenomena that
accompanied the centuries-long slave trade, and even persisted after its
abolition in di√erent forms, we can consider them in light of their trans-
formation across space and time in order to get a clearer sense of the
ruptures and continuities in the passage from Africa to the Americas. One
of the assumptions of an analysis based on provenience groups is that
individuals and groups, even in captivity, had before them infinite possibil-
ities of social arrangements and rearrangements. While the importance of
slave populations’ native social organizations and cultural practices prior
to dislocation is not eliminated, the principal focus is on how these ele-
ments were placed alongside others to be redistributed and reorganized
once in the New World. I believe this method best suits the situations that
historians of African slavery confront, in which the new forms of social
organization articulated by Africans in the diaspora have more to do with
the immediate conditions of captivity than with memories of the past. The
past is not e√aced but is brought back, as it were, to the degree that present
conditions permit. That means, among other things, that the criteria used
to determine who belonged to this or that group were defined in the
specific locality and social context in which the group organized, and not
by some sort of ironclad tradition from Africa that even the disruptions of
New World slavery were not enough to unbind.
It should be obvious from the foregoing discussion that I believe the
provenience group called Mina in Rio de Janeiro was not identical to the
groups called Mina in Bahia, Pernambuco, or Maranhão. The Mina identity
was associated with a variety of smaller nations or ethnic groups, such as
(in nineteenth-century Rio alone) Mina-Calabar, Mina-Mahi, and Mina-
Nagô, all subgroups that included Gbe and Yoruba speakers deriving from
di√erent proveniences. Studies have not been done on the diversity of Mina
in Bahia, but related research indicates that there the Nagô were seen as
distinct. The Jêje nation of Bahia might correspond in broad outline to the
peoples called Mina in Rio, but in Rio the Yoruba speakers were classified
within Mina identity while in Bahia the Yoruba speakers (Nagô) were
separate from the Gbe speakers (Mina and Jêje). Of course, what was
described as Mina in nineteenth-century Rio might have been di√erent
from the Mina in that city a century before. We know that in the eighteenth
century, the group called Mina was predominately Gbe speaking, but dur-
ing the nineteenth century that changed as Yoruba speakers became more
numerous. I am using linguistic di√erence here for the sake of argument as
a sort of grand divider of the ethnic waters, but even this is far too simplis-
tic. Literature on African history demonstrates the pervasive and complex
exchanges between Gbe and Yoruba speakers in the region of the Bight of
Benin, both in culture and in social relations—not to mention the recorded
cases of bilingualism, intermarriage, and collective migrations. And I do
not want to give the impression that any and all change in these spheres of
identity was a direct consequence of the slave trade. Be it in Africa or in the
diaspora, in regions more or less touched by the Atlantic slave trade, in
every epoch, such changes arose according to wider logics or strategies of
social arrangement that were manifested di√erently in every situation.
Thus, rather than looking at ethnicities in the sense of ‘‘pure’’ origins or
lineages with steadfast cultural characteristics, I see the nations and prove-
nience groups as social configurations engaged in permanent processes of
reorganization and redefinition.
Following this reasoning it comes to mind that the di√erent possibili-
ties for an ethnic group to appropriate aspects of cultural traditions from
90 | Chapter Three
other groups within the same overall nation, or even from other nations,
makes an isolated definition of ethnicity quite di≈cult, principally when
cultural criteria are used to try to distinguish between groups. And most of
the extant literature follows and reiterates that problematic perspective,
which means that we have to draw more widely on all the available primary
sources and theoretical models. At times the di√erences between nations
and ethnic groups are clear, at other times not, but even when they are
clear, their meanings might be diverse. All that said, it remains important
to keep in view the distinction, whether one analyzes nations at the macro-
level or the micro-level. I view the nation as an assemblage of ethnic
groups, engaged in continual historical processes of change, rather than as
a constellation of cultural manifestations, traits, or heritage stripped from
their social and historical contexts. It is this dynamic, and not only the
transfer or di√usion of African cultures around the Atlantic, that underlies
the processes of identification and di√erentiation of the African popula-
tions in captivity.
A focus on provenience groups places at least as much significance on
circumstantial criteria, such as ports of embarkation, as on cultural aspects
such as language. But even cultural components of identity may not be
ethnic. For example, the Mina in both Rio de Janeiro and Minas Gerais all
spoke what was called in eighteenth-century documents the ‘‘general lan-
guage of Mina.’’∂∏ Advanced linguistic studies of this ‘‘language’’ remain to
be performed, but evidence suggests that it emerged from the coexistence
of several languages, likely of the Gbe family. And when Mina in Rio de
Janeiro founded the Brotherhood of Santo Elesbão and Santa Efigênia in
1740, they did so not as a purely ‘‘ethnic’’ exercise but in the company of
Africans from Cape Verde, São Tomé, Mozambique, and even more dispa-
rate regions. Indeed, the brotherhood was a sort of supra-organizational
unit that comprised various African nations. I could find no standard name
for this level of social organization in the period documents. The word
Africans existed in colonial vocabulary, but it did not enter into daily usage
to encapsulate peoples otherwise recognized as diverse until the nine-
teenth century.∂π During the eighteenth century, the usual expressions were
‘‘all nations’’ and ‘‘the assembled nations.’’
But if what many historians of Latin America and the Caribbean call
‘‘ethnic groups’’ or even ‘‘tribes’’ are, in fact, colonial-era nations and
provenience groups, how do we arrive at a coherent definition of an ethnic

group? A nation can be recognized within the archives of slave society by a
series of distinctive traits—provenience, language, physical markings, rit-
uals, material culture. Proceeding along this line of analysis takes us right
back to the methodology of Melville Herskovits. Similarly, most historians
explain the existence of an ethnic group by arguing that its members share
a ‘‘common origin.’’ The problem here is the assumption that cultural
traits have the mysterious power to not merely represent but actively pro-
duce connections between the past and the present. African slaves in the
New World are often portrayed (even by scholars with honorable inten-
tions) as being joined at the hip with an all-powerful past, and even when
other cultural influences become absorbed into social life that the African
past is still present, whether the Africans and their descendants know it or
not. This is the attitude pervading Roger Bastide’s arguments in his book
As Religiões Africanas no Brasil, and it can often seem vaguely condescending
since Westerners are implicitly constructed as less dominated by their
pasts. There is a corollary to this view, more normative in character, which
quietly (sometimes loudly) roots for Africans and their descendants in the
New World who chose to adhere to some idea or expression of an authentic
African past. But those choices have to be considered in each context. In
contrast to those views, the utility of provenience groups as a base for
analysis is that they throw into bold relief the processes of constructing
connections, exercised directly or indirectly by diverse institutions of slav-
ery in the lives of African slaves in the New World. ‘‘Traits’’ of particular
groups have not persisted through the centuries; the cultural backgrounds
of ethnic groups are continually reworked and reelaborated through time
and space.
I am attempting to move the debate away from focusing on ‘‘origins,’’ in
terms of an original or ideal culture, to focus on provenience groups and
the constantly changing cultural background and processes of complex
social change that define them. We might be able to move away from the
fretting over cultural loss and degradation or the squabbles over authen-
ticity that are understandable but have been increasingly shown to be
unproductive. João Pacheco was writing of change in indigenous groups
when he observed: ‘‘Social units abandon old cultural forms, receive (and
remember) forms from other societies, and beyond that create new and
92 | Chapter Three
distinct forms.’’∂∫ Ethnic groups should be seen as a kind of toolbox from

which people may draw to make innovative and practical use of their social
organization, politics, kinship, and language, in particular historical con-
texts. They are in a constant state of transformation, although changes are
always complicated and usually do not occur quickly. That is why one can
no longer establish a list of particular cultural traits and march o√ to
discover the relics and shadows that have survived of them in the New
World, as Herskovits and his followers (such as Bastide) did. Once we
understand that ethnic groups cannot be considered outside of specific
historical contexts, it is easier to understand that they will produce and
negotiate di√erent responses and interjections within their particular his-
torical circumstances. In the case of African slaves in Brazil, one segment
of such a generally defined ethnic group from Africa was captured, en-
slaved, and carried to Brazil, where some of those people ultimately man-
aged to reunite and configure newly meaningful ways to coexist. They
reestablished older forms of sociability while reworking them and com-
bining them with others created in the milieu of a Portuguese colonial
society. I suggest that this perspective is the most historically defensible
way to begin evaluating identity construction amongst Africans in the New
World, and I use it in the following section, in which I analyze the Mina
provenience group of eighteenth-century Rio de Janeiro in more depth.
the mina provenience group

The distribution of principal provenience groups in Brazil derived from
Portuguese naval routes (which themselves depended on a range of actors,
including nautical and shipbuilding skills as well as the winds and tides)
and the commercial accords between people in Africa, Portugal, and Por-
tuguese America. In the eighteenth century, Akan slaves from modern
Ghana (who were called Fante-Ashante in Brazil) were more commonly
traded by the English, and most of them were sent to English colonies,
although some were traded into Brazilian ports. Portuguese and Brazilian
merchants found their niche in the relatively close-by Bight of Benin. The
substantial contingent of Mahi (and other Gbe-speaking) slaves who dis-
embarked in Brazilian ports—making their way to Rio de Janeiro, Bahia,
Maranhão, Minas Gerais, and Goiás—were obtained through networks of
both African traders and European agents and intermediaries.
In a pioneering study of the Brotherhoods of the Rosário in Minas

Gerais, Julita Scarano showed that these associations predominantly con-
sidered themselves Mina, followed by groups calling themselves Benguela,
Nagô, and Angola. There were also many smaller groups, some with only a
handful of representatives, known as ‘‘Dagomé (Dahomey), Tapa (Nupe),
Kongo-Cabinda, Mozambique, Maqui (Mahi), Sabará (Savalu), Timbu
(Tibu?), Cobu, Xambá (Chamba), Malê (Muslims).’’ The indigenous were
also represented, under the name Carijó.∂Ω In his well-known book The
Golden Age of Brazil, focused on the eighteenth century, the historian Charles
R. Boxer mentioned Mina, Angola, ‘‘luangos (Loango), caboverdes (Cape
Verde), (Mozambique), creoles of the kingdom (crioulos do reino, from Por-
tugal), creoles of Rio, bastards, and some random Chinese.’’∑≠ Neither
Scarano nor Boxer was concerned with ethnic groups or African prove-
nience, but both helpfully registered the presence of Africans of various
nominal identities. Their lists help us comprehend both the diversity of
groups articulating an identity, and the multiplicity of criteria that underlay
that identity. Scarano was able to probe additional documents in Minas to
attain a level of detail that the baptismal records in Rio could not provide,
but that are conversant with the documents of Rio’s Brotherhood of Santo
Elesbão and Santa Efigênia, which demonstrated that Mahi, as well as the
Savalu, were present.∑∞ In the documents of both Rio and Minas, it is clear
that these were internal subgroups of the Mina.
The proveniences listed by Boxer are similar to those identified in Rio’s
baptismal records, while the references Scarano provided are similar to
those in the documents of the Mina brotherhood in Rio. Boxer referred to
various sources but did not mention documents pertaining to a religious
brotherhood, although these might be one the best resources for compre-
hending the diversity of African ethnic groups in captivity in colonial Bra-
zil. The level of organization attained by the Mina in Brazil was high, even
earlier than the late eighteenth century, when the importation of Yoruba
speakers (the so-called Nagô of Bahia) began to accelerate.∑≤
According to the available baptismal records, the entrance of Mina
slaves in Rio was most prominent during the years 1718–26, with a total of
681 adults baptized. The periods of greatest ritual activity were 1722, 1723,
and 1724. In the Parish of Sé, the numbers fell dramatically, from a total of
681 adults baptized between 1718 and 1726, inclusive (nine years) to 235
between 1744 and 1750, inclusive (seven years), which shows a drop in
94 | Chapter Three
Table 7. Mina Adults Baptized in Rio de Janeiro, 1718–1760
year parish total

Sé Candelária São José Santa Rita
# % # % # % # % # %
1751 23 2.0 35 3.1 9 0.8 0 0 67 5.8
1752 14 1.2 34 3.0 18 1.6 14 1.2 80 7.0
1753 9 0.8 27 2.4 14 1.2 28 2.4 78 6.8
1754 13 1.1 44 3.8 22 1.9 21 1.8 100 8.7
1755 28 2.4 52 4.5 24 2.1 26 2.3 130 11.3
1756 28 2.4 49 4.3 47 4.1 24 2.1 148 12.9
1757 34 3.0 41 3.6 35 3.1 21 1.8 131 11.4
1758 22 1.9 68 5.9 31 2.7 39 3.4 160 13.9
1759 13 1.1 56 4.9 35 3.1 25 2.2 129 11.2
1760 37 3.2 43 3.7 1.8 1.6 26 2.3 124 10.8
Total a 221 19.3 449 39.1 253 22.1 224 19.5 1,147 100
Source: acmrj,Livros de Batismo de Escravos, Rio de Janeiro: Sé, 1718–26, 1744–61;

a
annual rates from 75 to 34. Together, these two nonconsecutive periods in

Sé Parish saw a total of 916 adult Mina baptized, or 57 per year. In terms of
the numbers of Mina slaves arriving in Rio, there were many more who
passed on through the city to Minas Gerais without appearing in the
baptismal sources.
Table 7 demonstrates that the Mina were concentrated in the Parish of
Candelária (carved out of the original Sé in 1632), with 39.1 percent of the
total baptisms performed on adult Mina slaves. Candelária covered the
port of Rio, along with a busy commercial district.
Manolo Florentino has argued that the supply of slaves to Brazil was
elastic.∑≥ The commerce linking the Mina Coast to Rio de Janeiro o√ers a
rich case study showing that this elasticity needs to be reassessed within
each geographical and historical aspect of the slave trade. Low investment
and high risk characterized the Mina trade, carried out by small boats,∑∂
often through clandestine operations, as opposed to the more broad-based
Angola commerce that involved higher investment but relatively stable
conditions of supply and demand. In the mid-1750s the number of Mina

adult baptisms rose slightly, implying an equivalent increase in imports,
but the sources available suggest that Rio’s trade with Mina never again
reached the levels of the early 1720s, when the most extravagant quantities
of Minas gold poured through Rio and Bahia out to the slave markets of the
Bight of Benin.
Data on marriage between slaves provide another source of insight into
the Mina provenience group. The ecclesiastical compendium Constituições
Primeiras was in favor of slave marriage, but the eighteenth-century Jesuit
Andreoni suggested that slave masters were not enthusiastic proponents
of marriages between the men and women they owned.∑∑ That may be
true, but the proportion of slaves who actually went through with the
vows was small. It would be safe to say that of the minority of all slaves
who married, by far most of them married individuals belonging to the
same master. Baptism records of children include the civil status of the
mother, since it was necessary to state whether the child was legitimate
(conceived in marriage) or natural (conceived out of marriage). Of the
6,609 records of children baptized—some of them children who, as I have
specified, might have had the same mother, so that the total number of
mothers might be somewhat less than that—only 253 noted that the
mother was married.
When slaves married in Rio (much as Andreoni and Schwartz found
in Bahia, and Faria reports for Campos), they usually married someone
within their provenience group: Guinea with Guinea, Mina with Mina, and
so on. This suggests that even if the general values placed on marriage in
colonial Brazilian society led it to be encouraged among slaves, the endog-
amy practiced by members of provenience groups perhaps represented the
exploitation of a small but significant space for the strategic alliance of
people with some shared social, cultural, and linguistic background. Fur-
ther study is needed in order to understand the nature of these unions and
what new options and constraints marriage may have presented to slaves in
Rio. Based on data published by João José Reis in Rebelião Escrava no Brasil,
the historian Kátia Mattoso cites 16 marriages between Africans, 13 of
which were within what Reis termed ethnic groups (9 Nagô couples, 2 Jêje,
1 Hausa, and 1 Bornu), and all of those 26 people involved came from the
Mina Coast. Still examining Bahia, Mattoso suggested that some ethnic
96 | Chapter Three
groups or nations tended to be more endogamous than others, and that

both endogamy and free unions (living together in a relationship outside
of marriage) are more consistently observed among freed Africans than
among slaves, particularly after 1850.∑∏ I am unaware of the situation in
Bahia in the eighteenth century, but the data for that period in Rio do show
a strong inclination for endogamy among provenience groups, whether or
not the slaves involved had been freed.
Given the risks in drawing comparisons between Bahia and Rio in this
context—di√erent geographic locations, holding di√erent commercial re-
lations with Africa, in two di√erent centuries—we can make at least one
observation. If in Bahia the groups from the Mina Coast showed the most
marked inclination for endogamy—13 cases, versus 3 for Angola—in Rio it
was Angola that seemed to pursue endogamy most commonly. But as has
been shown, Angola slaves were the dominant African group in Rio at the
time, as the Mina were in Bahia later. The critical calculation, then, is the
percentage of endogamous people within each provenience group. In Rio
in the eighteenth century, 77 percent of Angola women married Angola
men; 87 percent of so-called Guinea women married Guinea men; but only
33 percent of Mina women were endogamous. In both cases, the smaller
provenience group was less endogamous. The questions for future re-
search then are to investigate in detail why endogamy was more common
when the provenience group had a numerically larger population; and why
smaller provenience groups tended to be marked by a search for marriage
partners beyond their own group. It will also be important (if a particularly
di≈cult task) to understand how people actually sought and evaluated
potential partners, within or outside their groups.
I am unable to consider in depth the indices of marriage between freed
slaves, or free unions, and the way that provenience group demography
might have intersected with them. It is possible to begin such a project by
looking at the baptism records for children of natural (unmarried) parents
in which the identity of the father is adduced. But as a formal, socially
recognized status, marriage provides a particularly helpful perspective into
the types of unions and organizations available to provenience groups, and
as we have seen these options seem to vary with place, time, and circum-
stance. The idea of matrimonial strategies being carefully weighed and
assessed in each situation faced by slaves does not suggest that the result-
Table 8. Marriages of Female African Slaves in Rio de Janeiro, 1718–1760
man’s woman’s total

provenience provenience
Guinea Mina Angola
# % # % # % # %
Mina 10 4.0 1 0.4 6 2.4 17 6.7
Guinea 1 0.4 85 33.6 0 0.0 86 34.0
Angola 12 4.7 0 0.0 96 37.9 108 42.7
Others 4 1.6 5 2.0 23 9.1 32 12.6
No data 3 1.2 7 2.8 0 0.0 10 4.0
Total a 30 11.9 98 38.7 125 49.4 253 100.0
Source: acmrj, Livros de Batismo de Escravos, Rio de Janeiro: Sé, 1718–26, 1744–61; Candelária,
1745–74; São José, 1751–90; Santa Rita, 1751–99.
a
ing patterns of marriages seen in the documents should be shapeless and

incoherent. Rather, the tendency for endogamy within the largest prove-
nience groups in eighteenth-century Rio (Angola) and nineteenth-century
Bahia (Nagô) indicates that even in distant times and places, and indepen-
dent of the internal contours of identity construction and social organiza-
tion adopted by various provenience groups, homologous conditions in
Brazil’s slave regime could foster equivalent practices and solutions.
The data in table 8 show that in Rio de Janeiro, Mina women repre-
sented 12 percent of the total number of married African women. That
number itself was small, as noted earlier—of 6,609 births recorded to slave
mothers, only 253 of those mothers were married. Obviously, most African
women had natural children, and either did not marry or married after
giving birth.
The phenomenon of intermarriage between provenience groups, being
principally Mina/Angola, is significant not only in itself but because of the
wider new social relationships and intermingling of identities that such
conjugal groupings could create. The tenor of the coexistence of Mina and
Angola groups in lay brotherhoods takes on a new light from this angle. As
chapter 5 of this book shows, tensions within the Brotherhood of Santo
98 | Chapter Three
Elesbão and Santa Efigênia were attributed to the entrance of ‘‘Angola

blacks’’ and ‘‘creoles.’’∑π Mina women tended to marry outside the group
more than Mina men, and it was they (Angola women) who were typically
accused of provoking conflicts in the brotherhood by bringing in hostile
elements from outside.
The changing distribution of Rio’s slave population through the 1750s
tracks with the growth of the city, which had until recently been concen-
trated along the coast of Guanabara Bay but was rapidly spreading to the
hinterland, taking over the wetlands and smaller hills. The expanding
habitations to the north, clinging to the hillsides, became the jurisdiction
of the new Parish of Santa Rita in 1751. By 1760, the population of Rio de
Janeiro numbered somewhere around 30,000. Urban change can be seen
behind the distribution of younger slaves in Rio.
In the 1750s Mina and Angola slaves of child-rearing age, and their
children, were concentrated in the new parish districts—80 percent lived in
Santa Rita or São José, as opposed to 20 percent in the older parishes of
Sé and Candelária. Over the decade, 28.7 percent of baptisms were per-
formed in Santa Rita, while no less than 51.4 percent of the total were
registered in São José (table 9), where the most slave owners lived. A
significant portion of these mothers from São José must have enjoyed
relative freedom of transit, given what sources show was the frequency of
attending such churches as Rosário, Santa Efigênia, Lampadosa, São
Jorge, and São Domingos, all on the far side of the city. We can also observe
that the number of baptisms of Mina children (31.1 percent) versus Angola
children (68.9 percent) shows that the Mina still had a substantial presence
in the city’s slave population, relative to slaves from the central-western
African coast, as midcentury progressed. By comparing the baptism rec-
ords for innocents and adults, it is possible to conclude that adult recent
arrivals tended to be concentrated in the Parish of Candelária, but there is a
gender component, with women from both the Mina Coast and Angola
being more numerous, and men outnumbering women in Candelária
(where the port was located, an important location of work for male slaves,
especially Mina). Descriptions of daily life from the era lead us to surmise
that in the Parish of São José, both Angola and Mina women could be
found in the domestic realm but probably more frequently in the street,
since female Mina slaves had a recognizable presence as food vendors—so-
Table 9. Distribution of Child Baptisms according to Mother’s Proveniences, 1751–1760
parish mother’s provenience total

Mina Angola
# % # % # %
Sé, 1751–60 62 5.8 92 8.6 154 14.3
Candelária, 1751–60 34 3.2 26 2.4 60 5.6
Santa Rita, 1751–60 88 8.2 220 20.5 308 28.7
São José, 1751–60 150 14.0 402 37.4 552 51.4
Total a 334 31.1 740 68.9 1,074 100.0
a
called escravas do tabuleiro (tray-bearing female slave venders), who circu-

lated throughout the city carrying their wares.∑∫ The Parish of Sé, which
comprised the oldest parts of the city, was relatively dilapidated and had the
fewest numbers of slaves of the four parishes. Although both male and
female Mina slaves were spread over the other three parishes, men were
concentrated in Candelária and women in São José.
The total of 1,481 baptisms of Mina slaves between 1751 and 1760 (table
10) corresponds to the sum of Mina adult slaves baptized, and the number
of Mina mothers who took their children to be baptized. Since there were
probably a number of slaves not listed in those books, this is still an im-
perfect picture of the Mina population of Rio de Janeiro by the 1750s, but
there is not a better estimation for the period. There is no reasonable way to
estimate the numbers of adults who were not baptized, of women who did
not bear children, of women who had their children baptized outside the
city limits, or of women who did not baptize their children at all. Of these,
Mina women without children were probably the single most significant
portion of the demographic that remains invisible in the numbers recorded
in this chapter’s tables. It was also nearly impossible to identify adult men
who had been baptized in parishes outside the city of Rio. Both men and
women tended to move around a great deal between the captaincies of
Bahia, Rio de Janeiro, and Minas Gerais. Acknowledging these limitations,
100 | Chapter Three
Table 10. Distribution of Mina Baptisms by Parish, 1718–1760
baptisms parish total

Sé Candelária São José Santa Rita
Children 62 34 150 88 334
Adults 221 449 253 224 1,147
Total 283 483 403 312 1,481
then, the data we do have show that the greatest concentrations of Mina
men were in Candelária, and of Mina women in São José. Together, these
two parishes correspond to 60 percent of the Mina slaves in the city of Rio
de Janeiro between 1751 and 1760, as can be calculated from table 10.
Title page of the Constituições Primeiras do Arcebispado da Bahia (1707), the first o≈cial set of
regulations of the Brazilian Catholic Church. Announced in Bahia in 1707, it was published
in Portugal in 1719/1720. Courtesy of bn, Brazil. Published in a modern edition: Sebastião
Monteiro da Vide (Dom), Constituições Primeiras do Arcebispado da Bahia (Brasília: Senado
Federal, 2007).
Title page of Manoel Ribeiro Rocha’s Ethiope Resgatado, Empenhado, Sustentado, Corrigido, Instruído
e Libertado, on the debate about the conversion and indoctrination of African slaves in Brazil,
published in Portugal in 1758. Courtesy of bn, Brazil. Published in a modern edition: Manoel
Ribeiro Rocha, Etíope Resgatado, Empenhado, Sustentado, Corrigido, Instruído e Libertado, presented
and transcribed from the original by Silvia Hunold Lara (Campinas: ifch-unicamp, 1991).
Sketches of male Africans slaves from di√erent proveniences (called ‘‘nations’’), by Jean-
Baptiste Debret (ca. 1820). The artist focused here on the detail of tattoos and hairstyles.
Debret reproduced many of these sketches in his plates of street porters and dockworkers.
Courtesy of bn, Brazil. Reprinted in Jean Baptiste Debret, O Brasil de Debret, plate 56.
Sketches of female African slaves from di√erent proveniences (called ‘‘nations’’), by Jean-
Baptiste Debret (ca. 1820). Here, the artist’s eye was caught by the hairstyles and the use of
European-style clothing among the domestic slaves of elite households. Debret reproduced
many of these sketches in his scenes of the home life of Brazilian families. Courtesy of bn,
Brazil. Reprinted in Jean Baptiste Debret, O Brasil de Debret, plate 44.
Urban Transport in Rio de Janeiro, by Carlos Julião (ca. 1780). Indigenous hammock porters in Rio
de Janeiro. Until the second half of the eighteenth century, some indigenous free workers
shared with Africans the task of conveying people through the city streets. At the end of the
eighteen century most porters were Africans, but Carlos Julião—a soldier serving Portugal—
did not miss the few remaining indigenous porters. Courtesy of bn, Brazil. Printed in Carlos
Julião, Riscos Iluminados de Figurinhos de Brancos e Negros dos Uzos do Rio de Janeiro e Serro do Frio,
plate 7.
Poor urban dwelling in Rio de Janeiro, by Jean-Baptiste Debret (ca. 1820). Various
architectural developments came to the city of Rio de Janeiro following the installation of the
Portuguese court in 1808, although the old part of the city occupied by people of humble
means was a√ected less than other districts. The style and appearance of those poor houses
had changed little since the eighteenth century. Courtesy of bn, Brazil. Reprinted in Jean
Baptiste Debret, O Brasil de Debret, plate 54.
African Women Collecting Donations on the Street, by Carlos Julião (ca. 1780). Notable in this image
of African women in the folia of a Catholic Lay Brotherhood is the ornate, festive attire,
including that worn by a little girl accompanying the older women. Courtesy of bn, Brazil.
Printed in Carlos Julião, Riscos Iluminados de Figurinhos de Brancos e Negros dos Uzos do Rio de
Janeiro e Serro do Frio, plate 35.
Funeral of the son of an African king of a folia, by Jean-Baptiste Debret (ca. 1820). Although
Debret suggested that the folias no longer appeared in public processions on saints’ days
after the arrival of the Portuguese court (1808), he was able to observe this funeral cortege
depicted in detail. Courtesy of bn, Brazil. Reprinted in Jean Baptiste Debret, O Brasil de Debret,
plate 80.
Collection of Donations in the Church of the Rosário of Rio Grande, by Jean-Baptiste Debret (ca. 1820).
This interior scene of a church in Rio Grande, southern Brazil, shows the folia of the
Brotherhood of the Rosário. The king and queen lead the ceremonial collection of donations,
alongside the capitão da guarda (royal bodyguard), musicians, and other members of the court.
Courtesy of bn, Brazil. Reprinted in Jean Baptiste Debret, O Brasil de Debret, plate 90.
Queen of the Folia, by Carlos Julião (ca. 1780). A cortege featuring the queen, dressed in her
regal attire. Parasols, musical instruments, and dancing add to the vibrancy of the event.
Courtesy of bn, Brazil. Printed in Carlos Julião, Riscos Iluminados de Figurinhos de Brancos e Negros
dos Uzos do Rio de Janeiro e Serro do Frio, plate 37.
King and Queen of the Folia, by Carlos Julião (ca. 1780). The king and queen are the focal point of
this street procession, which is also distinguished by the presence of a parasol and the
standard of their reign. Note the presence of the capitão da guarda (royal bodyguard) following
the royal couple. Courtesy of bn, Brazil. Printed in Carlos Julião, Riscos Iluminados de Figurinhos
de Brancos e Negros dos Uzos do Rio de Janeiro e Serro do Frio, plate 38.
African girls dance in the folia, by Carlos Julião (ca. 1780). Note the stylized African costume,
a sheet around the waist combined with lace, which is quite unlike the adult attire. Julião’s
original plate reflects the artist’s detailed examination of the markings and tattoos on the
girls’ arms and chests. Courtesy of bn, Brazil. Printed in Carlos Julião, Riscos Iluminados de
Figurinhos de Brancos e Negros dos Uzos do Rio de Janeiro e Serro do Frio, plate 26.
4 | urban life and brotherhoods in the city
By the closing decades of the sixteenth century, the city of Rio de
\ Janeiro began to expand markedly. It strained its older limits at

the hills to push restlessly outward toward the fertile lowlands,
while the urban core—based on Morro do Castelo (Castle Hill), the Fort of
São Sebastião (Saint Sebastian), the Church of São Sebastião (where the
saint’s image was kept), and the College of the Jesuits—began a progres-
sive slide into decline. One indication of the increased population in the
region came in 1634, when the small Chapel of Candelária in the plains was
elevated to a parish. In 1640, Rio’s assembly acceded to a request articu-
lated by citizens to transform the city’s system of natural drainage, which
was overwhelmed and unpredictable, into a more e≈cient network of
trenches. Once completed, the aptly named Vala (drainage ditch) carried
away city rainwater as well as some waste, and also served as a popular and
functional demarcation of Rio’s urban space from the surrounding fields.
the construction of urban space

The principal social activities that had been centered on the Church of São
Sebastião began to change as the city’s reach was extended to the newly
important lowlands. The residents of Castle Hill, so-called homens bons
(respectable men), and their families relocated from the older, tightly
clustered neighborhoods on the Castle Hill streets in search of fresh air,
open space, and enhanced prestige. Meanwhile, they also gradually disap-
114 | Chapter Four
peared from the nocturnal processions that led up the hill from São Sebas-
tião to participate in the activities of the chapels along Rio’s plains.∞ The
bishop and the priests directly under him stayed behind on the hill, as did
two black brotherhoods (Our Lady of the Rosário and São Benedito) and at
least one brotherhood of pardos. It is unclear when each of these congrega-
tions formed, but already in 1639 the two black groups united to form the
Black Men’s Brotherhood of Our Lady of the Rosário and São Benedito.
The association’s compromisso was approved by Catholic authorities in 1669,
but evidence suggests that the two groups may have been in existence for
sixty years by then.
The Prelacy of Rio de Janeiro dated to 1576, and in 1659, the Church of
São Sebastião was elevated in status to the episcopal See of Rio’s prelacy.
But because that church was rather the worse for wear, the bishop peti-
tioned the king to transfer the See to the newly built Church of Saint Joseph
(São José) on the edge of the city.≤ That idea was immediately rebu√ed by
the brothers of São José, who had no interest in the bishop’s occupation of
their church. The religious life of the city center continued to su√er from
the slow but constant attrition of its nobler flock. Pope Innocent XI in-
tervened to create the Diocese of Rio de Janeiro in 1676, transforming the
prelacy to a bishopric;≥ and in 1685, São Sebastião was reconfigured into a
cathedral for the episcopal See. But the problems persisted. The poverty
encroaching more and more around the old church was also noticeable
within it, as its brotherhoods and lay activities were increasingly being
sustained almost entirely by enslaved or freed blacks and small numbers of
pardos.
But all spheres of Rio’s religious life, including the most humble, ab-
sorbed the bustling energy of a growing city in motion. The black brother-
hoods were increasingly visible in public parades, carrying images of the
saints, colorful banners, and other sacred objects and adornments.∂ As
they grew in numbers, the conflicts also were more marked between dif-
ferent brotherhoods—as well as between the brotherhoods and the Cath-
olic hierarchy. By the end of the seventeenth century, the priests of the See
resolved to expulse both the Brotherhood of Our Lady of the Rosário and
São Benedito (led by Angola Africans and creoles) and the Brotherhood of
São Domingos (led by so-called Guinea gentios) from the Church of São
Sebastião.∑ In essence, such disputes took two often related forms: on the
Urban Life and Brotherhoods | 115
one hand, conflict between the brotherhoods and the ecclesiastical author-
ities; and on the other, conflict between brotherhoods over the use of space
in the church and over the place of each in the local hierarchy. Public
processions were often the clearest manifestation of that ordering (since
sequence equated to status), as well as colorful stages for its (subtle or
direct) contestation.∏ This concern over the order of appearance in proces-
sions reflects a pervasive preoccupation with the way hierarchical order
was conceived in colonial society framed by the ancien régime.π According
to Schwartz, an ordered society makes viable in practice hierarchies of
grade, privilege, and honor. In such a system it is possible to define the
position of any individual according to his or her insignia, privileges, or
even obligations. In such social conditions, ‘‘protocol and order assume
important symbolic significance in public events,’’ serving to reinforce the
‘‘prerogatives of each group.’’∫
In June 1702, D. Fr. Francisco de São Jerônimo became the third bishop
of the Bishopric of Rio de Janeiro; his tenure lasted until his death in 1721.
Born in Lisbon, a doctor of theology and censor for the Santo Oficio (Holy
O≈ce), D. Fr. Francisco arrived seemingly intent on chafing local sen-
sibilities. Once established in Rio, he fired o√ a jeremiad to the crown
about the unacceptable condition of the Church of São Sebastião relative to
the disproportionate opulence of the newer churches. Like a sharp blow
to a hornets’ nest, this act aggravated the aggression that had been mildly
simmering between Rio’s churches (and even after the initial heat wore o√,
the peevishness between churches persisted for a century).Ω In his letter,
the bishop unleashed particular invective about the habits of Rio’s female
population.∞≠ Whatever his e√ects on local women’s comportment in gen-
eral, the bishop did attain permission in 1705 to build a convent, which
later would be called the Convent of Nossa Senhora da Ajuda (Our Lady of
Relief ).∞∞
The lay brotherhoods of Rio de Janeiro drew inspiration from older Por-
tuguese models, as well as from the local Santa Casa de Misericórdia (Holy
House of Mercy), but they represented diverse sectors of society. In the
eighteenth century, according to the regulations of the Tribunal of Orders,
each parish should have a Brotherhood of the Santíssimo Sacramento.
Since it had the o≈cial incentive of the ecclesiastical authorities, this devo-
tion was usually held in high regard locally, a fact which attracted people
116 | Chapter Four
and families of greater or lesser means. Still, its members had little auton-
omy, because the bishop kept close watch over its functions. Throughout
eighteenth-century Brazil, these were the most sought-after and exclusive
brotherhoods for white men; perhaps only the third orders of Carmo and
São Francisco were more competitive.∞≤ Most churches were constructed
through the contributions (financial and otherwise) of groups organized
around the devotion to particular saints, such as the lay brotherhoods; their
constitution typically reflected other nonreligious social factors, such as
common profession, familial ties, or a di√erentiating identity (the blacks
and pardos). A brotherhood that built a church e√ectively owned it, and
could open it to the use of smaller and poorer congregations who might
need to promise in their written statutes to perform certain obligations for
the more powerful brotherhood. In Rio, the Church of São José, con-
structed in the lowlands in the seventeenth century, was the domain of the
brotherhood of carpenters; as time went on, historical factors and urban
change led some of the wealthiest families of the city to worship in that
church. Africans, African-descended creoles, and other pretos (blacks) were
devoted to Saint Benedict and Our Lady of the Rosário, while the pardos
united around Our Lady of the Conceição. Each group had a saint, and a
charter or set of vows and statutes (the compromisso) binding them to the
local church and to the larger Catholic edifice with its roots in Rome. They
also displayed the color and insignia associated with their saints on their
flags and banners. These colors and insignia referred most directly to
saints, but at the same time they were ‘‘read’’ by people in the society as
another part of the symbolic universe that inscribed their humble bearers in
a discrete social position and rank.
Among the principal public events of the city were the festive or funeral
processions organized by lay brotherhoods. In the hierarchy of religious
associations, the black and pardo groups were always made to embody
their lowly position by appearing in the processions’ final ranks. But even
here, at the end of the parade, there were finer distinctions to be made.
Mulattos and freed blacks enjoyed somewhat higher status, especially if
they had served time in the military (in, for instance, the Ordinance Com-
pany of Freed Blacks, created in 1698). The lowest of the low were the slaves
recently arrived from Africa, referred to as pretos novos (new blacks).
Compared to European cities and even to Spanish American cities of the
era, it is clear that Rio de Janeiro in the seventeenth century was poor.∞≥
During the light of day, the city was harsh and unappealing to the eye (and
to the nose, as some period documentation emphasizes). One entered
Rio’s urban core along unpaved and craggy roads, lined with improvised
stalls and low houses clustered so densely as to prevent the coastal breezes
from ventilating their musty interiors. Malodorous trash was everywhere.
Pedestrians did their best to tiptoe and hop through the refuse, as did a
remarkable number of ambulatory vendors selling just about anything one
might want to eat or drink. Here and there in the plazas, impromptu slave
markets o√ered the spectacle of male and female slaves, in varying states of
undress and often still reeling from the middle passage, on display before
skeptical and haggling buyers. But at night the fetid gloom deepened to
murkiness, and the city took on a macabre aspect. The darkness was nearly
total, broken only by the glimpse of an occasional greasy candle burning
within a house, or the lamps placed as a gesture of grandeur above the
signs for a few private commercial establishments. More often, it was the
bobbing lanterns of nocturnal funeral processions that o√ered a bold but
ephemeral challenge to the darkness; their light revealed sinister-looking
men and women, idling alone or in small groups along the narrow streets.
The reek of the trash seemed to gather its pungency at night, broken every
now and then by the sudden rainstorms that emptied the streets of litter but
turned them to mud.
Not farther away from the urban core, along the seaside, the situation
was di√erent. A few roads were paved with stone, and they were lined with
houses (often used simultaneously for residence and commerce). Al-
though the city’s freed blacks tended to work in the warrens of central Rio,
where they could also a√ord cheap housing, it was in the less populated
areas beyond the Vala trench that the black brotherhoods began to receive
small donations of land to build their own chapels in the first half of the
eighteenth century. In 1700, the Church of the Rosário began construction
near a stretch of the Vala, on its far side.∞∂ And in 1706, the Chapel of Saint
Domingo was inaugurated near the Campo da Cidade (City Field), which
came to be called the Campo de São Domingos.
A century and a half after the first French invasion of Rio de Janeiro had
been successfully turned back in 1565, French corsairs attacked again, this
time more decisively, in 1710 and 1711. They took Rio not for its own sake
118 | Chapter Four
but for its strategic port, through which so much gold from the mines of
Minas Gerais was known to pass.∞∑ That the city fell so quickly indicates
how poorly it was guarded. In fact, the problems were less of a military or
tactical nature than with a profoundly ine√ective city administration, in
which all the wealth was maintained in the hands of a cadre of powerful
mill owners and merchants. Perhaps fittingly, it was they who paid the
ransom to retrieve the bruised city from the French; the payment accepted
by René Duguay-Trouin to sail away from Rio consisted of 600,000 cruzados,
100 chests of sugar, and 200 head of cattle (to feed the French sailors).∞∏
After this embarrassing and costly episode, more emphasis was placed
on keeping the city secure. Plans were debated in 1713 for a protective wall,
and construction finally began two years later, only to be halted and finally
abandoned at an unknown date. However, the sketches made by João
Massé, coordinator of the project, did survive, and they give a view of the
city’s dimensions as well as some principal features. The sketches also
depict the chapels of the Rosário and São Domingos, although these are
shown to be outside the wall’s zone of protection.∞π
The eighteenth century was marked, throughout the Christian West, by
the crisis of Constantine Christianity in parallel with the rise of the modern
state. The events analyzed in this book, focusing on the Mina in Rio de
Janeiro, occurred within a set of historical circumstances whose connec-
tion to the new relations between church and state are complex, multiple,
and deserving of more attention than I can provide here.∞∫ In the Por-
tuguese Empire, the bishops maintained a significant autonomy through
the e√ects of the Padroado, which kept the ecclesiastical administration
under the orientation of the Tribunal of Conscience and Orders in Portugal
and acted as a bu√er against the forces for both secular and religious
change. The bishops strove to maintain their stature as the highest author-
ity in the territory of their jurisdiction, and were notorious for evading
certain guidelines from the Council of Trent (1564), which stipulated that
they should yield to papal authority in questions of ecclesiastical procedure
(such as norms for indoctrination, administration of sacraments, and
filling out of parish record books).
It may be recalled that in 1707, a synod of the diocese was held in Bahia.
The synod was poorly attended, with the bishops of Olinda, Maranhão, and
Rio de Janeiro absent. The synod proceeded with its Bahian representa-
tives, and its most enduring result was the publication in 1719 of the
Constituições Primeiras, the already-mentioned ecclesiastical code and regula-
tion that was first written for Bahia and later applied to Brazil. Given the
lack of other interventions to define such guidelines for other bishoprics in
the ensuing years, the code from Bahia gradually became adopted into the
bishoprics throughout Brazil. The need for these regulatory structures was
both urgent and delicate, because Catholic religiosity in eighteenth-century
Brazil—what some scholars have aptly termed baroque Catholicism∞Ω —
depended on wide lay participation. The diverse ceremonies that the public
instigated, in larger or smaller groups—in their homes, in churches, or in
chapels they themselves had built—were important for the church’s vitality
and reach. These events and practices also became dynamic spaces for
sociability in colonial society.≤≠ In Brazil, researchers exploring the history
of the church have tended to see the profusion of lay religious activity as an
expression of the distance between ecclesiastical structures and the general
population, especially given the sheer scale of the territory that was under
the jurisdiction of a few scattered church o≈cials.≤∞ That factor should not
be ignored, but neither should we discount the predisposition among
bishops in the eighteenth century to avoid the more stringent mandates of
roman clericalism that had been propagated by the Council of Trent.≤≤ The
issue is less one of inaccessible regions, or a lack of religious functionaries
dedicated to pastoral work, or a supposed indi√erence toward the cate-
chism among slaves. The prevailing attitude of local authorities, even in
urban centers such as Rio, was to work around the recommendations of
the new ecclesiastical model. There were administrative complications too,
not least in the filtering of Rome’s guidelines through Portugal’s Tribunal
of Conscience and Orders, with the socially impractical result that they
were indiscriminately applied to slaves and free people, black and white,
cities and villages, lay congregations and clerics. From the perspective of
the Catholicism envisioned by the Council of Trent, the city of Rio de
Janeiro in the eighteenth century presented a picture of incomplete Chris-
tianity (but then so did most of Brazil).
In 1719, Aires Saldanha e Albuquerque Coutinho Matos e Noronha
became governor of the captaincy. Aires Saldanha had a special concern for
public works; many roads were improved and paved under his watch, and
work on the Carioca Aqueduct was accelerated (at least in part through
120 | Chapter Four
indigenous labor). It was also during his administration that the highest
numbers of baptisms of adult Mina slaves was recorded, suggesting that
more slaves were entering the city during this phase of urban development;
simultaneously, there were higher levels of Mina slaves entering the mine
regions of Minas Gerais. In 1725 there was a new governor, Luís Vahia
Monteiro,≤≥ and a new bishop—D. Fr. Antônio de Guadalupe—became
Rio’s fourth, replacing D. Francisco de São Jerônimo. Antônio was steeped
in more rigorous observation of Catholic traditions than his predecessor,
and quickly established the daily celebration of the Liturgy of Hours and
of the Holy Mass. Public prayers and catechisms were commonplace in
his administration, which was also marked by the broad endorsement of
Bahia’s Constituições Primeiras for Rio de Janeiro.≤∂ Through the financial
assistance and oversight of Monteiro and D. Fr. Antônio, some of the preto
and pardo brotherhoods of Rio were able to secure their hold on permanent
spaces for their congregations.≤∑
Already in 1725, the Church of the Brotherhood of Our Lady of the
Rosário and São Benedito of Black Men was inaugurated. This was a
genuine advance for the brotherhood, even though their church had to be
constructed near an undesirable spot along the Vala trench, where city
e√luent tended to accumulate in reeking pools until rainstorms would
carry it out to the ocean.≤∏ In 1734, the See cathedral was transferred from
the Church of São Sebastiáo on Castle Hill to the Church of the Santa Cruz
of the Militaries; three years later the See was relocated to the newer and
increasingly central Church of the Rosário. That meant relocating the
Cabido (chapter of the See) as well. But the priests of the Cabido who for
four decades had tried to rid the prestigious cathedral of its black brother-
hood suddenly found themselves with the blacks once again, although this
time in a situation where the priests were the guests and the blacks, the
hosts.≤π By this time the city boasted many churches and chapels con-
structed by lay brotherhoods, including Santa Luzia, São José, São Do-
mingos, and Santa Rita, and Candelária, the holy candle. Of course, not all
the lay associations were able to build their own spaces. Many of the
Africans arriving from the Mina Coast, for example, maintained their
congregations in the consecrated churches for blacks within the city, but it
was a period of flux. In 1715, one group of Mina slaves organized the
Brotherhood of Santo Antônio da Mouraria (said Glorious Saint Anthony
of the Moors) in the backrooms of the Church of Saint Sebastian, alongside

the brotherhoods of the Rosário and São Benedito, and of São Domingos.
They stayed behind when the other black brotherhoods moved on from the
old church to their own newer chapels. Little is known about this associa-
tion, and about the provenience of the Mina who composed the brother-
hood; the reference to Moors suggests that they were converted Muslims,
possibly brought from Cacheu and arriving in Rio by way of Cape Verde. In
1737, when the Cabido abandoned the Church of São Sebastiáo for the
Church of the Rosário, the brotherhood went with them to find their own
space in the newer church. Sometime before 1740, in the Church of São
Domingos, another group of Mina formed a congregation around the
worship of the Menino (Child) Jesus.≤∫ In the 1740s, another group of
devout Africans organized a congregation at the Church of the Rosário: the
Brotherhood of Our Lady of Lampadosa, which comprised Africans of
various nations, including Mina.
Thus, in 1740, when the Brotherhood of Santo Elesbão and Santa Efi-
gênia was created (about which more will be said in the following chapter),
Rio de Janeiro contained two churches that were owned by black brother-
hoods—Rosário and São Domingos—and another church where a brother-
hood of pardos, Our Lady of Assunçáo (Assumption) and Boa Morte (Good
Death), was organized.≤Ω Beyond these, there were many smaller brother-
hoods and congregations that had formed within the city’s less privileged
ranks. Many were not o≈cially documented or registered, and they united
to worship in whatever space they could find in the urban churches. While
most of the devoted in these congregations were slaves, the leaders were
almost always freed blacks; some were people of means, and the Mina were
prominent in these relatively elevated ranks.≥≠
The manner of dressing, down to sartorial details, was read as an
expression of one’s position in the hierarchy of colonial society. Clerics
wore longer robes than laymen. Freed black women used much of their
available income in gold jewelry and the bright, ornately embellished panos
da costa (large pieces of colored cloth originating in the African Atlantic
ports, usually worn by women around torsos or over their shoulders) to
maintain a distinctive visual impression.≥∞ The African cloth was used to
wrap and carry babies on the mother’s back, or to cover the head in the
manner of a shawl, which was adopted from the Portuguese habit. Because
122 | Chapter Four
these cloths were highly valued by both male and female Africans, they
were an increasingly prominent component of the trade between Brazil
and Africa. Palm oil, cola nuts, certain vegetables, beads, amulets, and
soaps were also sent in bulk from Africa to Rio, where they were sold at the
marketplace in stands that freed Africans themselves often owned.≥≤
Not until a later era did conditions in the city allow for wheeled means of
public transport. Elites rode in tasseled hammocks or padded chairs sus-
pended on poles, both carried by slaves; these were the same types of
conveyance Rio’s nobility had used a century before. Everyone else got
around on foot, although here too there was opportunity for the better
endowed to demonstrate their greater wealth and respectability than the
average pedestrian ri√ra√. This was of great importance, since the image a
person presented on the street for public scrutiny had ramifications for his
or her position in the social hierarchy. If one lacked the resources to be
carried about by slaves, one could still perhaps be accompanied by a slave,
who would often be festooned in elaborate costume and carrying a large
parasol to shade the master’s respectable head. There was some irony in
the hammock: as a way to carry a living person, it connoted deference and
material wealth. But when the passenger was dead, observers knew that the
deceased was an individual of modest means, since a hammock was the
cheapest way to carry the body to the cemetery.
In the early eighteenth century, most houses in Rio were single-story
structures. Because houses also served the dual function of residence and
place of commerce, they were often structured with the store or o≈ce close
to the street; the domestic area was in the more private back section,
connected to the front by a corridor. A small backyard was common. The
walls were built of stone or brick, secured with a mortar with a whale-oil
base. In poorer houses the floor might have been bare earth, stamped to a
hard evenness; wealthier houses had stone or brick floors. Whitewashing
the houses’ outer walls was mandatory, but infrequently enforced; many
poor people left them plain. Furniture was a rarity for all but the wealthiest
residents. A foreigner who passed some time in the city during this era
visited a home where, he later wrote, all the women were sitting ‘‘Moorish
style, on a hard bench.’’ Beside that, the house appeared to contain only
one table and chair, which were o√ered to him to use.≥≥
A chronic lack of skilled carpenters meant that nearly all the doors and
windows, as well as such furniture as existed, had a rustic look and feel.
Still, windows were remarkably uniform in size at around one meter wide
by two high. They opened to the inside, and could be locked shut with a
simple but sturdy wooden latch. Doors were usually a meter wide also, but
often were longer than the windows. They were topped with an awning, or
if possible a form of wooden shutter that allowed air to circulate. Peepholes
were standard, for the convenience of the residents. Some houses used
tightly layered palm fronds for a roof, while other dwellings incorporated
rounded shingles of ceramic tile. Smaller houses might have a two-plane
roof, with one sloping toward the street, and the other toward the rear of the
house. Larger houses grew to the rear, lengthwise, instead of up or laterally;
thus, to accommodate a lengthening structure, the roof on such houses
would be inverted, with the slopes to the sides. The nature and condition of
the roof were assessed by people in colonial society as a direct expression of
the resources belonging to inhabitants of the house. A plain or inadequately
tiled roof, a double-tiled roof, or a roof bearing yet a third layer of colored
tiles visible to passersby—these di√erent roofs related to the ancien ré-
gime’s hierarchical universe, and to the tiered public display of wealth
elsewhere in the city: they likely corresponded to (first) a man who had no
slaves to accompany his passage through the streets; (second) a better-o√
man, probably carried in a hammock by two reasonably dressed slaves; and
finally, a rich man, who would ride on a padded sedan hoisted by four slaves
attired in velvet, lace, and silk.
With time this standard form of house construction took on new varia-
tions, principally from the development of second-story lofts or attics into
habitable spaces. These new rooms were often rented out, turning a single
house into a multifamily home. With further subdivision into smaller
quarters, some houses also became collective habitations for freed blacks
or poor whites. One freed slave known to own his own house, Caetano da
Costa, died in 1749; he left behind his two-story house and three slaves to
his wife.≥∂ As the city grew, domestic architecture also became more am-
bitious. Two-story, even three-story, houses became commonplace in some
neighborhoods, with elegant wood floors and steep staircases. In the back,
low fences demarcated the yards belonging to di√erent houses.≥∑
The Parish of Sé, which made up the oldest part of Rio de Janeiro,
contained in the 1780s some 1,600 houses (table 11). It is as yet impossible
124 | Chapter Four
Table 11. Distribution of Houses by Parish in Rio de Janeiro, 1779–1789
parish house type total

1-Story 2-Story 3-Story
# % # % # % # %
Sé 1,600 27.6 457 7.9 15 0.3 2,072 35.7
Candelária 480 8.3 676 11.6 188 3.2 1,344 23.1
São José 860 14.8 338 5.8 26 0.4 1,224 21.1
Santa Rita 646 11.1 450 7.7 71 1.2 1,167 20.1
Total a 3,586 61.8 1,921 33.1 300 5.2 5,807 100.0
Source: ‘‘Memórias Públicas e Econômicas da Cidade de São Sebastião do Rio de Janeiro

para Uso de Vice-rei Luiz de Vasconcellos por Observação Curiosa dos Anos de 1779 até
o de 1789,’’ 31.
a
to decipher the documents to know how many of these houses kept slaves,
or how many slaves.≥∏ However, if the Parish of São José had the most slaves
in the decade of the 1750s,≥π it had only 26 two-story houses in the same
period. The parish with the most two-story houses was Candelária, where
the prevalent commercial activity in those districts took over the house’s
bottom floor and the family moved up to the second story. In that sense,
these houses should not be thought of as two-story residences, but as
commercial buildings in which the owner and his family lived above the
business.
Manolo Florentino has compiled 1,067 postmortem inventories of the
owners of rural and urban households in Rio de Janeiro from the years
1790–1835. Based on that research, he argues that ‘‘almost all the free men
inventoried were the masters of at least one slave.’’≥∫ We can only wonder
how many of those slaves were Mina, recently arrived and cheap in relation
to other slaves who would be already acquainted with the Portuguese
language (thus making them a likely purchase for free men or freed blacks
working up the socioeconomic ladder). The larger problem is to compre-
hend who had the means to buy slaves at the time, how many they bought,
and what the di√erent social ramifications were for owning one slave, or
more, or not owning slaves. That is, the wills and testaments of freed Mina
suggest that, if they were able to buy their freedom (usually from money
made providing services on the street), some among them also saved
money and bought one or more African slaves, sometimes by pooling their
resources. The value placed on owning a slave seems to have been universal
in colonial Brazil’s slave society; this is a subject that needs further study.
what was done with the dead

The corpses deriving from well-o√ families were buried within the churches’
walls.≥Ω For slaves, of course, church floor space was elite real estate, too
exclusive socially and physically inadequate anyway to absorb the high vol-
ume of defunct slaves. By the seventeenth century, the flow of slave corpses
had overwhelmed Rio’s existing open-air cemeteries and had led to the estab-
lishment of newer, ampler ones. Two of these were the 1613 Cemetery of the
Rocio (Dew) of the City, also called the Cemetery of Mulattos, and the 1623
Cemetery of the Santa Casa de Misericórdia (Holy House of Mercy). Convents
also increasingly used parts of their grounds for small cemeteries dedicated
to the burial of slaves.
The bodies of slaves, as well as poor whites, were often carried to their
final resting places in co≈ns rented for this purpose by the Church of the
Santa Casa da Misericórdia, which held a monopoly on this service. In
1687, the Brotherhood of Our Lady of the Rosário and São Benedito of
Black Men received a special license to possess their own co≈ns. That way,
they had to pay only for funerals themselves and not for the transportation
of the cadaver to the burial site. This privilege was expanded to the Brother-
hood of Our Lady of Boa Morte of Pardo Men in 1688, and to the Brother-
hood of São Domingos in 1699. This made the process cheap enough for
Rio’s slaves that many could have an actual burial ceremony, and escape the
fate of the anonymous collective pits, owned by the Santa Casa de Miseri-
córdia, where those of the humblest means were interred. The use of ham-
mocks to carry the dead, mentioned previously, was the usual way to make
the process less expensive. Since just a few slaves and free blacks were able
to gather enough resources to pay for a more proper burial independent of
Misericórdia’s rental co≈ns and pits, Misericórdia could still dedicate
much of its attention to the less frequent but still lucrative and socially
significant funerals of Rio’s wealthier residents. The attempts of Miseri-
126 | Chapter Four
córdia to control the funeral market derived not only from financial inter-
ests but, perhaps more significantly, desire for the prestige and recognition
that such a position would bring.∂≠
The Church of São Domingos was inaugurated in 1706, and soon it
was interring slave bodies both inside and outside its walls. In 1709, the
Franciscans there obtained a separate piece of land by the convent that they
dedicated as a cemetery for slaves (the land had already been used as an
informal dumping ground for deceased slaves). In 1722, the brotherhood
of the newly built Church of Santa Rita also built a cemetery on its grounds;
as its name implies, the Cemetery of New Blacks was dedicated to Africans.
The Cemetery of the Rosário was opened at roughly the same time. The pits
at Misericórdia remained the standard option for poor blacks who, for
whatever reason, had not entered into a contractual agreement with one of
the brotherhoods to provide a more dignified end for their earthly remains.
The preferred cemeteries for blacks who had the capacity to choose seem to
have been those alongside the churches of Our Lady of the Rosário and São
Domingos. Others were buried in the cemetery of the Candelária Church,
which was the seat of Candelária Parish. The church had worked out a deal
with slave owners, who could deposit their dead there for prearranged
prices. According to wills, obituary records, and other documents, it seems
that black people in Rio at the time were never buried in co≈ns. They were
lowered into the grave in a net, usually wrapped in white cloths or a white
sheet. Some blacks did pay extra for special habiliment—freed Mina with
enough disposable income tended to prefer being buried in the charac-
teristic habit of São Francisco—but the body was always transported in a
hammock.
One of the most-cited justifications for the creation of black brother-
hoods in colonial Brazil was the observed tendency of masters to abandon
slaves who had become old, sick, or injured. Slaves who died with no
system of support beyond the attentions of their owners typically had their
cadavers tossed unceremoniously on the beaches or in the fields outside
the city, or, in the best possible case, dropped at the door of a church, where
it was asked that they be buried ‘‘for the love of God’’ (i.e., at no charge to
the master). It would not be thought untoward for a slave master of limited
means to present himself to the priests with the slave’s corpse, requesting a
charitable burial. But as a 1735 petition from a brotherhood in Salvador to
the crown suggests, this might be distasteful to the master: ‘‘Owners prefer
to discard the bodies of slaves secretly, rather than endure the priests’
drawn-out and embarrassing investigations into their capacity to pay for
the ritual. And even if a charitable burial is finally deemed appropriate,
enough time has gone by that the remains of the slave are too frightful to
handle.’’∂∞
But, in fact, it was not only the ostensibly flint-hearted slave masters
who abandoned black cadavers to the weeds and gulls. Even brotherhoods
would resort to the practice when they did not have enough funds to
provide for a burial. This unpleasant fact was admitted in petitions from
brotherhoods in Bahia, and it is reflected in the documentation of the
Church of Santo Elesbão and Santa Efigênia in Rio. When reading the
petitions, it is important to recall that whenever anyone wrote to the crown
asking for special support or privileges, dramatic elaborations of one’s
unfortunate circumstances was a normal and expected part of the rhetoric.
But often, particularly in the case of black brotherhoods, the vivid descrip-
tions of dire straits can be verified by other documents and do not seem to
depart greatly from reality.
The preoccupation with death, particularly understandable in a slave
regime characterized by high mortality, led many blacks (even those with
limited or no property) to record a testament that specified their final
wishes and their preferred conditions of interment. In other cases, where
there was no testament, the obituary often recorded a few details of the indi-
vidual’s life and how he or she was preparing for death. We see from such
sources that, to cite several examples, Tereza de Jesus, a freed Mina slave,
married to a freed black man, was dressed in the habit of Santa Rita and
buried in the grounds of Santa Efigênia. Antônia de Jesus, another freed
Mina, was buried at the Church of the Rosário. João Batista, a freed Mina,
married to the freed Mina Ana Maria, was attired in the habit of São Fran-
cisco and laid to rest at Santa Efigênia.∂≤ This apparent mixture of devo-
tions, locations of burial, and funeral garb suggests the diverse options
available to blacks as they planned for death, and does away with the idea
that there might have been exclusive or consistent practices of devotion for
slaves, freed blacks, or ethnic groups. The manner of religiosity was above
all flexible, in that it combined the devotion to a given saint with a range of
religious services that could be contracted, according to various factors
128 | Chapter Four
(cost, the spouse’s faith, etc.), all of which were administered in a form to
maximize each person’s salvation within the limits of available resources.
On the day of the event, the burial procession only set out late in the
afternoon. Groups of friends, supporters, and brotherhood members
would gather early in the morning at the house where the dead lay. African
musicians often brought instruments and would keep the music and sing-
ing going for hours. The burials of men or women were proclaimed with
di√erent bell tolls: one toll on a large bell for men, two on a smaller bell for
women. Jean-Baptiste Debret, a French artist who arrived in Rio de Janeiro
in 1816, described the burial of a black female from Mozambique at the
Church of the Lampadosa. He noted that most of the accompanying mourn-
ers were women, who seemed also to be charged with collecting the money to
pay for the funeral, and he noted the phrase repeated throughout the wake:
‘‘We are weeping for our kin [parente].’’ The idea of kinship in this context
raises the strong possibility that the procession Debret observed was made
up, not of one extended family but of the Mozambique provenience group
within which these women organized socially and articulated their religious
practices. Among Africans, the use of the term kinship commonly applies to
wider social groups than in Western, blood-relation cases, and its use in the
context of a provenience group in the New World would be consistent.
Africans from Mozambique could consider themselves ‘‘kin’’ in Rio de Ja-
neiro, as could Africans from Angola or Africans from the Mina Coast.∂≥
In the following section I present a brief survey of the obituaries of
slaves from the Parish of Candelária, between the years of 1724 and 1736.
Candelária at the time had seventeen o≈cial burying grounds: eleven at
churches, four discrete cemeteries, and two places (Carmo and Hospício
de São Francisco) that are unclear as to their location or a≈liation. The vast
majority of burial sites were churches, which shows the close proximity
even at this late date in Rio’s urban milieu between the living, the dead, and
the saints.∂∂
The drawings by Debret, dated between 1816 and 1830 in Rio de Janeiro,
were made a century after the period examined here, but suggest that little
had changed in funerary practice. Significantly, Debret mentioned that the
grounds most prized by the black lay brotherhoods included the cemetery
of the ‘‘Old Sé’’ (most likely in reference to the Church of the Rosário), and
the Churches of Our Lady of Lampadosa, Our Lady of Parto (Childbirth),
and São Domingos.∂∑ The absence of separate cemeteries for slaves indi-
cates that such places were not deemed necessary. It is notable that even
those eighteenth-century masters who insisted on having their slaves bap-
tized to save their souls showed little preoccupation with the destiny of
their human remains. Although o≈cial church doctrine mandated that the
corpses of converted, baptized heathen should be given Christian burials,
colonial society viewed the bodies of their slaves with rather more indif-
ference. That explains the common practice of dumping slave corpses in
the fields or on the beach, instead of bothering to bury them properly.
Many obituary records indicate that the bodies were interred in the patio
area in the grounds at the front of the church. Given the high numbers of
burials, the smaller cemeteries that appeared alongside churches were
most likely the necessary lateral extensions of those front burial grounds.
Around the Church of Candelária alone in the twelve years of records
surveyed, space had to be found for 115 dead slaves. This church was the
seat of the parish, and thus well known and distinguished, but perhaps
something else helps explain the slaves’ a≈nity for it: the Church of Can-
delária housed images of Crispim and Crispiniano, the patron saints of
shoemakers and the open-minded protectors of underdogs and the victims
of discrimination. There was a brotherhood of shoemakers devoted to
these two saints there.∂∏
In general obituaries were registered in the parish where the deceased
resided, but in the case of slaves the record was made where the master
resided. The obituary should include the locale of the burial, which could
be in any church or cemetery independent of where the record was made.
Sometimes a description of funerary attire was made; records tended to
become more detailed and standardized with the passing of time. Al-
though I found 623 obituaries, I analyze only those that presented com-
plete information.
In addition to the cemetery at Candelária, most Africans were buried in
two other cemeteries—those at the Church of the Rosário and the Church of
São Domingos, both on the outskirts of Rio proper (and São Domingos was
extra-muros, or beyond the proposed borders of the city’s protective wall). Of
the 71 burials of Mina slaves, 32 were in the Rosário, 23 in the Candelária,
and 8 in the São Domingos cemeteries (table 12). These numbers bolster
the argument made in chapter 3 that the Mina coexisted with Angola slaves
130 | Chapter Four
Table 12. Distribution of Slave Obituaries from Candelária Parish, by Locale of Burial in
the City of Rio de Janeiro, 1724–1736
burial provenience others total

Mina Angola Guinea Benguela Kongo Cabo Crioulo Pardo
Verde
Rosário Church 32 29 24 3 1 2 26 5 122
São Domingos Cemetery 8 28 22 3 2 1 5 4 73
Candelária Church 23 29 9 11 5 4 24 10 115
Santa Rita Church 3 7 3 0 1 0 2 1 17
Bom Jesus Church 0 1 0 0 0 0 1 0 2
Sé Church 3 3 7 1 0 1 2 5 22
Carmo Cemetery 0 0 0 0 0 0 0 12 12
São Pedro Church 0 1 0 0 0 0 0 0 1
São Cristóvão Church 0 2 0 0 0 0 0 0 2
João Dias Cemetery 1 3 0 2 1 0 0 0 7
São Francisco Hospício 0 0 1 0 0 0 0 0 1
Santo Antônio Cemetery 0 1 0 0 0 0 0 1 2
Hospício Church 1 6 0 0 1 0 0 1 9
São Bento Church 0 1 0 0 0 0 0 2 3
N.S.a Outeiro Church 0 0 0 0 0 0 0 1 1
N.S.a do Parto Church 0 0 0 0 0 0 3 0 3
Misericórdia Cemetery 0 4 0 0 0 0 1 0 5
Total 71 115 66 20 11 8 64 42 397
Source: acmrj, Livro de Óbito de Escravos, Freguesia da Candelária, 1724–1736.

a
‘‘N.S.’’ in Portuguese stands for Nossa Senhora (Our Lady).
in the Church of the Rosário and the Church of São Domingos, which is not
to say that theirs was a harmonious relationship. The numbers also show
that Mina were concentrated in Candelária Parish, where the Brotherhood
of Santo Elesbão and Santa Efigênia was established in 1740; but despite the
limited association of Mina with the Church of São Domingos, that church
would provide the meeting place for two congregations of Mina in the
1740s (Menino Jesus, which might have existed as early as the 1720s, and
Santo Elesbão and Santa Efigênia). The presence of twenty Benguela slaves
should be noted in the data, although only three were buried at the Rosário
cemetery. The prominent African demographic in the cemeteries of both
São Domingos and Rosário was the so-called Guinea heathen, probably a
part of the group that left the Sé in 1704. Their o√spring, a first generation
of creoles, would rise to prominence in the Brotherhood of the Rosário not
long after the period studied here.∂π
The burial records show that in death as in life, Rio’s Africans shared
certain spaces in varying combinations, which may help future investiga-
tions of their intergroup cooperation and conflicts over time in colonial
society. It is a basic premise of this book that even though Africans may
have come to Brazil as captives and slaves, they were not entirely stripped of
the capacity for self-determination; religious practice and a≈liation were a
vital aspect of this self-determination, as was the construction and mainte-
nance of provenience groups. Obviously there were constraints on the
exercise of those volitions, and researchers confront the echoes of those
constraints in the documents. Here, due to the simple scarcity of places
available for slave burial, corpses representing diverse provenience groups
and local social associations were jostled together in the same cemeteries
in a manner that obscures the groups’ diversity and modes of organiza-
tion.∂∫ We are often limited to painting with a broad brush. Data on the
dead and the living Africans in Rio suggest that Angola were the majority in
the churches of Rosário and São Domingos. They had departed the See
some years before and built those two churches; in the case of Saint
Domingos, they had done so together with Africans labeled ‘‘Guinea hea-
then.’’ Other Mina stayed at the See, and had by 1715 started the Brother-
hood of Santo Antônio of the Moors (mentioned in chapter 2). This broth-
erhood left the Church of São Benedito and moved to the Church of the
Rosário together with the Cabido in 1737. Another group of Mina united at
Rosário within the Brotherhood of Our Lady da Lampadosa. A third seg-
ment of Mina organized around the devotion of the Menino Jesus (some-
times called Menino Deus) at the Church of São Domingos. Yet a fourth
group founded the Brotherhood of Santo Elesbão and Santa Efigênia, also
at São Domingos, in 1740. It is possible to a≈rm that Santo Antônio of the
Moors and Menino Jesus were the first two Mina brotherhoods in Rio de
Janeiro. Research suggests that the former was linked with slaves from
Cape Verde, whereas all indications suggest that the latter emerged from
Africans from the Bight of Benin who, already in the 1720s (seven Mina
were buried in 1724), had installed themselves at the Church of São Do-
132 | Chapter Four
mingos. Their multiple subdivisions into di√erent congregations might

imply that the Mina were a dispersed, di√use minority. However, they were
both a numerically significant sector of Rio’s African demographic and, as
initial research into their wills and testaments is revealing, their separation
into distinct religious fraternities did not dissolve their overarching, mu-
tual social ties. Chapter 6 shows various instances of collaboration be-
tween di√erent Mina brotherhoods during the eighteenth century.
It is worth considering the distributions of these burial records from the
perspectives of time and gender.∂Ω Of the 499 records from the period
1724–36 deemed suitable for basic analysis, 251 referred to women and
248 to men (table 13). This is a curious finding given the nearly unanimous
scholarly opinion about gender ratios in Atlantic slave commerce: more
men were taken from Africa and made it to the New World than women.∑≠
Of course, these obituaries point to Christian burials, and do not tell us the
total number of African deaths. This suggests that if more men than
women were brought from Africa, more African women attained the better
socioeconomic conditions necessary for a Christian burial than African
men did in early eighteenth-century Rio. These conditions involved not
merely belonging to a religious organization, but frequent devotional con-
tributions to its charitable fund, as well as to other causes; the personal
accumulation of some patrimony that was saved for the funeral; and good
relations with the masters. All of these considerations in turn provide an
argument that women might have had a notable influence in the admin-
istration of the brotherhoods to which they belonged—even though men
were prone to limit what they saw as excessive female assertiveness, as
chapter 6 shows in more detail.
The feminine profile was especially marked in Mina burials, with thirty-
five males recorded versus forty-four women for the period. Luciano Fig-
ueiredo has highlighted the substantial proportion of black and pardo
women participating in brotherhoods in Minas Gerais (a tendency reversed
in the white population), which he explains by characterizing the women
as more stable than men in that social milieu.∑∞ But a broadly similar
finding in Rio de Janeiro indicates that some explanation beyond local
factors is necessary.
Another way to view the slave population through these figures is by
manner of identification; that is, a basic distinction can be made between
slaves characterized by their African provenience (described as belonging

to a nation), and slaves described in other ways. The first group is domi-
nated by Angola (117), followed by Mina (79) and Guinea (68). The second
becomes a catchall, containing creoles (68), pardos (47), and slaves with-
out any qualifying description of provenience or color (47). The gender
di√erence observed above is consistent in this second group, particularly
for creoles (19 men vs. 49 women) and pardos (15 vs. 32). This reinforces
the idea that women, African or locally born (and birthrates for this genera-
tion are equivalent), ultimately had greater access to Christian burial prac-
tices than men. It is di≈cult to evaluate how representative these numbers
are of the whole population of enslaved and freed blacks in the city, but we
can attempt a tentative projection.
Lacking a population census in the eighteenth century, we have to rely
on records of baptisms and obituaries; these provide an imperfect but
workable notion of the distribution of births and deaths. Taking the three
most significant African nations in colonial Rio, and looking more specifi-
cally at the Parish of Sé between 1718 and 1750, documents show the
following: 958 baptisms of children of Guinea mothers, compared to 66
obituaries for Guinea men and women; 238 baptisms of children of Angola
mothers, compared to 115 total Angola obituaries; and 296 baptisms of
children of Mina mothers, compared to 71 obituaries of Mina.∑≤
It is possible to conclude from this that only a small contingent of
African slaves had access to Christian burials. Given that the cadavers of
most slaves were either left at the mercy of the elements outside the city, or
dropped in the communal pit at Santa Casa da Misericórdia, or discarded at
the door of churches to be disposed of anonymously, it becomes reason-
able to imagine that the performance of a burial ritual based on ecclesiasti-
cal norms probably accrued almost exclusively to slaves a≈liated with a lay
Catholic brotherhood. Even for this comparatively elite group, the norms
were probably rarely fulfilled in every particular. Having the sacraments
administered promptly after death, acquiring appropriate funerary attire
for the corpse, providing transport to the cemetery, including a priest in the
procession, burning all those candles—these individual expenses added up
fast to what could seem a small fortune. Even if a modest hammock, rather
than a pricier rented wooden co≈n, was used to carry the dead, the costs
were still steep.
134 | Chapter Four
Table 13. Distribution of Slave Burials by Year and Sex in the City of Rio de Janeiro,
1724–1736
identifier annual distribution

1724 1725 1726 1727 1728 1729 1730
m f m f m f m f m f m f m f
Mina 5 2 1 4 1 1 1 4 2 2 4 2 2
Guinea 2 20 22 11 8 1 2 1
Angola 4 7 5 6 6 5 3 4
Cabo Verde 1 2 1
Benguela 1 1 1
São Tomé 1
Kongo 3
Ganguela
Quissamã
Muxicongo
Mozambique 1
Monjolo 1
Ambaca 1
Cacheo
São Lourenço
Massangano
mulatto 1
boçala
Rebelo
creole 3 6 6 1 5 6 1 1
pardo 4 4 4 1 4 1 1 1 1 2 2 5
slave 1 1 1 1 1 1
cabra 1 1
ladinob 1
Indian 1
black 1
freed
Total 11 18 35 37 15 15 5 3 11 16 14 18 10 15
Source: acmrj, Livro de Óbito de Escravos, Freguesia da Candelária, 1724–1736.

a
‘‘Boçal’’ refers to recent arrived African slaves who did not know how
to behave within the colonial world.
b
‘‘Ladino’’ refers to African slaves who had learned Portuguese and had
adapted to local habits.
of burials total
1731 1732 1733 1734 1735 1736 Sex General
m f m f m f m f m f m f m f
4 5 2 7 2 6 7 4 2 4 2 3 35 44 79
1 33 35 68
9 1 13 9 2 6 3 6 11 3 10 4 66 51 117
1 1 1 1 7 1 8
2 1 2 1 1 5 1 1 3 12 8 20
1 1 1 2 3
1 1 3 4 1 12 1 13
1 1 1 1 2 3
1 1 0 1
1 1 0 1
1 0 1
1 1 1 3 1 4
0 1 1
1 1 0 1
1 0 1 1
1 1 0 1
1 1 3 0 3
1 0 1 1
1 1 0 1
2 5 1 5 1 5 4 2 3 3 1 7 19 49 68
1 3 1 2 1 2 1 1 1 4 15 32 47
1 3 1 1 2 10 8 10 4 29 18 47
1 1 1 1 4 5
1 0 1
1 2 0 2
1 0 1
1 1 0 1
22 17 25 26 8 21 23 16 40 22 28 27 248 251 499
136 | Chapter Four
The data for Guinea slaves stand out from the rest: 958 baptisms with a
Guinea mother present, while only 35 Christian burials of Guinea women
were recorded over the same period (table 13 on pages 134–35). I noted
earlier the probability that many of these women were domestic slaves,
given their high incidence of marriage. But if even half of them were, or a
third, the marginal rate of Christian burials they attained detracts from the
common image of the protective, caring mistress who looked out for her
slave’s interests in the end. Whatever advantages were given a female
domestic slave over other types of servitude in eighteenth-century Rio,
obtaining a Christian burial was evidently not among them.
For Angola and Mina slaves, the numbers are more balanced, likely an ex-
pression of their more active organization into lay brotherhoods throughout
the first half of the century. At the same time, the Guinea identity—as dis-
cussed in chapter 3—was undergoing a sort of social and conceptual vanish-
ing act. The only substantial reference to forms of association between en-
slaved and freed Guinea comes from the observation from Vieira Fazenda
that Guinea were represented in the Brotherhood of São Domingos. But the
rarity of specific information about what Guinea did and where they did it
almost undoubtedly derives from the fact that although the so-called Guinea
heathen had been bundled under a purportedly geographic and cultural des-
ignation, for the convenience of Rio’s colonial slaveholding society, these
African peoples did not develop the mechanisms for local forms of sociability
that other, more prominent provenience groups constructed in the city of Rio
de Janeiro.
The diminishing number of obituaries through time corresponds with
what a range of other documents makes clear—even brotherhoods, not
only poor Africans on their own, faced increasing hardship in paying for
the Christian burials. It was common for brotherhoods to carry the corpse
of a deceased member to the door of a church, and to try to collect enough
charitable contributions there to pay for a ceremony. If su≈cient funds
were not raised, the corpse would often be left behind to be buried ‘‘for the
love of God,’’ a tactic identical to that used by poor or miserly slave owners.
At one point, the judge of the Brotherhood of São Domingos of the Convent
of São Francisco in the city of Salvador petitioned the crown to build a tomb
where brothers could be buried. Acknowledging the brotherhood’s lack of
means, he said the objective in creating their own space was to avoid having
to ‘‘leave our defunct ones in the entryways of churches to be buried for the
love of God.’’∑≥ All of this implies that belonging to a brotherhood was
perhaps the only resource available to slaves and freed blacks hoping to
escape the ignoble anonymity of mass graves for a socially recognized
Christian burial. The brotherhoods could attempt various measures to meet
this request, from collecting members’ contributions, to asking for public
donations, to writing a letter to the king; but clearly many times even these
e√orts were not enough to provide for a funeral for each member.
The testaments show that when freed blacks, and more commonly freed
black women, had some money saved at the time of death, they often left
most of it to their brotherhoods. If the philanthropic desires expressed in
those testaments were common knowledge in advance among other peo-
ple in the association, it raises the possibility that such individual benefac-
tors might have carried more influence or prestige while alive (as well as
ensuring a burial befitting their means). All the brotherhoods depended on
charity, from members and nonmembers alike. The Brotherhood of Our
Lady of the Rosário was one of the beneficiaries listed in Maria do Rosário’s
will. Maria, a freed slave, requested to be buried in the cemetery of the
Church of the Rosário attired in the habit of São Francisco—‘‘I ask the
brothers of my association to accompany my corpse to its final resting
place according to custom.’’ Among her possessions were listed four
slaves, although one ran away at the first opportunity and she specified that
another, a young boy, should be freed upon her death. The will also men-
tioned several pieces of gold jewelry, most of them chain necklaces, that
were to be found with her early master, and a cross adorned with seven
diamonds that was in hock. It was noted that she still owed eighteen patacas
for the purchase of some dressmaking cloth.
By the early 1740s, the Church of the Rosário provided shelter to at least
three brotherhoods who counted Mina among their members: that of Our
Lady of the Rosário, of the Glorious Santo Antônio of the Moors, and of Our
Lady of Lampadosa. This probably explains the high proportion of Mina
buried at the cemetery of the Rosary church. Of 79 Mina funerals (table 13),
32 were carried out there, versus only 8 at the cemetery of the Chapel of São
Domingos (table 12). In the second half of the century, with the inaugura-
tions of the churches of Santa Efigênia and Lampadosa, many of the Mina
who received funerals had them performed at these newer churches. The
138 | Chapter Four
brotherhoods would also continue to depend on the freed, elite class of

Africans represented by Maria do Rosário. Beyond their comparative finan-
cial power, these influential individuals could also serve the critical function
of mediating between a congregation of blacks—most of them Africans and
slaves, on the lowest rung of colonial society—and the colonial church that
recognized the congregation’s legitimacy.
provenience groups and their

processions and festivities
In Lisbon, the Board of the Brotherhood of Our Lady of the Rosário of the
Monastery of São Domingos, a group founded in the sixteenth century,
was composed of both administrative functionaries, and a court—with a
king, duke, count, and other nobles—representing a microcosm of the
hierarchical Portuguese royal court. The brotherhoods in Brazil were draw-
ing on this tradition when they elected, from within their own ranks, kings
and queens as their own ruling figureheads. Among black brotherhoods,
the practice was even more common than among white ones. The principal
di√erence between the Portuguese tradition and its adaptation in Brazil
was that in the colony, black brotherhoods had two separate boards: one
for executive administration, and one, often called a court, for the folia
(revelry).
When a brotherhood decided to create a folia, the court was charged
with organizing its o≈cials and subjects in a form based on well-known
hierarchies. If the court called itself a Reinado (Reign), it elected kings and
queens; an Estado Imperial (Empire) elected emperors and empresses.
Several times a year, the folia would parade through the city streets, dressed
in regal splendor, collecting charitable contributions for the festival of the
saint after which their church was named. Just as the regular members of a
brotherhood marched in public adorned with special capes and emblems,
the nobility of the folia were attired for the event: kings wore crowns
and mantles, and carried scepters; the queens wore crowns and elaborate
gowns; and the rest of the folia carried walking sticks, banners, flags, and
musical instruments (not necessarily of African derivation). The instru-
ments were used, in the case of African brotherhoods, to play for dances
and to support the singing of songs in di√erent African languages.
The kings and queens of the folia never walked in the open air without
the protection of an enormous parasol, borne by a liveried attendant; this
was another nod to the standard markers of prestige in the ancien régime.∑∂
A brotherhood and its folia bore identical insignia and color schemes on
their capes and banners, demonstrating that one pertained to the other.
Important processions and festival days reunited folias, marchers, specta-
tors, and often some of colonial society’s more distinctive personages. The
folklore scholar Luís da Câmara Cascudo noted that the processions of the
brotherhoods of São Gonçalo, Divino, and others could typically count on
the presence of high city o≈cials.∑∑
The folias were considered by government and church authorities to be
peaceful organizations. Still, occasional commotion was not unheard of,
and this was particularly true for the events associated with the elections of
kings and queens. Across colonial Brazil, di√erent black brotherhoods
chose kings representing various African nations. For instance, the com-
promisso of the Brotherhood of Our Lady of the Rosário in Recife men-
tioned an Angola king,∑∏ but in 1674 they elected a Kongo king and a Kongo
queen. At the Church of the Lampadosa in Rio de Janeiro, kings from
di√erent nations were elected at various times: a Rebolo-Tunda king
(1760s) and a Cabunda king (1811).∑π
Slave owners were known to enhance the pomp of these parades by
loaning jewelry, costumes, and other decorations to the elected kings and
queens for their promenade to the church, where the vicar was waiting to
crown them. In 1729, the Brotherhood of the Rosário in Salvador was
prohibited from participating in public events because of what were termed
their excesses. In 1786, they petitioned the crown for permission to parade
for a few days with all their nobility, dances, and songs (which, it was
specified, were in the ‘‘Angola language’’).∑∫ As a rule, though, whatever
excesses occurred during the parades appeared to take place outside the
folia proper. The courts and their retinues took most public events se-
riously and maintained a solemn comportment. What happened afterward
might be a di√erent story, and perhaps that was why many of the brother-
hoods agreed in principle in their founding statutes to return from any
festive event (including funeral processions) in quiet order. The informal
disbanding of a parade could easily lead to bumptious exchanges between
di√erent groups, or between marchers and spectators, as well as to noise,
140 | Chapter Four
music, or dance deemed threatening or improper; confusion; and fights—

particularly given that strong drink was nearly always in abundance.
In Rio de Janeiro, the São Domingos Field was a favored route for black
folias in the second half of the eighteenth century. In the Chapel of San-
tana, the folias often paraded with the Portuguese motif of the Império do
Divino (Empire of the Divine), while those at the Church of the Lampadosa
were known for their festivals for King Baltazar, associated with the King
of Kongo. At the Church of Santa Efigênia, the brotherhood’s folia had for
their theme the Império de Santo Elesbão (Empire of Santo Elesbão),
amplifying the tiers of royalty to include several kings serving under an
emperor. One notable aspect of the various folias is that they commonly
visited each other’s churches and paid homage to each other. For many
years, during the festival of the Chapel of Santana, the folia from Lam-
padosa marched to Santana to salute the folia of the Império do Divino.
This again suggests that the underlying connections between the members
of di√erent brotherhoods, most of them Africans and slaves (or former
slaves), were at least as strong as the individual loyalties to particular
brotherhoods.∑Ω
To attempt a rough family tree for these folias, it seems that the brother-
hood at Lampadosa was the first to adopt the processional traditions
associated with the Church of the Rosário. The folias of Santana and Santo
Elesbão were both based at the Church of São Domingos. The brother-
hoods of Rosário and São Domingos date back, as noted earlier, to the
seventeenth century. They were located at Rio’s first See church, São Sebas-
tiáo, where the increasing ostentation of their folias appears to have been
one of the reasons for their expulsion. It is possible that even at that time
the Brotherhood of the Rosário elected kings and queens, while the Broth-
erhood of São Domingos elected emperors and empresses, and these two
practices were passed on to the associations that came later.
In 1763, the year that Rio de Janeiro took Salvador’s place as the capital
of the colony of Brazil, the aforementioned Rebolo-Tunda king waited to
assume his o≈ce as the leader of the brotherhood’s folia at the Church of
the Lampadosa. This king was a slave who happened to belong to Count da
Cunha, the first viceroy. Notwithstanding several loud voices of opposition
raised against the folias in Rio’s elite social circles, da Cunha went ahead
and authorized the public crowning of his slave, as well as all the ‘‘typical
celebrations’’ that would be expected to precede it. But the city’s high
criminal judge moved against da Cunha and outlawed the ritual, arguing
that the folias presented too much risk of physical danger and general
chaos to be permitted.∏≠ The brotherhoods’ parades highlighted a murky
area of the Constituições Primeiras, the ecclesiastical rules written in Bahia and
applied throughout Brazil. Those regulations specifically authorized re-
ligious processions which were ‘‘honest and decent,’’ and made mention
of the Império do Divino association as an example to be followed. How-
ever, they expressly prohibited ‘‘disorder’’ and ‘‘immoral acts,’’ with no
further interpretive detail o√ered. The problem was that in practice, the
folias were usually viewed as decent on the way to the church or funeral,
but characterized as indecent the moment the parade was over and the
members were going home or wandering around. In the absence of a viable
principle, the legal status of each event was the product of local negotia-
tions between the group and the o≈cials charged with deciding the matter.
The Brazilian historian Martha Abreu, analyzing the festivals of the Divino,
found a 1780 document from the Oversea Council that recommended
tolerance with the dances of the ‘‘blacks,’’ and with the licentiousness that
was presumed to be a natural part of those practices.∏∞
Although the folias are represented in several colorful images by the
eighteenth-century Portuguese artist and draftsman Carlos Julião, they
appear not to have been registered by other artists. Later, in the nineteenth
century, the French artist Debret wrote a short account of them:
We do some justice to the history of the black brotherhoods by recalling that,

with the arrival of the Portuguese court in Rio de Janeiro, there was a
complete prohibition on the blacks’ costumed revelries, which were under-
taken at certain times of the year to remember their homelands. These
entertainments were deemed too clamorous and intemperate. The prohibi-
tion applied equally to other festive behaviors, quite tranquil by contrast but
still performed in costume, which the blacks had introduced into local
Catholic custom. This is why it is only in other provinces of Brazil, but not
here in the capital, that one can still observe the election of a king, a queen, a
captain of the guard, and other nobles from among a company of blacks.∏≤
His statement implies that during his stay in Brazil (1826–31), most of it
spent in Rio de Janeiro, Debret never personally witnessed the parades or
142 | Chapter Four
other festivities. What happened during the commemorations that Debret

suggested were celebrations of the black Africans’ homelands? There is
very little evidence of their cultural background, and much less that could
show how such practices and attitudes might have informed the nuances of
the folias. The one thing that is clear is that whatever happened ‘‘extra-
muros,’’ beyond the wall that separated the urban core of Rio from its
wilder environs, was not observed by the artists and writers who left ac-
counts of the city’s social and cultural life. If Debret had seen any of these
events, dedicated and prolific artist that he was, one might assume he
would have sketched them.
Another aspect of the festivals and processions that should be empha-
sized was the distinctive participation of men and women. In a certain
sense, the folias brought the di√erentiated gender roles at work in the
internal operations of the brotherhood into the light of day. The brother-
hoods were a nominally masculine space in which women’s participation
was typically viewed as necessary, but simultaneously as a potential threat
to men’s power that had to be constantly scrutinized and contained. Be-
yond any material resources that women might provide, a unique dimen-
sion of women’s identity was their association with witchcraft. Available
sources on this theme emerge mostly from inquisitorial records, and until
now the evidence does not provide enough information for a persuasive
broader analysis of how witchcraft intersected with the organization of
ethnic groups and other social associations. But there are suggestive de-
scriptions of festive gatherings called calundus or batuques, such as the one
in this 1780 formal correspondence from Martinho de Mello e Castro to the
king of Portugal about tensions over the festival of the Church of the
Rosário in Recife:
I received the notice dated 09 June from Your Majesty in which Your Majesty
ordered me to report about the letter from the Holy O≈ce and the Governor
of Pernambuco, and to examine the situation in Recife with respect to
superstitious dances (a concern of the Holy O≈ce) as well as dances that
may not be the most saintly and yet may not be worth punishing (as the
Governor maintains). The latter form refers to the celebrations of blacks
who are divided into nations, and each group with its own instruments,
dances and gambols with harlequins, or writhes about with diverse corporal
movements that are perhaps not as innocent as those characterizing the

fandangos of Castela, or the fofas and lunduns danced by whites and pardos in
Portugal. These are regrettable yet do not appear dangerous. But what I am
told are the most despicable dances, the ones perhaps needing swift prohibi-
tion, are those that the Mina blacks perform in their own spaces, hidden
away from the public eye in houses or fields, led by a female black who
stands before altars that are covered in idolatrous objects. The people wor-
ship live goats, as well as clay likenesses of goats. They rub themselves with
diverse oils and blood from a chicken they first slaughter for the purpose,
and eat corn cakes over which bizarre blessings have been uttered, phrases
which make the rustics believe that eating the cakes will bring good fortune
in wealth or romance . . . I believe thus that the Holy O≈ce speaks of certain
dances, and the governor of others, and I cannot persuade one authority to
disallow what another approves. This is my statement, and Your Majesty, in
the shining lights of his wisdom, will resolve the matter to its most just
solution.∏≥
The rather harried distinction between dances that were ‘‘superstitious’’

versus ‘‘not the most saintly’’ demonstrates the flexibility (and complexity)
of the criteria used to determine what was socially legitimate behavior.
Houses often received particular attention from authorities because of their
capacity to shelter religious practices, not merely dance, from outside
observers. The reference in this letter to Mina blacks hiding out among
private alters, worshipping goats, and sacrificing chickens presents a de-
liberately extreme case of the practices that were thought prudent to pro-
hibit. Most of the cases that called for o≈cial deliberation lay somewhere
between the poles of a public parade and a purportedly pagan ceremony
(replete with idols and blood sacrifice).
The city of Rio de Janeiro experienced substantial changes in infrastruc-
ture in the 1770s, during the administration of the viceroy Marquês de
Lavradio. He implemented substantial projects, such as paving roads, con-
structing a new slaughterhouse, draining and filling over with earth the
lagoon of Pavuna, and the clearing of many new footpaths. One of his
initiatives that would a√ect daily life for many people in Rio was the
relocation of the slave market from the Rua Direita, a principal street in the
business district, to a distant region of the city commonly referred to as
144 | Chapter Four
Valongo.∏∂ Now, Africans recently arrived at port were no longer kept in the
streets in the center of the city awaiting buyers, but were removed to this
locale; and slaves sold to work in Minas Gerais or other places outside Rio
tended thereafter to gather in Saint Domingos Field—where the Marquês
claimed they had ‘‘all the comforts’’—before they were marched out.∏∑ It
was also during the Marquês’s rule that the idea was floated to demolish
the small chapels constructed by black brotherhoods in the territories
beyond the boundaries of the (unfinished) city walls. Once distant and
considered irrelevant, those lands were newly prized in the face of Rio’s
urban expansion. Documents show that this plan was supported by the
viceroy and the bishop, as well as by the black brothers of the Rosário, who
stood to benefit. The idea was to simply cut o√ support for maintaining
and reforming those chapels until they deteriorated beyond functional use;
the time involved would not be considerable, given their already poor
condition. Once e√ectively condemned, they could be taken over by the
church (a provision included in the Constituições Primeiras),∏∏ which was
empowered to order their demolition. In such a case the several black
brotherhoods would be transferred to the Church of the Rosário, which in
turn would receive more generous dispensations and could accumulate
funds to pay for its own renovations. The Brothers of the Rosário sent a
missive to the king arguing their case:
Various black brotherhoods occupy undignified and indecent little chapels,

which barely qualify for the name, among them the Brotherhoods of Mercês
and São Domingos, São Philip, São Tiago, the Menino Jesus, Santa Efigênia
and Santo Elesbão, Our Lady of Lampadosa, São Mateus, São Benedito and
Santo Antônio, Bom Jesus do Cáliçe, Our Lady of Belém and Santo Antônio
of the Moors. It would well serve Your Majesty to annex them, and in
demolishing them and their gruesome cemeteries Your Majesty would fur-
ther be providing a great service to God. The brotherhoods would be wel-
comed here, and the small sums they can o√er to this church will bring to a
happy end our desperate need for support to finish the many works to
improve this house of God.∏π
In the opinion of the viceroy Marquês de Lavradio, these chapels were

the sites of ‘‘base and indecent acts,’’ which were eagerly performed by the
‘‘depraved, vile people’’ who composed the congregations. This allegation
was adduced as reason enough for the chapels’ destruction.∏∫ But the list of
black brotherhoods compiled in the petition above provides a sense of how
the fields beyond Rio’s Vala trench must have looked after 1750 when the
brotherhoods departed the principal churches of the Rosário and of São
Domingos to create their own spaces. The region had grown full of small,
very poor chapels, surrounded by cemeteries in precarious conditions. It is
also possible to conclude from this list that several chapels resisted the
pressure and ultimately transformed into more substantial churches (São
Domingos, Santa Efigênia, Lampadosa, and Bom Jesus do Cálice). The
others seem to have disappeared quickly, and some of these produced no
further documentary evidence after being named in the petition for de-
struction written by brothers of the Rosário. Among these was the Brother-
hood of Santo Antônio of the Moors, whose actual physical site I could
never locate.
5 | constructing a religious norm
It is increasingly clear from a range of scholarship (not only this
\ book but much recent work in the history of slavery) that the
social formations, alliances, and institutions adopted by urban
slave populations are far more complicated than have previously been
imagined.∞ Research on African slaves, in particular, is showing how
prominent and diverse this complexity was across the colonies of the New
World. Underlying this scholarship is one basic perception: although so-
ciety presents established rules and limits for identity construction and
group organization, individuals learn to maneuver dynamically within the
circumstances of each particular case to create a strategic balance between
contestation and acceptance. That is, there exists neither an absolute deter-
minism of social regulation on human action, nor the boundless autonomy
of individual volition.≤ On the one hand, colonial slave society imposed
rigid constraints on the pretos; on the other hand, this same society opened
to them a myriad of other, often quite unintended pathways to distinction
and dignity across a rich cultural terrain. And as I have argued in the
preceding chapters, the principal route to social prominence for African
slaves—and particularly for freed African slaves—was membership in a lay
religious brotherhood.
Devotional processions and funeral parades o√ered important oppor-
tunities for fraternal orders of pretos and pardos to be represented publicly
within the dominant religious hierarchy, even if they were typically left to
bring up the rear in Catholic ceremonial parades. The brotherhoods’ par-
Constructing a Religious Norm | 147
ticipation was nonetheless coherent and meaningful despite being rele-

gated to the last ranks. I take the lay brotherhood to be a ‘‘modern’’
institution (very di√erent from the medieval orders), whose internal orga-
nization is based on the same models and principles of ancien régime
hierarchical societies and European absolutist states. In that sense the
‘‘court society’’ analyzed by Norbert Elias can be understood to embrace
not only the state but the church as well. In his discussion of the architec-
ture and social organization of the French court, Elias observed that almost
every feature one encounters in majestic scale in the palace of the king, one
finds in small scale in the house of the nobleman.≥ The same might be said
in relation to church ritual: almost everything one witnesses in the gran-
deur of Vatican spectacles, one sees in reduced form in Brazil’s many local
bishoprics striving with fewer resources to convey the full symbolic power
of Catholic pomp and pageantry. The poor brotherhoods’ exertions to
reproduce the ceremonial patterns and gestures of prestigious, better-
endowed religious groups—such as the Santíssimo Sacramento—demon-
strate this process of hierarchical influence.
It should also be remembered that in eighteenth-century Brazilian so-
ciety the abolitionist mentality that coalesced in the next century did not yet
exist. The concept of liberty was not seen as a universal value that could or
should be extended to everyone, and among slaves freedom was more
immediately comprehended and desired. Freedom in this context was an
individual victory, represented by the definite but often intangible advance
of one step up the ladder of the social hierarchy. Even at the moment of
buying his or her freedom, the slave, upon presenting the agreed-upon
sum to the master, would have to ask the master’s permission; and the
master was legally and socially understood as ‘‘conceding’’ freedom to the
slave. And freed slaves could be reenslaved through a variety of mecha-
nisms in colonial society. Still, a slave who saw no hope of emancipation
through broad legislative means or a master’s act of beneficence could seek
out the small personal freedoms a√orded by daily life, in such areas as were
permitted or ignored by authorities. In the world of the urban slave, these
modest spheres of self-determination typically included the choice of con-
jugal partners, the chance to work for oneself to build up savings, the
ability to frequent recreational gatherings, the possibility to traverse and
learn from the city (even to find a place to live outside the purview of the
148 | Chapter Five
master), and the opportunity to join a lay brotherhood. Within the stric-
tures of slave life, brotherhoods provided special access to the experience
of liberty, of social recognition, and of administrative activity. In the city of
Rio de Janeiro, the Brotherhood of Santo Elesbão and Santa Efigênia was
one of these rare spaces where the Mina slaves could construct their own
sociability.
In this era, brotherhoods were at once a locus of the exercise of liberty
among black slaves, and a medium of indoctrination for African peoples
within Portuguese society. They were among the few venues for association
that the Portuguese state tolerated among the pretos, and in fact from the
point of view of the church, they had a genuine appeal. Far from being
disparaged as misguided vestiges of medieval religiosity, the brotherhoods
were understood to o√er a congenial space for collective indoctrination as
well as for encouraging the arduous sacramental obligations prescribed by
the Council of Trent. An examination of the brotherhoods’ statutes and
other internal documents reveals that in one form or another, all of the
groups traced their spiritual and formal lineage back to the Brotherhood of
Our Lady of the Rosário of the Monastery of São Domingos, founded in
Lisbon in 1565.
The two fundamental pillars that have traditionally supported lay re-
ligious fraternities are devotion and charity; or, in the words of Russell-
Wood, ‘‘propagation of doctrine’’ and ‘‘social philanthropy.’’∂ Documents
suggest that in the Brotherhood of Santo Elesbão and Santa Efigênia,
greater emphasis appears to have been given to devotion than to charity. In
fact, the scope of charity is interpreted narrowly enough within the docu-
ments, addressing primarily the group and its members rather than the
larger community around it. This recalls John Bossy’s definition of ‘‘medi-
eval charity’’ as a type of beneficence in which giving money is less es-
teemed than performing ‘‘acts of physical charity’’ such as providing food
and clothing, visiting the sick or imprisoned, and participating in fu-
nerals.∑ The idea of medieval charity reappears in a highly particular way in
the compromisso of the Brotherhood of Santo Elesbão and Santa Efigênia:
the principal charitable act is to attend the funerals of group members.
Even the giving of alms, when the total sum exceeds a stipulated amount, is
considered an act of devotion but not of charity.
It is significant that in the compromisso of this Brotherhood, the word
charity appears only one time. Chap. 11 declares that the brotherhood was
to unanimously appear, ‘‘incorporated,’’ on the occasion of a funeral of one
of the group members.∏ This obligation, almost a contractual agreement in
tone, is considered an act of charity. Such a definition of charity stands in
stark contrast to the notion that philanthropic acts are practiced to benefit
others without the expectation of retribution or quid pro quo. But charity,
in the philanthropic sense, could occur between brotherhoods of pretos. In
these instances, works of beneficence were administered by a brotherhood
to other needy people and not to group members themselves. For example,
the Santa Casa de Misericórdia included in its compromisso a summary of
fourteen such acts of charity—seven spiritual, seven physical.π They are all
philanthropic insofar as they are intended to serve the poor. However, the
compromisso of Santo Elesbão and Santa Efigênia makes no mention of
philanthropy, only of charity—and even then, it is charity tailored to the
needs of the group itself. With its stringent, contractual language, the
compromisso carries an implicit threat of exclusion from the brotherhood
if precepts are broken.
In this document, whose first chapters were composed in 1740, ‘‘devo-
tion’’ is articulated as an individual commitment that each brother needed
to make to the group (unlike ‘‘charity,’’ which always refers to a group
activity). It appears in numerous chapters, always in the context of encour-
aging a member, ‘‘for devotion,’’ to bring even more to the brotherhood’s
endeavors than is called for in the compromisso. Thus, to show ‘‘zeal and
devotion,’’ a brother can o√er ‘‘munificent alms’’ (chaps. 5, 12, 22, and 5 in
the folio). Also for ‘‘devotion,’’ white, pardo, and ecclesiastical brothers
are accepted into the fraternity (chaps. 10, 11). One chapter (26) suggests a
link between devotion and celebrations, but that discussion ends with a
remark that the expenses of sacred festivities not be paid from the brother-
hood’s co√er.
Although it is customary to view the twin ideals of devotion and charity
as integral to the constitution of all lay brotherhoods,∫ in the associations
studied here these values were subjugated to preoccupations of a practical
nature, relative to the internal needs and organization of each group.
Faithfulness to the saints, made manifest in acts of devotion (alms), en-
abled initiates to attain general membership in the brotherhood but did not
make them eligible to take on other, more formal responsibilities. Theoret-
150 | Chapter Five
ically, a white person could have been accepted as a devotee by a black

brotherhood, but his membership would have been limited to the fraternity
as a social body. Ethnic groups or provenience groups were a di√erent
category, and would have been o√ limits. In other words, devotion was the
necessary point of entry to the brotherhood, but it alone did not create
mechanisms of ethnic or group identification within the brotherhood.
Devotion per se obviously mattered because it substantiated both the com-
promisso (the contractual spirit uniting the brothers through the commu-
nal enactment of devotional practice) and the institutional profile of the
brotherhood. In that sense, numbers were important. The more members,
the greater recognition a brotherhood received from religious authorities
and from society at large, and for that simple reason the standards or rules
for a≈liation were often wide. From the perspective of the pretos, either
slave or freed—and whatever their religious orientation—participation in
lay fraternal orders was desirable on purely secular grounds: such groups
presented the only social space in colonial Brazil that was protected from
unwarranted interventions by the dominant institutions of society. The
state had to respect a brotherhood’s organization and activities (since these
were presumably sanctioned by the church); similarly, the church had to
recognize a brotherhood’s legitimacy as a lay association of Catholic faith,
as long as its compromisso had been approved. And the brotherhoods of
white men, who unproblematically excluded blacks on social grounds, still
were bound by the rules of the church to respect black associations.
The goal of this chapter is to lay a framework for understanding how the
Brotherhood of Santo Elesbão and Santa Efigênia articulated its identity
between two poles: the formal Portuguese Catholic concept of lay religious
organizations, and the historical contours of daily life in a colonial city
based on slavery, peopled by a diverse composition of Africans and their
descendants, as well as by Europeans. (Of particular help to me in this
undertaking were documents containing regulations and correspondence
exchanged between the brotherhood and the Tribunal of Conscience and
Orders in Portugal, which was the sphere of Portuguese administration
most directly involved in enforcing the Padroado.) In the second half of the
eighteenth century, the Church of Santo Elesbão and Santa Efigênia housed
a number of smaller brotherhoods. Some constituted their identity around
ethnic or provenience groups, such as the Mahi Congregation or the Mina
Congregation; others, while involving ethnicity and provenience, were

based on devotion to particular saints or other entities, such the Almas do
Purgatório (Souls of Purgatory) or Our Lady of Remédios (Our Lady of
Remedies).Ω
The Brotherhood of Santo Elesbão and Santa Efigênia had a relatively
low profile among Rio de Janeiro’s black brotherhoods, a condition that
persisted into the nineteenth century (and more recent times as well). Trav-
elers or correspondents visiting the city of Rio de Janeiro rarely made
mention of it;∞≠ nor does it appear in the documentation of the Tribunal of
Conscience and Orders after 1808, when that institution established its
base in Brazil.∞∞ In general the brotherhood seems to have interacted only
minimally with the various o≈cial institutions of the era, which reduced
the possibility of discovering hidden veins of complementary documenta-
tion. In light of the paucity of substantive historical documents from other
sources, an analysis of compromissos is arguably the most promising
approach for comprehending the discursive development of this brother-
hood’s identity, as well as the tangible practices (quotidian, festive, or
otherwise) involved in negotiating and expressing it.
approval of the brotherhood of santo elesbão

and santa efigênia
The ecclesiastical procedures to organize and gain approval for the broth-
erhood began in 1740. The documentary trail of this process had to be
pieced together through persistence and luck, because the brotherhood’s
name was not uniformly recorded in the beginning. The text of the first
proposed compromisso refers merely to a ‘‘fraternity of the saints of our
brotherhood,’’ also calling them ‘‘glorious saints,’’ without specifying
which saints these were. Nor was the name of the brotherhood itself
included, which would have conveyed the names of the patron saints. It is a
separate but related document, prepared by Bishop D. Antônio de Guada-
lupe to authorize the creation of the ‘‘Brotherhood of Santo Elesbão and
Santa Efigênia,’’ that makes the connection. Another anonymous docu-
ment from the period refers to the Brotherhood of the Glorious Saints
Elesbão and Efigênia. A further variation was found in the insignia of the
brotherhood itself, which bears the inscription ‘‘Venerable Brotherhood of
152 | Chapter Five
Santo Elesbão and Santa Efigênia.’’∞≤ Nowhere was it explained how these
two saints came to be paired; usually orders formed around the devotion of
a single saint. In other cases of multiple devotions within the same brother-
hood, however, such a brotherhood typically emerged from the fusion of
several smaller, informal devotional associations that shared some com-
mon ground, spiritual or secular.
Early in this process, the vicar of the parish of Candelária, the parish
where the group hoped to establish their brotherhood, expressed his sup-
port. He viewed it as a noble endeavor that would encourage other local
blacks to devote themselves to dark-skinned saints under the purview of
the church. The vicar did raise a skeptical eyebrow at what he felt was the
surprisingly large number of lay orders in the area associated with Africans
of the Mina nation. But he concluded that because this petition for a new
brotherhood included a substantial list of supplicants, it deserved the
attention and approval of higher authorities.∞≥ The response by other re-
gional ecclesiastics to a petition from Bahia in 1765 begins to suggest the
absence of any overarching guidelines for evaluating the worthiness of
proposed black brotherhoods. In that case, church o≈cials opposed the
constitution of the Fraternity of Senhor Jesus dos Martírios (Lord Jesus of
the Martyrs) of Vila de Cachoeira, which had been formed by Africans of
the Jêje nation. The o≈cials did forward the brothers’ petition on to the
Tribunal of Conscience and Orders in Lisbon, but they appended a note
suggesting that the petition be dismissed because the Jêje ‘‘are too recently
and imperfectly withdrawn from their pagan traditions in Africa, and are
tenacious in their superstitions.’’ For this reason, the note continues, it
would be more ‘‘convenient’’ to leave this group subject to the oversight
and discipline of the local parish.∞∂
The two predominant themes of the compromisso of the Brotherhood of
Santo Elesbão and Santa Efigênia are the administration of death (and the
treatment of the dead), and the management of resources that were dedi-
cated to festivals, funerals, and acts of service or assistance to the brothers.
A close reading of the entire document suggests that the first twenty chap-
ters were probably an early version of the complete text, while the ensuing
chapters were additions or alterations annexed to the main corpus during
the long petition process (1740 to 1764). The first approval notice, in 1740,
refers to twenty-four chapters, which indicates that chaps. 21 through 24
must have been composed soon after, or on the occasion of, the brother-
hood’s formalization in 1740. Of the twenty-four chapters the brotherhood
submitted to the Tribunal of Conscience and Orders, twenty-three were
approved and one rejected.∞∑ Over the next ten years, eight more chapters
were added by the brothers, and a 1748 correspondence from the bishopric
refers to the o≈cial acceptance of four more chapters into the compro-
misso approved in 1740.∞∏ Monsenhor Pizarro mentioned two provisions of
the compromisso, one dated 24 January 1747 and the other 28 August 1754,
although he did not describe their contents.∞π In 1764, five more chapters
were sent to Lisbon, numbered separately from the earlier chapters and
bound together.∞∫ Soon after, the total of thirty-seven chapters written by
the brothers and approved by the local bishop were sent to Portugal for the
king’s consideration.∞Ω Included with them was also an undated acrescenta-
mento, or addition, probably composed before 1764.≤≠ Royal approval was
granted to the brotherhood in 1767.≤∞
The first chapter of the compromisso is a formal introduction to the
brotherhood, indicating its judge, scribe, ‘‘and other humble brothers’’ as
authors of the document. It states that the group was located in the Church
of São Domingos,≤≤ and justifies its presentation of compromisso as a way
to spread the faith and increase its ranks of ‘‘loyal Christians.’’ Rather than
listing the contents of each of the ensuing chapters, in the following
discussion I consider the major subjects and concerns of the whole docu-
ment. I refer to the original numeration or dates of the chapters when
possible to indicate how specific changes in the brothers’ arguments devel-
oped over time—changes which would have derived, in whole or in part,
from the negative reactions of ecclesiastical authorities to particular chap-
ters, ideas, lines, or words in the proposed statutes.
festivals
The annual festival dedicated to the patron saint or saints of each brother-
hood represented the high point of the yearly religious cycle, and was
celebrated sumptuously for days, even a week or longer. The inside of the
church was decorated with icons, candles, flowers, and bunting, while
outside on its grounds a makeshift village of tents and stands sprung up to
sell food and drink, and o√er ra√le giveaways of livestock. Musicians
154 | Chapter Five
played and sang, and dancers twirled and leapt, all to celebrate the event
and attract passersby. The apogee of the festival was the day of the saint,
and it was marked by quiet, solemn rituals as well as animated festivities.
The Brotherhood of Santo Elesbão and Santa Efigênia wrote in their com-
promisso that they celebrated their devotion on the day associated with
Elesbão, 27 October (although, sometime in the nineteenth century or early
twentieth century, they changed to Efigênia’s day, 21 September). Well
before that time, the brotherhood would begin organizing the festival and
gathering funds for its realization. Within the group there were obligatory
contributions (specified as charity in the statutes) as well as voluntary ones
(given for devotion). Under the former category were membership fees and
annual dues, as well as supplemental contributions for specified expenses
(these o√erings were called esmolas, and were considered donations, not
alms). Because the operating expenses of a brotherhood often dwarfed the
sums collected from within the group, and indeed certain uses of a broth-
erhood’s internal funds (such as the purchase of food to serve during
festivals) were prohibited by the church, special processions to beg for
donations from the public were authorized by the bishop. These were
usually undertaken on city streets, with the image of the saint carried by
some of the brothers alongside a platter or sack in which to store the
collected donations. A brotherhood that asked for money from the public
in this way without the bishop’s permission would be severely punished.
Lay brotherhoods begging for money in the streets was already a long-
standing tradition in the 1740s. It dates back at least to 1549, when King
João III allowed the black brotherhood of the Rosário to collect financial
donations on the streets of Lisbon (some sixteen years before the brother-
hood had formalized its compromissos). In the Constituições Primeiras the
practice was deplored and dismissed as an ‘‘archaic style’’ of needy brother-
hoods to get help. But the writers of the Constituições Primeiras were actually
in favor of allowing brotherhoods to beg for money, with the proper per-
mission and regulation. Their foremost preoccupation was the juxtaposi-
tion of saintly Roman Catholic images with the African ‘‘archaic style’’ of
vigorous dances and syncopated music played on drums and other percus-
sion instruments. They evidently decided that it would be easier to prevent
this untoward mingling of traditions by banning the European element,
because their work contains a prohibition on taking to the streets the
saints, images, or even sacred paintings when a brotherhood was going out
to beg.≤≥ This is rather more surprising because as a historical practice,
using music to accompany and enhance the request for charitable dona-
tions, was associated with both whites and blacks (i.e., it was not imme-
diately assumed to be some kind of pagan abomination introduced by the
Africans). A common theme in Brazilian lithographs from the nineteenth
century is a group of white men carrying images of saints and asking for
public donations, surrounded by black musicians playing drums and
horns. But black brotherhoods, particularly when the majority of members
were African, carried more of a cultural threat. They sang their songs in
African tongues; played their drums, rasps, marimbas, and other percus-
sion; and dressed in approximations of their native African attire. All of this
was legal in the church-sanctioned context and only needed prior authori-
zation from the local authorities.
Elections for positions on the Mesa (brotherhood’s board) were usually
held in conjunction with the annual festival of the saint. The night before
the saint’s day, the judge, scribe, procurator, and chaplain would convene
the other members at the church for the election. The o≈cials would
present the judge with a list of three names of contenders for each position,
and ‘‘in secret the Judge would ask each of the Brothers which of the three
proposed subjects the Brother would elect.’’≤∂ Some other brotherhoods
carried out elections through voting with fava bean pods. Each elector
would select a candidate by depositing a bean pod in the candidate’s bag;
the candidate who received the highest number of pods was the winner.
Whatever the method, whether votes were conveyed by whispers or bean
pods, the major judges would tally them and communicate the results to
the brotherhood (honestly, it was hoped). In the case of a tie, the judge had
the discretion to choose which of the candidates was better suited to the
brotherhood’s needs.≤∑
Within the brotherhood, the ultimate power authorities held to resolve
issues important to the group was a characteristic of broader social rela-
tions in the ancien régime. The judge and other administrators viewed
overt questions from brothers regarding o≈cial decisions as a grave threat
to order, that is, far more than as insubordination. The brotherhood’s
board was formed of twelve elected males. Of these, five held o≈ces:
judge, procurator, scribe, treasurer, and andador (courier). The roster of
156 | Chapter Five
o≈cials was di√erent in the older Brotherhood of Our Lady of the Rosário
of the Monastery of São Domingos in Lisbon. There, all twelve of the
elected board members held a position, with six considered higher and six
lower. The high positions included judge, procurator, scribe, treasurer,
and two administrators. The low positions were king, prince, duke, count,
marquis, and cardinal; these did not entirely disappear from Brazilian
brotherhoods but were mainly incorporated in the folias.
If the saint’s festival was the overall responsibility of the judge, it fell to
the procurator to plan, organize, and ensure su≈cient donations to fund it
in many brotherhoods, including that of Elesbão and Efigênia. Through
the festival the procurator was to ‘‘strive to grow and preserve the Brother-
hood,’’ and an important aspect of the arrangements was the visual impact
of the decorations inside and entertainments outside the church. Every-
thing needed to look clean, bright, sturdy, well made, and inviting.≤∏ The
rituals leading up to the saint’s day festival began earlier, often nine days
before, when a novena (performance of repeated orations at the same place
and time, for nine days) was begun. The novena was followed by the
elections, and then the main celebrations on 27 October. During the morn-
ing hours a closed mass was held, and the souls of deceased brothers were
prayed for. Following that was a Solemn Mass, with the brotherhood
dressed in their ritual finery. Once the second mass ended, the group
paraded out on a public procession through the streets of the city.
The compromisso does not provide details on the composition or pro-
ceedings of the celebratory cortege, perhaps because they are described in
detail for the funeral processions. In this the brotherhood may have been
following closely the regulations of the Constituições Primeiras, which ex-
pressed greater interest in the nuance of funeral processions. At any rate,
the overall structure was probably similar. In front of the rest of the march-
ers strode the judge, carrying the tall metal sta√ symbolizing his o≈ce. He
would typically be accompanied by the chaplain. Behind them, brothers
dressed in their ceremonial cloaks, called opas, carried a cross and images
of the patron saints.≤π Then came the rest of the procession, with one or two
members of greater prestige carrying a banner with the insignia of the
brotherhood; there would be several flags, and various additional forms of
adornments and alfaias (ritual objects and ornaments) used by all brother-
hoods in such contexts. In the citywide processions sponsored by the
bishop that united multiple brotherhoods, such as the Corpus Christi, each
brotherhood would make up one flank of the long parade, carrying again its
saints, crosses, and other decorations. According to custom, the black
brotherhoods took up the rear, from the richest and most prestigious to the
most humble. The existence of internal hierarchical divisions (correspond-
ing to rank or social position) within processions of the Brotherhood of
Santo Elesbão and Santa Efigênia is suggested by chap. 13 of the compro-
misso. At the end of the parade, the cortege would return to the church,
replace the saints on the altar, and put away all the other items that had been
brought out to enhance the brotherhood’s public presentation. Finally, the
annual event wound down with a large afternoon meal, and informal ac-
tivities and camaraderie involving dances, auctions, and lotteries.
A saint’s festival was thus long and diverse, with di√erent manifesta-
tions occurring both inside and outside the church walls. As a collective
event, the festival began inside the church (with the two masses), and
passed outward to the church grounds and beyond to the city; it also ended
inside the church with the dinner gathering. Were contemporary observers
concerned with a black brotherhood apparently taking over part of a
church for period of hours or days? Even in the text of the Constituições
Primeiras, there is ambiguity over what types of behavior are to be allowed
within a church: according to those ecclesiastical guidelines, ‘‘indecent’’
acts were prohibited, while ‘‘honest’’ ones were accepted. At the dinner,
food was abundantly supplied and probably voraciously consumed, while
diners chatted, sang, played instruments, and danced.≤∫ Devotion to a saint
was the festival’s justification, but such activities within a church blurred
the lines between the sacred and profane. Occasionally accusations of
improper behavior were made, but because the tone of the event was
religious and worshipful, such momentary scandals were usually resolved
quickly or forgotten and the festival was allowed to proceed as if nothing
had happened.
For colonial authorities, one way of trying to minimize such distur-
bances before they occurred was through scrutinizing and regulating the
festival preparations. There was likely some of this strategy in Bishop D.
Antônio de Guadalupe’s position towards the Brotherhood of Santo Eles-
bão and Santa Efigênia with respect to what he called ‘‘licit’’ (versus illicit)
expenses. Since his arrival in 1725, Guadalupe had fought for more strin-
158 | Chapter Five
gent observance of the Constituições Primeiras and of Catholic rituals and

precepts in general. The bishop wanted the brotherhood’s compromisso
(he drew special attention to chap. 2) to agree clearly with the regulation
that only ‘‘licit’’ purchases could be made with the funds collected within
the order. Those resources—based on dues, internal contributions, and so
on—were expressly forbidden from going toward ‘‘dinners and other such
things.’’ In other words, the success of the dinner (indeed, its very exis-
tence) depended on the brotherhood’s ability to obtain donations from the
public to underwrite it, through processions undertaken for this purpose,
overseen by the church. A brotherhood with a reputation for disorderly or
immoral acts within its church during festival days of previous years would
have far less success collecting funds from a suspicious citizenry, not to
mention gaining the church’s permission to take to the streets to beg in the
first place.
mortuary rites
Lay orders and devotional practices associated with death were common in
the eighteenth century. A prominent example is the image of Christ de-
ceased, lying in a co≈n, that was hidden from view throughout the year
except for Holy Week, when it was featured in the Procession of the Senhor
Morto. The Brotherhood of Our Lady of the Rosário also engaged themes of
death and resurrection, and one derivation of this order was the Brother-
hood of Our Lady of Boa Morte (Our Lady of the Good Death). A popular
saint throughout colonial Brazil was São Miguel Arcanjo (Saint Michael the
Archangel). His role, receiving souls of the departed in heaven, made São
Miguel the focus of considerable attention among diverse sectors of the
population united in their preoccupation with salvation. Devotions such as
these are expressions of the baroque Iberian Catholicism that was rooted in
the seventeenth century, but that reached a new level in the next century with
the discovery of gold and the lavish ornamentation of churches. For the
brotherhoods, this baroque aesthetic and the concern with death that were
traits of society as a whole intersected in the practices and rituals associated
with caring for dead brothers. It is suggestive that in both the Constituições
Primeiras and many of the brotherhoods’ compromissos from the era, de-
tailed guidelines for the order and texture of public processions are laid out
in the context of funerals, but not of festivals.≤Ω
We might well wonder why slaves, most of them African born, and all of
them with a supposedly fragile grasp on Catholic doctrine, would dedicate
such time, energy, and resources to the goal of obtaining Catholic burials.
This subject was raised in chapter 4, but merits further elaboration here
since it was both the paramount objective of the brotherhoods and the
most direct way for individual slaves to demonstrate their human equality
within the hierarchical confines of colonial society. In 1694, an accord was
signed between the governor of the captaincy and the Santa Casa de Miseri-
córdia dealing with slave burials. The price of a burial was set at 960 réis,
and it was stipulated that when a master could prove his or her inability to
pay, the church would absorb the expense by providing a mortuary cloth (in
practice, a white sheet), a co≈n (actually a cloth or rope hammock), and
the use of two slaves to transport and bury the body. It was not unusual even
for masters who obviously had the funds to pay, but lacked the inclination,
to rid themselves of the corpse in some secluded spot on the coast or
outside the city. Dead slaves secretly abandoned in this way would not be
given an obituary register, which funerals provided, as well as the Catholic
grave. They were also denied, as Debret quotes some women from Mozam-
bique as observing, a place ‘‘under the ground until Judgment Day.’’≥≠ But
even the funerary services o√ered for a fee by Santa Casa de Misericórdia
would seem to have left much to be desired, according to this nineteenth-
century account of them:
The poorer people, and certainly the blacks, are treated with far less cere-
mony in these rites than others more advantaged. Soon after death, the
corpse is sewn into a rough sack, and a message is sent to the cemetery to
make room for one more. Two men appear at the house of the deceased.
They load the corpse into a hammock, suspending it by a pole running along
the hammock from head to feet, and carry it with no noticeable delicacy
through the streets. If during this walk they encounter other corpses being
likewise trundled o√ to the same horrible resting place, the bearers will roll
up all the bodies into one bulging heap, and carry them in the same ham-
mock, each bearer taking a share of the weight. Arriving at the final destina-
tion, the men approach a long hole, some six feet wide and four or five deep.
Into that pit the corpses are tossed, with no rites performed. The bodies pile
up every which way, bent and twisted. Then when the hole is filled near to the
very top with corpses comes the handiwork of the black sacristan, who is
160 | Chapter Five
apparently devoid of any human thought or feeling. He shu√les about with

an empty gaze, shoveling dirt on top of the wretched dead to level out the
hole with the surrounding earth.≥∞
The fear of having one’s corpse left to the elements, or to Santa Casa’s
unseemly collective graves, motivated the slaves to pursue the only other
dignified option—a formal, paid Catholic funeral. The brotherhoods’ fu-
neral rites did not merely pay homage to the individual dead person, but in
their pomp and exuberance they willfully displayed to the rest of society
their commitment to proper care of the dead (for Catholics, a body left
unburied or interred without proper rites was a supreme o√ense).≥≤ It was
imperative that the whole brotherhood turn out to mourn and accompany
the corpse to the grave, and if a brother was absent from these events it was
considered a serious infraction. The corpse was taken first to the church
and placed in front of the altar, where a mass would be held; later, after
burial, another mass would be held for the souls of brothers long departed.
It is clear that the brotherhoods took the public dimensions of these
rituals very seriously, since it was a way to express their prestige and
dignity.≥≥ But detailed descriptions of the funeral practices of Rio’s black
brotherhoods do not seem to exist, which makes accounts such as the one
above—from the early nineteenth century—so important. Other sources
indicate that there was little overall change in funerary practice in Rio de
Janeiro from the eighteenth century until around 1850, when new legisla-
tion regarding burials was approved. The compromisso of the Brother-
hood of Santo Elesbão and Santa Efigênia indicates that a sequence of
measures was to be taken upon the death of a member (or a spouse or
child). To begin with, the brotherhood needed to be formally informed of
the death and convene the assembly to begin the rites. The deceased’s final
wishes with respect to property needed to be found; typically these would
have been recorded in a will. It should be noted that all these proceedings
could reinforce the solidarity of a brotherhood, but disputes and jealousies
were also to be expected as the private holdings of the brother were dis-
closed. Brotherhoods hoped to be listed as beneficiaries in the wills of their
members, and if the deceased happened to belong to more than one
brotherhood, those groups could argue over how much each was owed (in
obligatory payments, if not inheritance).
The process started with informing the treasurer of the death. He, in
turn, would tell the courier. The courier then told the judge, the chaplain,
and the other brothers, walking from house to house announcing the news
and advising that, as the compromisso says, ‘‘all gather together to accom-
pany the body of the dead.’’ That the courier knew where all the brothers
lived suggests that the brotherhood involved close social ties.≥∂ The judge
led the funeral procession from the church (or from the deceased’s house,
if there were no mass), carrying his sta√ of o≈ce in hand. The brothers
followed, carrying the body and walking ‘‘in well-composed ranks.’’ If the
deceased had belonged to several brotherhoods, these would all be present
in order of age and prestige, from the oldest to the newest. Sensitivity to
prestige was also conveyed in where the corpse was laid to rest. Prominent
members might be buried within the church, with more humble ones
buried outside on the church grounds. The members were to return from
the burial site in an ordered fashion, just as they had arrived there, and after
a final blessing the group would disperse.
One detail that was impossible to derive from the compromisso was the
precise location of burials. Until 1754, the brotherhood was located at the
Church of São Domingos, which had a cemetery. That year, the brother-
hood’s own Chapel of Santo Elesbão and Santa Efigênia was inaugurated,
and over the second half of the century numerous obituaries specify the
cemetery at that chapel as the place of burial.≥∑ Presumably, in all cases, the
mass for the deceased would be held at the church housing the brother-
hood; the public procession would therefore start and end at the same
church, assuming that the body was buried at the church where the broth-
erhood met. Another question might be raised about the brotherhood’s
use of hammocks to carry its corpses, just as was done in the uninspired
service o√ered by Santa Casa de Misericórdia. If the brothers invested so
many capital and symbolic resources in the funeral, why wouldn’t they
have arranged more formal modes of conveyance? Perhaps because death
rites had to accomplish several things at once. Beyond the simple issue of
getting a body into the ground, the procedure had to be dignified and
thorough enough that all the brothers could remain confident that the
burial guidelines in their compromisso were being met. But the street
procession—with its displays of fancy attire, flowers, and other pomp—
o√ered a moment for the rest of the city to recognize the brotherhood
162 | Chapter Five
and admire its vivacious devotion. Because resources were always tight, it
would have been reasonable to calculate that the appearance of the living
counted just a bit more than that of the dead in this context. The more regal
the cortege in its entirety looked, the less onlookers might have noticed or
dwelled upon the poor aspect of the corpse, swaying along in its hammock
amongst the finery.
Beyond the obligations specific to burials, the brotherhood maintained
other death-related commitments to its members as put forth in the com-
promisso. When a brother died, the group would guarantee that his widow
(remaining unmarried) and any children younger than fourteen would
have ‘‘the same privileges they enjoyed during the life of their husbands
and fathers, without being asked to make any contributions.’’≥∏ The em-
phasis on widows hints that the brotherhood valued marriage and deemed
it appropriate that the brothers be married. At the same time, a widow
without children (Mina women seem to have borne fewer children than
women of other African groups) or parents living close by—and African
slaves usually had no Catholic parents to mention—became the heir to
whatever wealth the deceased left behind. From the brotherhood’s per-
spective, there was another reason to take care of her: perhaps she would
eventually marry someone from the brotherhood and keep the inheritance
in play.
The brotherhood could also choose to waive the annual dues from a
longstanding member who had fallen on hard times or become ill. Poverty
and infirmity, often acting in concert, a√licted slaves and freed alike.
Through small gestures the brotherhood could ameliorate these condi-
tions, but not reverse them, and the life expectancy of this population was
notoriously short.
freed slaves
Brotherhoods could also help to free members who were slaves. To con-
sider what freedom meant or brought to a slave in these contexts, we have
to keep in mind how society in that era viewed slavery. Scarano’s work was
the first to demonstrate that the black brotherhoods, composed basically
of slaves and freed slaves, themselves owned slaves that they either bought
or inherited from members through their wills.≥π However, the meanings
and ramifications of slavery go far beyond immediate ownership or the use

of slave labor. Stuart Schwartz argued forcefully that in colonial Brazil,
slavery created the fundamental facts of life, and indeed no one was truly
free in the presence of slavery.≥∫ In the 165 sets of compromissos from
black brotherhoods analyzed by Patricia Mulvey, 11 made reference to the
goal of freeing members from slavery: 5 from Rio de Janeiro, 4 from Bahia,
and 2 from Minas Gerais, all of them written before 1800.≥Ω The particular
challenge here is to understand the social conditions in which a given
brotherhood could buy or inherit a slave, while at the same time create
mechanisms to liberate another.
The 1565 compromisso of the Brotherhood of Our Lady of the Rosário
of the Monastery of São Domingos prohibited slaves—along with ‘‘white
moors,’’ mulattos, and Indians—from holding any administrative posi-
tion. The document recommends that the brotherhood help enslaved
members obtain the resources needed to buy their freedom, but also sug-
gests that the application of funds from the brotherhood’s co√ers for such
purposes be limited.∂≠ The compromisso of the Brotherhood of Santo Eles-
bão and Santa Efigênia was even less concerned with the processes and
transactions involved in attaining freedom. It assumes in its profiles of the
model brother a freed African man, who has a wife and children. Such
individuals would not have been the numerical majority among the broth-
ers, but one senses from the compromisso and other documents that they
were more prominent and occupied more positions of authority than peo-
ple from the larger population of slaves. And although the rates of mar-
riage associated with the brotherhood were low, a survey still in progress
indicates that the preeminent brothers were all married.∂∞
It is striking that the theme of obtaining freedom does not appear in any
of the thirty-two chapters of the compromisso of the Brotherhood of Santo
Elesbão and Santa Efigênia. But this theme was raised by two separate lay
brotherhoods of Mina-Mahi Africans who formed out of the Mahi Congre-
gation at the Church of Santo Elesbão and Santa Efigênia in the second half
of the eighteenth century: the brotherhoods of the Devotion of the Almas do
Purgatório (Souls of Purgatory) and of Our Lady of Remédios.∂≤ In both
cases, the mention in the compromisso includes proposed methods of fi-
nancing the purchase of freedom. These initiatives might be understood as
a form of mutual aid, similar to the emancipation societies of the nine-
164 | Chapter Five
teenth century.∂≥ But even these systems of financial support had clear
limits, being conceded to members of the congregation only in special
circumstances. They represented neither a right for every brother nor a
contractual act of charity among all brothers, but were a privilege granted to
the most illustrious enslaved brothers obligated to repay the organization
as soon as possible (and it must have been assumed that they could do so).
power and resources

However tempting it might be, given the lack of complementary sources
and of better contextual information on the black brotherhoods, the com-
promisso cannot be read as an ironclad script or template for social be-
havior. Not only is it true that they were the final products (revised and
appended) of disputes, alliances, and compromises between di√erent in-
terest groups at the core of the brotherhoods, but in practical terms some
of the strength of such documents (which bridged the Catholic Church and
organizational provenience groups of Africans) must have resided in their
flexibility. Certainly some of what was written was probably not obeyed, or
was not always obeyed in the same way. At the same time, we should also
try to appreciate the networks of power and social relations in which
detailed guidelines such as these were enforced and obeyed. The compro-
misso of the Brotherhood of Santo Elesbão and Santa Efigênia had the
critical dual function of constructing social norms for the group, and of
establishing and embodying the authority of those entrusted with guard-
ing the norms through the exercise of power. The delicate relation between
these two spheres was often a√ected by changes in power relations within
the group. The compromisso was altered five times between 1740 and
1764, and it is possible to get a sense in each case of the di√erent interests
and strategies involved. The most common form of dispute dealt with
finances and with administrative positions (their roles, and how they were
appointed). Early in the document the administrative functions and the
procedures for nominating brothers to them are laid out, but the fact that
this information keeps recurring in increasingly detailed form throughout
later additions demonstrates that it was a continual source of contention.
The first twenty chapters of the compromisso describe the attributes of
the judge, which was the position of greatest importance and highest
authority on the brotherhood’s board. The judge would care for the images
of the saints, look after the financial and other material holdings of the
brotherhood, and ‘‘observe the demands’’ (supervise the proceedings of
any legal cases in which the brotherhood was a beneficiary). The main
judge could be given the responsibility of appointing other judges (the
complete board had a total of twelve judges) or could also accept nomina-
tions from among the brothers for a general position called juiz da Mesa
( judge of the board). This seems to have been a ceremonial title that could
be open to any prestigious brother with enough money to pay the hefty
donation associated with it; the compromisso refers to the ‘‘advantageous
size’’ of that sum.∂∂ The Board’s four o≈cials—judge, procurator, trea-
surer, and scribe—were to meet at the church every Sunday of the year to
address internal a√airs of the brotherhood. The mandatory meetings were
to assure that the board members kept up with the necessary work (‘‘para
evitarem os descuidos que podem haver em cada um dos oficiais que
servem’’). Chap. 22 introduces a striking addition to the original docu-
ment: ceremonial board positions for women, including a female ‘‘judge of
the board.’’ These also carried substantial fees, although whether or not
they were the same as the men paid, women did not gain equivalent power.
The same chapter also states that the board was to be composed of twelve
men and twelve women, which, given the limited number of o≈cial ad-
ministrative functions, suggests that the expansion of board titles was
intended at least in part to bolster the annual esmola receipts. But another
part of the motivation for all those people on the board came from the need
to sta√ the royal hierarchy to organize the folias.
In the original twenty chapters of the compromisso, participation by
women in the brotherhood is mentioned only in chap. 10. The intention
there had been to limit membership to women from the Mina Coast, Cape
Verde, the island of São Tomé, and Mozambique. Membership was denied
to creole women (born of Africans in Brazil), cabras (mixed-race), and pretas
d’Angola (women from Angola). In Portugal, most of the brotherhoods
allowed only men to join. A particular exception was the black Brotherhood
of Our Lady of the Rosário of the Monastery of São Domingos in Lisbon,
in which women were permitted membership if they were married to a
brother of the order.∂∑ In Brazil, brotherhoods of white men tended to
include women only as dependents of male members; they gained some
166 | Chapter Five
small benefits but were not permitted to join outright. The black brother-
hoods allowed women to join as members (called irmãs, sisters), whether
or not they were already married to a member in good standing. But this
could be a source of conflict. The sisters paid dues and were theoretically
equal to the other brothers, but as a practical matter they were excluded
from participating in decision making or holding high administrative posi-
tions. The changes made over time to the compromisso’s chapters reflect
women’s progress toward gaining more power inside the brotherhood.
This probably derives from their recognized ability to meet the men on
equal financial ground—that is, to pay the higher donations associated
with authority and prestige. For example, chap. 26 creates the positions of
a juiz and juiza (male judge and female judge) for each patron saint, with the
donation set at 12,800 réis per person.∂∏
According to the compromisso’s twenty initial chapters, the main job of
the procurator was to organize the festival of the saints. Chaps. 28 and 31
introduce new responsibilities to the position: caring for sick brothers,
distributing monetary contributions to needy brothers, and performing
religious services for dying brothers.∂π Beyond this, since part of the job
was to collect and distribute donations within the order, the procurator
found himself caught up in a variety of internal disputes which he was
expected to help resolve. These social and financial functions were critical
to the well-being of the group, and the procurator, understood to be
second only to the main judge in the brotherhood’s hierarchy, needed to be
a man of patience, wisdom, and influence.
Under the procurator in rank came the treasurer, who, the compro-
misso specified, needed to be a white man. This would seem to follow the
model established by the Brotherhood of Our Lady of the Rosário of the
Monastery of São Domingos in Lisbon, in their 1565 compromisso. Admis-
sion into the brotherhood had been open to white people, as well as to
pardos, since its founding. I have been unable to identify any whites or
pardos a≈liated with the Brotherhood of Santo Elesbão and Santa Efigênia,
but all the indications from a range of evidence suggest two hypotheses:
these people would have been acquainted in one form or another with the
founders of the order, and they would probably have been linked to the
commerce of slaves in Rio de Janeiro. (That would make the source of their
financial contributions to the brotherhood, which must have been signifi-
cant, somewhat ironic.) The compromisso also states that if a brother

capable of handling the position of treasurer was not to be found, someone
from outside the order may be chosen if he was meticulous and honorable.
The last of the four board o≈cials was the scribe, a person able to take
care of the books and keep them up to date, ‘‘benemérito assim no zelo
como na inteligência das contas’’ (distinguished in his zeal and his exper-
tise in arithmetic). He recorded critical information, including money
going in and out of the treasury from dues, donations, and payments for
masses. The scribe was fourth in rank, but in the judge’s absence, he would
preside over the assembly. This was to keep the procurator, as second in
command, from concentrating too much power, and also to prevent con-
trol of the order from passing, even for a moment, to a white man (the
treasurer). A common fifth position in rank was that of courier, although in
the Brotherhood of Santo Elesbão and Santa Efigênia this position was not
part of the board, which was limited to four o≈cials.
The brotherhood made money from essentially two sources: internal
dues and contributions, and public donations. Chap. 27 does not precisely
mandate, but assumes, that the two female judges of the patron saints were
directly responsible for asking for money in the street processions. Collect-
ing public money is not mentioned in a detailed way, but it is well known
that this was one of the principal functions of the folias. All of this means
that women holding places of prestige within the brotherhood’s hierarchy,
along with the king of the folia, were responsible for a significant part of
the order’s income. We are left to wonder how the folias di√ered in their
strategies for gathering money in the earlier years, before a parallel hier-
archy of women was established in the brotherhood. It should be remem-
bered that monies collected in this way went to a fund, not authorized by
the bishop, to be used for dinners and other special needs. The compro-
misso does not spell out how these funds would be spent or divided, and I
found no documents pertaining to the brotherhood’s accounting. But
clearly, the control of resources was a growing point of concern for the
board, especially after 1746, when plans were laid by the order to construct
their own church. Taken together, these points demonstrate the dynamic
process by which the need for income intersected with other concerns
about gender and the concentration of power.
In the first twenty chapters, the annual donations associated with mem-
168 | Chapter Five
bership and administrative functions are listed. Among the o≈cials, the
judge owed 12,000 réis; if he wished to stay past his mandate, he needed to
o√er another esmola avantajada (distinguished contribution). The scribe
owed 10,000 réis. It was notable that the treasurer, procurador, and courier
were exempt from paying elevated dues, given the amount of time involved
in these positions. Of course, if these o≈ce holders chose to contribute to
por sua devoção (by showing their devotion), their o√ering would be ac-
cepted. Other brothers a≈liated with the board in festive, nonadministra-
tive roles owed 2,000 réis a piece. Rank-and-file brothers owed 480 réis.
We gain some perspective on these amounts by considering that at the
time, cheap slaves could be bought and sold in Rio’s markets for 50,000
réis. When new judge positions were created for the patron saints in chap.
26, this represented a significant enhancement to the brotherhood’s inter-
nal receipts.
The order needed to augment its income, but it was not acquisitive to the
point of sacrificing what its o≈cials deemed harmonious social relations.
In chap. 30, the compromisso acknowledges that it was not uncommon to
encounter an unruly element or two among the sincere and worshipful
brothers.∂∫ To avoid the intrusions of these ‘‘enemies of peace,’’ the board
(led by the judge) had the obligation to interview new brothers to deter-
mine whether they had been expelled from another brotherhood. If a
prospective member was discovered to have a history of disobedience, he
was to be immediately and forever expelled from the Brotherhood of Santo
Elesbão and Santa Efigênia, no matter what amount of money he had
agreed to or o√ered to pay. ‘‘We are only interested in quiet, peaceful
people here,’’ the compromisso states, ‘‘not rabble-rousers.’’∂Ω
The compromisso also devotes several chapters to the relations between
the brotherhood and the chaplain. Early in the text it is stated that the judge
and the other o≈cials would select this person, and his main duty was to
lead fifty masses a year, including those for the living and for the dead. He
would be paid a donation of 25,000 réis for the year, and needed to arrange
his own substitute if he were to miss any of his commitments to the
brotherhood. It is instructive to compare the way the compromisso ad-
dresses the responsibilities of the chaplain versus those of the board mem-
bers. The latter are described initially in basic form, although with the
passing of time and the growth of the brotherhood’s ambitions, somewhat
more of the o≈cials’ duties and powers (or the limits of those powers) are
spelled out. But from the very first, the compromisso is quite specific about
what is expected of the chaplain. For one thing, he needed to sign a
contract for his term. The contract would list the masses he must lead, and
also would require that he be present at all burials and all solemn occa-
sions, whether in the church, on the street, or elsewhere. This formality
and precaution reflected the distrust that characterized most interactions
between black brotherhoods and ecclesiastical authority.
Chap. 23 directs newly detailed demands and rules to the chaplain. It
stipulates that every brother who died during the term had a right to no
fewer than ten masses. But more boldly, the matter of payment is now
spelled out to remove any ambiguity. The chaplain would be paid at the end
of the year, and no longer should expect a fixed sum: for every mass he had
performed until then, he would receive the equivalent of one pataca (a silver
coin worth 320 réis). If for no other reason, this is why the scribe’s func-
tion of recording all the masses was so critical. Yet even these measures
seem to have been insu≈cient to inspire consistent attention from the
chaplain. By chap. 28, the brothers complain about the frequent absence of
the chaplain from the bedsides of sick or dying brothers, grumbling that
chaplains ‘‘mais procuram a sua conveniência que cumprirem com a sua
obrigação’’ (pursue their own convenience more avidly than their obliga-
tions). This chapter authorizes the brotherhood to contract another priest
whenever the chaplain, or his substitute, fails to appear. The expense for
those services would be deducted from the amount payable to the chaplain
at the end of the year. The theme of the unreliable chaplain is taken up yet
again, in the last chapter of the compromisso, and here the tone is more
bitterness than anger. These abuses of the chaplain against his own word
and o≈ce are an o√ense ‘‘o que Deus não o tal permite’’ (not permitted by
God), because the dead are not to be punished for the squabbles of the
living.
The compromisso also reflects change over time in the frequency and
distribution of masses called for. The first twenty chapters do not establish
a number of masses that every dead brother should receive, and even in the
case of death of one of the four o≈cials of the board, the only special assem-
bly called for is one mass at the church with the corpse. Chap. 23, however,
requires ten masses for every member of the brotherhood who passes
170 | Chapter Five
away.∑≠ The number of masses was not set for an administrator of the order
who dies, but one assumes it would have been in accord with the ‘‘zealous
service’’ and ‘‘advantageous contributions’’ of those distinguished individ-
uals. Later, in the undated addition to the compromisso, it is declared that
upon death judges of either sex should receive 20 masses; scribes and
treasurers, 18; couriers, 12; and other nono≈cial members of the board, 16.
Over time, the gradual entrenchment of the internal hierarchy of the broth-
erhood is expressed by the increasing privileges accruing to members of the
board, and especially to the four o≈cials. The number of masses for a rank-
and-file brother stayed at ten, while that of the judge was twice as many.
Certainly the judge paid a larger contribution to the order than other broth-
ers did. But the compromisso reveals a marked trend in the concentration
of the brotherhood’s expenditures among the board members and their
families.
The integration of women into the brotherhood seems to have been
viewed as a source of both advantages and challenges. As noted earlier,
women provided a source of new income, and the statutes are much more
upfront about the donations associated with each female board position
than they are about the rights and duties that such positions would entail.
(For instance, we are left to wonder about the precise responsibilities of the
so-called juiza de ramalhete, whose title would seem to indicate caring for
flowers and bouquets.) Elsewhere, women are occasionally mentioned as
widows in the event of a married brother’s death. The text of the compro-
misso reflects its era, so that the masculine identity of members is assumed
throughout. But it is worth remarking that chap. 10, which denies entry in
the order to men or women from Angola, as well as to male and female
creoles and cabras, justifies that prohibition based on the alleged ‘‘bad
behavior’’ of ‘‘sisters’’ from these populations. Despite this attempt to
exclude them, women from Angola could take advantage of another way to
enter that the statutes made possible: if they were married to a member in
good standing. Although there were few marriages between Mina men and
Angola women, the possibility of some such unions was always present.∑∞
On the other hand, creole and cabra women would have been far less likely
to marry an African man.
The compromisso is eloquent in its aversion to female Angola, as well as
to female creoles and cabras. It is recommended that, when one term of
o≈ce is ending for a group of board o≈cials, a reckoning should be made

of any inappropriate admissions granted to such women, whose identity
might have been hidden or unverified. The o≈cial responsible for allowing
them to become members should be prohibited from ever again ‘‘servir
cousa alguma na dita Irmandade’’ (holding an administrative position in
the brotherhood)). And the women, their cover blown, would be expelled.
They would receive their entry dues back in full (again demonstrating that
the brotherhood’s interest in income flow could be sharply subjugated to
their social concerns). The registers of the women’s membership would be
revoked with notations placed at the margin of the record book (to ‘‘fique
de nenhum efeito seus assentos, de que se fará declaração à margem dos
livros deles’’). Nowhere does the compromisso define what is meant by
creole; the Brazilian-born children of Mina in the brotherhood would have
fit an objective definition. The compromisso refers to young children or-
phaned by the death of a father who was a member, but other than that
have little to say about children in general.
As the size of the board expanded, finally encompassing a total of 24
people (12 men and 12 women), the potential for internal debate and
tension grew along with it. But the brotherhood still had to answer to
ecclesiastical and royal authority as well. In 1767, a royal decree sent from
the Tribunal of Conscience and Orders in Lisbon to the bishop of Rio de
Janeiro requires the revision of chap. 10 of the compromisso, discussed
above. All of the groups previously prohibited from membership were now
to be allowed to enter.∑≤ The specific alteration asked for in the name of
King D. José was the removal of any reference to ‘‘a diferença da natu-
ralidade dos pretos’’ (a di√erence in the nature or essence of black peo-
ple).∑≥ In light of this directive, the prohibition against membership could
not stand. But the board also maneuvered to limit the ramifications of the
change by restructuring itself through new wording. The rewritten chapter
states that the board would now be composed of twelve people, of whom
six must be irmãos criadores (founders)—from Mina Coast, Cape Verde, São
Tomé, or Mozambique; and the other six spots would be dedicated to the
nonfounders, including people from outros admitidos, the other groups who
had just won the right to enter. This allowed the current members and the
nations they represented to safeguard half the positions on the board,
while also holding out the possibility that some of the remaining positions
172 | Chapter Five
would go to whites or pardos, who were obviously not founders of the

brotherhood but who had still been accepted as members from the begin-
ning. That a white man, possibly a slave merchant, could in the eyes of the
order’s African founders be preferable to black people from Angola as
o≈cials of the board speaks to the complexity of social relations and
strategic alliances in this period.
Chap. 29 deals with discipline. It states that the founders and the rest of
the board should enjoy the guarantee of bom regime (proper comportment)
from the brothers, especially with respect to matters of paying contribu-
tions and entry dues. Infractions could be punished, up to and including
expulsion from the order. This chapter emphasizes what chap. 24 began
to make clear: principal decision making in every sphere of the brother-
hood’s activity was to be left in the hands of the judge and the other
o≈cials, as well as the chaplain (who was included through pressure from
the church). This solidified the order’s internal hierarchy and increased the
prestige associated with the founders’ African identities. By chap. 31, the
graduated ranking of the four board o≈cials was replaced with a single
high administrative level; the procurator, treasurer, and scribe, formerly
under the judge, were now also to be treated as superiors equal to the judge,
under threat of punishment. At the first o√ense, the guilty party was
required to pray on bended knee to the Holy Mother. A second show of
disrespect meant the o√ender would have to carry ‘‘uma pedra que haverá
na Irmandade,’’ a rock (of undisclosed size and weight), owned by the
brotherhood and kept nearby. A third time would lead to chastising in front
of the assembled board; after this, only expulsion was left as a reproof for
impertinence to authority. Decisions about expulsion were the prerogative
of the board, but unfortunately they seem not to have recorded how many
times it (as well as disciplinary action in general) was used, or under what
conditions.
the boundaries of group identity

Since the fifteenth century, first in Portugal and then also in Brazil, histor-
ical changes led to the creation of a multiplicity of lay Catholic brother-
hoods with distinct, even contrasting identities.∑∂ In Portugal, the expand-
ing population of Africans and their descendants was not welcome in the
earlier, ‘‘white’’ brotherhoods; those institutions both excluded the blacks

and encouraged them to form their own orders, such as that of Our Lady of
Rosário.∑∑ In the ensuing years, after the first black brotherhoods were
established in Portugal, parallel processes of inclusion and exclusion in
their own admission procedures (based on a variety of social criteria)
helped lead to the proliferation of diverse black brotherhoods, notably in
Brazil. In Rio de Janeiro, Mina blacks had been prohibited from attaining
positions of any power in the Brotherhood of the Rosário, and they decided
to found their own orders. As noted in chapter 4 of this book, from 1715 to
1740 those Mina left the Church of Rosário, separated into three sub-
groups, and subsequently organized three new brotherhoods: one created
the Brotherhood of Santo Antônio da Mouraria (Saint Anthony of the
Moors), another the Brotherhood of the Menino Jesus, and the third the
Brotherhood of Santo Elesbão and Santa Efigênia. In their own organiza-
tions they could hold the positions of authority and prohibit access to
creole and Angola blacks, who had been the ones preventing them from
advancing in the Brotherhood of Rosário.∑∏ Beyond these three orders
linked to the Mina ‘‘nation,’’ there was a related group, the Mahi Congrega-
tion (founded 1762). This congregation created two more devotions: one to
the Almas (Souls) of deceased Mina, and the other to Our Lady of Remédios
(to aid the poor and sick Mina in Rio).
Taken as a whole from diverse sources, the array of documents pertain-
ing to these groups weaves a fundamental ambiguity around the word
nação (nation). As detailed in earlier chapters, this term could refer to
geography, to ethnic networks, or both: it was used to designate the place
of provenience of African peoples, and also to identify African peoples’
ethnic background. The double meaning of such a common word in period
documents complicates any attempt to distinguish the brotherhoods who
based membership on provenience groups (such as that of Santo Elesbão
and Santa Efigênia, centered on Africans from Mina Coast, Cape Verde, São
Tomé, and Mozambique) from those that formed as subgroups within a
provenience group (such as the Mina Congregation, a subset of Mina
blacks)—and from those that cohered around an ethnic identity within a
provenience group (such as the Mahi Congregation and their derivations).
In any analysis of African slave identity in Brazil, it is critical to keep in
mind the distinction between nação as provenience identifier (Angola na-
174 | Chapter Five
tion, Mina nation) and as ethnic identifier (Rebolo-Tunda nation, Mina-

Savalu nation, and so on). Both types of nations were involved in the
Brotherhood of Santo Elesbão and Santa Efigênia, although the associated
identities could be expressed di√erently in di√erent situations.
The procedure for evaluating potential members is listed in chap. 10 of
the compromisso, where it is admonishes that if the candidate is ‘‘preto’’ or
‘‘preta’’ (male or female black, but in the vernacular context, specifically
African), ‘‘se examinarão com exata diligência a terra e nação donde
vieram’’ (he/she should be closely examined to determine his/her land and
nation they came from). Those individuals who seem to be from Mina
Coast, Cape Verde, São Tomé, and Mozambique were rushed through the
process, since they were from the nations that founded the brotherhood.
Once they paid the entry dues, a record was made of their membership
(‘‘logo se fará assento nela dando de sua entrada quatro patacas’’), and they
were brothers. O≈cial positions on the brotherhood’s board could only be
held by African people of these nations—except for the role of treasurer,
which, as noted earlier, presented a special case where even a white man
was preferable to Africans of other nations.∑π The fact that Angola women
were prohibited (more emphatically than others) from joining the brother-
hood did not seem to create or reflect openly antagonistic relations between
Mina and Angola in the city of Rio de Janeiro. To the contrary, although
there was tension and occasional conflict, these groups interacted reg-
ularly. They shared churches, and cemeteries; they united for festivals and
funerals. And one suspects that the members of the Brotherhood of Santo
Elesbão and Santa Efigênia would not have been so preoccupied with the
admission of Angola women if this were not a common infraction.
the imperial state of santo elesbão

The major theme of the five new chapters added to the compromisso in
1764 was the creation of a folia in the brotherhood, to be called the Imperial
State of Santo Elesbão. The name recalls the Catholic kingdom of Elesbão
(who had been a prince, son of an Ethiopian king) in the European Middle
Ages.∑∫ The appearance of the folia is a direct expression of the growth of
the brotherhood. Adding new positions to the board (even when they were
more festive than administrative) was good for the overall organization,
discipline, and camaraderie of a large group; the justification o√ered for the
folia was that it would maintain the ânimos (good spirits) of the brother-
hood. And of course it would be good for the bottom line, since the kings of
the folia, of which there were several, would each pay a donation of 15,000
réis. An impressive folia on the streets would also enhance public dona-
tions to the brotherhood during the special processions authorized for that
purpose. The donations collected from the public were handed by the folia
over to the treasurer, who should in turn place them in the cofre Divino
(divine safe). No record was kept of the use of these funds, although it is
reasonably certain that they underwrote most of the costs of the annual fes-
tival of the saint. They likely were also drawn upon to help defray the costs
of attending to sick and dying brothers, as well as various ritual and festive
expenses. These chapters do not specify that such funds could be used
toward the purchase of freedom for enslaved brothers, although the Mahi
Congregation’s compromisso related to folias did mention that possibility.
The hierarchy would be elected for a period of three years, and while the
new positions included yet another judge, most titles were royal: emperor,
empress, kings, prince, princess, and so on. As with the o≈cial positions
on the board, an individual’s term could be extended if the properly advan-
tageous donations were furnished. It is apparent that the role of emperor
needed to go to a man of some prestige and means, and not only because
this was the top rank of the folia. The emperor himself would be responsi-
ble for providing the rest of the court with suitably regal attire, as well as all
the small decorations and alfaias. The kings and queens would pay for
their own outfits, but in recognition of the financial burden the emperor in
particular was expected to experience, he was exempt from any other
personal contributions to the brotherhood during the three years of his
term. The fittings and other arrangements for costumes must have taken
place soon after elections were held, since the statutes stipulate that the
entire court of the folia was to report to the board on the assigned day to
take o≈ce already wearing their cloaks, crowns, mantles, and other finery.
Although not part of the brotherhood’s actual administration, the emperor
was to be treated with due deference by the rest of the brothers. When he
was brought to address the board, the judge needed to o√er him the best
position at the table. The folia held their assemblies in the church, al-
though access to space was limited in the church and constantly needed to
176 | Chapter Five
be negotiated among the brotherhood, the folia, and the several smaller
devotional orders headquartered there. That the folia was o√ered the use of
the church consistory is an indication of the prestige accorded the group.
There were diverse ethnic groups within the dominant ‘‘nation’’ of Mina
blacks, as well as smaller populations of other nations, united in the
Brotherhood of Santo Elesbão and Santa Efigênia. But despite the overarch-
ing structure of the lay brotherhood with its hierarchy and multiple stat-
utes—and the layers of ecclesiastical authority above it—the congregation
was hardly locked in some sort of formal stasis. Conflicts over power and
identity continually reshaped the internal social alignment of the group,
creating strategic alliances among certain factions and subgroups while
inciting division among others. Just as in the Brotherhood of the Rosário,
when the Angola blacks had imposed criteria to distinguish themselves
from Mina blacks (and make the Mina ineligible to hold o≈ce), a similar
impulse to separate ‘‘us’’ from ‘‘them’’ operated here. The election of kings
to the folia became one of the key activities around which alternate identi-
ties could cohere and find a venue for expression, as the next chapter
demonstrates with respect to the Mahi subgroup.
The presence of multiple provenience groups in the brotherhood dates
back to its founding in 1740. It may be that after 1748, when work began on
the Church of Santo Elesbão and Santa Efigênia, and certainly after the
inauguration of the church in 1754, the coexistence of these di√erent
groups began to su√er strains. The land for the new church had been
obtained in 1747, and it was during the period of final planning and
groundbreaking for construction that the first reforms of the compro-
misso were completed, in 1748. These reforms brought more detail to the
functions of the treasurer, established the number of board positions, and
described the conditions for women’s participation in the order. There are
no transcripts of the debates and disputes that led to these alterations, but
almost certainly the principal issue driving them centered on the need to
collect and carefully spend lots of money to make the desire of the brother-
hood’s own church a reality. At the same time, at least some of those who
were giving money toward this end might have expected to receive in-
creased power, prestige, or other advantages within the brotherhood. In
that light, because women were the one group that benefited in such a
marked way from the compromisso reforms of the period—gaining not
only a position of judge, but a complete parallel board for women—it

seems reasonable to conclude that their financial contributions had been
substantial.
Construction of the church represented an unusual accomplishment for
a black brotherhood, and in general this project had served to unify the
diverse sectors of the order. But after 1754, once ensconced in its new
church, the brotherhood’s various component nations (here understood as
both provenience groups and ethnic groups) began to chafe and quarrel
over what one faction termed ‘‘preferências e maiorias’’ (preferences and
majorities).∑Ω But with respect neither to gender nor nation were alliances
set in stone, because the push and pull of short-term and long-term inter-
ests created dynamic, unpredictable subsets of constituents within the
order. Women might band together at one juncture to argue a certain
cause, while later in a di√erent context some women would join a group of
men to argue another point. The plasticity of these alliances is perhaps not
what a modern observer would expect, and yet it appears characteristic of
how such groups interacted.
Ever since its founding in 1740, the Brotherhood of Santo Elesbão and
Santa Efigênia comprised several African nations, and admitted pardos and
whites but excluded Angolans and creoles. From within this first order a
group of Mina organized, calling themselves the Pretos Minas (Congrega-
tion of Mina Blacks) and electing Pedro Costa Mimozo as the King of their
folia.∏≠ By the time of their next election, a subgroup within the new
congregation had emerged as opposed to Mimozo. That subgroup exerted
itself to get the faction led by Mimozo voted out of o≈ce, but they were
unsuccessful. The winner of the vote was Clemente Proença, an ally of
Mimozo,∏∞ but the disputes for power within the congregation soon multi-
plied in complex new forms. Just as the Congregation of Mina Blacks had
arisen through the formalization of a split between Mina and other nations
within the brotherhood, the very same Congregation of Mina Blacks would
soon undergo its own schism based on identities and interests. The sub-
group that formed from within it was called the Mahi Congregation, and
its first elected leader was Captain Ignacio Gonçalves do Monte, praised as
a ‘‘true Mahi.’’∏≤ It turns out that Monte had enemies in the Congregation of
Mina Blacks, and they all happened to be Dahomeans. He rebelled against
them and their congregation, taking his followers with him to derive yet
178 | Chapter Five
another order from the original Brotherhood of Santo Elesbão and Santa
Efigênia. This sequence of events indicates that Pedro Costa Mimozo and
Clemente Proença were both from Dahomey. Monte’s faction was made up
of several Mina subgroups—including Savalu, Agonli, and Iano—but
Monte’s place of honor as leader, and the name of the new order, suggests
that the Mahi were the largest. And notably, the very first page of the Mahi
Congregation’s Manuscript constructs the somewhat embellished image
of a ‘‘kingdom of Mahi’’ towering over West Africa, a kingdom described as
‘‘one of the most excellent and powerful of the whole Mina Coast,’’ ready at
any moment of its choosing to subdue the nearby kingdom of Dahomey.
From the first, the Brotherhood of Santo Elesbão and Santa Efigênia
contained various groups whose disputes for power led to a persistent
process of segmentation through the creation of strategic alliances. A
breakaway and subsequent founding of a new order was the ultimate
articulation of this clash of identities, but even within an order a contrast-
ing identity could be asserted. This was the case with the Agonli and Savalu
‘‘nations,’’ minority factions in the Mahi Congregation, who finally elected
their own ‘‘kings’’ within that order. Smaller groups wanted avenues to
greater participation and prestige in the larger assembly. It was in recogni-
tion of this basic fact that the Brotherhood of Santo Elesbão and Santa
Efigênia ultimately created the Imperial State. The court of this principal
folia grew to comprise as many as seven kings, each one the figurative
leader of a ‘‘nation’’ within the brotherhood; each king’s court in turn was
made up of the so-called kin or parentes (relatives, actually subjects) of that
ethnic or provenience group. Not counting the category of ‘‘Mina,’’ eight
such provenience or ethnic groups have so far been identified within the
Brotherhood of Santo Elesbão and Santa Efigênia: São Tomé, Mozam-
bique, Cape Verde, Dahomey, Mahi, Savalu, Agonli, and Iano. We can
speculate, but do not know for sure why, the total number of kings was
smaller than the number of nations present within the order. This may be
because the Dahomeans of the Mina Congregation already controlled the
Imperial State and did not need a king.
Most of the historical literature that mentions the folias treats them as
folkloric ensembles, pleasant and colorful but merely festive appendages
to the main entity of the lay brotherhood. To the contrary, I propose that the
folias were intimately and strategically linked to the brotherhoods; they
were a fundamental aspect of the brotherhoods’ internal mechanisms of

power, social di√erentiation, and identity construction. They had impor-
tant roles, and represented important notions of belonging and participa-
tion, for all the members of the brotherhoods. But the festival or street
procession was their most public, visible face, and it was this dimension of
their activities that has endured to animate the folias’ o√shoots in contem-
porary Afro-Brazilian culture, from Rio to Minas Gerais to Recife. These
cultural manifestations are often pigeonholed as ‘‘folklore,’’ in everything
from Brazilian state policy to foreign travel guides; but the concept of
folklore has been recuperated from its lowly status by the French historian
André Burguière, who perceives in its marginality a discreet, unmapped
space of power (rather than a lack of meaning or importance). ‘‘Long
abandoned to the lovers of the picturesque and exotic, folklore should be of
interest to the historian,’’ he argues. ‘‘Its apparent insignificance in the
social realm indicates that in it was invested, and is still preserved, some-
thing of value. A characteristic of power is that it is never exactly where it
appears to be.’’∏≥
Analyzing the place of the Imperial State in the Brotherhood of Santo
Elesbão and Santa Efigênia allows us to see that the exercise of power
within the brotherhood was directly linked to the capacity to resolve the
tensions and disputes between and among smaller, component groups.
Making possible the election of seven kings from seven nations should be
understood as an attempt both to minimize conflict and to improve the
receipt of donations in the brotherhood’s co√er. If we accept Burguière’s
view of the nature of folklore, perhaps the folias are intriguing to folklor-
ists because, in one form or another, they contain hidden keys to the
comprehension of some of Brazil’s later religious practices.
negotiating norms and royal approval

The first twenty chapters of the compromisso of the Brotherhood of Santo
Elesbão and Santa Efigênia treat a range of subjects but in a very general way.
The ensuing revisions and additions appeared in response to specific practi-
cal problems, and also more clearly spell out the powers and responsibilities
of the board.∏∂ The initial casualness of the document was an expression of
the group’s smaller and less formal structure; at the time, the brotherhood
180 | Chapter Five
met in a private house that the compromisso did not identify. After the
brotherhood transferred to the Church of São Domingos the number of
members increased, and along with it the need for more complete and
rigorous rules. The four chapters added in 1740 brought new detail to what
had only been outlined before, especially with respect to how money was to
be collected and spent, women’s participation, masses for the dead, disci-
plinary measures, and board positions. The eight chapters that appeared later
speak to those themes, and others, but noticeably concentrate power and
benefits among the board members. The final addition to the compromisso
reinforces the brotherhood’s hierarchy through the di√erential distribution
of masses. But during this period the kingdoms of the folia had been in-
stituted as new, alternative hierarchies in the order—in part to help alleviate
tensions wrought by the growing power of the board.
Interrogated in a slightly di√erent way, the compromisso reveals that the
principal di≈culties the brotherhood had to confront over the period were
these: the role and methods of the treasurer, women’s participation, inter-
ference or unreliability from the chaplain (or bishop), clashes between
subgroups within the order, and the absence from a brother’s funeral of his
fellows.∏∑ The series of solutions created for these problems show that over
time, the brotherhood transformed into a relatively well-o√ and confident
organization, building its own church, tackling internal discipline, and
confronting the ecclesiastical authorities over their obligations to the
brotherhood.
In 1740, when the brotherhood initiated the process for formal recogni-
tion by the church, it had around seventy members. That total certainly
grew over the years, and one indication of how much comes from a refer-
ence to the folia. The folia held meetings of the emperor and the multiple
kings, each with a court drawn from his nation, and these must have been
bustling a√airs. They should be held at the church, according to the com-
promisso, to avoid ‘‘convocar tanta gente em sua casa . . . que faz suspeitar
entre a vizinhança’’ (loading so many people into the emperor and king’s
houses . . . as to create suspicion among the neighbors). The Mahi Manu-
script records one reunion of the Mahi Congregation’s folia that brought
200 people to the church.∏∏
Daily life changed for many in Brazil after 1750, with the progressive
implementation of the policies of the Portuguese minister Sebastião José
de Carvalho e Mello (later the Marquis of Pombal). The tone of colonial

administration grew more interventionist and centralizing. This a√ected
di√erent sectors of the population in di√erent ways, of course, but the
black brotherhoods were soon being scrutinized for their alleged spirit of
independence. Both ecclesiastical and secular authorities demanded trans-
parency, order, and humility. In the 1760s, many compromissos sent by
nascent brotherhoods in Brazil to Lisbon were denied in what seems to
have been a demonstration of monarchical control. During this same pe-
riod the crown also produced various provisions and other legal instru-
ments rea≈rming regal authority over the brotherhoods; the two provi-
sions from King D. José for the Brotherhood of Santo Elesbão and Santa
Efigênia are examples.∏π
In 1765, a royal letter was sent to all the brotherhoods within the Bishop-
ric of Rio de Janeiro informing them that they were to send their compro-
missos to Lisbon to be evaluated and confirmed by the Tribunal of Con-
science and Orders. When the compromisso of the Brotherhood of Santo
Elesbão and Santa Efigênia arrived at the bishop’s chambers in Rio, a
functionary there, the purveyor of chapels and residuals, found evidence of
a previous oversight: the document had not been sent on to Lisbon for royal
approval in 1740. It had made it as far as the desk of Rio’s Bishop D. Antônio
Guadalupe, who had approved it, and there the matter had stopped. This
was not an uncommon occurrence; many Brazilian brotherhoods’ compro-
missos had not been placed on board ships to Portugal before the 1760s,
even though o≈cial procedures mandated that the crown had the final
word on a brotherhood’s petition for recognition.∏∫ So in 1765, the revised
compromisso—all thirty-two chapters, including the addition and the five
chapters describing the folia—embarked across the Atlantic for the first
time.∏Ω They received o≈cial approval in a provision signed by King D. José
on 11 March 1767; the provision a≈rmed the brotherhood’s creation and
revalidated its license, adding one note and several changes to the compro-
misso.π≠
King D. José’s note emphasizes that the crown had full power over this
and every other lay brotherhood in Brazil (which was a standard announce-
ment from Lisbon at the time). Comparing the provision’s modifications
with the version of the compromisso that had been revised to accommodate
them, it is clear that the king’s changes had been only partially adopted by
182 | Chapter Five
the brotherhood. But this was enough to get that revision of the compro-
misso decisively approved, finally, by Bishop D. Frei Antônio do Desterro in
Rio de Janeiro. The major change appears in a new formulation of chap. 10
to allow Angola and creoles to become members of the brotherhood. An-
other royal correction, stipulating a reduction in the amounts of esmolas
charged, was not incorporated into the last revision of the compromisso.π∞
6 | conflict and ethnic identity among mahi
In its last five chapters, the compromisso of the Brotherhood of
\ Santo Elesbão and Santa Efigênia authorized the election of up

to seven kings to compose the court of the Imperial State of
Santo Elesbão. The only condition imposed for the selection of kings was
that each reign should represent su≈cient numbers of members within its
own court, who in turn should hold su≈cient amounts of personal wealth
to support the reign’s material needs during the term in o≈ce. The docu-
ment cites the importance of ‘‘bens móveis como de raiz,’’ or both portable
and stationary property; a good indication of what those terms probably
referred to can be found in the wills of freed Mina slaves.∞ Beyond that
criterion, those five chapters, added in 1764 to sanction the creation of the
folia within the brotherhood, did not provide details about how the reigns
should be formed or organized. I did not find evidence of the existence of
‘‘empires’’ previous to 1764, only reigns. However, based on a small but
rich body of documentation, it is certain that even by the 1740s, the election
of kings and queens was common practice in the brotherhood. The Mahi
Manuscript mentioned two kings of the Mina Congregation (one of them
was Pedro da Costa Mimozo, already a king in 1748, and the other Cle-
mente Proença, who followed Mimozo). The first king of the Mahi Con-
gregation was elected in 1762.
The Mahi Manuscript is perhaps the most important document avail-
able to shine light on the lives of slaves who had departed from the Bight of
Benin to enter the world of slavery in colonial Brazil. The document was
184 | Chapter Six
written in or around 1786, in the structure of a dialogue. Its authors were

two men identified as Mahi: Francisco Alves de Souza, a freed slave and the
leader of the Congregation, and Gonçalo Cordeiro, who held an alferes
(second lieutenant) in the Black Regiment as well as being Souza’s secre-
tary and right-hand man. This Manuscript also transcribes the compro-
misso of a subgroup of the Mahi Congregation, the Devotion to the Souls
of Purgatory, but it is particularly useful for helping us comprehend the
presence of small, self-identifying ethnic groups within the Brotherhood
of Santo Elesbão and Santa Efigênia. A second document critical to my
analysis was the 1788 compromisso of the Fraternity of Our Lady of the
Remedies, which was written by the principals of the Devotion to the Souls.
Although constituted by the same people, the two organizations had dif-
ferent objectives and rules. The earlier group, the devotion, focused on the
salvation of the souls of deceased brothers, as well as praying for ancestors
who had not been converted. The later group, the fraternity, was dedicated
to caring for any Mina Africans in the city who were poor, aged, or infirm. It
may appear odd at first glance, but the creation of parallel, distinct group-
ings out of a single constituency was part of a strategy by Francisco Alves de
Souza and his allies to orient and articulate the Mahi as a formidable
faction within the Brotherhood of Santo Elesbão and Santa Efigênia.
Whatever else the devotion and the fraternity had in common, they both
explicitly commingled the formulation of o≈cial norms with the infor-
mality of daily life. In Brazil, according to Schwartz, both government and
society are structured according to ‘‘two interlinked systems of organiza-
tion’’: metropolitan administration (based on bureaucratic norms), and a
‘‘web of primary interpersonal relations’’ that is not o≈cially recognized.≤
The coexistence of established, institutional rules and informal relation-
ships permeates social calculation and social action in Brazil, creating a
universe of possibilities in which o≈cial versus uno≈cial, formal versus
informal, impersonal versus personal, and collective versus individual cir-
cumstances are unpredictably hybridized.
This final chapter focuses on the informal groups that cohered around
ethnic criteria inside the Brotherhood of Santo Elesbão and Santa Efigênia.
In particular I am concerned with the kingdoms of the folia, seven of which
were allowed (and regulated) by the statutes of the brotherhood. I hope to
show how the Mahi reconstructed its identity as an ethnic group within the
Conflict and Ethnic Identity among Mahi | 185
strictures of a milieu dominated by captivity, but in which new options for

social organization and cultural expression were also present—largely (and
it would seem paradoxically) through the avenue of Catholic religious
activity.
Francisco Alves de Souza’s narrative of the Mahi Congregation begins
with his arrival from Bahia in Rio de Janeiro in 1748, where he found a
number of Mina blacks organized in the Mina Congregation.≥ This con-
gregation had already formalized its separation from the other nations that
made up the Brotherhood of Santo Elesbão and Santa Efigênia (Mozam-
bique, Cape Verde, São Tomé), which was still based at the Church of São
Domingos. The Mina nation gathered Africans from di√erent ethnic
groups, with such names in the documents as Savalu, Mahi, Agonli, Daho-
mey, and Iano, all of whom spoke what colonial observers called the ‘‘gen-
eral tongue of Mina.’’ At the time, the congregation had as its king Pedro da
Costa Mimozo; at the end of his term he was followed by his ally, Clemente
Proença, in a highly contested election. Proença’s time in o≈ce was frac-
tious, as the previous chapter shows, with subgroups increasingly jostling
for power within the Mina Congregation.∂
As Souza told it, the troubles within the congregation began to manifest
because the Dahomeans, the most powerful subgroup and hence in control
of the leadership, began to abuse their majority position at the expense of
other subgroups. There were clashes in which, it was alleged, ditos picantes
(heated words) were uttered by the Dahomey majority. Scandalized by the
a√ronts and excesses of the Dahomeans, Souza wrote, the Mahi and the
other smaller subgroups saw no dignified option other than to abandon
the Mina Congregation. In 1762 they created their own Mahi Congregation,
and elected Ignacio Gonçalves do Monte as their king.∑ Sometime after
that, Monte composed a Termo (a written document) that has regrettably
been lost, but references to it were made elsewhere because Monte appar-
ently included the bold declaration that he would serve as leader of the
Mahi Congregation until his death. These events are the first substantive
indication that the later reigns of the folias were constituted according to
ethnic groups organized within the Brotherhood of Santo Elesbão and
Santa Efigênia. For a while, at least three smaller subgroups—the Savalu,
Agonli, and Iano—remained united within the Mahi Congregation, under
its Mahi leader. But at a certain moment, the Savalu and Agonli each
186 | Chapter Six
created their own reigns. Why the Iano seemed not to do so remains
unclear.∏ The so-called alas (wings or flanks, actually factions) of the fu-
neral corteges and festive processions mentioned in the compromisso of
the Brotherhood of Santo Elesbão and Santa Efigênia were probably com-
posed by these groups, which had organized around their shared devo-
tional and ethnic sensibilities. Of these subgroups, the only one for which I
encountered significant information was the Mahi, and it is clear that they
had not only their congregation but, within it, the two devotions discussed
in chapter 5 (Almas do Purgatório and Our Lady of Remédios).
The complex distribution of new titles as part of the folia, the growth in
membership of the brotherhood, and the construction of the Church of
Santo Elesbão and Santa Efigênia were contemporary and interrelated
occurrences. Just as the creation of a separate board for women was in-
tended to resolve some of the brotherhood’s growing pains, the creation of
the Imperial State also was an attempt to redistribute participation and
prestige to ameliorate conflicts within the brotherhood. In that sense the
inclusion of the folia and its titles in the compromisso in 1764 gave formal
recognition to the presence of the (or at least seven of the) subgroups,
giving each of them the guarantee of an election for their own king. But it
would seem that the first election of the folia’s emperor and kings took
place before ecclesiastical authorities gave their approval to the Imperial
State in 1764. The election of the Mahi king in 1762 was most likely an
expression of this new configuration of power in the brotherhood, suggest-
ing that other kings would also have been elected at the same time.
Negotiations between the brotherhood and church o≈cials over the
content and phrasing of their compromisso were obviously important for
arriving at a document that would serve as a point of reference for the
future. But as I have argued, many of the particularities of the compro-
misso were included after the fact—that is, once new needs and problems
had already arisen, demanding the implementation of solutions that only
later would be written down as a prescriptive set of regulations. The Impe-
rial State o√ered solutions to many immediate internal conflicts in which
di√erent groups sought material and symbolic advantage: roles and titles,
space in the church, access to donation funds, the ordering of public
processions, and so on. In their increasingly fine segmentation, the reigns
accompanied the processes of social organization of Africans in the city of
Rio de Janeiro over the years: initially composed of Africans from diverse
regions (but excluding Angola), as the brotherhood grew its members

articulated and reinforced more restrictive boundaries of identity and in-
corporation; finally the folia, with its separate kingdoms, embodied the
assertion of ethnic di√erence.
Given that the Dahomeans were prominent enough in the brotherhood
for the rest of the subgroups to unite against them, it is reasonable to
suspect that they had first held the position of king when there was only
one such title available. The other subgroups would have had to dispute the
hierarchy of lower ranks (counts, dukes, etc.) among them. With the
institution of the Imperial State, the majority Dahomeans would have
reserved the privilege to elect the emperor, and the title of king was dis-
tributed as a gesture of equivalence among the other subgroups. Unfortu-
nately, little is known about the kings in general, or about how each
subgroup might have used the Imperial State di√erently to communicate
an ethnic identity. Perhaps more research will uncover new sources, but at
present the only king whose story can begin to be told is the Mahi Con-
gregation’s first king, Ignacio Gonçalves do Monte.
the reign of the ‘‘mina mahi’’

Monte was elected in 1762 amid praise of his ‘‘true Mahi’’ character, and the
Manuscript suggests that he was responsible for not merely inspiring but
growing the congregation. The Manuscript also establishes lines of con-
tinuity between the Mahi Congregation and its two devotions, and presents
Francisco Alves de Souza as Monte’s heir as the political, material, and
symbolic leader of the group.π More specifically, we learn that the two kings
who led before Monte had spoken Portuguese with di≈culty, and had
governed bocalmente—that is, orally—without producing written documen-
tation of their priorities and mechanisms of governance. Monte had writ-
ten a declaration of his term and a brief outline of regulations for the
congregation. But Souza had produced a more complete set of vows and
statutes, in a manner that e√ectively displayed his capacity with the Por-
tuguese language. Souza had been involved with the Mina Congregation
since 1748, and by 1762 (when the Mahi Congregation was created) he was
obviously a prominent member since he had been chosen as an imediato, or
adjunct, of King Monte.∫
At a date that was not recorded in the Manuscript, Monte fell mortally
188 | Chapter Six
ill. On his deathbed, he summoned Souza and placed in his hands the
leadership of the congregation. This transfer of power was carried out in
the presence of witnesses, said pessoas de crédito (with reputations for hon-
esty) and considered appropriate company for the spectacle of a king’s
death.Ω Souza accepted Monte’s wishes and served, with the title of sub-
stitute, as the congregation’s leader until Monte’s death on 25 December
1783. The Mahi Manuscript indicates that the passing of Monte opened up
a series of conflicts within the Mahi Congregation that persisted until 1788,
and probably later.
Unlike many other compromissos or statutes that merely state rules,
and thus camouflage the tensions that gave rise to them, the dialogues of
the Mahi Manuscript more openly address the concerns and disputes that
pervaded the congregation. Of course, we must keep in mind that Souza
and Gonçalo Cordeiro—his co-author, secretary, and confidante—might
not have applied the strictest standards of objectivity to their presentation
of these subjects. The two men were old friends; Souza referred to Cordeiro
as ‘‘my dear and loyal comrade, to whom I have given the most loyal
friendship since infancy.’’∞≠ This statement raises immediate questions.
When was this infancy? How long was it understood to last, and how and
where was it passed (Africa or Bahia)? There is also no mention of time
spent in captivity, and whether this separated the two men or how it
otherwise a√ected them. Cordeiro stayed close to Souza during the trou-
bled years after Monte died and the succession to Souza was challenged; he
took on other powers as well, being elected secretary of the Devotion of
Almas and was involved in initiating the composition of statutes for the
Fraternity of Nossa Senhora dos Remédios.
who should have the keys to the safe?

A critical source of conflict was control of the congregation’s co√ers.
When Monte died, this control passed to his widow. The safe held the
king’s own donation as well as other contributions toward the expenses
associated with the king’s responsibilities to the brothers. Unlike the
Brotherhood of Santo Elesbão and Santa Efigênia, in which these internal
funds derived from contributions from individual brothers according to
the obligatory donations whose amounts were specified in the compro-
misso, gathering funds for the Imperial State and the kingdoms were
largely the responsibility of the emperor and the kings themselves. That is
why the five chapters dedicated to the folia in the brotherhood’s compro-
misso make clear that those brothers elected to these high positions in the
folia should have access to patrimony su≈cient to sustain the activities of
the o≈ces. In the Mahi Congregation, looking after the safe was the
responsibility of the queen; Monte’s wife was therefore probably in charge
of the safe during the whole twenty years of his royal tenure. When he died,
she refused to turn it over to the congregation. The circumstances of that
conflict encapsulate the tension between personal and group allegiances
when property of both financial and symbolic value was at stake.
Her opponents pointed out that with the king’s death, her own term as
queen ended, and therefore the safe and its key should be returned to the
group. At play in this dispute was not only the safe, although it was the
fundamental object of interest; there were other items, including books, a
cloth purchased in the store of one Antonio Ramalho Lisboa, and a large
African pano da costa (left to the congregation in the will of Ignacio da Silva
Roza, it was used to cover the table of the church consistory on the days the
members paid their donations).∞∞ For her part, Monte’s widow countered
that all the source of all those items (and most of the money) was dinheiro da
finta, or voluntary contributions, during the time of Monte’s reign. She
argued that since she was the heir of King Monte, she stood to inherit the
products of voluntary contributions made to him or in his name.
In the compromisso of the Brotherhood of Santo Elesbão and Santa
Efigênia, chapter 5 of the folia section refers in explicit terms only to the
king’s donation but not to any such obligatory payment from other mem-
bers, and that is why a donation from a brother was considered finta, or
voluntary. This case shows the growing centralization of power in the
hands of the king (and his queen), but also reveals the absence, from both
the larger brotherhood and for that matter the church, of formal mecha-
nisms of control over the congregations or reigns. The informality of
norms with respect to the kings’ responsibility and accountability provided
Monte’s widow with the space to maneuver against the congregation who
elected him. But she was too smart to merely argue. Soon after Monte’s
death, Souza removed himself from the congregation for a period of four-
teen days, due to his own bout with a disease: erizipela.∞≤ Taking advantage
190 | Chapter Six
of his absence, the queen convened an assembly of the congregation in the

consistory of the Church of Santo Elesbão and Santa Efigênia. Helped, it
seems, by a few loyal operatives, she managed to bring others to the
reunion from outside the congregation—representatives of the Mina and
other nations. The outcome of this meeting was her election as queen of
the Mahi Congregation, in substitution of the king. In 1762, she had been
crowned because her husband was elected; now, she was elected queen
herself with no king alongside.∞≥
The diversity of nations present shows that the dispute for succession
among the Mahi was of interest to Mina and other sectors of the Brother-
hood of Santo Elesbão and Santa Efigênia. The queen apparently counted
representatives from various nations among her allies too, not just some of
the Mahi. Souza was careful to allege that she had called the meeting under
false pretenses, which might have gone far to explain its high attendance:
to collect donations for the soul of the fallen king. Once all the brothers
were present, she turned the tables and proposed her own election. He
noted that many of those present quickly perceived what was happening
and distanced themselves from her socially (‘‘fugiram dela no mesmo
dia’’), although one man unnamed in the sources (a creole from Bahia)
assisted her with the plan and ultimately lived with her.∞∂
The fact that prominent individuals such as the queen, and Souza, had
networks of alliances reaching beyond the Mahi Congregation suggests
that members of the brotherhood organized themselves not only in formal
groupings (ethnically based reigns) but also in political factions that could
cut across the lines of ethnicity or nation. The dispute over the safe reveals
that, in this occasion, the Mahi were divided into at least two factions—
those supporting the king’s widow as queen, and those more aligned with
Souza, still serving as director but representing a legitimate and complete
change in leadership. Each faction sought support outside the circle of
Mahi in the larger brotherhood. The tumult was probably substantial. Al-
though the Manuscript does not record the number of people in attendance
at the assembly called by the queen, Souza (author of the Manuscript) did
later write that at his own election 200 members of the Mahi Congregation
were present, along with people from other nations in the brotherhood.∞∑
The Manuscript presents a dialogue between Souza and Cordeiro that
depicts Souza as reluctant to assume the promise he made to Monte on the
king’s deathbed to take over leadership of the congregation. He is shown

as wanting to avoid confronting Monte’s widow, who is portrayed as an
‘‘obstreperous enemy of peace and success’’ responsible for the earlier
schism between the Mahi and the Savalu and Agonli. Despite Cordeiro’s
pleas, Souza insisted on remaining the director and not pursuing the
crown of the king of the congregation. Souza maintained that while he
respected Monte’s wishes, he also knew that the queen had many support-
ers in the congregation; and of these, many were involved in ‘‘abuses and
superstitions.’’ Although this startling accusation is not pursued in the
text, it would seem that Souza did not want to confront the queen precisely
because he feared the e√ects of what in context referred to witchcraft. If
Souza tended to believe in witchcraft or curses, his own sudden sickness
once he became Monte’s successor might have given him cause to worry
about his fate if he confronted political opponents too directly.∞∏ But he
would have to avoid making too pointed an accusation in this respect. By
accusing the queen of witchcraft (or any other ‘‘heathen’’ practices and
beliefs), he would in e√ect be implicating a number of her closest allies as
well, and that could invite problems for not only the congregation but the
entire brotherhood. Instead, he took the subtle approach of raising the
subject in a di√erent way: he declared in the Manuscript that all Mina are
God fearing, and claimed it was the Angola who, by insisting on carrying
out their primitive pagan practices, led the white community in Rio de
Janeiro to think that all pretos (Africans) were given over to superstitious,
backward nonsense.∞π
As the dialogue proceeds, Souza proposed that the congregation should
select another person from within its ranks to serve as king. Cordeiro
lamented his friend’s demurral, and warned that the queen could draw on
her allies to help her take control of the congregation. Significantly, he
argued that a woman had never occupied such a lofty position; this con-
firms the notion that the queen wanted to govern in the place of the king,
and not merely at the side of a successor to her husband. Cordeiro went on
to explain that women were permitted to hold the o≈ce of judge, because
of their number within the group and their ability to pay the substantial
donation, but that this contained no guarantee that they could ever govern
the congregation. Drawing on a story of the origins of the church itself to
back up his claim, Cordeiro said that women lack the robusteza (robustness)
192 | Chapter Six
of men for high leadership.∞∫ Souza countered that Cordeiro may as well
dispense with the arguments, because he would not pursue the kingship.
Not sitting idly back, other members of the congregation got involved.
Frustrated with both the coronation of the queen and the continued re-
fusals of the director to confront her, a group of brothers calling them-
selves the Notables (maiores) sent a solicitation to the highest colonial court
in Rio de Janeiro, the Tribunal da Relação (High Court), with two petitions:
first, that Souza be obliged to assume leadership of the congregation; and
second, that the queen return the safe and the other disputed items to the
congregation.∞Ω The court was amenable to the Notables’ request, or at
least the first part of it. On 9 March 1784, less than three months after
Monte’s death, a lower o≈cial from the court (the meirinho das cadeias)
appeared before Souza to notify him of the proceeding initiated by the
Notables, and to inform him that a minister of the court had ordered Souza
to report to him in person to resolve the matter. Souza then agreed to
accept the kingship, which prevented the indignity of his being brought by
force to the High Court. The court o≈cial also informed Souza that the
queen had sent her own solicitation for a hearing, and the matter was
pending.
Lest we infer that the Notables were simply admirers of Souza within the
congregation, a sort of fan base anxious to see him ensconced in power, a
small conflict would soon arise between them and Souza himself that
revealed their interests in the matter. But first, Souza a≈rmed to his loyal
friend Cordeiro that he would take the leadership of the congregation; he
also stated that he wanted to address the assembled brothers, to discuss
the ‘‘poor style’’ and ‘‘malice’’ to which recent events had exposed him.≤≠
This seems not to have been an allusion to the queen (whom he deeply
feared) but to his own supporters, who had grown increasingly strident
and even threatening toward him as his reluctance to accept the leadership
had persisted. Just before taking power, Souza stunned the Notables—who
had brought him into o≈ce—by announcing that he accepted his new
o≈cial responsibilities but intended to refuse the title of king and stay with
the less formal charge and title of regente (regent), or associate king. Cor-
deiro, aware of the Notables’ ardent disapproval of this idea, tried to reason
with Souza by pointing out that the use of the title of king was a tradition
extending to dos primeiros fundadores (the founders of the brotherhood).
Souza countered that he preferred another title, one more consonant with
his ‘‘profession’’ (which, regrettably, remains unknown to us). And he
made the remarkable argument that the kingly title ‘‘perturbs good har-
mony and devotion.’’ But it emerges from the Manuscript that the Notables
were far less interested in Souza the individual than in having a king in
front of the congregation. Without a king, the entire hierarchy of royal
titles in the folia (princes, counts, dukes, marquis, generals, and so forth)
had no basis for existence. The court structure descended from the king as
the major title and authority—and not from the queen. Without the king (as
the head) and his court (the body), a critical mechanism of social distinc-
tion within the congregation and the brotherhood would disappear. The
Notables asserted that the whole purpose of the hierarchy of postos e nomes
(positions and titles) in the court was to ‘‘distinguish the large from the
small, the greater from the lesser, the exalted nobleman from the lowly
mechanic; and to maintain the proper respect between them all.’’ Souza
and the Notables were formidable opponents. The crisis was finally re-
volved with the o≈cial extinction of the title ‘‘king,’’ which was replaced by
‘‘regent,’’ but it was clarified that all the aspects of the king’s position
would be maintained—including the court. The Notables were satisfied
with this compromise since they did not lose, as they put it, ‘‘any of the
rights and small regal privileges we have enjoyed for so long.’’≤∞
According to the Mahi Congregation’s hierarchy, just beneath the regent
was his substitute. These two high o≈cials were followed by brothers who
held titles or duties; after them came brothers who carried only ‘‘names.’’
Finally, at the bottom was the large group of brothers who lacked any of the
congregation’s formal social distinctions. Titles and duties were, as the
terms suggest, rather di√erent. A duty was a function or position that was
awarded by election and evaluated according to regulations, such as judge
or secretary. A title was an honorary distinction conferred upon individuals
deemed worthy by someone holding a superior rank; worthiness was a
quality typically measured in personal wealth and prestige. The actual titles
used drew from the ranks of European nobility (‘‘como se dá cá na terra dos
brancos’’), but sometimes added an African element as well. In contrast,
duties and names were usually given in the general tongue of Mina ‘‘à
imitação dos fidalgos de nosso reino de Mahi’’ (in imitation of the nobles of
our Mahi Kingdom).≤≤ Altogether the combination of titles, duties, names,
194 | Chapter Six
and languages created an intricate hierarchy, part of which, from around

the time of Souza’s acceptance of leadership, is reproduced here.≤≥ Note
that the same person could hold several attributes at once; and one person,
Boaventura Fernandes Braga (6, 12), held attributes corresponding to two
di√erent ranks. Rita Sebastiana was Souza’s wife.
name rank
1. Francisco Alves de Souza Regente (Regent)
2. Rita Sebastiana Regenta (Female Regent)
3. João Figueiredo Vice Regente (Vice Regent)
4. Antônio da Costa Falcão 2\ Vice Regente (2nd Vice Regent)
5. Gonçalo Cordeiro Secretário (Secretary)
6. Boaventura Fernandes Braga 2\ Secretário (Secretary)
4\ do Conselho (4th Counsel)
7. Luiz Rodrigues Silva Procurador (Procurator)
Aggau (General)
8. José da Silva Aggau (General)
9. José Antônio dos Santos 1\ do Conselho (1st Counsel)
1\ Chave (1st Key)
10. Alexandre de Carvalho 2\ do Conselho (2nd Counsel)
2\ Chave (2nd Key)
Eiçuûm Valûm (Duke)
11. Marçal Soares 3\ do Conselho (3rd Counsel)
3\ Chave (4th Key)
Aleolû Belppôn Lifoto (Duke)
12. Boaventura Fernandes Braga Chave de Dentro (Inner Key)
Aeolû Cocoti de Daçâ (Duke)
13. José Luiz 5\ do Conselho (5th Counsel)
Ajacôto Chaul de Za (Marquis)
14. Luiz da Silva 6\ do Conselho (6th Counsel)
Ledô (Count)
In the company of Cordeiro and another brother,≤∂ Francisco Alves de

Souza reported to the consistory of the Church of Santo Elesbão and Santa
Efigênia on 13 March 1786 for the ceremony marking his rise to the highest
o≈ce of the Mahi Congregation. Seeing so many people gathered there,
including the Notables, he feigned surprise. His modesty was disingenu-
ous, given his involvement in the intense mediations behind the scenes to
arrive at this juncture. As secretary of the congregation Cordeiro went so
far as to highlight four of the Notables’ names in his account of the
occasion, noting respectfully that all were freed men. (Their names appear
as seventh, eighth, ninth, and tenth in rank in the list above.) The names of
Souza and his wife were registered as Regents in the record book of the
Brotherhood of Santo Elesbão and Santa Efigênia, and some forty of the
brothers in attendance signed the inscription. Before signing the book
himself, Souza delivered some brief remarks on the original motivations
for the founding of the congregation.
From the beginning, this land has brought very many African blacks from
the Mina Coast and Angola. But because of the cruelty and inhumanity of
some of the men who bought and owned them, when the slaves became
wounded or sick or elderly, they could do nothing but lie on the ground
naked and unprotected to await a death from hunger and the cold. They had
no one to bury them. If the owners refused to call on the Santa Casa da
Misericórdia, with its well-known zeal and charity, to collect and bury these
unfortunates[,] their cadavers remained exposed and illegitimate. For this
reason the blacks themselves entered into a group or corporation with the
intention to properly serve their compatriots at such a time. That is, when
one of our nation dies, his parentes will collect money to bury him properly,
and arrange for masses for his soul.≤∑
The usual translation for ‘‘parentes’’ would be relatives, but in the

context the word means people of the same nation, indicating that burials
were under the responsibility of the provenience group. Although inflected
with sarcasm in its description of the interment practices of Santa Casa da
Misericórdia, in its overall themes Souza’s speech was a standard one for
the occasion of taking high o≈ce in a black lay brotherhood at the time.≤∏
Souza also was communicating his personal intention to focus attention
and resources on the brothers in greatest need. Once in o≈ce Souza spent
his spare time outlining statutes, not only for the Mahi Congregation but
for the two groups about to be created within the congregation—the Devo-
tion to the Almas do Purgatório (1786), and the Fraternity of Our Lady of
Remédios (1788). As noted at the beginning of this chapter, the former
group was concerned with the souls in particular of deceased Mahi broth-
196 | Chapter Six
ers, while the latter hoped to ameliorate the e√ects of poverty among all
Mina in the city of Rio de Janeiro. There is no way to know whether these
two groups were Souza’s innovation, or whether they had been a vision of
King Monte’s that Souza was now striving to fulfill. Because the groups
were within the umbrella Mahi Congregation, both sets of statutes would
refer to its larger organizational structure; they were careful to a≈rm that
the congregation’s safe should possess three locks with three di√erent
keys, to be distributed among members of the congregation’s board (to
ensure that only in the presence of all three people could the safe be
opened). The current safe, held by the widow, had only one lock and key.
The fate of that safe had still been hanging in the deliberations of the
High Court with respect to the petition of the former queen, Monte’s
widow. The court had taken longer on this problem than on Souza’s
responsibilities to the congregation. Finally, the Mahi Congregation’s rep-
resentative informed Souza that the court hearings were about to be con-
cluded and that a judgment was expected soon. Several days later the
representative appeared again before Souza, this time much distressed.
The court in its wisdom had sided with the widow, he reported. She would
stay in possession of the safe and the other associated items, and there was
no possibility of appeal. Delighted and inspired with her victory, the widow
let it be known that she was interested in the position of empress of the
Mina Coast, the highest position a woman could attain in the Imperial
State of Santo Elesbão. She also made a copy of the High Court judge’s
verdict and delivered it to the viceroy, alleging that the decision was being
disrespected by allies of Souza in the Mahi Congregation who were im-
properly withholding the delivery to her of donations that had been piling
up since Monte’s death. This statement suggests that the widow had ambi-
tions to control not only the safe, which ostensibly contained gifts to her
husband, but the congregation’s new income—and hence its capacity for
any financial transactions—as well. She went further, accusing Souza of
being the cabeça de motim (head of an insurrectionary movement) to foist
himself on the majority of the Mahi Congregation, who in fact wanted her
as regent. That was a gravely serious accusation, one of the most serious
that could be made against someone within not only the Mahi Congrega-
tion, or the Brotherhood of Santo Elesbão and Santa Efigênia, but the
political imaginary of Portuguese colonial society itself. According to the
brotherhood’s compromisso, a member accused and found to be guilty of

mutiny was immediately subject to a range of disciplinary measures from
the board (from public reprimand, to the humiliation of carrying the rock
back and forth, to expulsion). The viceroy was su≈ciently persuaded by the
widow of Souza’s malfeasance for him to prohibit Souza from appearing
with the rest of his folia during the street processions of both the Church of
Santo Elesbão and Santa Efigênia and the Church of Our Lady of Rosário.
Souza’s absence stunned his supporters and seems to have been a blow
unprecedented within any folia in the brotherhood. It e√ectively prohibited
his folia, the Mahi court of the Imperial State of Santo Elesbão, from
carrying out the traditional collection of public donations during the pro-
cessions.
It should be noted that Souza had recently clashed with his superiors on
the brotherhood’s board precisely on the issue of public donations, or how
they were collected. One of his declarations soon after accepting the posi-
tion of regent of the Mahi Congregation was that the standard practice of
collecting donations on the street to the por toque de tambores (rhythmic
accompaniment of drumming) should be done away with. But this contra-
dicted the compromisso of the brotherhood, which specified that acquir-
ing donations from the public should be carried out in the estilo antigo (early
style), that is, with the drums—and under the direction of women. Souza’s
instructions to his folia meant that he was publicly opposing the board of
the brotherhood, making him theoretically subject to o≈cial reproof.≤π
Adding to the complexity of this situation is the ethnic dimension, since
the Mahi had originally broken away from the Mina Congregation some
twenty years before to get away from what they regarded as the domineer-
ing behavior of the Dahomeans, and now, the Dahomeans were still ex-
tremely influential in the brotherhood and seem to have been the most
formidable faction of support for the widow. Souza’s attempts to introduce
changes in the Mahi Congregation that went against the norms of the
Brotherhood of Santo Elesbão and Santa Efigênia may have been, as he
maintained, in the name of a more rigorous Catholic orientation for the
congregation. But those attempts, and the responses they received from
the brotherhood at large, cannot be analyzed without keeping in mind
ethnic identities as one factor.≤∫
Even after prohibiting Souza from appearing with his folia, the viceroy
198 | Chapter Six
expressed a persistent interest in the case. He ordered both Souza and the
king of the Brotherhood of Our Lady of Rosário to appear before him to
explore the matter further. Souza took the opportunity to ask for pity
because of his own pouquidade (unimportance) and inocência (in this case,
lack of familiarity with subtle codes of behavior) instead of punishment for
the baseless contrivances of the widow. His self-description was figurative.
Souza was certainly a prominent man in his community, he held a pres-
tigious position in the congregation, and we see from the documents that
he was intimately aware of his corresponding rights and responsibilities.
Some time after this meeting, in the face of the viceroy’s continuing skepti-
cism, Cordeiro made a shrewd and unexpected tactical maneuver to get his
friend o√ the hook. He explained that the sentence was invalid because the
‘‘congregation is a devotion that all participate in of their own choice, with
no obligation, because we have no statutes’’—in other words, the group
was informal and unrecognized in the eyes of secular or church law, hence
not subject to this type of intervention. This elegant argument took advan-
tage of the fact that even though the Mahi Congregation had internally
approved the statutes of the Devotion to the Souls (on 31 January 1786), the
document had never been sent on to the ecclesiastical authorities. The
Mahi Congregation had no o≈cial recognition. And the compromisso of
the Brotherhood of Santo Elesbão and Santa Efigênia, which provided for
the creation of the Imperial State and by extension the Mahi court, was of
little help to the viceroy in nailing down Souza’s status and obligations in
this nuanced conflict.
One notable aspect of social life within the brotherhood that this long
dispute helps bring to light is the strength and influence that women could
exert within a milieu where power was o≈cially concentrated among men.
In the Mahi Manuscript women were described as ‘‘vain connivers, disrup-
tors of peace and tranquility’’ (orgulhosas, amigas de enredos, pertur-
badoras da paz e do sossego) to be endured as a source of donations. But
women brought diverse strategies to bear on the limited space for par-
ticipation that men allowed them, and could make significant waves. These
strategies might involve collective mobilization, or for individual women
the playing on men’s fear of witchcraft; the widow probably leveraged that
against Souza, as well as her own strong personality and ambitions.≤Ω
In 1788 Souza and his allies decided to complete the statutes for the
Fraternity of Nossa Senhora dos Remédios. This second smaller group

drawn from within the Mahi Congregation was dedicated to assisting
those poor, sick, and aged Mina blacks who had no system of support
beyond wandering the streets begging. Once completed, the statutes were
sent directly to Dona Maria I, Queen of Portugal and Brazil, for her ap-
proval. This document, like the one drawn up earlier for the Devotion to the
Souls, took pains to state that the Mahi Congregation should have a safe
requiring three di√erent keys to be opened (to prevent the ‘‘inconveni-
ences’’ brought to the congregation by individual access).≥≠ But unusually,
the statutes also go on to mention the conflict over the present safe, which
was still held by Monte’s widow. This was done in a careful manner, in
language that, if the compromisso were accepted, would get the safe back
from the widow once and for all, with royal approval.
Chap. 12 described the safe that Souza and his supporters intend to
have. Its three keys would be divided among the congregation’s regent,
secretary, and procurator. Chap. 13 outlined the conflict with the widow
and asked for its resolution through whatever legitimate steps were neces-
sary. The widow had gained unjust control over the congregations’ fi-
nances, the chapter states, and the situation was the more grievous because
she was a woman and hence naturally incapable of fulfilling such a func-
tion (one must wonder what the queen of Portugal thought upon reading
that). More to the point, chap. 13 argues that Monte’s widow had no claim
or right to the inheritance of the king of the Mahi Congregation, only to the
personal e√ects of her husband.≥∞ Given these terms, if the queen approved
the statutes of the fraternity, the widow’s hold on the safe would be auto-
matically rendered illegitimate and it would have to be returned to the
leaders of the congregation, namely Souza and his allies.
The statute emphasizes that stewardship of the safe was an attribute of
the congregation’s queen, a privilege that Souza as regent wanted to pass
on to his wife. But the crisis of succession, unprecedented in twenty years
of the Mahi Congregation’s existence, focused on the ambiguity of the
social regulations of the brotherhood regarding the queen’s personal
rights and her responsibilities to the congregation independent of the
presence of a king. Chap. 13 was also phrased to raise the suspicion that
since Monte’s widow had used external means to maintain her hold on the
safe against the will of the congregation’s leadership (and she possessed
200 | Chapter Six
the only key), the contents of the safe might be at risk. While the safe
belonged to the Mahi Congregation, and all the members’ donations
should be paid into it, each smaller organization associated with the con-
gregation would have its own funds and objects of value which would have
been placed there as well. That made the oversight and administration of
the safe especially complicated, since di√erent groups would pool their
property within it. Thus whatever its other objectives, the creation of the
Fraternity of Our Lady of Remedies in 1788 seems to have represented a
strategy to increase the number of organizations relying on shared access
to the safe, adding weight to the argument that the safe belonged not with
Monte’s widow but with Souza’s wife, the female regent of the Mahi
Congregation. As can be verified from the signatures of the authors and
directors included at the end of the statutes, the fraternity was led by some
of Souza’s key allies and was designed to enhance his power in the Mahi
Congregation against the widow’s supporters. The new fraternity was
undoubtedly a response to the plight of Mina blacks who scraped out an
existence begging on the streets of Rio de Janeiro, but it was no less a
rejoinder to the (equally sophisticated) machinations of the widow and her
victory in the High Court two years earlier.
It is unclear whether these statutes, and their built-in appeal for royal
intervention, were sent to Lisbon before or after the widow sought out the
viceroy in Rio de Janeiro. According to standard procedures at the time, the
crown should be the last recourse in any dispute. This conflict had passed
through several levels of authority: beginning within the congregation, it
was carried to Rio’s High Court (by supporters of Souza); then to the
viceroy (by the widow and her supporters); and finally to the queen of
Portugal and Brazil (again by Souza and his people). The safe was at the
heart of the dispute, but much more than a small box of money and cloth
was at stake: the safe was the symbol of the social structure, accomplish-
ments, and ambitions of the congregation itself. Souza and Monte’s
widow, ably assisted by their operatives, were wrangling over control of
power within the Mahi Congregation—but also, and more fundamentally,
over the nature of that power.≥≤
the statutes of the devotion to

the souls of purgatory
The founding statutes for this subgroup of the Mahi Congregation con-
sisted of sixteen chapters. It was written by Souza and Cordeiro, formalized
within the Mahi Congregation on 31 January 1786, and transcribed in the
Mahi Manuscript.≥≥ In tone, the rules and regulations for this devotion drew
inspiration from the compromisso of the Brotherhood of Santo Elesbão
and Santa Efigênia, but there were divergences; these owed largely to how
the devotion fit into the larger organizational structure of the congregation
and the brotherhood.
According to the brotherhood’s compromisso, the election of an em-
peror was associated with the folia (the Imperial State of Santo Elesbão). It
is implied that every new emperor should establish a congregation of some
sort that would have its own complete board of o≈cials, although the
emperor would still be the dominant authority in political, symbolic, and
financial terms for that congregation.≥∂ The rank of king, one step below
emperor, was a position with di√erent priorities; kings were responsible
for payment of a substantial donation, but were not obligated to organize a
congregation.≥∑ The five chapters of the compromisso describing the folia
and the duties of the royal titles were approved in 1764, but the Mina
Congregation dates back at least to 1748 and perhaps even to 1740, the year
of the founding of the brotherhood. The statutes thus were written to
formally incorporate a preexisting organization within the brotherhood,
not to mandate its creation.
The appearance of the statutes of the Devotion to the Souls of Purgatory
in 1786 represented a means of institutionally recognizing the Mahi Con-
gregation, similar to the way the Mina Congregation had been. The Mahi
Congregation would have its own folia, which over time asserted increas-
ing independence from the Imperial State of Santo Elesbão (including
holding street processions at di√erent times). The Mahi Congregation
dedicated itself in general terms to ‘‘charity for the living and prayers for
the souls of the dead’’ (fazer caridade com os vivos e sufragar as almas dos
mortos), but in practice it was more inwardly focused on its ancestors and
it ethnic network, whether enslaved or freed. In special cases it would even
loan money to an enslaved member for the purchase of freedom, but again
only Mahi could hope to take advantage of this resource.
202 | Chapter Six
Recall that although the compromisso of the Brotherhood of Santo

Elesbão and Santa Efigênia indicated that a king was not required to
organize a congregation, it did permit up to seven kings, compared to one
emperor for the whole brotherhood. This meant that the man elected king
would typically be a prominent member of one of the eight or so compo-
nent nations (ethnic group or provenience group) within the brotherhood;
but internal conflicts were inevitable, since the members of a given nation
did not hold identical beliefs and values. Souza, elected the king and later
using the title of regent of the Mahi Congregation, was an unusually
energetic proponent of Catholic doctrine and ritual. It is instructive to
compare the statutes he helped write for the devotion with the compro-
misso of the brotherhood. The latter document makes no reference to
heathen practice or superstitions, but Souza was determined in his statutes
to cleanse the improper elements (‘‘tirar o mau estilo,’’ as he put it) he saw
in the norms of the brotherhood. The tenor of the dialogue between Souza
and Cordeiro exemplifies this preoccupation. Cordeiro suggested that it
causes him great su√ering to see his parentes still behaving as pagans,
resisting Catholicism, and fleeing the salvation that was their proper des-
tiny (‘‘fim para o que fomos nascidos’’). Souza responded by praising the
wonders they had received from the generous hand (‘‘liberal mão’’) of God.
He stressed that patience was necessary because ‘‘from the very birth of the
World [these wonders] are slow in arriving,’’ and he a≈rmed his faith that
one day all of them would be Christians.≥∏
For Souza, the underlying objective of the devotion, and by extension the
larger Mahi Congregation, was the displacement of any hint of paganism
and its substitution with Catholicism. Two points need to be made here.
First, Souza’s deep Christian sensibilities (and his willingness to enshrine
them in statutes) likely explain some of the vigorous opposition he faced
within the congregation. The fact that the compromisso of the Brother-
hood of Santo Elesbão and Santa Efigênia carefully avoid any reference to
paganism, primitive practice, or superstition is another indication that the
issue was polemical and needed to be negotiated carefully. Second, Souza’s
overt Christianity sat oddly with his overriding fear of the dark powers of
Monte’s widow. But again, Souza’s particular manner of expressing his
opposition to superstitions and witchcraft was not incompatible with be-
lieving in their force.
There were other di√erences between the statutes of this devotion and
the compromisso of the brotherhood. For instance, while only in chap. 10
of the statutes of the brotherhood’s compromisso are criteria for member-
ship outlined, this subject is addressed early and directly in the statutes of
the devotion. Chap. 2 of the devotion’s statutes emphasize that only Mahi
can serve on the board, while chap. 3 prohibits people from Angola from
joining the Mahi Congregation of which the devotion was a subset. These
statutes date from 1786, nearly twenty years after the royal request had been
received to alter chap. 10 of the brotherhood’s compromisso and remove
the prohibition of Angola and creoles.≥π The Brotherhood of Santo Elesbão
and Santa Efigênia had needed to make the change in their regulations,
becoming o≈cially more inclusive, in order to receive approval of their
compromisso from the Tribunal of Conscience and Orders and the Portu-
guese Crown. Despite having no clear documentary evidence, one can as-
sume that the early prohibition against Angola was not merely copied over
from the compromisso to the devotion’s statutes. Its reappearance in the
statutes years afterward suggests that the social exclusivity impossible to
maintain at the formal level of the brotherhood had quietly been transferred
to the separate congregations which made up the brotherhood (and which
did not require royal approval of their statutes). The result was the same in
any case, since Angola blacks would still have trouble joining any of the eth-
nic congregations based at the Church of Santo Elesbão and Santa Efigênia.≥∫
Souza was reluctant even to enter in contact with Angolans, claiming
that they were incorrigibly mired in heathenism, indecency, and sorcery. At
one point he leveled a specific accusation:
The Angola blacks have the abominable habit of removing the cadavers of
their dead kin from the tombs of the Santa Casa da Misericórdia, laying them
about the entryways of churches, and serenading them with pagan chants.
The Angola go on to exploit the deplorable spectacle of their dead to beg
money from passing churchgoers, claiming that the money is needed to
provide a proper burial. This is so common in the city that white men accuse
all blacks, whether Mina or Angola, of the same o√ense, and that is why I
strive to properly govern and protect my people.≥Ω
Elsewhere in the Manuscript he repeated the charge, highlighting the

Angola blacks’ duplicity and sacrilege while adding that an o≈cial had
204 | Chapter Six
ordered them to be apprehended and punished. The discrimination against

Angola blacks came to be applied to other African nations from the central-
western coast; Souza’s secretary Cordeiro admonished brothers for merely
‘‘socializing with Benguela people.’’∂≠
Although the writers of the Mahi Manuscript commonly used the words
abuso (profanity or indecency), gentilismo (paganism), and superstição (super-
stition), they expended much greater e√ort in detailing the punishments
such practices deserved than in actually explaining what those practices
were or how they could be recognized. But this was characteristic of texts
from the era. The compilation of regulations from Brazilian bishops in the
Constituições Primeiras also made use of the terms without defining them.
Reading these documents, one infers an implicit urging from the authors
that there was a broad public consensus or shared common-sense under-
standing of what was meant, rendering explanations unnecessary. There
was consistency in their usage: the context was always accusatory, and
involved circumstances in which ecclesiastical norms had been broken,
mocked, or otherwise deviated from.∂∞ But their persistent ambiguity in the
very documents that were intended to provide rules for social behavior left a
wide margin for the flexible interpretation and negotiation of individual
cases. And it remains unclear, despite the ominous tones, how often such
practices were identified and punished.∂≤ The Inquisition was more vexed
by heresy and Judaism than by paganism. Even still, many in Rio de Janeiro
in the 1780s would have remembered how Bishop D. Francisco de São Jer-
ônimo had mobilized against the New Christians (converted Jews and their
descendants) throughout the captaincy.∂≥ In the first volume of the Consti-
tuições Primeiras, the discussion of paganism conveys less a call for persecut-
ing than for indoctrinating the guilty parties. Souza (who was a devout
Catholic, regent of the Mahi Congregation, and de facto leader of its two
derivative orders), appears to have treated allegations of paganism among
the brothers under him in a way similar to how other authorities in colonial
society handled allegations of apostasy among the New Christians in Bra-
zil. Expulsion (or excommunication) should be the last recourse, turned to
only if the infraction against the church was too severe or if all reasonable
e√orts to instruct the accused in Christian doctrine should fail.∂∂
The compromisso of the Brotherhood of Santo Elesbão and Santa Efi-
gênia was written almost fifty years before the statutes of the Devotion to
the Souls of Purgatory and included no reference at all to paganism or

superstition. That fact, when considered in light of the internal controver-
sies sparked by Souza’s espoused e√orts to ‘‘remove the bad elements’’
from the devotion in 1786, reinforces the idea that Souza himself, not the
brotherhood at large, was responsible for bringing the issue to the fore-
front. The profile of these ‘‘bad elements’’ emerges from Souza’s writings
—drums in the public processions, witchcraft and sorcery, consorting with
Africans from certain nations (Angola, Benguela), desecration of crypts,
and pagan rituals involving dead bodies. One set of practices Souza did not
describe, but with which he was likely familiar, were the religious rituals
involving music, dance, and animal sacrifice (often called by the general
name calundu). These were very common in West Africa, especially among
the peoples around the Bight of Benin who provided significant numbers
of slaves to Brazil, but they have not been adequately explored in the
historiography of colonial Brazil.∂∑ Of the prominent cases known to
scholars of such practices in Brazil, few were in Rio de Janeiro. A famous
narrative from Nuno Marques Pereira (1625–1733) describes his visit to a
large rural property in Minas Gerais, where the singing, drumming, and
dancing emanating at a slaves’ calundu kept the tired traveler awake all
night. An intriguing case of a calundu in Bahia in 1785 involved Africans of
di√erent nations—Dahomey, Mahi, and Tapa/Nupe. Police records refer to
similar practices in Pernambuco.∂∏
That the calundus were a feature of life in Rio de Janeiro can hardly be
doubted, and it is suggestive that the (occasional) persecution promoted by
civil and religious authorities was internalized within the Mahi Congrega-
tion and its derivative groups. The congregation itself took on the obliga-
tion to identify such behavior among potential members, and prohibit
their membership if found guilty. Chap. 3 of the devotion’s statutes de-
scribes the procedure: ‘‘Everyone who wishes to enter this adjunct or
congregation (except blacks from Angola) will be examined by the secre-
tary and the aggau [general], who is also the procurator. These o≈cials will
verify whether or not the candidates, men or women, are involved with
heathen customs, paganism or superstition. Discovering that they are, they
will not be received here.’’∂π Again, what remains unclear is precisely what
would be prohibited, and the circumstances under which a brother (poten-
tial or active) might be denounced for such behavior.
206 | Chapter Six
Recent research into the interactions of the Inquisition with lay brother-
hoods has not been of significant help. We see from the records that
whatever the Inquisition’s concern with specific expressions of heresy or
profanity, when dealing with the brotherhoods its accusations tended to be
di√use—so vague even that crimes went undescribed, and the guilty un-
identified.∂∫ These delicate matters were usually resolved among the o≈-
cials of the brotherhoods. Details of expulsions, if they occurred, were kept
secret, and accusations within the brotherhoods were rare (Monte’s widow
was never accused of such practices, despite the certainty of Souza and
Cordeiro). Social and political considerations could increase one’s reluc-
tance to accuse a fellow brother of engaging in pagan practices, but the
consistency with which indications of guilt were allowed to discreetly
disappear also suggests that the brotherhood as a whole preferred to
distance itself from the whole subject in its documents. This could be
equally true for practitioners, and for those who feared them; Souza’s
attitude toward the widow of the deceased king reflected his belief in
witchcraft, and although he desired to win the case he did not want to risk
evoking what he feared might be her full wrath in doing so.
Francisco Alves de Souza publicly challenged witchcraft and pagan rit-
ual in the name of Christian faith, but he did so in part because inwardly he
trembled at their power. Souza stood at the intersection of several social
experiences, perspectives, and values that influenced his attitude. He was
an African, had been a slave, had attained his freedom, and had come to
serve the highest position of authority in the Mahi Congregation. Clearly he
was among the elite sector of Mina Africans in Rio de Janeiro in the
eighteenth century. He possessed personal wealth, was literate in Por-
tuguese (often demurring that he wrote poorly), knew arithmetic, and
adorned his speech with biblical citations in Latin. In one of his dialogues
with Cordeiro in the Mahi Manuscript, the topic under discussion was
geography. Cordeiro asked him to define the concept of a zona (zone) in
this context, and his answer is impressive: ‘‘It is a space on the known
terrestrial globe between two parallel circles . . . The globe contains diverse
climates, and geographers divide them into five zones or bands, drawn
according to their temperature. There are one torrid, two temperate, and
two cold or glacial zones.’’∂Ω When asked about the conquest of Africa, he
o√ered an expansive response.
Already during the reign of Senhor Dom Afonso V, the most faithful mon-
arch of Portugal, there was commerce along the coasts of Africa in Guinea
for ivory and gold . . . The discoverer of these riches was another Portuguese,
Senhor Infante Dom Henrique, or Henry the Navigator, third son of the first
King João. It was the desire of Dom Henrique to convert the infidels and
spread the sacred faith of Catholicism, as well as extend the reach of the
Portuguese crown. He was an astute geographer and mathematician, and
revealed the ignorance of many other scientists by showing the existence of
antipodes on the earth, and also showing that the torrid zone of the globe
was inhabited. When he went to Ceuta, Dom Henrique learned from the
Moors of a great desert region in Africa they called Caharâ or Sahara, and of
people there the Moors called azenegues. Those people, and blacks called
Ialof, marked the end of the territory of the Moors and the beginning of a
place the Moors called Guinacolo, or in the Portuguese tongue Guinea,
whose name comes from its principal city Genna which was famous among
traders and craftsmen for its fine gold. It sat just inland from the coast in a
part of Africa that was very remote from the kingdoms of Fez and Morocco.∑≠
The dialogues provided Souza with an opportunity to demonstrate the

scope of his knowledge in a range of subjects. At one point he mentioned
that he had a friend who was a helmsman, well acquainted with the water-
ways of western Africa, and this friend had taught him many details about
the Mina Coast; he went on to share a vivid description of the Volta River.∑∞
Souza showed himself to be familiar with Ilustração (Illustration), a Por-
tuguese version of Enlightenment that combines the new European think-
ing with Portuguese holy fervor. He reflected on his own conversion:
Your Merciful awoke in me the potential of the soul, with the recollection of
death, that is the ultimate end we all face. Consider the wisdom of São
Basílio, the bishop of Capadocia, who was asked by the gentio philosopher
Eubolo for the definition of philosophy. São Basílio replied that its basic
principle was the thought of death. Eubolo was so taken with this response
that he soon converted to the faith of Jesus Christ our Lord. There is much to
learn from the way these few words could a√ect a heathen, and it is a lesson
for me, a miserable sinner that still in my infancy came to know the faith of
Your Holiness and am a Christian by the grace of God. And none of the light I
now see was revealed to me by my father or my mother.∑≤
208 | Chapter Six
This excerpt still leaves some doubt about Souza’s early life. Both he and
Cordeiro stated in the Manuscript that they spent their infancy together, but
it is never stated where. Souza’s reference to a conversion at such a tender
age encourages the interpretation that he and Cordeiro were shipped to
Bahia when still quite young, and Souza would have been converted there.
Souza arrived in Rio de Janeiro later, in 1748, but there is no hint that
Cordeiro arrived with him or even spent much time in Bahia. Nonetheless,
we see in Souza’s account of his own faith a reflection of the rigor with
which the statutes of the Devotion to the Souls of Purgatory speak to enforc-
ing Christian doctrine among the brothers. The statutes are expressly con-
cerned with fostering the spirituality of the devotion’s members in positive
terms, but there is a steely undertone: disciplinary measures and punish-
ments for those whose commitment was irresolute are included through-
out. Chap. 4 determined that all members should be devoted to God, the
Virgin Mary, and all the saints, especially the saints associated with their
names and guardian angels. The saints of the souls in Purgatory deserved
special attention, and masses would be held daily for them with special
services held on Monday, the day dedicated to them. For members who
could not attend mass—and these would mostly have been slaves, with
generally inflexible demands on their time—it was recommended to pray
on bended knee before an image of the crucified Christ and recite ‘‘six Our-
Fathers, six Ave-Marias, with Gloria Patris included for the Souls of Purga-
tory.’’∑≥ Freed slaves were held to another standard. Chap. 10 declares that
freed members who were absent ‘‘without just motive’’ from a funeral
procession for a deceased brother would have to pay a penalty (euphe-
mistically called a donation) of 120 réis into the safe to compensate their
rebeldia e frouxidão (insubordination and weakness). Those who could justify
their absence from the all-important funeral procession by showing ‘‘legiti-
mate cause,’’ such as a commitment related to work, needed to recite ‘‘the
Our Father and Ave Maria with the Gloria Patris, o√ered to the sacred
passion of God for the soul of our fallen kin.’’ A freed member of the
devotion who did accompany the procession to the cemetery needed to
o√er special prayers to ‘‘the Sacred Death and Passion of our Lord, for the
soul of the dead’’ at graveside. In accordance with chap. 14, every member
of the devotion, excepting only laborers and the elderly, should fast on the
Mondays of Lent, and attend masses. Those who were literate were required
to read the nine prayers of São Gregório (Saint Gregory), which were collo-
quially called the Novena of Souls, while members unable to read were
instructed to recite nine each of Our Fathers, Ave Marias, and Gloria Patris
for the souls in Purgatory.
As with the discussion they introduced of paganism and superstition,
the statutes of the Devotion to the Souls of Purgatory also depart from the
model of the compromisso of the Brotherhood of Santo Elesbão and Santa
Efigênia in their detailed focus on Catholic doctrine, practices, and obliga-
tions. Even though the Constituições Primeiras had recommended the consis-
tent regulation of Catholic doctrine within all levels of the lay brotherhoods,
the Brotherhood of Santo Elesbão and Santa Efigênia did not address the
subject in their founding documents. So once again Souza’s innovation,
this time in emphasizing religious instruction, would seem to derive from
his own personal preoccupations and initiative. But we must not assume
that the absence of explicit ecclesiastical norms from the brotherhood’s
statutes meant that the brotherhood was trying to ignore them or keep
them out. (According to the Ordenações Filipinas, in fact, indoctrination of
slaves was the sole responsibility of their owners.) Their appearance in the
statutes written by Souza for the devotion raises the possibility that one of
the functions of such smaller groups within the brotherhood was precisely
to reinforce the ecclesiastical norms, while the brotherhood’s compro-
misso established the overarching structures of organization and pro-
cedure. The absence of norms from any one document must not be given
too much weight, since eighteenth-century Rio de Janeiro was a baroque
society, with all its spheres of social life pervaded by an intricate profusion
of norms. Rather, it is more important to note how the norms were articu-
lated, how di√erent norms interacted, and, when broken, how they were
reformulated or reinforced.
The Mahi Congregation’s statutes do adhere somewhat more closely to
the tone of practical regulations characterizing the compromisso of the
brotherhood. Chap. 1 determines that the congregation should be led by a
regent and vice-regent elected by vote; chap. 2 limits eligibility for those
positions to ‘‘natives of the Mina Coast and of the kingdom of Mahi.’’ The
regent was to work in the best interests of the congregants and meet with
the Notables, or the council of prestigious brothers holding bureaucratic
positions and royal titles, whenever necessary for the resolution of prob-
210 | Chapter Six
lems or decisions. The following chapters address other topics, but chap.
10 returns to the regent and the proper veneration he is due from other
congregants; his authority included the right to determine how to punish a
member who showed disrespect to his position. The guilty member would
be ‘‘chastised according to the judgment of the regent,’’ who would need to
confer with his companion ‘‘the female regent, and others who hold posi-
tions of authority in the same Congregation.’’ This centralization of deci-
sion making among the small group of congregation leaders may have been
modeled on similar provisions in the compromisso of the Brotherhood of
Santo Elesbão and Santa Efigênia. Chap. 31 of that earlier document had
stipulated that the judge, procurator, and treasurer must be treated as
‘‘superiors’’ (although here, unlike in the congregation’s statutes, specific
punishments were also outlined for disobedience to authority).∑∂
The procurator received special attention in the Souza’s statutes for the
Mahi Congregation. The duties of this position were to stay abreast of any
legal proceedings involving the devotion; keep track of the members (espe-
cially the old, sick, or poor) and report news of their condition to the
regent; visit the sick or dying; and mediate internal disputes. This last
function was seen as perhaps the most critical of the position, ‘‘because
many times, from a small spark a roaring flame is created’’ (porque muitas
vezes, por um pequeno incêndio se levanta uma grande labareda). Accord-
ing to chap. 15, another responsibility of the representative was to look
after the four books the congregation was to maintain: one each for the
membership registry, for income and expenses, for the statutes, and for
lists of masses performed. Keeping accurate records of the masses realized
by the congregation was less an expression of spiritual probity than of
necessary accounting, since the chaplain contracted by the Brotherhood of
Santo Elesbão and Santa Efigênia was paid annually for the number of
masses he performed during the year.∑∑
Like all lay religious organizations at the time, the Mahi Congregation
had an institutional preoccupation with death and the fate of the dead. Its
creation of a separate devotion to address the needs of the dead was not
without precedent among Brazil’s lay brotherhoods; groups similar to its
own Devotion to the Souls of Purgatory are to be found within various other
brotherhoods. It was unusual, however, in focusing its activity strictly
among deceased brothers of the Mahi nation. This ethnic exclusivity pro-
moted comradeship with other congregations, as chap. 5 of the statutes

suggests: even if the deceased in question had identified himself or herself
principally as a member of a separate brotherhood in the city, and was only
secondarily a member of the Mahi Congregation, the Devotion to the Souls
of Purgatory within the Mahi Congregation was required to accompany the
funeral procession led by the other brotherhood. The regent had the obli-
gation to collect donations within the devotion to pay for masses for the
irmão falecido (deceased brother).
The subtle distinctions in responsibilities and priorities between the
Brotherhood of Santo Elesbão and Santa Efigênia and the Mahi Congrega-
tion permit us to observe that the sphere of charity was much more the
jurisdiction of the congregation than of the brotherhood. This was true in
the contexts of both funerals and freedom. In the former case, the con-
gregation had a privileged place alongside the chaplain, whose various
services for the dead were incorporated more clearly and fully into the
statutes of the congregation than of the brotherhood. The chaplain was to
hear the confession of the sick and dying, give the Holy Sacrament, and
assist in the preparation of the will and the distribution of the bequests; he
would ready the sinner to ‘‘die of the present life,’’ as the Constituições
Primeiras put it. The congregation was intimately associated with these
rituals and services through its statutes, which formally called for them;
through the solicitous actions of its o≈cials to assist the chaplain; and
through the presence and support of its members.∑∏
Members of the Mahi Congregation could, under special conditions,
turn to the congregation’s co√ers for financial help in buying their free-
dom (completing the ajuste de sua alforria). A congregant making use of that
service would have to formally accept the obligation to repay the amount
borrowed.∑π Although the privilege was not accessible to every member—
whether because of intrinsic limits to the congregation’s purse, or (also)
because it was provided only selectively to members meeting unwritten
qualifications—its prominent appearance here and in other congregations
within the brotherhood suggests that it was viewed as a special respon-
sibility of the smaller ethnic organizations.
Visiting Brazil in the first half of the nineteenth century, the French artist
Debret recorded the following observations of sociability, social di√eren-
tiation, and the acquisition of freedom among slaves. His depiction corre-
212 | Chapter Six
sponds closely with what the eighteenth-century documents imply about

social relations within the ethnic reigns of the black lay brotherhoods.
When two blacks of unequal rank encounter each other on the street during
the performance of their errands, the vassal will respectfully salute the
sovereign of his caste, kiss his hand, and ask for his blessing. Dedicated to
his king, trusting in his superior knowledge, it is to him the subject turns for
consultation during di≈cult circumstances. And the noble slaves, thanks to
their lofty position, can obtain from their subjects the means su≈cient to
buy their freedom. Once freed, the noble blacks will then scrupulously direct
all their activity towards the repayment of that sacred debt.∑∫
The mention of ‘‘noble slaves’’ with royal titles seems a clear reference to
the nobility of the folias. The description of how the nobles bought their
freedom with money borrowed from their ‘‘subjects’’ also likely refers to
the brotherhoods, since these were the only institutions that commonly
engaged in such transactions.∑Ω If the specificity and emphasis demon-
strated by the Mahi Congregation’s statutes are any indication, the pur-
chasing of freedom with money borrowed from a pool of contributions
was taken extremely seriously and created formidable bonds between the
freed slave and the congregation. Among the Mina Africans who gathered
at the Church of Santo Elesbão and Santa Efigênia, this special distribution
of resources was limited to the tighter network within the provenience
group (those they called parentes).
The social category of parente was something altogether di√erent from
irmão (brother). Everyone belonging to the brotherhood was a ‘‘brother.’’
Their conception of parente, or ‘‘kin,’’ is a more fundamental and exclusive
quality of identity. Belonging to a lay brotherhood implied the immediate
but abstract sense of joining a religious group, in the context of the rules
and rituals of a Catholic fraternity of brothers. The collective phrase ‘‘broth-
ers by compromisso’’ was commonly used in all brotherhoods, whether of
blacks, pardos, or whites, and nicely connotes the formality upon which
membership in this religious ‘‘family’’ was predicated. For blacks, and
most especially Africans, parente could be used in any situation, not only in a
religious context. The word suggested social links rooted in shared identi-
ties (ethnic or provenience) that were built from the active reconstruction of
a common past that had nothing to do with genetic kinship or blood ties. It
addressed a social grouping conceived with historical qualities, whose

parameters were refined through socioreligious organization in the pres-
ent. For example, the Devotion to the Alms of Purgatory accepted members
from various African nations, but the assertion of its core identity was Mahi,
reflected in the regulation that only a Mahi could lead it (and reflected also
in the devotion’s overarching concern for deceased Mahi). The folia should
be understood as, among other things, an expression of ethnic identity that
highlighted the brotherhood’s diversity. But each flank of the folia was also
a mechanism to attract more parentes to the congregation and catalyze
their indoctrination; at least that is how the Catholic Church saw the matter.
The Constituições Primeiras understood the folia to be useful for conversion
and squarely within the missionary tradition, and that is why they kept it o√
their list of ‘‘very indecent actions’’ to be prohibited.∏≠ This perception was
echoed in chap. 13 of the statutes of the Devotion to the Souls of Purgatory,
which argued for the creation of a folia because, the writers stated, ‘‘experi-
ence has shown us’’ that folias are advantageous for ‘‘exercitar os ânimos
dos pretos’’ (raising the spirits of the blacks) and attracting new mem-
bers.∏∞ Those statutes also recommend that the folia of the Mahis should
march to the Church of the Rosário in a show of support whenever the
Brotherhood of the Rosário there elected a Mina king.
The folia of our Mahi nation should proceed to Church of Our Lady of the
Rosário on the day of that saint to celebrate when the King who has been
chosen there is from the Mina Coast. The Mahi folia will accompany him
(but if the King is of any nation but Mina, our folia will not accompany him)
to the palace of His Illustrious Excellency the Viceroy of this State; and after
the festivities every member of the folia will return to his house in the
peaceful and orderly manner such occasions require.∏≤
This show of solidarity with the Mina from another brotherhood is not
surprising. Indeed the forceful insistence on Mahi primacy in their own
congregation (and its associated groups) might lead one to the impression
that fraternizing with orders of Africans of other nations would have been
viewed as untoward, even deviant behavior.∏≥ But not only did it undoubt-
edly occur; in this case it was explicitly encouraged, perhaps to help main-
tain the bonds of the Mina provenience group across the city. Angola people
were the majority in the Brotherhood of the Rosário, and typically domi-
214 | Chapter Six
nated its leadership, but Mina were present in considerable numbers (and
many of them also belonged to the Brotherhood of Santo Elesbão and Santa
Efigênia). Similar to the marriage practices discussed earlier in this book,
despite very straight notions of belonging there were intermarriage and
intergroup cooperation among di√erent provenience groups, suggesting
how social norms were continually reconstructed and reenacted through
the testing of their limits.
Documents pertaining to the Mahi Congregation are an especially rich
resource for helping us understand the complex relations between social
norms and the exercise of power in daily life. The various conflicts that
developed within the Brotherhood of Santo Elesbão and Santa Efigênia
often led to changes at the formal level of the association and its regula-
tions, but the tensions would persist in new forms and call for new solu-
tions. Women progressively gained more and more space to participate in
the brotherhood, but their presence was contested all along the way. An-
gola won the right to enter the brotherhood through a royal provision, but
were consistently excluded both from positions of power and from the
quotidian activities of the lay community at the church, which were seg-
mented and organized according to ethnic group. The Mahi Congregation
reproduced many of these characteristics of the larger brotherhood, but it
also asserted its autonomy—whether in terms of a separate ethnic identity,
or in the direct engagement with paganism and superstition implemented
by Souza.∏∂ Even within these smaller organizations, there were disputes of
diverse forms; the story of the Mahi Congregation’s safe indicates how
unpredictable and serious they could become. A look at the statutes of the
Fraternity of Nossa Senhora dos Remédios will also show how conflicts
within the Mahi Congregation could lead to transformations of the organi-
zation and profile of the association itself.
the reconstruction of identity

On 4 June 1788, two years after the emergence of the Devotion to the Almas
do Purgatório, but still during the period of legal tussling over possession
of the Mahi Congregation’s safe, Francisco Alves de Souza and a group of
elite congregants in his circle completed the statutes for the Fraternity of
Our Lady of Remédios. The statutes were promptly sent to Lisbon for
approval, which was granted in October by a royal procurator. It is di≈cult

to miss the connection between this new association and the Devotion to
the Almas do Purgatório, at least if one looks at the individual brothers’
signatures that appear on the statutes (alongside those of Souza; his wife,
Rita Sebastiana; and the secretary Gonçalo Cordeiro). Of the nine brothers
who helped found the new fraternity, five also held positions of authority in
the devotion;∏∑ and of those five, two had been present at the deathbed of
King Monte when leadership was passed to Souza back in December
1783.∏∏ This continuity is an indication of the unity Souza had managed to
maintain among a segment of the Mahi elite throughout the course of the
conflict over the safe. (Given the lack of complementary documentation, it
is di≈cult to measure the impact of that conflict over time on the rank-and-
file brothers.)
The statutes of the new fraternity insist on strict observance of Catholic
doctrine. They recommend that the group enlist the services of a ‘‘capelão
instruído e de bons costumes’’ (learned and honorable chaplain), which
might also be a subtle acknowledgment of the erratic performance of the
chaplain contracted by the Brotherhood of Santo Elesbão and Santa Efi-
gênia. Funds are allocated for the purchase of wine, as well as flour for the
holy bread and even ‘‘an iron of excellent make’’ for shaping the conse-
crated host. Chap. 4 is especially emphatic in its account of how the
behavior of members was to be evaluated; it was left to the o≈cials to hear
accusations, weigh evidence, and determine the fate of the accused.
Any person who wishes to join the Fraternity shall agree to an annual
contribution of six hundred and forty . . . and will present himself to the
Procurator, who will welcome him and enter the membership in the Book of
Brothers. However, if the new brother makes it known that he is a black of
o√ensive habits, given to vice or villainy, or to witchcraft and superstitions,
the Procurator will report to the Board, who will then decide based on the
information about the brother’s conduct whether or not to exclude him from
the Fraternity.∏π
The fraternity would have its own folia, which, being under Souza’s
direction, was more explicitly linked to Catholic rituals than was the folia
of the Brotherhood of Santo Elesbão and Santa Efigênia. It should partici-
pate in the festival of the Rosário, as well as carrying out its own festivities
216 | Chapter Six
on the day of the Holy Spirit.∏∫ Still, the basic premise of the fraternity was
charity, and its choice of patron saint clearly communicated its priorities.
As stated in the petition accompanying the statutes that were sent to Queen
Maria I, the fraternity’s mission was to improve the health and dignity of
the poor Mina Africans in Rio de Janeiro by providing ‘‘remedies, nursing,
food, and even funeral shrouds.’’ Beyond this, the fraternity would pray for
the souls of dead brothers, which was an objective common to all lay
religious associations.∏Ω
When Souza o≈cially took over as Regent of the Mahi Congregation in
1786, he stated before the assembled brothers his own priorities as leader
and Catholic: burying the dead and caring for the poor—or, in a word,
charity.π≠ The two smaller groups within the congregation that Souza
helped found undertook charity in distinct ways. As practiced by the Devo-
tion to the Souls of Purgatory, charitable acts were directed inward, to
members of the group; but the Fraternity of Our Lady of Remédios would
reach out to all the Mina in the city. While the first group focused on souls
of the dead, the second looked after the living who were sick or hungry, and
even provided shelter to the neediest. And while the devotion assisted its
members (primarily the noble ranks of the folia) in buying their freedom,
the fraternity used its funds to hire legal mediation when an enslaved
member raised a just complaint against his or her master.π∞ Over this same
period, the number of positions of authority on the board of the Mahi
Congregation was clearly increasing, which indicates that the congrega-
tion was expanding its membership and its influence among the Mina of
the Church of Santo Elesbão and Santa Efigênia. I am not suggesting that
the charitable work of the Mahi Congregation was solely responsible for its
growing stature (assuming the work was e√ectively carried out), or even
that the types of activities assumed by the devotion and fraternity had not
already been occurring in other forms. Rather, there was a shift or reorien-
tation at the very basic level of identity conceptualization and articulation—
what might be called the norms of identity construction.π≤
As they narrate the rise of the Mahi Congregation within the Mina
Congregation of the Brotherhood of Santo Elesbão and Santa Efigênia,
Souza and his secretary, Gonçalo Cordeiro, depicted a winnowing process
catalyzed by intergroup conflict. That is, a series of what the Manuscript
treats as ‘‘us versus them’’ conflicts involving two sets of antagonists (first
the Dahomeans, then the widow and her allies, all of them portrayed as
superstitious malcontents) helped produce a particular sense of Mahi iden-
tity that Souza and Cordeiro traced all the way back to a purportedly
glorious Mahi Kingdom, whose dominion was supposed to have stretched
across the whole Mina Coast. The regent and his secretary drew a contin-
uous line between this powerful representation of an African kingdom and
themselves, the legitimate and humble heirs of their ancestors, for whose
salvation they now prayed to God.
Ultimately, control over the Mahi Congregation, not just in terms of
money and procedure but the power to articulate its very identity, was at
stake in the dispute between Souza and the widow. The two smaller asso-
ciations were created, at least in part, to formalize Souza’s status as leader.
That is why, for example, chap. 3 of the statutes of the Fraternity of Our
Lady of Remédios includes the a≈rmation that the congregation—the
larger Mahi Congregation, not merely the fraternity—wished to keep Souza
as their regent. And within the statutes of the fraternity were two requests
regarding matters of great importance to the congregation as a whole:
first, that widow be forced to turn over the safe and other valuable items to
the congregation;π≥ and second, that the fraternity be exempted from pay-
ing taxes on processions and funerals.π∂ An exemption granted to a small
fraternity could quickly be used as a precedent to argue for similar exemp-
tions, not only for the Mahi Congregation but for all the associations based
at the Church of Santo Elesbão and Santa Efigênia.
The Mahi Manuscript was not a passive record of the struggles between
Souza and the widow (and their cadres of operatives) for control over the
organization and identity of the Mahi Congregation. Rather, it was a tactic
in the fray, intended by Souza and Cordeiro to further their cause by
recasting the terms of the dispute. Nowhere is that more evident than in the
final part of the Manuscript, when the dialogue turns to a description of the
Mina Coast provided by Souza at Cordeiro’s request.π∑
Both of these men had been taken to Brazil at a very young age from
regions located well inland from the Bight of Benin, so they would have
had extremely little personal familiarity with the coast. Souza alleged that
everything he knew had been told to him by his friend the unnamed
helmsman, who had reportedly accrued impressive practical knowledge of
the area. But what strikes the reader is that in its historical and social
218 | Chapter Six
dimensions, Souza’s narrative reproduces the perspective of fifteenth-

century and sixteenth-century Portuguese writings in which a Christian
Mina Coast (anchored by the Mina Castle) was framed in opposition to the
heathen kingdom of Benin. At no time did Souza or Cordeiro mention their
own African pasts, the slave trade that carried them across the Atlantic, or
their experiences in captivity. Rather, Souza dwelled on the conversion of
heathens, which was usually carried out in the environs of the Portuguese
castle. He then turned to a description of the coast as one travels from the
castle to the kingdom of Benin. In general form, his account—replete with
references to geographic features, prominent European forts, and native
villages—is indistinguishable from those of the scribes who were approv-
ing witnesses to a conquest 200 years before.
Souza’s depiction of Benin, which again is supposed to derive from the
direct observations of his friend the pilot, accentuates the fantastic. Benin
was a mysterious territory permeated with rivers, lagoons, and bogs, where
ferocious beasts large and small, of types unknown in Brazil, lurked in
hiding to attack the unwary. Even the similar bore a touch of the strange:
sheep there ‘‘grew hair instead of wool.’’ But there were valuable discov-
eries to be made as well, Souza noted, such as the peppers, cotton, and
palm oil that attracted Dutch traders. The kingdom had as its capital a
‘‘beautiful and grandiose city,’’ also called Benin, that ‘‘é do tamanho de
uma légua sem muros, mas está cercada de uma grande cava’’ (is a league
across, encircled not with walls but with a deep trench). Its king, who was
called Bâ Benin by the deferential natives, captured untold numbers of
people to be traded as slaves ‘‘for bracelets and other things.’’π∏ Souza
described the natives of Benin as daring and generous, as well as zealously
loyal to the king: ‘‘para mostrarem que o são fazem em seus corpos umas
cicatrizes, ou marcas’’ (to show their identity and allegiance, they cut scars
and marks into their bodies). The natives displayed other elaborate but
curious customs. The men could only wear garments given to them by the
king, and young women were restricted to the use of clothing given to them
from their future husbands, so ‘‘não se vê pela rua senão ranchos de
homens e mulheres nus’’ (the street corteges were full of naked men and
women). The king appeared in public only once a year, in an enormous
festival whose grim centerpiece was the sacrifice of over one hundred
slaves (‘‘mais de cem pessoas, todos escravos’’). Human sacrifice was an
integral part of funerals, as well, and when the funeral was for a king there
was an especially cerimônia bárbara (barbarous ceremony):
When a king dies, the nobles of his court, who are called Ômon, slaughter 16
slaves, then carry them to the crypt to be buried along with the king and the
king’s robes and personal property. Some of these same princes and nobles
may also be killed at the gravesite to accompany the corpse of the fallen king,
under the notion that they will serve him in the other world. Seven days later
yet another sacrifice of slaves who are called Ovem is carried out; there is
dancing atop the sepulcher of the king to the throbbing of drums, and
celebrations spring up all around the grave, the people exulting in their grief
and apparitions. It is said in this place that God, being good and pure by
nature, does not require human sacrifice; so they devote their sacrifices to
the Devil, to please him, while at the same time they worship their idols.ππ
Souza and Cordeiro were emphatic in denying any similarity between

these savage customs and the habits of the Mahi, although such practices
were common among all the Yoruba and Fon speakers of the region (in-
cluding the Mahi). The objective of this second dialogue is to reconstruct
the identity of the Mahi within the Christian universe, as a small bright
light of Catholicism surrounded by the darkness of the kingdom of Benin
and other heathen groups. Profane practices were left out of their discus-
sion of the Mahi, except insofar as the Mahi were shown confronting these
o√enses with compassion and the word of God.
A more tangible example of this reelaboration of the past is perceptible
in the way that the pano da costa was put to new use by the Mahi Congrega-
tion. The particular cloth in question was probably from the Mina Coast,
since it had been o√ered to the congregation by a brother. It was kept with
the congregation’s safe not only because of its symbolic connection to a
homeland across the Atlantic, but because of the important role accorded it
within the proceedings of the congregation: this cloth was used to cover
the table in the consistory of the Church of Santo Elesbão and Santa
Efigênia on the day each year that brothers approached the table to contrib-
ute their annual donation. It was used instead of the customary white lace
cloth of Portuguese make.π∫ In this context, meaning on the church table,
the African cloth represented a new use of the pano da costa in the Christian
universe. The pano da costa, and other African objects and memories were
220 | Chapter Six
brought to the scene to link their lives and social practices with their African
past. As the Mahi Manuscript shows, this is the same feeling that made
them strongly concerned about registering and telling the history of the
Mina Coast, in particular of the Mina Castle, and of conversions and a
growing Christian presence there. No one denied that there were ancestors
who had never been converted, and lived as heathens; and the souls of
these ancestors received prayers for redemption in the Devotion to the
Almas do Purgatório, where they assumed would be their heathen and
Catholic ancestors and all deceased people.
This Catholic Mahi identity was constructed in opposition to a heathen
Benin, where naked people wandered the streets and human life had little
value other than to provide bloody sacrifice to idols (Souza’s more direct
treatment of the theme of human sacrifice in his writing suggests it was
more heinous to him than witchcraft—or more containable). In Souza’s
account of the Mahi, the fight between good and evil, between God and the
devil, began in Africa but persisted into the Church of Santo Elesbão and
Santa Efigênia, where they were so ‘‘scandalized and a√ronted’’ by the
Dahomeans that the Mahi broke away to form a new congregation. It
remains unclear whether the ‘‘heated words’’ of the Dahomeans involved
witchcraft, or perhaps accusations of black magic against the Mahi. But
with Souza as regent of the Mahi Congregation, the Mahi identity was
increasingly refined through the negation of heathen elements in their
African past, and the development of a Christianized identity based on
redemption and Catholic doctrine in the new setting of Brazil’s colonial
slave society.
I could find no documents dating after 1788 to provide a hint of how
these stories continued to play out. Souza’s personal zeal and popularity, as
well as his intellectual attributes, probably helped maintain his influence
for some time in the Mahi Congregation, and even in the larger Brother-
hood of Santo Elesbão and Santa Efigênia. It is curious to note that across
Rio de Janeiro, however, the Mina Africans as a group were increasingly
associated with witchcraft and boisterous pagan rituals. In 1835, the city’s
chief of police led a lengthy investigation of several Mina, including one ‘‘a
quem muitos outros rendem o maior respeito, e que ali vão iniciar-se em
princípios religiosos (to whom many others show the greatest deference,
and who apparently initiates them in religious principles).πΩ Such initiation
ceremonies would have been carried out like the early calundus, later called
‘‘houses that o√er good fortune’’ (casas de dar fortuna), where it was
alleged that the festivities were accompanied by fantastical visions and
apparitions. In referring specifically to visagens (apparitions), the police
record echoes Souza’s description, given fifty years earlier, of the death
rites performed at the sepulcher of the kings of Benin. Additional period
documentation shows that these houses of good fortune in Rio de Janeiro
were notorious sites of dancing, chanting, and animal sacrifice, all carried
out to the pulsing rhythms of the drums. But this part of the story must be
told another day. . . .
postscript
The writing phase of the work involved in transforming my
\ doctoral dissertation into a book was completed in 1997. Much

has changed since then. Not only have many exciting new works
been published (or others brought to my attention) that contribute much-
needed context, nuance, or counterpoint to my arguments in the book, but
my own research has continued as well—motivated by the desire to learn
more about the Mahi, and inspired by new sources as well as new conversa-
tions. I have two goals in this essay: first, to introduce more detail to the
book’s sketch of the daily life of Mina Africans in the Brotherhood of Santo
Elesbão and Santa Efigênia (a subject with increasingly evident links to the
history of Africa and the slave trade); and second, to o√er some final
reflections on what this case tells us about culture and ethnicity.
As the book went to press, I was particularly interested in trying to
discover connections between the scenario I had found in Rio de Janeiro
and the circumstances in Minas Gerais and Bahia. It was something of a
surprise to discover that very few of the studies based in those places dealt
in a sustained way with the question of African slave proveniences in the
eighteenth century. But drawing from them, and from many of the primary
sources on which they were based, I could conclude with some certainty
that Minas Gerais was the dominant destination of the slave trade between
Brazil and the Bight of Benin in the first half of the eighteenth century. In
other words, although the slave trade to Rio and Bahia di√ered in volume
and composition, its primary objective was the same in each city—to feed
224 | Postscript
the gold mines of Minas. And notwithstanding the demographic di√er-

ences in the trade to Rio and Bahia, I was more struck by the distinct forms
of identification of slave proveniences, often with completely di√erent
names for the same nations, utilized by the primary sources (and histo-
rians) a≈liated with each place. Still, the data provided specific insights,
such as the fact that between 1700 and 1750 slaves arriving from the Mina
Coast represented up to 10 percent of Rio’s slave population, while over 30
percent of the slaves in Minas Gerais were Mina.∞
The eighteenth-century (1737–1800) slave inventories analyzed by Pierre
Verger included 140 slaves from the Mina Coast: 121 Jêje, 16 Mina, 1 Lada,≤ 1
Mahi, and 1 Savalu.≥ Such small numbers are suggestive but not statistically
meaningful. But in Rio de Janeiro and Minas Gerais, there was more consis-
tency in the designation of slave proveniences. Beyond the Mina category,
there were prominent references in the documents in each of these places to
Mahi,∂ Savalu,∑ Chamba,∏ and Nagô. In Minas Gerais mention was also
occasionally made to Lada and Jaquem slaves;π Coura and Cobu were more
common there, but less common in Rio de Janeiro.∫ I have not yet been able
to identify the slaves called Craberá (possibly the Kabre?) or Fuam.
It should be noted that especially for Rio de Janeiro, the nature of the evi-
dence complicates the conclusion that one group appeared more frequently
than another, because certain types of documents foreground particular
groups. Wills, manumission records, and marriage records all contain
more Coura than Mahi, but in the Mahi Manuscript Mahis are predominant
(perhaps predictably) over the Coura. And I want to emphasize the fact that
a group bearing the same designation—that is, purportedly of the same
‘‘nation’’—could assume, in di√erent places and circumstances, di√erent
configurations of social and ethnic identity.
The richest vein of evidence I encountered was in the Arquivo da Cúria
Metropolitana do Rio de Janeiro (acmrj), where I could find new docu-
ments, in particular the Livro de Batismo de Escravos, Freguesia da Sé, 1726–33,
and more than one hundred wills transcribed in the books for registering
the burials of deceased people (Livro de Óbito). Many Mina were able to buy
their freedom and accumulate personal property, which would have been
listed in wills. Prominent individuals are relatively easier for the historian
to trace this way because the date of their death would have been recorded
elsewhere, and indeed the first will I could locate was that of Ignacio
Postscript | 225
Monte, Souza’s predecessor as king of the Mahi Congregation. After that, I

began to scour the archives of obituaries, as well as participate with net-
works of colleagues (especially Sheila de Castro Faria). With time I was able
to accumulate over 100 wills of Mina Africans in Rio de Janeiro dating from
the eighteenth century. Information about their husbands or wives led to
marriage records as another useful source. Working backward from the
freed slaves’ manumission letters, I could learn about their owners, who
were often Mina as well, and whose wills listed their slaves as property. By
compiling the names of witnesses, small business partners, and other
people who appeared tangentially in the wills, the social universe of these
Mina expanded even more; I am presently compiling all of this material
into one database.Ω
A Mina would typically identify himself or herself as such on legal
documents, where the declaration of provenience was required. Although
some of the freed slaves would instead refer to themselves as pretos forros
(freed blacks), in most cases the presence of complementary documenta-
tion permitted me to ascertain whether or not they would otherwise have
used the term Mina. Most of the freed Mina were members of a lay brother-
hood (or of more than one), such as that of Santo Elesbão and Santa
Efigênia, Our Lady of Rosário, São Domingos, or Santo Antônio da Moura-
ria. Few of these people declared a profession, but among them the princi-
pal occupations were barbers, shoemakers, and street venders.
It was perhaps inevitable that my research would lead me to the history
of Africa, but I could not have imagined how much I would learn in that
area, from both archival material and scholarly exchange, through trips to
the United States and Canada between 2000 and 2007. It quickly became
obvious that my approach to studying the Mahi should be expanded, be-
cause they presented a special opportunity to comprehend the historical
connections drawn by a people between an African past and the new milieu
of slavery and a colonial slave economy. Recent works that locate Brazil
within the African diaspora were a critical part of this process. But I also
read carefully the work of Robin Law, which focuses on African coastal
kingdoms, and the work of Robert Cornevin on the history of Benin and
Togo. Cornevin’s intimate knowledge of French colonial Africa brings a
distinctive nuance to his writing. If Law helped me understand the Atlantic
slave trade based at the Bight of Benin, Cornevin contributed both histor-
226 | Postscript
ical perspective on the Mahi and frameworks for understanding their rela-
tions with neighboring peoples of the hinterland in the eighteenth century.
In broad terms, the goal of my present research is to follow the Mahi (and
the groups often united with them, such as the Cobu and Coura) from
Africa to Minas Gerais, exploring the transformations in lifestyle, social
practices, and organizational strategies that they underwent in this long
journey. I am also interested in how the Mahi, once in Brazil, may have used
aspects of that colonial society to their own advantage.
Still, looking closely at the Cobu, Coura, and other groups is important,
and not only for their interactions with the Mahi.∞≠ Their stories provide
additional insight into the complex processes of identity construction and
identity ascription that occurred as Africans and non-Africans interacted in
the New World. Their stories also further highlight the utility of a research
perspective based on provenience groups rather than a vaguely defined
ethnicity. When a slave self-identified as ‘‘Mina-Mahi’’ or ‘‘Mina-Coura,’’
such designations combined the general name of a nação (nation) with the
more specific name of a terra (homeland). These are categories or levels of
identity that have nothing inherently to do with some kind of frozen and
immutable ethnicity.∞∞ The allusion to a homeland was meaningful not
only in the strict geographical sense, but more fundamentally as a refer-
ence to a localized past that was shared with others and which thus could
serve as the basis for the construction (or reconstruction) of a group
memory and social organization. I perceived from studying the Mahi that
there may have been other small groups coexisting within the umbrella
Mina provenience group—socializing, collaborating, and struggling with
each other in diverse ways, for many reasons, but principally over ques-
tions of power and identity. Although colonial documents reinforce the
idea that all the Mina spoke the so-called general tongue of Mina, there was
a pronounced linguistic diversity among them, and even bilingualism or
multilingualism, just as there was among Africans all around the Bight of
Benin and its hinterlands.∞≤ It is increasingly clear that a ‘‘Mina identity’’ in
Rio de Janeiro was complex and dynamic, continually absorbing di√erent
social aspects (including linguistics) from both Africa and Brazil.
Perhaps the most promising way to get a handle on these large ques-
tions is through history at the microlevel. Biographies, to the degree that
we can reconstitute them, reveal much about how the a≈rmation of a
Postscript | 227
nation or homeland intersected with self-determination, sociability, and

survival strategies in the complex push and pull of identity construction.
Based on ongoing research, I have begun to assemble biographies of
several freed Mina slaves; as much as possible, I follow these individuals’
social networks to trace the nature of the ties—religious, professional,
financial, kinship, or otherwise—that link them to others across the city. I
am especially interested in how these people articulated a sense of both
nation and homeland, and in what circumstances such identities were
expressed. Below I present three of these biographies, spanning three
Mina subgroups: Ignacio Monte, king of the Mahi Congregation (Mina-
Mahi); his wife, and later widow, Victoria Correa da Conceição (Mina-
Coura), and Antônio Luiz Soares (Mina-Cobu).
Earlier literature on these groups is scarce. The first and most detailed
account of the Mahi was written in 1920 by French colonial administrator
J. A. M. A. R. Bergé. This writer observed that the Mahi territory was a
preferred ‘‘hunting ground’’ for the Dahomeans in search of captives. A
second important work came from Robert Cornevin, another French colo-
nial administrator, whose Histoire du Dahomey (1962) o√ers systematic data
on the Benin region. Cornevin averred that the Mahi had migrated north-
ward from areas closer to the coast, coming to occupy a mountainous
territory north of the kingdom of Dahomey alongside Yoruba speakers who
had arrived from the east.∞≥ The only academic work on the Mahi is a
doctoral dissertation that remains unpublished (and that only briefly treats
the years before 1770).
None of the Africanist literature I encountered dealt at any length with
the Coura or Cobu. Pierre Verger linked the word Courano/Coirano to a
lake called Curamo near the city of Lagos (or Onim, in modern Nigeria).
Verger transcribed the description of an attack by Dahomeans on the Fort
of Ouidah (Ajudá in Portuguese) that ended with the imprisonment of the
fort’s director, João Basílio, in 1743; the presence of Couranos in the fort is
noted, along with the observation that they were ‘‘enemies of the King of
Dahomey.’’∞∂ Another document cited by Verger announced in 1767 that
Ouidah was being invaded by ‘‘coiranos.’’ Verger implied that the Coura
were a coastal people, centered around the lake region, but the lack of
corroborating evidence (in his work or elsewhere) casts doubt on this
notion. I suspect that the place called the ‘‘land of the Coura’’ by the Coura,
228 | Postscript
or Couranos, lay to the north of the Mahi territory, in modern Benin. And
based on African cartography and Brazilian slave trade data, I hypothesize
that the region of the Cobu lay even farther to the north.
Documentation from Minas Gerais includes references to the Mahi,
Coura, and Cobu by the 1720s, if not earlier.∞∑ Ignacio Monte and Victoria
Correa must have embarked on the Middle Passage after that, probably
around 1740. Both were baptized in 1742—Ignacio in Rio de Janeiro and
Victoria in Vila Rica, Minas Gerais—in their midtwenties. Antônio Luiz’s
arrival in Brazil in the hold of a slaver must have been before theirs,
however, because he died an elderly man in 1755.
the ‘‘nations’’ in the context of rio de janeiro

Victoria Correa was approximately twenty-five years old,∞∏ and the property
of Domingos Correa Campos, when she was baptized in the Church of Our
Lady of the Conception in Vila Rica, Minas Gerais, on 27 January 1742. Her
ceremony was recorded by the vicar Leão Sá, who made a point to comment
that the slave ‘‘had her face cut, in the style of her land; she was short, and
comely of figure.’’∞π We know nothing about her during the years between
that date and 13 December, 1755, when in Rio de Janeiro she bought her
freedom from her master at the time, Domingues Rabello de Almeida, for
180,000 réis.∞∫ Monte’s baptism notice was succinct in comparison to
Correa’s, but priests in Rio who recorded baptisms demonstrated a charac-
teristic taste for brevity. It was observed only that two adult slaves owned by
Domingos Gonçalves do Monte, Ignacio Mina and José Mina, were bap-
tized. Serving as Ignacio Monte’s godfather was another José, a slave
owned by Antônio Gonçalves; and his godmother was Tereza, slave of
Domingos Francisco.∞Ω In 1757, Antônio Gonçalves da Costa (the same
person who owned Ignacio’s godfather?) paid Domingos Gonçalves do
Monte the substantial sum of 350,000 réis to buy Ignacio’s freedom.≤≠ His
price was high because Ignacio was a barber and a bleeder, and probably a
dentist as well, according to his letter of freedom.≤∞ I do not know why
Costa would have been willing to pay this amount of money to free Ignacio,
but I did determine that Antônio Gonçalves da Costa was also a barber.
The marriage of Ignacio Monte and Victoria Correa was announced in
the Parish of Candelária in early 1759, and the ceremony itself took place on
Postscript | 229
the 27 February that year, at eight o’clock in the evening.≤≤ They would
spend the next quarter century together, in Rio. Ignacio died on 27 Decem-
ber, 1783. Victoria served as principal executor of his will.≤≥ Although some
of her activities in the ensuing period can be traced through other docu-
ments, I have not yet found a will or obituary record under her name.
For some time, Monte had cultivated close relations with members of his
extended ethnic group (people he called patrícios), first in the Mina Con-
gregation and later in the Mahi Congregation.≤∂ That would help him rise to
prominence in the Church of Santo Elesbão and Santa Efigênia in the early
1760s. In 1762, the brotherhood wrote to Lisbon, requesting permission
from the Tribunal of Conscience and Orders to amend their statutes to
create a folia called the Imperial State of Santo Elesbão (see chapter 5 of this
book). Permission was granted in 1764, and the folia began to hold annual
elections for an emperor and empress, seven kings and queens, and various
dukes, counts, and other ranks. In 1763, soon after Monte had been elected
king of the Mahi within the brotherhood (and while the proposed changes
to the statutes, including the postulate that emperors and kings should be
able to demonstrate the possession of personal patrimony, were being
mulled over in Portugal), he sat down to write a will.≤∑ He named three
executors: his wife, Victoria; Francisco do Couto Suzano (a ‘‘Mina black’’
and member of the brotherhood); and José dos Santos Martins (another
Mina brother, as well as fellow barber).≤∏ The will is quite detailed, but given
what we may speculate were Monte’s ambitions within the brotherhood, it
should be interpreted with care. Here, in fact, Ignacio made no mention of
Antonio Gonçalves—his godfather and the man who paid for Ignacio’s
freedom (whether or not this was the same man remains unclear), asserting
instead that he bought his own freedom.
The will was written twenty years before Monte’s death, and seems not
to have been updated, so we cannot know the state of his personal fortunes
in 1783.≤π We do have the will of José dos Santos Martins, who died in 1800,
to provide a general idea; he had served as one of Monte’s three executors.
Martins was the owner of no fewer than eleven slaves, valued in total at
934,000 réis, as well as a collection of instruments from his profession.≤∫
The wills and other personal documents provide insight into how the
members of the Brotherhood of Santo Elesbão and Santa Efigênia created
diverse networks of connections and relationships. Beyond the formal
230 | Postscript
hierarchies of the brotherhood itself, there were many informal relations

between individuals or groups involving loans, or the paying of debt
through a mix of money and services. There were trade associations (such
as among barbers or greengrocers) and smaller groups organized around
particular devotions. The wills of members of the Mahi Congregation
clearly reveal that some were creditors, and others borrowers, in transac-
tions internal to the group. Ignacio Monte made many small loans, and
engaged in the renegotiation and servicing of debts to others, creating an
intricate web of debts, favors, and obligations that were carefully recorded
in a small book to which (according to his will) his wife, Victoria, had
access.≤Ω The recording of these transactions helps us perceive how re-
sources flowed back and forth between group members: elite brothers
channeled their funds into the congregation by making loans to buy the
freedom of brothers, or investing in small businesses, as well as providing
small loans or contributions on an individual basis for emergencies (such
as sickness).
All this activity took place under the watchful eyes of the leaders of the
Brotherhood of Santo Elesbão and Santa Efigênia, who often appeared in
the wills of prestigious brothers as executors, witnesses, or editors (most
brothers, even the better o√, were illiterate). José dos Santos Martins, the
successful barber who died in 1800, was an executor of Monte’s will and a
witness to the will of Luiz Francisco do Couto, a Mina member of the
brotherhood and barber who died in 1777. In his will Couto stated that he
had been freed upon his master’s death saying ‘‘I was a slave of João
Francisco do Couto, who did me the goodness of liberating me when he
died.’’ It is unlikely that Luiz would have been able to save or borrow
enough money to pay for his own freedom. When he died, he left behind a
long list of debts, most of them to other members of the brotherhood or
the Mahi Congregation. These creditors included Rosa da Cruz, wife of
Manoel Lobo dos Santos, both of the brotherhood; and Alexandre de
Carvalho, Luiz’s godson, and Boaventura Fernandes Braga, his godfather,
both noted members of the Mahi Congregation.≥≠
The Mina population in Rio de Janeiro I studied was a prominent part of
the city’s African workforce in the eighteenth century. An African with a
trade or vocation enjoyed advantages, even if he or she were a slave, that a
freed but unemployed African did not. A trade o√ered sociability and the
Postscript | 231
potential for financial gain, as well as di√ering degrees of recognition and

prestige; influence outside the brotherhood was often reflected by influ-
ence within it. I have been able to identify with certainty four barbers and
bleeders within the Mahi Congregation: Ignacio Gonçalves do Monte, José
dos Santos Martins, Gonçalo Cordeiro, and Luiz Francisoo do Couto. It
hardly seems coincidental that all four would sign their names to the
statutes that created the Fraternity of Our Lady of Remédios, a devotion
dedicated to improving the physical condition of Rio’s poor and sick Mina
blacks through medicine, botanicals, and other treatments. The barbers
may have represented a core power group within the Mahi Congregation,
since of these four, two held high positions (Monte was king, and Cordeiro
secretary) and Martins was a close ally of Monte (and also served as an
executor of Monte’s will). These three were also literate.
Merchants were another important professional sector within the
brotherhood. The will of Luiz Fagundes, who died in 1751, shows that he
had acquired substantial patrimony: two small houses near the Church of
São Domingos. Fagundes was also a member of the brotherhood there and
asked to be buried at that church. His houses would seem to be a measure
of his success as a vendor in the market in front of the part of the city port
known at Praia do Peixe.≥∞ A second Mina entrepreneur in the brotherhood
was Antônio Vieira da Costa, who had been baptized in Portugal, in the
Parish of São João da Foz in Porto, on his circuitous trajectory to Brazil.
Until his death in 1800, he worked at his own stand in the market in front of
the part of the city port known as Praia dos Mineiros.≥≤
A di√erent activity that could also foster enduring social bonds was
military service. We know that Ignacio Monte held the rank of captain in
the Regimento de Homens Pretos (Black Men’s Regiment), and his old
friend and confidante Gonçalo Cordeiro was a alferes (second lieutenant)
there. Many other members of the brotherhood can be identified as having
served in the regiment, which was popularly known as the Henriques
Regiment in reference to Henrique Dias (a hero of the seventeenth-century
war against the Dutch in Pernambuco). Although it lacked the financial
promise of other trades, participating in the regiment brought recognition
not only internally but from wider society; it should not appear extraordi-
nary that Monte, who held high rank in the Mahi Congregation, would also
attain a position of prominence in the regiment.
232 | Postscript
the ‘‘lands’’ in the context of rio de janeiro

It was standard procedure of the era to specify provenience in wills, and ex-
slaves needed to be able to prove that they had been freed in order to leave a
will at all. To satisfy those requirements, former slaves hoping to create a
will were often intensely questioned for details on both subjects, even
when a manumission letter could be helpfully produced (for instance, it
was important for authorities to know whether freedom had been be-
stowed as a gift or had been purchased). Of the wills I surveyed, nearly
every one declared the provenience as ‘‘natural of the Mina Coast.’’ Despite
their legal formality and the pragmatic lists of possessions, executors,
heirs, witnesses, and so forth, wills could display a significantly individu-
alized character because they permitted the inclusion of personal state-
ments or autobiographical information.
Antônio Luiz Soares, who had survived the ownership of multiple mas-
ters before gaining his freedom, was a member of the Brotherhood of
Santo Elesbão and Santa Efigênia, although he considered the Brother-
hood of São Domingos his primary order. He asked to be buried at the
Chapel of São Domingos. Still, when he died an old man in 1755, Soares
left 3,400 réis toward the construction of the Church of Santo Elesbão and
Santa Efigênia, which had broken ground a year before. In his will, the
story is told—in awkward but confident prose—of his life, from birth in
Africa to freedom in Rio de Janeiro.
I declare that I was born in a brutish land of heathens as was the Mina
Coast . . . I am of the Cobu nation, and left that kingdom from there when the
Lord remembered me . . . around seven years, a little more or less, to be sold.
I went to the City of Bahia [Salvador], where I was sold to the first master I
had in that city, by name Antônio de Bastos Mendes, who instructed me in
the Holy Doctrine and wished to have me baptized. So I was baptized at the
Church of Nossa Senhora da Conceição da Praia of Salvador. I left to be sold
in this City of Rio de Janeiro and was bought by the dead man, in his life
called Antônio Soares, who was a foreigner and apothecary who lived with
his wife behind the Convent of Nossa Senhora on Carmo Hill. I freed myself
from that slavery of the dead man for two hundred and some thousands of
réis that were given to the Reverend Father Teodósio de Souza, who was his
executor and who passed to me the Manumission Letter for the above-
Postscript | 233
mentioned price, and everything was done with the approval and consent
and goodwill of the woman who was the widow of the above-mentioned
dead man.≥≥
For his part, Soares had served as the executor of at least two other
people’s wills; it is worth mentioning them because they help illustrate his
wider social circle, which included some elite Mina. One of them was
Tereza Gomes da Silva, who died in 1752 and was buried at São Domin-
gos’s cemetery in an extravagant funeral.≥∂ The second executor of Tereza’s
will, along with Soares, was Antônio Pires dos Santos. One of the founders
of the Brotherhood of Santo Elesbão and Santa Efigênia, Santos helped to
purchase the land for the brotherhood’s church in 1744.≥∑ And the other
will Soares was named to execute was that of Luiz Fagundes, the vender
mentioned above. In his will, Fagundes asserted: ‘‘I am of the Mina nation,
and came from my land to this little America; I was baptized in the city of
Bahia, and was the slave of various masters.’’≥∏ He supposedly came from
Clará, a region unidentified in the transcript or elsewhere. We are left to
wonder precisely what ‘‘land’’ it was that he had come from, as well as to
puzzle over the balance of African social identities and local relationships
that united Fagundes with Soares, Silva, and Santos.
The existence of such connections between individuals is an intriguing
fact that raises more questions than it answers. What sorts of relations did
prominent Mina such as Antônio Luiz Soares, or Ignacio Monte, maintain
with their parentes (also called patrícios)? How did participation in the broth-
erhood with other Mina a√ect one’s multiple dimensions of daily life and
sociability? Soares died in 1755, while Monte was still a member of the Mina
Congregation but before he would be elected king of the Mahi Congrega-
tion; at the time, the Brotherhood of Santo Elesbão and Santa Efigênia was
based at the Church of São Domingos. Evidence suggests that they knew
each other for at least ten years, outside as well as inside the brotherhood.
Perhaps Monte, younger and ambitious, spent some time with Soares,
whose business acumen and social sensibilities had established him in
Rio’s community of freed Mina. Monte’s will, like Soares’s, furnished an
autobiographical sketch:
I declare that I am a freed African, born in the Mina Coast. As a slave I was
owned by Domingos Gonçalves do Monte, to whom I gave in exchange for
my liberty three hundred and fifty thousand reis, as my Manumission Letter
234 | Postscript
states. I declare that I am married to Victoria Correa da Conceição, freed

black, and until the writing of this my will we have not had children; nor do I
have children with any other woman, when I was single or since I was
married. I further declare that my wife is also my blood kin, three steps
removed, because she was the daughter of my grandfather [Eseú] Agoa, a
renowned king of the heathen on the coast of the Kingdom of May, or
Maqui.≥π
Of note here is Monte’s reference to his grandfather as a Mahi king, as

well as his remark that he and his wife were blood relations. The study of
kinship is an especially demanding specialization of anthropology, and it is
even more di≈cult in a historical case with limited data. However, if the
kinship of Monte (of Mahi derivation) and Conceição (of Coura derivation)
was based on several generations of intermarriage, we can hypothesize that
an exchange of women between the Mahi and Coura might not have been
uncommon. This would be in accordance with the idea that Mahi and
Coura were actually lands that were neighboring or close to one another,
whereby the locals had regular interchange (based on war, slaving, mar-
riage, or other activities). At the same time, Monte’s allusion to a Mahi
Kingdom—similar to Soares’s reference to a Cobu Kingdom—deserves
particular attention.
The notion that identity was based on the coexistence and interaction of
peoples and villages in a given territory conflicts with the practice of cen-
tralized political authority and political borders—and with kingship and
kingdoms, such as those of Allada, Dahomey, and Hueda.≥∫ Indeed, the
assertion of a longstanding Mahi Kingdom seems not to hold up to histor-
ical analysis: the first attempt to unify a Mahi people under the centralized
political power of a king seems to have occurred only in the last quarter of
the eighteenth century (and it was due largely to the continued pressure
from nearby Dahomey). Monte’s evocation of a Mahi Kingdom, articulated
when he was in Rio de Janeiro, happened before those events and seems
more than anything else to be a response to his immediate circumstances.
In the Portuguese Catholic tradition, which underlay colonial Brazilian
society, so-called African nations had been organizing into brotherhoods
and electing their kings and queens since at least the fifteenth century.≥Ω
References to Mahi or Cobu Kingdoms in the wills demonstrate that these
Mina Africans were reinterpreting their past according to the terms and
Postscript | 235
modalities they had learned from the Portuguese absolutist worldview.

That is, using the term kingdom for the Land of the Mahi, or king for one’s
grandfather in Africa, was a translation of the past that recast it in mean-
ingful terms in the present. Without a doubt, the ‘‘ethnic’’ reigns of the
folias in the black brotherhoods were important spheres of political reor-
ganization for Africans in captivity, but when they spoke (or wrote) about
ethnic reigns, this does not mean they were necessarily organized under
kingdoms in Africa—even when they referred to their homelands as king-
doms. It should be noted that Monte himself penned his own will,∂≠ and the
word Agoa was the best way he found to convey the name, or title or
position, of his grandfather in Portuguese. Agoa immediately recalls the
narrative of the founding of the Mahi as a people, which we know from the
work of Cornevin. That author shows that in the seventeenth century, the
Bight of Benin and its immediate hinterland were crosscut with migra-
tions; some of these movements of people provided the origins of the
kingdom of Dahomey. During these movements, a group under the leader-
ship of a man named Agoua-Guédé was dislocated from the coastal la-
goons and pushed northward. The group settled in Ouakin, where mem-
bers mixed with local peoples as well as with Yoruba-speaking groups
arriving from the east (among them was a group that established itself in
Dassa, a place cited by the secretary of the Mahi Congregation as the
homeland of the notable brother Boaventura Braga).∂∞ In Portuguese there
is a notable phonetic similarity between Monte’s Agoa and the historical
Agoua.
The unavoidable conclusion from all this is that the name of a terra (land
or place), and not necessarily the name of an ethnic group, provided the
necessary reference around which to construct and reconstruct identities in
the new setting of colonial Brazil. In what seems a play of words, geo-
graphic location and ethnic group (here again I follow Fredrik Barth’s
analysis of ethnicity) are not the same thing, yet they interweave and articu-
late each other in the process of formulating identities. There is little
reason to doubt that this was already occurring before the crossing of the
Atlantic, due to the frequent migrations within Africa. Certainly it occurred
afterward in the reconstruction of identities in the New World—now with
reference to a past much more distant in space, if not (also) in time. Such a
spatial reference to the past is underscored in the use of such ancestral
phrases as ‘‘her brutish land,’’ ‘‘when I came from my land,’’ and ‘‘the god
236 | Postscript
of her land,’’ which run through the colonial documentation I consulted in

Rio de Janeiro. But this terminology was also common among native
Portuguese speakers of the era, as with the priest who wrote of Victoria
Correa da Conceição that she bore facial markings ‘‘in the style of her
land.’’ Without a substantial ethnic or cultural reference—bolstered by the
memories, evocations, and myths of an actual African place where a given
people was said to be ‘‘from’’—ethnic nations could neither consolidate
nor perpetuate themselves in the New World.∂≤ Both references were neces-
sary. When the last generation of Mina Africans died out in Rio, their
African nations died out with them.∂≥
It should be emphasized that whatever the myriad challenges and trans-
formations experienced by the Mina Africans brought to Brazil as slaves,
the documents that some of them left in Rio de Janeiro make clear that they
identified themselves and interacted as Dahomeans, or Mahi, Coura, Cobu,
or Savalu. Nonetheless, the strong link with their African past should not
mislead us into believing that their discourses and social practices in Rio
were somehow a direct reproduction of African social organization or
territorial politics. Subsets of Mina could claim to come from the kingdom
of the Mahi, or the land of the Coura, or of the Cobu, but this manner of
defining their proveniences was reshaped into larger nations through a
colonial worldview. Certainly some of the places they mentioned—such as
Savalu, Dassa, and Za—did exist (and still do); we need to better under-
stand these places, the historical relations between them, and their inser-
tion in the circuitous routes of the slave trade. The important point, how-
ever, is that in colonial Brazil, the ethnic identity of these Africans was
principally designated according to territoriality, by names that referred to
their homelands. All the personal identification documents an African
could hope to attain in colonial society—baptism and marriage records, a
manumission letter, a will—required a declaration of provenience. The
Portuguese word used in that formal context, naturalidade, or birthplace,
left no room for ambiguity.
This line of study has ramifications for the concept of African ‘‘nations’’
more broadly. The meanings, markers, and discourses that identified a
given nation could not be gathered and passed down, like a bag of valu-
ables, to the later generations of African descendants born in Brazil. Only
Africans brought to Brazil could belong to a nation, because the whole
Postscript | 237
premise of the nation—which was an aspect of Portuguese Imperial power

—depended on one’s place of birth in the colonial universe, not on lineage.
At the same time, as we saw with the term naturalidade, the colonial concept
of ‘‘nation’’ carried the built-in implication that something, some inherent
ethnic quality linked to the homeland, could be inherited. But the argu-
ment for an unchanging ethnic heritage is complicated, both theoretically
and historically, by the emergence of African-descended creoles in Brazil
who developed other social practices.
The institutionalized aversion to creoles in the Mahi Congregation (as
well as the larger Brotherhood of Santo Elesbão and Santa Efigênia) should
be understood in light of the strong links created among Africans who
survived both the Middle Passage and the travails of slavery to reconstruct
their lives in Brazil. The base of shared identity among the Mahi took on
geographical dimensions as well as temporal ones, embodied in the notion
of a land known to them but left behind. The two devotions created within
the Mahi Congregation were concerned with the well-being of Mina Afri-
cans, whether in Africa (the Devotion to the Almas do Purgatório was,
among other things, an ancestral cult) or in Rio de Janeiro (the Fraternity of
Our Lady of Remédios featured the skills of barbers, bleeders, dentists, and
other healers). The Brotherhood of Santo Elesbão and Santa Efigênia re-
created some Mina traditions and references, and lost or transformed many
others, in the e√ort to reinforce their own ties of sociability. It is ironic that
when the Mina in Rio had children—and they, particularly the Mahi, had
few—these children would be part of the very demographic of creoles that
the Brotherhood strove to exclude in order to maintain its own identity.
If the blanket identity of the Mina nation was capable of obscuring the
identities of its various component groups, it also provided new forms and
avenues for those identities to be expressed. The Mahi engaged in recon-
structing their African past as part of a dialogic process that involved the
expansion of their presence and participation in the brotherhood. The
boundaries between past and present could dissolve a bit through the
elaboration of the folia—with its Western royalty, solemn rituals, festivals,
and ornate processions. The folia represented nothing less than a new
dynasty, recasting the indeterminate land of the Mahi as a powerful king-
dom and its gentios as charitable Christians. This crossing of temporal lines
is obvious in the title of the Mahi Congregation’s founding document:
238 | Postscript
Rules or statutes in the form of a dialogue, where are announced the vows
and charitable functions of the Mina blacks and their compatriots in Brazil,
especially in Rio de Janeiro, where they must conduct and organize them-
selves without the disgraces of heathen superstition; composed by Francisco
Alves de Souza, black, of the kingdom of Mahi, one of the most excellent and
powerful of the Mina Coast.∂∂
Taken together, the various documents I encountered in Brazil and

Portugal demonstrate that the Mahi, Coura, Cobu, and other Mina sub-
groups reinvented earlier forms of sociability—and devised new ones—in
Rio de Janeiro. They developed ties within their groups and across groups,
from the relatively permanent (marriage, religious a≈liation) to the rela-
tively mercurial (political factions and strategic alliances, a characteristic of
the Brotherhood of Santo Elesbão and Santa Efigênia’s internal dynamics).
The memory of an African past that these groups constructed enabled them
all to be identified as ‘‘Mina blacks,’’ as their colonial slave society cate-
gorized them, but also as Mahi, or Coura, and so on, when it better suited
their individual and collective needs and interests.
The research that led to my doctoral dissertation, and later this book,
continues today down paths unforeseeable to me a decade ago. It reveals in
increasingly nuanced terms that African ‘‘nations’’ were the result of colo-
nial processes of representation and organization of both places and peo-
ple; the category drew from ethnic configurations without being them-
selves actual ethnic groups. While the cultural heritage of a group or
individual is vital to the construction of identities and memories, that
cultural heritage is not frozen in timeless purity but is in a constant state of
adaptation and transformation. Going back and forth between narrower
and wider historical frames (as in this postscript, from individual biogra-
phy to African history) has underscored that the techniques and perspec-
tives of microhistory are fundamental to the comprehension of historical
processes at the macrolevel, while the reverse is also necessarily true.
One of the most promising developments in the study of Atlantic com-
merce and the African diaspora is the Trans-Atlantic Slave Trade Database,
which allows researchers to estimate with unprecedented precision the
number of slaves sent to di√erent parts of the Americas, as well as to
determine the ports used by the ships carrying them.∂∑ But those data must
be incorporated into studies of colonial documents from the Americas and
Postscript | 239
the Caribbean, as well as African sources, in order to arrive at a more

complete understanding of from where in Africa particular slaves came,
and where they actually went in the New World. Only then can work of the
sort I have attempted in this book—to analyze sociability, organization, and
identity construction among captive and freed Africans—be undertaken
with more grace and cohesion. This is beyond the reach of any one histo-
rian, and I hope that like-minded scholars will collaborate in research
groups across specializations (and national borders) to write what might
be called the microhistory of the African diaspora.
appendix
The following tables provide general demographic data for the
\ city of Rio de Janeiro and for each parish, based in the collection
of Livros de Batismo de Escravos housed by the acmrj. They show
the total of 9,578 records collected (table 14) and the 9,269 records used for
further calculations (table 15), according to parish. Until the seventeenth
century the city of Rio de Janeiro had only one parish. In 1632 it was
subdivided in two parishes: Sé (See) and Candelária. For decades between
1718 and 1750 the only available volumes were those of the Parish of Sé;
those of Candelária were lost or destroyed (tables 16 and 17). In 1751 the
city was subdivided into four parishes: Sé, Candelária, São José, and Santa
Rita. For the decade of 1751–60 I worked on the four urban parishes (tables
18, 19, 20, 21).
242 | Appendix
Table 14. Total Slave Baptisms in Rio de Janeiro, 1718–1760
parishes baptisms total

With Without Adults No age No
mother mother given information
Sé, 1718–26 983 21 127 728 50 1,909
Sé, 1744–60 1,381 0 274 9 49 1,713
Sé, 1751–60 893 7 311 13 67 1,291
Candelária, 1751–60 1,648 5 571 15 15 2,254
São José, 1751–60 1,117 3 315 6 49 1,490
Santa Rita, 1751–60 587 11 288 3 32 921
Total 6,609 47 1,886 774 262 9,578
Source: acmrj, Livros de Batismo de Escravos, Rio de Janeiro: Sé, 1718–26, 1744–60,
1751–90; Candelária, 1745–74; São José, 1751–90; Santa Rita, 1751–60.
Table 15. Total Slave Baptisms Valid for Analysis, 1718–1760
parishes baptisms total

Baptism with mother Baptism of adults
Sé (1), 1718–26 983 855 1,838
Sé (2), 1744–60 1,381 283 1,664
Sé (3), 1751–60 893 324 1,217
Candelária, 1751–60 1,648 586 2,234
São José, 1751–60 1,117 321 1,438
Santa Rita, 1751–60 587 291 878
Total 6,609 2,660 9,269
Source: acmrj, Livros de Batismo de Escravos, Rio de Janeiro: Sé, 1718–26, 1744–60,
1751–90; Candelária, 1745–74; São José, 1751–90; Santa Rita, 1751–60.
Appendix | 243
Table 16. Slave Baptisms in Sé Parish, Rio de Janeiro, 1718–1726
year baptisms total

With Infants Adults No age No
mother given infor-
mation
# % # % # % # % # % # %
1718 108 5.7 1 0.1 21 1.1 42 2.2 7 0.4 179 9.4
1719 111 5.8 1 0.1 12 0.6 66 3.5 5 0.3 195 10.2
1720 106 5.6 4 0.2 5 0.3 63 3.3 7 0.4 185 9.7
1721 116 6.1 0 0.0 10 0.5 114 6.0 7 0.4 247 12.9
1722 136 7.1 2 0.1 9 0.5 111 5.8 5 0.3 263 13.8
1723 115 6.0 0 0.0 0 0.0 87 4.6 3 0.2 205 10.7
1724 119 6.2 12 0.6 28 1.5 143 7.5 5 0.3 307 16.1
1725 136 7.1 1 0.1 36 1.9 76 4.0 5 0.3 254 13.3
1726 36 1.9 0 0.0 6 0.3 26 1.4 6 0.3 74 3.9
Total a 983 51.5 21 1.1 127 6.7 728 38.1 50 2.6 1,909 100
Source: acmrj, Livro de Batismo de Escravos, Rio de Janeiro: Sé, 1718–26.

a
year baptisms total

With mother Infants Adults No age No
given infor-
mation
# % # % # % # % # % # %
1744 171 10.0 1 0.1 16 0.9 5 0.3 10 0.6 203 11.9
1745 220 12.8 0 0.0 15 0.9 0 0.0 10 0.6 245 14.3
1746 187 10.9 0 0.0 40 2.3 0 0.0 6 0.4 233 13.6
1747 208 12.1 0 0.0 27 1.6 1 0.1 4 0.2 240 14.0
1748 192 11.2 0 0.0 44 2.6 0 0.0 5 0.3 241 14.1
1749 182 10.6 0 0.0 57 3.3 0 0.0 10 0.6 249 14.5
1750 220 12.8 2 0.1 75 4.4 3 0.2 2 0.1 302 17.6
Total a 1,380 80.6 3 0.2 274 16.0 9 0.5 47 2.7 1,713 100.0

a
244 | Appendix
year baptisms total

mother given infor-
mation
# % # % # % # % # % # %
1751 92 7.1 1 0.1 28 2.2 4 0.3 20 1.5 145 11.2
1752 80 6.2 0 0.0 27 2.1 0 0.0 6 0.5 113 8.8
1753 104 8.1 2 0.2 23 1.8 0 0.0 7 0.5 136 10.5
1754 94 7.3 0 0.0 20 1.5 0 0.0 1 0.1 115 8.9
1755 76 5.9 3 0.2 42 3.3 2 0.2 7 0.5 130 10.1
1756 93 7.2 1 0.1 43 3.3 1 0.1 9 0.7 147 11.4
1757 96 7.4 1 0.1 42 3.3 3 0.2 3 0.2 145 11.2
1758 96 7.4 0 0.0 26 2.0 1 0.1 4 0.3 127 9.8
1759 76 5.9 0 0.0 22 1.7 0 0.0 9 0.7 107 8.3
1760 86 6.7 0 0.0 38 2.9 1 0.1 1 0.1 126 9.8
Total a 893 69.2 8 0.6 311 16.0 12 0.9 67 5.2 1,291 100.0

a
Appendix | 245
Table 19. Slave Baptisms in Candelária Parish, Rio de Janeiro, 1751–1760
year baptisms total

given infor-
mation
# % # % # % # % # % # %
1751 194 7.8 0 0.0 60 2.4 1 0.0 2 0.1 257 10.4
1752 181 7.3 0 0.0 57 2.3 0 0.0 1 0.0 239 9.6
1753 214 8.6 0 0.0 52 2.1 3 0.1 2 0.1 271 10.9
1754 182 7.3 1 0.0 56 2.3 2 0.1 1 0.0 242 9.8
1755 177 7.1 2 0.1 64 2.6 4 0.2 2 0.1 249 10.0
1756 165 6.7 0 0.0 71 2.9 1 0.0 2 0.1 239 9.6
1757 179 7.2 0 0.0 60 2.4 3 0.1 3 0.1 245 9.2
1758 161 6.5 2 0.1 83 3.3 1 0.0 1 0.0 248 10.0
1759 195 7.9 0 0.0 68 2.7 0 0.0 1 0.7 107 8.3
1760 175 7.1 0 0.0 49 2.0 2 0.1 0 0.0 226 9.1
Total a 1,823 73.5 5 0.2 620 25.0 17 0.7 15 0.6 2,480 100
Source: acmrj, Livro de Batismo de Escravos, Rio de Janeiro: Candelária, 1751–60.

a
246 | Appendix
Table 20. Slave Baptisms in São José Parish, Rio de Janeiro, 1751–1760
year baptisms total

given infor-
mation
# % # % # % # % # % # %
1751 101 6.8 1 0.1 12 0.8 0 0.0 2 0.1 116 7.8
1752 105 7.0 1 0.1 24 1.6 0 0.0 0 0.0 130 8.7
1753 127 8.5 0 0.0 26 1.7 0 0.0 1 0.1 154 10.3
1754 115 7.7 1 0.1 29 1.9 3 0.2 4 0.3 152 10.2
1755 121 8.1 0 0.0 30 2.0 0 0.0 5 0.3 156 10.5
1756 101 6.8 1 0.1 52 3.5 2 0.1 3 0.2 159 10.7
1757 121 8.1 2 0.1 40 2.7 1 0.1 4 0.3 168 11.3
1758 111 7.4 1 0.1 36 2.4 0 0.0 8 0.5 156 10.5
1759 109 7.3 2 0.1 43 2.9 0 0.0 7 0.5 161 10.8
1760 106 7.1 0 0.0 23 1.5 0 0.0 9 0.6 138 9.3
Total a 1,117 75.0 9 0.6 315 21.1 6 0.4 43 2.9 1,490 100
Source: acmrj, Livro de Batismo de Escravos, Rio de Janeiro: São José, 1751–60.
a
Appendix | 247
Table 21. Slave Baptisms in Santa Rita Parish, Rio de Janeiro, 1751–1760
year baptisms total

mother given infor-
mation
# % # % # % # % # % # %
1751 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0
1752 25 2.7 0 0.0 17 1.8 0 0.0 5 0.5 47 5.1
1753 72 7.8 1 0.1 36 3.9 0 0.0 7 0.8 116 12.6
1754 60 6.5 1 0.1 28 3.0 1 0.1 3 0.3 93 10.1
1755 73 7.9 5 0.5 37 4.0 0 0.0 7 0.8 122 13.2
1756 78 8.5 3 0.3 41 4.5 2 0.2 2 0.2 126 13.7
1757 63 6.8 0 0.0 25 2.7 0 0.1 4 0.4 92 10.0
1758 73 7.9 1 0.1 41 4.5 0 0.0 1 0.1 116 12.6
1759 64 6.9 0 0.0 31 3.4 0 0.0 0 0.0 95 10.3
1760 79 8.6 0 0.0 32 3.5 0 0.0 3 0.3 114 12.4
Total a 587 63.7 11 1.2 288 31.3 3 0.3 32 3.5 921 100.0
Source: acmrj, Livro de Batismo de Escravos, Rio de Janeiro: Santa Rita, 1751–60.
a
notes
introduction
1 The classic work on the subject is Bastide, O Candomblé da Bahia.
2 Arquivo Nacional, Rio de Janeiro, ‘‘Estatutos da Congregação dos Pretos Minas
Makii no rj (1786). Cópia de documento da Biblioteca Nacional (1907).’’
3 Biblioteca Nacional, Rio de Janeiro, ‘‘Regra ou estatuto por modo de um
diálogo onde, se dá notícia das Caridades e Sufragações das Almas que usam
os pretos Minas, com seus Nacionais no Estado do Brazil, especialmente no
Rio de Janeiro, por onde se hão de regerem e governarem for a de todo o abuzo
gentílico e supersticioso; composto por Francisco Alves de Souza preto e
natural do Reino de Makim, um dos mais excelentes e potentados daquela
oriunda Costa da Mina.’’ (Rules or statutes in the form of a dialogue, where are
announced the vows and charitable functions of the Mina blacks and their
compatriots in Brazil, especially in Rio de Janeiro, where they must conduct
and organize themselves without the disgraces of heathen superstition; com-
posed by Francisco Alves de Souza, black, of the kingdom of Mahi, one of the
most excellent and powerful of the Mina Coast.)
4 Catálogo da Exposição de História do Brasil (1882). The exhibition, organized by the
National Library, was launched in December 1881.
5 The full citation to the manuscript is ‘‘Estatutos da Congregação dos Pretos
Minas Maki no Rio de Janeiro (1786),’’ Biblioteca Nacional, Rio de Janeiro. I
will refer to it simply as the Mahi Manuscript.
6 The modern definition of the Portuguese word derrota is ‘‘defeat’’ or ‘‘over-
throw,’’ but the word also has a nautical usage, referring to a ship’s route, and
in this sense derrota is a written account of a voyage.
7 C. Ginzburg, ‘‘Sinais.’’
8 The Mahi appear in Brazilian documents as Maki, Makim, Maqui, Maquim,
and Maí, and are di√erent from the Mahim, a group located on the coast of
modern Nigeria.
250 | Notes to Introduction
9 On the Nagô in Bahia, see Juana Elbein dos Santos, Os Nagô e a Morte; and Maria
Inês Cortes de Oliveira, ‘‘Quem Eram os ‘Negros da Guiné’?’’
10 The former indicates a union in Brazil of two linguistic and cultural traditions,
while the latter is a subgroup of the Jêje. But it should be emphasized that these
terms usually derive from the anthropological studies of candomblé, and not
from the historiography of slavery.
11 Verger, Os Orixás.
12 I. A. Akinjogbin, Dahomey and Its Neighbours, 1708–1818 (I first encountered this
in João José Reis, ‘‘Identidade e Diversidade Étnica nas Irmandades Negras no
Tempo da Escravidão’’); Karl Polanyi, Dahomey and the Slave Trade.
13 Verger, Fluxo e Refluxo do Tráfico de Escravos entre o Golfo do Benin e a Bahia de Todos os
Santos dos Séculos XVIII a XIX, 9, 91. The Mahi are included among the slave
groups that were known as ‘‘Mina’’ in Rio de Janeiro and Minas Gerais.
14 On the língua geral (general tongue), see Antonio da Costa Peixoto, Obra Nova de
Lingua Geral da Mina.
15 Jêje is a Bahian nation for African slaves who spoke Gbe languages, at the time
usually called Ewe or Eves in Portuguese.
16 Rodrigues, Os Africanos no Brasil, 105–6.
17 Translator’s note: The most basic definition of compromisso, the standard Por-
tuguese term in this context, is ‘‘commitment’’ (whether publicly or privately
assumed). Its applicability to the organizing principles of a religious brother-
hood resides in the word’s formal connotations of structure, ritual, and au-
thority, as well as in a separate spiritual or devotional sensibility. Clearly, the
formal statutes or laws that established a brotherhood were understood to
derive their meaning from their underlying relationship to Catholic beliefs and
hierarchy. An English translation might be ‘‘statutes and vows,’’ although the
actual term compromisso is preferable in the text.
18 The first is in Fitzler and Enes, A Seção Ultramarina da Biblioteca Nacional; the
second, is ‘‘Inventário de códices e de documentos avulsos do ahu referentes
ao Rio de Janeiro—316 códices abrangendo milhares de documentos com
datas entre 1548 e 183.—326 caixas com 27.446 documentos avulsos de 1614–
1853.’’ I thank Francisco Silva Gomes for bringing this to my attention. He is
also cited in Mulvey, The Black Lay Brotherhoods of Colonial Brazil, 297.
19 The catalog of the Conselho Ultramarino informs: ‘‘Compromisso da Irman-
dade de Nossa Senhora dos Remédios dos homens pretos de Minas (1788), a
qual se acha colocada na capela de Santa Efigênia no Rio de Janeiro. 1 vol.; 21 x
35, 5 cm; 26 fls. (13 fls. Br.) enc. Int. veludo: b. est. (16 imagens) ahu /cu-
cód.1300.’’ There is an error in the description when it mentions that the
request came from ‘‘pretos de Minas,’’ that is, blacks from the captaincy of
Minas Gerais in Brazil,’’ instead of ‘‘pretos da Mina,’’ that is, blacks from the
Mina Coast.
20 For the concept of an ‘‘ethnic text,’’ I relied on Joutard, ‘‘Un Projet Régional de
Notes to Introduction | 251
Recherche sur les Ethnotextes.’’ I have continued to work with Joutard’s essay;
some of my more recent reflections are included in this book’s postscript.
21 For more on this archive, see Antônio Alves Ferreira (Monsenhor) Santos,
Arquidiocese de São Sebastião do Rio de Janeiro; and José Carlos de Macedo Soares,
‘‘Fontes da História da Igreja Católica no Brasil.’’ According to the treaty
between the Catholic Church and Portugal, baptismal records were regarded as
o≈cial documents of the kingdom. For more on this point, see Maria Eulália
Lahmeyer Lobo, ‘‘Historiografia do Rio de Janeiro,’’ 45.
22 On commemoration and memory, see Pierre Nora, ‘‘Entre Mémoire et Histo-
ire.’’ Of course, Nora focused on national sites of memory in France, and never
delved into African slavery.
23 The full citation for the Portuguese general regulation promoted by Felipe I is
Código Philipino ou Ordenações e Leis do Reino de Portugal recompiladas por mando D’El
Rey D. Philipe I.
24 Constituições Primeiras do Arcebispado da Bahia.
25 By ‘‘register’’ I mean the individual biographical information. A record in the
baptism book can refer to one or more registers. Each page of a book contains
anywhere from four to twelve records, depending on the size of the script and
the size of the folio.
26 Debret, Viagem Pitoresca e Histórica ao Brasil, plate 35: ‘‘Negras cozinheiras,
vendedoras de angu’’; Agassiz and Agassiz, Viagem ao Brasil, 68.
27 Schwartz, Segredos Internos; Russell-Wood, The Black Man in Slavery and Freedom in
Colonial Brazil; Karasch, Slave Life in Rio de Janeiro, 1808–1850; Verger, Fluxo e
Refluxo do Tráfico de Escravos entre o Golfo do Benin e a Bahia de Todos os Santos dos
séculos XVIII a XIX; Reis, Rebelião Escrava no Brasil.
28 Scarano, Devoção e Escravidão; Boschi, Os Leigos e o Poder.
29 According to Braudel, certain structures resist change: ‘‘I reflected on the
di≈culty of breaking certain geographical frames, certain biological realities,
certain productivity limits, even this or that spiritual coercion: mental frames
can also be long-term prisons.’’ Braudel, Escritos sobre a História, 50.
30 Still according to Braudel, ‘‘a history with brief, rapid, nervous oscillations. By
definition ultra-sensitive, the smallest step throws into alert all its gauges and
instruments. As such, it is more passionate and rich in humanity, but more
dangerous as well.’’ Braudel, Escritos sobre a História, 14.
31 Oliveira, Identidade, Etnia e Estrutura Social; Oliveira, ‘‘Os Instrumentos de
Bordo’’; Bartolomé, Gente de Costumbre y Gente de Razón, 75–98. See also Yvonne
Maggie, Medo do Feitiço.
32 Elias, A Sociedade de Corte.
33 As Marc Bloch remarks: ‘‘The question, in a word, is not to know whether or
not Jesus Christ was crucified and came to life again. What we are trying to
learn at this point is how there come to be so many people who believe in the
Crucifixion and the Resurrection.’’ Bloch, Introdução à História, 33.
252 | Notes to Introduction
34 The notion of the ‘‘historical situation’’ was utilized by João Pacheco de Oliveira
in his analysis of the insertion of indigenous communities into national so-
ciety. Oliveira, ‘‘O Nosso Governo,’’54–59.
35 Fundamental reading on these issues includes Charles Boxer, A Idade de Ouro do
Brasil; Schwartz, Burocracia e Sociedade no Brasil Colonial, and Segredos Internos; and
Alfredo Bosi, Dialética da Colonização.
36 Boschi, Os Leigos e o Poder, 59–60.
37 I gained some insight on extant sources and bibliographies for this topic from
Eulália Lobo, História do Rio de Janeiro, and ‘‘Historiografia do Rio de Janeiro.’’
38 João Lúcio de Azevedo, Épocas de Portugal Económico; Maurício Goulart, A Es-
cravidão Africana no Brasil; Verger, Fluxo e Refluxo; Manolo Garcia Florentino, Em
Costas Negras; Luis Felipe Alencastro, ‘‘La Traite Négrière et l’Unité Nationale
Brésilienne’’; Klein, ‘‘The Portuguese Trade from Angola in the 18th Century’’;
Corcino Medeiros dos Santos, O Rio de Janeiro e a Conjuntura Atlântica.
39 Castro, Das Cores do Silêncio.
40 Maurício, Templos Históricos do Rio de Janeiro; Coaracy, Memórias da Cidade do Rio de
Janeiro; José de Souza Azevedo Pizarro Araújo, Memórias Históricas do Rio de
Janeiro; Moreira de Azevedo, O Rio de Janeiro; Luiz Gonçalves dos Santos, Memó-
rias para Servir à História do Reino do Brasil; Fazenda, Antiqualhas e Memórias do Rio
de Janeiro.
41 This is probably the moment to observe that the origin of this book, my
doctoral dissertation, was written side by side (and perhaps in some intertex-
tual argument) with two very important dissertations on the history of Rio de
Janeiro: Maria Fernanda Bicalho, ‘‘A Cidade e o Império: O Rio de Janeiro na
Dinâmica Colonial Portuguesa, Séculos XVII e XVIII’’ (Ph.D. diss., University of
São Paulo, 1997); and Nireu Oliveira Cavalcanti, ‘‘A Cidade do Rio de Janeiro:
As Muralhas, Sua Gente, os Construtores (1710–1810)’’ (Ph.D. diss., Federal
University of Rio de Janeiro, 1998). As both were defended after my own, they
could not be included here in the formal sense.
42 Ferrez, O Rio de Janeiro e a Defesa do seu Porto; and As Cidades do Salvador e Rio de
Janeiro no Século XVIII; Barreiros, Atlas da Evolução Urbana da Cidade do Rio de
Janeiro; Julião, Riscos Iluminados de Figurinhos de Brancos e Negros dos Uzos do Rio de
Janeiro e Serro do Frio; two editions of Debret, Viagem Pitoresca e Histórica ao Brasil
and O Brasil de Debret (the first is the complete text and images in black and
white, and the second contains selected colored plates).
43 Reis, Rebelião Escrava no Brasil.
44 On paleography, see Euripedes Franklin Leal and Ana Regina Berwanger,
Noções de Paleografia e Diplomática.
45 When I had finished the dissertation in 1997, I read Jeux d’Échelles, edited by
Jacques Revel. This book provided theoretical insights that I had been lacking
and, although it is not cited in the body of the text, has shaped subsequent
revisions of my work.
Notes to Chapter One | 253
1 | from ethiopia to guinea

1 This sketch of early visions of Africa is adapted from the documents and
commentary in Catherine Coquery-Vidrovitch, A Descoberta da África, 19–28.
2 Vitorino Magalhães Godinho, Ensaios II, 117–26.
3 Both gold and slaves were highly valued merchandise in African trade long
before the Portuguese arrived. Claude Meillassoux, Antropologia da Escravidão.
4 The fifteenth parallel crosses the coast of what is today Senegal, and the
Equator is close to the island of São Tomé. A. Teixeira Mota, Topónimos de Origem
Portuguesa na Costa Ocidental de África.
5 The historiography of slave commerce in Portuguese domains has often re-
ferred to Zurara. See João Lúcio de Azevedo, Épocas de Portugal Económico; and
Maurício Goulart, A Escravidão Africana no Brasil.
6 Luis Felipe Barreto, Descobrimentos e Renascimento, 67. The initial publication was
by J. P. Aillaud. I draw from a commemorative edition published 150 years after
Denis located Zurara’s manuscript. J. Bragança, ‘‘Introdução.’’
7 Zurara, Crônicas de Guiné, 46–47.
8 The idea of ransom was also linked to the salvation of pagans. The Portuguese
would buy Africans supposedly condemned to death, o√ering them their lives
and the possibility of saving their souls. Suess, Paulo, ‘‘Introdução Crítica,’’ in
A Conquista Espiritual da América Espanhola.
9 Azevedo, Épocas de Portugal Económico, 173.
10 Blake, European Beginnings in West Africa, 1454–1578, 22.
11 Zurara, Crônicas de Guiné, 146.
12 The symbol of the two palms is a recurring motif in letters from the period.
João de Barros Asia, 1a. Década referred to them as marking the transition
between the lands of the azenegues (southern moors) and of the negroes. They
also are drawn festooning the grounds of Mina Castle in some sixteenth-
century maps.
14 According to him, after 1448 ‘‘we dedicated our e√orts to merchandizing, not
to the responsible safeguarding and defense of our holdings.’’ This phrase was
highlighted by Pierre Chaunu in Expansão Européia do Século XIII ao XV, 113n134.
15 This a≈rmation, and what historians have made of it, takes on new meanings
in light of the debates around the formation of the modern Portuguese state.
Antônio Manoel Espanha, Às Vésperas do Leviathan, 22–36.
16 Kenneth Baxter Wolf, ‘‘The ‘Moors’ of West Africa and the Beginnings of the
Portuguese Slave Trade,’’ 449–69.
17 Suess, A Conquista Espiritual da América Espanhola, 225–32.
18 Luis Felipe Barreto, Descobrimentos e Renascimento, 65–66.
19 According to Bragança, the Romanus Pontifex of Nicholas V reflects a familiarity
with the first edition of the Chronicles of Guinea. Bragança, ‘‘Do Título desta
Crônica.’’ See also Godinho, Ensaios II, 122–26.
254 | Notes to Chapter One

21 Ibid., 124.
22 Pina, Crônicas de Rui de Pina; Pereira, Esmeraldo de Situ Orbis; Barros, Ásia, 88–89.
23 Cadamosto, a Venetian merchant, went there on an expedition sponsored by
Henry the Navigator. See excerpts from the Relation de Voyages à la Côte Occidentale
d’Afrique in Coquery-Vidrovitch, A Descoberta da África, 70–78.
24 Barros, Ásia.
25 Pereira, Esmeraldo de Situ Orbis, 68, 64.
26 Ibid., 69–70.
27 Barros, Ásia, 88–89. It seems that the next organized e√orts to convert natives
in the region were not until the second half of the sixteenth century, when
missionaries were sent to the kingdom of Warri in present-day Nigeria. J.
Cuvelier and L. Jadin, L’Ancien Congo d’après les Archives Romaines (1518–1640), 73.
28 Boxer highlights two factors for this—first, that the program for conversion set
up in Kongo was unsubstantial and weakly pursued, and second, that the
peoples of Angola were seen by the Portuguese as too poorly developed to
benefit from such ministrations. Boxer, Relações Raciais no Império Colonial Por-
tuguês (1415–1825), 27.
29 Pina, Crônicas de Rui de Pina, 992.
30 ‘‘É porém pelas Doações, e concessões Apostólicas, que os Reis seus ante-
cessors tinham do dito Senhorio, com é legitamente se poderam dele também
intitular; mas porque em seus dias, e até no tempo d’El Rei, foi Guiné coisa mui
pequena, e de pouca estima para Reis dela se intitularam, o deixaram por
ventura de fazer.’’ Pina, Crônicas de Rui de Pina, 934.
31 D. Afonso I wrote a letter of obedience to D. Manoel, king of Portugal in 1512.
Coquery-Vidrovitch, A Descoberta da África, 93–94.
32 Boxer, Relações Raciais no Império Colonial Português (1415–1825), 52.
33 Chaunu, Expansão Européia do Século XIII ao XV, 117.
34 Among many publications, Martellus’s and Cantino’s maps are in Max Justo
Guedes and Gerald Lombardi, Portugal Brazil. Pages 88–89 for Martellus’s; and
147 for Cantino’s.
35 Goulart, A Escravidão Africana no Brasil, 98–99. For more on the trading of slaves
from Kongo through São Tomé, see Jean Nsondé, ‘‘Les Relations Culturelles et
Commerciales entre Populations de Langue Kongo et Européens du XVIe au
XVIIIe Siècle dans la Région du Bas-Congo’’; and Isabel de Castro Henriques,
‘‘L’Invention Sociale de São Tomé et Principe au XVIe Siècle.’’
36 Goulart, A Escravidão Africana no Brasil, 101. See also Boxer, A Idade de Ouro do
Brasil, 29.
37 Azevedo, Épocas de Portugal Económico, 81.
38 Goulart, A Escravidão Africana no Brasil, 115–16.
39 He wrote, ‘‘One can say that the importation of black colonial strongbacks
[colonos] to Brazil involved members of all the nations.’’ Later, he clarified: ‘‘The
Notes to Chapter One | 255
best known in Brazil were those from Guinea (in whose number were Berbers,
Jalofos, Felupos, and Mandingas), Kongo, Mozambique, and the Mina Coast,
whence came the majority of slaves entering Bahia. The easy mode of travel
between Mina and Bahia led to the curious fact that slaves there rarely bothered
learning Portuguese and communicated with each other in nagô.’’ Varnhagen,
História Geral do Brasil, 1:224.
40 A. J. R. Russell-Wood, ‘‘United States Scholarly Contributions to the Histo-
riography of Colonial Brazil.’’
41 Rodrigues’s Os Africanos no Brasil was his last work of dozens. He died, in 1906,
before completing it; it was not published until 1933.
42 Romero, ‘‘O Brasil Social,’’ transcribed in Arthur Ramos, As Culturas Negras no
Novo Mundo, 185. The same passage with small alterations appears in Romero,
História da Literatura Brasileira, 199–200.
43 Azevedo, Épocas do Portugal Económico, 71.
44 Sérgio Buarque de Holanda, ‘‘Prefácio.’’
45 Goulart, A Escravidão Africana no Brasil, 185.
46 This and other maps from the era are in Mapas Históricos Brasileiros.
47 In his book about race relations, Boxer considered the Portuguese presence in
northwest Africa (from Morocco to Mina) but dwelled on the analysis of Kongo
and Angola. Boxer, Relações Raciais no Império Colonial Português, 16–43.
48 Mauro, Le Portugal et l’Atlantique au XVII Siècle (1570–1670), 158.
49 There is yet to be written a book about African and Bahian history that comes
close to what José Roberto do Amaral Lapa attained with respect to India in A
Bahia e a Carreira da Índia.
50 There were no studies about Africans from the Mina Coast in the rest of the
captaincy of Rio de Janeiro at the time this volume was first published. For
further information, see Soares, Rotas Atlânticas da Diáspora Africana.
51 This terminology, characteristic of the cultural school of anthropology, will be
analyzed more closely in the pages that follow in this chapter. The first edition
of this work by Ramos appeared in 1937; that makes it a contemporary of the
publications of Herskovits’s group, whose style it approximated. According to
Ramos, the term Mina referred specifically to the Fanti-Ashanti. See Ramos, As
Culturas Negras no Novo Mundo, 181, 185–87.
52 Herskovits, Antropologia Cultural.
53 Polanyi, Dahomey and the Slave Trade, ix.
54 I could not find reliable information as to whether the books were written
separately or in conjunction, but the fact that they appeared only two years
apart suggests that they were produced more or less contemporaneously.
55 Bastide echoed Nina Rodrigues on the extension of Guinea from Senegal to
Orange. Bastide’s analysis of the seventeenth century is curiously limited to the
assertion that most slaves deviam ser (should be) Bantus.
56 I take in account here the explorations and conquests of Henry the Navigator
256 | Notes to Chapter Two
(1434–60). Dom Afonso V was the first to incorporate the African possessions
into the titles of the Portuguese Crown
57 The island of Príncipe was named in 1470 in homage to the young prince and
future king Dom João II, who began his rule in 1481.
58 Rodrigues, Os Africanos no Brasil, 23.
59 Bastide, As Religiões Africanas no Brasil, 68.
2 | commerce with the mina coast

1 In Brazil, the subject of Indian labor and religious education has intrigued both
historians and anthropologists. See Manuela Carneiro da Cunha, História dos
Índios no Brasil; and John M. Monteiro, Negros da Terra.
2 Coaracy, O Rio de Janeiro no Século Dezessete, 33–34, 44.
3 Delgado de Carvalho, História da Cidade do Rio de Janeiro, 32.
4 Pizarro alleged that the contracts for slave imports to Brazil, and specifically to
the captaincy of Rio de Janeiro, represented the first o≈cial Portuguese trans-
actions for slaves in Mina Coast. But at one point Pizarro seemed confused
about the dates for the 1615 permit and the contract that followed it, suggesting
that they emerged on the same date. Araújo, Memórias Históricas do Rio de Janeiro,
239.
5 Coaracy observed that this was the beginning of regular stowage service, in the
soon-familiar figure of the black Brazilian stevedore. Coaracy, O Rio de Janeiro no
Século Dezessete, 46.
6 Ibid., 66.
7 According to Mauro, the durations of Atlantic trips from Angola were usually
as follows: 35 days to Pernambuco; 40 days to Bahia; 50 days to Rio de Janeiro.
Mauro, Le Portugal et l’Atlantique au XVII siècle (1570–1670), 171.
8 On the relations between Rio de Janeiro and Angola, see Boxer, Salvador de Sá e a
Luta pelo Brasil e Angola (1602–1686).
9 Coaracy, O Rio de Janeiro no Século Dezessete, 72.
10 Boxer, Salvador de Sá e a Luta pelo Brasil e Angola (1602–1686).
11 Schwartz, Segredos Internos, 162.
12 abn , vol. XXXIX, doc. 245, 28.
13 The capitão-mór was an administrative position beneath, and typically preced-
ing, that of governor. As a captaincy grew in population and developed greater
political and commercial complexity, a governor would be named; captain-
majors usually saw themselves as the most suitable candidates for that role.
The position was common in Portuguese territories in Brazil, Angola, and the
Atlantic islands.
14 Boxer, Salvador de Sá e a Luta pelo Brasil e Angola (1602–1686), 202.
15 Coaracy, O Rio de Janeiro no Século Dezessete, 121.
16 The allegations against him are heated, and perhaps because of this are occa-
Notes to Chapter Two | 257
sionally oddly worded: ‘‘And proof of all this is that when Salvador Correia
came to this realm to buy the post of capitão-mór, he had nothing of his own,
and in fact was owing, as his own mouth confessed, many times 38,000
cruzados, and he had more than 300, but in salary less than 200,000 réis, to serve
5 years.’’ abn , vol. XXXIX, doc. 268, 30.
17 Boxer, Salvador de Sá e a Luta pelo Brasil e Angola (1602–1686), 224; Schwartz,
Segredos Internos, 160.
18 Polanyi, Dahomey and the Slave Trade, 101.
19 Ibid., 102.
20 The Dutch stayed at Mina Coast until 1872, when they sold to England the forts
still under their power. These included Mina Castle, Fort Komenda (Vreden-
burg), and Fort Saint Anthony (which the Portuguese had called Axim when
they built it in 1515). From that point, the Cape Coast colony acquired bound-
aries approximating modern Ghana.
21 Coquery-Vidrovitch, A Descoberta da África, 130, 111.
22 Fynn, Asante and Its Neighbours, 1700–1807.
23 The patacho was a common commercial vessel, if not as large as some. Docu-
mentos Históricos—1681–1686 (collection of documents published by the bn /
rj, vol. XXVIII, 1934); António Carreira, As Companhias Pombalinas de Navegação,
140.
24 Florentino, Em Costas Negras, 123–29.
25 Arquivo Público do Estado da Bahia, 7, 108. Cited by Verger, Fluxo e Refluxo do
Tráfico de Escravos entre o Golfo do Benin e a Bahia de Todos os Santos dos séculos XVIII a
XIX, 39–40.
26 Boxer, A Idade de Ouro do Brasil, 68.
27 Palanquim or liteira refers to a sedan chair, typically curtained, suspended on
two sturdy poles, the four handles of which slaves would hoist on their shoul-
ders in order to carry the chair through the streets. It was a form of conveyance
for the elite in colonial times and can be seen in drawings from the period.
28 abn , vol. XXXIX, doc. 2.815, 295–96.
29 abn , vol. XXXIX, doc. 2.917, annexed doc. 2.913, 302–4.
30 Manolo Florentino suggests that there was regular trade between Mina Coast
and Rio de Janeiro early in the eighteenth century, carried out by resident
merchants: ‘‘To attend the stupendous demand for captives caused by the
discoveries at the mines, they used the gold to buy slaves from diverse parts of
Africa. They often bought from the Dutch and English, using gold dust or gold
bars, outside of o≈cial control.’’ He believes this trade no longer existed after
1816. Florentino, Em Costas Negras, 123, 86.
31 We know in one case that the wealthy Portuguese trader Francisco Pinheiro
operated concerns in both Angola and Mina Coast, but his relations with
traders in Rio remains mostly unknown.
32 Statement dated 1 September, 1706. abn , vol. XXXIX, doc. 2.913, 301–2.
33 João António Andreoni, Cultura e Opulência do Brasil; Goulart, A Escravidão Africana

no Brasil, 165; ‘‘Diário da Jornada que fez o Exmo. Senhor Dom Pedro desde o
Rio de Janeiro até a cidade de São Paulo, e desta até as Minas ano de 1717,’’
Revista do Serviço do Patrimônio Histórico e Artístico Nacional no. 3 (1939): 295–316.
34 Goulart, A Escravidão Africana no Brasil, 153, 217.
35 This might be explained by the fact that in Angola, group baptisms were
commonly performed on slaves in the barracks before boarding ship. See
Boxer, Salvador de Sá e a Luta pelo Brasil e Angola (1602–1686).
36 This point receives more detailed argument in Polanyi, Dahomey and the Slave
Trade, 134.
37 The reference to Akinjogbin and his book Dahomey and Its Neighbours is in Reis,
‘‘Identidade e Diversidade Étnica nas Irmandades Negras no Tempo de Es-
cravidão,’’ 18.
Santos dos séculos XVIII a XIX, 669.
39 Schwartz, Segredos Internos, 282. After that time, the Portuguese continued to
hang on to the fort (often at great cost) until the twentieth century.
41 ‘‘Contrato novo dos direitos que pagam os escravos que entram no Rio de
Janeiro, vindos da Costa da Mina e Cabo Verde, arremetado a Jerônimo Lobo
Guimarães por tempo de 3 anos e por preço em cada um deles de 50.00
cruzados. Lisboa, 9 de fevereiro de 1725.’’; icabn, vol. XXXIX, doc. 5.325, 465.
42 ‘‘Contrato novo dos direitos dos escravos, que vão para as Minas do Porto do
Rio de Janeiro, que se fez no Conselho Ultramarino com Jerônimo Lobo Gui-
marães, por tempo de três anos e por preço em cada um deles de 36.000
cruzados e 300,000 réis. para a Fazenda. Lisboa, 28 de março de 1725.’’ abn ,
vol. XXXIX, doc. 5.460, annexed to 5.459, 474.
43 abn , vol. XXXIX, doc. 4.212, 397.
44 ‘‘Informação do Provedor da Fazenda em que expõe as suas dúvidas sobre a
execução do contrato de Jerônimo Lobo Guimarães e os direitos que deveriam
pagar os escravos. Rio de Janeiro, 03.07.1725.’’ abn , vol. XXXIX, doc. 4.941,
annexed to 4.932, 441.
45 ‘‘Ata da Junta,’’ 5 July 1725. abn , vol. XXXIX, doc. 4.942, 441.
46 The petitions are available in abn , vol. XXXIX, doc. 4.932–4.937, 441.
47 ‘‘Termo de fiança em que os referidos comerciantes se responsabilizaram pelo
contrato dos direitos dos escravos, se o renunciasse Jerônimo Lobo Gui-
marães. Rio de Janeiro, 14.07.1725.’’ abn , vol. XXXIX, doc. 4.940, 441; ‘‘Rep-
resentação dos comerciantes da Praça do Rio de Janeiro Domingos Martins
Breto, Bernardo Alves da Silva e Domingos Gonçalves Barreiros sobre os
direitos dos escravos que exigia o contratador Jerônimo Lobo Guimarães, de
24.07.1725.’’ abn , vol. XXXIX, doc. 4.939, annexed to 4.932, 441.
48 There were two o≈cial decrees to that e√ect, issued on 13 October 1725 and 20
May 1726. abn , vol. XXXIX, docs. 5.326 and 5.327, annexed to 5.324, 465.
Notes to Chapter Two | 259
49 abn , vol. XXXIX, doc. 5.234, 459.

50 Boxer referred to this tension in ‘‘Brazilian Gold and British Traders in the First
Half of the Eighteenth Century,’’ 460.
51 abn , vol. XXXI, doc. 347, 29–30.
52 Goulart, A Escravidão Africana no Brasil, 168.
53 As Schwartz noted, ‘‘The impact of the magistracy on colonial society should
be seen not only in terms of its professional attitudes but, also, in light of the
lifestyles and personal motivations of the magistrates and of the reactions or
initiatives of certain elements of the colonial population.’’ Schwartz, Burocracia
e Sociedade no Brasil Colonial, 251.
54 Verger, Fluxo Refluxo do Tráfico de Escravos entre o Golfo do Benin e a Bahia de Todos os
Santos dos séculos XVIII a XIX, 28–29. The designation ‘‘Portuguese’’ here is
uncertain; does it include ships from Rio that might not have had contracts or
licenses from the Ultramarine Council?
55 ‘‘With the discovery of gold in Minas Gerais in the final decade of the seven-
teenth century, it became urgently necessary to obtain slaves that were stronger
and more capable for mining work and mine conditions than the Bantus of
Angola and Kongo. This was resolved by reopening the commerce of slaves
between Brazilian ports and the ‘Mina Coast,’ as the Portuguese called Lower
Guinea.’’ Boxer, Relações Raciais no Império Colonial Português (1415–1825), 22.
56 He cited a letter from the viceroy to the secretary of state to that e√ect. Arquivo
Público do Estado da Bahia, ordens régias, vol. 34, doc. 15. Russell-Wood,
Fidalgos e Filantropos, 33.
57 Boxer, A Idade de Ouro do Brasil, 68.
58 Manuscript from the Municipal Library of Porto, apud Scarano, Devoção e Es-
cravidão, 107.
59 Boxer, ‘‘Brazilian Gold and British Traders in the First Half of the Eighteenth
Century,’’ 461.
60 In 1734, the kingdom of Dahomey prohibited the transport of gold outside the
borders of the dominion. See Fynn, Asante and Its Neighbours, 1700–1807, 15.
61 Consultation of the Oversea Council about a petition from Cristóvão Pereira de
Abreu, tobacco contractor residing in the city of Rio de Janeiro, 10 March 1716.
abn , vol. XXXIX, doc. 3.474, 351. Florentino, compiling a list of slave traders
active between 1811 and 1830, includes the name of Joaquim José Pereira de
Abreu. Florentino, Em Costas Negras, 281.
62 Not all the tobacco was exported. Bahian tobacco was again most prized by
colonial consumers, especially when it had been treated with a sweet paste
derived from sugar cane.
63 Translator’s note: An arroba was a unit of weight that is estimated today at 15
kilograms.
Santos dos séculos XVIII a XIX, 26. Rio’s role as a broker of tobacco from elsewhere
in Brazil seems to have begun o≈cially in 1695. In 1700, its 7,750 cruzados of
earnings put it in third place on the list of the captaincy’s most valuable
contracts, behind the royal tenth and whaling. Jean Baptiste Nardi, O Fumo
Brasileiro no Período Colonial, 286–94.
65 abn , XXXIX, annex 5.975, 513. This information also appears in Fazenda,
‘‘Antiqualhas e Memórias do Rio de Janeiro.’’ rihgb , book 89, vol. 143, 490.
66 ‘‘Capitania do Rio de Janeiro—correspondência de várias autoridades e avul-
sos—ano de 1757.’’ rihgb , book 65, pt. 1 (1902), 88. Monsenhor Pizarro
dated the opening of the contract at 1757. Araújo (Monsenhor Pizarro), Memó-
rias Históricas do Rio de Janeiro, 2:247.
67 The 1794 almanac mentions the arrival of slaves from ‘‘Angola, Benguela, and
the Mina Coast,’’ but the 1798 edition does not include the Mina Coast as a site
of provenience for slave imports into the city. Nunes, ‘‘Almanaque histórico da
cidade de S. Sebastião do Rio de Janeiro, 1799’’ rihgb , vol. 267 , 1965: 93–
214.
68 ‘‘Sempre é preciso e justo que V.M. faça contribuição annual pelos mesmos 4
anos para a edificação da nova Sé, principalmente desta, cujos dízimos cobra a
fazenda real; e atendendo à conjuntura presente em que tem a Real fazenda
tantas despesas, se me representa que será boa e suficiente a aplicação de
30.000 cruzados nos ditos 4 anos. Para que as rendas atuais de V.M. não
recebem desta contribuição prejuízo algum, basta que V.M. conceda licença
para os contratadores de tabaco desta cidade poderem tirar da Bahia mais 700
arrobas de tabaco de fumo, além das que pelo contrato tiram, porque esta
licença importará mais que aquela contribuição e será muito conveniente e
justa, porque o tabaco, que tiram por contrato, não basta para o gasto que tem
na cidade e está muitos meses o estanco sem ele com grande prejuízo dos
brancos e maior parte dos pretos, que no fumo do tabaco se sustentam e
vivem.’’ Letter dated 3 February 1709. abn , vol. XXXIX, doc. 3238, annexed to
3.236, 326–27.
69 The tobacco contract in the city of Rio de Janeiro during this period allotted 1
percent of profits for obras pias, or church-run charities. Nardi, O Fumo Brasileiro
no Período Colonial, 290.
70 Workers involved in the tobacco contract were divided into three categories:
slaves, soldiers, and laborers. Most were paid by the day. Nardi, O Fumo Brasi-
leiro no Período Colonial, 296–98.
71 There were other brotherhoods composed of Mina blacks in the churches of
São Domingos and Our Lady of the Rosário, but very little information has
been found about them. I will return to them in chapter 4.
72 Silveira, ‘‘Conferência,’’ 4 (photocopy); Verger, Fluxo e Refluxo do Tráfico de Es-
cravos entre o Golfo do Benin e a Bahia de Todos os Santos dos séculos XVIII a XIX, 191.
73 On 12 January 1752, the High Court judge João Eliseu de Souza began an
inquiry commanded by royal decree into the connections between Teodósio
Rodrigues and João Dias Cunha, both administrators of tobacco, and the
Notes to Chapter Three | 261
viceroy Conde de Athouguia, who was accused of having undisclosed personal

interests in the tobacco trade. Through the mediation of Teodósio, Conde de
Athouguia received presents from the king of Dahomey. abn , vol. XXXI, doc.
730, annexed to doc. 729, 57; doc. 794, 60.
74 Campos, Procissões Tradicionais da Bahia, 78.
75 At an unknown later date the other two brotherhoods, Menino Jesus and Santo
Antônio of the Moors, also constructed chapels.
76 Maurício, Templos Históricos do Rio de Janeiro.
78 Some of them also spoke Yoruba.
3 | african ‘‘nations’’ and provenience groups

1 Dating from the twelfth century, when it called for a common authority for
religious and military orders, the Padroado was conceded to the Order of Christ
in the papal bull Inter Coetera (1456). Leadership of the Order passed to the king
of Portugal, rather than a religious o≈cial, starting with the reign of D. Manoel
(1469–1521), which established a link between ecclesiastical and temporal
powers. In 1522, King João III became the head of both the Order of Christ and
the church administration in Portugal; when he attained the leadership of the
religious orders thirty years later, the consolidation of religious and civil power
in the Portuguese throne was e√ectively complete. The crown’s dual functions
were especially apparent in the colonies and occupations. After 1822, the
Padroado continued nominally in the person of the emperor of Brazil. It was
abolished in 1890 with the proclamation of the republic. For more details, see
Graça Salgado, Fiscais e Meirinhos, 113–21.
2 Depending on many circumstantial factors, more or less information might
have been recorded. Sometimes the entry for an adult slave includes a descrip-
tion of the slave’s religious character and doctrinal preparedness for the event.
3 I recall here what Schwartz called ‘‘metropolitan administration’’ versus ‘‘webs
of primary interpersonal relations.’’ Schwartz, Burocracia e Sociedade no Brasil
Colonial, xi–xii.
4 I am aware of the argument that the Portuguese state perceived Brazil as a
complete whole, while the perspective from the Brazilian colony of itself was
one of fragmentation in which all manner of connections had to be imagined
and created. For another look at this idea, see Afonso Carlos Marques dos
Santos, No Rascunho da Nação, 23.
5 In 1707, through the e√orts of the archbishop of Bahia, a synod of the arch-
diocese was convened with the goal of adapting the determinations of the
Council of Trent to Brazil. The meeting was downgraded to a synod of the
diocese when so many bishops failed to turn out, including the bishop from
the Diocese of Rio de Janeiro. The results of this meeting were compiled in the
262 | Notes to Chapter Three
Constituições Primeiras do Arcebispado da Bahia, a work subsequently used to regu-

late ecclesiastical proceedings in the colony.
6 This point is also raised by Sheila Siqueira de Castro Faria, ‘‘A Colônia em
Movimento,’’ 283–339.
7 Variations in the form of taking down information seem rarely to have both-
ered church authorities. Only once (in the acmrj, Livro de Batismo de Escravos,
Freguesia da Sé, 1718–26) did I see a written reprimand; it was placed alongside a
record of a group baptism in which the individual registers were jumbled and
di≈cult to interpret.
8 According to Schwartz in Bahia, a varied terminology was used to identify
children older than infants but younger than adults: cria de peito (still breast-
feeding); cria de pé (already walks); menino/menina (until around age two); mole-
que/moleca (until age fourteen); moleque grande (until age twenty); and rapaz/
rapariga (undetermined age, depending on appearance). Schwartz, Segredos
Internos, 288. In the baptism records from Rio de Janeiro, almost all children
were identified by reference to the mother, and called inocente. This makes it
essentially impossible to determine the age of a given child, although most
were between newborn and seven. In Sé Parish, starting in 1751, the curate
began to regularly include the child’s date of birth. Almost all of the records he
wrote were for ceremonies performed when the child was around a month old,
so that all of them were called ‘‘inocentes.’’
9 There is not a reliable estimate for childbearing among female slaves in Rio de
Janeiro in the eighteenth century. For Salvador in the nineteenth century, Katia
Mattoso derived an average of two children per mother. Katia M. Mattoso,
Bahia, Século XIX, 168.
10 According to Mary Karasch, in Rio de Janeiro in the nineteenth century ‘‘the
term negro alone often implied African slave.’’ Karasch, Slave Life in Rio de Janeiro,
1808–1850, 5. I could not verify this claim for the eighteenth century. In table
13, I included the ‘‘preto’’ (black, in this case man or woman) among the non-
Africans. The terms negro (negro man) and negra (negro woman) do not appear
in baptism records of this era.
11 acmrj, Livro de Batismo de Escravos, Freguesia da Sé, 1718–26. The two cases of
freed women were maintained to show that emancipation was not enough to
qualify being recorded in the book of freed blacks, rather than the book of
captive blacks.
12 acmrj, Livro de Batismo de Escravos, Freguesia da Sé, 1744–61, 5 May 1745.
13 The Brotherhood of Santana was created in 1735 by a group of creoles, but by
the nineteenth century it was identified as an association of soldiers. The story
of that relationship, or transformation, remains to be told.
14 In 1871, with the Lei do Ventre Livre (Law of the Free Womb), children born to
slave mothers were legally free.
15 acmrj, Livro de Batismo de Escravos, Freguesia da Candelária, 1745–74, fols. 99v, 100.
16 I was unable to perform a careful survey of the Books of Baptisms of Whites to

know how many children of slaves were recorded there.
17 ‘‘In the bustling ports of Recife, Salvador, and Rio de Janeiro, as well as in the
immediate interior of these centers, negros and mulattos predominated, with
pure whites in second place, and Indians and caboclos in the third.’’ Boxer, A
Idade de Ouro do Brasil, 6.
18 Schwartz, Segredos Internos, 288.
19 Law dated 1 April 1680. See also the Regimento das Missões do Estado do
Maranhão e Pará, 1 December 1686; and Soares, ‘‘O Missionário e o Rei.’’
20 Antonio Moraes e Silva, Dicionário de Língua Portuguesa.
21 Paul Augé, Larousse du XXème Siècle.
22 Carreira, Notas sobre o Tráfico Português de Escravos, 22.
23 Mota, Topónimos de Origem Portuguesa na Costa Ocidental de África, 23. Mota also
noted that French cartography, the most prevalent in the eighteenth century,
made use of Portuguese toponymy.
24 The theme is present in the writings of Padre Antônio Vieira and João Francisco
Lisboa. In the recent historiography, see Schwartz, Segredos Internos, 58.
25 According to Schwartz, in Bahia the expression ‘‘native negro’’ (negro da terra)
was used by both Jesuits and colonists, but fell into disuse in the sixteenth
century as indigenous slavery waned and the number of Africans in the area
increased. Ibid. The expression lived on in Rio de Janeiro and São Paulo until
the eighteenth century. Monteiro, Negros da Terra.
26 The 1849 city census, the earliest reliable source for local demographics, re-
vealed that Africans made up 66.4 percent of the total slave population. Ka-
rasch, Slave Life in Rio de Janeiro, 1808–1850, 8.
27 Page 275 of the 1718–26 book contains five entries. Two, referring to ‘‘gentio
from Guinea,’’ are by Barbosa. The other three are from Rodrigues Cruz: two
pardos and one ‘‘Mina,’’ in place of Barbosa’s ‘‘gentio from Mina.’’ I could not
compare these registers with those of Candelária from the same time period
because they are described as lost by the acmrj.
28 Mina, Angola, and Guinea slaves were responsible for 796 births recorded
between 1718 and 1726; only 62, or 8 percent, were legitimate. The census of
1872, the first o≈cial census of the Brazilian population, indicated that 10
percent of slaves were married or widowed. Mattoso, Ser Escravo no Brasil, 126.
29 However, there was an unusual register worth mentioning in 1754 for the
baptism of Ignacia: she was the legitimate child of João and Suzana, ‘‘ambos de
Guiné, nação benguela’’ (both of Guinea, Benguela nation). acmrj, Livro de
Batismo de Escravos, Freguesia de São José, 1754.
30 S. M. Baldé, ‘‘L’Esclavage et la Guerre Sainte au Funta-Jalon,’’ 192.
31 One eighteenth-century Coura woman, Rosa Egipcíaca—who went from being
a prostitute to a devout—is the subject of Luiz Mott’s Rosa Egipcíaca. The same
author also mentions Coura slaves in Minas Gerais in his article ‘‘Acotundá.’’
264 | Notes to Chapter Three
The four Coura identified in Rio de Janeiro’s baptism records were adults
undergoing baptism between 1751 and 1760; in the same period, two children
were born to a Coura woman.
32 Most of the records used the term ‘‘Mina’’ without further qualification, al-
though the occasional reference to a Mina subgroup demonstrated the grow-
ing perception of diversity. On the other hand, the central part of the West
Coast is represented by a grand variety of names of ports, locales, or commu-
nities, which correlates with the more intense scale of commerce there.
33 To explore behind these names a bit more, Loanda (Luanda) was a Portuguese
city, founded in 1575. Kongo and Kasanje were kingdoms. Cabinda was a port.
Massangano and Ambaca were locations of seventeenth-century Portuguese
forts that fought o√ the Dutch. Boxer, Salvador de Sá e a Luta pelo Brasil e Angola,
182. Both Luanda and Massangano had a Santa Casa da Misericórdia. Russell-
Wood, Fidalgos e Filantropos, 28. On the variety of African groups in nineteenth-
century Rio de Janeiro, see Karasch, Slave Life in Rio de Janeiro, 1808–1850
(including map on 16); and Boxer, O Império Colonial Português (1415–1825)
(especially maps 3 and 6). Spelling of these names is not consistent in the
documents, although there are some standard variations: the letters e and a can
be switched, as in Embaca/Ambaca or Benguela/Banguela. Similarly, o and u
can alternate, as in Loanda/Luanda or Cacheo/Cacheu.
34 The wording of such registers was generally like this one: ‘‘On the 26th of
August of 1753, in the Church of São José in this city I baptized and placed the
Holy water on Maria, adult, baptized sub conditione, as she had requested of her
master (Senhor So-and-So) to arrange her baptism, because she had not had
that sacrament and could not remember having it, although her master sus-
pected that she had . . . and it being asked that I examine said slave on the
matter, I did so and determined that she should be baptized sub conditione,
which I performed. The godparents were——’’ acmrj, Livro de Batismo de Es-
cravos, Freguesia de São José, 1751–90.
35 Patricia Ann Mulvey, The Black Lay Brotherhoods of Colonial Brazil, 112.
36 See Livro de Batismo de Escravos, Freguesia da Candelária, 1745–74, pp. 99, 118v,
154v, 109v. At the time the bishop of Rio de Janeiro was D. Frei Antônio do
Desterro Malheiros O.S.B. [Ordem de São Bento] (1746–73). His episcopate
marked a new phase in the religious life of Rio de Janeiro. The new bishop was
a Benedictine doctor of theology from Coimbra. He visited Rio briefly in 1740
en route to Luanda, where he was the bishop of Kongo and Angola. He
returned to the city in 1747, and was responsible for creating the two new
parishes, of São José and Santa Rita. He renewed and reinvigorated the com-
mitments of parish priests to centrally defined rules and standards, including
the norms for writing baptismal records.
37 Constituições Primeiras do Arcebispado da Bahia, Livro 1\, III, 6, 8; and Livro 1\, XIV,
47, 50, 53, 57.
38 Boxer, A Idade de Ouro do Brasil, 29. Elsewhere Boxer described the typical
baptismal proceedings for slaves in seventeenth-century Luanda: ‘‘The slaves
waiting to be exported had been packed into large sheds . . . When the morning
broke of the day they were to set sail, they were taken to a nearby church, or to
some other adequate place, where a priest would baptize them in bunches,
several hundred at a time. The ceremony was quick. Each captive appeared
briefly before the priest, who said to him} ‘Your name is Pedro,’ or ‘João,’ or
‘Francisco,’ and so forth, and who handed each one a scrap of paper with the
name written on it before quickly tossing a pinch of salt in the captive’s mouth.
That part finished, the priest then flung holy water on the crowd, and a negro
interpreter cried out the message ‘Alright, you are all now children of God,
setting out on the path to Portuguese (or Spanish) lands where you will learn all
about the faith. Forget everything having to do with where you come from; stop
eating dogs, rats, or horses. Now you may go, and be happy.’ ’’ Boxer, Salvador
de Sá e a Luta pelo Brasil e Angola, 243.
39 Andreoni, Cultura e Opulência do Brasil.
40 About India, see Lapa, A Bahia e a Carreira da Índia.
41 acmrj, Livro de Batismo de Escravos, Freguesia de Candelária, 1718–26.
42 acmrj, Livro de Batismo de Escravos, Freguesia da Candelária, 1745–74, fols. 49v,
69v. In the Constituições Primeiras do Arcebispado da Bahia, Livro 1\, XIV, item 53
specifically addresses the baptism of children of so-called infidel mothers.
43 In 1740, the Brotherhood of Santo Elesbão and Santa Efigênia was founded by
peoples deriving from four ‘‘nations’’: Mina, São Tomé, Mozambique, and
Cabo Verde.
44 Fredrik Barth, Ethnic Groups and Boundaries, 26n5.
45 Bartolomé, ‘‘Bases Culturales de la Identidad Étnica.’’
46 The idea of the general language became particularly widespread after a written
vocabulary was compiled in the mid-eighteenth century. Peixoto, Obra Nova de
Língua Geral da Mina. Luiz Carlos Villalta appears to have been unaware of the
existence of purported general African languages in ‘‘O Que Se Fala e O Que Se
Lê.’’
47 To be precise, use of the word African to refer all the peoples who had come to
Brazil from Africa became more common practice after 1830 and was associ-
ated with new legislation that dealt with so-called free Africans. For more on
this period, see Leslie Bethell, A Abolição do Tráfico de Escravos no Brasil.
48 Oliveira, ‘‘Os Instrumentos de Bordo,’’ 118.
49 Scarano, Devoção e Escravidão, 108.
50 Boxer also suggested that most Mina slaves were Yoruba speakers, an error in
light of recent research. Yoruba speakers began to be tra≈cked as slaves near
the end of the eighteenth century. Boxer, A Idade de Ouro do Brasil, 195–96.
51 In the Rio documents, maki and sabaru are mentioned; in the Minas documents,
maqui and sabará.
266 | Notes to Chapter Four
52 See Verger, Fluxo e Refluxo do Tráfico de Escravos entre o Golfo do Benin e a Bahia de
Todos os Santos dos séculos XVIII a XIX. The stereotype of the Mina woman as
statuesque and noble, reproduced by both Elizabeth Agassiz and Christiano Jr.
in the second half of the nineteenth century, is due more to the internal
tra≈cking of Mina slaves within Brazil at the time (especially between Bahia
and Rio) than to Rio’s commerce with the Mina Coast, already moribund by
this point. Agassiz and Agassiz, Viagem ao Brasil: 1865–1866; Azevedo and
Mauricio Lissovsky, Escravos Brasileiros do Século XIX na Fotografia de Christiano Jr.
53 ‘‘The supply of African arms and backs had to be elastic, and cheap.’’ Floren-
tino, Em Costas Negras, 85.
54 Several types of vessels were specified throughout the sources on Rio’s small-
scale naval commerce—patachos, sumacas, and galeras—although o≈cial fleets
would utilize larger ships. Between 1726 and 1728, I found reference to eleven
di√erent departures from Rio de Janeiro to Lisbon: four naus, six navios, and
one galera. abn , XXXIX, 469–517.
55 ‘‘Some slave masters were in favor of marriage, and not only did not prevent it
but openly suggested it to couples they observed, saying in essence ‘You, slave,
should marry that one in good time.’ They would treat such slave couples as if
they had been received in Brazil thus, as husband and wife. But they were
reluctant to impose marriage, fearing that the slaves would ultimately become
annoyed and despondent within the bonds of the institution and might try to
kill each other through hexes, spells, or stronger measures.’’ Andreoni, Cultura
e Opulência do Brasil, 160–61.
56 See Mattoso, Kátia M. de Queiróz. Bahia, Século XIX: Uma Província no Império. Rio
de Janeiro: Editora Nova Fronteira, 1992: 161–169. For the analysis based on
Reis’ data see in particular pages 164–165.
57 These were the terms used in the statutes addressed in chapter 5.
58 In 1505, King D. Manoel I gave permission to the freed or ‘‘redeemed’’ black
women of the Brotherhood of Rosário in Lisbon to sell herbs and other goods
in the markets, alongside white women. In 1515 they were also allowed to work
as vendors in the plazas. Mulvey, ‘‘The Black Lay Brotherhoods of Colonial
Brazil,’’ 22. As Luciano Raposo has shown in O avesso da Memória, the ‘‘black
women of the tray’’ were a common category in the o≈cial documents of the
colonial administration; they sold a variety of items, including tarts, cakes,
sweets, honey, bread, bananas, tobacco, and beverages. See Luciano Figuei-
redo, O Avesso da Memória, 33–34, 41–42.
4 | urban life and brotherhoods in the city

1 Among the most popular and upscale churches here were the Santa Casa da
Misericórdia, the Brotherhood of São José, and the Third Order of Carmo.
2 The See included the religious functionaries that lived and worked with the
Cabido (bishop). Sé (See) refers to a bishop’s episcopal jurisdiction.
Notes to Chapter Four | 267
3 According to Monsenhor Pizarro, the bishopric encompassed all the area

occupied by Portuguese from Espírito Santo to Prata, and it was answerable to
the archbishop of Bahia. It was a creation of the papal bull Romani Pontificus
Pastoralis Solicitudo of 22 November 1676. That year also saw the commence-
ment in the See of the ‘‘Third Book of the Dead—Book and Notations of White
People Expired in this city . . .’’ (Livro 3\ dos Falecidos—Livro e cadernos das pessoas
brancas falecidas nesta cidade . . . ), whose title suggests that already the parish
records separated whites and blacks. Araújo, Memórias Históricas do Rio de Jan-
eiro, 2:217n27. I could not find the books from the eighteenth century in the
acmrj.
4 A group devoted to Carmo found space in the Chapel of Nossa Senhora do Ó in
1590 (and is still present in the city today). In 1607, Franciscans constructed a
convent on a nearby hill for the Carmelites; that hill was first called Morro do
Carmo, and later referred to as the Morro of Santo Antônio (Saint Anthony).
The Benedictines, present in the Chapel of Nossa Senhora do Ó in 1589, left
that Chapel to the Carmelites to found a convent just outside the city on a hill
that would be called Morro de São Bento. The churches of São Sebastião and
Santa Luzia date from the sixteenth century. In the seventeenth century the
churches of São José and Candelária arrived, as well as the chapels of Our Lady
of the Conceicão, Our Lady of Parto, and Our Lady of Bonsucesso.
5 Fazenda, ‘‘Antiqualhas e Memórias do Rio de Janeiro,’’ tomo 86, 348. If it were
not for the quick aside by Vieira Fazenda referring to Guinea blacks associated
with Saint Domingos, that connection might not have come to light, because I
did not find mention of it anywhere else.
6 Conflicts between brotherhoods and the local Catholic authorities were ana-
lyzed with admirable rigor by Carlos Boschi. An understanding of the tensions
between brotherhoods is best obtained through the period chroniclers, such as
Vieira Fazenda and Padre Perereca, who colorfully described the pugilistic
encounters, theft of ceremonial objects, and processions interrupted, along
with other incidents. See Boschi, Os Leigos e o Poder; and Fazenda, Antiqualhas e
Memórias do Rio de Janeiro, 5 volumes.
7 Sabotage and delays in the processions led to accords being signed between
groups, as well as to the mandatory signing of documents by leaders before the
procession began that signified their acceptance of the place given to each
brotherhood. Coaracy, O Rio de Janeiro no Século Dezessete, 85.
8 Schwartz, Segredos Internos, 210. The pomp and ostentation of each brotherhood
also served as a form of propaganda to attract new adherents. Russell-Wood,
Fidalgos e Filantropos, 157.
9 D. Francisco wrote the letter in 1709 cited in chapter 2 of this book.
10 On this polemic, see Silvia Hunold Lara, ‘‘Sob o Signo da Cor.’’
11 Leila Mezan Algranti, Honradas e Devotas.
12 According to Scarano, the congregations were to be of the ‘‘most esteemed
people . . . not being permitted the entrance, without prior investigation of a

white person born in Brazil, since the status of pure blood was required.’’ Julita
Scarano, Devoção e Escravidão. In Portugal, the Brotherhoods of the Santíssimo
Sacramento were perhaps the most numerous of all the lay associations, along-
side the Confrarias das Almas.
13 Holanda, ‘‘O Semeador e o Ladrilhador’’; Angel Rama, ‘‘La Ciudad Letrada.’’
14 It is unclear when the Brotherhood of the Rosário left São Benedito for the new
church. The building was inaugurated in 1725, but the transfer took place
before that date.
15 But the French were persecuting the Portuguese all along the route to India, on
land and at sea, at the time. They burnt Benguela in 1705, and sacked the
Islands of Príncipe (1706), São Tomé (1709), and Cape Verde (1712). Whatever
their immediate concern for gold in Rio, taken as a whole these actions were
part of larger disputes surrounding the War of Spanish Succession between
1702 and 1714. Fernando Novais, Portugal e Brasil na Crise do Antigo Sistema
Colonial, 26, 40.
16 Mello Barreto Filho and Hermeto Lima, História da Polícia do Rio de Janeiro, 65.
17 Fazenda, ‘‘Antiqualhas e Memórias do Rio de Janeiro,’’ tomo 86, 367–72.
Expressions referring to locations in the city as extra-muros (outside the walls)
and dentro dos muros (within the walls) are common in documents from the era
in which the protective wall was a matter of public awareness and debate.
18 In his pioneering work, Caio Boschi explored one aspect of this question—the
relations of lay worshippers with the church. Boschi, Os Leigos e o Poder.
19 See Boschi, Os Leigos e o Poder, 178; and João José Reis, A Morte é uma Festa, 49.
20 My view of sociability has been influenced by Roger Chartier and D. Roche,
‘‘Social (História),’’ 573.
21 See in particular the 1977 work by Hoonaert et al., História da Igreja no Brasil.
22 Boschi, Os Leigos e o Poder, 64–65.
23 Known as Onça (Panther), Monteiro was deposed in 1732 and died a year later.
24 Rupert, A Igreja no Brasil III, 3:44.
25 The governor helped finance the last stages of construction of the Church of
the Rosário, and brokered an arrangement for the Brotherhood of Our Lady of
the Conception of Pardo Men to congregate in the See (Saint Benedict), even
though this was against the wishes of the priests.
26 An act established by the judge of the captaincy, Dr. Agostinho Pacheco Teles,
prohibited any person, free or slave, from throwing ‘‘trash, rubbish, or filth’’
there, but the act was weakly enforced. João da Costa Ferreira, ‘‘A Cidade do Rio
de Janeiro e seu Termo.’’
27 There were many tense years of coexistence until 1808, when the See cathedral
was transferred to the Largo do Paço on the occasion of the transfer of the
Portuguese Crown to Rio de Janeiro.
28 Maurício, Templos Históricos do Rio de Janeiro, 109, 215.
29 If the Cabido priests directed a formidable degree of animosity toward the black
brotherhoods located in the See cathedral, the brotherhood of pardos also
based there (first dedicated only to Our Lady of the Conception) was not spared
the priests’ ire. They stayed at Castle Hill until 1729, when Governor Monteiro
deemed it proper that they should move to the Church of the Hospício; this was
a chapel belonging to the Third Order of São Francisco da Penitência, which
had on its grounds a cemetery for slaves. In 1734, that group united with the
congregation of the Boa Morte to form the Brotherhood of Nossa Senhora da
Conceição e Boa Morte dos Homens Pardos.
30 The last wills and testaments of some of these freed Mina slaves, listing their
material patrimony, are transcribed in the funerary books of the parishes,
which are today held in the acmrj.
31 Various wills include detailed descriptions of gold jewelry, while the pano de
costa is visible in many of the drawings from the period.
32 For more on the African goods traded in Rio de Janeiro, see Carreira, As
Companhias Pombalinas de Navegação, Comércio e Tráfico de Escravos entre a Costa
Africana e o Nordeste Brasileiro, 190–242.
33 ‘‘Diário Anônimo de uma Viagem à Costa d’África e as Índias Espanholas.’’ The
report was prepared for publication by Gilberto Ferrez.
34 acmrj, Livro de Óbitos e Testamentos, Freguesia da Sé, 1746–58, 3 July 1749, 135.
35 A. M. Rios Filho, ‘‘Evolução Urbana e Arquitetônica do Rio de Janeiro nos
Séculos XVI e XVII (1567–1699).’’ The work presents a series of photographs of
historic houses to illustrate the developments detailed by the author.
36 We can still try to apply the sources on houses to the parish records to get an
idea about the occupation of dwellings, patterns of commerce, and distribu-
tion of the slave population. Helpful data are available in ‘‘Memórias Públicas e
Econômicas da Cidade de São Sebastião do Rio de Janeiro para Uso de Vice-rei
Luiz de Vasconcellos por Observação Curiosa dos Anos de 1779 até 1789.’’
rihgb , book 47, pt. 1, 1884, 25–51.
37 Many of the female slaves living there might have worked as ambulatory
vendors, the so-called escravas de tabuleiro. See Figueiredo, O Avesso da Memória,
60–71.
38 Manolo Garcia Florentino, Em Costas Negras, 82–83.
39 The available obituary records are much less complete than the baptismal
records for early eighteenth-century Rio de Janeiro, although reading them
provides a useful portrait of the conditions and options for burial at the time.
Could it be that these circumstances were roughly the same as those studied by
Katia Mattoso and João Reis in Salvador? Mattoso, Ser Escravo no Brasil; Reis, A
Morte é uma Festa.
40 Russell-Wood, Fidalgos e Filantropos, 174.
41 Extract of a petition to the crown from the Brotherhood of São Domingos of the
Convent of São Francisco, in the city of Salvador, Bahia. ahu, Bahia, 1735,
transcribed by Mulvey, in ‘‘The Black Lay Brotherhoods of Colonial Brazil: A

History,’’ 197–99.
42 acmrj, Livro de Óbitos de Livres, Freguesia de Candelária, 1797–1809, registers
from March 1798, 5 June 1799, and 2 August 1799.
43 Chapter 6 of this book provides a case study of how an African kinship group
was organized in the Brotherhood of Santo Elesbão and Santa Efigênia. The
term parentes also appears in the nineteenth century, when Antônio Ferreira, a
freed slave of the Benguela nation, petitioned for authorization to perform
‘‘together with his kin [parentes] the customs and pastimes of that nation.’’ See
Martha Campos Abreu, ‘‘O Império do Divino’’; and Reis, A Morte é uma Festa,
55.
44 For a thoughtful analysis of the theme(s) of death that is useful to the present
discussion, see Philipp Ariès, O Homem Diante da Morte. The close relationship
between churches and sepulchral practice continued in Brazil, especially in
Rio, until 1850. That year, decree number 583 established the creation of public
cemeteries and prohibited burials in or next to churches. See Cláudia Rodri-
gues, ‘‘Lugares dos Mortos na Cidade dos Vivos’’ (esp. chap. 3).
45 Debret, Viagem Pitoresca e Histórica ao Brasil, tomo 2, 3:186. Judging by his
sketches, Debret seemed to have been intrigued by the variety of co≈ns he saw
on the streets, from the most splendid to the well worn and rough hewn. The
use of hammocks and inexpensive co≈ns are understood here as the inversion
of what Michel Vovelle had in mind with respect to ‘‘funerary art,’’ but they are
just as eloquent on a given people’s social conditions and death practices.
Chapter ‘‘Iconografia e História das Mentalidades.’’ In Vovelle, Ideologias e
Mentalidades.
46 According to Vieira Fazenda, the frequently expressed idea that the Church of
the Lampadosa contained the images of São Crispim and São Crispiniano was
a misconception. I encountered one 1667 testament from a black shoemaker
who was a member of both the Brotherhood of the Rosário and of São Crispim
and São Crispiniano.
47 See Padre Perereca’s comments to that e√ect in Luiz Gonçalves dos Santos,
Memórias para Servir à História do Reino do Brasil, 42.
48 In the nineteenth century, the Parish of the Santíssimo Sacramento absorbed
the old São Domingos Field and the nearby chapels built by black brother-
hoods during the previous century. Cláudia Rodrigues analyzes the records of
all 584 burials of Africans in the parish between 1812 and 1885 and found the
following distribution of interment at di√erent churches: 17.3 percent were
buried at the Church of São Domingos, 16.4 percent at Lampadosa, 12.5 per-
cent at Rosário, 10.1 percent at Santo Elesbão and Santa Efigênia, 9.8 percent at
Santíssimo Sacramento, 2.9 percent at Lord Bom Jesus do Calvário, 2.6 percent
at São Gonçalo Garcia, 2 percent at Our Lady of Boa Morte (Good Death), and
less than 1 percent at diverse locations. Rodrigues, Lugares dos Mortos, 220, table
21. The churches of Lampadosa and Santa Efigênia were constructed in the
1750s.
49 The first table (table 12) of obituaries contains 397 cases, while the second
table (table 13) contains 499; the higher number includes records that did not
specify the locale of burial. When obituaries recorded that the deceased was
buried without the sacraments, which was quite frequent, the cause of death
was usually an unidentifiable sudden event.
50 Sheila Castro Faria provocatively revisits this theme in light of important early
works by Manning and Klein. Faria, ‘‘A Colônia em Movimento,’’ 269–70.
51 Figueiredo, O Avesso da Memória, 164.
52 The number of baptisms is derived from adding the data from the first two
periods of baptisms in Sé Parish; see chapter 3, table 5.
53 Mulvey, ‘‘The Black Lay Brotherhoods of Colonial Brazil,’’ 198.
54 This leaves aside the use of parasols as royal symbols in African kingdoms,
notably the kingdom of Kongo. Detailed study is needed of the relations
between the cultural and political practices associated with African kings, and
those with the kings of African folias in colonial Brazil. Regrettably, this
fascinating topic is beyond the scope of this book.
55 Descriptions of the festivals are included under the headings ‘‘Divino’’ and
‘‘São Gonçalo’’ in Cascudo, Dicionário do Folclore Brasileiro, 356–57, 432–36.
56 Mulvey, ‘‘The Black Lay Brotherhoods of Colonial Brazil,’’ 104–5, 122.
57 Maurício, Templos Históricos do Rio de Janeiro, 112–13.
58 Mulvey, ‘‘The Black Lay Brotherhoods of Colonial Brazil,’’ 114–15.
59 About Lampadosa, see Vieira Fazenda, ‘‘Antiqualhas e Memórias do Rio de
Janeiro,’’ tomo 95, 149:123–27; and Maurício, Templos Históricos do Rio de Janeiro,
109–17. About the festivities of Império do Divino in the Church of Santana,
see Abreu, ‘‘O Império do Divino’’; and Henrique José do Carmo Neto, ‘‘Rec-
ordações e Aspectos do Culto de Sangt’Anna.’’
60 Fazenda, ‘‘Antiqualhas e Memórias do Rio de Janeiro,’’ tomo 95, 149:123–27.
61 Abreu, ‘‘O Império do Divino,’’ tomo 1, chap. 2, note 56. The relative tolerance
that characterized Portuguese administration in these contexts di√ers greatly
from the stance of the English in their African colonies, where anything as-
sumed to be related to witchcraft was discouraged or punished. The 1890 penal
code of Brazil distinguished between so-called black magic, which was pro-
hibited and punishable, and white magic, which was held to be beneficial,
particularly as a supplement to other cures in matters of physical and mental
health. For a comparison of the nineteenth-century penal systems in Brazil and
English Africa, see Maggie, Medo do Feitiço, 24–30.
62 Debret, Viagem Pitoresca e Histórica ao Brasil, tomo 2, 225.
63 The full Portuguese text goes on to greater lengths than the translated extract,
which focuses on the dances themselves:
‘‘Recebi o aviso de Vossa Excelência de 9 de Junho em que Sua Majestade
ordena dê o meu parecer a vista das Cartas do Santo Oficio e do Governador de

Pernambuco, do Santo Oficio vejo tartar de danças que ainda não sejam as mais
santas não as considero dignas de uma total reprovação; estas considero eu e
pela carta do Governador, vejo serem as mesmas aquelas que os pretos divid-
idas em nações e com instrumentos próprios de cada uma dançam e fazem
voltas como arlequins, e outros dançam com diversos movimentos do corpo,
que ainda que não sejam as mais inocentes são como os fandangos de Castela,
e fofas de Portugal, e os lunduns dos broncos e pardos daquele país; os bailes
que entendo serem uma total reprovação são aqueles que os pretos da Costa da
Mina fazem às escondidas, ou em casas ou roças com uma preta mestra com
altar dos ídolos adorando bodes vivos, e outros feitos de barro, untando seus
corpos com diversos óleos, sangue de galo, dando a comer bolos de milho
depois de diversas bençãos supersticiosas, fazendo crer os rústicos que na-
quelas unções de pão dão fortuna, fazem querer bem mulheres a homens, e
homens a mulheres, e chega tanto a credulidade de algumas pessoas, ainda
daquelas que pareciam não serem rústicos como frades e clérigos, que chega-
ram a vir presos a minha presença, em os cercos que mandava botar a estas
casas, que querendo-os desmaginar me foi preciso em as suas presenças lhes
fazer confessar o embuste aos pretos donos das casas; e depois remetê-los aos
seus prelados para que este os corrigissem como mereciam, e aos negros fazia
castigar com vigorosos açoites, e obrigava aos senhores que os vendessem para
fora. Estas são as duas castas de bailes que vi naquela capitania em o tempo que
a governei e me persuado que o Santo Oficio fala de uns e o Governador fala de
outros, pois não me posso persuadir que o Santo Oficio reprove uns, nem que o
Governador desculpe outros. Este é meu parecer, e Sua Majestade com mais
claras luzes resolverá o mais justo.’’
‘‘Deus Guarde a Vossa Excelência ms. ans. Lisboa 10 de Junho de 1780.
Conde de Povolide Ilmo. e Exmo. Senhor Martinho de Mello e Castro.’’ bep,
letter manuscript in ‘‘Correspondência da Corte, 1780–1781,’’ f01.23r–v.
64 The name Valongo was given to several places in Portugal. It seems to derive
from the phrase vale longo (long valley), which would be consistent with its
application in Rio to a place that was best reached through the curving plain
between two hills, Conceição and Livramento. The first recorded reference to
the name in Rio de Janeiro seems to be in 1701. I thank Maurício Abreu for this
information.
65 Mascarenhas, ‘‘Relatório.’’
66 Constituições Primeiras do Arcebispado da Bahia, Livro 4\, Tit. 19, 692.
67 ‘‘Várias irmandades de pretos com igrejinhas indignas e indecentes que nem
devem ter este nome como são a Irmandade das Mercês e São Domingos, São
Felipe, São Tiago, o Menino Jesus, Santa Efigênia e S. Elesbão, Nossa Senhora
da Lampadosa, São Mateus, outra de São Benedito em Santo Antônio, o Senhor
Jesus do Calix, Nossa Senhora de Belém e Santo Antônio da Mouraria, as quais
Notes to Chapter Five | 273
sendo vm servido ficarem anexas, e recolhidas a esta Igreja demolindo-se os

alpendres em que existem para cemitérios faria vm um grande serviço a Deus,
e grande aumento desta Igreja, e irmandades, pois as dispersas despesas que
fazem, reunidas, e incorporadas nela ficaria cessando a sua grande neces-
sidade para a conclusão da obra.’’ ahu, Petition from the Brotherhood of the
Rosário to the king, 27 July 1774. I thank Fernanda Bicalho for a copy of this
correspondence.
68 ahu, Response from Marquês de Lavradio to the king’s request for further
information on the matters raised by the petition of the Brothers of the Rosário
(cited above), dated 17 July 1775.
5 | constructing a religious norm

1 Even colonial legislation at the time was mostly preoccupied with devising
solutions to immediate problems, and not with the establishment of an inte-
grated juridical framework. See Salgado, Fiscais e Meirinhos.
2 For a pioneering work regarding this approach in the history of slavery in
Brazil, see João José Reis and Eduardo Silva, Negociação e Conflito. For more on
the brotherhoods in Brazilian colonial society, see Russell-Wood, The Black Man
in Slavery and Freedom in Colonial Brazil.
3 Norbert Elias, A Sociedade de Corte.
4 Russell-Wood, Fidalgos e Filantropos, 154.
5 John Bossy, A Cristandade no Ocidente, 1400–1700. The notion of charity among
family members or ethnic groups is taken up again in chapter 6.
6 The term incorporated was common in period ecclesiastical documentation and
carried a double meaning. It called for the group to be present as a whole, but
beyond that, the brothers should all be dressed in the brotherhood’s formal
attire and carrying all the flags, sta√s, and other objects that represent the
brotherhood’s overall identity as well as the di√erent roles of the individuals
within it. In festive processions, the icons of the protective saints were carried
along, but during funerals these were left at the church.
7 The seven spiritual acts: instruct the ignorant, give good counsel, punish
transgressors, console the sorrowful, forgive personal a√ronts, accept human
imperfections, pray for the living and the dead. The seven physical acts: liberate
captives and visit the imprisoned, care for the sick, clothe the naked, feed the
hungry, give drink to the thirsty, give shelter to travelers and the poor, bury the
dead. Russell-Wood, Fidalgos e Filantropos, 14–15.
8 Anderson José Machado de Oliveira, ‘‘Devoção e Caridade.’’
9 Principal documents relating to these organizations are as follows: Compro-
misso of the Brotherhood of Santo Elesbão and Santa Efigênia, copy located in
the archives of the brotherhood; proposed statutes for the Devotion of the
Almas do Purgatório, transcribed in the Mahi Manuscript; and proposed stat-
274 | Notes to Chapter Five
utes for the Devotion of Our Lady of Remédios, document located in the
Arquivo Histórico Ultramarino.
10 Before the completion of my doctoral dissertation, two other works about the
Church of Santo Elesbão and Santa Efigênia had been written (but still remain
unpublished). The first, Vânia Penha Lopes’s ‘‘À Venerável Irmandade de Santo
Elesbão e Santa Efigênia, ou uma Tentativa de Entendimento da Questão Étnica
no Brasil,’’ is a present-day study of the brotherhood devoted to the church’s
patron saints. The second, Oliveira’s ‘‘Devoção e Caridade,’’ o√ers a historical
account of the church in the nineteenth century. Together, these works pro-
vided a provocative starting point from which to formulate my own project.
11 The Tribunal of Conscience and Orders was transferred to Rio de Janeiro along
with the royal family in 1808 and was dissolved in 1822, when Brazil was
declared independent. Its documentation from these years (today held at the
National Archive) refers to four distinct brotherhoods of Santo Elesbão and
Santa Efigênia in Brazil, but does not specify a brotherhood of that name in Rio
de Janeiro. Nor, for that matter, does it mention the existence of any brother-
hood based in Rio’s Church of Santo Elesbão and Santa Efigênia. See Sérgio
Chahon, ‘‘Aos Pés do Altar e do Trono.’’
12 For more on how various types of cases were processed by the Tribunal of
Conscience and Orders, see Guilherme Pereira Neves, ‘‘E Receberá Mercê.’’
Although the author primarily addresses the nineteenth century, at times he
considers the eighteenth. And there seems to have been little change in the
evaluative methods or bureaucratic nature of these processes from the earlier
period to the later one.
13 aisese, doc. 1. My analysis throughout this chapter draws on the compro-
misso of the Brotherhood of Santo Elesbão and Santa Efigênia, included in the
covenenant book belonging to the archives of the Church of Santo Elesbão and
Santa Efigênia (aisese). This book contains fifteen other documents to which
I refer in sequential order as aisese docs. 1, 2, and so on.
14 Maurício, Templos Históricos do Rio de Janeiro, 215; Mulvey, ‘‘The Black Lay Broth-
erhoods of Colonial Brazil,’’ 113.
15 aisese, doc. 2.
16 aisese, doc. 3.
17 I could not find these two provisions, and it was not entirely clear what Mon-
senhor Pizarro was referring to. It is possible that the so-called 1747 provision
was actually the complete collection of documents sent to reformulate the
statutes in 1748. Pizarro’s second provision, dated 1754, might refer to the
inauguration of the Church of Santo Elesbão and Santa Efigênia that year.
18 aisese, doc. 5.
19 aisese, doc. 9.
20 aisese, doc. 16.
21 aisese, docs. 11 and 12.
22 Chap. 6 of the compromisso mentions that the brotherhood owned some

furniture and decorations housed in the church, and recommends against
loaning out items belonging to the group. This would suggest, among other
things, that several associations shared the same space in the church.
23 Constituições Primeiras do Arcebispado da Bahia, Livro 40. , LXIV, 882. Today it is a
common sight to see people, especially women, traversing the streets of Rio de
Janeiro begging for money with one hand and holding an image of their saint
in the other. It is common too among devotees of candomblé, who, upon
initiation, ask for donations for their ‘‘saints’’ or protecting deities.
24 Alfredo Mendes de Gouveia, ‘‘Relação dos Compromissos de Irmandades,
Confrarias e Misericórdias do Brasil.’’
25 Of course, even a vote based on the palpable transfer of fava beans was suscep-
tible to rigging. From this custom derives the modern Brazilian expression
‘‘são favas contadas’’ (the beans are already counted), used when it is assumed
that a dispute will secretly be resolved in a premeditated manner.
26 Peter Burke, A Fabricação do Rei, 13–25.
27 The ceremonial cloaks were roughly standard across the brotherhoods, but
their colors di√ered in accordance with the particular saint. The cloak of Santo
Elesbão is red, and that of Santa Efigênia violet. Each cloak also bore the
insignia of the brotherhood to which it pertained. A simpler white cotton cloak
could be worn by brothers outside festive or other ritual occasions.
28 A detailed description of this aspect of the festival, known as comilança, is to be
found in Campos, Procissões Tradicionais da Bahia.
29 On death and death rituals, see Ariès, O Homem Diante da Morte. For the Brazilian
case, see José de Souza Martins, A Morte e os Mortos na Sociedade Brasileira; and
Reis, A Morte é uma Festa.
30 Debret, Viagem Pitoresca e Histórica ao Brasil, tomo 2:184.
31 John Luccock, Notas sobre o Rio de Janeiro e Partes Meridionais do Brasil, 39.
32 I recall here the preoccupations of Guilherme Marechal with his own death.
Georges Duby, Guilherme Marechal ou o Melhor Cavaleiro do Mundo.
34 It was not possible to locate the habitations of the brothers in the eighteenth
century. I found many hints, but nothing definitive. In the brotherhood’s
books from the nineteenth century, available in their archive, these data are
more easily obtained. According to a survey by Anderson Oliveira, the 298
registers of members between 1843 and 1930 indicate that 69 lived on the same
street as the church and 49 on an adjacent street. Oliveira, ‘‘Devoção e Car-
idade,’’ 162. In the eighteenth century, the courier must have had farther to
walk. Even though the city was smaller, slaves were dispersed throughout, as
most likely were members of the brotherhood.
35 The cemetery was closed in 1850 when new laws mandated burial in public
cemeteries. Rodrigues, ‘‘Os Lugares dos Mortos na Cidade dos Vivos.’’
276 | Notes to Chapter Five
36 aisese, c.
38 Schwartz pointed out that colonial Brazil deserves to be called a ‘‘slave society’’
due not only to its wide reliance on forced captive labor, but also to the other
related social hierarchies that expanded and rigidified through and with slavery
there. He also explored the juridical distinctions created between free and
slave, and the hierarchies associated with slavery, race, and class. Schwartz,
Segredos Internos, 209–15.
40 The document contains these lines: ‘‘Acordaram que os irmãos desta confraria
são muito importunados de muitos escravos que ficam meio forros, e assim
escravas que lhe dêem para se acabarem de forrar, em tal caso a Confraria não
será obrigada a mais que a favorecer os ditos escravos a falar por eles, e não a
demandas, salvo alguma pessoa a que a Confraria tiver muita obrigação, e
quiserem tirar pelos irmãos algumas esmolas para isso, mas a custa da Con-
fraria querendo lhe fazer alguma esmola será até a quantia de quinhentos réis e
isto porque a Confraria não fique desfraldada.’’ Quoted ibid., 258–61.
41 From research underway at the acmrj, I can draw the still-tentative conclu-
sion that most (if not all) the marriages took place after freedom was achieved.
And although having children was prized in the abstract, I have not found
evidence of children for the majority of married brothers.
42 For sources, see note 9 of this chapter.
43 See Kátia Mattoso’s discussion of the Sociedade Protetora dos Desvalidos da
Bahia, for whom ethnic identity was part of the criteria involved in accessing
support for the purchase of freedom. Mattoso, Ser Escravo no Brasil, 151, 163.
44 The statutes say of worthy candidates for this position, ‘‘por seu zelo e de-
voção’’; they could be expected to o√er ‘‘esmola de grandeza avantajada.’’
46 A royal decree had mandated reductions in the esmolas paid within lay brother-
hoods, but this seems to have been ignored with respect to several positions
described in the statutes, including the king of the folia. aisese, doc. 11.
47 Among other groups, notably the Brotherhood of the Rosário do Mosteiro de
São Domingos shared this preoccupation with assisting poor and sick mem-
bers.
48 ‘‘É costume haver nas Irmandades principalmente na dos pretos, Irmãos re-
voltosos inimigos da paz,’’ the statute a≈rms, adding that such people make
themselves known by using ‘‘palavras descompostas.’’
49 ‘‘Só se cuida nesta [irmandade] que haja paz e quietação e não distúrbios.’’
50 Considering all the black brotherhoods of colonial Brazil, the number of
masses for dead brothers varied by era and di√ered from order to order. The
Brotherhood of Nossa Senhora do Rosário do Serro in Minas Gerais stipulated
six masses per brother (Scarano, Devoção e Escravidão, 53), while the Jêje Broth-
erhood of Bom Jesus dos Martírios da Vila de Cachoeira in Bahia held twenty-
five for any brother. Reis, A Morte é uma Festa, 207.
51 As chapter 3, table 8, of this book shows, only six Mina men married Angola
women during the period under study, while twelve Mina women married
Angola men.
52 ‘‘Que suposto no capítulo 10\ recuse não sejam admitidos pretos e pretas de
Angola, crioulos e mestiços e cabras. Contudo agora é contente em toda a
Irmandade se admitam por Irmãos todos estes recusados.’’
54 I refer to the notion of contrasting identity developed by Oliveira in Identidade,
Etnia e Estrutura Social.
55 The 1618 statutes of the Brotherhood of Misericórdia in Salvador established
seven conditions for admission; one of them was to be limpo de sangue (of clean
blood). According to Russell-Wood, the period expression sangue religioso (re-
ligious blood) derived from usage in Portugal, where it referred to Moors and
Jews. Russell-Wood, Fidalgos e Filantropos, 93, 108, 40.
56 It is possible to identify three brotherhoods in Bahia constructed around a base
of ethnic identity from which Angola blacks had been excluded: two were
groups of Jêje Africans, and the third was a group of Nagô-Yoruba Africans.
The Brotherhood of Senhor Bom Jesus das Necessidades e Redenção dos
Homens Pretos was established in the Chapel of the Corpo Santo in Salvador
(1752); the Brotherhood of Lord Bom Jesus, Lord of Martírios dos Homens
Pretos was in the Convent of Our Lady of Monte do Carmo, in Cachoeira (1765);
and the Brotherhood of Our Lady of Boa Morte was founded in the Church of
Barroquinha, in Salvador (date unknown). Mulvey, ‘‘The Black Lay Brother-
hoods of Colonial Brazil,’’ 296.
57 Other brotherhoods present variations as to the identity of o≈cials, as Mulvey
showed. For example, the board of the Brotherhood of Nossa Senhora do
Rosário of the Monastery of São Domingos had, among their six highest
elected o≈cers, five from Kongo and one from Bahia. On the other hand, the
Brotherhood of Nossa Senhora do Terço, which was founded by creoles at the
Church of the Rosário in Recife, established that certain groups from West
Africa were to compose the board, while the kings of the folia would be Kongo
or Angola. Mulvey, ‘‘The Black Lay Brotherhoods of Colonial Brazil,’’ 27, 86.
58 Oliveira, ‘‘Devoção e Caridade.’’
59 Mahi Manuscript, 22.
60 According to the baptism register for this man, first called Pedro Mina (and
later Pedro Costa), his owner was the chief judge of Rio de Janeiro, Manoel da
Costa Mimozo. acmrj, Livro de Batismo de Escravos, Freguesia da Sé, 1726–33, 38.
This book was lost at the acmrj and I found it after the Brazilian editon of the
book. For this reason it is not mentioned in the tables.
278 | Notes to Chapter Six
62 Ibid.
63 André Burguière, ‘‘A Antropologia Histórica,’’ 43.
64 Russell-Wood reported a similar phenomenon in the statutes of the Brother-
hood of Santa Casa de Misericórdia. The early version of that document, from
1516, seems stubbornly vague. But by 1618 the revised statutes are character-
ized by ‘‘particularity and the absence of ambiguity.’’ Russell-Wood, Fidalgos e
Filantropos, 75.
65 In the Brotherhood of Santa Casa de Misericórdia, the penalties for missing a
funeral were severe, but brothers still displayed resistance to accompanying
the funeral processions. Russell-Wood, Fidalgos e Filantropos.
66 Mahi Manuscript, 14. According to a survey performed for Viceroy D. Luiz de
Vasconcellos, Rio de Janeiro’s slave population in the 1780s was around 9,700
men and 7,100 women.
Vasconcellos, ‘‘Memórias Públicas Económicas,’’ 27. A note indicates that
this was the population within the walls of the city, which would have left a
substantial part of São Domingos Field and some of the new developments
south of the city proper. The drawings made by Francisco João Roscio in 1769
provide a good sense of this expansion.
68 For a more detailed look at this phenomenon, see Mulvey, ‘‘The Black Lay
Brotherhoods of Colonial Brazil,’’ especially appendix E, 289–303.
69 aisese, doc. 10.
70 aisese, docs. 11 and 12. In 1767 at least eleven more brotherhoods in Brazil
had their statutes confirmed in Portugal. Scarano, Devoção e Escravidão, 21–22.
71 Provision from D. Antônio do Desterro, dated 18 August 1767. The copy of this
approval notice archived in the Church of Santa Efigênia bears the incorrect date
1797; Desterro arrived in Rio de Janeiro in 1747 and died in 1773. aisese, doc. 15.
6 | conflict and ethnic identity

among mahi
1 aisese, compromisso of the Brotherhood of Santo Elesbão and Santa Efigênia,
chap. 5 of the section sanctioning creation of the folia. The most common
form of portable property in the wills was human property, or slaves. Women’s
wills might also include gold jewelry. Stationary property was far more rare,
and typically took the form of small houses that had been used either as
residences or rentals.
2 Schwartz, Burocracia e Sociedade no Brasil Colonial, xi–xii.
3 Coincidentally or no, the date Souza provided for his arrival, 1748, corresponds
with the year of the first reformulation of the compromisso of the Brotherhood
of Santo Elesbão and Santa Efigênia.
4 ‘‘Começaram os pretos a zingar as nações umas com as outras, buscando
preferências de maiorias.’’ Mahi Manuscript, 22.
Notes to Chapter Six | 279
5 ‘‘Ao que deu ocasião a que as nações Mahi, Agolin, Iano, Savalu saírem do jugo
Dagomé escandalizados e afrontados de alguns ditos picantes que os Dagomés
lhes diziam, procuraram fazer o seu Rei e com efeito o fizeram, na pessoa do
Capitão Ignacio Gonçalves do Monte no ano de 1762 por ser verdadeiro Mahi e
este foi o primeiro que fez termo e endireitou e aumentou esta congregação.’’
Mahi Manuscript, 22.
6 Ibid., 22.
7 Ibid., 21–22.
8 The term imediato had the connotations of a military position, and perhaps it
was applied to Souza because Monte seems to have been a colonel in a local
black regiment. I found references to Monte’s rank as such but no o≈cial
evidence of it.
9 These included Luiz Roiz Silva, Antonio da Costa Falcão, and Roza de Souza de
Andrade. Ibid., 13.
10 ‘‘Meu muito fiel e prezado amigo, a quem professei desde a minha infância, a
mais intima e cordial, amizade.’’ Mahi Manuscript, 2.
11 Ibid., 27–28. This is an additional use of the already described pano da costa.
12 Erizipela was a common but poorly diagnosed disease at the time. Symptoms
included high fever, generalized pain, and redness and swelling of the legs. It
seems to have been a bacterial infection due to germs attacking wounds that
were poorly treated, and (now as then) was associated with poor hygiene.
Slaves—even freed ones—often lived in deplorable conditions of hygiene,
sleeping in dirt-floor houses or yards. A belief among some in contemporary
Brazil is that the disease has no medicinal cure, and is linked to curses or black
magic, making prayers the most e√ective remedy.
13 According to Souza, the widow ‘‘fez por uma coroa na cabeça dizendo que era a
rainha.’’ The expression ‘‘fez por’’ indicates that she was crowned by others.
14 The compromisso of the Brotherhood of Santo Elesbão and Santa Efigênia still
prohibited creoles from entering; this was before chap. 10 was rewritten by
royal mandate. Thus Souza’s comments are especially critical of the morality of
the queen.
16 As Yvonne Maggie has shown, a combative posture against black magic is one
of the possible responses that characterize believers, not unbelievers. Maggie,
Medo do Feitiço, 22.
17 Souza wrote, ‘‘Tudo poderei fazer, visto a desculpa dos prudentes, sábios
exceto ser regente . . . Tenho teimado que não quero, porque conheço que a
viúva não faz gosto que eu seja sem o seu consentimento.’’ The polemical
situation between him and the queen is visible in various parts of the Mahi
Manuscript, e.g., 2, 7, 14, 38.
18 The text reads that women may serve as judges ‘‘por razão de seu número e por
contribuírem com suas esmolas e não a mais servem.’’ Ibid., 14, 39.
19 As with the case of the unreliable chaplain mentioned in chap. 32 of the

compromisso of the Brotherhood of Santo Elesbão and Santa Efigênia, a
serious impasse within the order could be resolved by resorting to external
authority.
21 Ibid., 23. This language recalls the hierarchy of posts in the Brotherhood of the
Santa Casa de Misericórdia, which were divided among irmãos de maior condição
(better-o√ brothers), which in Portugal would all have been nobles; and irmãos
de menor condição (poorer brothers), also called mechanics or mechanical work-
ers. Russell-Wood, Fidalgos e Filantropos, 15. The e√orts by the maiores to
protect their positions of prestige is reminiscent of the French nobles de-
scribed by Saint-Simon. Elias, A Sociedade de Corte.
23 Adding to the complexity of this organization, there were some titles and
positions awarded on the occasion of Souza’s election, according to the Manu-
script—but those names do not match up with the list that the Manuscript
includes soon afterward. It should also be noted that there may be errors in
orthography in the transcription of non-Portuguese words.
24 By custom, the king or master should never walk alone; at least two of his
fellows should always be present. But this was another tradition which Souza
chose not to uphold consistently. On one occasion, a group of brothers insisted
on accompanying him, stating ‘‘we are here to go with you because it is our
duty’’ (estamos prontos para o acompanhar pois temos de obrigação), but Souza dis-
missed them: ‘‘Thank you for such an honor, but the secretary is enough
company’’ (Obrigado a Vossas Mercês por tanta honra . . . basta que venha o secretário).
25 ‘‘Desde o princípio desta terra em que entraram a conduzir os pretos de África
que vem da Costa da Mina e de Angola, e pela desumanidade de alguns
senhores que os compravam, todas as vezes que adoeciam de moléstias in-
curáveis e envelheciam, os deitavam fora para morrer de fome e frio nus por
estas praias sem ter quem os mandasse enterrar. Se a Santa Casa da Misericór-
dia os não mandasse buscar para os enterrar com aquele zelo e caridades que
costuma, aí ficariam os cadáveres com o seu invalidez. E por esta razão intro-
duziram os pretos entre si a fazerem este adjunto ou Corporação a fim de
fazerem bem aos seus nacionais, a saber, que a Nação que morrer seus par-
entes tirar esmola para es sepultar e mandar-lhe dizer missas por sua alma.’’
Mahi Manuscript, 20–21.
26 It was common for documents produced by black lay brotherhoods to include
language about the su√ering of slaves as the basic justification for the creation
of brotherhoods. Of course, such su√ering did exist, in multitudinous forms.
What is notable is the rhetorical utilization of that su√ering by brotherhoods in
the argument for recognition and benefits.
27 Chap. 27 of the brotherhood’s statute reads as follows: ‘‘Aos irmãos que assim
forem eleitos serão obrigados a tirar esmolas pelos cantos das ruas onde
melhor lhe estiver aos domingos e dias santos como se costuma nas mais
Irmandades por estilo antigo . . . Recusando estes tais Irmãos assim nomeados
fazê-los serão admoestados.’’ See also the Mahi Manuscript, 22.
28 It should be clarified that the playing of drums in general in this context was not
considered a heathen practice. Brotherhoods of white men took to the streets in
public processions to collect esmolas accompanied by drumming and wind in-
struments, always played by blacks. The di√erence seems to lie in the type of
instruments used by black versus white brotherhoods, the rhythms played on
them (their nature and origin), and the dances that went along with them. Many
such small distinctions are clear if one compares the available drawings and
paintings of white brotherhoods with those of black brotherhoods.
29 Mahi Manuscript, 34. Luciano Figueiredo has shown that in Minas Gerais,
women were always the minority in lay brotherhoods, due largely to discrimi-
nation. Based on his research he hypothesizes that batuques (informal recre-
ational gatherings with music and dance) were more important spaces of
sociability and power for them. Figueiredo, O Avesso da Memória, 164, 171–81.
30 Chap. 12 of the statutes of the Fraternity of Our Lady of Remedies includes the
following line: ‘‘Haverá um cofre com três chaves, terá a Regente uma, o
Secretário outra, e o Procurador outra, mas nunca o cofre se abrirá para coisa
alguma, sem estarem presentes todos três pelos inconvenientes, que do con-
trário se podem seguir.’’ ahu /cu-cód 1300, Estatuto da Confraria de Nossa
Senhora dos Remédios, chaps. 12 and 13.
31 Chap. 13: ‘‘Mas porque atualmente está . . . de tesoureira das esmolas uma
senhora que o fora do Regente passado, hoje falecido, por se ter valido da
posse do cofre recorrendo à Justiça para ser conservada naquela mesma posse,
como se a administração de semelhantes bens fosse compatível com aquele
sexo, ou como, se esta poderá passar por morte do marido à mulher, como
herança: Por evitar semelhante abuso, em conseqüência da Graça Régia supli-
cada será tirado o cofre do poder da depositária.’’ Ibid.
32 Despite its fractious origins, the Fraternity of Our Lady of Remédios existed
within the Brotherhood of Santo Elesbão and Santa Efigênia for many years.
Today, elder members of the brotherhood recall that the Mahi were associated
with mutiny and witchcraft, and prefer not to speak of them.
33 Mahi Manuscript, 30–36.
34 aisese, Compromisso of the Brotherhood of Santo Elesbão and Santa Efigênia,
chap. 1 of the Folia.
35 The kings and their courts were often referred to as adjuntos (adjuncts) at the
time, underscoring their informality relative to the emperor.
36 Cordeiro: ‘‘Não se enfade com as minhas importunações porque a maior
paixão que tenho, é de não ver os nossos nacionais todos católicos fazendo
serviços a Deus que é o fim para a que fomos nascidos.’’
Souza: ‘‘Algum dia nos fará Deus essa vontade, porque ele sabe muito bem o
que faz, melhor do que imaginamos; . . . todas as maravilhas que temos
recebido e recebemos da sua Onipotência e liberal mão, desde o princípio
do Mundo sempre foram devagar. Pois virá tempo, que Vossa mercê veja
cumprido este seu gosto.’’
Mahi Manuscript, 45 (65). This exchange is reminiscent of the Letter of Obe-

dience sent by the heir to the throne of Kongo, D. Afonso I, to the pope in 1512
(cited in chapter 1 of this book).
37 aisese, doc. 11.
38 Provision dated 11 March 1767, aisese, doc. 11. The statutes of the Fraternity
of Nossa Senhora dos Remédios specifies that members should be ‘‘Mina
blacks,’’ but does not overtly restrict other nations.
39 ‘‘Os de Angola tem por costume tomarem da tumba da Santa Casa da Mis-
ericórdia os cadáveres de seus parentes para os porem nas portas das Igrejas
com cantigas gentílicas, e supersticiosas tirando esmola dos fiéis para os
enterrarem, o que é constante nesta cidade, e por esta razão, os senhores
brancos, entenderam que todos os pretos usavam do mesmo, quer que seja
Mina, ou de Angola, e essa é a razão porque me eximo de reger e proteger os
meus parentes.’’ Mahi Manuscript, 7.
40 Ibid., 21, 44.
41 On the question of sources for the study of deviant behavior in the colonial
period, see Ronaldo Vainfas, ‘‘Moralidades Brasílicas’’; and Mott, ‘‘Cotidiano e
Vivência Religiosa.’’
42 It is worth mentioning the case of police repression in the village of Cachoeira,
Bahia, which as João Reis showed, represented a strategy by o≈cers to guaran-
tee good relations with ecclesiastical and other authorities. Reis, ‘‘Magia Jêje na
Bahia.’’
43 On this movement, and especially the autos-da-fé that incorporated city resi-
dents, see Azevedo, ‘‘Judaísmo no Brasil.’’
44 Ecclesiastical regulations tended to dwell on the extreme cases, involving ex-
communication. But the Constituições Primeiras also acknowledges that smaller
cases of individual infractions would be responded to with attempts to instruct
in the Christian faith—even in heathen languages, if need be (Saint Paul and his
epistle to the Corinthians are cited). The second book of the Constituições Pri-
meiras portrays a continuum: there are many ‘‘nations and diverse tongues, that
range from heathenism to this State’’ of spiritual enlightenment; and it was
incumbent on religious authorities to make every e√ort to bring the light of
God to the blind. In title 1 of the fifth book, persecution is discussed in the
context of ‘‘heresy and Judaism.’’
45 The practice of animal sacrifice still occurs in Brazil today (notably in candomblé
and umbanda), influenced by traditional culture as well as more recent contact
with African religions.
46 Pereira, Nuno Marques. Compêndio Narrativo do Peregrino da América (1728); Reis,

‘‘Magia Jêje na Bahia’’; Mott, ‘‘Acotundá’’; State Library of Pernambuco, ‘‘Cor-
respondência a Corte (1780–1781).’’
47 ‘‘Que todos as pessoas que quiseram entrar nesse adjunto ou congregação
(exceto pretos de Angola) sejam examinados pelo secretário deste adjunto e
pelo aggau que é o Procurador Geral. Vejam que não sejam, pretos ou pretas,
que usem de abusos e gentilismos ou superstição. E que em achando, ou tendo
notícias que usam, não os podem receber.’’ Mahi Manuscript, 31.
48 For another perspective on this issue, see Vainfas, A Heresia dos Índios.
49 ‘‘É o espaço do globo terrestre compreendido, entre os dois circulos paralelos
entre si . . . Este espaço é como uma banda ou faixa, que contém muitos climas
que os geógrafos dividem em cinco zonas ou partes, considerando segundo os
diferentes graus de frio ou de calor, estas zonas são a tórrida, duas temperadas
e duas frias ou glaciais.’’ Mahi Manuscript, 49.
50 ‘‘Já no tempo do Reinado do Senhor Dom Afonso V fidelíssimo monarca de
Portugal, havia comércio nas costas de África em Guiné de marfim e o ouro
como adiante se mostrará, sendo autor destes descobrimentos o Senhor In-
fante Dom Henriques, filho 3\ de El Rei Dom João, o primeiro, também de
Portugal, que com o desejo que tinha da conversão dos infiéis e propagação da
Santa Fé católica e do cresentamento da coroa deste Reino, e como bom
geógrafo e matemático, que era alcançou tanto desta ciência que mediante a
sua profunda erudição mostrou ao Mundo que havia Antípodas, e que, a Zona
Tórrida era habitada, cousas naqueles tempos ignorada de todos os matemáti-
cos e cosmógrafos. Contra os que trazem autoridades de Santo Agostinho e
Latâncio Firmino que nega em muitos lugares haver Antípodas que são os
habitadores das terras que o Infante queria descobrir. E informando-se o dito
Infante dos Mouros de Ceuta quando lá esteve, veio a ter notícias dos desertos,
de África a que eles chamam Caharâ e dos povos que eles chamam azenegues,
que confinam com o pretos Ialof aonde se começa a região que os Mouros,
chamam Guinacolo, e em português quer dizer Guiné, tomada o seu distintivo
da cidade Genna, que pelo muito bem ouro que tem é celebrado o seu Com-
ércio; situada não muito longe do mar daquelas partes mui remotas em África
aos Reinos de Fez a Marrocos.’’ Mahi Manuscript, 49–50.
51 Souza went on to describe the size and ‘‘fury’’ of the Volta River, which he
called Voltas: ‘‘Este rio é muito largo na entrada mas corre com tanta força que
se conhece a sua corrente estando 3 léguas do mar, trás tantas árvores de
dentro do sertão arrancadas que detendo-se e embarcando-se umas com as
outras; causam na boca do rio grandes ariciros, de sorte que senão pode passar
em uma canoa, mas que duas vezes no ano que é ordinariamente desde o mês
de abril até o de novembro, mas deste mês por diante em que começam as
chuvas cresce muito o rio, e corre com muita fúria, quem partir da Mina para o
Rio das Voltas ponha-se 3 ou 4 léguas ao mar e faça o caminho de les nordeste,
e irá dar na aldeia deste rio e haverá na derrota quarenta e seis léguas pouco
mais ou menos.’’
52 ‘‘Espertou-me Vossa Mercê agora as potências desta alma, com a lembrança da
morte, que é o último fim em que havemos parar, e se não veja o que diz São
Bazílio bispo da Capadócia pelo filósofo gentio Eubolo, qual era a definição da
filosofia: respondeu que a primeira era o pensamento da morte e ficou tão
convencido o filósofo, com esta resposta que logo se converteu à fé de Jesus
Cristo Senhor Nosso, e é muito para considerar fazer uma só palavra tanto
abalo em um gentio, que farei eu miserável pecador que desde minha infância
conheço a Sua Santíssima fé e sou cristão pela graça de Deus. E não tendo eu
merecimentos nem de meu pai nem de minha mãe.’’ Mahi Manuscript, 57.
53 Ibid., 31.
54 This was the only area in which punishments were clearly prescribed in the
statutes. They ranged from praying on bended knee to Our Lady, to carrying a
large rock back and forth, to (upon the third o√ense) expulsion from the
brotherhood as an amotinador (agitator). aisese, Compromisso of the Brother-
hood of Santo Elesbão and Santa Efigênia, chap. 31.
55 Masses were very expensive. It is not uncommon to find masses included in
wills prepared by members of lay brotherhoods; in such cases, the person
writing the will would commit to having the masses paid for out of his or her
private funds upon death.
57 Ibid.
58 ‘‘É comum, quando dois pretos se encontram a serviço na rua, o súdito saudar
respeitosamente o soberano de sua casta, beijar-lhe a mão e pedir-lhe a bên-
ção. Dedicado, confiando nos conhecimentos de seu rei consulta-o nas cir-
cunstâncias difíceis. Quanto aos escravos nobres, graças à sua posição, con-
seguem de seus súditos os meios suficientes para comprar a própria liberdade;
e desde então empregam escrupulosamente toda a sua atividade no reembolso
da dívida sagrada.’’ Debret, Viagem Pitoresca e Histórica ao Brasil, 185.
59 Along with the purchase of freedom came a certificate, called a ‘‘letter of
enfranchisement,’’ that described in florid prose the meanings of freedom. The
letter presented to Feliciana Antonia do Desterro when she bought her freedom
from Bishop D. Antônio do Desterro was composed well within the conven-
tions of the era: ‘‘You can now go wherever you well choose, as master of
yourself, as if you had been born free and autonomous from the womb of your
mother,’’ and so forth. (It goes on ‘‘ . . . Lhe damos de hoje e para sempre
liberdade para que fique sendo forra, e vá para onde muito bem quiser, como
senhora, que ficou sendo de sí, como se livre e liberta nascesse do ventre da sua
mãe.’’) acmrj, Livro das Portarias e Ordens Episcopais, 1779–1830, bk. 3, fol. 43v.
60 The writers of the Constituições Primeiras explained that they saw no reason to
interfere, even when there was dancing on church grounds, as long as the
folias were ‘‘honest’’ and ‘‘decent.’’ Constituições Primeiras do Arcebispado da Bahia,

Livro 4, XXX, 742.
61 ‘‘Por quanto vemos que a experiência nos tem mostrado, que um estado de
folias, nas irmandades pretas [seria] de muita utilidade assim de exercitar os
ânimos dos pretos, como para acudirem de novo muitos de fora, assentarem
pé na Congregação, a fim de os ir atraindo com aquela suavidade, para os por
prontos para as caridades e tudo quanto for de serviço de Deus, nosso Senhor.’’
62 ‘‘Queremos que no dia de Nossa Senhora do Rosário, haja um Estado de Folias
desta nação Mahi, que acompanharão ao rei de Nossa Senhora do Rosário
sendo da Costa da Mina e não o sendo, o não acompanharão, somente se
permita as suas saídas, para o palácio do Ilustríssimo Excelentíssimo Senhor
Vice-Rei deste Estado, e depois de brincarem, recolher-se cada um para sua
casa, com toda quietação, e sossego, que se requer em semelhantes funções.’’
Ibid., 35.
63 A provocative analysis of deviance in this context comes from Erving Go√-
man’s exploration of ‘‘intergroup deviances’’ in Erving Go√man, Estigma, esp.
chaps. 4 and 5.
64 See the discussion of negotiation and conflict in Reis and Silva, Negociação e
Conflito.
65 The five were Boaventura Fernandes Braga (6, 12), Marçal Soares (11), Antônio
da Costa Falcão (4), Luiz Rodrigues Silva (7), and Alexandre de Carvalho (10).
The numbers in parentheses refer to the roster of ranks in the section ‘‘Who
Should Have the Keys to the Safe?’’ of this chapter.
66 These two were Luiz Rodrigues Silva and Antônio da Costa Falcão.
67 ‘‘Toda a pessoa que quiser entrar para a Confraria dará de entrada seiscentos e
quarenta de annual . . . , e apresentando-se ao Irmão Procurador ele o aceitará,
e fará assinar nos Livros dos Irmãos. Porém se o novo Irmão notoriamente
constar que é Preto de péssimos costumes, vicioso, infame, ou que usa de
feitiçarias e superstições de nenhuma sorte o aceitará o Procurador, ficando . . .
excluído de se queixar à Mesa, a qual informando-se também de sua conduta,
ou o excluíra ou o aprovará, conforme o que lhe constar pela informação
tomada.’’ ahu /cu-cód. 1300, Estatuto da Confraria de Nossa Senhora do
Remédios. fol. 7v.
68 Ibid., chap. 10.
69 ‘‘Com botica, enfermeiro, comida e até mortalha.’’ Ibid., 5.
70 ‘‘Como apresente Irmandade se dedique por seu principal Instituto ao exercitar
atos de caridade.’’ Ibid., chap. 14.
71 Ibid., chaps. 9 and 23.
72 Here I depart somewhat from the very useful framework of Reis and Silva in
Negociação e Conflito, which is concerned with negotiations of the exercise of
power in daily life. I argue (particularly in chapter 5 of this book) that such
negotiations are also part of, and expressed in, the institutional sphere of rules
and hierarchies
73 Estatuto da Confraria de Nossa Senhora do Remédios. Fol. 10v, 11; Estatuto da
Confraria de Nossa Senhora do Remédios. Chap. 13: ‘‘But because presently
the Treasurer of Esmolas is a lady who was the wife of the previous leader, now
deceased, who used the secular courts to maintain herself in that position, as if
the administration of such things was compatible with that sex, or as if such
things could pass from husband to wife, as an inheritance. To avoid such
abuse, Her Royal Grace is begged to remove the safe from the person who
holds it for its right return to the house of the present leader, or another
Brother of the board.’’
’’Mas porque atualmente estão . . . de Tesoureira das Esmolas uma senhora
que o for a do Regente passado, hoje falecido, por se ter valido da posse do
cofre recorrendo a Justiça para ser conservada naquela mesma posse, como
se a administração de semelhantes bens fosse compatível com aquele sexo,
ou como, se esta poderá passar por morte do marido à mulher, como
herança: Por evitar semelhante abuso, em conseqüência da Graça Régia
suplicada será tirado o cofre do poder da depositária onde se acha para a
casa do Regente atual, ou de outro Irmão da Mesa.’’
74 Estatuto da Confraria de Nossa Senhora do Remédios. Fols. 10v, 15, chap. 24:
‘‘The Brotherhood of Mercy in this city has won a privilege, that does not have
legal standing, to tax every other brotherhood a sum of four thousand reis each
time that they appear on the street, whether for a festive procession or a
funeral, in fulfillment of their own statutes. This is a severe and unjust burden
to pay. We ask that, if the present statutes are granted royal permission, our
Fraternity will not have to pay the Brotherhood of Mercy any penalty what-
soever in order to perform the functions prescribed in these statutes.’’ (A
Irmandade da Misericórdia desta Cidade tem advogado o privilégio, que não
tem para Lei alguma ou Graça Régia, de multar todas as mais confrarias em
quarto mil réis, cada vez que elas saem por conseqüência dos seus mesmos
Estatutos a exercitar qualquer ato, ou seja de Procissão, ou de enterro, Quando
. . . por gravames pecuniários, é um Direito inerente à Pessoa dos Príncipes
Soberanos, e por isso em virtude da Aprovação Régia dos presentes Estatutos
pretende a Confraria ficar isenta de contribuir à da Misericórdia multa alguma
por exercitar todas e quaisquer funções prescritas neste Compromisso.)
75 The title of this section is ‘‘Second dialogue, in which are o√ered details
regarding the founding of the grand fort of São Jorge of the Mina Coast,
constructed on the African coast of Guinea, and its ports; and of the Kingdom
of Benin; and other curious facts, by Francisco Alves de Souza, black, native of
the Kingdom of Mahi—one of the most excellent and powerful of the Mina
Coast.’’ (Diálogo segundo em que se dá notícias da fundação da grandiosa
Notes to Postscript | 287
fortaleza de São Jorge da Costa da Mina edificada nas costas de África em

Guiné e dos seus portos e o reino de Benin, e outras notícias curiosas, por
Francisco Alves de Sousa, preto e natural do reino de Mahi um dos mais
excelentes e potentados daquela oriunda costa da Mina.) Mahi Manuscript, 47.
76 Ibid., 60.
77 ‘‘Já disse acima que o Rei que presentamente governa se chamava a Dâlicâ, e
estes quando morrem, lhe fazem os fidalgos da sua Corte, a que eles chamam
Ômon, um sacrifício matando 16 mais escravos, e partes dos mesmos fidalgos
e príncipes o acompanham à sepultura aonde se enterram com o cadaver do Rei
defunto, com grande número de seus trastes e vestidos, matando muita gente,
e com elas o sepultam dizendo que é para ele se servir no outro Mundo. E cousa
de sete dias lhes fazem outro sacrifício, matando tantos escravos a que cha-
mam Ovem, dançando em cima do sepulcro do dito Rei a toques de tambores,
saltando ao redor dele fazendo muitas festas e visagens. E dizem que sendo o
Deus por natureza bom não necessitava de sacrifícios porem [o fazem] ao
diabo, para o aplacar adorando ao mesmo tempo a Ídolos.’’ Ibid., 60–61.
78 Ibid., 28.
79 National Archive, códice 334 (1833–40), ‘‘Correspondência reservada recebida
pela Repartição de Polícia.’’ Apud Abreu, O Império do Divino, bk. 1, n. 58, 220.
postscript
1 Adalgasia Arantes Campos et al., ‘‘O Banco de Dados Relativo ao Acervo da
Freguesia da Nossa Senhora do Pilar do Ouro Preto,’’ 24.
2 Lada probably referred to Allada, an important kingdom on the coast. Docu-
mentation from the seventeenth century in Brazil called the natives of this place
Arda.
Santos dos Séculos XVIII a XIX, 669–75.
4 According to Robin Law, the Mahi people were located north of Dahomey; they
derived from a fusion of Gbe speakers who came from the south and Yoruba
speakers from the east. Law, The Slave Coast of West Africa, 1550–1750, 19, 23–26.
There is not a discrete place called Mahi, and the territory of the Mahi people
included specific locales that were associated primarily with the Gbe language
(Savalu) and the Yoruba language (Dassa). For more on the Mahi language, see
Gbéto, Le Maxi du Centre-Bénin et Centre-Togo.
5 Savalu was a location within the Mahi territory. See Law, The Slave Coast of West
Africa, 1550–1750, 19.
6 Chamba was a place within modern Togo, but the term was often used for Gur
speakers as a group. Law, The Slave Coast of West Africa, 1550–1750, 189.
7 Jaquem, also called Jakin and Jeken, was one of the slave ports in the Bay of
Benin.
288 | Notes to Postscript
8 Earlier, I suggested that perhaps the slaves called Cobu (or Cabu) referred to
the Kaabu of Senegambia, but present research indicates that these peoples
may have come from the north of modern Benin.
9 This is a project in collaboration with Mauricio Abreu, with support from the
Fundação de Amparo a Pesquisa do Estado do Rio de Janeiro (faperj), 2007–
9.
10 The Cobu have yet to receive significant attention from scholars. Work has
begun on the Coura, however, with the completion of an important thesis on
the presence of Couras in the Brotherhood of Nossa Senhora do Rosário in the
town of Mariana, Minas Gerais, in the second half of the eighteenth century.
Pinheiro, ‘‘Confrades do Rosário.’’
11 With a perspective of identity construction that is quite close to what I have
called provenience groups, the historian Paul Lovejoy has suggested that the
forced migration of millions of Africans to the Americas might have resulted in
more inclusive ethnic identities. New forms of solidarity may have emerged in
the New World between groups that had previously been separated by social,
cultural, or geographic distance. Lovejoy, ‘‘Identifying Enslaved Africans in the
African Diaspora.’’
12 Law, ‘‘Ethnicity and the Slave Trade.’’
13 On the Mahi, see J. A. M. A. R. Bergé, ‘‘Étude sur le Pays Mahi’’; Robert Corne-
vin, Histoire du Dahomey avec 10 cartes, 1 croquis et 35 photographies; I. A. Akinjogbin,
Dahomey and Its Neighbours, 1708–1818; Jessie Gaston Mulira, A History of the Mahi
Peoples from 1774–1920; Law, The Slave Coast of West Africa, 1550–1750; and Edna
G. Bay, Wives of the Leopard.
14 ‘‘E se seguiu pretender o mesmo Cabo que se lhe entregassem uns negros
Couranos inimigos do Rey Daumê, que se dizia estarem na dita Fortaleza . . .’’
The Cabo here was an agau (a general of the Dahomean army). Verger, Fluxo e
Refluxo do Tráfico de Escravos entre o Golfo do Benin e a Bahia de Todos os Santos dos
séculos XVIII a XIX, 204–9.
15 Moacir Rodrigo de Castro Maia, ‘‘Quem Tem Padrinho Não Morre Pagão.’’
16 Database of the Parish of Pilar, cited by Patrícia Porto de Oliveira in ‘‘Batismo de
Escravos Adultos e o Parentesco Espiritual nas Minas Setecentistas,’’ 11.
17 ‘‘Rosto coartado à moda de sua terra, era baixa e refeita de corpo.’’ Coartado is
an archaic Portuguese word that meant ‘‘cut or divided in parts,’’ but was also
used for simply ‘‘cut’’ or ‘‘scarified.’’
18 I have yet to find an indication from the documents of how, when, or why Victoria
wound up in Rio de Janeiro, or of how she obtained the funds to buy her letter of
manumission. Her marriage record listed (or should have listed) all of the names
she had used up to that moment: at baptism, Victoria Courana, slave of Domingos
Correa Campos; upon buying her freedom, Victoria Correa Campos. Victoria
does not appear in the list of freed black and parda women meticulously studied by
Sheila de Castro Faria in Sinhás Pretas, Damas Mercadoras.
19 acmrj, Livro de Batismo de Escravos da Freguesia de Nossa Senhora da Candelária,

1745–74, from where the record dated 22 September 1742 was reproduced in
the Habilitações Matrimoniais: Ignacio Monte e Victoria Correa, doc. 22119,
caixa 1648.
20 Even more curious than the uncertain identity of Antônio Gonçalves and An-
tônio Gonçalves da Costa is Ignacio’s assertion, thirty years later in his will,
that he himself paid his master to purchase his freedom. acmrj, Livro de Óbitos
e Testamentos da Freguesia da Sé, 1776–84, fols. 442v–44.
21 This price seems high, and yet it corresponds to the price paid in 1753 for
another slave who was trained as a barber. This slave, Antônio, was freed by his
wife, Rita da Silva, who delivered 256,000 réis to his owner with the promise to
pay the remaining 94,000 réis in monthly installments of 4,000 réis. For his
part, Ignacio had his freedom purchased with a combination of paper currency
and gold and silver coins. Antônio’s manumission letter was registered in the
1\ Ofício de Notas do Rio de Janeiro, Livro de Notas 123, p. 130 (12 January 1753).
Ignacio’s was registered in the 2\ Ofício de Notas do Rio de Janeiro, Livro de
Notas 76, fol. 17v (12 January 1757). Both Ignacio and Antônio appear in the
database of freed slaves compiled by Sheila de Castro Faria, whom I thank for
the information.
22 acmrj, Habilitações Matrimoniais: Ignacio Monte e Victoria Correa, doc.
22119, caixa 1648.
23 acmrj, Livro de Óbitos e Testamentos da Freguesia da Sé, 1776–84, will of Ignacio
Gonçalves do Monte, fols. 442v–444.
24 Patrício was used in his will to convey the same meanings as parente, a term that
became more widespread in the nineteenth century. The notion of kin that both
words connote is one of social group rather than family ties per se. This is
particularly clear in the later documentation in Rio, as well as in Bahia; see João
José Reis, A Morte É uma Festa, 55.
25 In accordance with ecclesiastical norms, Ignacio Monte’s 1763 will was an-
nexed to his death record. acmrj, Livro de Óbitos e Testamentos da Freguesia da Sé,
1776–84, will of Ignacio Gonçalves do Monte, fols. 442v–44.
26 Based on the names, it may be that José dos Santos Martins and Manoel dos
Santos Martins—a freed Mina slave and a partner in the trio of freed brothers
who bought the land for the Church of Santo Elesbão and Santa Efigênia—had
both been owned by the same master.
27 Ignacio Monte was bedridden for a period of days before death; seemingly, he
had time to rewrite his will. The stakes were even higher in 1783 than in 1763,
not least because it was clear that his health was failing. It is curious that the
parish death registry shows that he died on 27 December, but an o≈cial
declaration of the Mahi Congregation specifies that it was the twenty-fifth.
Given this, and the lack of a new will, one suspects that the death had been kept
secret from the church authorities (and even from the Brotherhood of Santo
Elesbão and Santa Efigênia?) for a couple of days in order to get a handle on the
ensuing conflicts regarding the future of the Mahi Congregation.
28 Six slaves were men; of these, two were working barbers and one was an
apprentice. The tonsorial implements were evaluated at 8,000 réis and in-
cluded the following: three used razor sharpeners of Bahian stone, mounted
on rollers (900 réis each); three shaving basins (500 réis each); two other worn
basins (250 réis each); one small anvil and iron hammer (800 réis); twelve well-
used shaving razors (60 réis apiece); two sets of tongs for pulling teeth, and a
small pincer (1,400 réis); one wooden razor sharpener (400 réis); and two
leather razor cases (160 réis). an, Inventário de José dos Santos Martins, doc.
7129, caixa 628, galleria A, Juízo dos Orfãos.
29 ‘‘I record these transactions in a little book that I keep in my drawer of valu-
ables; my wife has complete knowledge of what is in there. The book has
seventy-five pages, each marked with my name or nickname. Everything and
anything written there by me is the absolute truth. I keep these records in the
same manner as any businessman.’’ (Cujos assentos e declarações faço em um
livrinho que tenho na minha gaveta, aonde trago as mais cousas de valor de que
minha mulher tem perfeito conhecimento e o dito livrinho tem setenta e cinco
folhas, rubricados com o meu nome ou apelido, Monte. Tudo quanto estiver
assentado e declarado nele por minha letra, é a mesma verdade. Os ditos
assentos e declarações de dívidas, os faço em uma página conforme o número
das folhas e as saídas em fronte como livros de deve, e o de haver dos homens
de negócio.)
acmrj, Livro de Óbitos e Testamentos da Freguesia da Sé, 1776–84, will of
Ignacio Gonçalves do Monte, fols. 442v–44.
30 Ibid., fols. 42v and 391. Carvalho and Braga and mentioned in the Mahi
Manuscript as well as other documents from the Mahi Congregation.
31 acmrj, Livro de Óbitos e Testamentos da Freguesia da Sé, 1746–58, fols. 211–12.
32 acmrj, Livro de Óbitos e Testamentos da Freguesia da Sé, 1797–1809, fol. 90v.
33 Declaro que fui nascido em terras de brutos e de gentilidade como foi, [ . . . ]
a Costa da Mina e sou da nação Cobu e por [ . . . ] o Senhor se querer lembrar
de mim, passei daquele reino, dele [ . . . ] há sete anos pouco mais ou menos
a vender [ . . . ] da Cidade da Bahia onde fui vendido ao primeiro senhor que
tive na dita [terra cidade] por nome Antônio de Bastos Mendes, o que me
ensinou a Santa Doutrina e me mandou batizar e, com efeito, fui batizado
na Igreja de Nossa Senhora da Conceição da Praia da dita cidade qual parti
também a vender nesta Cidade do Rio de Janeiro e fui comprado pelo
defunto em sua vida Antônio Soares homem estrangeiro e boticário que foi
e morou com sua mulher detrás do Convento da Nossa Senhora do Monte
do Carmo e desta escravidão me libertei depois do dito defunto por du-
zentos e tantos mil réis que dei pela minha pessoa ao defunto o Reverendo
Padre Teodósio de Souza como testamenteiro do mesmo defunto o qual
testamenteiro me passou Carta da Liberdade em notas pelo sobredito preço

tudo feito a beneplácito e consentimento e vontade da mulher viúva do
sobredito defunto.
acmrj, Livro de Óbitos e Testamentos da Freguesia da Sé, 1746–58, will of Antônio
Luiz Soares, deceased 27 January 1755, fols. 298v–301.
34 Silva’s wealth is indicated in her request that her corpse be dressed in the habit of
São Francisco, and that her funeral procession be accompanied by twelve priests.
35 The price was 233,330 réis. an, ‘‘Escritura de venda de chãos que faz Maria
Correia de Abreu, viúva de Bernardo Tavares, aos irmãos de Santo Elesbão e
Santa Efigênia Manoel dos Santos Martins, Antonio Pires dos Santos e Fran-
cisco Gonçalves Nunes, pretos forros . . .’’ 4\ O√ício de Notas. Livro de Notas:
24/2/1744–7/9/1744. N 36, p. 109.
36 acmrj, Livro de Óbitos e Testamentos da Freguesia da Sé, 1746–58, fls. 211–12.
37 Declaro que sou natural da Costa da Mina preto forro, e liberto, e fui escravo
de Domingos Gonçalves do Monte, a quem dei por minha liberdade trezen-
tos, e cinqüenta mil réis como constará da minha Carta de Alforria. Declaro
que sou casado com Vitória Correa da Conceição, preta forra, e até o fazer
deste meu testamento não temos tido filhos, e nem os tenho de outra
qualquer mulher em solteiro, e nem depois de casado. Declaro que a dita
minha mulher é minha parenta por sangüinidade [sic] em terceiro grau, por
ser ela filha do meu avô [Eseú] Agoa; bem conhecido rei que foi entre os
gentios daquela costa do Reino de May, ou Maqui.
acmrj, Livro de Óbitos e Testamentos da Freguesia da Sé, 1776–84, will of Ignacio
Gonçalves Monte, fols. 422v–44.
38 Slippage between the concepts of territory/region/land on the one hand, and
country on the other, is common not only in the assertions in these wills but
also in the documents of Portuguese and French colonial administrations. See,
for example, the title of the work by Bergé—‘‘Étude sur le Pays Mahi’’—where
the word pays (country) is used instead of the more common, and theoretically
distinct, terre (land).
39 Soares, ‘‘O Império de Santo Elesbão na Cidade do Rio de Janeiro, século
XVIII.’’
40 He specifies: ‘‘O qual testamento por assim o querer, o fiz da minha letra e
Sinal, hoje Cidade de São Sebastião do Rio de Janeiro dia mes e ano ao principio
declarado. Ignacio Gonçalves do Monte.’’ acmrj, Livro de Óbitos e Testamentos
da Freguesia da Sé, 1776–84, will of Ignacio Gonçalves Monte, fols. 422v–444.
41 See the roster of directors of the Mahi Congregation in chapter 6, where Braga
is identified as a duke, or ‘‘aeolû cocoti de daçâ’’ The su≈x de daçâ likely meant
‘‘from Dassa.’’ Also, José Luiz, a marquis, was called ajacôto chaul de za,’’ and
Za was the name of a city west of Dahomey. For more on these cities see Law,
The Slave Coast of West Africa, 1550–1750.
42 The ‘‘god of her land’’ phrase mentioned above emerged from a court hearing
against one Josepha Coura, which has been analyzed by Luiz Mott. According
to the o≈cial narrative of the case, Josepha engaged in something called the
Acotundá (Dance of Tunda), uttered chants in the Coura language, and o√ered
sacrifices to the god of her land, which was represented by an idol: ‘‘A clay
figure with its facial features clearly in imitation of the Devil, wearing a white
black cape, impaled on an iron stake. This pagan doll was placed on a rug in the
middle of her house, surrounded by clay pots of water and metal pans contain-
ing o√erings such as cooked and raw herbs, shells, African money, a dead
chicken, beans, et cetera. Luiz Mott, ‘‘Acotundá.’’
43 On the lands and nations most prominent in eighteenth-century discourse, see
Soares, ‘‘A ‘Nação’ que Se Tem.’’ For more on the disappearance of the Mina
nation in Rio de Janeiro, see Soares, ‘‘From Gbe to Yoruba.’’
44 As said before, the full title of the Mahi Manuscript at the overture of the
document is ‘‘Regra ou estatuto por modo de um diálogo onde, se dá notícia
das Caridades e Sufragações das Almas que usam os pretos Minas, com seus
Nacionais no Estado do Brazil, especialmente no Rio de Janeiro, por onde se
hão de regerem e governarem for a de todo o abuzo gentílico e supersticioso;
composto por Francisco Alves de Souza preto e natural do Reino de Makim, um
dos mais excelentes e potentados daquela oriunda Costa da Mina.’’
45 David Eltis, Stephen D. Behrendt, David Richardson, and Herbert Klein, The
Trans-Atlantic Slave Trade: A Database.
bibliography
abbreviations
abn —Anais da Biblioteca Nacional
acmrj —Arquivo da Cúria Metropolitana do Rio de Janeiro
ahu /cu —Arquivo Histórico Ultramarino/Coleção Ultramarina
aisese —Arquivo da Irmandade de Santo Elesbão e Santa Efigênia
an —Arquivo Nacional
apeb —Arquivo Público do Estado da Bahia
bep —Biblioteca do Estado de Pernambuco
bn(rj)—Biblioteca Nacional do Rio de Janeiro
cnsr —Confraria de Nossa Senhora dos Remédios
ihgb —Instituto Histórico e Geográfico Brasileiro
rihgb —Revista do Instituto Histórico e Geográfico Brasileiro
archives and manuscripts consulted

Arquivo da Cúria Metropolitana do Rio de Janeiro (acmrj)
Habilitações Matrimoniais: Ignacio Monte e Victoria Correa, doc. 22119, caixa
1648
Livro de Batismo de Escravos, Freguesia da Sé, 1718–26
Livro de Batismo de Escravos, Freguesia da Sé, 1744–61
Livro de Batismo de Escravos, Freguesia da Candelária, 1745–74
Livro de Batismo de Escravos, Freguesia de São José, 1751–90
Livro de Batismo de Escravos, Freguesia de Santa Rita, 1751–99
Livro de Óbito de Escravos, Freguesia da Candelária, 1724–36
Livro de Óbito de Escravos, Freguesia da Candelária, 1793–33
Livro de Óbito e Testamentos, Freguesia de Santa Rita, 1751–17[?]
Livro de Óbitos e Testamentos da Freguesia da Sé, 1746–1758
294 | Bibliography
Livro de Óbitos e Testamentos da Freguesia da Sé, 1776–1784

Livro de Óbitos e Testamentos da Freguesia de Nossa Senhora da Candelária, 1797–1809
Livro de Portarias Episcopais, #3, 1779–1830
Arquivo Histórico Ultramarino, Lisbon

‘‘Estatuto da Irmandade de Nossa Senhora dos Remédios,’’ ahu /cu-cód. 1300
‘‘Representação da Irmandade do Rosário,’’ ahu, Rio de Janeiro, caixa 107,
doc. 31
Biblioteca do Estado de Pernambuco

‘‘Correspondência da Corte (1780–81)’’
Biblioteca Nacional, Rio de Janeiro (bn [rj])

‘‘Regra ou estatuto por modo de um diálogo onde, se dá notícia das Caridades e
Sufragações das Almas que usam os pretos Minas, com seus Nacionais no
Estado do Brazil, especialmente no Rio de Janeiro, por onde se hão de regerem
e governarem for a de todo o abuzo gentílico e supersticioso; composto por
Francisco Alves de Souza preto e natural do Reino de Makim, um dos mais
excelentes e potentados daquela oriunda Costa da Mina. (1787)’’
Museu do Negro: Arquivo da Irmandade de Santo Elesbão

e Santa Efigênia (aisese)
‘‘Compromisso da Irmandade de Santo Elesbão e Santa Efigênia’’ (copy,
including ecclesiastical correspondence pertaining to the compromisso’s
approval)
‘‘Estatutos da Congregação dos Pretos Minas Maki no rj (1786)’’
published sources, books, and articles

Abreu, Martha Campos. ‘‘O Império do Divino: Festas Religiosas e Cultura
Popular no Rio de Janeiro (1830–1900).’’ Ph.D. diss., Universidade de
Campinas, 1996.
Abreu, Maurício. Evolução Urbana do Rio de Janeiro. Rio de Janeiro: iplanrio,
1988.
Agassiz, Luiz, and Elizabeth Cary Agassiz. Viagem ao Brasil: 1865–1866. Belo
Horizonte, 1975.
Akinjogbin, I. A. Dahomey and Its Neighbors, 1708–1818. Cambridge: Cambridge
University Press, 1967.
Alden, Dauril. ‘‘The Population of Brazil in the Late Eighteenth Century: A
Preliminary Study.’’ Hispanic American Historical Review 43 (May 1963): 175–205.
Bibliography | 295
Alencastro, Luis Felipe. ‘‘La Traite Négrière et l’Unité Nationale Brésilienne.’’

Revue Française d’Histoire d’Outre-Mer 66 (1979): 244–45.
Algranti, Leila Mezan. Honradas e Devotas: Mulheres da Colônia. Estudo sobre a Condição
Feminina nos Conventos e Recolhimentos do Sudeste do Brasil (1750–1822). Rio de
Janeiro: José Olympio Ed., 1993.
Almeida, Eduardo de Castro, ed. ‘‘Inventário dos Documentos Relativos ao Brasil
Existentes no Archivo da Marinha e Ultramar Organizado por Eduardo de
Castro Almeida: Bahia 1613–1762.’’ Annaes da Biblioteca Nacional do Rio de Janeiro.
Vol. 31 (1909). Rio de Janeiro: O≈cinas Graphicas da Bibliotheca Nacional,
1913.
———. ‘‘Inventário dos Documentos Relativos ao Brasil Existentes no Archivo da
Marinha e Ultramar Organizado por Eduardo de Castro Almeida: Bahia 1613–
1762.’’ Annaes da Biblioteca Nacional do Rio de Janeiro, vol. 34 (1917). Rio de
Janeiro: O≈cinas Graphicas da Bibliotheca Nacional, 1921.
Andreoni, João António (André João Antonil). Cultura e Opulência do Brasil:
Introdução e Vocabulário por A. P. Canabrava. São Paulo: Companhia Editora
Nacional, 1967.
Araújo, José de Souza Azevedo Pizarro (Monsenhor Pizarro). Memórias Históricas do
Rio de Janeiro. 10 vols. Rio de Janeiro: Imprensa Nacional, 1948.
Ariès, Philipp. O Homem Diante da Morte. 2 vols. Rio de Janeiro: Francisco Alves, 1989.
Augé, Paul. Larousse du XXème Siècle. 6 vols. Paris: Larousse, 1930.
Azevedo, João Lúcio de. Épocas de Portugal Económico: Esboços de História. Lisbon:
Livraria Clássica Editora, 1929.
———. ‘‘Judaísmo no Brasil.’’ Revista do Instituto Histórico e Geográfico Brasileiro 91, no.
682.
Azevedo, Moreira de. O Rio de Janeiro: Sua História, Monumentos, Homens Notáveis,
Usos e Curiosidades. 2 vols. Rio de Janeiro: Livraria Brasiliana Editora, 1969.
Azevedo, Paulo Cesar, and Mauricio Lissovsky, eds. Escravos Brasileiros do Século XIX
na Fotografia de Christiano Jr. Essays by Jacob Gorender, and Manuela Carneiro da
Cunha e Muniz Sodré. São Paulo: Editora Ex Libris, 1988.
Baldé, S. M. ‘‘L’Esclavage et la Guerre Sainte au Fuuta-Jalon (Maccugaaku au
Funta-Jaloo).’’ In L’Esclavage en Afrique Précoloniale, edited by Claude Meillassoux.
Paris: Françoise Maspero, 1975.
Barreiros, Eduardo Canabrava. Atlas da Evolução Urbana da Cidade do Rio de Janeiro:
Ensaio—1565–1965. Rio de Janeiro: ihgb, 1965.
Barreto Filho, Mello, and Hermeto Lima. História da Polícia do Rio de Janeiro: Aspectos
da Vida Carioca 1565–1831. 3 vols. Rio de Janeiro: Editora A Noite, 1939.
Barreto, Luis Felipe. Descobrimentos e Renascimento: Formas de Ser e de Pensar nos Séculos
XV e XVI. Lisbon: Imprensa Nacional, Casa da Moeda, 1982.
Barros, João de. Ásia: Dos Feitos Que os Portugueses Fizeram no Descobrimento e Conquista
dos Mares e Terras do Oriente. 1\ Década. 6th ed. Lisbon: Agência Geral das
Colônias, 1945.
296 | Bibliography
Barth, Fredrik. ‘‘Introduction.’’ In Ethnic Groups and Boundaries: The Social

Organization of Culture Di√erence, edited by Fredrik Barth. Boston: Little, Brown,
1969.
Bartolomé, Miguel Alberto. ‘‘Bases Culturales de la Identidad Étnica.’’ In Gente de
Costumbre y Gente de Razón: Las Identidades Étnicas en México, edited by Miguel
Alberto Bartolomé. Mexico City: Siglo Vientiuno Editores, 1997.
Bastide, Roger. As Religiões Africanas no Brasil: Contribuição a uma Sociologia das
Interpenetrações de Civilizações. São Paulo: Livraria Pioneira Editora, 1989.
———. O Candomblé da Bahia. São Paulo: Companhia Editora Nacional, 1978.
Bay, Edna G. Wives of the Leopard: Gender, Politics, and Culture in the Kingdom of
Dahomey. Charlottesville: University of Virginia Press, 1998.
Bergé, J. A. M. A. R. ‘‘Étude sur le Pays Mahi.’’ Bulletin du Comité d’Études Historiques
et Scientifiques de l’A O F II (1928).
Berger, Paulo. Bibliografia do Rio de Janeiro de Viajantes e Autores Estrangeiros, 1531–
1900. Rio de Janeiro: Livraria São José, 1964.
Bethell, Leslie. A Abolição do Tráfico de Escravos no Brasil: A Grã-Bretanha, o Brasil e a
Questão do Tráfico de Escravos, 1807–1869. Rio de Janeiro: Expressão e Cultura,
1976.
Biblioteca Nacional. Anais da Biblioteca Nacional do Rio de Janeiro. Vol. 31. Rio de
Janeiro: Biblioteca Nacional, 1909.
———. Anais da Biblioteca Nacional do Rio de Janeiro. Vol. 34. Rio de Janeiro: Biblioteca
Nacional, 1910.
Bicalho, Maria Fernanda, ‘‘A Cidade e o Império: O Rio de Janeiro na Dinâmica
Colonial Portuguesa, Séculos XVII e XVIII.’’ Ph.D. diss., University of São
Paulo, 1997.
Blake, John William. European Beginnings in West Africa, 1454–1578: A Survey of the
First Century of White Enterprise in West Africa, with Special Emphasis upon the Rivalry of
the Great Powers. Westport, Conn.: Greenwood Press, 1937.
Bloch, Marc. Introdução à História. Publicações Europa-América, n.d.
Boschi, Caio César. Os Leigos e o Poder: Irmandades Leigas e Política em Minas Gerais.
São Paulo: Editora Ática, 1986.
Bosi, Alfredo. Dialética da Colonização. São Paulo: Companhia das Letras, 1992.
Bossy, John. A Cristandade no Ocidente, 1400–1700. Lisbon: Edições 70, 1990.
Bott, Elizabeth. Família e Rede Social: Papéis, Normas e Relacionamentos Externos em
Famílias Urbanas Comuns. Preface by Max Gluckman. Rio de Janeiro: Francisco
Alves, 1976.
Boxer, C. R. ‘‘Brazilian Gold and British Traders in the First Half of the
Eighteenth Century.’’ Hispanic American Historical Review 19, no. 3 (1969): 455–
72.
———. A Idade de Ouro do Brasil: Dores de Crescimento de uma Sociedade Colonial. São
Paulo: Companhia Editora Nacional, 1969.
———. O Império Colonial Português (1415–1825). Lisbon: Edições 70, 1981.
Bibliography | 297
———. Relações Raciais no Império Colonial Português (1415–1825). Porto, Portugal:

Afrontamento, 1988.
———. Salvador de Sá e a Luta pelo Brasil e Angola (1602–1686). São Paulo: Editora
Nacional, 1973.
Bragança, J. ‘‘Introdução’’ and ‘‘Do Título desta Crônica.’’ In Crónicas de Guiné, by
Gomes Eanes Zurara. Barcelos, Portugal: Livraria Civilização Editora, 1973.
Braudel, Fernand. Escritos sobre a História. São Paulo: Editora Perspectiva, 1992.
Burguière, André. ‘‘A Antropologia Histórica.’’ In A Nova História, edited by
Jacques Le Go√. Coimbra: Almedina, 1990.
Burke, Peter. A Fabricação do Rei: A Construção da Imagem Pública de Luis XIV. Rio de
Janeiro: Jorge Zahar Editor, 1994.
Campos, Adalgasia Arantes, et al. ‘‘O Banco de Dados Relativo ao Acervo da
Freguesia da Nossa Senhora do Pilar de Ouro Preto: Registros Paroquiais e as
Possibilidades de Pesquisa.’’ Proceedings of the 10th Seminar on the Economy
of Minas Gerais. Diamantina, Brazil. 2002
Campos, S. J. Procissões Tradicionais da Bahia. [Published posthumously.] Preface by
Arnaldo Pimenta da Cunha. Salvador: Museu da Bahia: Secretaria de Educação
e Saúde, 1941.
Carmo Neto, Henrique José do. ‘‘Recordações e Aspectos do Culto de Sangt’Anna.’’
Revista do Instituto Histórico e Geográfico Brasileiro 94, no. 142, pt. 2 (1927).
Carreira, António. As Companhias Pombalinas de Navegação, Comércio e Tráfico de
Escravos entre a Costa Africana e o Nordeste Brasileiro. Porto, Portugual: Centro de
Estudos da Guiné Portuguesa, 1969.
———. Notas sobre o Tráfico Português de Escravos, Circunscritos à Costa Ocidental Africana.
Lisbon: Universidade Nova de Lisboa, 1978.
Carvalho, Delgado de. História da Cidade do Rio de Janeiro. Rio de Janeiro: Secretaria
Municipal da Cultura, 1994.
Cascudo, Luís da Câmara. Dicionário do Folclore Brasileiro. Rio de Janeiro: Ediouro,
n.d.
Castro, Hebe Maria Mattos de. Das Cores do Silêncio: Os Significados da Liberdade no
Sudeste Escravista—Brasil, Século XIX. Rio de Janeiro: Arquivo Nacional, 1995.
Castro, Therezinha de. História Documental do Brasil. Rio de Janeiro: Record, n.d.
Catálogo da Exposição de História do Brasil. Realizada pela Biblioteca Nacional do Rio
de Janeiro a 2 de Dezembro de 1881. Rio de Janeiro: Typographia G. Leuzinger
and Filhos, 1881. (3 volumes)
Cavalcanti, Nireu Oliveira, ‘‘A Cidade do Rio de Janeiro: As Muralhas, Sua Gente,
os Construtores (1710–1810).’’ Ph.D. diss., Federal University of Rio de Janeiro,
1998.
Chahon, Sérgio. ‘‘Aos Pés do Altar e do Trono: As Irmandades e o Poder Régio no
Brasil (1808–1822).’’ M.A. thesis, Universidade de São Paulo, 1996.
Chartier, Roger. ‘‘Textos, Impressões, Leituras.’’ In A Nova História Cultural, edited
by Lynn Hunt. São Paulo: Martins Fontes, 1992.
298 | Bibliography
Chartier, Roger, and D. Roche. ‘‘Social (História).’’ In A Nova História, edited by

Jacques Le Go√. Coimbra: Almedina, 1990.
Chaunu, Pierre. Conquista e Exploração dos Novos Mundos: Século XVI. São Paulo:
Edusp, 1984.
———. Expansão Européia do Século XIII ao XV. São Paulo: Pioneira, 1978.
Coaracy, Vivaldo. Memórias da Cidade do Rio de Janeiro. Belo Horizonte: Itatiaia,
1988.
———. O Rio de Janeiro no Século Dezessete. Rio de Janeiro: Livraria José Olympio
Editora, 1965.
Código Philipino ou Ordenações e Leis do Reino de Portugal recompiladas por mando D’El Rey
D. Philipe I. Facsimile 1603 and 1821. Brasília, Senado Federal. 2004.
Constituições Primeiras do Arcebispado da Bahia. Feitas e Ordenados pelo Illustrissimo e
Reverendissimo Senhor D. Sebastião Monteiro da Vide 5\ Arcebispo da dito Arcebispado, e
do Conselho de sua Magestade: Propostas, e Aceitas em o Sinodo Diocesano, Que o dito
senhot celebrou em 12 de junho do anno 1707. São Paulo: Typographia 2 de
Dezembro de Antonio Louzada Antunes, 1853.
Coquery-Vidrovitch, Catherine, ed. A Descoberta da África. Lisbon: Edições 70, 1981.
Cornevin, Robert. Histoire du Dahomey avec 10 Cartes, 1 Aroquis et 35 Photographies.
Paris: Éditions Berger-Levrault, 1962.
Cunha, Manuela Carneiro da. Antropologia do Brasil. Mito–História–Etnicidade. São
Paulo: Brasiliense, 1987.
———, ed. História dos Índios no Brasil. São Paulo: fapesp, 1992.
Cuvelier, J., and L. Jadin. L’Ancien Congo d’après les Archives Romaines (1518–1640).
mircb, xxxv I, 2. Brussels, 1954.
Debret, Jean-Baptiste. O Brasil de Debret. Essay by Sérgio Milliet, Rubens Borba de
Morais, and Antônio Carlos Villaça. Belo Horizonte: Editora Vila Rica, 1993.
———. Viagem Pitoresca e Histórica ao Brasil. Essay by Sérgio Millet. São Paulo. Livraria
Martins, 1940.
———. Viagem Pitoresca e Histórica ao Brasil. Preface by Antônio Carlos Villaça. Belo
Horizonte: Itatiaia, 1989.
Delumeau, Jean. A Civilização do Renascimento. Lisbon: Editorial Estampa, 1984.
Devisse, Jean, ed. Les Assises du Pouvoir: Temps Médiévaux, Territoires Africains. Saint-
Denis, France: Presses Universitaires de Vincennes, 1994.
Duby, Georges. Guilherme Marechal ou o Melhor Cavaleiro do Mundo. Rio de Janeiro:
Graal, 1988.
Elias, Norbert. A Sociedade de Corte. Lisbon: Editorial Estampa, 1987.
Eltis, David, Stephen D. Behrendt, David Richardson, and Herbert Klein. The
Trans-Atlantic Slave Trade: A Database on cd-rom . New York: Cambridge
Espanha, Antônio Manoel. Às Vésperas do Leviathan: Instituições e Poder Político,
Portugal—séc. XVII. Coimbra, Portugal: Livraria Almedina, 1994.
Faria, Sheila de Castro. ‘‘Sinhás Pretas, ‘Damas Mercadoras.’ As pretas minas nas
Bibliography | 299
cidades do Rio de Janeiro e de São João Del Rei (1700–1850).’’ Manuscript,

2004. In author’s possession.
Faria, Sheila Siqueira de Castro. ‘‘A Colônia em Movimento: Fortuna e Família no
Cotidiano Colonial (Sudeste, Século XVIII).’’ Ph.D. diss., Universidade Federal
Fluminense, Niterói, 1994.
Fazenda, Vieira. ‘‘Antiqualhas e Memórias do Rio de Janeiro.’’ Revista do Instituto
Histórico e Geográfico Brasileiro 86, no. 140 (1919), 1921.
———. ‘‘Antiqualhas e Memórias do Rio de Janeiro.’’ Revista do Instituto Histórico e
Geográfico Brasileiro 89, no. 143 (1921), 1924.
Geográfico Brasileiro 93, 147 (1923), 1927.
Geográfico Brasileiro 88, 142 (2nd ed.), 1940.
Geográfico Brasileiro 95, 149 (2nd ed.), 1943.
Ferreira, Aurélio Buarque de Holanda. Dicionário da Língua Portuguesa. Rio de
Janeiro: Nova Fronteira, 1975.
Ferreira, João da Costa. ‘‘A Cidade do Rio de Janeiro e Seu Termo.’’ Revista do
Instituto Histórico e Geográfico Brasileiro 164 (1931). O√print.
Ferrez, Gilberto. As Cidades do Salvador e Rio de Janeiro no Século XVIII: Álbum
Iconográfico Comemorativo do Bicentenário da Transferência da Sede do Govêrno do Brasil.
Rio de Janeiro: ihgb, 1963.
———. ‘‘Diário Anônimo de uma Viagem às Costas d’África e às Índias Espanholas.’’
Revista do Instituto Histórico e Geográfico Brasileiro 267 (April–July 1965).
———. ‘‘O Rio de Janeiro.’’ Revista do Instituto Histórico e Geográfico Brasileiro 288 (July–
September 1970). Published in 1971.
———. O Rio de Janeiro e a Defesa do Seu Porto—1555–1800. Rio de Janeiro: Serviço de
Documentação Geral da Marinha, 1972.
Figueiredo, Luciano. O Avesso da Memória: Cotidiano e Trabalho da Mulher em Minas
Gerais no Século XVIII. Preface by Laura de Mello e Souza. Brasília: EdUnB, 1993.
Fitzler, M. A. Hedwig, and Ernesto Enes. A Seção Ultramarina da Biblioteca Nacional:
Inventários. Lisbon: Biblioteca Nacional. 1928.
Florentino, Manolo Garcia. ‘‘Alforrias e Etnicidade no Rio de Janeiro Oitocentista:
Notas de Pesquisa.’’ Topoi 5, 2002.
———. Em Costas Negras: Uma História do Tráfico Atlântico de Escravos entre a África e o Rio
de Janeiro (Séculos XVIII e XIX). Rio de Janeiro: Arquivo Nacional, 1995.
Fynn, J. K. Asante and Its Neighbours, 1700–1807. Evanston: Northwestern
Gbéto, Flavien. Le Maxi du Centre-Bénin et du Centre-Togo: Une approche Autosegmentale
et Dialectologique d’un Parler Gbe de la Section Fon. Cologne: Köppe, 1997.
Ginzburg, C. ‘‘Sinais: Raízes de um Paradigma Indiciário.’’ In Mitos, Emblemas,
Sinais: Morfologia e História. São Paulo: Companhia das Letras, 1989.
300 | Bibliography
Godinho, Vitorino Magalhães. Ensaios II: Sobre a História de Portugal. Lisbon:

Livraria Sá da Costa Editora, 1968.
Go√man, Erving. Estigma: Notas sobre a Manipulação da Identidade Deteriorada. Rio de
Janeiro: Editora Guanabara, n.d.
Goulart, Maurício. A Escravidão Africana no Brasil: Das Origens à Extinção do Tráfico.
Preface by Sérgio Buarque de Holanda. São Paulo: Editora Alfa-Omega, 1975.
Gouveia, Alfredo Mendes de. ‘‘Relação dos Compromissos de Irmandades,
Confrarias e Misericórdias do Brasil: Existentes no Arquivo Histórico Colonial
de Lisboa, Que Pertenceram ao Cartório do Extinto Conselho Ultramarino
1716–1807.’’ Anais do IV Congresso de História Nacional 7 (1949): 201–38.
Gudeman, Stephen, and Stuart Schwartz. ‘‘Purgando o Pecado Original:
Compadrio e Batismo na Bahia no Século XVIII.’’ In Escravidão e Invenção da
Liberdade, edited by João José Reis. São Paulo: CNPq, 1988.
Guedes, Max Justo, and Gerald Lombardi, eds. Portugal Brazil: The Age of Atlantic
Discoveries. Lisbon: Bertrand Editora, 1990.
Guia Brasileiro de Fontes para a História da África, da Escravidão Negra e do Negro na
Sociedade Atual. Vol. 2: Rio de Janeiro-Sergipe. Rio de Janeiro: Arquivo Nacional,
1988.
Henriques, Isabel de Castro. ‘‘L’Invention Sociale de São Tomé et Principe au XVIe
siècle.’’ In Les Assises du Pouvoir: Temps Médiévaux, Territoires Africains. Saint-Denis,
France: Presses Universitaires de Vincennes, 1994.
Herskovits, Melville J. Antropologia Cultural. São Paulo: Editora Mestre Jou, 1973.
———. ‘‘On the Provenience of New World Negroes.’’ Social Forces 12 (1933): 247-62.
Holanda, Sérgio Buarque de. ‘‘Prefácio.’’ In A Escravidão Africana no Brasil: Das
Origens à Extinção do Tráfico. São Paulo: Editora Alfa-Omega, 1975.
———. ‘‘O Semeador e o Ladrilhador.’’ In Raizes do Brasil. Rio de Janeiro: José
Olympio Editora, 1993.
Hoormaert, Eduardo, Riolando Azzi, Klaus van der Grijp, and Benno Brod.
História da Igreja no Brasil: Ensaio de Interpretação a Partir do Povo. Primeira Época.
Petrópolis: Edições Paulinas, 1992.
Jancsó, István. Cronologia de História do Brasil Colonial (1500–1831). São Paulo:
Departamento de História, 1994.
Joutard, Philippe. ‘‘Un Projet Régional de Recherche sur les Ethnotextes.’’ In
Annales: Économie Sociétés Civilisations 35, no. 1 (January–February 1980): 176–82.
Julião, Carlos. Riscos Iluminados de Figurinhos de Brancos e Negros dos Uzos do Rio de
Janeiro e Serro do Frio. Historical introduction and descriptive catalogue by Lygia
da Fonseca Fernandes Cunha. Rio de Janeiro: Biblioteca Nacional, 1960.
Karasch, Mary. Slave Life in Rio de Janeiro, 1808–1850. Princeton: Princeton
Klein, Herbert S. ‘‘The Portuguese Trade from Angola in the 18th Century.’’ In The
Middle Passage: Comparative Studies in the Atlantic Slave Trade. Princeton: Princeton
Bibliography | 301
Lacerda, José Maria de Almeida e Araújo Correa de. Diccionario Enciclopédico. 2 vols.
Lisbon: Escritório de Francisco Arthur da Silva, 1879.
Lapa, José Roberto do Amaral. A Bahia e a Carreira da Índia. São Paulo: Companhia
Editora Nacional, 1968.
Lara, Silvia Hunold, ed. Revista Brasileira de História, 16, file ‘‘Escravidão.’’ São
Paulo: anpuh, 1988.
———. ‘‘Sob o Signo da Cor: Trajes Femininos e Relações Raciais nas Cidades de
Salvador e do Rio de Janeiro, ca. 1750–1815.’’ Paper presented at the Latin
American Studies Association, 1974.
Law, Robin. ‘‘Ethnicity and the Slave Trade: ‘Lucumi’ and ‘Nago’ as Ethnonyms in
West Africa.’’ History in Africa 24 (1997): 205–19.
———. The Slave Coast of West Africa, 1550–1750: The Impact of the Atlantic Slave Trade on
an African Society. Oxford: Clarendon Press, 1991.
Leal, Euripedes Franklin, and Ana Regina Berwanger. Noções de Paleografia e
Diplomática. Santa Maria: Universidade Federal de Santa Maria, 1992.
Le Go√, Jacques, ed. A Nova História. Coimbra: Almedina, 1990.
Leite, Miriam Moreira, ed. A Condição Feminina no Rio de Janeiro no Século XIX:
Antologia de Textos de Viajantes Estrangeiros. São Paulo: Editora Hucitec, 1984.
Lisboa, Balthazar da Silva. Annaes do Rio de Janeiro, Contendo a Descoberta e Conquista
deste Paíz, a Fundação da Cidade com a História Civil e Eclesiástica, até a Chegada d’el-Rei
Dom João VI; Além de Notícias Topográphicas, Zoológicas e Botânicas. Bk. 7. Rio de
Janeiro: Typ. Imp. E Const. De Seignot-Plancher e Cia, 1835.
Lloyd, P. C. ‘‘The Political Structure of African Kingdoms—an Exploratory
Model.’’ In Political Systems and the Distribution of Power, edited by Michael
Banton. London, 1966.
Lobo, Maria Eulália Lahmeyer. História do Rio de Janeiro (do Capital Commercial ao
Capital Industrial e Financeiro). Rio de Janeiro: ibmec, 1978.
———. ‘‘Historiografia do Rio de Janeiro.’’ Revista Brasileira de História, file
‘‘Historiografia Propostas e Práticas,’’ 15, 30 São Paulo: anpuh Contexto,
1995: 45–62.
Lopes, Vânia Penha. ‘‘À Venerável Irmandade de Santo Elesbão e Santa Efigênia,
ou uma Tentativa de Entendimento da Questão Étnica no Brasil.’’ B.A. thesis,
Universidade Federal Fluminense, Niterói, 1981.
Lovejoy, Paul E. ‘‘Identifying Enslaved Africans in the African Diaspora.’’ In
Identity in the Shadow of Slavery, edited by Lovejoy. London: Continuum, 2000.
Luccock, John. Notas Sobre o Rio de Janeiro e Partes Meridionais do Brasil. Foreword by
Mário Guimarães Ferri. Belo Horizonte: Editora Itatiaia, 1975.
Maggie, Yvonne. Medo do Feitiço: Relações entre Magia e Poder no Brasil. Rio de Janeiro:
Arquivo Nacional, 1992.
Maia, Moacir Rodrigo de Castro. ‘‘ ‘Quem Tem Padrinho Não Morre Pagão’: As
Relações de Compadrio e Apadrinhamento de Escravos numa Vila Colonial
(Mariana, 1715–1750).’’ M.A. thesis, Universidade Federal Fluminense, Niterói,
2006.
302 | Bibliography
Malheiro, Perdigão. A Escravidão no Brasil: Ensaio Histórico, Jurídico, Social. 2 vols.

Petrópolis: Vozes, 1976.
Mapas Históricos Brasileiros. São Paulo: Abril Cultural, n.d.
Martins, José de Souza, ed. A Morte e os Mortos na Sociedade Brasileira. São Paulo:
Hucitec, 1983.
Mascarenhas, Luiz de Almeida Soares Portugal Alarcão Eça Mello Silva e
(Marquês do Lavradio). ‘‘Relatório.’’ Revista do Instituto Histórico e Geográfico
Brasileiro 4 (1842): 453–76.
———. ‘‘Relatório.’’ Revista do Instituto Histórico e Geográfico Brasileiro 76 (1913).
———. ‘‘Memórias Públicas e Econômicas da Cidade de São Sebastião do Rio de
Janeiro para Uso do Vice-rei Luiz de Vasconcellos por Observação Curiosa dos
Anos de 1779 até o de 1789.’’ Revista do Instituto Histórico e Geográfico Brasileiro 47,
pt. 1 (1884): 25–51.
Mattoso, Kátia M. de Queiróz. Bahia, Século XIX: Uma Província no Império. Rio de
Janeiro: Editora Nova Fronteira, 1992.
———. Ser Escravo no Brasil. São Paulo: Editora Brasiliense, 1988.
Maurício, Augusto. Templos Históricos do Rio de Janeiro. Rio de Janeiro: Gráfica
Laemmert Limitada, 1947.
Mauro, Frédéric. Le Portugal et l’Atlantique au XVII Siècle (1570–1670): Étude
Économique. Paris: École Pratique des Hautes Études Sixième Section
S.E.V.P.E.N., 1960.
McEvedy, Colin. Atlas da História Moderna (até 1815). São Paulo: Editora Verbo, 1979.
Meillassoux, Claude. Antropologia da Escravidão: O Ventre de Ferro e Dinheiro. Rio de
Janeiro: Jorge Zahar Editor, 1995.
———, ed. L’Esclavage en Afrique Précolonial. Paris: Maspero, 1975.
Mello Junior, Donato. Rio de Janeiro: Planos, Plantas e Aparências. Rio de Janeiro:
Edição da Galeria de Arte do Centro Empresarial Rio, 1988.
Monteiro, John M. Negros da Terra: Índios e Bandeirantes nas Origens de São Paulo. São
Paulo: Companhia das Letras, 1994.
Mota, A. Teixeira. Topónimos de Origem Portuguesa na Costa Ocidental de África: Desde o
Cabo Bojador ao Cabo de Santa Caterina. Bissau: Centro de Estudos da Guiné
Portuguesa, no. 14, 1950.
Mott, Luiz. ‘‘Acotundá: Raízes Setecentistas do Sincretismo Religioso Afro-
Brasileiro.’’ In Escravidão, Homossexualidade e Demonologia. São Paulo: Ícone
Editora, 1998.
———. ‘‘Cotidiano e Vivência Religiosa: Entre a Capela e o Calundu.’’ In História da
Vida Privada no Brasil I, edited by Laura de Mello e Souza, ed. São Paulo:
Companhia das Letras, 1997.
———. Escravidão, Homossexualidade e Demonologia. São Paulo: Ícone Editora, 1998.
———. Rosa Egipcíaca: Uma Santa Africana no Brasil. Rio de Janeiro: Bertrand Brasil,
1993.
Bibliography | 303
Mulira, Jessie Gaston. ‘‘A History of the Mahi Peoples from 1774–1920.’’ Ph.D.
diss., University of California, Berkeley, 1984.
Mulvey, Patricia Ann. ‘‘The Black Lay Brotherhoods of Colonial Brazil: A History.’’
Ph.D. diss., City University of New York, 1976.
Nardi, Jean Baptiste. O Fumo Brasileiro no Período Colonial: Lavoura, Comércio e
Administração. São Paulo: Brasiliense, 1992.
Nascentes, Antenor. Efemérides Cariocas. Rio de Janeiro: Prefeitura do Distrito
Federal, 1965.
Neves, Guilherme Pereira. ‘‘E Receberá Mercê: A Mesa da Consciência e Ordens e
o Clero Secular no Brasil (1808–1828).’’ Ph.D. diss., Universidade de São Paulo,
1995.
Nora, Pierre. ‘‘Entre Mémoire et Histoire: La Problématique des Lieux.’’ In Les
Lieux de Mémoire: La République, edited by Nora. Paris: Gallimard, 1984.
Novais, Fernando A. Portugal e Brasil na Crise do Antigo Sistema Colonial (1777–1808).
São Paulo: Editora Hucitec, 1981.
———, ed. História da Vida Privada. Vol. 1: O Cotidiano da Vida Privada na América
Portuguesa. São Paulo: Companhia das Letras, 1997.
Nsondé, Jean. ‘‘Les Relations Culturelles et Commerciales entre Populations de
Langue Kongo et Européens du XVIe au XVIIIe Siècle dans la Région du Bas-
Congo.’’ In Les Assises du Pouvoir: Temps Médiévaux, Territories Africains, edited by
Jean Devisse. Saint-Denis: Presses Universitaires de France, 1994.
Nunes, Antonio Duarte. ‘‘Almanaque Histórico da Cidade de São Sebastião do Rio
de Janeiro.’’ Revista do Instituto Histórico e Geográfico Brasileiro 267 (April–June
1965): 93–214.
Oliveira, Anderson José Machado de. ‘‘Devoção e Caridade: Irmandades
Religiosas no Rio de Janeiro Imperial (1840–1889).’’ M.A. thesis, Universidade
Federal Fluminense, Niterói, 1995.
Oliveira, João Pacheco de. ‘‘Elementos para uma Sociologia dos Viajantes.’’ In
Sociedades Indígenas and Indigenismo no Brasil, edited by João Pacheco de Oliveira.
Rio de Janeiro: Universidade Federal do Rio de Janeiro, 1987.
———. ‘‘Os Instrumentos de Bordo: Expectativas e Possibilidades do Trabalho de
Antropólogo em Laudos Periciais.’’ In A Perícia Antropológica em Processos Judiciais,
edited by Orlando Silva et al. Florianópolis: Associação Brasileira de
Antropologia, 1994.
———. ‘‘O Nosso Governo.’’ Os Ticuna e o Regime Tutelar. São Paulo: Marco Zero, 1988.
Oliveira, Maria Inês Cortes de. ‘‘Quem Eram os ‘Negros da Guiné’? A Origem dos
Africanos na Bahia.’’ Afro-Ásia 19/20 (1997): 37–73.
Oliveira, Patrícia Porto de. ‘‘Batismo de Escravos Adultos e o Parentesco Espiritual
nas Minas Setecentistas.’’ Anais da V Jornada Setecentista (Curitiba), 26–28
November 2003.
Oliveira, Roberto Cardoso de. Identidade, Etnia e Estrutura Social. São Paulo: Livraria
Pioneira Editora, 1976.
304 | Bibliography
Peixoto, Antonio da Costa. Obra Nova de Língua Geral da Mina. Manuscript of the
Biblioteca Pública de Évora and the Biblioteca Nacional de Lisboa published by
Luis Silveira, with commentary by Edmund Correia Lopes. Lisbon: Agência
Geral das Colônias, 1945.
Pereira, Duarte Pacheco. Esmeraldo de Situ Orbis [ca. 1506]. Lisbon: Imprensa
Nacional, 1892.
Pina, Rui de. Crônicas de Rui de Pina: D. Sancho I, D. Afonso II, D. Sancho II, D. Afonso
III, D. Dinis, D. Afonso IV, D. Duarte, D. Afonso V, D. João II. Introduction and
revision by M. Lopes de Almeida. Porto: Lello e Irmãos Editores, 1977.
Pinheiro, Fernanda Aparecida Domingos. ‘‘Confrades do Rosário: Sociabilidade e
Identidade Étnica em Mariana, Minas Gerais (1745–1820).’’ M.A. thesis,
Universidade Federal Fluminense, Niterói, 2006.
Polanyi, Karl. Dahomey and the Slave Trade: An Analysis of an Archaic Economy. Seattle:
University of Washington Press, 1968.
Portugal e os Descobrimentos: O Encontro de Civilizações. Foreword by Vasco Graça
Moura. Commissioned for the Universal Exposition of Seville in 1992 [n.d.].
Prado Junior, Caio. Formação do Brasil Contemporâneo: Colônia. São Paulo: Editora
Brasiliense, 1972.
Rama, Angel. ‘‘La Ciudad Letrada.’’ In América Latina: Palavra, Literatura, e Cultura.
Vol. 1: A Situação Colonial, edited by Ana Pizarro. Campinas: Ed. Unicamp, 1993.
Ramos, Arthur. As Culturas Negras no Novo Mundo. São Paulo: Companhia Editora
Nacional, 1979.
———. Introdução à Antropologia Brasileira. 2 vols. Rio de Janeiro: Casa do Estudante
do Brasil, 1961.
Reis, João José. ‘‘Identidade e Diversidade Étnica nas Irmandades Negras no
Tempo da Escravidão.’’ Tempo 2, no. 3 (1997): 7–33.
———. ‘‘Magia Jêje na Bahia: A Invasão do Calundu do Pasto de Cachoeira.’’ Revista
Brasileira de História 16 (1988). São Paulo.
———. A Morte é uma Festa: Ritos Fúnebres e Revolta Popular no Brasil do Século XIX. São
———. Rebelião Escrava no Brasil: A História do Levante dos Malês (1835). São Paulo:
Brasiliense, 1987.
Reis, João José, and Eduardo Silva. Negociação e Conflito: A Resistência Negra no Brasil
Escravista. São Paulo: Companhia das Letras, 1989.
Revel, Jacques, ed. Jeux d’Échelles: de la Micro-analyse à l’Expérience. Paris: Seuil, 1996.
Ribeiro Filho, J. S. Dicionário Biobibliográfico de Escritores Cariocas (1565–1965). Rio de
Janeiro: Livraria Brasiliana Editora, 1965.
Rios Filho, A. M. ‘‘Evolução Urbana e Arquitetônica do Rio de Janeiro nos Séculos
XVI e XVII (1567–1699).’’ Revista do Instituto Histórico e Geográfico Brasileiro 288
(1970): 22–254.
Rocha, Manoel Ribeiro. Etíope Resgatado, Empenhado Sustentado Corrigido, Instruído e
Libertado. Introduction by Paulo Suess. Petrópolis: Vozes, 1992.
Bibliography | 305
Rodrigues, Cláudia. ‘‘Lugares dos Mortos na Cidade dos Vivos: Tradições e

Transformações Fúnebres na Corte.’’ M.A. thesis, Universidade Federal
Fluminense, Niterói, 1995.
Rodrigues, Nina. Os Africanos no Brasil. Brasília: EdUnB, n.d.
Romero, Sílvio. História da Literatura Brasileira: Contribuições e Estudos Gerais para o
Exato Conhecimento da Literatura Brasileira. 4th ed. organized and with a preface by
Nelson Romero. Rio de Janeiro: Livraria José Olympio Editora, 1949.
Rupert, Arlindo. Expansão Territorial e Absolutismo Estatal (1700–1822): A Igreja no
Brasil, III. Santa Maria: Editora Pallotti, 1988.
———. A Igreja no Brasil III: Expansão Territorial e Absolutismo Estatal (1700–1822). Santa
Maria: Editora Pallotti, 1988.
Russell-Wood, A. J. R. The Black Man in Slavery and Freedom in Colonial Brazil. New
York: St. Martin’s Press, 1982.
———. Fidalgos e Filantropos: A Santa Casa de Misericórdia da Bahia, 1550–1755. Brasília:
EdUnB, 1981.
———. ‘‘Prestige, Power, and Piety in Colonial Brazil: The Third Orders of
Salvador.’’ Hispanic American Historical Review 69 (1989): 61–89.
———. ‘‘United States Scholarly Contributions to the Historiography of Colonial
Brazil.’’ Hispanic American Historical Review 65, no. 4 (1985): 683–723.
Salgado, Graça, ed. Fiscais e Meirinhos: A Administração no Brasil Colonial. Rio de
Janeiro: Nova Fronteira, 1985.
Salvador, Vicente do (Frei). História do Brasil. New edition revised and annotated
by Capistrano de Abreu. São Paulo: Weisz Flog Irmãos, 1918.
Santos, Afonso Carlos Marques dos. No Rascunho da Nação: Inconfidência no Rio de
Janeiro. Rio de Janeiro: Secretaria Municipal da Cultura, 1992.
Santos, Antônio Alves Ferreira (Monsenhor). Arquidiocese de São Sebastião do Rio de
Janeiro. Rio de Janeiro: Tipografia Leuzinger, 1914.
Santos, Corcino Medeiros dos. O Rio de Janeiro e a Conjuntura Atlântica. Rio de
Janeiro: Expressão e Cultura, 1993.
Santos, Juana Elbein dos. Os Nagô e a Morte: Pàde, Asèsè e o Culto Égun na Bahia.
Petrópolis: Vozes, 1976.
Santos, Luiz Gonçalves dos (Padre Perereca). Memórias para Servir à História do
Reino do Brasil. Preface and annotations by Noronha Santos. 2 vols. Rio de
Janeiro: Livraria Editora Zelio Valverde, 1943.
Scarano, Julita. Devoção e Escravidão: A Irmandade de Nossa Senhora do Rosário dos Pretos no
Distrito Diamantino no Século XVIII. São Paulo: Companhia Editora Nacional, 1978.
Schwartz, Stuart B. Burocracia e Sociedade no Brasil Colonial: A Suprema Corte da Bahia e
Seus Juízes. 1609–1751. São Paulo: Editora Perspectiva, 1979.
———. Segredos Internos: Engenhos e Escravos na Sociedade Colonial, 1550–1835. São
Paulo: CNPq, 1988.
Serrão, Joel, ed. Dicionário de História de Portugal. 3 vols. Porto: Livraria
Figuerinhas, n.d.
306 | Bibliography
Silva, Antonio Moraes e. Dicionário de Língua Portuguesa. Lisbon: Oficina de Simão

T. Ferreira, 1889.
Silveira, Américo Bispo da. ‘‘Conferência.’’ Photocopied Manuscript in author’s
possession. N.p. N.d.
Slenes, Robert. ‘‘Malungu ngoma vem! A África Coberta e Descoberta do Brasil.’’
Revista da usp 12 (December–January–February 1991/1992): 48–67.
Soares, José Carlos de Macedo. ‘‘Fontes da História da Igreja Católica no Brasil.’’
Revista do Instituto Histórico e Geográfico Brasileiro 20 (July–September 1953).
Soares, Mariza de Carvalho. ‘‘A Biografia de Ignácio Monte; O Escravo Que Virou
Rei.’’ In Retratos do Império: Trajetórias Individuais no Mundo Português nos Séculos XVI
a XIX, edited by Ronaldo Vainfas et al. Niterói: Eduff, 2006.
———. ‘‘Can Women Guide and Govern Men? Gendering Politics among African
Catholics in Colonial Brazil.’’ In Women and Slavery. Vol. 2: Americas, edited by
Gwyn Campbell, Suzanne Miers, and Joseph C. Miller. Athens: Ohio University
Press, 2008.
———. Devotos da Cor: Identidade Étnica, Religiosidade e Escravidão no Rio de Janeiro, Século
XVIII. Rio de Janeiro: Civilização Brasileira. 2000.
———. ‘‘From Gbe to Yoruba: Ethnic Changes within the Mina Nation in Rio de
Janeiro.’’ In The Yoruba Diaspora in the Atlantic World, edited by Toyin Falola and
Matt Childs. Bloomington: Indiana University Press, 2004.
———. ‘‘Indícios para o Traçado das Rotas Terrestres de Escravos na Baía do
Benim, Século XVIII.’’ In Rotas Atlânticas da Diáspora Africana: Entre a Baía do
Benim e o Rio de Janeiro, edited by Mariza de Carvalho Soares. Niterói: Eduff,
2007.
———. ‘‘O Império de Santo Elesbão na Cidade do Rio de Janeiro, Século XVIII.’’
Topoi 4 (2002): 59–83.
———. ‘‘O Medo da Vida e o Medo da Morte. Um Estudo da Religiosidade
Brasileira.’’ M.A. thesis, Universidade Federal do Rio de Janeiro, 1990.
———. ‘‘O Missionário e o Rei.’’ B.A. thesis, Pontifícia Universidade Católica do Rio
de Janeiro, 1978.
———. ‘‘A ‘Nação’ Que Se Tem e a ‘Terra’ de Onde Se Vem: Categorias de Inserção
Social de Africanos no Império Português, Século XVIII.’’ Estudos Afro-Asiáticos
26 (May–August 2004): 303–30.
———. ‘‘Nos Atalhos da Memória: Monumento a Zumbi.’’ In Cidade Vaidosa: Imagens
Urbanas do Rio de Janeiro, edited by Paulo Knauss. Rio de Janeiro: Sette Letras,
1999.
———, ed. Rotas Atlânticas da Diáspora Africana: Entre a Baía do Benim e o Rio de Janeiro.
Niterói: Eduff, 2007.
Souvenir: Lembranças do Brasil. Introduction by Lygia da Fonseca Fernandes da
Cunha. Rio de Janeiro: sedegra, n.d.
Souza, Laura de Mello e. Declassificados do Ouro: A Pobreza Mineira no Século XVIII. Rio
de Janeiro: Graal, 1986.
Bibliography | 307
Spix, J. B., and C. F. P. Martius. Viagem ao Brasil, 1817–1820. São Paulo:

Melhoramentos, 1975.
Suess, Paulo. A Conquista Espiritual da América Espanhola: 200 Documentos—Século XVI.
Introduction by Paulo Suess. Petrópolis: Vozes, 1992.
Vainfas, Ronaldo. A Heresia dos Índios: Catolicismo e Rebeldia no Brasil Colonial. São
———. ‘‘Moralidades Brasílicas: Deleites Sexuais e Linguagem Erotica na Sociedade
Escravista.’’ In História da Vida Privada no Brasil I, edited by Laura de Mello e
Souza. São Paulo: Companhia das Letras, 1997.
Vainfas, Ronaldo, and Marina de Mello e Souza. ‘‘Catolicização e Poder no Tempo
do Tráfico: O Reino do Congo da Conversão Coroada ao Movimento
Antoniano, Séculos XV–XVIII.’’ Tempo 3, no. 6 (1998): 95–118.
Varnhagen, Francisco Adolfo de. História Geral do Brasil: Antes da sua Separação e
Independência de Portugal. Revision and notes by J. Capistrano de Abreu and
Rodolfo Garcia. 3 vols. São Paulo: Edições Melhoramentos, n.d.
Velloso, Mônica Pimenta. ‘‘As Tias Baianas Tomam Conta do Pedaço . . . Espaço e
Identidade Cultural no Rio de Janeiro.’’ Estudos Históricos 6, file ‘‘Cultura e Povo’’
(1990): 207–28.
Verger, Pierre. Dieux d’Afrique: Culte des Orishas et Vodouns à l’Ancienne Côte des Esclaves en
Afrique et à Bahia, la Baie de Tous les Saints au Brésil. Preface by Théodore Monod and
Roger Bastide. Photographs by the author. Paris: Éditions Revue Noire, 1995.
———. Fluxo e Refluxo do Tráfico de Escravos entre o Golfo do Benin e a Bahia de Todos os
Santos dos séculos XVIII a XIX. São Paulo: Editora Corrupio, 1987.
———.Os Orixás: Deuses Iorubás na África e no Novo Mundo. São Paulo: Círculo do Livro,
1981.
Vieira, Antônio (Padre). Sermões. Authorized edition adapted from the o≈cial
orthography, with summaries preceding each chapter, and with textual
revisions and corrections by Frederico Ozanam Pessoas de Barros under the
supervision of Padre Antônio Charbel, S.D.B. [Salesians of Don Bosco] (of the
Instituto Teleológico Pio XI) and of Prof. A. Della Nina. 24 vols. São Paulo:
Editora das Américas, 1957.
Villalta, Luiz Carlos. ‘‘O Que Se Fala e o Que Se Lê: Língua, Instrução e Leitura.’’
In História da Vida Privada. Vol. 1: O Cotidiano da Vida Privada na América Portuguesa,
edited by Fernando A Novais. São Paulo: Companhia das Letras, 1997.
Vovelle, Michel. ‘‘Iconografia e História das Mentalidades.’’ In Ideologias e
Mentalidades. São Paulo: Brasiliense, 1991.
Wolf, Kenneth Baxter. ‘‘The ‘Moors’ of West Africa and the Beginnings of the
Portuguese Slave Trade.’’ Journal of Medieval and Renaissance Studies 24, no. 3
(1994): 449–69.
Zurara, Gomes Eanes de. Crônicas de Guiné. Based on the Paris manuscript,
modernized; introduction, notes, commentary, and glossary by José de
Bragança. Barcelos, Portugal: Livraria Civilização Editora, 1973.
index
Page numbers in italics refer to illustrations; ‘‘t’’ signifies table.
Afonso I (Kongo), Dom, 28, 254n31, toriography of, 93, 254n28; king of,
282n36 139; in Mahi Manuscript, 191, 203–5;
Afonso V (Portugal), Dom, 37, 255– marriage and, 97; Mina and, 6, 127–
56n56; in Mahi Manuscript, 207 28, 165, 170–77, 182, 187, 213–14;
Agassiz, Elizabeth, 10, 14, 266n52 obituaries for, 130t; population of, 77–
Agassiz, Louis, 10, 266n52 79, 96, 98, 131, 133, 136, 263n28; slave
Agonli, 66; provenience group of, 178, baptism and, 54t, 58t, 82, 84, 99t
185, 191 Angola slave trade, 13, 30, 41, 47, 53, 76,
Ajudá (Dahomey), 5, 55, 227 255n47; baptism and, 258n35; colo-
Akim, 47 nial documentation of, 50–51, 81, 94–
Alfaias, 156, 175. See also Folias 95; inadequacy of, 58–59, 259n55;
Algarve, 22 voyages of, 256n7
Allada, kingdom of, 45–46, 55 Ashanti: historiography of, 34, 255n51;
Ambaca provenience group, 84, 134t kingdom of, 47; as slaves in Brazil
Angola, 28; bishop of, 264n36; distance and, 92
to Brazil from, 264n7; Dutch occupa- Avis dynasty, 37
tion of, 42–45; group baptism in, Azevedo, João Lúcio de, 32–33
258n35; historiography of, 33–37;
maps of, 76; Portuguese and, 27, 32; Bahia: Africans in, 4–5, 34, 54, 64, 95–
smallpox epidemic in, 48 96, 152, 205, 250n15; black lay broth-
Angola provenience group, 84, 128; erhoods in, 276–77n50, 277n56; can-
alleged paganism in, 191; Bahia, domblé in, 1, 4, 35, 85, 250n10;
277n56; brotherhood expulsed from, Church of Nossa Senhora da Con-
114; burials of, 134t; colonial docu- ceição da Praia in, 232; indigenous
mentation of, 39, 62, 69–70, 78, 90, slavery in, 263n25; pardos in, 74; slave
133; church construction and, 131; his- baptism in, 262n8; slave marriage in,
310 | Index
Bahia (cont.) 114–15, 147, 267nn7–8; purchases of

95–96; slaves to Minas Gerais and, freedom and, 212; rhetoric and,
223–24; slave trade and, 34–35, 48, 57– 280n26; social significance of, 11–12,
58, 254n39, 264n52; synod of 1707 in, 64–65, 93, 146, 148, 150. See also Folias
118–19, 261n5. See also Jêje provenience Black Men’s Regiment, 184, 231, 279n8
group; Ladá provenience group; Nagô Boçal, 134t
provenience group; Tobacco trade Bomsucesso, Nossa Senhora do, 267n7
Baltazar, King, 140 Book of Baptisms of Whites, 73–74,
Baptism: ages of, 262n8; freed slaves 251n25
and, 73–74, 262n11; slaves and, 9t, 81, Book of Black Captives, 73–74, 224,
242t–47t, 258n35, 264n34 251n25, 262nn7–8, 264n34
Baptism records, historical significance Book of the Freed, 73
of, 7–8, 52–53, 67–68, 80, 269n39 Boschi, Caio, 12
Baroque Catholicism in Brazil, 119, 158 Bosman, William, 46–47
Barros, João de, 25 Boxer, Charles R., 33, 59–60, 93
Barth, Fredrik, 86, 235 Brancos, 68, 193, 260n68, 282n39. See also
Bartolomé, Miguel Alberto, 87 Livro de Batismo de Brancos
Bastide, Roger, 35–36, 85, 91–92 Braudel, Fernand, 11, 251nn29–30
Batuques, 142, 205, 221, 281n29 Brotherhood of Boa Morte, Bahia,
Begging, 199–200, 275n23 277n56
Benedictines, 267n4. See also Desterro Brotherhood of Boa Morte of Pardo Men,
Malheiros, Dom Frei Antônio do 68, 121, 125, 158, 269n29, 270n48
Benguela: French attack on, 268n15; as Brotherhood of Bom Jesus do Cálice,
slave trade hub, 54, 76, 79, 81–82, 144–45, 272n67
260n67; spelling of, 264n33 Brotherhood of Our Lady of Assunçáo
Benguela provenience group, 93, 130, and Boa Morte, 121
134t, 263n29, 270n43; burials of, 134t; Brotherhood of Our Lady of Lampadosa,
in Mahi Manuscript, 204–5; slave bap- 64, 121, 131, 137, 144, 272n67
tism and, 54t, 58t, 84 Brotherhood of Our Lady of Lapa dos
Benin, Bight of, 3–4, 9–10; cultural diver- Mercadores, 64
sity of, 89, 226; European slave trade Brotherhood of Our Lady of the Con-
in, 45, 47; in Mahi Manuscript, 218– ceição of Pardo Men, 68, 116, 268n25,
20; migrations and, 235; Portuguese 269n29
alliances and, 26–27, 34, 92; slaves to Brotherhood of Our Lady of the Rosário,
Minas Gerais and, 49 Recife, 139, 142–43
Black brotherhoods in Brazil, 172–73; Brotherhood of Our Lady of the Rosário
approval process and, 152, 181, and São Benedito of Black Men, 68,
278n70; authorities vs., 267n6; beg- 114, 120, 137; funeral practice of, 125;
ging and, 154, 158; church con- supports annexation, 144
struction and, 117, 120–21, 177, Brotherhood of Our Lady of the Rosário
261n75, 268n25; cloaks and, 275n27; do Serro, Minas Gerais, 276n50
expulsion e√orts and, 62; funerals of Brotherhood of Our Lady of the Rosário
members of, 126–27, 137, 160, 195, of Black Men, 68, 114, 116, 140; Angola
276n50; identity in, 84, 277nn56–57; blacks in, 6, 62, 130–31, 139, 176; on
public processions of, 11, 100, 108–9, death and resurrection, 158
Index | 311
Brotherhood of Our Lady of the Rosário Martírios, Bahia, 64, 152, 276–77n50,
of the Monastery of São Domingos, 277n56
Lisbon, 138, 140, 148, 156, 165–66; Brotherhood of the Menino Jesus, 64,
begging and, 154; charity and, 276n47; 130–31, 144, 173, 261n75, 272n67
female vendors and, 266n58 Brotherhood of the Rosário, creoles in,
Brotherhood of Santo Antônio da 131
Mouraria: at Church of the Rosário, Brotherhood of the Rosário, Recife, 142–
131, 137; Mina in, 64, 120–21, 144–45, 43
173, 261n75 Brotherhood of the Rosário, Rio Grande,
Brotherhood of Santo Elesbão and Santa 107
Efigênia, 5, 63–66, 148; approval of, Brotherhood of the Rosário, Salvador,
151–52, 181–82; authority structure of, 139
165–66; church and, 64; conflict Brotherhood of the Santa Casa de Mis-
among provenience groups, 6, 97–98, ericórdia, 149, 280n21
176–78, 185, 203, 213–14; creation of, Brotherhood of the Santíssimo Sacra-
90, 121, 131, 265n43; creoles in, 237; mento, 115–16, 147
elections for, 155; folias of, 179; Brotherhoods in Brazil, 147, 212; begging
funerals and, 127, 160–61, 169–70, and, 154–55; churches and, 116; lin-
273n6; income and, 167; irmãos eage of, 148, 280n21; merchant con-
criadores of, 171; marriage and, 162; nections with, 63–64; organization of,
professions in, 231; saint’s day festival 115–16, 166; Portuguese court re-
and, 154; white men in, 172, 174; wills created by, 138; processions and,
of members of, 230; women in, 214. 281n28
See also Brotherhood of Santo Elesbão Brotherhoods in Portugal, 115, 148,
and Santa Efigênia; Church of Santo 266n58, 267–68n12; Africans and
Elesbão and Santa Efigênia; descendants and, 172–73
Compromisso Brotherhoods of the Rosário, Minas
Brotherhood of Santo Pedro Gonçalves, Gerais, 93
64 Boxer, Charles R., 93
Brotherhood of São Benedito, 68, 144; Burguière, André, 179, 278n63
Angola in, 62; at Church of São Burials: of Africans, 128, 132, 134–36,
Sebastião, 114, 121 195, 203, 270n48; brotherhoods and,
Brotherhood of São Domingos, 64, 121, 137, 160–62, 169; of Catholics, 159–
225; Antônio Luiz Soares and, 232; 60; charitable, 126–27; freed slaves
church and, 144, 241; folia of, 140; and, 224, 134t–35t; locations of, 129–
funeral practice of, 125; Guinea in, 62, 31, 270n44, 275n35; obituaries and,
114, 136 269n39, 271n49; processions for, 107,
Brotherhood of São Domingos of the 117, 128, 156; Santa Casa de Misericór-
Convent of São Francisco, Salvador, dia and, 159; slaves and, 134t; slave
136 women and, 133; transportation of
Brotherhood of São José, 68, 116 corpse for, 125, 159, 161–62, 270n45
Brotherhood of Senhor Bom Jesus das
Necessidades e Redenção dos Homens Cabido, 62, 120–21, 131, 269n29
Pretos, Bahia, 277n56 Cabinda, port of, 79, 264n33
Brotherhood of Senhor Bom Jesus dos Cabinda provenience group, 9, 81, 93
312 | Index
Cabo Verde: Cacheu in, 121; French attack Casas de dar fortuna, 221
on, 268n15; as part of Guinea, 76; as Cascudo, Luís da Câmara, 139, 271n55
slave trade hub, 6, 26, 28, 30, 41, 66, Cassenge provenience group, 81
80 Catholic Church in Africa: baptisms of
Cabo Verde provenience group, 84, 93, Africans into, 81, 258n35, 265n38;
134; apportionment contract and, 56; bishopric in Kongo and, 30, 264n36;
in Brotherhood of Santo Elesbão and influence of, in Mahi Manuscript, 218–
Santa Efigênia, 90, 130, 165, 171–74, 20; missionary e√orts of, 27–28,
178, 185, 265n43; burials for, 134t; 254nn27–28; Santa Casa de Midericór-
obituaries for, 130t; slave baptism and, dia and, 264n33
54, 58t, 84 Cemeteries in Rio de Janeiro, 125–31,
Cabras, 72, 134, 165, 170; Brotherhood of 133, 137, 144–45, 159, 269n29; of
Santo Elesbão and Santa Efigênia and, Church of Santo Elesbão and Santa
277n52; burials of, 134t Efigênia, 161; public, 270n44, 275n35
Cacheo, 51, 264n33. See also Cacheu prove- Ceuta, 20, 22; in Mahi Manuscript, 207
nience group Chamba provenience group, 224, 287n6
Cacheu: as slave trade hub, 41, 81, 121; Church of Candelária, 81, 113, 120; slave
spelling of, 51, 264n33 burials at, 126, 129
Cacheu Company, 46, 51 Church of Our Lady of Lampadosa, 98,
Cacheu Council, 51 145, 270n46, 270–71n48, 271n59;
Cacheu provenience group, 51, 264n33; black lay brotherhood festivals at, 139–
burials for, 134t; slave baptism and, 40; Debret describes funeral at, 128
54t, 58t Church of Our Lady of Parto, 128
Cacheu River, 41 Church of Our Lady of the Rosário. See
Calundus, 142, 205, 221, 281n29 Church of the Rosário
Candelária parish, 53, 63, 78–79, 126, Church of Our Lady of the Rosário and
128, 228; house distribution of, 124; São Benedito of Black Men. See Church
slave baptisms and, 9t, 71t, 94t, 100t, of the Rosário
242t, 245t; slave burials and, 126; slave Church of Santa Rita, 126
obituaries from, 130t; slave population Church of Santo Elesbão and Santa
of, 77t–78t, 94, 98–100 Efigênia, 1, 5–6, 161, 167, 289n26,
Candomblé, 1, 4, 35, 85 291n35; conflict among provenience
Candomblé, da Bahia, Le (Rite Nagô) (Bas- groups of, 176–77; congregations of,
tide), 85 150–51; construction of, 64
Cão, Diogo, 27 Church of São Benedito, 131, 268n14,
Cape Bojador, 22, 24 268n25
Cape Coast, 45, 257n20 Church of São Domingos, 98, 117–18,
Cape of Good Hope, 29 121, 130, 145; annexation plan for, 144;
Cape Verde. See Cabo Verde Antônio Luiz Soares and, 232; Brother-
Carmo, Order of, 116, 266n1, 267n4 hood of Santo Elesbão and Santa
Cartography, African, 20–21, 28–31, 33, Efigênia and, 64, 153, 161, 180, 185,
39, 76, 80 233; folias at, 140; Mina and, 132, 137,
Carvalho e Mello, Sebastião José de (Mar- 260n71; slave burials at, 126, 128–31,
quis of Pombal), 180; control of broth- 270n48
erhoods by, 181 Church of São José, 114, 116, 120, 266n1
Index | 313
Church of São Sebastião, 267n4; black Donations; Brotherhood of Santo Ele-

brotherhoods of, 121, 140; decline of, sbão and Santa Efigênia
113, 115; Episcopal See and, 114, 120; Compromisso, Devotion to the Souls of Pur-
pardo brotherhoods of, 62, 114, 269n29 gatory, 201, 250n17; Brotherhood of
Church of the Rosário, 145, 197; cemetery the Church of Santo Elesbão and Santa
of, 126–27, 129–30; construction of, Efigênia and, 202–5, 209; discipline
117–18, 268n25; Debret account of, 128; and, 208, 210, 284n54; folias and, 213
inauguration of, 120, 268n14; in Mahi Compromisso, Fraternity of Our Lady of
Manuscript, 213, 215; provenience Remedies, 184, 199, 250n17; approval
groups, 6, 64, 121, 131, 137; slaves and, and, 214–15; charity in, 216; member-
98; supports annexation, 144 ship and, 282n38
Church of the Rosário, Recife, 142–43, Compromisso, Mahi Congregation, 175,
277n57 250n17; Africa in, 235; authority struc-
Church of the Sé, 62–63; relocation con- ture in, 209–10; compromisso of Santo
troversy of, 114, 120. See also Church of Elesbão and Santa Efigênia and, 211;
São Sebastião lack of definition in, 204–5
Cobu: historiography of, 80, 227–28, Conceição, Victoria Correa da, 188–89,
288n8; kingdom of, 234, 244 288n18; baptism of, 228; marriage of,
Cobu provenience group: 224, 226–27, 234; queen, Mahi Congregation and,
236, 238, 290n33; Antônio Luiz Soares 190, 196; statutes, Fraternity of Our
and, 232; Minas Gerais and, 93, 228 Lady of Remedies and, 199–200;
Color, in parish records, 68 witchcraft and, 191, 198, 202
Compromisso, 116, 150, 212 250n17; black Congo River, 37
brotherhoods and, 163; Brotherhood Conquest, Spanish, 87
of the Rosário of the Monastery of São Conselho Ultramarino. See Oversea
Domingos (Lisbon), 166 Council
Compromisso, Brotherhood of the Church Constituições Primeiras do Arcebispado da
of Santo Elesbão and Santa Efigênia, Bahia, 101, 261–62n5; baptism regula-
5–6, 164, 188, 250n17, 275n22, 278n3, tions in, 7, 69–70, 83, 265n42; black lay
280n19; approval process and, 150–53, brotherhoods and, 154–55; control over
181–82; authority positions in, 167; churches and, 144; festival behavior in,
changes to, 165–66, 169–70, 179–80; 157–58, 284n60; folias and, 213; funeral
charity vs. devotion in, 148–50; chil- processions and, 156, 158; lack of defi-
dren in, 171; death rites and, 160–61; nition in, 204; languages and, 282n44;
Devotion to the Souls of Purgatory and, publication of, 119; Rio de Janeiro and,
202–5, 209; discipline and, 172, 120, 157–58; slave marriage in, 95
284n54; elections and, 183, 275n25; Cordeiro, Gonçalo: Mahi Manuscript
esmolas and, 168, 182, 188–89, 276n46; and, 184, 188, 197–98, 215–17; profes-
folias and, 175; freedom not discussed sion of, 231
in, 163; funeral cortege and, 156; Cornevin, Robert, 225, 227, 235, 288n13
masses regulated in, 169–70, 284n55; Costa, Dom Rodrigo, 49
membership and, 174, 279n14; mutiny Costa Mimozo, Pedro da, 177, 185
and, 197; processions and, 157–58, Council of Trent, 118–19, 148, 261n5
186; white men in, 165; widows and Coura, 226–28; historiography of,
women in, 162, 166, 176–77. See also 288n10
314 | Index
Coura provenience group, 84, 234, 236, 165–68; compromisso and, 182, 189;
238, 263–64n31, 292n42; Minas folias and, 179, 186, 201; funerary from,
Gerais and, 80, 224, 226; region, 228; 190; penalty and, 208; public and, 154–
slave baptism and, 84 55, 158; for saint’s day festival, 156;
Craberá provenience group, 224 women members and, 170, 198
Creoles, 74, 133, 190, 262n13; African- Dum Diversas, 24
born, 69, 116; Angola, 98, 114; in Dutch: Portuguese holdings and, 42–44,
Brotherhood of the Rosário and, 131; 46; forts of, sold to England, 257n20
burials of, 134t; European-derived, 93;
maternal derivation of, 70–72; Mina Egypt, 20
brotherhood and, 165, 170–71, 173, Elias, Norbert, 11, 147, 280n21
177, 182, 203, 237; obituaries of, 130t Emancipation at baptism, 7, 73
Crioulos. See Creoles Empire, brotherhood, 138
Crispim and Crispiano, 129, 270n46 Endogamy, African, 96–97, 170, 277n51
Equatorial Africa. See Terra dos negros
Dagomé. See Dahomey entries (Zurara)
Dahomey (Herskovits), 35 Erizipela, 189, 279n12
Dahomey, kingdom, 3, 234–35, 259n60, Esmeraldo de Situ Orbis (Pereira), 25
261n73; Brazilian tobacco and, 61; Por- Estado Imperial, brotherhood, 138
tuguese fort in, 55–56; slave trade in, Ethiopia, 23, 27, 31; European concep-
5, 35, 47, 53, 65, 81, 227, 288n14 tions of, 20; Santo Elesbão and, 21, 174
Dahomey and the Slave Trade (Polanyi), 35, Exhibition of Brazilian History (1882), 2
65 Expansionism, Portuguese, in Africa:
Dahomey provenience group, 34, 54, 93, baptisms of Africans and, 81, 258n35,
236; Bahia, 205; Brotherhood of Santo 265n38; conversion attempts and, 24,
Elesbão and Santa Efigênia and, 177– 27–28, 254nn27–28; early exploration
78, 185, 187, 197, 217, 220; histo- and trade and, 20, 22, 25–29; histo-
riography of, 35 riography of, 31–39; House of Guinea
Debret, Jean-Baptiste, 10, 14, 103, 105, and, 23; House of Slaves and, 26;
107, 270n45; burial description by, 128; Kongo bishopric and, 30, 264n36
folia description by, 141, 211–12
Desterro Malheiros, Dom Frei Antônio Fanti-Ashanti. See Ashanti
do, 182, 264n36, 278n71, 284n59 Felipe III, 40
Devil, 219–20, 292n42 Festivals. See Folias
Devotion to the Almas of Purgatório. See Fez, 207, 283n50; in Mahi Manuscript,
Devotion to the Souls of Purgatory, 207
Mahi Finance Council, Lisbon, 44
Devotion to the Souls of Purgatory, Mahi, Florentino, Manolo, 48, 94; household
6, 163, 186; Christian doctrine in, 208– inventories by, 124
9; Mahi Congregation and, 198; in Folias, 108–10, 140, 156, 213, 237;
Mahi Manuscript, 184; provenience in, accoutrements for, 139, 175, 271n54;
210–11, 213; strategy of, 201 alas of, 186; Debret on, 141, 211–12;
Donations, Brotherhood of Santo Elesbão gender in, 142; hierarchical structure
and Santa Efigênia, 175, 188; annual, of, 138, 193; missionary tradition and,
217; association of, with positions, 213; provenience in, 176, 178, 185, 235;
Index | 315
public donations for, 106–7, 167, 189, Gold Coast: English slave trade on, 92;
197, 201, 205, 237; strategy of, 179–80 European expansionism and, 45–46
Folklore, 39, 139, 179 Golden Age of Brazil, The (Boxer), 93
Fon speakers, 219; historiography of, 35 Gold mines in Brazil, 48–49; Mina slaves
Franciscans, 126, 267n4 in, 59, 259n55; slave trade and gold
Fraternity of Nossa Senhora dos from, 57, 60, 117–18, 257n30, 259n60
Remédios. See Fraternity of Our Lady of Goulart, Maurício, 33, 52, 57
Remedies, Mahi Guadalupe, Dom Antônio de, Bishop,
Fraternity of Our Lady of Remedies, Brotherhood of Santo Elesbão and
Mahi, 6, 163, 186, 281n32; concept of Santa Efigênia and, 157–58, 167, 171,
charity in, 216; Devotion to the Souls of 181
Purgatory and, 184; professions in, Guimarães, Jerônimo Lobo, 56
230–31, 237; statutes and Mahi Con- Guinea: European conceptions of, 19–21,
gregation crisis and, 198–200, 217 23, 39; historiography of, 31–33, 36,
Freedom, 147, 228, 230, 288n18; man- 259n55; importance of, to Portuguese
umission letters and, 225, 233–34, Empire, 26–30; in Mahi Manuscript,
236, 289n21, 284n59; purchasing, 207, 286n75; moors and, 24; trade
163–64, 201, 211–12, 228, 232, with, 22, 25. See also High Guinea
289n21; records of, 224 Guinea-Bissau. See High Guinea
Freed slaves: baptism of, 73–74, 262n11; Guinea provenience group, 9, 133, 136,
burials of, 224, 134t–35t; professions 254–55n39, 264n29; brotherhoods in
of, 129, 225, 228–29, 237, 289n21; Brazil and, 114, 131, 267n5; burials of,
trade associations with, 230; wills of, 134t; colonial documentation of, 29–30,
137, 160, 231–32, 284n55, 291n34 37, 39, 62, 68–70, 75–77, 263nn27–29;
Freyre, Gilberto, 35 marriage in, 95–97; obituaries of, 130t;
Fuam provenience group, 224 population of, 77t, 78–79, 84; slave bap-
Fula provenience group, 34 tism and, 54t, 58t, 82, 84
Funeral practices, 125–32, 270n45; Guinea River, 30
masses and, 160–61, 169–70
Funeral records, gender imbalance in, Hammocks, 104, 122–23, 126, 133; funer-
132–33 ary use of, 159, 161–62, 270n45
Heathens, 2, 9, 24; burial of converted,
Ganguela provenience group, 84; burials 129; ‘‘nations’’ of, 37, 69, 74–76, 80;
of, 134t as pagans, 15; principal ethnic catego-
Gbe language group, 4; in Brazil, 14, 89– ries of, 83–84
90, 92, 250n15; Mahi subgroup of, Henriques Regiment. See Regimento dos
287n4 Homens Pretos
General Company of the State of Brazil, Henry the Navigator, 22–25, 254n23; in
45 Mahi Manuscript, 207
Gentio, 15, 71; indigenous, 24, 40, 69; Herodotus, 19
slave designations of, 37, 70, 77–79. Herskovits, Melville, 16, 35, 85; on cul-
See also Heathens tural traits, 87, 91–92
Ghana: borders of, 257n20; European High Guinea, 33, 41, 46, 51
forts in, 26, 45; Portuguese arrival in, Histoire du Dahomey avec 10 Cartes, 1 Aroquis
25; slaves from, 92 et 35 Photographies (Cornevin), 227
316 | Index
Holy House of Mercy, 126, 266n1; broth- João I (Kongo), Dom, 28

erhoods and, 115; cemetery of, 126, João I (Portugal), 207
130t, 161, 203; funeral monopoly and, João II (Portugal), Dom, 25, 28, 37
125, 159, 195, 280n25 João III (Portugal), Dom, 154, 261n1
Holy O≈ce, 115, 143, 272n63 João V (Portugal), Dom, 62
Household inventories, 123–25 José (Portugal), Dom, 171, 181
House of Bragança, 23, 25
House of Guinea, 23 King Baltazar, 140
House of Slaves, Lisbon, 26 King of Kongo, Brazil, 139–40
Houses, Rio de Janeiro, 117, 147; con- Kinship, 128
struction of, 122–23; distribution of, 124 Kongo: 31, 264n33; conquest of, 25; con-
Hueda, kingdom of, 45–47, 53, 55, 234 versions in, 28, 254n28; historiogra-
phy of, 34–37, 254–55n39; kings of,
Iano: in Mahi Manuscript, 279n5; prove- 271n54; population subgroups (colo-
nience group of, 178, 185–86 nial) of, 81; Portuguese arrival in, 27;
Ilustração, 207 Portuguese bishopric in, 30, 33; slave
Imperial State of Santo Elesbão, 174, 201; trade in, 27–28, 41–42, 47, 53,
approval of, 229; creation of, 178; 268n35. See also Angola; Afonso I
empress of, 196; funds for, 189; hier- (Kongo), Dom; João I (Kongo), Dom;
archical structure of, 175, 193; kings’ Kongo slaves
requirements in, 183; Mahi court of, Kongo slaves: baptism of, 54t, 58t; 81–
197; strategy of, 186–87 82, 84; burials of, 134t; obituaries of,
Império do Divino, 140 130t; population of, 83–84, 93
India, 51, 83, 255n49, 265n40; House of,
35. See also West India Company Ladá provenience group, 224, 287n2
Indians, 24, 48, 75–76, 93, 163, 256n1; Ladino, 134t
burials of, in Rio de Janeiro, 134t; labor Lagos, Nigeria, 227
of, in Rio de Janeiro, 40–42, 104, 119– Lagos, Portugal, 23–24
20, 134, 263n17 Lavradio, Marquês de, 143–44, 273n68
Infante Dom Henrique, 22–25, 254n23; Law, Robin, 225
in Mahi Manuscript, 207 Livro de Batismo de Brancos, 73–74, 251n25
Innocent XI, Pope, 114 Livro de Batismo de Pretos Cativos, 73–
Inquisition, 204, 206 74, 224, 251n25, 262nn7–8, 264n34
Inter Coetera, 24, 261n1 Livro de Óbitos, 224
Islam in Africa, 3–4, 23, 27, 38; Muslim Livro dos Forros, 73
slaves in Rio de Janeiro and, 121 Loanda. See Luanda
Luanda, 28, 79, 82, 264n33; Dutch
Jaquem provenience group, 224, 287n7 occupation of, 43–44; São Paulo de
Jêje provenience group, 87, 89; Bahia Luanda founded, 30, 264n33
and, 4–5, 35, 54, 64, 152, 250n15; Luanda provenience group, 54, 81, 84;
brotherhood, 276–77n50, 277n56; slave baptism in, 54t, 58t, 84, 265n38
endogamy in, 95–96. See also Gbe lan-
guage group Macumba, 36
Jesuits, 40, 42, 47–48, 51–52, 95, Madagascar, 29, 54, 83. See also São
263n25; College in Rio de Janeiro, 113 Lourenço provenience group
Index | 317
Mahi: historiography of, 227–28; king- Mattoso, Katia, 95–96

dom of, 2, 4–5, 66, 234–35, 287n4; Mauro, Frédéric, 33–34
Mahi Manuscript and, 178; Mina Con- Mello e Castro, Martinho de, 142–43
gregation and, 185; provenience group Mendonça, Father Dr. Lourenço de, 42
of, 224, 249n8, 265n51, 287n4; slaves Mesa de Consciência e Ordens, Lisbon, 6;
in, 92 brotherhoods in Brazil and, 115, 150–
Mahi Catholic Organizations, 173, 184, 53, 171, 181, 203, 229, 274n11; Por-
195–96, 209, 215, 217. See also Devotion tuguese empire and, 118–19, 274n12
to the Souls of Purgatory, Mahi; Frater- Mestiços, 41, 277n52
nity of Our Lady of Remedies, Mahi; Mina: burials of, 131–32, 134t; childbirth
Mahi Congregation and, 162; freed, 225; ‘‘general tongue’’
Mahi Congregation, 2–3, 6, 12, 16, 150, of, 5, 66, 90, 265n46; marriage of, 97;
238; Brotherhood of Santo Elesbão ‘‘nation’’ of, 3, 255n51; obituaries of,
and Santa Efigênia and, 214, 216; cre- 130t; paganism and, 220; provenience
ation of, 177, 185; creoles in, 237; group of, 89, 92, 95, 213; slave bap-
focus of, 195, 201, 216; folias and, 180, tism of, 54t, 58t, 77–78t, 81–82, 84,
197, 237; kings of, 183; leadership of, 93–94t, 99t–100t; slave population of,
184; names of members of, 291n41; 8, 13–14, 59, 98, 230
pano da costa and, 189, 219–20; profes- Mina Coast: Flemish presence on, 26;
sions of members of, 225, 228–29, Fort of São Jorge on, 25–26, 28;
231, 237, 290n28; provenience in, 173, Guinea and, 30, 39; Mahi Congrega-
178, 203, 209, 213; purchasing free- tion statutes and, 3–4, 238; Portuguese
dom in, 175; succession conflict of, on, 21; slave trade between Rio de Jan-
188–96, 199–200, 217, 289–90n27; eiro and, 13, 15, 57–58, 65, 94–95,
uno≈cial status in, 198; wills of mem- 259n55, 260n67, 264n52
bers of, 230 Mina Congregation, 173, 185; Daho-
Mahi Manuscript, 2, 6, 15, 178, 238; means in, 185, 187, 197, 217, 220;
authors of, 183; Benin in, 218–20; con- provenience groups and, 177–78, 185,
text of, 188; paganism in, 204–5; 190, 264n32
provenience in, 203, 216; strategy of, Minas Gerais: baptism of Victoria Correia
217; structure of, 3, 184–85, 187, 190, da Conceição, 228; Brotherhoods of
201; women in, 191–92, 198 the Rosário and, 93; Rio de Janeiro
Maí. See Mahi and, 54; slave provenience of, 224,
Maki. See Mahi 226; slave trade routes for, 48–49, 51–
Malê: provenience group of, 93; uprising 52, 92, 259n55
of, 14 Mondobi provenience group, 54
Mali, 20 Monjolo provenience group, 84, 134t
Mande, 80 Monte, Capt. Ignacio Gonçalves do: bap-
Manoel, Ignacio, 73, 81–82 tism of, 228; death of, 188, 289n27;
Maqui. See Mahi Mahi Congregation and, 177, 185, 187;
Maranhão, 34, 75, 89, 92, 118 will of, 224–25, 229–30, 233. See also
Maria I, Dona (Portugal), 199 Conceição, Victoria Correa da
Martinique, 43 Monumenta Missionária Africana, 83
Massangano provenience group, 84, 134t Morocco, 20, 257n47; in Mahi Manu-
Mattos, Hebe, 13 script, 207
318 | Index
Mozambique, as slave trade hub, 6, 28, Outline of Dahomean Religious Belief

33, 51, 54, 66, 83, 85 (Herskovits), 35
Mozambique provenience group, 90, 93, Ovem, Benin, 219
165, 171, 173–74, 178, 265n43; burials Oversea Council: appeals to, 49–51; cre-
of, 128, 134t; ‘‘nation’’ of, 254–55n39; ation of, 43; slave traders and, 56
slave baptism in, 54t, 58t
Mulatos, 68; burials of, 134t Padroado, 67, 118, 150, 261n1
Muxicongo provenience group, 134t Paganism, 3, 15, 32, 143, 152, 191, 202;
brotherhood and, 206; dolls and,
Nação. See ‘‘Nation’’ 292n42; Mina and, 220; ‘‘nations’’
Nagô provenience group, 4, 87, 93, 224; and, 75, 203; poorly defined, 204–5;
Bahia and, 95, 97, 250n9, 277n56; his- salvation and, 253n8
toriography of, 31, 36, 254–55n39; Palanquim, 50, 257n27
language of, 14, 89 Panos da costa, 121–22, 189, 219–20
‘‘Nation’’: brotherhoods and, 171–74; Papal bulls, 24
colonial documentation of, 37, 39, 74– Pardo brotherhoods, 68, 116, 146, 212,
76, 78–80, 84, 264n29, 292n42; coun- 269n29; Church of São Sebastião and,
try vs. territory and, 234, 291n38; cul- 62, 114; financial assistance to, 120–21,
tural traits of, 91; endogamy and, 96– 268n25; in Minas Gerais, 132; Our
97, 170; ethnicity and, 88, 90, 224, 226, Lady of Boa Morte of Pardo Men and,
235–36, 238; Mina identity and, 89 125
Naturalidade, 171, 236–37 Pardos, 13, 72, 74, 130; Brotherhood of
Negros, 26, 29, 32–33, 36, 39, 50; Rio de Santo Elesbão and Santa Efigênia and,
Janeiro documentation of, 68, 76, 149, 166, 172, 177; burials of, 125, 134t;
262n10; smallpox and, 48. See also obituaries and, 130t; population of,
Guinea; Pretos; Terra dos negros (Zurara) 133, 263n27; in Portugal, 143
Negros da terra, 69, 263n25. See also Indians Parentes, 195, 202, 235, 270n43, 282n39,
Niger River, 20, 27 289n24; Debret account of, 128; folias
Noronha, Aires Saldanha e Albuquerque and, 178; Ignacio Gonçalves do Monte
Coutinho Matos e, 119–20 and, 229, 233; religious family of,
Nossa Senhora da Ajuda, 115 212–13
Nossa Senhora da Lapa, 64 Parto, Nossa Senhora do, 128, 267n4
Nubia, origin of Santa Efigênia and, 21 Patachos, 48, 264n54
Nupe, 4; provenience group of, 93, 205 Patrícios. See Parentes
Pereira, Duarte Pacheco, 25–26
Obaluaê-Xapanã, 4. See also Smallpox Pina, Rui de, 25, 28, 31
Ómon, Benin, 219 Pizarro, Monsenhor, 13, 41, 153, 256n4,
Orange River, 32 266n4, 267n3, 274n17
Ordenações Filipinas, 7, 209 Polanyi, Karl, 53, 250n2, 255n53,
Ordinance Company of Freed Blacks, 116 257n18, 258n6, 258n40, 261n77; on
Our Lady of Remedies, 6, 151 Dahomean slave trade, 55; on Gold
Our Lady of the Conception, 116, 228, Coast, 46; Herskovits’s influence on,
267n4 35; on slave-trading peoples, 65
Our Lady of the Rosário, 68, 116 Portuguese Empire, 11, 16, 21, 33, 40, 67,
Ouro Preto and Vila Rica, 51, 287n1 69, 80; bishops in, 118; Dutch attack
Index | 319
on, 43; Guinea and, 28–30; ‘‘nation’’ Regimento dos Homens Pretos, 184, 231,
and, 88, 236–37 279n8
Pretos, 53, 68–69, 73, 116, 190, 262n10; in Reigns and Reinados, 138
brotherhoods, 120, 146, 148–50, 161, Reis, João José, 10; Rebelião Escrava no
174, 213, 249n3. See also Livro de Brasil and, 95
Batismo de Pretos Cativos; Pretos forros Religions Africaines au Brésil, Les. Vers une
Pretos forros, 225. See also Freed slaves sociologie des Interpénétrations de Civiliza-
Pretos novos, 116 tion (Bastide), 85, 91
Príncipe: French attack on, 268n15; Religiosity, 11–12, 35, 119, 127–28, 148
origin of name of, 256n57; Portuguese Resgate, 82, 102, 253n8
arrival at, 25 Rio de Janeiro: administration in, 118–20;
Procession of the Dead Lord, 158 annexation of chapels in, 144–45; City
Processions, 11, 116, 139, 141, 146, 186, Field in, 117; as colonial capital, 58;
217; conflict with, 114–15, 267n67; French occupation of, 117; house con-
festive, 142, 154, 158; funeral, 117, struction in, 105, 122–23; Minas
156; icons in, 273n6; music at, 138, Gerais and, 223–24; population
281n28; public donations for, 137, 167, growth in, 62, 78–79, 98; port of, 83,
175, 197, 201, 205, 237; saints’ days, 94; protective wall of, 118, 129, 142,
156–57 144, 268n17, 278n66; slave burials in,
Procedência, 16 134t; Vala trench of, 113, 117, 120, 145
Proença, Clemente, 177 Rodrigues, Nina, 5, 16, 32, 34, 38, 85; on
Provenience, 16, 86, 90, 132–33, 212–13, racial origins, 87
226; kings of folia and, 202; marriage Romanus Pontifex, 24, 253n19
within, 95–97; territory and, 236, Romero, Sílvio, 32, 34, 38–39
264n32; wills and, 232. See also individ- Royal fifth, 49, 60
ual provenience groups
Provenience groups of Brotherhood of Sá, Salvador Correia de, 43–45
Santo Elesbão and Santa Efigênia, 178, Sabará provenience group, 93, 265n51
265n43; burials for, 195 Sahara Desert, 19–21, 23
Santa Casa da Misericórdia, 126, 266n1;
Quinto, 49, 60 brotherhoods and, 115; cemetery of,
Quissamã provenience group, 54, 81; 126, 130t, 161, 203; funeral monopoly
burials of, 134t; slave baptism and, 54t, and, 125, 159, 195, 280n25
58t, 84 Santa Rita, habit of, 127
Santa Rita parish, 79, 97–98, 120, 241,
Race. See Brancos; Cabras; Creoles; Mestiços; 264n36; Church of Santa Rita and, 126;
Mulatos; Negros; Pardos; Pretos house distribution of, 124t; slave bap-
Ramos, Arthur, 34 tisms and, 9t, 71t, 94t, 99t–100t, 242t,
Rebelião Escrava no Brasil (Reis), 95 247t; slave population of, 77t–78t
Rebolo provenience group, 81, 84, 139– Santo Basílio, 207
40, 160; burials of, 134t; slave baptism Santo Elesbão and Santa Efigênia: African
and, 84 origins of, 21, 174; brotherhood’s pair-
Recife, 59–60, 263n17, 277n57; folias at, ing of, 152; festival days for, 154, 156
179; Our Lady of the Rosário and, 139, Santo Ofício, 115, 143, 272n63
142–43, 279n57 São Benedito, 116
320 | Index
São Domingos Field, 140, 144, 270n48, Smallpox, 41, 48; Obaluaê-Xapanã and, 4
278n66 Soares, Antônio Luiz, 227–28; will of,
São Domingos River, 81 232–33
São Francisco, habit of, 126–27, 137 Social hierarchy: in brotherhoods, 114–
São Jerônimo, Dom Francisco de, 62–63, 16, 146–47, 157, 170, 180, 193, 280n21;
115, 204, 214 clothing and, 121; Debret account of,
São José parish, 79, 84, 241, 264n36; 211–12; funerals and, 125–26, 160–61;
house distribution of, 124; slave bap- house details and, 123; transportation
tisms and, 9t, 71t, 94t, 242t, 246t; slave and, 104, 122, 257n27; urban space
population of, 77t–78t, 98–100, 124 and, 113
São Lourenço provenience group, 54t, Souza, Francisco Alves de: Angola and,
58t, 84; burials of, 134t 203–4; on Catholicism and paganism,
São Tomé (island): French attack on, 191, 198, 202, 206, 209, 219–20; on
268n15; lack of data for, 34; latitude charity, 216; conversion of, 207; Frater-
of, 253n4; Portuguese arrival on, 25; as nity of Our Lady of Remédios and,
slave trade hub, 6, 28, 30–31, 42, 50, 214–15; history of, 5, 208, 278n3,
57, 90, 254n35 279n8; illness of, 189; as Mahi Con-
São Tomé provenience group: Brother- gregation leader, 3, 187–88, 190, 193–
hood of Santo Elesbão and Santa 95, 280n24; Mahi folia and, 197; Mahi
Efigênia and, 165, 171, 173–74, 185, Manuscript and, 184–85, 217, 238,
265n43; burials of, 134t; ‘‘nation’’ of, 81 279n13, 279n14, 286n75; Tribunal da
Savalu, 66; provenience group of, 93, 174, Relação and, 192, 196
178, 185, 191, 224, 236 Strabo, 19
Scarano, Julita, 93, 162 Sub conditione, 81–82, 264n34
Schwartz, Stuart, 10, 95, 115, 163, 184 Superstition: brotherhoods and, 191, 202,
See, 62, 120–21, 131, 269n29 205, 214–15, 217, 238, 249n3; dances
Senegal River, 21 and, 142–43; Devotion to the Souls of
Sé parish, 53, 56–58, 68, 79, 83–84, 94, Purgatory and, 202; Jêje and, 152
266n2; Church of São Sebastião and,
62, 114; Debret on, 128; housing stock Tapa, 4; provenience group of, 93, 205
and, 123; slave baptisms and, 9t, 54t, Terra dos brancos, 193
58t, 71t, 94t, 99t–100t, 242t–44t, Terra dos negros (Zurara), 21–23, 31
262n8; slave population and, 77t–78t, Territoriality, in ‘‘nation’’ designation,
93, 98–99, 133 236
Silveira e Albuquerque, Dom Álvaro da, Timbuktu, 20, 59
49–50 Timbu provenience group, 93
Sintra, Gonçalo de, 23 Tobacco trade, 60–63, 259n62, 259n64,
Slave Coast, 27 260n69, 260n70, 260n73
Slave marriages, 95–98, 136, 162, 170, Trans-Atlantic Slave Trade Database, 238
214 Treaty of Alcáçovas, 26, 28
Slave obituaries, 127–33, 271n49 Tribunal da Relação, Rio de Janeiro, 192,
Slave trade, Brazilian, 15, 65; ethnic 196, 200
groupings and, 4, 9–10, 31–38; Rio de Tribunal of Conscience and Orders, Lis-
Janeiro and, 3, 7, 13, 39, 41, 57–58, bon, 6; brotherhoods in Brazil and,
94–95 115, 150–53, 171, 181, 203, 229,
Index | 321
274n11; Portuguese empire and, 118– West India Company, 42, 46, 59
19, 274n12 Wills, 224–25; of freed slaves, 137, 160,
231, 284n55, 291n34; of Ignacio Gon-
Union of the Iberian Crowns, 30, 40 çalves do Monte, 227; property in,
269n30, 269n31, 278n1; provenience
Vala trench, 113, 117, 120, 145 in, 232, 236
Varnhagen, Francisco Adolfo de, 31–32, Witchcraft, 191, 198, 271n61
38
Verger, Pierre, 4–5, 10, 12, 53–54, 58–59, Xambá provenience group, 93
224, 227
Vila Rica and Ouro Preto, 51, 287n1 Yoruba speakers, 4, 227, 235, 261n78,
Volta River, 27, 207, 283n51 265n50; historiography of, 34–35,
Voyage to Guinea (Bosman), 46–47 265n50; in Mahi Manuscript, 219;
Mahi subgroup of, 287n4; provenience
War of Castilian Succession, 26 group of, 14, 54, 89, 93, 277n56
War of Spanish Succession, 268n15
West Africa, 21, 30, 57; land trade routes Zurara, Gomes Eanes de, 27, 31, 33, 36,
of, 20. See also Guinea; Mina; Mina 253n5; Chronicles of Guinea, 21–25
Coast
mariza de carvalho soares is an associate professor of history at Uni-
versidade Federal Fluminense and a researcher funded by cnpq-Conselho
Nacional de Desenvolvimento Científico e Tecnológico. She is the author of
Devotos da Cor: Identidade Étnica, Religiosidade e Escravidão no Rio de Janeiro, Século
XVIII (2000) and co-author (with Ricardo Henrique Salles) of Episódios da
História Afro-Brasileira (2005). She also edited the collection Rotas Atlânticas
da Diáspora Africana: Entre a Baía do Benim e o Rio de Janeiro (2007), and co-
edited (with Nielson Rosa Bezerra) the collection A Escravidão Africana no
Recôncavo da Guanabara (2011).
Library of Congress Cataloging-in-Publication Data
Soares, Mariza de Carvalho.

[Devotos da cor. English]
People of faith : slavery and African Catholics in eighteenth-century Rio de Janeiro /
Mariza de Carvalho Soares ; translated by Jerry Dennis Metz.
p. cm. — (Latin America in translation/en traducción/em tradução)
Originally published as: Devotos da cor : identidade étnica, religiosidade e escravidão
no Rio de Janeiro, século XVIII (Rio de Janeiro : Civilização Brasileira, 2000).
Includes bibliographical references and index.
isbn 978-0-8223-5023-1 (cloth : alk. paper)
isbn 978-0-8223-5040-8 (pbk. : alk. paper)
1. Slavery—Brazil—Rio de Janeiro—History—18th century.
2. Africans—Brazil—Rio de Janeiro—Ethnic identity.
3. Africans—Religious life—Brazil—Rio de Janeiro.
4. Slaves—Religious life—Brazil—Rio de Janeiro.
I. Title.
II. Series: Latin America in translation/en traducción/em tradução.
ht1129.r53s6313 2011
306.3%62098153—dc22
2011015701

People of Faith

Uploaded by

Copyright:

Available Formats

People of Faith

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

People of Faith

Uploaded by

Copyright:

Available Formats

People of Faith

A book in the series

mariza de carvalho soares

duke university press

Printed in the United States of America on

Library of Congress Cataloging-in-Publication

40 | 2 Commerce with the Mina Coast

67 | 3 African ‘‘Nations’’ and Provenience Groups

101 Gallery of Illustrations

146 | 5 Constructing a Religious Norm

183 | 6 Conﬂict and Ethnic Identity among Mahi

9 | 1 Total Slave Baptisms Valid for Analysis, 1718–1760

54 | 2 Baptisms of Adult Africans in Sé Parish, 1718–1726

58 | 3 Baptisms of Adult Africans in Sé Parish, 1744–1750

71 | 4 Total Baptisms of Slaves in Rio de Janeiro, 1718–1860

77 | 5 Principal Proveniences of Slaves in Rio de Janeiro, 1718–1760

78 | 6 Principal Proveniences of Slaves in Rio de Janeiro, 1718–1760

94 | 7 Mina Adults Baptized in Rio de Janeiro, 1718–1760

97 | 8 Marriages of Female African Slaves in Rio de Janeiro, 1718–1760

99 | 9 Distribution of Child Baptisms according to Mother’s

100 | 10 Distribution of Mina Baptisms by Parish, 1718–1760

124 | 11 Distribution of Houses by Parish in Rio de Janeiro, 1779–1789

130 | 12 Distribution of Slave Obituaries from Candelária Parish, by Locale

242 | 14 Total Slave Baptisms in Rio de Janeiro, 1718–1760

242 | 15 Total Slave Baptisms Valid for Analysis, 1718–1760

243 | 16 Slave Baptisms in Sé Parish, Rio de Janeiro, 1718–1726

243 | 17 Slave Baptisms in Sé Parish, Rio de Janeiro, 1744–1750

244 | 18 Slave Baptisms in Sé Parish, Rio de Janeiro, 1751–1760

245 | 19 Slave Baptisms in Candelária Parish, Rio de Janeiro, 1751–1760

246 | 20 Slave Baptisms in São José Parish, Rio de Janeiro, 1751–1760

247 | 21 Slave Baptisms in Santa Rita Parish, 1751–1760

The initial version of what would become this book was my

\ doctoral dissertation for the Department of History at the Fed-

I paid my ﬁrst visit to the Church of Santo Elesbão and Santa

\ Eﬁgênia one Wednesday morning, back in 1989, and it was an

conversations, speaking in hushed voices as we sat in the benches of the

in the eighteenth century. Almost without exception, Brazilian studies of

no centralized authority or government. It also seems that during this

The number of Jêjes [Ewe or Eves] has markedly diminished in Bahia.∞∑

My own research also beneﬁted from the kindness of informants. At the

in Portuguese—as well as the complete ecclesiastical correspondence be-

But I had a more immediate question to address. What could I make of

graphics in Rio de Janeiro, and, proceeding from there, in evaluating the

Table 1. Total Slave Baptisms Valid for Analysis, 1718–1760

parish and years baptisms total

individual identiﬁcation the names were partly a strategy to classify Afri-

The diversity of sources I encountered enabled me to follow a relatively

Studies of slavery in Brazil were another framework to be considered. I

commerce of slaves in the city of Rio de Janeiro). Still, none of these

intertextuality has to be kept in mind to analyze or contextualize any given

As Renaissance Europe probed its southern frontier through

\ trade networks branching across the Mediterranean, its mer-

a place of Christian kingdoms where Santo Elesbão (Ethiopia) and Santa

the land of the black moors

lauding the notable deeds of Infante Dom Henrique (1394–1460, also

This helps us understand why he would come to insist in the Chronicles

the mina of guinea

establish Portuguese-controlled trade in a region favored by slave hunters,

case, the discovery of these territories unknown to previous explorers and