Slavery and Identity

Slavery and Identity
Blacks in the Diaspora

Founding Editors
Darlene Clark Hine, John McCluskey, Jr., and David Barry Gaspar
Editor
Claude A. Clegg III
Advisory Board
Kim D. Butler
Judith A. Byfield
Leslie A. Schwalm
Tracy Sharpley-W hiting
M IEK O N IS H ID A

Ethnicity; Gender, and
Race in Salvador, Brazil,
1808-1888
INDIANA
NDIANA
University Press
Bloomington & Indianapolis
This book is a publication of
In d ia n a U n iv e r s it y F u s s
601 North Morton Street

Bloomington, IN 47404-3797 USA
http ://iupre$s.indiana.edu
Telephone orders 800-S42-6796
Fax orders 812-855-7931
Orders by e-mail iuporder€indiana.edu
C 2003 by Mieko Nishida
Ail rights reserved
No part o f this book may be reproduced or utilized in any form or by any means, electronic or
mechanical, including photocopying and recording, or by any information storage and retrieval
system, without permission in writing from the publisher. The Association o f American University
Presses* Resolution on Permissions constitutes the only exception to this prohibition.
The paper used in this publication meets the minimum requirements o f American Nation*]
Standard for Information Sciences— Permanence o f Paper for Printed Library Materials» ANSI
Z394«»984-
Manufactured in the United States o f America
Library o f C o n g r a i Cataloging-ln-Publication Data
Nishida, Mieko, date

Slavery and identity: ethnicity, gender, and race in Salvador, Brazil, 1808-18&8 / Mieko Nishida.
p. cm. — (Blacks in the diaspora)
Includes bibliographical references (p .) and index.
ISBN 0-253-34209-0 — ISBN 0-253-21581-1
1. Slavery— Brazil—Salvador— History— 19th century. 2. Slavery— Brazil—Salvador— Psychological
aspects. 3. Slaves— Brazil— Salvador— History—19th century. 4. Slaves—Brazil—Salvador— Social
conditions— 19th century. 5. Slaves— Brazil—Salvador— Psychology. 6. Slaves— Brazil—Salvador—
Emancipation. 7. Africans— Brazil— Salvador— Ethnic identity. 1. Title. II. Series.
HTH29.S2 N57 2003
3o6.3#62'o98i —d cii
2002010944
12 3 4 5 08 07 06 05 04 03
To the memory o f my grandmother
Kikuno Nishida (1909-1989)
All m y life I had been looking for something, and everywhere I turned
someone tried to tell me what it was. I accepted their answers too, though
they were often in contradiction and even self-contradictory. I was naive.
I was looking for m yself and asking everyone except myself questions
which I, only I, could answer. It took me a long time and much painful
boomeranging o f my expectations to achieve a realization everyone else
appears to have been born with: That I am nobody but myself.
— Ralph Ellison, Invisible Man
Contents
Acknowledgm ents xi
List o f Abbreviations xv
Introduction 1
1. A “ Capital o f A frica” in Brazil 11
P A R T O N E : T O BE A F R I C A N - B O R N A N D E N S L A V E D , C I R C A 1808-1831
2. T he Creation o f N ew Identity, 1808-1831 29
3. T h e Representation o f Identity, 1808-1831 48
PART T W O : T O BE A F R I C A N - B O R N A N D F R E E D , C I R C A 180 8-1880

4. T he Re-creation o f Identity, 1808-1831 73
5. T h e Convergence o f Identity, 1831-1880 92
PA R T T H R E E : T O BE B R A Z I L I A N - B O R N , C I R C A l 8 o 8 - l 8 8 8
6. T h e Creation o f Disparate Identity, 1808-1851 123

7. T h e Labyrinth o f Identity, 1851-1888 142
C onclusion 157
G lossary 167
Notes 169
Bibliography 227
Index 251
Acknowledgments
This book narrates a peculiar sort o f history o f the New World’s “peculiar
institution.” It is not about slavery per se; instead it presents a new interpre
tation o f urban slavery in an Atlantic port city from the vantage point o f
enslaved Africans and their descendants during the slavery regime and ex
amines these people’s self-perceptions and self-identities in a variety o f situa
tions. Drawing on appropriate available primary sources, both archival and
printed, this book discusses the perspectives o f slaves, ex-slaves, and free-
born people o f color and explores a number o f factors that affected their
lives and self-perceptions most critically.
Enslavement in Africa had detached these people from their communities
and, together with the following hardships o f the Middle Passage, had de
prived them o f considerable honor and dignity as human beings. They had
been forced to give up their original identities upon enslavement. Having
arrived in the New World, every one o f them was placed in a situation o f
slavery. Inevitably all the enslaved human beings o f African birth suffered
enormously and sometimes despaired. But despite the hopelessness o f their
situation, they exercised their slight remaining power over themselves to
create and re-create their own distinctive identities through their struggles
for the recovery o f their lost freedom and humanity. W hile avoiding roman
ticizing the past, this book reconstructs these people’s stories o f (re)creating
identities within the context o f New World history.
This book involves my own search for m y identity. I was bom and grew
up in Japan, and eventually left for the United States to be enrolled in a
Ph.D. program, without imagining what fate would await me. My geographi
cal move from the Old World to the New World coincided with m y discipli
nary departure from anthropology for history. I have since sailed a long
voyage to re-create myself as a historian o f Brazil beyond and above my
multiple “otherness.” Although my name appears in print as the author o f
this book, I know that the book would never have been completed without
many wonderful teachers, colleagues, friends, and family in three countries
where I have lived: the United States, Brazil, and Japan. I remain immensely
grateful to all o f them.
First o f all, I would like to thank tw o anonymous readers for their me
ticulous readings o f the manuscript, followed by sharp critiques, insightful
comments, and thoughtful suggestions for the successful publication o f this
book. I am very grateful to Professor Franklin W. Knight for his strong sup
port, generous help, and warm friendship over the years. I thank Professor
Anani Dzidzieyno for his help, friendship, and support, and all o f our inter-
cultural conversations on race, ethnicity, and gender in Brazil and beyond.
I owe most special thanks to Professor W illiam B. Taylor for teaching me
history as a “series o f miracles” and for reading several different versions o f
my manuscript w ith much care. Professor Joseph L. Love generously found
the time to critique my entire manuscript. I thank Professor W illiam £.
Jackson for his unfailing faith in my work and me. James Sidbury has been
an excellent critic and a good friend.
M y thanks are due to Dona Edy Aleluia, Judith Lee Allen, Professor
Iraci Del Nero da Costa, the late Professor Peter L. Eisenberg, Professor
Jack P. Greene, Professor M ary C. Karasch, Professor Richard Graham,
Dona Helena R. Guim araos, Professors Takashi and Jandyra Maeyama,
Jacira Almeida Mendes, Professor Joseph C. Miller, the late Sr. Teizan Nishi-
oka, Tejumola Olanyian, Professor Anne P^rotin-Dumon, the late Professor
Armstead L. Robinson, Sr. Aloisio Conceiâo Rocha, M ary F. Rose, Profes
sor A. J. R. Russell-Wood, Professor Consuelo Novais Sampaio, Sr. Daniel
Azevedo dos Santos, Dona Arlinda Slras, the late Sr. Tetsuya Tajiri, Professor
Luis Henrique Dias Tavares, and Betty L. W hildin. I also thank David J.
Bachner, Thomas C. and Muriel E. Beattie, Regan Brumagen, M ark Erick
son, Lee and Joanne Fisher, and Am y Rosner for their friendship and sup
port.
I am indebted to the wonderful staff o f the public and private archives in
Salvador and Rio de Janeiro, Brazil; the Carter G. Woodson Institute o f
the University o f Virginia; and the Benson Latin American Collection and
the Institute o f Latin American Studies, both o f the University o f Texas at
Austin.
M y research and writing were funded by The Johns Hopkins University
graduate fellowships; a predoctoral research fellowship at the Carter G.
Woodson Institute o f the University o f Virginia; and a Rockefeller Founda
tion postdoctoral fellowship at the Institute o f Latin American Studies o f
the University o f Texas at Austin.
M y thanks are due to the editorial staff o f the Indiana University Press:
Robert J. Sloan, M arvin Keenan, Kendra Boileau Stokes, Jane Lyle, and copy
editor Carol Kennedy. 1 would like to acknowledge permission to reprint
xii Acknowledgments
portions o f several chapters that have been previously published in differ
ent forms: “Manumission and Ethnicity in Urban Slavery: Salvador, Brazil,
1808-1888” Hispanic American Historical Review 73, no. 3 (1993): 361-391;
and “ From Ethnicity to Race and Gender: Transformations o f Black Lay So
dalities in Salvador, Brazil,” Journal o f Social History 32, no. 2 (1998): 329-348.
Several quotes in the text have been taken from four great works o f Ameri
can fiction with permission, and 1 thank Random House Inc. for Ralph El
lison, Invisible Man, and Caryl Phillips, Crossing River; Atlantic M onthly
Press for David Mura, Turning Japanese; and International Creative M an
agement, Inc. for Toni Morrison, The Bluest Eye.
I would like to express my heartfelt thanks to Kevin P. Rauch, an astro
physicist who patiently guided me through to find my own “star” in the very
dark sky. V\\ always remember him saying to me, “Yes, you can do it!”
Lastly but not least importantly, m y profound gratitude goes to m y fam
ily in Japan for their unfailing support: Takako, Taketoshi, and Hiroaki
Nishida, my parents and brother; and K ikuno and Kiyokazu Nishida, my
late grandparents. This book is dedicated to my grandmother, who always
believed in me.
M ieko Nishida
Acknowledgments xiii
Abbreviations
Archives
ACM S Arquivo da Cùria Metropolitana de Säo Salvador da Bahia, Salva

dor, Brazil
AINS Arquivo da Igreja da Nossa Senhora do Rosàrio dos Homens
Prétos, Salvador, Brazil
AM CS Arquivo Municipal da Cidade do Salvador, Salvador, Brazil
ANRJ Arquivo Nacional do Rio de Janeiro, Rio de Janeiro, Brazil
APB Arquivo Público do Estado da Bahia, Salvador, Brazil
ASPD Arquivo da Sociedade Protetora dos Desvalidos, Salvador, Brazil
BNRJ Biblioteca Nacional do Rio de Janeiro, Rio de Janeiro, Brazil
Journals
AAEB Anais de Arquivo Público do Estado da Bahia

ABNRJ Anais de Biblioteca Nacional do Rio de Janeiro
AH R American Historical Review
HA HR Hispanic American Historical Review
Introduction
Identity is not given. It is one’s situational synthesis o f self in relation to

others. Hence, one defines and redefines his or her identity in each specific
situation/context, particularly in relation to other people’s perception(s) o f
him or her. Yet, even i f one does not seem to possess any political power to
influence others, one does not necessarily always have to accept a label or
category that the larger society imposes, most likely based on ostensible ap
pearance or phenotype. The creation o f identity is a constant negotiation
o f varying impressions. In other words, the creation o f identity is by no
means one's passive reaction to a specific environment. It is as well an ac
tive process o f self-creation. Collective identity emerges only when some
individuals successfully recognize their common traits as a group, beyond
differences, and connect with one another in opposition to others. There
fore, identity, either individual or collective, evolves, transforms, disappears,
and/or reemerges over time in relation to a changing context.'
These are the issues o f identity that this book scrutinizes and emphasizes.
In order to do so, I choose as my subject the powerless, who were simulta
neously the ethnic/racial “other” in New World history, namely enslaved
peoples o f African birth and descent. This study discusses how such power
less individuals and their descendants, women and men, enslaved and le
gally free (freed and free-born), created and re-created distinctive identities
in diverse situations during the slavery regime. M y understanding is that
a specific sense o f self, namely identity, stimulates, encourages, and even
forces the individual to make concrete decisions and to take certain actions.
This study examines very carefully one’s actions and behavior as the out
come or representation o f his or her specific identity or identities, not the
other way around.2
The city o f Salvador, Bahia, Brazil provides us with a most provocative
setting for our discussions on the creation o f identity during the slavery re
gime. By the mid-i770s Salvador had grown larger than any city o f British
America and possessed a larger population than major British cities such as
Bristol, Liverpool, Birmingham, and Manchester.3As an important Atlantic
port city, Salvador maintained its lead over Rio de Janeiro as the most popu
lated city o f colonial Brazil until the late 1800s.4 Urban slavery had devel
oped in Salvador and had existed throughout the city since the early colonial
period, and people o f African descent have constituted the majority o f the
population in Salvador from the early colonial period until the present day.
The unique urban setting o f Salvador enabled enslaved Africans and their
descendants to come into contact w ith each other on various occasions in
the course o f everyday life.
For this study I have chosen to focus on a specific time period, from 1808
to 1888, which may well be regarded as a most critical transitional period o f
Brazilian history, following the transfer o f the Portuguese court from Lis
bon to Rio de Janeiro. Upon arrival, at the demands o f Great Britain, whose
N avy escorted their move to Brazil immediately after the Napoleonic inva
sion in Portugal, the Portuguese court was forced to immediately open all
Brazilian ports directly to the world market. This brought an economic
boom to Salvador as well as other port cities o f Portuguese America. Tran
sition from mercantilism to free trade (1808) brought Salvador new eco
nomic growth; Bahian exports increased from a total value o f 1,418 contos
in 1808 to 8,178 contos in 1824, whereas sugar production increased from
20,000 caixas (boxes or crates) in 1808, to 29,628 caixas in 1818, and 33,000
caixas in 1833. Imports to Bahia also increased by 50 percent w ithin the first
year after the opening o f the ports to free trade.5 During the given period
1808 to 1888, the transformation o f Brazilian society took place at tw o dif
ferent levels. One was socioeconomic: transition from slave society to legally
free society (1888). The other was political: transition from colony to m on
archy (1822) and then to the Republic (1889). As a result o f such drastic and
rapid political and socioeconomic changes in the larger society, the charac
teristics o f those o f African descent in Salvador continued to change in
accordance with the illegalization o f the slave trade in Brazil (1831); the ter
mination o f the transatlantic slave trade (1851); and the destruction o f Bra
zilian slavery, beginning with the enactment o f the Free Womb Law (1871)
and moving toward total abolition (1888) with the enactment o f the Golden
Law (Lei Atirea) o f May 13,1888.
By focusing on a specific local setting for a specific time period, this book
raises theoretically broad and historically important questions on slavery
and identity in New World history. In the course o f our discussions on the
creation o f identity, three important factors emerge and form the book’s
three major intersecting axes o f analysis. They are ethnicity, gender, and
race.
2 Slavery and Identity

Studies on Identity
This historical case study focuses on the creation o f individual iden

tities; inter-relatedness among those who came to share the same identity;
and the emergence, transformation, and disappearance o f collective identi
ties. This study follows such important questions on identity as those raised
by British social anthropologist A. L. Epstein, concerning by what processes
a “sense o f collective identity” was “generated, transmitted, and perpetu
ated”; how new identities came to be formed and how they interacted with
preexisting ones; and what were the “circumstances in which established
identities” were “abandoned or simply disappeared.”6
Anthropological Studies on Ethnic Identity
M y approach to ethnic identity among the African-born popula

tion in nineteenth-century Salvador reflects urban studies o f South-Central
Africa by British social anthropologists at the University o f Manchester.
J. Clyde Mitchell discusses urban ethnicity in the name o f “tribalism” from
a “phenomenological approach, which takes into account the meaning o f
actions from the actor's point o f view coupled with a situational approach
in which the meanings are related to the actor's definition o f the social situa
tions” 7 An urban environment creates anonymity, which is the keystone o f
categorical interaction since it operates in public places or wherever the ac
tors know little o f one another.8 Am ong the urban migrant population on
the Copperbelt o f Northern Rhodesia in the 1950s, more than one hundred
“tribal” groups were reduced to a handful o f “tribal” categories in accor
dance with tw o principles: geographical propinquity and cultural affinity.9
Unlike “tribes” in rural areas, “tribalism” in urban areas was a situational
phenomenon in response to urban African life, a category o f interaction
based on broad cultural differences but within an urban labor system. Thus
Mitchell distinguishes “tribe” in rural areas as a structure from “tribe” in
cities as a category. According to Mitchell, categorization is “a common re
action in a situation where social relationships are o f necessity transitory
and superficial while at the same time multitudinous and extensive.” 10 In
such circumstances individuals seek means o f reducing the complexity o f
social relations with which they are confronted. Urban migrants achieve this
goal by classifying the people around them in a limited number o f catego
ries. They draw a “cognitive map,” using a variety o f indicators, such as eth
nic badges, to which individuals respond with the appropriate behavior.11
Introduction 3
Historical Studies on the Creation o f Collective Identity
M y discussions on the historical creation o f collective identity have

greatly benefited from the renowned historian E. P. Thompson’s classical
work, The Making o f the English Working Class (1963).12 Thompson defines
class as “an historical phenomenon, unifying a number o f disparate and
seemingly unconnected events, both in the raw material o f experience and
in consciousness” and “something which in fact happens (and can be shown
to have happened) in human relationships.” 13Thompson continues: “ (C]lass
happens when some men, as a result o f common experiences (inherited and
shared), feel and articulate the identity o f their interests as between them
selves, and as against other men whose interests are different from (and usu
ally opposed to) theirs.” 14 Accordingly, between the years 1780 and 1832,
most English working people came to share a collective identity in common
as a distinctive class against their rulers and employers.
The U.S. historian Herbert G. Gutm an, highly influenced by T hom p
sonian theories o f class formation, in his The Black Family in Slavery and
Freedom, 1750-192$ (1976), discusses historical processes by which slaves in
the American South, “especially in the eighteenth century, had to forge new
institutions and beliefs to sustain them in their oppression,” and “some—
not all, by any means— o f these institutions and beliefs were formed and
transmuted.” 15 He presents his monumental book as “a special aspect o f
American labor history: those men and women who labored first in bondage
and then mostly as half-free rural workers,” 16and maintains that “the same
kind o f historical questions can be asked o f slaves that historians could ask
about any other exploited population.” 17This raises the question o f whether
slaves in the American South formed a social class as a working people, as
was the case in England.
Anthropologists Sidney W. M intz and Richard Price, in their book en
titled An Anthropological Approach to the Afro-American Past, which was
published coincidentally in the same year as Gutm an’s book, discuss the
historical creation o f African American culture, namely the creolization
process.18 By taking such a similar combination o f phenomenological and
situational approaches, as suggested by J. C. M itchell,19M intz and Price per
ceive the “creolization” o f diverse African cultures under slavery not as
the simple transplantation o f Old World cultures in the New World but as
a creative process.20 Yet their model, largely based on the case o f colonial
Surinam, seems to suggest that creolization could be synonymous with the
convergence o f diverse African cultures under a slave system into a unified,

singular “African American culture.” M intz and Price maintain that “the
Africans in any New World co lo n y. . . began to share a culture only insofar
as, and as fast as, they themselves created them.”21
Thompson, Gutman, and M intz and Price all emphasize that human ac
tions mattered in making history. By sharing the same experiences o f be
ing socioeconomically exploited (and sometimes even culturally deprived),
people connected with one another, reduced differences among themselves,
and eventually created a common collective identity. This process resulted
in the formation o f a distinctive social class, or the creolization o f diverse
cultures, in accordance with each context. This book will test whether these
models o f collective identity formation are applicable to the case o f African-
born slaves and their descendants who lived in a specific New World city
during the slavery regime. Did all such exploited individuals actually de
velop a collective identity over time by being oppressed commonly as slaves
and/or by being part o f the population o f African descent? Was it rather an
outsider’s relatively shallow observation and stereotyping o f others based
on their seemingly shared superficial appearance(s), or even on a romanti
cized imagination o f the powerless as a collective entity?
In order to answer these questions we have to investigate and study the
individual’s self-perception(s) on individual terms, not as a member o f a
group to which she or he is (or was) supposed to belong. After all, identity
often differs from a category imposed on the individual by the larger so
ciety, thereby inevitably creating a critical distance between self-perceptions
and ostensible appearances. This is well exemplified by the anthropologist
Stephan Palmi£ in his monograph on the development o f new collective
identity as “black” Hispanics in M iami in that group’s relations with A fri
can Americans, as well as with lighter-skinned Hispanics who may pass as
whites in the U.S. society. Palmi£ skillfully demonstrates that sharing the
same phenotypes as “ blacks” (by U.S. standards) does not, in and o f itself,
provide sufficient foundation for “black” Hispanics to share the same racial
identity w ith African Americans.22 For, as Anthony P. Cohen demonstrates
in his theoretical monograph on the notion o f community, “appearances
are deceptive”23 Suzanne Oboler discusses similar, intriguing conflicts be
tween ethnic labels and individual and collective identities among the di
verse populations o f Latin American origins in her meticulously researched
monograph Ethnic Labeb, Latino Lives (1995).24
W hile greatly benefiting from the aforementioned major monographs by
Thompson, Gutman, and M intz and Price, this study challenges their linear
models o f collective identity formation by demonstrating that the histori-
Introduction 5
cal creation o f collective identity by no means took just one single process
for the population o f African descent who lived in Salvador during the
nineteenth century. Despite the rapid creolization process after the mid
nineteenth century, all the free-born population o f African descent in Sal
vador did not come to perceive/identify themselves as a homogeneous racial/
ethnic group. This book w ill attempt to illustrate carefully the most intri
cate historical process o f identity formation by African-born slaves and
their descendants in New World history.
Race and C o lo r in Brazil
Perceptions o f race and color in Brazil differ considerably from the

situation in the United States, despite the fact that they share the long his
tory o f African/black slavery in the New World.25 Definition o f who is black
or not varies and is only culturally defined; there is no cross-cultural crite
rion. Sidney W. M intz elucidates: “ (A)nyone here [in the U.S.] who has any
black ancestry is defined as black.”26As for Brazil, as the Brazilian historian
Emilia Viotti da Costa cogently explains, when Brazilians and Americans
speak o f blacks, they mean different things. The former emphasize the in
dividual phenotype, whereas the latter define race in hereditary terms: par
entage, not appearance, decides who is white or black in the United States.
Accordingly for Viotti da Costa, in Brazil the mulatto has been seen as evi
dence that the population was becoming white, whereas in the United States
as becoming black.27 In short, as M intz accurately summarizes, black is “a
social line, not a racial line” in both the U.S. and Brazil.28
In Brazil, unlike the case o f the United States, recognition o f variations
in the physical appearance o f the population o f African descent has been
varied and complex since the early colonial period. In New World slave so
cieties, as David W. Cohen and Jack P. Greene suggest, during the early phase
o f the emergence o f the free population o f color (the first several genera
tions), the complex, multitiered, color-coding system may have been opera
tive in terms o f marriage reference and social status. However, unlike many
other New World societies, where these categories became vague and con
fused even during the first century o f the institution o f slavery, in Brazil
this color-coding system continued to function throughout the colonial and
imperial periods and has survived to a considerable extent until today.29 For
instance, according to Charles R. Boxer, in Brazil, as well as in Portuguese
Asia and Portuguese Africa, negro (black), prito (black), and cafre (Kaffir)
were all pejorative terms, often synonymous with escravo (slave).50 During

the colonial period, Brazilian-born people o f color were occasionally de
scribed with such ill-defined phrases as trigueiro, corado bastamente (brown,
fairly brown), de côrfechada (o f a closed color), de uma côr equivoca (o f a
dubious color), ao parecer branco (white to all appearances), or de côr fula
(o f the color o f the Fulah).31 In present-day Brazil hundreds o f terms o f
color exist, which are impossible to translate into English. The census o f
1980 found 136 color labels for self-description, while in the National House
hold Survey o f 1976, Brazilians responded with 200 color terms.32
In Salvador, throughout the slavery regime, the Brazilian-born individual
o f African descent not only was always referred to in terms o f legal status
but also was identified with a color, and such usage o f color continued to
divide the Brazilian-born population o f African descent. In chapter i> I
w ill introduce to the reader the color term inology that was prevalent in
nineteenth-century Salvador. Chapter 7, focusing on the free Brazilian-born
population o f African descent in the late nineteenth century, should enable
us to understand race, color, and identity in present-day Brazil. In this
book, whenever possible, I choose to avoid categorizing and referring to
African-born peoples and their Brazilian-born descendants together as
“blacks,” which inevitably carries the perception o f race and color prevalent
in the United States. Furthermore, whenever necessary, I specify a place o f
birth, ethnic origin, legal status, or color in accordance with the original lo
cal (Brazilian and/or Bahian) usage. Only in the case o f lay sodalities for the
population o f color and the colored militia regiments do I follow the pre
ceding studies to employ the usage o f “black” and “mulatto” sodalities and
militia regiments to describe and distinguish them from each other and also
from their white counterparts.
Structure o f the Book
This book is divided into three parts, in accordance with the chang
ing composition o f the population o f African descent: from being predomi
nantly enslaved to free, and from being predominantly African-born to
Brazilian-born. I present empirical evidence that clearly demonstrates that
enslaved people o f African birth and their Brazilian-born descendants cre
ated and re-created distinctive identities during the slavery regime, and I
show when, how, and why the creation o f identity took place.
Chapter 1 serves as the foundation for the entire study. It reveals the char
acteristics o f Salvador, Bahia, and the transatlantic slave trade; discusses
Introduction 7
urban slavery and slaveholding in nineteenth-century Salvador; and catego
rizes people o f color in Salvador.
Part I (chapters 2 and 3) focuses on the creation o f identity by the en
slaved African-born women and men o f diverse ethnic origins as they in
teracted with the continuing influx o f new arrivals from Africa. Chapter 2
examines processes by which African-born slaves created ethnic and gender
identities in New World urban slavery and emphasizes that the creation o f
gender identity undermined ethnic solidarity among the enslaved popula
tion to a considerable degree. Chapter 3 discusses how new gender and eth
nic identities came to be represented in various collective forms: gatherings
and groupings, mating and consensual unions» voluntary associations, and
slave flights and uprisings.
The subject o f part II (chapters 4 and 5) is African-born peoples who
managed to regain their freedom in the New World. Chapter 4 examines the
re-creation o f identity as ex-slaves. African-born ex-slaves, who could not
benefit from the practice o f unpaid manumission, most commonly obtained
their freedom through self-purchase, occasionally by manipulating their
ethnic identities. Upon obtaining freedom they attempted to follow the
elite's behavioral patterns, such as slaveholding and multi-memberships in
lay sodalities, but their newly acquired freedom in the New World was never
to be the same as the freedom that many o f them had experienced as free-
born people and had taken for granted in their native lands o f the Old
World. Once they obtained freedom, the former slaves o f African birth
sharply distinguished themselves from their enslaved counterparts.
Chapter 5 examines the rapid convergence o f diverse ethnic identities
into a broader ethnic identity as African-born. The 1831 decree officially
banned the slave trade in Brazil, putting an end to Salvador’s direct impor
tation o f African-born slaves, although a great number o f enslaved Africans
continued to be imported to Bahia, albeit illegally, until the mid-nineteenth
century. The resident population o f African-born ex-slaves faced new dis
illusionment and frustrations. Their newly acquired freedom as ex-slaves
did not approximate the freedom that most o f them had known as free-
born in their homelands o f Africa; the larger society continued to perceive
African-born ex-slaves as a critical threat to social order because o f their
cultural otherness derived from their foreign birth. Ultimately in 1835, under
the common banner o f Islam, African-born ex-slaves united with African-
born slaves in 1835 to stage a well-organized, large-scale uprising. Known as
the Mal£ revolt, this uprising was predominantly a male phenomenon, as so
many previous rebellions had been. But because it involved both slaves and

ex-slaves almost equally, it was a unique kind o f social movement. And
it paved a way to other unique events. Some African-born ex-slaves volun
tarily returned to their homelands o f Africa, while others remaining in Sal
vador began to connect with one another through kinship and Active kin
ship networks.
Part 111 (chapters 6 and 7) discusses the creation o f a disparate identity
by the Brazilian-born population o f African descent. Chapter 6 shows how
the Brazilian-born population created their identities in relation to the
larger society's favorable treatment o f Brazilian-born over African-born, o f
lighter-skinned over darker-skinned, and o f women over men among the
Brazilian-born population. The proportion o f the Brazilian-born free popu
lation began to expand rapidly once urban slavery started to decline in Sal
vador during the 1840s. I examine how the Brazilian-born population o f A f
rican descent re-created their identity in freedom, in relation to a changing
socioeconomic and cultural context in the 1830s and 1840s.
Chapter 7 treats the creation o f a racial identity by a group o f free-born
black men o f Brazilian birth. In the last few decades o f the slavery regime,
free-born people o f African descent distinguished themselves from the slave
population, most o f whom had been labeled as “blacks” by the larger so
ciety, and tended to identify themselves as “mulattoes,” regardless o f their
visible skin colors and other ostensible phenotypes. Nevertheless, a group
o f free-born men o f African descent with artisanal skills and substantial
economic means chose a collective racial identity as blacks and transformed
one o f the tw o newly established free black lay sodalities into a new type o f
mutual-aid association exclusively for black Brazilian male citizens. But
they did not allow women, as their wives and mothers, to become full mem
bers o f their association. Their better socioeconomic standing enabled them
collectively to choose to identify themselves as blacks, albeit by excluding
others in terms o f class and gender.
The concluding chapter summarizes and reexamines the major argument
about the creation o f ethnic, gender, and racial identities in slavery and free
dom in the case o f nineteenth-century Salvador, Brazil. The way Africans
and their descendants behaved in Salvador clearly indicates that they were
a highly resilient population who never lost their creativity either in the ex
perience o f the Middle Passage or in the centrifugal experience o f slavery
in the New World: those who speak o f “way o f death” and “social death”
greatly exaggerate.331 end the study by examining its implication for our
own “global age” a time when the creation— and loss— o f racial, ethnic, and
national identities is arguably our most important problem.
Introduction 9
i A “Capital of Africa” in Brazil
The city o f Salvador, commonly called Bahia after the name o f its state, is
located on the southeastern shore o f the Bay o f All Saints (Bahia de Todos
os Santos), and is situated approximately thirteen degrees south o f the
equator. Geographically speaking, the city is divided into distinct parts:
the cidade alta (upper city) and the cidade baixa (lower city). Throughout
the colonial period and the nineteenth century, the upper city, built on an
escarpment about six hundred feet high, was a largely residential district for
wealthy merchants as well as for Bahian sugar planters who resided in the
capital city for part o f the year. Many wealthy British merchants resided in
Vitória parish with its church o f O ur Lady o f Grace. The upper city also
contained the cathedral (formerly the Jesuit College), the governor’s palace,
the Casa de Càmara (municipal offices), the Treasury, the Santa Casa da
Misericòrdia (Holy House o f Mercy), and the monasteries o f the Francis
cans, Carmelites, and Benedictines.1 The lower city, composed o f the two
parishes o f C o n c e d o da Praia and Pilar, is built on alluvial soil, with a
rocky substratum, and occupies a very narrow strip along the bay. Tradi
tionally, the lower city was the commercial district, containing the dock
yard, the marine arsenal, the Alfandega or custom house, and markets, as
well as the stores o f merchants and spacious warehouses for the port.2 In the
nineteenth century, these two parts o f Salvador were connected only by
narrow alleys, which were “passable with great difficulty by carriages or
teams.” 3
Visiting Salvador in 1856, the German traveler Robert Abé-Lallemant de
scribed the overwhelming presence o f “blacks” in the city:
No sooner has the traveler set foot in Bahia [Salvador], than he/she
is struck by the inescapable fact that the public roaming the streets is
exactly in keeping with the maze o f houses and alleys. Indeed, there can
be few cities so uniquely peopled as Bahia. If one did not know that this
city is in Brazil, one would think it to be a capital o f Africa and the seat
o f a mighty black prince. This is a city where it is easy for the newcomer
to overlook the all-white population. Blacks appear to be everywhere.
There are blacks at the beach, blacks in the city, blacks in the area o f the
lower city, and blacks in the districts o f the upper city. Everyone that
runs, yells, works, carries, and fetches, is black. Yes, even the carriage
horses in Bahia are black. At least, this was m y impression. There are the
unavoidable Bahian sedan chairs, the caderinhas used here as cabriolets
are elsewhere, and for which blacks play at being horses.4
Abé-Lallemant, who characterized the city o f Salvador as “a capital o f A f

rica” in Brazil, was, in fact, one o f many visitors from Europe and the British
North American mainland who noted in their travel journals the prominent
blackness o f the population o f Salvador during the nineteenth century. In
their accounts o f Brazil, many such foreign travelers expressed amazement
at the numerically notable “black” population in Salvador and presented a
stereotypical description o f these “blacks” as possessors o f a collective men
tality, w ith common physical characteristics, who were eminently suitable
as a labor force in such street activities as chair carriers working in gangs.5
Small wonder all these foreign travelers did not overly exaggerate the his
torical reality. In 1807, Salvador's population amounted to 51,112, which rapidly
doubled by 1872, to 108,138.6 Throughout the ensuing eighty years until the
abolition o f slavery in Brazil (1888), some 70 percent o f the whole popula
tion o f Salvador was officially classified as “persons o f color (pessoas de cór).”
Salvador, Bahia, and the Transatlantic

Slave Trade, 1549-1851
The C ity o f the Savior (cidade do Salvador), founded by Governor

Tomé de Sousa in 1549, was the seat o f the captaincy o f Bahia and the capital
o f Portuguese America until 1763, when Rio de Janeiro became the capital.
Salvador was the seat o f the governor-general or viceroy, as well as o f the
royal treasury and high court. The High Court o f Appeals (Relaçào) started
to operate in Salvador in 1609/ Salvador became a bishopric in 1551, and an
archiépiscopal see in 1676, and came to be known for its reputed religiosity;
it is said there are 365 churches in the city, one for each day o f the year. Al
though Salvador was the name o f the city and the captaincy was named
Bahia, nomenclature has come to confuse city with captaincy, and the des
ignation “ Bahia” has come to be loosely applied to city, captaincy, and Bay
o f All Saints. Departing from this common practice o f referring to both the
city and the captaincy with the same term, I distinguish Salvador (city)
from Bahia (captaincy, or province after independence in 1822) to avoid con
fusion.

Salvador developed as a major port city for the export o f sugar and later
(in the eighteenth century) o f tobacco as well as for the import o f human
cargoes from Africa, with the Bahian Recóncavo as its agricultural hinter
land. From the colonial period, the Bahian Recóncavo has commonly been
called “the interior,” despite the fact that many Recóncavo towns are located
along or near the coast and are within a day's travel from Salvador. Stuart B.
Schwartz interprets Salvador's relationship with its hinterland as a concep
tual dichotomy; the city o f Salvador has represented urban, cosmopolitan
life, whereas the Bahian Recóncavo has been perceived as rustic, removed,
and aristocratic.8Although wealthy Bahian sugar planters spent a large part
o f the year in residence on their plantations in the Recóncavo and developed
tightly knit networks among themselves to form a dominant class charac
terized as the sugar oligarchy, they maintained their two-storied city man
sions (sobrados) in Salvador for their own and their respective families' part-
time residence and for their children's formal education.9 For such planters,
the city o f Salvador was supposed to intellectually represent the entire cap
taincy (or province) o f Bahia, thereby automatically causing the Recóncavo,
the “ interior,” to be regarded as culturally marginal.
The overwhelming “blackness” in the population o f Salvador undoubt
edly resulted from the long history o f African slavery in Salvador and Bahia,
but African slavery had not been introduced into Bahia until the m id
sixteenth century, when the colony’s agro-export economy began to expand
rapidly with the large-scale cultivation o f sugar crops. Neither at the time
o f their “discovery” o f Brazil (1500) nor immediately thereafter had Portu
guese colonists found in Bahia the golden kingdom o f El Dorado they had
expected. Therefore their earliest commercial activities were limited to the
cutting and exporting o f logs from brazilwood (pau brasil) trees. The red
dye extracted from the brazilwood was highly valued in Europe, especially
at the French court. The Portuguese colonists increasingly turned to the in
digenous population, namely the Tupinambás and the Tbpiniquins, for dye-
wood logs, manioc flour, and labor. The Portuguese called these indigenous
populations gentios (gentiles) or negros da terra (blacks o f the land). These
Portuguese terms reflect Portugal’s historical experience o f the slave trade
and the institution o f slavery in Portugal and the Atlantic Islands. By the
time o f Portuguese colonization o f South America, the term negro had
become almost synonymous with “slave” in the Portuguese language and
was used in contrast to the description o f African slaves as negros de Guiñé
(blacks from Guinea). Such parallel term inology as negros da terra and ne
gros de Guiñé suggests that the Portuguese perceived both the legally free
A “Capital o f Africa” in Brazil 13

indigenous population o f Portuguese Am erica and enslaved Africans as
sources o f slave labor.10
The barter system started to shift to chattel slavery in the 1530s with the
introduction o f the donatary system, which granted property rights to Por
tuguese noblemen. The donataries, with the obligations to develop their
grants by settling colonists and establishing a secure economic basis, made
new demands on indigenous populations. Barter could never supply enough
labor for the cultivation o f the new commercial crop, namely sugar, and the
Portuguese began to enslave Indians to secure labor.11 Whereas the Portu
guese Crown made a distinction between “good” and “bad” Indians and
permitted the colonists to take slaves obtained in “just wars” the Jesuits and
eventually other religious orders were allowed to create aldeias (missionary
villages), where the indigenous population were to be converted to be avail
able for useful activities such as subsistence agriculture and work in the cane
fields. From 1540 to 1570, Indian slavery was most commonly observed on
sugar plantations on the coast o f Brazil, and particularly in Bahia.12
It was the cultivation o f sugar that led to the colony’s prosperity in the
sixteenth and seventeenth centuries. We are not certain exactly when sugar
was introduced to Brazil, but as early as the 1530s and 1540s, sugar agricul
ture was established on a firm basis.13 Fortunately, the fertile black and red
massapi soil, together w ith abundant rain, made the Bahian Rec6ncavo
ideal for large-scale sugar cultivation. As a result, the wealth o f the colony
was concentrated in the hands o f a landed aristocracy o f Bahian sugar
planters and merchants.14
Cultivation o f sugar in a plantation regime requires a large labor force.
Indigenous labor was essential for the colony’s growth in the sixteenth cen
tury, but could not meet the colonists’ increasingly inexhaustible demands
for labor on sugar plantations.15 The Portuguese procured an alternative la
bor supply by extending African slavery from the Atlantic islands to Portu
guese America. Africa became the source o f labor for the sugar plantations
o f Bahia, and enslaved people were imported en masse from Guinea to Sal
vador as early as the 1570s. Under the very harsh working conditions, there
was a high mortality rate, which made necessary a constant replenishment
o f the labor force with slaves from Africa.16During the seventeenth century,
Bahia absorbed almost all human cargoes from Angola.
During the eighteenth century, the major slave source for Bahia shifted
from Angola to West Africa, a change that was attributed to the fall o f Bra
zilian sugar in the world market, as well as the emergence o f a new local
economy in central Brazil. By the end o f the seventeenth century, Brazilian

sugar had lost its predominance in the world market. Brazil's sugar produc
tion fell from 2.5 million arróbas in 1730 to 1.4 million arrdbas in 1776. In
terms o f the share o f the world sugar market, this represented a decrease
from one-third to less than one-tenth. By 1787, according to one English
man’s estimate, Brazil’s share had gone down further, to 7 percent.17 At the
same time, w ithin the colony the discoveries o f alluvial gold (1690s and
subsequently) and diamonds (1728) in Minas Gerais shifted the center o f
prosperity from the sugar-producing Northeast (Bahia and Pernambuco) to
the Central-South. In 1763, in recognition o f the economic and strategic
importance o f the South, the Brazilian capital was transferred from Salva
dor to Rio de Janeiro, which had emerged as the main port o f entry for
Angolan slaves, whose ñnal destination was the gold-bearing regions o f
Minas Gerais, Goiás, and Matto Grosso. This, coupled with changing con
ditions w ithin Africa, prompted Bahia (and Pernambuco) to turn to the
Gold Coast, whose demand for Brazilian tobacco was increasing. Bahia con
tinued to be the dominant producer and supplier o f tobacco, and Bahian
tobacco became an important article o f the slave trade.18 By 1790, Brazil’s
sugar had begun to recover on the world market. The major export centers
o f sugar remained Pernambuco (plus Paraiba), Bahia (and the subordinate
captaincy o f Sergipe), and Rio de Janeiro, but sugar was also becoming a
major crop in Sáo Paulo.19In 1791, a slave uprising in Saint Domingue, which
was the largest sugar producer in the world, grew into what has come to be
known as the Haitian Revolution. During the ensuing war for independence,
Saint Domingue was virtually eliminated from the market, and, as a result,
sugar prices rose dramatically. To meet the vast demand for sugar, Europe
ans turned to the traditional suppliers, such as Brazil, as well as new ones.
The Bahian sugar economy expanded enormously during the following de
cades.20The level o f exports o f Bahian sugar increased by 54.6 percent be
tween 1757 and 1798, and advanced another 9.3 percent during the following
decade. Since approximately one-tenth o f the sugar produced in Bahia was
locally consumed, yearly production rose from nearly 360,000 arrdbas in
1759 to 880,000 arrdbas by about 1807.21 As the result o f the recovery o f the
sugar industry, the total population o f Bahia, which had rapidly decreased
from 288,848 in about 1776 to 193,598 in 1780, recovered to 247,000 by 1799.22
The major suppliers o f slaves to Bahia during the resurgence o f Brazilian
sugar in the last decades o f the eighteenth century were the Bight o f Benin
and the Bight o f Biafra o f West Africa, although limited numbers o f slaves
continued to be imported from other parts o f Africa.23The newly imported
slaves, who were from different regions o f West Africa, arrived in Bahia

on vessels o f Brazilian ownership. This was a reflection o f important changes
that took place after 1793 in the organization o f the transatlantic slave trade,
which affected the slave population in Bahia. First, French and Dutch were
driven from the high seas as a consequence o f war in Europe. A fter 1807,
British and U.S. ships also withdrew from the slave trade, which left only
the Portuguese to operate in the Bight o f Benin. Spanish ships also pur
chased slaves in the Bight o f Benin by the end o f the first decade o f the
nineteenth century, but most o f these slaves went to Cuba. A fter 1807, Brit
ish pressure to end the transatlantic slave trade made it difficult for other
countries to operate between Africa and Bahia. Consequently Brazilians did
most o f the transatlantic slave trade from Africa to Brazil during the sugar
resurgence.24
The transfer o f the Portuguese court from Lisbon to Rio de Janeiro
(1808) exercised a significant influence on the transatlantic slave trade to
Brazil; Great Britain exerted ever-greater pressure to abolish the slave trade
to Brazil. Slaves who had been captured while on a slave ship en route to
Brazil were called africanos livres (free Africans) or emancipados (the eman
cipated). Upon landing in Brazil, they were dejure “liberated” by the British-
Brazilian Mixed Commission in Rio de Janeiro on the basis o f the treaties
o f 1817 and 1826, and by the Brazilian authorities by the terms o f the law o f
November 7,1831, which finally banned the slave trade.25 Although legally
free, “free Africans” were de facto used in virtually the same way as slaves.
Most were forced to w ork on public projects, for charitable institutions
such as the Santa Casa da Misericòrdia, and in army garrisons and naval
arsenals. Many were also hired out to individuals, until this practice was
forbidden after 1850.26The decree o f 1831 appears to have been only tempo
rarily effective. The total number o f slaves imported from Africa to Brazil
decreased dramatically from 43,000 in 1830 to 3,500 in 1831, but it rapidly
regained previous levels. The number o f imported slaves for the period 1836-
1840 (240,600) nearly matched that for 1826-1830 (250,200). In sum, a sub
stantial number o f Africans continued to be imported into Brazil and sold
as slaves until the end o f the transatlantic slave trade in 1851.27
Enactment o f the anti-slave trade law o f 1831 had a crucial impact on A f
rican slavery in the city o f Salvador; slaves were no longer taken directly into
the port o f Salvador. Slave ships from Africa unloaded their illegal cargoes
clandestinely on the myriad o f islands o f the Bay o f All Saints or, later, at
the mouth o f the Rio d’Una. The west side o f the island o f Itaparica became
the main depot from which Africans were shipped farther in a clandestine
coastal trade or taken directly to the slave market o f Salvador.28The number

o f enslaved Africans imported into Bahia for the period 1811 to 1830 was
142,300. Surprisingly, over the next tw o decades (1831-1850), imports still
amounted to 98,600. However, examination o f the distribution o f these
slave imports for each year and by decades does demonstrate that the law
effectively decreased the number o f Africans imported into Bahia. (The
single exception to this trend is a period between 1846 and 1850, when
45,000 slaves were imported.) The number o f newly arrived Africans sud
denly dropped from 7,000 in 1830 to 1,000 in 1831.29 Since the plantations o f
the Bahian Recdncavo demanded a constant number o f slaves for the agri
cultural labor force, there are reasons to believe that the city o f Salvador
accepted a lower percentage o f new arrivals after 1831. Few documents trace
the exact change in numbers, but, not surprisingly, baptismal registers o f
Santo Antdnio parish for the period from 1808 to 1869 show that the number
o f baptized African-born slaves was already drastically declining in the
1830s.30
The anti-slave trade bill became law (known as the Queir6s Law) on Sep
tember 4,1850; in 1851, the transatlantic slave trade was formally abolished,
after which fewer and fewer slaves continued to be imported to Bahia, as
well as to Brazil as a whole.31 The effective termination o f the transatlantic
slave trade was to promote the rapid decline o f slavery in Brazil.
Urban Slavery in N ineteenth-C entury Salvador:

Slaveholding Patterns and the Decline o f Slavery
Urban slavery had prevailed in the city o f Salvador for more than
two centuries by the turn o f the nineteenth century. Salvador thus matured
fully as a “slave society,” in which, according to the definition o f the late Sir
Moses 1. Finley, slavery was established as the most dominant institution,
both socioeconomically and culturally.32 The distinctive patterns o f urban
slavery and slaveholding had been well established in Salvador.33
Sex Ratios among the Urban Slave Population
O ne o f the major characteristics o f urban slavery in Salvador is

found in the sex balance among the slave population. As is well known by
now, throughout the transatlantic slave trade from Africa to the New World,
males consistently outnumbered females by at least a ratio o f 2:1.* This was
in response to the vast demand for male labor on plantations in the Am eri
cas, based on the Europeans' gender perception o f manual labor as male, but
A uCapital o f Africa” in Brazil 17

it was also a reflection o f the local demands for female slaves in sub-Saharan
Africa.35 According to David Eltis’s calculation, 64.1 percent o f slaves im
ported to Bahia for the period 1811-1850 were males.36 Despite this sex im
balance among African-born slaves imported to Bahia, the urban slave
population in nineteenth-century Salvador was nearly balanced by sex for
both African-born and Brazilian-born, with females even slightly outnum
bering males. Among slaves purchased and sold in Sé parish for the period
1838-1888, male-female ratios were 48:52 for African-born, and 47:53 for
Brazilian-born.37 Male-female ratios in the slave population were 44:56 in
the surviving censuses o f 1855 and 49:51 in the first national census taken in
1872.38 The city contrasted with rural plantation areas, such as the Bahian
Recôncavo, which recorded a much higher proportion o f male slaves as field
hands on the sugar plantations. For example, six Bahian engenhos (sugar
plantations) in 1816 showed 275 males for every 100 females.39
Sex balance in the urban slave population is mainly attributable to the
fact that the city demanded abundant female domestic slaves; almost every
free household owned or rented at least one slave woman as a domestic.
Among all slaves registered in their owners* inventories, which the historian
Maria José de Souza Andrade studied for the period 1811-1888,19.5 percent
o f males and 74.1 percent o f females were domestic servants.40 I consulted
370 inventories o f slave owners registered in Salvador for 1808-1888; 27.1 per
cent of men were shown as engaged in domestic activities for the period
1808-1849 and 20.9 percent for 1850-1888, whereas 76.6 percent (1808-1849)
and 63.1 percent (1850-1888) o f slave women were indicated as domestics.41
The high demands for female domestic slaves in urban slavery were well
reflected in the varying slave prices. Regardless o f sex, the prices o f skilled
domestic slaves were higher than those o f field hands. For instance, in Pilar
parish from 1850 to 1888, the average slave prices o f domestic slaves were
8825096 réis for men and 7725915 réis for women, considerably higher than
those o f field hands, which were 685S071 réis for men and 6155319 réis for
women.42 Even more importantly, the prices o f female domestic slaves were
much higher than those o f male field slaves, despite the fact that men were
more expensive than women in the same occupational category. In the city,
where slave women were often used in skilled domestic services, their prices
increased and occasionally exceeded those o f male slaves. Brazilian-born
women, who usually had acquired more domestic skills than African-born
women, often cost more than their male counterparts, as clearly demon
strated by the data on the prices o f slaves in Sé parish for the period 1838-

i888.43 By contrast, in plantation slavery, in which gang labor was the basis
for a slave system where physical strength was the fundamental criterion
for slave labor, male slave prices were usually higher than those o f female
slaves.44
The Scale o f Urban Slaveholding
Urban slaveholding took place on a much smaller scale than on the

plantations. For instance, in 370 inventories o f slave owners registered in
Salvador for 1808-1888,86.2 percent (319 individuals) owned no more than
ten slaves. W hen further broken down, the figures are even more reveal
ing: 16.4 percent owned only one slave; 13.5 percent owned tw o slaves; and
17.3 percent owned three slaves.45 According to Joâo José Reis, who used
the same source for the period 1811-1850, among 395 individuals who reg
istered inventories, 67.1 percent (256 individuals) owned only one to ten
slaves, whereas 13.2 percent (342 individuals) did not own any slaves.46 In
nineteenth-century Salvador, every household with the financial means
owned or rented at least one slave woman as a maid-for-all-service. There
were exemptions from payment o f taxes on slaves unless the number o f
slaves for the service o f a single person exceeded two; for a married couple,
four; for a big family, six; and where all slaves were under the age o f fifteen.47
Larger households not infrequently included slaves born as the owners' il
legitimate children. Often there was a slave hierarchy, with special maids
called mucamas ( mucumbas) at the top. The mucamas, who were often mu-
lattoes, and might be mistresses or common-law wives o f the owners, served
as supervisors o f other slaves, washerwomen, cooks, housekeepers, and wet
nurses (amas de leite) for the owners’ legitimate children.48
The small scale o f urban slaveholding was partly attributable to the criti
cal limitations o f space in urban households. Furthermore, even for those
families who owned only a few slaves, not all slaves o f the same owner could
lodge under the same roof. In fact, many urban slaves, except for small num
bers o f full-time, live-in female domestics available for all services in the
household, were live-out slaves who often rented the lojas (basements, or
cellars under the house) with other slaves o f the same sex (but often o f dif
ferent owners), or with ex-slaves, often in the same occupations as slaves.49
In Salvador, where no racially segregated section originally existed, a slave
lived side-by-side with, and in the same building as, other tenants who could
be whites, free-born people o f color, and ex-slaves.
A “Capital o f Africa ” in Brazil 19

Wage Earning
Urban slavery in Salvador was also characterized by its wage-earning

(ganho) system, characterized as a “means o f subsistence” for the owner. In
1848, the British Vice-Consul James Wetherell described slavery in Salvador
as follows:
A Brazilian has slaves, he sends them out to work at different trades—

to cultivate the land> to sell vegetables, to hire as servants, as boatmen,
& c.— in fact employs them in every way that servants or workmen are
required. The master directs the slave to pay him at the rate of, it may be,
about one shilling a day; this frequently is the case, and all the slave can
raise above that sum which his master demands, belongs to himself. In
the process o f time those who are industrious raise sufficient money to
pay the price their master values them at, and when such is the case, the
slave can claim his freedom.50
The German travelers Jonathan B. von Spix and Carl Friedrich P. von Mar-
tius stated in the late 1810s that each wage-earning slave (escravo de ganho)
was obliged to return the daily wage o f 240 réis to his or her owner.51 Ac
cording to another source, in the first third o f the nineteenth century, it was
customary for the wage-earning slave to pay the owner one pataca (320 réis)
per day or six patacas on Saturday, although there was no standard sum or
wage authorized by the Bahian authorities.52 In fact, among 194 African-
born ex-slave women who resided in Santana parish in 1849, every one o f
nine washerwomen earned the smallest amount o f 320 réis a day, while oth
ers with specific occupational skills earned much more.53 These data may
confirm that one pataca was an unofficial but widely determined minimum
daily wage for unskilled laborers.
Slaves themselves, as another British traveler, Alexander Marjoribanks,
observed in both Salvador and Rio de Janeiro in the mid-nineteenth century,
applied this wage-earning system o f urban slavery.54To earn extra cash, they
“hired their time out” on a part-time basis in a variety o f urban occupa
tions, such as peddlers o f both sexes, male porters and transporters in work
gangs, and female market-stall keepers.55 This practice should not be inter
preted as suggesting that slavery in the city was less harsh than plantation
slavery. The owner punished the slave terribly should the slave fail to return
to the owner the wage that the owner and slave had mutually agreed upon.
But the wage-earning system o f urban slavery at least enabled the enterpris
ing slave to accumulate enough money eventually to purchase freedom, even

though it took many years. This also led slaves o f both sexes in the city to
be usually multifunctional. A female domestic might also peddle cooked
foods after work and/or involuntarily or voluntarily engage in prostitution.
A male sedan chair carrier (carregador de cadeira) might also work in his
owner’s vegetable garden in the suburbs o f the city. Although notarial rec
ords, such as inventories and registers o f purchases and sales o f slaves, usu
ally mentioned the primary or official use o f slaves, many urban slaves had
more than one occupation or function.56
Decline o f Urban Slavery in Salvador in the 1840s:

From “Slave Society” to “Slave-owning Society”
The above-mentioned distinctive characteristics o f urban slavery

and slaveholding began to change, and some even disappeared, by the m id
nineteenth century. By 1840, Salvador had been ousted by Sao Paulo from
her position as the second most important commercial city o f Brazil; the
economic boom in Bahia o f the late colonial and early imperial period had
already given way to a deep recession. It was in the 1840s that the inter
provincial slave trade started on a large scale, after the slave trade was
banned in Brazil in 1831. As a result, according to the British official James
Bandinel, in Brazil a newly arrived slave in 1840 cost four times more than
had been the case in 1821.57 By the mid-nineteenth century, free labor had
replaced slave labor to a great extent in Salvador, as well as in other major
Brazilian cities, whereas slaves were increasingly concentrated in the coffee
regions through the interprovincial slave trade.58Slaves in Salvador, particu
larly males, were transferred en masse as agricultural laborers to rural re
gions, especially the booming coffee plantations o f the Southeast (Minas
Gerais, Rio de Janeiro, and Sao Paulo).59Interestingly, there was greater like
lihood that urban slaves in Salvador would be sold to the internal slave trade
than would plantation slaves from the Rec6ncavo. This was because urban
slavery in Brazil was “an escape valve for further remunerative slave activity
in the face o f cyclical export stagnation,” as in mainland British America.
W hen the demand for slave labor declined in export-oriented rural-based
activities, the growing excess in the labor supply in plantations and mining
was transferred, rented, or sold for urban activities. Once new export cycles
came into being, there was a significant transfer o f urban slaves back to
rural-based activities.60 This is what took place in Salvador in the 1840s
with the transfer o f urban slaves to the booming coffee plantations in the
Southeast.

The historian Herbert S. Klein, comparing the data on slaves transferred
to Rio de Janeiro in 1852 w ith the national census data o f 1872, demon
strates that skilled and more urbanized slaves, such as male artisans and fe
male domestics, predominated among those being transported in the inter-
provincial slave trade. Am ong the 1,835 slaves who were exported from the
province o f Bahia in 1854,836 were from urban centers, 583 had been in ag
ricultural labor, and 416 were o f unknown occupation. According to K leins
1852 data for Rio de Janeiro, two-thirds o f such slaves were Brazilian-born,
and 73 percent were aged between 15 and 40. He asserts that, given the equal
costs o f transportation, highly priced slaves were more likely to be preferred
as commodities for the internal slave trade.61
It is in this socioeconomic context o f the 1840s that urban slavery in Sal
vador began to decline rapidly, which can be characterized as a crucial struc
tural transformation, namely a transition from a “slave society” to a “slave-
owning society” in Finley’s definition. He maintains that “ [i]f the economic
and political elite depended primarily on slave labor for basic production,
then one may speak o f a slave society.”62 Hence, a slave-owning society is
defined as the one that owned slaves and used slave labor but did not rely
on slaves as the main labor force. Such a structural transformation o f the
larger society took tw o notable manifestations in Salvador: a decline in the
proportion o f the slave population and the larger society’s shared percep
tions o f slaveholding.
The numerical strength o f the slave population was a most important fac
tor in defining the nature o f the slaveholding society.63 Before the 1840s, the
enslaved population had been a numerical majority in the whole population
o f Salvador, and slaves were employed in every major economic activity o f
the port city. The slave proportion o f the population in Salvador declined
rapidly, from 42 percent in 1835 to 26 percent in 1855.64 Thus, we may con
clude that before or around the mid-nineteenth century, free labor replaced
slave labor to a considerable degree.
The change in the form o f labor simultaneously also took place at a cul
tural and ideological level. In a “slave society,” slaveholding did not simply
mean the predominant use o f slave labor, but it also carried status or social
prestige for the individual slave owner. This explains why in a “slave society”
those, including ex-slaves or sometimes even slaves themselves, who did
not own any other form o f property owned slaves.65 Once the slave society
transformed itself into a slave-owning society, slaves were reduced to be
coming either part o f the labor force or commodities to trade. Therefore, in
response to the prospering internal slave trade starting in the early 1840s,

many owners in Salvador chose to sell their slaves to turn a quick profit.
Even after the mid-nineteenth century, the elite in Salvador continued to
own slaves and use slave labor, but the urban society no longer depended on
the institution o f slavery, either economically or culturally. Thus, the trans
formation o f a slave society into a slave-owning society signifies that slavery
was not as important a cultural institution for free urban residents in Sal
vador after the 1840s as before.
Slave prices in Salvador showed a rapid rise at mid-century, after the
interprovincial slave trade started to take place, based on the assumption
that inflation does not significantly distort the raw data. For instance, among
the slaves purchased and sold in Sé parish for 1838-1888, average slave prices
started to escalate in the 1850s, reached their peak in the period 1858-1862,
and started to drop quickly.66 This price movement was largely reflected in
average prices for self-purchase: they rose sharply at mid-century, reached
their peak in 1861-1862, but quickly dropped later in the 1860s.67
This overall movement o f slave prices in Salvador corresponds to that
recorded in the major plantation areas o f the Americas (the U.S. South,
Cuba, and Brazil), where slave prices increased rapidly during the 1850s
and declined slightly in the early 1860s. Slave prices in the mid-i86os re
mained consistently and substantially above those for the years before 1850,
as Manuel Moreno Fraginals, Herbert S. Klein, and Stanley L. Engerman
demonstrate. This price movement is attributable to the importance o f slave
production for the expanding consumer markets in Europe and the United
States in the 1850s.68 Therefore, we may conclude that the movement o f
slave prices in nineteenth-century Salvador did reflect changes in the world
economy, although we cannot deny the interprovincial slave trade's effect as
a contributing factor to, as well as its secondary effect on, rising slave prices
in local markets.
The data on slave prices in Sé parish for the period 1838-1888 reveal dif
ferentiation by sex. The average price for an adult male rose from 4315538
réis (1848-1852) to 753$750 réis (1853-1857), and then to i:069$040 réis (1858-
1862) for African-born; and from 4425941 réis (1848-1852) to 9445964 réis
(1853-1857), and then to 9355830 réis (1858-1862) for Brazilian-born.69 By
contrast, the average price for an adult female did not go up as much as that
for her male counterpart. The average price for an African-born adult fe
male rose from 395S029 réis (1848-1852) to 671S849 réis (1853-1857), but then
fell to 6005581 réis (1858-1862), whereas that for a Brazilian-born adult fe
male rose from 4935571 réis (1848-1852) to 7665763 réis (1853-1857), and then
to 994S964 réis (1858-1862).70 This differentiation by sex in slave price in-
A “Capital o f Africa ” in Brazil 23

creases is attributable to higher demands for male slaves in the interprovin
cial slave trade. Klein’s study reveals a higher proportion o f males among
slaves transported in the internal slave trade than among the general slave
population. For instance, the male-female ratio among slaves transported to
Rio de Janeiro in 1852 was 2:1, whereas sex ratios among the nonmigratory
slave population were 126 males to 100 females in the province o f Sao Paulo
(1836), 145 males to 100 females in the city o f Rio de Janeiro (1849), and 151
males to 100 females in the province o f Rio de Janeiro (1850).71
Around mid-century» when slave prices began to rise, the percentage o f
paid manumissions in every category o f the manumitted (male/female
and African-bom/Brazilian-born) increased, whereas that o f unpaid manu
missions decreased.72 These data on nineteenth-century Salvador concur
with the overall manumission data in nineteenth-century Brazil analyzed
by Thomas W. Merrick and Douglas H. Graham.73 This significant change
in the general patterns o f manumission in Salvador correlates with the im
portant transformation o f urban Bahian society from a slave society to a
slave-owning society, which took place in the 1840s. This structural trans
formation involved cultural and ideological changes. As long as slaveholding
itself symbolized individual social status and prestige in society, many slave
owners preferred to manumit some slaves whom they usually favored, such
as Brazilian-born slaves and women and children, without payment, often
unconditionally, noting in the manumission letters their appreciation o f the
slaves' “good services.” By so doing, a slave owner made a public gesture o f
generosity or humanity, which would, in return, enhance the owner’s pres
tige as a Christian. But, after the 1840s, due to the aforementioned structural
change, slaveholding was no longer regarded as a status symbol. Once the
price o f slaves rose, slave owners were much more open to the idea o f mak
ing profits through sales o f their human commodities and to acceptance o f
cash for self-purchase o f their freedom.
Free labor became the norm in the 1840s. The interprovincial slave trade,
which expanded rapidly at mid-century, transferred large numbers o f young
male urban slaves from Salvador to the coffee-booming Southeast. This had
an impact on slaveholding patterns in Salvador. W hile the enslaved African-
born population continued to age, young Brazilian-born male slaves were
highly valued for sale to the interprovincial slave trade. The urban slave
population in Salvador began to experience an imbalance in terms o f sex.
The use o f slaves was reduced to almost exclusively the domestic domain,
where female servants were most needed. The familiar figures o f street
slaves o f both sexes gradually disappeared from daily life in Salvador.

During the last decades o f the slave regime, Salvador kept a larger part o f
the urban slave population o f both sexes as domestics. Am ong the slaves
registered in legal records for purchase and sale in Sé parish for the period
1861-1888, 66.3 percent o f males (492 individuals) and 81.5 percent o f fe
males (626 individuals) were domestics.74 Domestic slaves o f both sexes ap
peared frequently for sale, purchase, and hire in the slave advertisements o f
contemporary Bahian newspapers.75
Categories for People o f A frican Descent

in N ineteenth-C entury Salvador
The population o f African descent in nineteenth-century Salvador

was divided into tw o categories regarding the place o f birth: African-born
and Brazilian-born; and three categories in terms o f legal status: slave (es-
cravo), ex-slave (forro or liberto)9and free-born (livre).
All people o f African birth, with the exception o f a small number o f “free
Africans” or the “emancipated” after the enactment o f the decree o f 1831,
were brought to Salvador as slaves during the slavery regime. The Brazilian
elite always preferred to import new slaves from Africa rather than repro
duce the existing slave population. As a result, as long as the massive influx
o f human cargoes from Africa continued, the majority o f Salvador’s slave
population remained African-born. Hence, for the first four decades o f the
nineteenth century, slaves constituted around 40 percent o f the whole popu
lation o f Salvador, and two-thirds o f the slave population were African-
born. In Salvador, at least during the early nineteenth century, prêta (black)
meant African-born, either enslaved or freed, and negro (black) referred to
an African-born slave, and neither o f them was applied to the Brazilian-
born population.
The term forro (freed) was originally the generic word for ex-slave, re
gardless o f how the slave acquired freedom, but in colonial Brazil, the term
forro did not always refer to ex-slave; it was used broadly and somewhat con
fusingly. In the sixteenth century, for instance, indios forros were not only
ex-slaves but also the indigenous population who were not enslaved but un
der Portuguese control, especially, though not exclusively, that o f the Jesu
its.76 Until around the end o f the eighteenth century, forro referred to all
legally free people o f color, both free-born and ex-slave, without any clear
distinction. For instance, in eighteenth-century Brazil, colored militia regi
ments were composed o f the legally free population o f African descent, both
free-born and freed, under the same term forro; there is no document left

regarding whether the African-born population was included or not.77 On
the other hand, the term “liberated” (liberto) came to carry a new meaning,
which originated from the British antislavery movement during the first
decades o f the nineteenth century. It became a part o f the official lan
guage in Brazil, and later in the 1840s dominated common usage among
the ex-slave population for self-identification.78 In nineteenth-century Sal
vador, forro and liberto were used interchangeably, except that a child born
to a slave woman and baptized as legally free was always called forro, never
liberto. During the colonial period and the nineteenth century, the city o f
Salvador witnessed the significant presence o f the ex-slave population. No
matter how small a percentage they constituted, it is important to note that
slaves were manumitted with much greater frequency in the cities than in
the rural, agricultural areas; an urban environment was much more condu
cive to the opportunities o f freedom.
The term livre was used to clarify the legal status o f the Brazilian-born
individual o f color, and was not used to refer to the status o f any white per
son; the combination o f branco livre (free-born white) is never found in his
torical documents. To whites, being called livre must have been offensive, as
was also the case o f the antebellum U.S. South, where the term “free-born”
referred only to free-born people o f color, to distinguish them from slaves
and former slaves.
In nineteenth-century Salvador, in notarial records and other official
documents, the Brazilian-born population o f African descent, either slave,
ex-slave, or free-born, were divided into several categories o f color (cdr),
such as crioulo (black); pardo (mulatto); mulato (mulatto w ith pejorative
connotation); cabra (mulatto, often darker-skinned); and mestizo (mixed-
breed in general, but hardly used in the nineteenth century). The term
caboclo referred to a person o f Indian-European ancestry, but the city o f
Salvador included a very small caboclo population, unlike the rural regions
o f Bahia.

Part One
To Be African-Born and Enslaved ,
circa 1808-1831
There are no paths in water. No signposts.

There is no return. To a land trampled by the
muddy boots o f others. To a people encour
aged to war among themselves. To a father con
sumed with guilt. You are beyond. Broken-off,
like limbs from a tree. But not lost, for you
carry within your bodies the seed o f new trees.
Sinking your roots into different soil.
— C aryl Phillips, Crossing the River
2 The Creation of New Identity,
1808-1831
An African-born individual’s creation o f new identity began to take place

at the very moment o f enslavement in the homeland o f Africa. Many o f
these enslaved people were captives o f warfare, pawns, victims o f kidnap
ping, or captives for the services o f oracles. Some parents, because o f debts,
laziness, and insubordination, even sold their own children.1 Whatever the
reasons, when those people lost their freedom, they were forced to give up
their original identities, which were closely associated with their being free-
born and belonging to their own families, lineages, and larger local groups.
They were soon sold by local slavers/merchants into the hand o f European
slave merchants, who shipped them from ports on the coast o f Africa, where
they often had to wait together for days or months for slave ships to show
up. These people were loaded as human cargoes onto slave ships; they forc
ibly endured the long, terrible hardship o f the Middle Passage, which many
o f their fellow passengers could not survive; and after a passage o f a month
or even longer they eventually arrived as commodities for trade in the New
World.
This chapter examines processes by which African-born enslaved men
and women o f diverse ethnic origins created their new ethnic and gender
identities at the workplace. O ur special emphasis will be placed on critical
tensions between attempts by the powerful to categorize/label the powerless
for the sake o f slave control and the latters meticulous manipulations o f
such imposed categories/labels for the purpose o f their daily survival o f
New World slavery.
Enslavement and the Imposition o f

N ew “Cards o f Identity”
Being sold into the transatlantic slave trade did not merely mean the
deprivation o f one’s original identities, but also the imposition o f tw o new
“cards o f identity” by power holders on all enslaved human beings. One was
a Portuguese/Christian name given at baptism, and the other was an African
“ nation” {na$tio).
Enslaved individuals o f African birth were all forced to incorporate them
selves into a New World slave system in a double sense: as human com m odi
ties in a labor system, and also as pagans who had to be converted into the
new religious faith o f Christianity. The State and the Church were essen
tially inseparable from each other in establishing the institution o f African
slavery in Portugal and its overseas colonies; the Portuguese had sought to
justify the trade in slaves and the institution o f slavery on the grounds o f
conversion to Christianity and the salvation o f souls that would other
wise have been lost.2 Therefore every African-born slave was theoretically
obliged to be baptized as a Catholic before or upon arrival in Brazil, whereas
Brazilian-born slaves were baptized at birth.
Henry Koster, who resided in northern Brazil as a sugar planter during
the first decades o f the nineteenth century, gave us a good description o f the
baptismal ceremony o f the enslaved Africans:
The Africans who are imported from Angola are baptized in lots before
they leave their own shores, and on their arrival in Brazil they are to learn
the doctrines o f the church, and the duties o f the religion in which they
have entered. These bear the mark o f the royal crown upon their breasts,
which denotes that they have undergone the ceremony o f baptism, and
likewise that the king’s duty has been paid upon them. The slaves which
are imported from other parts o f the coast o f Africa, arrive in Brazil
unbaptized, and before the ceremony o f making them Christians can
be performed upon them, they must be taught certain prayers, for the
acquirement o f which one year is allowed to the master, before he is
obliged to present the slave at the parish-church. This law is not always
strictly adhered to as to time, but it is never evaded altogether.3
In Angola, on the day before embarkation, slaves were usually assembled

in a church, or perhaps in the main square o f the port, to be baptized. As
a prelude to the following ceremony, a Catholic priest walked among the
slaves, assigned a Christian name to each, and handed the individual a paper
with the name written on it.4As the anthropologist Jack Goody elucidates,
the renaming o f slaves was nothing new: “such new names served to cut the
individuals o ff from their kinsfolk, their society, from humanity itself and
at the same time emphasized their servile status.”5 Goody continues: “ How
ever the renaming was not in itself sufficient, the new name had to be w rit
ten down and handed to the individual.. . . The paper is intended to convey

information to the slave’s purchaser in the New World. But at the same time
it provides a card o f identity, a card to change an identity, which at one level
affects the change it purports to record”6 Thus each Portuguese/Christian
name came to symbolize the individuals lost identity through enslavement.
The other card o f identity, namely an African “nation,” was given to en
slaved individuals o f African birth collectively but functioned to divide
them into a limited number o f categories. The African “nation” was a legacy
inherited from the European custom o f identifying slaves in Africa by “na
tionalities,” regardless o f their specific places o f origin or ethnic affiliations.
Henry Koster mentioned that African-born slaves in Brazil were “known
under the names o f Angola, Congo, Rebolo, Anjico, Gabon, and M ozam
bique.” 7 These names were African “nations” that European slave traders in
Africa imposed on enslaved individuals to reduce them to a limited number
o f categories, before forced departure for the New World. But it fitted into
the European scheme, where people had such identities.
According to Philip D. C urtin, such “ nations”/“nationalities” can be
broadly divided into two group labels. The first was the name o f the port
from which slaves were shipped. For example, the term “M ina” was origi
nally applied to a slave shipped from the old Portuguese fort o f El M ina on
the Gold Coast, but its usage was extended in the nineteenth century to
mean anyone from West Africa. The other was the ethnic or linguistic term,
which was identified with a much larger group. For example, “Nag6 ” which
had originally referred to a subgroup o f the northern Yoruba, was broadly
applied to all Yoruba-speaking peoples. Furthermore, some “nationalities”
were loosely linked to geographical regions o f origin, such as “Congo” and
“Angola.” 8 It should be added that some “nations”/“nationalities” in Africa
were originally the group names by which they were known to other ethnic
groups. The Yoruba called the Nupe “Tapa.” The Yoruba were known to the
Ewe (G£ge) as “ Nagd ” Such group labels for reference by others, o f course,
did not necessarily correspond to exact ethnic group boundaries.
European colonists identified these “ nations” with certain physical types
and cultural heritages, and expressed preferences for the one over the other
when it came to the purchase o f plantation labor for the Americas. Such
stereotyping and perceived qualities, or lack o f them, were reflected in
higher or lower prices o f such slaves.9These characterizations o f slave types
had been present in colonial Brazil and survived throughout the nineteenth
century. Visiting Salvador in 1803, the British merchant Thom as Lindley
characterized slaves from Angola and Benguela as “a sturdy kind o f negroes,
docile to a degree, and very active and lively.” 10 Koster described “Angolas”
The Creation o f New Identity 1808-1831 31

as the best slaves because o f their great attachment, fidelity, and honesty,
whereas “ Minas” were “represented as possessing great firmness o f mind
and body, and ferociousness o f disposition” “ The German travelers J. B.
von Spix and C. F. P. von M artius noted that those from Congo and A n
gola were suited for domestic service because o f their docility and language
skills.12
Thus, with the imposition o f these tw o new cards o f identity, both o f
which signified their new social and legal status as slaves, women and men
o f African birth were forced to learn to perceive themselves as the absolute
powerless in the New World. However, they were not entirely receptive to
such labels/categories. Instead, as we shall discuss, they took an active in
volvement in the creation o f their new identities.
The Creation o f Ethnic Identity
Enslaved people o f African birth in Salvador gradually adopted such

group labels/categories as African “nations” to create and re-create their eth
nic identities, through the Middle Passage experience and in the process o f
being incorporated into New World slave systems.13 In this sense, the asso
ciation o f those who were transported on the same ship and called one
another malungos (shipmates) was a part o f this process o f constructing
and reconstructing ethnic groups. Koster mentioned that the term malungo
was much regarded among African-born slaves and that these individuals
showed “much attachment” to their malungos, as great as to their wives and
children.14 One o f the participants in the Maid revolt (1835), a Nagd slave
who was called Matheos in Portuguese but whose African name was Dad£,
testified that tw o other participants, Belchior and Gaspar da Silva, were his
malungos.15
Historian Paul E. Lovejoy suggests that in the case o f early nineteenth-
century Salvador such “shipmates formed allegiances, and the ethnic back
grounds o f slaves further consolidated these ties” since they were originally
from the same geographical area o f Africa and arrived in Salvador on the
same Portuguese-Brazilian ships.16 As M intz and Price discuss, this “ship
mate” relationship was o f a dyadic (two-person) nature between individu
als o f the same sex, partly because o f the general policy o f keeping men and
women separate on a slave ship. M intz and Price conclude that among the
population o f African descent in the New World these same-sex dyadic ties
that were derived from the Middle Passage experience “became a major
principle o f social organization and continued for decades or even for cen

turies to shape ongoing social relations” and that “the development o f these
social bonds, even before the Africans set foot in the New World, already
announced the birth o f new societies based on new kinds o f principles.” 17
Enslaved people o f African birth, who had begun to develop such same-
sex dyadic ties with one another on slave ships, were eventually brought into
the port o f Salvador. Interestingly, Salvador, the oldest port city o f Brazil,
did not develop such a commercially organized, large-scale slave market as
the Valongo, a long winding street that runs from the sea, located at the
northern edge o f the newer port city o f Rio de Janeiro. O n the Valongo o f
Rio de Janeiro, almost every house was a large warehouse, where the slaves
were deposited and customers came to purchase them, and each house was
sometimes filled with three or four hundred slaves o f all ages and both
sexes.18Perhaps the geographical features o f the port o f Salvador hampered
the establishment o f a large slave market. Instead o f one main street, many
streets o f Salvador served the function o f a slave market. The British mer
chant Thomas Lindley observed the streets and squares o f Salvador in 1803
being “thronged with groups o f human beings, exposed for sale at the doors
o f different merchants to whom they belong, five slave ships having arrived
within the last three days.” 19 There were numerous small houses on vari
ous streets o f the city that functioned as slave markets. Such houses were
particularly concentrated on Rua Nova do Com^rcio o f the lower city, as
well as Baixa dos Sapateiros street,20 near the Black Church o f O ur Lady
o f the Rosary (Igreja de Nossa Senhora do Rosdrio dos Homens Pr£tos) in
the Pelourinho (which means pillory), an old public square where owners
brought their slaves for public whipping, in accordance with laws and regu
lations.21 This public square, located at the administrative center o f the up
per city, had become identified with a pelourinho, which stood at the center
o f this busy square always crowded with street slaves.22
Upon arrival in Salvador, many slaves were traded individually on the
streets. Slave merchants stood with their slaves for sale at busy street cor
ners o f the commercial districts, and touted their wares by yelling to pas-
sersby. Others even peddled their human commodities around residential
districts. Slaves were most commonly sold and purchased one by one on the
streets. This practice continued despite the edict o f the municipal council
forbidding it: “Those who sell and purchase slaves must do so only in shops
situated in houses or stores. The penalty is 30$000 rOs and eight days im
prisonment ” 23Slaves in the city o f Salvador thus saw other slaves being sold
and purchased on the street, and had the opportunity to learn how this busi
ness o f trading in slaves worked. Some may have made a single purchase o f
The Creation o f New Identity, 1808-1831 33

a new arrival at a cheaper price for investment, whereas others got involved
in the business themselves.
Unlike the case o f plantation slavery, in urban slavery in Salvador, the
majority o f slaves worked and even lived outside their owners' houses. The
slaves constantly faced the danger o f kidnapping; slaves were often stolen
and subsequently resold on the streets. O n July 22,1830, for instance, a Por
tuguese man, a resident o f the suburban parish o f Brotas, was found guilty
o f having sold slaves who had been kidnapped in the city.24On M ay 20,1834,
when the police inspected a house in the Barroquinha o f Sâo Pedro Velho
parish, they found two slaves who had been kidnapped and hidden by the
Portuguese Francisco Antônio da Fonseca. These slaves were a newly arrived
Nagô slave boy owned by José Barboza Madureira and an African-born
woman, also Nagô, a slave o f Domingos Jorcome Ferreira.25 In and around
the port city o f Recife o f Pernambuco during the early nineteenth century,
the Brazilian historian Marcus J. M . de Carvalho documented fascinating
cases in which slaves actually bargained with their prospective owners on
the condition o f being kidnapped,26but there are no such documented cases
in the city o f Salvador. Yet, there was indeed a thriving trade in kidnapped
slaves from Salvador to the ports o f southern Brazil.27 Not only whites but
many o f African descent, including African-born and enslaved, were en
gaged in this interprovincial slave trade. In November 1831, in Rua do Passo
parish, tw o prêtos, José Francisco and a certain Bonfim, the latter the slave
o f Dom ingo Ferreira Bastos, were imprisoned for having kidnapped and
sold a prêta named Vitôria, who was a slave o f Bento Martins da Costa.28
O n July 19,1831, in Vitôria parish, a crioulo, Luiz Xavier, sold a prêta, Joanna.
Joanna proved to be the slave o f Antônio Diego de Sé Barreto, and the mu
nicipal police arrested Luis and his accomplice, a prêto by the name o f Joâo.29
Thus, survival in the urban slave society entailed, among other things, es
caping the danger o f being kidnapped and resold, and avoiding falling vic
tim to other urban crimes.
Daily interactions among urban slaves on the streets transcended the
boundaries o f both ownership and geography. This makes a very sharp con
trast to plantation slavery, in which each plantation functioned as a virtu
ally independent entity for human interactions under an organized labor
system. Urban slavery gave slaves considerable latitude to move as wage-
earners. Newly arrived African-born slaves continued to receive a m ini
mum amount o f guidance or occupational training from their owners. They
were usually hired out without any special occupational training in domes

tic service or craft skills, or even comprehension o f the Portuguese lan
guage. M uch o f this their Brazilian-born counterparts generally learned
from birth. Among slaves registered in their owners’ inventories for the pe
riod 1808-1849» all o f the men whose primary occupations were transporta
tion or porterage were African-born. They were even more noticeable in
fieldwork and as full-tim e wage-eamers than their Brazilian-born counter
parts, many o f whom were engaged in artisanal occupations and domestic
work. Among slave women, one-fifth o f the African-born were hired out as
wage-eamers on a full-tim e basis, whereas the majority o f the Brazilian-
born were primarily domestic servants. By contrast, very few Brazilian-born
slaves o f either sex were engaged in full-tim e wage earning.30
As discussed in chapter i, urban households did not require large-scale
slaveholding. Their needs were limited to full-tim e domestic servants, many
o f whom could also be engaged in part-time wage earning in the streets.
Thus, slaveholding in the city largely took the form o f cash earning. People
may have purchased African-born slaves for personal service but found hir
ing them out on a full-tim e or part-time basis could be income enhancing.
For those new arrivals who did not comprehend the Portuguese language or
Luso-Brazilian culture, it was especially important to associate w ith other
slaves who at least spoke the same language in order to enter an urban slave
system.
In Salvador, many slave women and men were hired out for wage-earning
activities on the street without the direct supervision o f owners. For the en
slaved individual, every street o f Salvador, which was ill-paved with irregu
lar, sharp, and large stones, was the workplace.31 Slaves o f both sexes worked
barefoot, suffering from the annoying bicho da pi, a tiny insect that easily
penetrated the skin o f their feet.32 For those slaves, the street served as the
most im portant m eeting place on a daily basis. W henever African-born
slaves encountered other African-born slaves on the street, all tried to dis
tinguish “their people” from others by superficial physical features, as well
as fecial marks received at their initiation rites in Africa.33 James Wetherell
observed:
Most o f the blacks have their feces marked in a peculiar manner with
scars or cicatrices, and each nation or tribe differs in some respect from
another. The most numerous, the Nago blacks, for their national mark
have three small cuts in the centre o f the forehead. The Benguela have 5,
7, or 11 small nodules of flesh in the centre of the forehead, forming a line
The Creation o f New Identity1 1808-1831 35

of warts from the roots of the hair to the nose. A very difficult and pain
ful mask to produce, I shall imagine. The cuts, being so much simpler,
appear to be more generally adopted__ 34
Small wonder that new arrivals from Africa could be easily distinguished by
others from those who had lived in Brazil for many years and were there
fore fairly acculturated, since the former's external appearance and cultural
behaviors were not Luso-Brazilianized. However, even among the latter, the
ethnic affiliation o f the individual could be identified by phenotypical, be
havioral, and cultural traits; the African-born population were keenly con
scious o f ethnic diversity among themselves.
If a stranger on the street appeared to be from a certain ethnic group o f
Africa, another individual would talk to the stranger in their native lan
guage, or might greet the person in terms and actions appropriate to their
native culture to express mutual respect. Even when chained with other
slaves in the same work gang, slaves would not pass by their fellow persons
without making some friendly gesture. This was described by outsiders and
foreign visitors as the “politeness o f blacks ” Wetherell observed this trait in
1842: “The Blacks seem naturally polite; they never pass without saluting by
removing their hats, and when they meet, they crack the two first fingers o f
the right hand, each taking hold o f the other’s hand for the purpose, in the
same way as if they were going to shake hands. If a black woman passes a
man seated, w ith whom she is acquainted, he not only takes o ff his hat but
rises from his seat.”35 Wetherell also noted cultural continuity among A fri
cans in Salvador: African-born people paid special respect by kneeling be
fore those who had been “princes” in their homelands.36 W hen they met
their people on the street, they used the special terms o f salutations, such as
“ Selamat” (“ I congratulate you on your safety” ), “Teijibeen” (“ I hope you
are well” ), “Ogirai” (“G ood m orning” ), and “Occuginio” (“ I hope you have
risen in health” ), none o f which Wetherell identified with a specific African
language.37 In New World slavery, differences o f status and some hierarchi
cal orders originating from within native African cultures were also stressed,
modified, and maintained among the African-born slave population o f di
verse ethnic origins, as well as in relation to the larger society. The American
James Henderson observed in 1821 that the slaves in Salvador, assembling in
a plaza o f the upper city on Sundays and holidays, “frequently selected one
from amongst the rest who was dignified with the title o f chief, and received
all the homage bestowed upon a chief in their home country.”38 Henderson
stated further that a friend o f his, passing through the plaza, had seen slaves

conducting “the ceremony o f executing, or putting to death, white men,
which were represented by effigies dressed for the purpose; this was in
tended for the amusement o f their ch ief”3’ Activities such as this are patent
substantiation o f persisting African cultural characteristics and therefore
refute the notion that the experience o f the Middle Passage left Africans de
void o f humanity and common sense.
In Salvador, many o f the African-born population continued to commu
nicate w ith one another in their native languages in their daily lives.40 In
1814, after a major slave uprising took place, slave owners sent a petition to
the governor o f Bahia (1810-1818), D. Marcos de Noronha, Count o f Arcos,
in which they noted: “ Finally, negros crowd at night into the streets as be
fore; they freely converse in their own languages; and they continuously use
whistles and other passwords. They even talk openly in our language which
they have come to understand.”41 W hile discussing the diversity o f the en
slaved African-born population and their native languages, which was sup
posed to lessen the danger o f a slave revolt, I. B. von Spix and C. F. P. von
M artius noted in the late 1810$ that slaves in Salvador understood one an
other at least by means o f certain expressions despite the fact that they lived
distant from one another. They interpreted many African languages as hav
ing great similarities to one another and as forming an extreme contrast to
those o f the indigenous populations.42 O f course, we do not know exactly
to what degree these foreign travelers* statements were valid. Diverse A fri
can languages, none o f which bore any resemblance to any European lan
guages, could have sounded similar to those German travelers, who may
have been struck, above all, by the superficial “ racial” homogeneity o f the
large enslaved population in Salvador, as “blacks.” The Bahian historian
Nina Rodrigues believes that the Gêge (Ewe) language was used throughout
the eighteenth century and until at least the middle o f the nineteenth cen
tury. Tapa and Hausa were spoken during the entire nineteenth century.
Above all, Nagô continued to be a lingua geral (general language) among the
African-born population, even in the early twentieth century.43As late as the
late 1930s, the American sociologist Donald Pierson stated that he had heard
Nagô spoken in Salvador and had met people whose command o f Nagô was
comparable to their command o f Portuguese.44
Furthermore, the African-born population also commonly referred to
one another by their African names, not by their Christian/Portuguese
names given at Catholic baptism before or upon arrival in Brazil, by which
they were widely known in the larger society.45 Their usage o f their A fri
can names may also suggest a shared attitude toward the larger society. A f
The Creation o f New Identity; 1808-1831 37

rican names had specific meanings in their native African languages, and
their continued usage in the New World could be essential for African-
born peoples’ (re)creation o f ethnic identity. O n the other hand, Portuguese
names officially bestowed on slaves by their owners were devoid o f any sym
bolic significance and could have been viewed by their fellow enslaved indi
viduals as tantamount to so many numbers. This is clear evidence that these
enslaved people did not lose their African identities, although those identi
ties were submerged in the New World.
Ethnicity in New World slavery was not simply a duplicate o f ethnicity
in their homelands o f Africa. Franklin W. Knight and Margaret E. Crahan
are accurate in stating that “Specific ethnic identifications did not necessarily
indicate the retention o f the formerly associated ethnic culture__ Repre
sentative ethnic cultures survived, but without necessarily reflecting closely
the basic original mix o f the African locality and religion.”46 Ethnicity needs
to be perceived as “a process o f interaction between ‘objective category* and
‘subjective identity* within a variety o f social environments ”47
In the case o f Salvador, at his trial after the Malê revolt o f 1835, one Nagô
defendant, the slave Antônio, whose profession was a chair carrier, testified
in court as follows: “ [W]e all are Nagôs, but each o f us has his/her own
homeland [ terra] .,ï48 This statement is worth noting, being uttered as it is in
the context o f his self-defense as Nagô, the African “nation” that represented
the majority o f participants in the uprising. Antônio identified his “nation”
as Egba and referred to the slave José o f the same owner as Oyo. Another
Nagô slave, José, testified that he was from the Jabü “nation” (Ijebu), whereas
the Nagô woman Edum was identified as Nagô-Ba (Egba).49 These state
ments exemplify the creation o f ethnic identity in New World slavery. Those
who were from the same African language group, but from different lineages
and/or subdivisions o f an ethnic group associated with one another, often
under the label o f an African “nation ”
These people sought and continuously created and re-created the com
mon ethnic symbols that they could share among themselves, as well as
some hierarchic order, perhaps based on their African lineage systems but
in relation to the larger society. Yet coming to share a specific ethnic identity
in common under New World slavery does not really mean that all such in
dividuals belonged to the same ethnic group or took collective actions as
members o f the same ethnic group. As J. C. Mitchell clarifies, we should dis
tinguish between “ ‘ethnicity’ as a construct o f perceptual or cognitive phe
nomena on the one hand, and the ‘ethnic groups’ as a construct o f behavior
phenomena on the other.”50

The Creation o f Gender Identity
Enslaved people o f African birth started to cultivate their gender

identities during the Middle Passage. We have already discussed that on
slave ships men and women were separated from each other, that therefore
shipmate relationships called malungos were same-sex dyadic ties, and that
such relationships played an important role in connecting enslaved peoples
o f African birth in New World slavery. Unlike the case o f plantation slavery,
in which men largely outnumbered women and sex did not critically deter
m ine the work an individual slave had to perform as a field hand, such
same-sex, dyadic ties as malungos might well be maintained rather easily
and were even strengthened for the creation o f gender identity in Salvador,
because o f the unique nature o f New World urban slavery. As discussed in
chapter 1, the slave population in Salvador was balanced by sex, mainly be
cause o f the city’s high demands for female domestic slaves. African-born
slave women were physically as “visible” in the public sphere as their male
counterparts. At the same time, one’s sex as a slave critically determined the
types o f work and activities in which one had to and/or wanted to engage,
which fostered the African-born people's creation o f gender identity in ur
ban slavery.
The historian Jacqueline Jones, in her study o f the U S. South, discusses
the gender division o f labor under slavery: “ (T]he more freedom the slaves
had in determining their own activities, the more clearly emerged a dis
tinct division o f labor between the sexes”51 Her statement would equally
apply to the case o f the enslaved population o f African birth in nineteentb-
century Salvador. It is true that one’s sex primarily determined how one
would be used as an urban slave, but the flexible labor arrangements in ur
ban slavery gave much freedom for the African-born population to (re d e
fine gender relations and create new gender identity.
In the city o f Salvador, enslaved men o f African birth were used as por
ters and transporters who worked in gangs. These male work gangs were
often composed o f slave men who shared the same “nation” and who usu
ally did not belong to the same owner because o f the small scale o f urban
slaveholding. Such male associations at work appear to have been voluntary
and may be interpreted as a notable manifestation o f a collective ethnic
identity among African-born slave men, but the associations could have
been initially devised as an effective labor arrangement by slave owners who
intended to make the best profits by hiring out newly arrived African-born

slave men to the streets. Relatively acculturated slave men o f African birth
with sufficient occupational skills could train newly arrived male slaves in
the same African languages. This arrangement did not threaten slave owners
because those slaves did not belong to the same owner.
In every busy street o f Salvador, at any time o f the day, there was the sight
o f tw o slave men carrying a hammock or sedan chair on their shoulders.
Sedan chairs for wealthy people consisted o f “a cane arm-chair, w ith a foot
board and a canopy covered w ith leather; curtains, generally o f moreen,
with gilt bordering and lined with cotton or linen, are contrived to draw
round, or open at pleasure; and the whole is slung by the top to a single pole,
by which tw o negroes carry it at a quick pace upon their shoulders, changing
occasionally from right to left.”52 Whereas, depending on the social status
o f the occupant, carriers o f sedan chairs might be in uniform, male porters
were nearly naked, dressed only in the scantiest pair o f casse cotton drawers.
They carried small things on the heads, but large objects were slung between
tw o poles which were carried on the shoulders.53 lames C. Fletcher and
Daniel R Kidder described these male street gangs (1839) as follows:
Immense numbers o f tall, athletic negroes are seen moving in pairs or

gangs o f four, six, or eight, with their loads suspended between them on
heavy poles. Numbers more o f their fellows are seen sitting upon their
poles, braiding straw, or lying about the alleys and corners o f the streets
asleep, reminding one o f black snakes coiled up in the sunshine. The
sleepers generally have some sentinel ready to call them when they are
wanted for business» and at the given signal they rouse up, like the ele
phant to his burden.54
As they moved with their heavy burdens, they sang a kind o f chorus, which
consisted o f “one o f the blacks chanting a remark on anything he sees, and
the others come in with a chorus o f ridiculous description, which is seldom
varied, however much the recitative solo part may.”55Each grouping o f these
street male workers was obliged to have a captain or leader, who organized
the work, set the fees, and took responsibility for payment o f municipal
taxes and licenses for his group.56
Slave men played a crucial role in communication between Salvador and
the Bahian Rec6ncavo, which was very difficult even after the turn o f the
nineteenth century. The roads were no better than “mule-tracks,” and all
merchandise was carried in small packages by horses and mules. There were
hardly any direct routes to any destination in the countryside, and the con

veyance often took hundreds o f extra miles. Carriage by road was thus very
costly and laborious.57 Sugar planters and their families spent part o f the
year in their houses in the upper city o f Salvador, and when they went to the
city some slaves from their rural properties who served as porters and do
mestic servants always accompanied them. The Bahian Recdncavo has sev
eral major rivers: Sao Francisco, Vermelho, Joannes, Jucuipe, Pojuca, Guassu,
Paraguassu, and Serigy, which played an important role for the daily trans
portation o f people and things throughout the nineteenth century. Slave
men were employed as canoeists on the rivers for the transport o f passen
gers. Slave paddlers propelled canoes while singing a monotonous chorus,
even sometimes raising the paddles completely out o f water and striking the
flat part o f them with their hands, keeping time.58
Slave men also transported agricultural products by water to the city. The
Bay o f All Saints and the SSo Francisco River were the two great, distinct
market zones o f Bahia.59 The urban inhabitants o f Salvador depended on
the Recdncavo for their daily diet. For instance, Itaparica, whose principal
product was sugar, supplied Salvador with poultry, vegetables, and fruit.60
Every day, numerous country boats, launches, sailing boats, and other types
o f vessels crossed the bay laden with produce for consumption in the city.
Canoes arrived from places twenty to thirty leagues away from Salvador.61
James Wetherell described in 1848 the arrival o f a barca in the city as follows:
Upon a barca reaching the city, a scene o f confusion ensues, boats put o ff
with market women to besiege with the new arrival, large crates o f fowls
are borne o ff by the fortunate purchasers, ananas [sic] and oranges are
piled in golden heaps, the shore boats are quickly laden with cabbages,
yams, sugar cane, pumpkins, or melons, each [sic] freighter hastening
to her market-stand to make the best o f her bargains by retailing. The
barca then draws near the quay to discharge the heavier part o f her
cargo, puncheons o f spirit, cattle, or empty packages sent down to be
returned filled with European productions.62
Yams, manioc flour, corn, rice, and many other vegetables, as well as tobacco,
wood, coral, fish, and even cattle were thus constantly brought from the
Rec6ncavo to Salvador by water by slave men.63 Chewing tobacco was very
popular among the population o f African descent in the city.64
Male stevedores, most o f whom were hired-out African-born slaves, were
the most common figures among workers o f African descent in the port o f
Salvador. They used large wooden carts for the conveyance o f sugar to the

place o f shipment.65These manual laborers also carried heavy loads on their
heads, or on dollies, from ships to waterfront warehouses and from ware
houses to the commercial district o f the lower city.
At the harbor, slave men moved commodities and people in canoes and
lighters from ship to shore. Besides small sailing vessels, one o f the most
common vessels was the falua> with tw o masts, each with a large sail, manned
by four to eight slave oarsmen, and with a covered stern to protect customers
from the sun.66It was also the role o f slave men to bring newly arrived slaves
ashore for auction in the slave markets.
Many male slaves engaged in fishing in Salvador, which was chronically
short o f meat but endowed with an abundant and varied supply o f fish from
the bay. Fish was sold at a high price, particularly during Lent.67 O n the
sandy beach, slaves were employed to catch shellfish, oysters, mussels, and
turtles.68 Every morning, in the bay near Rio Vermelho parish, slave fisher
men manned forty to fifty launches in search o f whales. These launches had
one large lugsail and were very fast. Every launch was composed o f ten crew
men: eight oarsmen, one patron, and one harpooner. According to the ob
servation by the British traveler Robert Elwes (1848), the crew o f a whale-
fishing launch was “often dressed in costume; one crew in red, red woolen
shirts, red trousers and red caps, another in blue, and & c W h a l e s were
brought into the warehouse called arma^Ho, where they were processed for
food and oil.69
African-born slave men also engaged in larger-scale marine activities.
These ranged from the coastal trade between Brazilian cities to the trans
atlantic slave trade. A considerable number o f men o f African descent chose
to engage in transatlantic trade, developed networks exclusively among
men, and, as a result, made a significant contribution to the transatlantic
communication among peoples o f African descent.70 In the case o f men o f
African birth, their ethnic origins and backgrounds seem to have been an
advantage for their profession as slave traders. Those who served as canoe
ists, oarsmen, or sailors were generally African-born, many working also as
wage-earning slaves. Captains o f slave ships hired slaves as cooks, cabin
boys, seamen, and barber-surgeons.
Many slave men o f African birth were also trained in various artisanal
skills, because o f the lack o f skilled free labor. Such enslaved men were used
as carpenters, joiners, masons, pavers, printers, sign and ornamental paint
ers, carriage and cabinet makers, coopers, sculptors in stone and carvers o f
saintly images in wood, silversmiths, lamp makers, jewelers, and lithogra
phers. Nearly a quarter o f the male slaves listed in their owners* invento

ries for the years 1811-1888 were artisans.71 As the historian Lyman John
son points out, in many colonial Latin American cities, purchasing a slave
to train him as an artisan guaranteed a master artisan a profitable return
on his investment. Since the artisan as slave owner turned a profit on the
difference between the cost o f the slave's subsistence and the value o f his
production, the owner’s profit increased as the slave’s skills improved. Fur
thermore, sales by a slave artisan provided the owner with a profitable sup
plement to his regular business activities.72 The use o f slave labor also gave
a master artisan a greater degree o f control over his workshop.73 In accor
dance with the degree o f skill, slave artisans were classified in their ow n
ers’ inventories as mestre (master), official (skilled), and apprendiz (appren
tice).74 The British traveler Thomas Ewbank stated that a vicar in the city
o f Rio de Janeiro had mentioned a slave who was a first-rate workman in
Bahia.75 Many such skilled slave craftsmen were hired out by their owners.
The variety o f street peddling and vending by slaves was not exclusive in
terms o f gender; both men and women were widely engaged in daily ped
dling and door-to-door vending. Some, such as butchers’ slaves who sold
fresh meat by peddling, were ordered by their owners to do so on a full-time
or part-time basis. Others hired themselves out on Sundays, holidays, or
late in the evening after their obligatory work as domestic servants or field
hands. Urban slaves used their free time to grow vegetables in their owner's
gardens and in provision gardens attached to suburban plantations, or to
cook foodstuffs in their owner’s kitchens and on the streets, or to engage in
prostitution as part-time hustlers. Should they be successful as vendors,
owners might send them out as full-tim e peddlers. Street peddlers offered
for sale whatever they could carry in a basket or tray on their heads: various
vegetables, fresh or cooked foodstuffs, oranges, limes, and tropical fruits,
cakes and candies, eggs, sugar canes, bundles o f firewood, and even shoes
and children’s toys.76 These street peddlers, most o f whom were African-
born, visited every house, calling out each item’s name in rhythmic chants.
The residents o f Salvador were also supplied w ith water from fountains by
these slave peddlers. J. B. von Spix and C. F. P. von M artius mentioned in the
late 1810s that the fountain o f Cam po de Sao Pedro was regarded as the best
in Salvador.77 Water was carried from the fountains in jars or small barrels,
generally on the shoulders o f men, on mules, or in carts, and sold to every
household at a modest price.78
Despite this prevailing practice o f street vending by slaves o f both sexes,
marketing in the city and all aspects o f petty commerce were virtually mo
The Creation o f New Identity,, 1808-1831 43

nopolized by African-born women. African-born women’s marketing ac
tivities were prevalent in Salvador at least by the beginning o f the 1830s.79
M ary C. Karasch points out elite perceptions in colonial and nineteenth-
century Latin America o f “those who sold groceries and liquor as lower
in status, especially i f they sold to the poor and enslaved.”80 In the British
Caribbean, which was described as where “ (n]egroes are the only market
people” and “ [njo body else dreams o f selling provisions,” slave marketing
activities might be divided into tw o categories: one was carried out by itin
erant slave peddlers, usually favored women who sold such household prod
ucts as beeswax candles, seeds, and vegetables from the manager’s large gar
den; and the other by slaves who bartered or sold their own commodities.81
In cities, slave women were generally in a much better position than their
male counterparts in terms o f market participation, since the former could
sell on the street what they cooked as domestics in the kitchen o f their
owner’s house. “House” and “street” thus formed a continuous, single living
space for slave women.82Unlike their male counterparts, many slave women
simultaneously belonged to tw o different social/cultural spheres: the slave
owner’s household and the street.83 This may explain the prevailing domi
nance o f women in the internal market system in the cities o f the Third
World, where poor women usually work as domestic servants.84 Further
more, men did not wish to engage in such a “woman’s job” as street vending,
which was not valued in their community. The anthropologist Marisol de la
Cadena states in her case study o f a contemporary “ Indian” com m unity
near Cuzco: “ Prevailing male view holds that the sale o f products is not
work because it is done ‘sitting down.’ For them this labor is secondary and
deviant— the fruit o f masculine work.”85
African-born slave women were prominent as quitandeiras in nineteenth-
century Salvador. The term quitandeira referred to women who sold items
at their own market stalls (quitandas or vendas) on the street or in the mar
kets. In Salvador the daily green-market located on the beach was well sup
plied w ith tropical fruits and vegetables, which were brought from the
nearby coast by slave men in small launches or boats. The market was filled
at an early hour o f the morning with coconuts, plantains, oranges, melons,
jacks, papayas, guavas, mangoes, and tamarinds, as well as yams, manioc,
peas, beans, and cucumbers.86According to a Bahian historian and professor
o f Greek named Luis dos Santos Vilhena, during the eighteenth century
market stalls were concentrated in three places in the city: the waterfront o f
Conceiâo da Praia parish, the square o f Terreiro de Jesus located on the

Street now called Nova in Sé parish, and outside the St. Benedict monastery
o f Santana parish.®7
Most market-stall keepers and market vendors (mercadejas) were women,
especially those o f African birth. Wetherell described market women in Sal
vador (1848) as follows:
Amongst heaps o f fruit, vegetables» 8c c., shaded by mats, which are some
o f them formed into something like huts, in others only propped up by
sticks, are seated the black market women. They are dressed in highly
characteristic but picturesque dresses o f many diversified colours, but all
o f the same fashion. Some have their infant children slung across their
backs with the “ Panno da Costa” (coast cloth shawl), others with heavy
baskets o f fruit and edibles on their heads. Little children, whose only
article o f clothing consists o f bracelets, ear-rings, and a band round the
body o f coral heads, squatted on flat wooden dishes, like Indian gods.88
Market-stall keepers and market women sold a diversity o f commodities:

cooked ethnic dishes such as fried cake o f cooked beans called acarajé o f
West African origin (akara in Dahomey), carurú (shrimp and okra stew),
and vatapá (fish-based paste), candies, cakes, fruits, vegetables, mingáo (pap),
half-salted meat, pork fat, and whale bone.89Fresh meat was very expensive
and usually sold on wooden platters and trays on the streets, despite the fact
that the municipal council prohibited such sales on penalty o f 30$000 réis
or eight days o f imprisonment (1847).90
African-born slave women virtually monopolized commodities at the
market. Fishermen did not sell to just anybody, even i f they could afford
the highest prices. Instead fishermen had only their relatives as customers.
Vegetable and fruit growers directly delivered their produce to the market
women.91 At the market, exclusively women sold fish. There was a commer
cial relationship between market-stall women and male workers o f African
descent at the whaling station in the suburban parish o f Rio Vermelho. After
extracting oil from the whales, the male workers sold whale meat to the
market-stall women, who cooked it and sold it on the streets.92 Wetherell
stated (1847) that blubber was brought to market already cooked, wrapped
in banana leaves.93These women sold fresh, salted, and dried fish from their
stalls, for example, outside the St. Benedict monastery or in the square in
front o f Barroquinha.94W hile there is no written documentation to support
the assertion, we can assume that women developed their own commercial
associations as independent entrepreneurs. These were not the domain o f

specific ethnic groups, but ethnic considerations did become a factor in dis
tribution networks, such as for fish, and determined locations o f stalls in
markets and on the streets. Newcomers underwent an informal apprentice
ship as peddlers or assistants to established market-stall women, and later
might be assisted by others with the practical experience, networking, skills,
and even capital to install their market stalls.
Sharing another special gathering place further strengthened association
among African-born slave women: the river and public fountains for laun
dry, which were exclusively for women. Slave women carried heavy laundry
in barrels or in bundles on their heads to a stream or fountain in the neigh
borhood. One district in Salvador where laundresses gathered is still called
“Barris” (barrels) after the number o f barrels sunk into the marshy ground
to form wells o f water. The laundresses, who were half-naked or entirely
naked at work, washed the clothes by beating them against large stones with
great force. The fruit o f a tree or horse dung was used in lieu o f soap. After
washing, they spread the clothes in the sun to dry. Throughout these busy
operations in the open air, these women talked to one another at the top o f
their voices, while their little children sprawled about playing, protected
from the sun’s heat by a cover made o f a propped-up mat.95 After the laun
dresses returned the clean laundry to the owners’ houses, they or other fe
male domestics known as engomadeiras, who specialized in starching and
ironing, would continue the arduous process o f bringing the final product
to perfection.
Such opportunities for association among African-born slave women
played critical roles not only for distributing and sharing information but
also in breeding a sense o f collective identity. An individual’s daily engage
ment in gossiping was essential for him or her to become and remain a
member o f a community.96 African-born slave women were most likely to
create and develop a much stronger collective gender identity, beyond ethnic
identity, in New World urban slavery than their male counterparts.
Conclusion
At the time o f enslavement in the O ld World, African-born women

and men were officially deprived o f their original identities as free, and the
powerful imposed the tw o new cards o f identity as enslaved in the New
World: Portuguese/Christian names and African “nations.” But it does not
mean that all such enslaved people o f African birth were entirely passive or
receptive to the larger society’s labeling/categorization o f them for the sake

o f slave control; they continued to communicate among themselves in some
o f their native languages and identify themselves and refer to one another
by their originally given African names.
The enslaved population o f African birth soon managed to find certain
effective ways to manipulate such labels as African “nations” for their own
benefit in an urban slave society. The urban slave population was com
posed o f diverse ethnic groups, and therefore they were quite heterogeneous.
W hen they came into contact as strangers they also needed to categorize one
another in urban slavery and chose to adopt African “nations” for categori
zation. Categorization is always a two-way process; in differentiating others,
one is also defining oneself. Epstein maintains that “ethnic categories always
have a dual aspect: they are at one and the same time both ‘objective,’ that
is external to, or independent of, the actor, and ‘subjective,’ that is internal
to the actor, a perception o f the self”97 Through this dialectic process the
enslaved population o f African birth created their ethnic identities in the
New World.
But the creation o f ethnic identity did not really embrace men and women
o f African birth exactly in the same way; gender divided people who shared
the same ethnic identity in Salvador. The uniqueness o f urban slave systems,
wage earning in particular, enabled the enslaved population o f African birth
to create and develop distinctive gender identities at the workplace. African-
born slave men fostered and reinforced their gender identity by sharing the
same ethnic identity w ith those o f the same “nation” in work groups. By
contrast, slave women created their own gender identity, which oftentimes
transcended divergent ethnic identity among the urban slave population o f
African birth.

3 The Representation of Identity,
1808-1831
The enslaved population o f African birth in Salvador, who shared the same
cruel “fate” o f enslavement in Africa and the hardships o f the Middle Pas
sage, soon began to get connected with one another as they were being in
corporated into urban slave systems in the New World. Oftentimes their
interpersonal connections were the clear reflection o f sharing common eth
nic and/or gender identities. Sharing o f the same identity clearly stimulated
and encouraged the enslaved African-born population to form groups, es
tablish voluntary associations, or rebel militantly against the power systems,
for instance.
This chapter w ill investigate the representation o f ethnic and gender
identities in a collective form: daily gatherings and groupings; mating, con
sensual unions, and fictive kinship; voluntary associations; and flights and
uprisings.
D aily Gatherings and Groupings:

Cantos , Batuques, and Candom bU
African slaves occupied the urban public space o f Salvador in ac

cordance with their collective ethnic identities. Those who came to share
the same ethnic identity tended to visit and stay at a specific meeting place,
namely a street com er called canto. On work days, the cantos functioned
as meeting places for wage-earning slaves. At these meeting places, while
seated on low stools with three legs, waiting to be summoned by regular and
casual customers, slaves made straw hats, small baskets, iron chains for par
rots, birdcages, and bead bracelets and polished leather or repaired umbrel
las, usually made o f yellow cotton cloth, which African-born slaves carried
wherever they went.1
The cantos, which generally took names from their locations, such as the
street or square, existed all over the city and represented specific ethnic
groupings. The contemporary historian Nina Rodrigues located the cantos
in Salvador. For instance, in the lower city, the cantos o f Rua do Comércio
and Rua das Princesas were exclusively for Nagôs. In the upper city, cantos
o f Nagôs included those o f Rua da Ajuda, Largo da Piedade, and Ladeira de
Sâo Bento. In the canto o f Cam po Grande, up from the Forte de Sâo Pedro,
Nagôs also gathered, but not exclusively, since three or four Gêges were in
cluded in the group. Hausas had a canto between Arcos de Santa Barbosa
and the Hotel o f the Nations. Minas tended to congregate in the canto o f
Sào Raimondo on Rua das Mercês. These canto locations were exclusively
for men, as gender identity was also clearly manifested in the cantos. African-
born women gathered in cantos on Rua da Vala near Sâo Miguel, on Rua da
Guadelupe, on Rua da Cabeça, in the Largo Dois de Junho, and on the
Ladeira do Boqueirâo o f Santo Antônio parish.2 According to Nina Rod
rigues’s observations, cantos for women generally manifested less o f collec
tive ethnic identities than those o f men.3 Considering the fact that slave
women o f African birth developed a common gender identity at the work
place across ethnic boundaries, as discussed in chapter 2, Rodrigues seems
to have been correct in pointing out gender differentiation in African-born
slaves’ participation in cantos.
Ethnic identity collectively shared by male African-born slaves gathered
at a canto was strengthened by the presence o f a few itinerant barber-
surgeons o f African birth who accompanied the canto. These African-born
men cut the hair o f male street slaves and performed rudimentary medical
procedures.4 Wetherell described the barber-surgeon in Salvador (1848) as
“not only expert at shaving and hair-cutting, but draws teeth and bleeds
with leeches, besides being musicians; thus whilst the master is performing
any o f the operations o f his profession, his companions will endeavor to
soothe the soul, or drown the cries o f pain, as the case may be.”5 Some
owned their own barbershops, but many others with less capital and prob
ably w ith only rudimentary skills as medical practitioners peddled their
services not only at the cantos but in the squares or streets as ambulatory
barbers.
These barber-surgeons were also employed as musicians for religious
holidays in the city. For instance, for the “ Day o f the Kings,” musicians with
guitars, drums, and other musical instruments roamed the streets in groups,
going from house to house throughout the night.6 They also played music
in bands for the arrivals and departures o f European merchant ships, as
Thomas Lindley observed in 1802:
The Representation o f Identity, 1808-1831 49

These musicians were entirely black, and are trained by the different
barber-surgeons o f the city, who are o f the same colour, and have been
itinerant musicians from time immemorial: they always command a
full band ready for service; and a variety o f young learners whose discor
dant tones are hatefully grating as you pass where they are practicing.
Numerous as these swarthy sons o f harmony are, they find constant
employment: not only as above mentioned, but also at the entrance o f
the churches on celebration o f festivals; where they sit playing lively
pieces, regardless o f the solemnities going forward within.7
These barber-surgeons/musicians contributed to communication networks

as messengers between those who belonged to the same canto as well as
among the overall African-born slave population beyond ethnic boundaries.
Interestingly, the cantos did not include female medical practitioners, de
spite the fact that in Salvador people o f African descent virtually monopo
lized the rudimentary medical professions, as barber-surgeons if males and
as midwives if females, throughout the slavery regime.8 In contrast to male
barber-surgeons, many o f whom were African-born and enslaved, the art
o f m idwifery in colonial Brazil was limited to free Brazilian-born women
o f African descent, who acquired their professional knowledge and skills
largely from Luso-Brazilian society.9 Once again, we see an interesting con
trast by gender. Whereas men’s rudimentary medical knowledge and prac
tice among the population o f African descent strongly relied on knowl
edge more or less directly derived from Africa, women widely depended on
Brazilian-born women o f African descent for one o f their most important
life events, namely reproduction.
Another representation o f collective ethnic identity was “African” drum
dances called batuques, which were a prominent feature o f the street scene
in Salvador.10O n Sundays and holidays and at night after work, the African-
born population gathered for such drum dances, singing in their native
languages, dancing, and attending festivals related to their native African
religions.11 According to Albert J. Raboteau, in West African religions the vi
brant pattern o f music is interwoven with the ritual experience; dancing,
drumming, and singing play a constant and integral part in the worship
o f gods and the ancestors.12 In Salvador, the practice o f batuque was di
vided by African “nations,” and those who identified themselves with the
same “nation” seemingly had their own gathering place for batuques on the
streets.13 For instance, in 1814, official correspondence from the governor o f
Bahia to the Portuguese royal court in Rio de Janeiro observed that “ in Rio

de Janeiro, 10,000 or 12,000 negros dance according to their nations but all
together in the Cam po de Santana, whereas here in Bahia [Salvador], one
hundred Nagds dance in the Largo do Teatro, fifty Gdges in the Piedade,
eighty Hausas in the Rua de Joao Pereira, and thus it is throughout all over
the city” 14
This correspondence gives us an interesting contrast between Salvador
and the other major port city, Rio de Janeiro, in which various ethnic groups
shared the same space. Salvador's longer history o f importing African-born
slaves o f diverse ethnic backgrounds directly though the transatlantic slave
trade and incorporating them in urban slavery may have fostered the crea
tion o f stronger ethnic identities among the urban slave population, who
therefore collectively claimed and occupied their own separate ethnic terri
tories in accordance with their shared identities.15 Thus ethnic groupings
and groups were formed, and these people made the funerals o f their people
the occasion for batuque, probably in accordance with their own traditions
and customs.16
Over time African-born slaves and their descendants had their African
religions syncretized w ith Roman Catholicism and developed their own
unique form o f religion called candombU. In addition to several well-known
terreiros (temple grounds or sacred meeting places) o f candombU, in which
leaders were officially given the title pax de santo (father o f the saint) or mae
de santo (mother o f the saint), small candombU houses {casas de candombU)
were dispersed all over the city.*7 These were usually modest homes in
working-class neighborhoods, and even the basements o f such houses, but
they turned out to be special gathering places for religious festivals in the
neighborhood. Even on an ordinary day, a neighbor would drop by to ask
for fortune-telling or to make wishes. Each candombU house came to func
tion as the African-born population's neighborhood center, where informa
tion could be exchanged. Each o f these houses allegedly had its own African
“nation” as its ethnic identity, and many were identified as Nagd-G£ge,18
Needless to say, slave owners were very attentive and even nervous, particu
larly about their slaves' cultural-religious associations with the batuques
and, by extension, candombU. Their major concern was that such congrega
tions o f slaves signified a degree o f loss o f control by owners. Owners tried
to control any opportunity for slaves to develop associations that might lead
to rebellion. But owners’ concern was no less attributable to their fear o f the
power o f pagan (or what were to them strange) religions and a spiritual
force and cohesion which they could not understand.19
As a result o f the daily ethnic gatherings and groupings by the African-

born enslaved population on the streets, communication networks devel
oped along ethnic lines throughout the city. M any African-born kinsfolk,
who had been separated from one another and transported to Salvador
as slaves, managed, through opportunities for congregation provided by
such groups, to be reunited and resumed their relationships. Some, such as
mother and infant child(ren), had been enslaved and shipped together, but
others looked for their enslaved parents, children, siblings, and other kins
folk in Salvador, or possibly even ran into them in the streets.
Let us take three examples from the wills o f ex-slaves. The first is the
African-born ex-slave Joana Fernandes. She and her son Jo3o Sararis were
enslaved in Africa, baptized as Roman Catholics in Lisbon, and shipped to
gether to Salvador, where they were sold to different owners. They later
somehow succeeded in being reunited, and in her will the mother named
her son as her executor, as well as her heir.20The second example is Maria
do Bonfim, an African-born ex-slave, whose specific “nation” was not men
tioned in her will. She had neither married nor borne any children in Sal
vador, but had given birth to a son in her homeland, before being enslaved.
Her son had also been enslaved, transported to Salvador, and lived with the
Portuguese name o f Joaquim do Nascimento. He had left Salvador for the
port o f Whydah. Maria do Bonfim found her siblings in Salvador: a brother
named Caetano de Miranda, and two sisters, named Maria de Jesus and
Maria das Moroas.21 The third is Domingas de Salles. Because o f her en
slavement, she had left her daughter in Africa and had been transported to
Salvador, where she eventually married Joaquim de Salles, an ex-slave o f the
same owner, and by whom she had a son named Manuel da Cruz, who died
at five years o f age. Sometime during this process, Domingas got acquainted
with an African-born ex-slave woman named Justina Maria Luisa and in
her will o f June 8,1842, recognized Justina as the daughter she had left be
hind in Africa.22No wonder these resumed kinship relationships in turn fos
tered and strengthened the creation o f collective ethnic identity by the en
slaved African-born population.
M ating, Consensual Unions, and Fictive K inship23
The African-born population, particularly those who had not yet

spent many years in the New World and were therefore relatively unacculu-
rated and unassimilated, naturally sought out their partners from among
those who shared the same ethnic identity. As discussed in chapter 1, the
urban slave population in Salvador was balanced by sex, unlike most cases

o f New World plantation slavery but like the case o f most cities o f Latin
America and the Caribbean. Therefore, theoretically, urban slaves had at
least no demographic constraints on mating. W hile the small scale o f urban
slaveholding made it difficult for a slave to find a suitable partner from
among the slaves owned by the same owner, urban slaves’ daily gatherings
and groupings gave them abundant opportunities to meet prospective part
ners w ho shared the same ethnic identity. A. J. R. Russell-Wood suggests
that slaves may have avoided choosing a partner o f the same ownership be
cause by doing so they avoided both being subject to the whims o f a single
owner,24 but we do not know if that was really the case in urban slavery;
most urban slaves were hired out to the streets for all sorts o f urban occu
pations, where they could mingle freely with one another regardless. For in
stance, a Hausa slave man named Francisco had mated and developed a re
lationship with a slave woman named Silvana, also a Hausa, who was owned
by another owner named Joào de Campos. Francisco and Silvana produced
tw o daughters, Fermiana and Fermina, who were inevitably born into slav
ery because o f the legal status o f their enslaved mother, Silvana. Perhaps
at the tim e o f his self-purchase, Francisco adopted his ex-owner*s family
name, da Rocha, as his own. As an ex-slave, Francisco da Rocha resided in
Pilar parish when he registered his will in Salvador on March 10,1830. His
will states that Silvana lived as a freedwoman across from the cathedral but
does not tell if he continued to maintain his relationship with her; Francisco
chose as executor o f his w ill not Silvana but Antônio Pereira de Almeida,
who was his godfather. Francisco da Rocha never contracted marriage. His
older daughter Florinda, fourteen or fifteen years o f age at the time he made
his will, had been sold to a male worker in a town o f the Bahian Recôncavo,
called Nazaré das Farinhas (“ Nazareth o f the flours” ), named after its ma
jor product, manioc flour (farinha). In his will, Francisco expected Florinda
to purchase her freedom from proceeds o f the sale o f the possessions he
was to leave to her, and named her as his heiress. He did not make any pro
vision in his w ill for his younger daughter Fermina because he had no
knowledge o f her whereabouts, except that she had been sold to Rio de
Janeiro.25
We have no way o f knowing how strongly the sharing o f the same ethnic
identity could unite African-born slaves as longtime partners, because o f
the critical limitations o f existent historical data on their mating and con
sensual unions. First o f all, both census records and parish marriage regis
ters are o f very little use. As illustrated by the case o f Francisco da Rocha,
the urban slave population in Salvador rarely contracted church marriage.26

This is attributable to various social and cultural factors: the high cost o f a
church marriage, which slaves could barely afford; slave owners’ reluctance
to allow their slaves to marry; slaves’, especially African-born slaves’, unwill
ingness to accept Catholic marriage;27 and general lack o f marriages among
the poor, namely non-property owners.28 Secondly, neither household rec
ords nor slave owners’ inventories reveal slave unions on paper; slave part
ners seldom co-resided.
Probably the only available data that enable us to trace the African-born
people’s choices o f partners based on collective ethnic identity are African-
born ex-slaves’ wills. Am ong the 325 o f them who registered their wills
in Salvador during the nineteenth century, 74 were married w ith African-
born spouses; 23 shared the same African “ nations” only 7 were o f different
“nations,” and the rest (44 cases) did not identify the “nations” o f either o f
such African-born spouses.29Regardless o f the limited quality and quantity
o f these data, they imply that among the African-born population in urban
slavery, sharing the same ethnic identity was a key determinant in choos
ing partners and forming consensual unions, and in some cases ultimately
contracting marriages at the church, usually only after both partners had
gained legal freedom.30
Most newly arrived African-born slaves were baptized as adults, at the
time o f their purchase, unless they had been converted to Catholicism be
fore landing in Brazil. Godparenthood usually was established at the time
o f baptism, and nearly 95 percent o f African-born slaves baptized in Santo
Antônio parish between 1809 and 1869, regardless o f gender, were bap
tized only with godfathers, the majority o f whom (more than 60 percent)
were free-born men.31 The owners o f newly arrived African-born slaves
made the selection o f godparents, and used tw o strategies to control their
slaves. For African-born slaves baptized with African-born godparents, it
was common for godparents to be o f the same “nation ” but usually those
godchildren and their godparents were under different ownership, or the
godparents were ex-slaves.32 Baptizing new arrivals with godparents o f the
same “nation” who had already been exposed to New World slavery and
Luso-Brazilian culture to a great degree, and who would instruct or supervise
their godchildren at the workplace, would certainly promote the smooth
and rapid integration o f newly arrived slaves into the work force. Yet owners
were afraid that too strong a solidarity might develop i f both godparents
and godchildren were under the same ownership. So slave owners chose
slaves o f different “nations” as godparents o f new arrivals in those cases
where godparents and godchildren had the same owner. African-born slaves

under the same ownership served as godparents only when they were not o f
the same “ nation” as their godchildren.33 O f course, African-born slaves,
particularly men, took advantage o f their owners’ strategies involving the
issues o f ethnic identity and used the opportunity for the development o f
their collective ethnic identity at the workplace, which became a good foun
dation for solidarity when taking various collective actions against the slave
systems.
Voluntary Associations:
Savings Associations and Lay Sodalities
Savings Associations
Male African slaves created and developed much stronger collective

ethnic identities than did their female counterparts. Their solidarity re
sulted in the formation o f unique savings associations, called juntas (unions),
and caixas de empristimo (boxes o f loan) for making loans. No primary data
on juntas are extant, since, according to the contemporary historian Manoel
Querino, nothing about their activities was written down.34We may attrib
ute this to general illiteracy among the African-born population, as well as
to the major characteristic o f the juntas as secret societies. Each junta had a
leader chosen from among the most respected members. The leader had a
wooden stick for each member o f the junta, and recorded the transactions
with cuts on the stick. Another person was appointed as a treasurer, who
kept the collected money, invested it in mortgages, and loaned it at interest
for the betterment o f the association. Every Sunday, members o f the same
junta met to make payments in copper coins and to discuss the question o f
loans. If a member needed a loan, he was entitled to withdraw a sum o f
money at a specified rate o f interest. The loan was sometimes used for self
purchase by the member.35 One o f the very few references to such juntas
comes in the w ill (1878) o f African-born freedman Marcos Gasper, who had
been treasurer o f the “ Junta dos Africanos.’>36French traveler Joseph Arthur
de Gobineau, the Count o f Gobineau, made another reference, in 1869, to a
savings association among Mina slaves.37
Black Lay Sodalities
Black lay sodalities functioned as a very special type o f voluntary as

sociation for the enslaved population o f African birth and their descendants
The Representation o f Identity; 1808-1831 55

during the slavery regime.38 Slave men and women held the memberships
o f black sodalities equally; lay sodalities did not limit their memberships in
terms o f gender and admitted both men and women almost equally.39Every
African-born slave was obliged to be baptized as a Roman Catholic before or
upon arrival in Brazil. Conversion to Christianity immediately gave en
slaved peoples o f African birth equal spiritual rights w ith the prosper
ous white laity in the eyes o f God; it allowed them to m arry in church, to
attend mass, and to receive confirmation in the faith by visiting bishops.
African-born slaves were also free to participate in various religious celebra
tions in honor o f Christian saints. Most importantly, such newly converted
Christians were entitled to acquire a membership in lay sodalities, which
guaranteed a decent Christian funeral and the saying o f masses for the de
ceased. Despite the spiritual equality that all enslaved Africans and their de
scendants shared with the well-to-do white laity, in reality the latter never
accepted the former into their white lay sodalities as their fellow mem
bers. Many white sodalities, especially the prestigious Santa Casa da Miser-
c6rdia (H oly House o f Mercy), which itself owned a number o f slaves,
limited membership to whites.40 Mulatto sodalities, which came into be
ing later than black sodalities, accepted and welcomed whites but excluded
both the African-born and the Brazilian-born black populations (see chap
ter 6). Therefore, lay sodalities were inevitably divided by race. African-born
slaves, being denied the membership to all white (and later also mulatto)
sodalities, had to establish lay sodalities o f their own with their priorities
and agendas. In a black sodality, most members were enslaved individuals
who not only were desperately poor but also were human commodities
owned by others. Black sodalities were naturally badly o ff and did not own
their independent churches or even chapels; white sodalities with their own
chapels sometimes allowed black sodalities to build altars for their patron
saints and to hold ceremonies in honor o f their saints.
The enslaved population o f African birth had obviously sought the for
mation o f lay sodalities for their own needs; in 1789 the presence o f at least
seventeen black sodalities was recorded in Salvador.41 Unlike the case o f the
aforementioned juntas, namely savings associations for slave men that func
tioned as secret societies, all black lay sodalities were officially recognized by
the larger society; every lay sodality, either black, mulatto, or white, estab
lished in Portuguese America was required to have its statutes approved
by the archbishop and by the Portuguese Crown. No matter how strongly
African-born slaves had initially reacted against forced conversion to Chris

tianity, they soon found it beneficial to belong to lay sodalities for their daily
survival in slave society.
One unique function o f black lay sodalities was as emancipation socie
ties. The statutes o f black sodalities stipulated the provision o f loans from
emancipation funds for their slave members. Once the slave obtained free
dom, he or she was obliged to return the money to the sodality, so that an
other slave member could use the loan.42Yet it seems that such manumission
funds in black sodalities were never in active use, at least in the case o f Sal
vador.43 It is possible that slaves did not mention the loan o f manumission
funds from sodalities in notarized documents. It is even more likely that the
archbishop rarely approved o f the petition from black sodalities regarding
the loan o f their manumission funds. W hatever the case, it would seem
most important for black sodalities to be officially equipped with manumis
sion funds. Their ideological, not necessarily practical, readiness to take ac
tions for the freedom o f slave members could appeal not only to the slave
population but also to whites and mulattoes who wanted to be good Chris
tians by lending a hand to the needy.
The formation and development o f black lay sodalities in Salvador inevi
tably reflected the African-born slave population's historical creation o f eth
nic identities; first, “Angolan” sodalities came into being, and later slaves
originally from West Africa established their own sodalities.
The oldest black sodalities in Salvador were not established until the last
decades o f the seventeenth century; the intensive influx o f slaves from A n
gola, many o f whom shared the same ethnicity, had not resulted in the im
mediate establishment o f lay sodalities. Many black sodalities had been
formed but quickly dissolved absent official approval; the dominant society
had successfully hampered the enslaved population in establishing their
own lay sodalities. The first approved black sodality was finally established
in 1685, mainly by slaves from Angola and their descendants, as the sodality
o f O ur Lady o f the Rosary (irmandade de Nossa Senhora do Rosdrio) in
Conceiîo da Praia parish. The archbishop eventually approved the statutes
o f 1686 in 1702. The black sodality o f O ur Lady o f the Rosary in Conceiîo
da Praia parish was soon followed by another: the black sodality o f Saint
Anthony o f Catagerona (irmandade de Santo Antdnio de Catagero) in the
chapel o f S2o Pedro parish in 1699, which was also established by Angolans
and their descendants. At the beginning o f the eighteenth century, members
o f the black sodality o f O ur Lady o f the Rosary in Conceiâo da Praia par
ish left their parish church because o f insults they had received as pritos

(African-born slaves) from the white sodalities housed at the same church.44
From 1703 until 1726, they constructed their own church at the present lo
cation in the Pelourinho by working at night, on Sundays, and on holidays,
while a famous white painter offered them help for the plan and construc
tion o f the church. They moved the image o f their patron saint to the new
church in the Pelourinho.45 Thus this black sodality has been called the
black sodality o f O ur Lady o f the Rosary in the Pelourinho.
Ethnic identity was represented clearly in the governing o f these tw o
oldest “Angolan” sodalities; only Angolans and crioulos were accepted and
equally represented on the governing body o f both o f these black sodalities.
Office positions alternated between Angolans and crioulos; whenever the
former held an office, a comparable post went to the latter.46This might have
been at least initially a device for the offspring o f Angolan-born members
to participate in the sodalities with their parents and kinsfolk, so that
the sodalities might represent the interests o f such people o f Angolan de
scent. Since the larger society did not have any specific term to define
Brazilian-born people o f Angolan descent, they were broadly described as
crioulos, together with those o f non-Angolan origin. Such am biguity o f de
scription may have subsequently encouraged some o f the general crioulo
population, not only those o f Angolan descent, to join the “Angolan” sodali
ties. We are not sure whether these sodalities were eager to be exclusively
“Angolan” or not, but whatever the case, the strong representation o f crioulos
on the governing body certainly suggests that the successful establishment
o f the oldest black sodalities near the end o f the seventeenth century had
something to do with the creolization o f the urban slave population. Black
sodalities relied on their crioulo members, who were born in Brazil, baptized
at birth, and much better integrated into the larger society, with sufficient
understanding o f Luso-Brazilian culture and the Portuguese language, as
intermediaries. In the last decade o f the seventeenth century, with the sig
nificant presence o f the crioulo population, the African-born slave popula
tion in Salvador finally found it possible to have their black sodalities ap
proved by the larger society.
West African slaves started to arrive in Salvador around the end o f the
seventeenth century, and many o f them joined in the existent “Angolan”
sodalities. As a result, the new members o f the black sodality o f O ur Lady o f
the Rosary in the Pelourinho for the period from 1722 to 1786 were predomi
nantly enslaved, and Brazilian-born only slightly outnumbered African-
born. Among the African-born members, West Africans (48.3 percent) out
numbered Angolans (29.0 percent).47 Nevertheless, crioulos outnumbered

Africans. The percentage o f West Africans among the new members o f A f
rican birth continued to grow; for the period from 1798 to 1865,73.7 percent
o f the African-born were from West Africa, while Angolans constituted only
20.5 percent.4* Despite the prominence o f West African members, maintain
ing a collective ethnic identity as “Angolan” continued to be important
for the members o f the black sodality o f O ur Lady o f the Rosary in the
Pelourinho. In 1786, the sodality asked the queen o f Portugal, Dona Maria,
for permission to wear masks and to dance while singing in the “Angolan
language” during festivals o f O ur Lady o f the Rosary, on the grounds that
this had been done in other Christian countries.49
In the late eighteenth century, slaves from West Africa also established
their own black sodalities. Formation o f these new black sodalities during
the eighteenth century reflected the creation o f new ethnic identity by West
African slaves in Salvador. The sodality o f Good Jesus o f the Necessity and
Redemption (irmandade de Sr. Bom Jesus das Necessidades e Redempâo)
in the Corpo Santo Church was established by G6ges (Ewes) in 1752. The
sodality o f O ur Lady o f the G ood Death (irmandade de Nossa Senhora da
Boa Morte) in the Barroquinha Church was established by Nag6s (Oyo-
Yorubas).50
As stated earlier, black sodalities did not discriminate against women
in their memberships; both women and men held memberships o f black
sodalities, almost equally. This is attributable to the fact that many slave
women o f African birth worked as market women in the city and were ca
pable o f earning more cash than their male counterparts, who worked as
porters and transporters.51 Such women members o f African birth could
have made a significant financial contribution to black sodalities. Nonethe
less, women were not allowed to hold offices in some black sodalities. For
instance, while no slaves o f either sex were entitled to become officeholders,
the statutes o f the black sodality o f the Rosary in the Pelourinho (both in
1698 and in 1820) stipulated that women were excluded from the governing
body “because o f the condition o f their sex.”52
Black sodalities were not exclusive in their memberships, either regard
ing ethnicity and gender or in terms o f race; they allowed, and in fact even
welcomed, well-to-do white and mulatto members because o f the poten
tial for bequests from them. Indeed, black sodalities were never exclusively
black. Furthermore, the offices o f treasurer and scribe o f black sodalities
were often reserved for white members, who usually paid higher entrance
fees and annual dues, because o f the alleged illiteracy and poverty o f their
black members. This custom was abolished only around the end o f the eigh
The Representation o f Identity; 1808- 153; 59

teenth century, by which time more and more black members had become
literate.53 Russell-Wood maintains that prosperous whites paid dues to be
members o f several sodalities, including black sodalities, all o f which, by
statute, guaranteed the attendance o f all members at their funerals; ostenta
tious display at a funeral was an indication o f the deceased s socioeconomic
standing in colonial Brazil.54 British vice-consul James Wetherell stated that
in Salvador a “curious custom exists at the funerals i f the deceased has been
a member o f one o f the Irmandades [lay sodalities] or clubs; most o f the
members attend the funeral procession, and both the corpse and the atten
dant brothers are habited in the peculiar dress o f their order”55 For instance,
in the 1830s, wealthy Portuguese merchant José Coelho Maia belonged to
seven sodalities, including the mulatto sodality o f O ur Lady o f Guadalupe.
During the same decade, Portuguese bricklayer Manoel Antônio da Costa
Rodrigues belonged to eight sodalities, including tw o black sodalities: the
sodality o f O ur Lord o f Martyrs (irmandade de Nosso Senhor dos M ar
tinos) in the Barroquinha church; and the sodality o f Saint Benedict in the
church o f Conceiçâo da Praia parish.56 Obviously, white members did not
mind having blacks attend their funerals. Perhaps black members stood in
the same line where the deceased's slaves did, which could appear as i f the
deceased had owned many more slaves. O f course, others joined black sodali
ties simply out o f Christian charity to help the needy, including slaves. Some
became members of black sodalities out o f curiosity about their slaves’
activities and/or black culture, or out o f their inevitable concern for slave
control.
Black sodalities chose to be devoted to the cult o f specific Christian pa
tron saints, often for the purpose o f worshiping African gods and goddesses;
in disguise o f Catholicism and Christian saints, enslaved African peoples in
colonial Brazil attempted to maintain their own African religious rites and
customs, and to create and re-create their collective ethnic identity. Many
black sodalities in Salvador were devoted to the cult o f O ur Lady o f the Ro
sary (Nossa Senhora do Rosârio). Roger Bastide maintains that the venera
tion o f O ur Lady o f the Rosary, initiated by Saint Dominic, had fallen into
disuse and was revived only when the Dominicans started to send their first
missionaries to Africa, which explains why it took root in and spread among
enslaved Africans.57 Later, Africans in Brazil may have found a way to iden
tify the Rosary with the African goddess Yemanjâ. In Brazil, O ur Lady o f
the Rosary was sometimes painted with a black face and hands.58 Black so
dalities o f O ur Lady o f the Rosary annually crowned African “kings” and
“queens” in their religious festivals. Henry Koster, a British planter who

lived in northern Brazil during the early decades o f the nineteenth century,
wrote: “ [TJhe Brazilian Kings o f Congo worship O ur Lady o f the Rosary,
and are dressed in the dress o f white men; they and their subjects dance, it
is true, after the manner o f their country; but to these festivals are admitted
African negroes o f other nations, creole blacks, and mulattoes, all o f whom
dance after the same manner; and these dances are now as much the na
tional dances o f Brazil as they are o f A frica”59Another most popular patron
saint o f color was Saint Benedict the Moor (S§o Benedito), who had been
known as a miracle worker immediately after his death in 1589, although his
cult remained marginal in orthodox Roman Catholicism until 1743, and he
was canonized only in 1807.60The black sodality o f Saint Benedict in Saint
Francis Monastery maintained the highest popularity in Salvador. Saint
Ephigenia and Saint Elesbao are also saints o f color.61 Saints worshiped by
the African-born population in Salvador included St. Barbara, St. John, Sts.
Cosme and Damaio, St. Jerome, St. Anthony, St. Lazarus, St. Roque, St. Ann,
O ur Lady o f the Conception, O ur Lady o f the Candles and, above all, O ur
Lord o f Good Ending.62 According to Albert J. Raboteau, Ibeji (in Yoruba),
the spirit o f twins, is syncretized w ith the tw in saints Cosme and Damaio,
and correlations are also found between Ogun, the god o f iron and war, and
St. John the Baptist; Oshun, the African Aphrodite o f voluptuous beauty,
and the Virgin o f Bobre; and Shopana, lord o f smallpox and St. Lazarus.63
Over time such identification o f African gods and goddesses with Christian
saints and accompanying religious festivals by African-born slaves resulted
in dynamic syncretisms between African religions and Roman Catholicism,
most notably in candomblL The backyard o f the Barroquinha Church, in
which the black sodality o f O ur Lady o f Good Death was established by
Nagds, functioned as a place for candombU festivals o f Nag6 origin for many
years until the church fell into ruin in the late twentieth century because o f
financial problems.64
Black sodalities were relatively free to bury the dead, in the name o f
Catholicism, with non-Christian rituals, especially in the case o f the black
sodality o f O ur Lady o f the Rosary in the Pelourinho, which had an inde
pendent church o f its own. In the backyard o f the church there was a ceme
tery for the burial o f their sodality members, as in any other church in co
lonial Brazil. However small the cemetery, it was indeed used for the burial
o f slaves who were members o f the sodality. Having an independent ceme
tery associated with their independent church guaranteed slave members a
decent burial on church property. In Salvador at least as late as the first de
cades o f the eighteenth century, it was not uncommon for African-born in
The Representation o f Identity* 1808-1831 61

dividuals to be buried in a non-Catholic mode.65 Burial rites have been very
important in West African religions, because o f the powerful position o f
ancestors. Ancestors, both those who passed away a long time ago and those
o f more recent memory, are revered as founders o f villages and kinship
groups throughout West Africa, where funeral ceremonies are long, com
plex, and expensive.66Therefore, in Salvador, the sites o f such burials served
as very important gathering places for the African-born population on spiri
tual occasions, through which they reaffirmed their collective ethnic iden
tity.67
W hile the individual’s ethnic identity did not always determine his/her
membership in a specific black sodality, collective ethnic identity continued
to be a critical factor for the association. It was expressed and maintained
by a black sodality through its worship o f African gods and goddesses in the
disguise o f Christian saints, its modified practice o f African burial rituals,
and its continued use o f African languages. Ironically, such an emphasis on
collective ethnic identity observed in black sodalities resulted in preserving
and reinforcing ethnic divisions among the enslaved African-born popula
tion, as long as the transatlantic slave trade continued to bring enslaved A f
ricans to the city o f Salvador.
Flights and Uprisings
Ethnic identity was collectively manifested in the frequent incidence

o f the slave flights and uprisings that took place in Salvador and the Bahian
Recdncavo during the first decades o f the nineteenth century.68British trav
eler George Gardner, expressing a commonly held view, stated in 1836 that
the slaves in Bahia were “more difficult to manage than those o f any other
part o f Brazil, and more frequent attempts at revolt have taken place there
than elsewhere.” Gardner ascribed this “ rebellious” character o f Bahian
slaves to their shared ethnicity: the “Gold Coast” o f West Africa.69 During
the sugar resurgence in the last decade o f the eighteenth century, an increas
ing number o f slaves continued to be imported from Africa to Bahia, and
its major component was those from the Bights o f Benin and Biafra o f
West Africa. The variety o f sources from which slaves were drawn reflected
the colonists’ attempts to offset the disruption stemming from West African
internal civil and religious conflicts, called jihad in Arabic, and to take ad
vantage o f increased numbers o f war captives sold into slavery. Usman da
Fodio launched a campaign w ith the help o f the Muslim Fulani to take
control o f the Hausa Kingdom. O yo collapsed in the 1820s and 1830s, and

Dahomey and various Yoruba warlords fought among themselves through
out the nineteenth century. The O yo nobility had relied on Muslim military
slaves, who became involved in O yo politics. The important split o f the Oyo
Empire took place in 1817. Many more slaves came onto the market between
1818 and 1823 as subject provinces struggled for autonomy. By the early 1830s,
when the O yo capital was destroyed, a new Islamic empire was built in the
savannah. As a result, a large number o f war captives were disposed o f by
way o f the slave trade to Bahia.70It is hard to ascertain to what degree slave
uprisings in Bahia can be interpreted as an extension o f the jihads o f West
Africa. True, many such individuals had been the captives o f the jihads and
therefore had been exposed to Islam in West Africa, but their degree o f con
version and devotion to the religion o f Islam varied. Islam in West Africa
might not have provided them with strong religious faith in the same God
but, as Jack G oody elucidates, the written religion o f Islam, whose text was
written in Arabic, could have given its adherents literacy and could have
functioned as the universal means o f communication among them.71
From the early colonial period there had been increasing formation o f
small fugitive slave communities, called at first mocambos and by the eigh
teenth century quilombos,72 in relative proxim ity to plantations and towns
and even within urban areas.73 Urban slaves had constantly run away from
Salvador to the suburbs o f the city, such as Soledade, Forte de S3o Pedro,
Agua de Meninos, Itapoam, Barra, and Rio Vermelho, and as far as Rio
Joanes in the Recdncavo.74Often bush captains (capit&es de mato), who were
usually free men o f color, were hired to search for runaways in the bush.73
These fugitive slave communities were usually conveniently situated near
the city, towns, and plantations on which they could depend and prey. The
mocambo economy was parasitic; those fugitive slaves lived on highway
theft, cattle rustling, raiding, and extortion.76
Slave flight gained prominence in the Bahian Recdncavo around the be
ginning o f the era o f the resurgence o f Brazilian sugar fortunes in the 1790s
and the early 1800s.77 In March 1807, the Bahian authorities found that two
quilombos had been formed near the villages o f Nossa Senhora dos Marts
and Cabula within tw o leagues o f the city center. O n March 29, the gover
nor o f Bahia, Joao de Saldanha da Gama, the Count o f Ponte, sent a troop
o f eighty soldiers who, on the following day, destroyed the quilombos and
arrested seventy-eight slaves and ex-slaves.78 On April 7, the Count o f Ponte
reported to the Portuguese overseas council that there had been numerous
quilombos in the suburbs o f Salvador and within the surrounding bush. Ac
cording to his statement, these maroon communities were “organized by

industrious charlatans, and attracted credulous» vagrants, the superstitious,
robbers, criminals, and the sick.” 79 In these quilombos, people were free to
“perform dances, wear capricious clothes, practice false medicine, recite
blessings and fanatical prayers, eat, and entertain themselves with the most
scandalous acts offensive to all o f the laws and public tranquility.”80 This
was the prelude to the slave uprisings o f 1807-1831. Interestingly, prior to 1807
no organized slave uprisings had been recorded either in Salvador or in
Bahia, in which the slave population was largely engaged.
The first uprising by African-born slaves was planned to take place on the
evening o f the 28th o f May, 1807. The plotters were thirteen African-born
slaves, all o f whom were allegedly Hausas. This had been carefully planned
on the assumption that many residents o f Salvador would be out o f town,
because the 29th o f May marked the religious festival o f Corpus Christi. The
participants intended to set fire to the customhouse and to a chapel in Naz-
aré, a suburban district o f Salvador, and then to escape to Africa by seizing
ships in the port. It seems to have been symbolically important to the rebels,
who were relatively new arrivals from Africa, to destroy the chapel that they
saw as a manifestation o f Catholicism, the religion o f slave owners. The reb
els had intended to be armed with bows, arrows, pistols, and rifles. But, on
the 22nd, an unidentified slave leaked these plans to the authorities.81 All the
rebels were arrested just before the uprising. The leader was a slave named
Antônio, resident o f Rua do Corpo Santo. His occupation as trader enabled
him to be in regular contact with Hausa slaves in the Recôncavo. He was
known by the title o f “ambassador” and was assisted by “captains,” each
assigned to a different district o f the city. Antônio and his principal lieute
nant, Balthazar, also a slave living on Rua do Corpo Santo, were sentenced
to death. The other plotters received 150 lashes each in the public square.
After this plot, the Count o f Ponte not only prohibited batuques, but also
placed restrictions on the physical movement o f slaves and ex-slaves. The
Count o f Ponte became conscious o f slaves’ African birth, especially their
ethnic origin, as a lethal threat to the larger society and, in his letter dated
June 16,1807, noted that in 1806 Bahia had imported 8,037 slaves from Costa
da Mina, and that these were more rebellious than those from East Africa.82
The following year o f 1808 witnessed the brief stopover o f the Portu
guese court in Salvador on their way to the colonial capital o f Rio de Janeiro.
We do not know the enslaved population’s general reaction to this histori
cal event, except through the short but interesting account by the British
traveler James Henderson. According to Henderson, the visit o f the Crown
Prince to Salvador led the slaves o f African birth to challenge the right o f

the governor o f Bahia to inflict the statutory 150 lashes, now that “the Lord
o f the land (Dom da Terra)” had arrived. The slaves expressed their feelings
in the following saying: wThe Lord o f the land arrived, the rule o f 150 lashes
over” 83 As soon as the royal family departed, the Count o f Ponte posted the
following lines all over the city and its suburbs: “The Lord o f the land has
passed, the rule o f 150 lashes rem ains”84 O f course, we have no way o f
knowing if this event actually took place; it might be an oral tradition be
cause Henderson only recorded it in 1821. Yet, this story still tells us how
conscious the enslaved population was o f political changes and the power
structure o f the colonial society; after all, the governor o f Bahia was an
agent o f the king, and not the embodiment o f real power as was the king o f
Portugal. In this story, the common punishment o f 150 lashes for slaves was
used as a symbol o f the harsh reality o f slavery.
Rapid commercial expansion resulting from the transition from mercan
tilism to free trade (1808) aggravated Bahia's longtime problem: food short
age. From the beginning o f the seventeenth century, the municipal council
had passed various regulations to ensure an adequate supply o f manioc
flour to the population o f Salvador and slaves o f the Rec6ncavo. Manioc was
a crop that merchants, planters, and even humble farmers disregarded in
their quest for high profits derived from sugar and tobacco. Larger allot
ments o f land were used for agro-export commodities, such as sugar, to
bacco, and cotton. Subsistence production, particularly o f manioc and
beans, continued to decline drastically, despite the fact that the population
was growing rapidly. As a result, the price o f basic commodities rose sharply,
and the material conditions for the slave population deteriorated. Bahia re
sorted to importation o f food and clothing from the other provinces and
from abroad, which contributed to drastic inflation through the first third
o f the nineteenth century.85 These socioeconomic factors could have con
tributed to the frequent uprisings among the ever-starving enslaved popu
lation in Salvador (and Bahia) from 1808 to 1830.
The last days o f 1808 and the first week o f the New Year witnessed two
separate incidents involving slave flight and the formation o f a quilombo in
the Recdncavo. The first started on Decem ber 26,1808, when unknown
numbers o f slaves escaped from some sugar plantations, especially in and
near the town o f N azari das Farinhas. W ithin tw o days, they formed a
quilombo in Riacho da Prata, located nine leagues from Salvador. On Janu
ary 4,1809, slaves, whose number was estimated at between tw o hundred
and three hundred, fled from the city to join the quilombo. At some three
leagues from Salvador, they began to engage in “robberies, acts o f arson,
The Representation o f Identity; 1808-1831 65

assaults, ferments, and all evil things” On January 6, the Bahian governor
sent troops to defeat these rebels. The troops not only killed many rebels but
also took as prisoners eighty-three males and twelve females. Others es
caped into the woods.86A fter these tw o slave uprisings o f 1807 and 1809, the
Bahian authorities began to pay close attention to a specific African “na
tion,” Hausa, as the group most threatening to slave control.87
The uprising o f 1814, involving many urban slaves, occurred in Itapoam.
Located only tw o leagues north o f the city, Itapoam was an area inhabited
predominantly by fisherfolk, who were principally people o f African descent
engaged in whale fishery.88Once again, the majority o f rebels were Hausas.89
This uprising was planned to occur on an evening o f semana santa (holy
week), when the whole population o f the city was occupied by numer
ous processions, such as that o f the Penitentiary Disciplinants and the fes
tival o f Judas Iscariot.90On the 27th o f February, some two hundred slaves
from Salvador and from sugar plantations in the Recdncavo converged out
side Itapoam. At about four o’clock in the morning o f the 28th, they be
gan to attack the amta^do (warehouse and processing plant for whale prod
ucts) and the home o f the future president o f Bahia, Manoel Ignlcio da
Cunha Menezes, and killed his foreman and family members. They liber
ated his slaves, who joined the uprising. Then the rebels burned down the
engenhos o f Joao Vaz de Carvalho and Francisco Lourenô Helclano, liber
ated more slaves, and proceeded to the village o f Itapoam, where they con
tinued on their rampage o f arson and killed “whites and pardos.” The rebels’
plans called for them to leave this area and march into the Recdncavo, but
the Bahian troops tracked them down. A fter a one-hour bloody battle in the
region o f Santo Amaro de Itapiranga near the Rio de Joanes, the rebels suf
fered defeat by six o’clock in the evening on the 28th. At least fifty-six rebels
lost their lives in combat. The authorities arrested and imprisoned thirty-
one in Salvador and apprehended eight others later at various places in the
Recdncavo. They hanged four slaves in public, and deported a further twelve
to the ports o f Angola, Benguela, and Mozambique.91
In response to a recommendation from the royal court in Rio de Janeiro
dated March 14,1814, the new governor o f Bahia (1810-1818), D. Marcos de
Noronha, Count o f Arcos, increased the number o f police in Salvador and
ordered the police to impose more severe restrictions on slaves. The Count
o f Arcos also reissued regulations prohibiting ttegros from carrying arms as
well as an edict that ordered punishment by 150 lashes o f any slave who was
out on the street after nine o’clock at night without an owner’s pass. Further
more, he issued a new regulation prohibiting slaves from gathering in groups

larger than four persons without their owners* supervision.92 The Count
o f Arcos took a much more liberal view o f slaves’ gatherings than had the
Count o f Ponte. In his decree o f April 10,1814, the Count o f Arcos stated:
The dances rtegros are accustomed to performing to the sound o f disso

nant and rattling instruments in the streets and public squares are totally
prohibited.. . . Acknowledging, however, that many slaveowners under
stand their needs and try to diminish the horrors o f slavery by permit
ting their slaves from time to time to divert themselves and forget for
some hours their sad condition, and acknowledging also that all civilized
cities in the world permit public entertainment even for the lowest classes
o f the nation, there will be no prohibition on slaves assembling at the two
public squares o f Graâ and Barbalho to dance until the sound o f Ave
Maria, after which they must return to the houses o f their owners.93
The Count o f Arcos did not regard the batuques as the reason for the upris
ing o f 1814. In his letter to the Portuguese court in Rio de Janeiro, the gov
ernor referred to the maltreatment o f slaves and the shortage o f meat and
beans in their diet. In his view, preserving the batuques among slaves o f di
verse “nations” that were supposed to be hostile to one another promoted
“disunion” among the slave population and robbed them o f that unity o f
purpose essential to a successful uprising.94
Bahian slave owners did not agree with the Count o f Arcos in his view
o f the batuques and insisted that the only way to control the slave popula
tion was with the threat and application o f rigorous punishment. In their
petition to the governor, slave owners noted that “ they know and speak o f
the fatal success o f the island o f Santo Dom ingo and other revolutionary
discourses and assert that by the day o f St. John [Sao Joao] no white nor
mulatto will remain alive.”95 In response to this petition, the Count o f Arcos
prohibited the use o f firearms in that year’s festivals on St. John’s Day or
Midsummer Eve (June 24).* St. John’s Eve, which, among other festivals,
was celebrated by bonfires, was an occasion when the enslaved could estab
lish godparentage relationships despite the Church’s consistent attempts to
limit the practice; by joining hands and leaping over a bonfire together,
people became “coparents o f the fig tree (compadres da figueira).”97 Aware
o f the conflicting views on slave control between the Count o f Arcos and
slave owners, the rebels elected Arcos their “prince” during the uprising o f
1814.98
Despite the Count o f Arcos’s assertion that the Bahian slave population
was engaged in revolutionary discourses, in the case o f slave uprisings for
The Representation o f Identity 1808-1831 67

the period 1807-1830 there is no historical document written by slaves them
selves. Therefore, it is uncertain to what degree the occurrence o f the Haitian
Revolution or any other uprising or revolt in other parts o f the New World
influenced these slave uprisings in nineteenth-century Salvador, as Bahian
planters and slave owners were convinced.99
The next slave uprising took place on the 12th o f February, 1816 in the
regions o f Santo Amaro and S2o Francisco do Conde. This too was sched
uled to coincide with a religious festival. Around thirty armed rebels, most
o f whom were Hausa slaves, burned down several engenhos and killed a
number o f people. This uprising lasted four days. This was the last slave
uprising during the colonial era in Bahia, and the last Hausa-dominated up
rising.100
During the war o f independence, slaves in Bahia rebelled only on a small
scale. These rebellions occurred in agricultural regions, such as in the village
o f Sao Mateus (February 1822), on the island o f Itaparica (June 13,1822),
and in the quilombo o f Pirajd (November 12,1822), but had no impact be
yond these immediate regions.
O n August 25,1826, the tobacco-producing Recdncavo town o f Cachoeira
suffered a large slave insurrection led by a Nagd slave who proclaimed him
self “ King.” In December 1826, more than one hundred slaves converged on
the quilombo o f Urubu, located in the region o f Pirajd. Urubu was estab
lished as a center for the clandestine practice o f candombli> and later became
a refuge for runaway slaves. They began to attack a nearby village on the
15th o f the same month. It took several days for the Bahian authorities to
defeat the rebels.101
Again in Cachoeira, on April 22,1827, slaves o f the Engenho da Victoria,
whose owner was the merchant Pedro Rodrigues Bandeira, together with
those from three nearby sugar plantations, rebelled and killed the foreman
and his brother. T hirty slaves were arrested. Pierre Verger’s research shows
that around this time the word “ Nag6-Mald” began to appear in police rec
ords. 102According to Verger, matt originates from iamle, the Yoruba term for
M uslim.103 It suggests that the Bahian authorities had become aware o f the
association o f the African “nation,” Nagd, and the religion o f Islam.
During the evening o f March 11,1828, slaves fled from Salvador into
the bush o f Pirajd to be united with those from engenhos in Cabrito and
Cabula. O n the following morning, they began to set fire to the engenhos o f
this area and proceeded to attack the buildings o f the former Bahian presi
dent, Manoel Igndcio da Cunha Menezes, and Francisco Lourenô Helclano,
both o f whom had been victim s o f the uprising o f 1814. Some eighteen other

slaves, predominantly negros novos (newly arrived African-born slaves),
joined the uprising. After a bloody battle, twenty slaves were killed and at
least live, including tw o female slaves, were taken as prisoners to Salvador.
All five prisoners were Nagôs. Throughout the 1820s, however, urban slaves
in Salvador do not seem to have had a major role in the frequent uprisings
in Bahia.'04
The uprising o f 1830 occurred in the center o f the city o f Salvador. On
the 10th o f April, around seven o'clock in the morning, a group o f eighteen
to twenty slaves who were identified as wage-earners (negros deganho) at
tacked the hardware store o f Francisco José Tupinambâ, located on Rua da
Fonte das Pedras in the lower city. They wounded the owner and his em
ployee and took twelve swords and five long-knives. After breaking into two
more hardware stores to obtain additional arms, they proceeded to Rua de
Juliâo, also in the lower city, to attack the slave market o f Wenceslau Miguel
de Almeida. The armed insurgents freed over one hundred newly arrived
slaves and wounded some eighteen others who refused to join the uprising.
Once the rebels, who now numbered over one hundred, arrived in sight o f
the police station o f Soledade, they were defeated by troops and civilians.
Police investigation revealed that the mission o f this uprising was to obtain
weapons for a larger revolt planned for the 13th o f April, 1830. The leaders
o f this uprising were identified as Nagôs.105
In the first three decades o f the nineteenth century, the enslaved African-
born population in Salvador frequently rebelled against the larger society.
On many o f these occasions, slaves carefully timed their uprisings for the
times when the city residents were preoccupied by religious festivals. Women
had virtually no participation in these slave uprisings. In each o f these up
risings, most o f the participants seem to have been drawn from a specific
“nation” : Hausa prior to 1816 and Nagô thereafter. As Schwartz points out,
this change in the representation o f ethnic identities among the participants
took place in accordance with the changing composition o f newly arrived
African-born slaves. Although Nagô was the most prominent African “na
tion” in Salvador throughout the nineteenth century, Hausa constituted the
majority o f new arrivals from West Africa in the first decade o f the nine
teenth century because o f the political turm oil within Hausaland. Nagô
slaves did not arrive in Bahia in substantial numbers until after 1815.106
The slave uprisings from 1807 to 1830 also reflected the development
o f collective ethnic identity across the urban-rural geographic boundaries
among the African-born male slave population who often worked in groups.
This is partly attributable to the physical mobility enjoyed by African-born

male urban slaves, such as chair carriers who accompanied their owners
from their plantations in the Recdncavo to their urban residences in Salva
dor, suburban field workers who regularly came to the city to sell their ag
ricultural products, and oarsmen and sailors who transported people and
goods by rivers and by sea between Salvador and the Recdncavo towns.
Conclusion
Ethnic identities were clearly manifested in the African-born popu

lation’s daily groupings and voluntary associations and in their participa
tion in slave uprisings, as well as in their choice o f partners for consensual
unions and o f their children's godparents. At the same time, distinctive gen
der identities effectively undermined the further development o f collective
ethnic identities. African-born slave men, who worked as porters, trans
porters, and artisans, tended to be strongly united by sharing the same
ethnic identity, which was reflected in the formation o f unique voluntary
associations, as well as in the incidence o f slave flights and uprisings. By
contrast, African-born slave women, who became successful in market
ing in the city, oftentimes chose to separate themselves from their male
counterparts who shared the same ethnic identity, and formed their own
groupings and associations beyond ethnic identities. Yet African-born slave
women were structurally hampered from establishing official power because
o f their gender; they were not eligible for holding offices in black sodalities
while participating equally with male members, for instance.
Through the process o f being integrated quickly into urban slave systems
and being assimilated gradually into Luso-Brazilian culture, the African-
born population got connected with one another in urban slavery by shar
ing ethnic and gender identities. Yet that does not mean that the creation o f
identity in the New World embraced all the enslaved African-born popula
tion strongly enough to unite them as a group or community. The majority
o f them had been free in the Old World, and they knew what freedom
meant. Their willingness to gain freedom often divided their solidarity. As
long as opportunities existed in urban slavery for the individual pursuit o f
freedom, the African-born slave men and women were strongly motivated
to recover their lost freedom in the New World, and for that purpose they
also found effective ways to utilize their ethnic and gender identities. How
did they (re)create their identities upon gaining freedom? How did their
identities change over time? That w ill be our major focus in Part II.

Part Two
To Be African-Born and Freed ,
circa 1808-1880
The man who emigrated— my grandfather—

carried within him the memory o f home, the
former world, the place where he was once
“real." It tore at him, that memory, and yet it
kept him anchored: he knew where his home
was, knew that he had lost it. The son o f that
man— my father— believed he could make the
new place home. The task was probably impos
sible, but it kept him occupied.
— David Mura,
Turning Japanese: Memories o f a Sansei
4 The Re-creation o f Identity,
1808-1831
Most enslaved Africans in the New World had something important in

common, which was the experiences and memories o f freedom in the Old
World; they had been born free in their homelands o f Africa and therefore
had once taken freedom for granted, no matter how varied their individual
experiences were. Such enslaved people knew “ what it was to be free, before
they were enslaved; they were captives who could remember freedom.” 1 For
whatever reasons, other human beings, including their parents and other
kinsfolk, betrayed their trust, denied their humanity, and reduced them to
chattels, to be traded as mere commodities. T he physical and emotional
trauma o f enslavement and the excruciating horror o f the Middle Passage
scarred their sense o f self-worth, damaged their dignities, and wounded
their souls. They remembered what freedom had once meant to them, and
they constructed and reconstructed their individual notions o f freedom in
their daily experiences o f oppression and exploitation in New World slavery.
They learned in a very hard way that other human beings, even their parents
and siblings, were able to deprive them o f their freedom, which they had
once taken for granted. Naturally these enslaved people desperately sought
the recovery o f their lost freedom in the New World.
This chapter will examine the African-born populations re-creation o f
identity as ex-slaves during the period in which the Atlantic slave trade con
tinued to bring abundant enslaved labor into the port o f Salvador. We shall
first discuss how the African-born population obtained their “letters o f lib
erty” (cartas de alforria) in urban slavery— since slave owners rarely manu
mitted their African-born slaves without payment, the slaves had to seek
self-purchase— and how some African women and men manipulated their
ethnic identities in their individual pursuits o f freedom in manumission.
We will also question whether gender differentiated their experiences. Then
we shall scrutinize the special character o f the “freedom” that African-born
ex-slaves could obtain. African-born ex-slaves usually kept the same occu
pations they had formerly acquired as slaves. These jobs were ones that the
white population could not fill because o f the shortage o f free labor or did
not want to engage in because o f the lack o f prestige, such as transportation
or craftsmanship, or those that only the African-born population could do
well because o f their ethnic backgrounds coupled with their multilinguistic
skills, such as the slave trade. Thus legal freedom did not change their o c
cupations. African-born ex-slaves naturally distinguished themselves from
their enslaved counterparts and made every effort to appear to be “free” by
adopting family names and taking up the elite's values and distinctive be
havior patterns, such as slaveholding and multi-membership in lay sodali
ties. Yet, no matter how hard they tried to prove themselves as free people—
to themselves as well as to others— their cultural otherness derived from
their foreign birth hampered them from being fully accepted or integrated
into the larger society as free. A fter all, all o f their otherness would remain
in the larger society’s perceptions o f them regardless o f their newly acquired
legally “ free” status.
Although this chapter focuses on the period from i$o8 to 1831 for our dis
cussions o f the re-creation o f identity by the African-born ex-slave popula
tion, the data o f the manumission o f African-born slaves that 1 use in this
chapter covers the entire period o f 1808 to 1888, which is essential for the
purpose o f contextualizing our arguments.
M anum ission and the M anipulation o f Ethnic Identity2
“T he slave can oblige his master to manumit him, by tendering to

him the sum for which he was first purchased, or the price for which he
might be sold, i f that price is higher than what the slave was worth at the
time he was first bought”3 British planter Henry Koster’s well-known as
sertion (1817) on the alleged legal rights o f slaves in Brazil became the basis
for Frank Tannenbaum’s (1947) and Stanley Elkins’s (1959) views. It spurred
a debate during the 1960s among U.S. historians, primarily regarding the
alleged “softness” and “ hardness” o f slavery in a comparative perspective.
In reality, the prevailing practice o f self-purchase by slaves in Brazil had no
basis in law, except in small numbers and on special occasions. Until the Free
Womb Law was enacted in 1871, no slave owner in Brazil was bound by any
legal obligation to liberate slaves who claimed their right to buy themselves
out o f slavery by offering a sum equivalent to their asserted values.4 Koster
himself reported malfunctions o f the laws:

(T]he master sometimes does refuse to manumit a valuable slave; and no
appeal is made by the sufferer, owing to the state o f law in that country,
which renders it almost impossible for the slave to gain the hearing; and
likewise this acquiescence in the injustice o f the master proceeds from
the dread» that if he was not to succeed he would be punished» and that
his life m ight be rendered more miserable than it was before.5
Koster even questioned the basis for such legal rights.6

In fact, self-purchase was a widespread practice only under customary
law; no owner had a legal obligation to come to terms with a slave who was
willing to purchase freedom. Thus, as the Brazilian anthropologist Manuela
Cam eiro da Cunha puts it: w[T]he law was not silent— it was silenced” 7 O nly
the owner could grant freedom to a slave; the State maintained the widely
prevailing practice o f self-purchase as a common law. This created a hierar
chy o f subordination and personal dependence o f slaves on individual ow n
ers even for those gaining their freedom legally. Not only did such private
control o f manumission maintain the subjection o f slaves as a system, but
it also successfully produced a group o f ex-slaves as dependent laborers.6
Thus, in the practice o f manumission, power was exercised over slaves in
extremely personalized forms; without having any official apparatus that
could protect their common-law rights in manumission, a slave who aspired
to gain freedom had to maneuver carefully in developing a personal rela
tionship with the owner who might give the slave easier access to the path
for freedom. Regardless o f the wide prevalence o f manumission, the accu
mulation o f sufficient capital to fully cover the price o f one’s own value as
a commodity could not free the slave from slavery without the owner’s con
sent. In other words, any slave who was eager to be manumitted had to be
have well in the eyes o f the owner; the slave had to know his or her “place”
and walk a thin line in the ongoing relationship with the owner. Freedom
was possible only w ith the consent o f the owner, who would continue to
exercise power over ex-slaves. Over time such highly personalized owner-
slave relations in the practice o f manumission seem to have contributed
greatly to the establishment o f distinctive patterns o f race relations in Bra
zil, which are characterized by patronage and clientage.
In Salvador during the first three decades o f the nineteenth century,
African-born slaves outnumbered Brazilian-born slaves at least by the ratio
o f 2:1, but the former were by no means favored over the latter by the larger
society in the practice o f manumission. Brazilian-born ex-slaves always
The Re-creation o f Identity; 1808-1831 75

outnumbered their African-born counterparts throughout the nineteenth
century.9 In urban slavery in Salvador, gender did not determine one’s op
portunity to gain freedom among the African-born population. T he male-
female ratio among the African-born ex-slaves was nearly equal, with a ratio
o f 47:53. By contrast, among the manumitted o f Brazilian birth, women out
numbered men by tw o to one, as we shall discuss in chapter 6.10Neither did
ethnicity become a crucial factor affecting the opportunities for individual
slaves o f African birth to gain freedom by manumission. The geographic
and ethnic distribution o f African-born ex-slaves in Salvador between 1808
and 1884 agrees with that o f African-born slaves purchased and sold in Sé
parish for 1838-1848 and 1852-1888, as well as o f slaves o f African birth reg
istered in their owners’ inventories, which I consulted for 1808-1849 and
1850-1888.11 It makes a sharp contrast to the mining towns o f Minas Gerais,
where twice as many slaves from “Costa da M ina” were freed as slaves from
the Congo-Angola region. This is attributable not to specific ethnicity but
to advanced m ining technology, which had been developed in the Costa
da Mina region o f West Africa; Kathleen Joan Higgins suggests that those
slaves from Mina had more experience in m ining and were successful in ac
cumulating quantities o f gold sufficient to buy slaves themselves.12 Such
technological advantage demonstrated by a specific (ethnic or regional)
group o f African-born slaves in their pursuit o f freedom was not seen in the
case o f urban slavery in nineteenth-century Salvador.
African-born slaves, who had been born free in their homelands, often
planned and participated in slave uprisings. They remained the cultural
“other” in the larger society’s view mainly due to their continuing usage o f
their native languages, w ith the relatively limited acquisition o f Portuguese,
as well as their distinctively “exotic” cultural behaviors, which were per
ceived as dangerous to the maintenance o f social order. African-born slaves’
undeniably ostensible otherness provoked a great amount o f suspicion in
slave owners, and it impeded them from cultivating and developing inter
personal and paternalistic relationships with their owners in the same m an
ner as their Brazilian-born counterparts could with little trouble.
Thus African-born slaves o f both sexes did (or could) not benefit from the
practice o f unpaid manumission as frequently as did their Brazilian-born
counterparts (46.1 percent for African-born and 74.1 percent for Brazilian-
born for the period 1808-1844; and 48.0 percent for African-born and 65.0
percent for Brazilian-born for the period 1851-1888). Therefore, for African-
born slaves, regardless o f gender and ethnicity, the most prevalent path to
legal freedom by manumission was through paid manumission; half o f the

manumitted o f African birth paid for their own freedom (47.6 percent for
males and 52.8 percent for females for 1808-1842; 63.3 percent for males and
51.1 percent for females for 1851-1884). By way o f contrast, a much smaller
percentage o f Brazilian-born slaves (17.8 percent for men and 21.4 per cent
for women for 1808-1842; 28.9 percent for men and 24.5 percent for women
for 1851-1884) were manumitted by self-purchase.13
Having learned that self-purchase was possibly the only means available
to them, African-born slaves were forced to devise their own strategies and
tactics. Was it even possible for them to accumulate enough capital for self-
purchase, first o f all? How could a slave and the owner come up with a m u
tually agreeable price for the former’s self-purchase? And how did the slave
pay the price for freedom? African-born slaves were desperate to free them
selves from New World slavery and were very eager to utilize all that might
work to their advantage. Small wonder that for the purpose o f exploitation
some African-born slaves did definitely look for the even more powerless
than themselves, namely newly arrived African-born slaves, whose market
values were much lower than theirs and who did not comprehend the Por
tuguese language and lacked familiarity with Luso-Brazilian culture.
Aspiring to self-purchase and possessing good knowledge o f his or her
own occupational skills and capacities, the slave had to make a realistic as
sessment o f how to achieve the goal. First, the slave had to know how much
it would cost. The price o f freedom did not necessarily correlate with the
commercial value o f an individual slave; price for self-purchase was deter
mined situationally, often based on the slave’s capacity to skillfully negoti
ate with the owner, who retained and exercised the absolute power over the
former.14 Chances were that, because o f their cultural alienation from the
larger society, as well as even from each other, and their lack o f interpersonal
ties with their owners, African-born slaves were easily forced to pay much
more for self-purchase than their Brazilian counterparts o f the same sex
with comparable ages, qualifications, and occupational skills, but there is no
way o f proving this point. Unfortunately, the data on slave prices in Sé par
ish for 1838-1888 and my sample o f manumission letters in Salvador for
1808-1884 are not compatible enough for us to make valid generalizations
on a possible correlation between the slave’s commercial value and the price
actually paid for self-purchase. Yet, based on the existing data in hand, no
matter how limited they are in terms o f both quality and quantity, we are
able to attempt to speculate on some general tendencies. On the one hand,
we certainly take note o f a clear parallel between slave prices and prices for
freedom among African-born slaves by gender; men were more expensive

than women, and they also paid more for self-purchase. In the data on slave
prices in S£ parish for 1838-1888, African-born males were usually more ex
pensive than African-born females.15 Letters o f liberty also indicate that,
w ith the exception o f the years 1871-1872, male ex-slaves o f African birth
indeed paid more for self-purchase than did their female counterparts.16On
the other hand, among slave women, when the given data are divided by
birthplace, African-born were generally forced to pay higher prices for their
freedom than their Brazilian-bom counterparts, despite the fact that the lat
ter, many o f whom were skilled domestics, usually cost more than the for
mer, who had fewer skills in domestic service.
W hen a slave came up with a proposal for self-purchase, the owner would
definitely attempt to raise the price o f freedom as high as possible, knowing
well that the slave had the capacity to pay more than his/her price as a com
modity. This is especially the case for African-born slave women, many o f
whom became successful entrepreneurs in informal markets o f the city, as
discussed in chapter 2, and could easily earn much higher wages than skilled
domestic slave women. This point is supported strongly by the data on daily
wages earned by African-born ex-slaves (133 men and 197 women) who lived
in Santana parish in 1849; successful market women earned much more than
skilled domestic servants. T he average daily wage o f eighty-nine market
women ( mercadejas) was 805 r&s, and that o f twenty-eight female market-
stall keepers (quitandeiras) was 981 r&s. Thirteen women in business (ne-
gdcio) earned a higher average daily wage o f i$283 rtis. By contrast, a female
domestic servant with special skills as a laundress (nine individuals) earned
only 320 r&$ a day.17 Thus, market women earned much higher wages and
could indeed afford to pay more for self-purchase than could skilled female
domestics.
How could the slave accumulate sufficient cash and eventually make a
payment to the owner for self-purchase? A fter all, the slave, who was prop
erty, had no property rights. W hatever a slave “owned” legally belonged to
his or her owner and could be taken away at any moment i f the owner had
not previously consented to the slave’s owning property. It is very likely that
the slave, who was hired out to the street every day on a full- or part-time
basis, saved cash little by little in complete secrecy from the owner, as well
as from fellow slaves, who might just steal it if there was any chance. The
slave could not be too careful about it. There were no authorities a slave
could appeal to if savings mysteriously disappeared. Some could eventually
manage to come up w ith sufficient savings in cash for self-purchase; in feet,
it was most common that the slave made a payment by cash. But cash pay-

ment was not always the case. An example found among the ex-slaves' wills
is payment by real estate. A Nagô ex-slave woman named Gertrudes Maria
do Espirito Santo obtained her freedom w ith the payment o f her value to
her owner, Silveiro da Silva, and his wife Joana da Silva, w ith a house located
on Rua do Genipapeiro o f Santana parish, which she owned among a few
others.18 Although it was not officially recorded in nineteenth-century Sal
vador, some slaves purchased freedom not only with real estate but also with
cattle in Paraiba from 1850 to 1888.19
The most extreme form o f payment was w ith another enslaved person
(or tw o persons, albeit rarely) through substitution, or “trade-in ”20 Let us
take one example first. A Nagô slave man named Francisco obtained his let
ter o f liberty on M ay 14,1852, by paying his owner the equivalent o f 7008000
réis, the price o f his freedom. Instead o f paying cash, Francisco substituted
another male slave called Joâo, whom he, Francisco, owned. Joâo was also
Nagô and subsequently took Francisco's place w orking on a small boat
known as an alvarenga, which is a lighter used for ferrying goods between
ship and shore.21 To our disappointment, Francisco’s letter o f liberty does
not tell how he had accumulated enough money to purchase Joâo, nor how
Francisco had come to own his own slave, albeit temporarily, presumably
with the permission o f his owner. Francisco, employed in an urban occupa
tion, possibly for many years, negotiated w ith his owner to arrive at a m u
tually agreed upon price for Joâo, a price equivalent to what the owner was
demanding for Francisco’s freedom, and the owner agreed to accept Joâo in
lieu o f cash payment. Interestingly, Francisco provided his owner with a
substitute slave not only o f the same sex, but also a Nagô, the same ethnicity
as himself.
In my sample o f 3,516 letters o f liberty for nineteenth-century Salvador,
there are a total o f 35 recorded cases o f substitution, including the aforemen
tioned one o f Francisco. These cases o f self-purchase transactions through
substitution constitute only one percent o f the total number o f letters o f
liberty, and 2.63 percent o f self-purchases (1,332), for the period 1808-1884.
The number and percentages are certainly small, but it is important to note
that substitution did exist as a special form o f self-purchase and that its
practice by the enslaved population was officially acknowledged and re
corded in notarial records. The scrutiny o f substitution will enable us to
discuss a fascinating aspect o f the re-creation o f identity by the African-
born population.
In nineteenth-century Salvador, the slave’s gender did not determine the
practice o f substitution; women and men were equally represented among
The Re-creation o f Identity, 1808-1831 79

those who purchased freedom by substitution. The 35 ex-slaves who ob
tained freedom through substitution numbered 17 males and 18 females. On
the other hand, it seems that substitution was predominantly an “African”
phenomenon; the 35 ex-slaves were composed o f 24 African-born o f diverse
ethnic origins (13 males and 11 females); 3 Brazilian-born (1 male and 2 fe
males); and 8 persons without any indication as to origin or color (3 males
and 5 females).22 Also, substitution took place when the transatlantic slave
trade was intact and therefore newly arrived African-born slaves were avail
able for sale; with one exception, which took place in 1861, all cases o f sub
stitution occurred between 1808 and 1852.
In the practice o f substitution, one-to-one replacement was common; in
only tw o cases were tw o slaves o f the same gender substituted for one per
son to be freed.23The common phrase used in cases o f substitution was “the
slave gave another slave in his/her place,” but specific occupations were
mentioned in four letters, including the letter o f Francisco.24The slave to be
freed usually chose a substitute slave o f the same sex (29 cases out o f 35). In
three cases the ex-slave provided not only a substitute but also cash to make
up the gap between the price o f the ex-slave and the lower price o f the sub
stitute. One example illustrates such a negotiated settlement between a slave
and an owner. An African-born slave man named Ventura, whose “nation”
was M ina and whose occupation was a bricklayer, paid 300$000 réis to his
owner: 200$000 réis represented by the value o f another African-born slave
man named Torcano, who was o f the same occupation and whose “nation”
was “gentio da Costa” (gentile from the coast), and ioo$ooo réis in cash.25
Certain naming practices in the procedure o f substitution reveal a fasci
nating aspect o f identity politics for the enslaved; the African-born popula
tion was well aware o f how they could play with their identity for their bene
fit. In tw o cases the slave to be freed and his/her substitute shared the same
Christian/Portuguese name.26 As discussed in chapter 2, the African-born
population in Salvador often identified themselves, and referred to one an
other, by their original African names, whereas their Christian/Portuguese
names, given at baptism, were employed by their owners and the non-African-
born population in general; such new names represented their continued
enslavement in the New World, not their personhood in the O ld World.
Therefore, the slave might well intentionally name the substitute w ith the
same Catholic/Portuguese name to show others his or her legal ownership
o f the substitute, while at the same time using the original African name
when communicating with the substitute slave in their native language.
Few letters o f liberty tell us how a slave obtained a substitute, as in the

case o f Francisco, and only tw o letters state who supplied the substitute. A
slave woman, Josefa, received her substitute slave, named Rita, from her
niece Marcelina Maria da Conceiçào. In the other case a crioula slave, Bar-
nardeira, obtained her letter o f liberty by providing her owner w ith an
other female slave, named Rola, African-born and Nagô, who had belonged
to Barnardeiras mother, Felicidade, also still enslaved but under different
ownership. In both cases, a kinswom an who herself had become a slave
owner made a significant financial contribution to the manumission o f a
slave woman.27
One wonders how some slaves and ex-slaves in Salvador could manage to
obtain and own newly arrived African-born slaves. For rural Bahia, Stuart B.
Schwartz presents interesting testimony made before the probate judge o f
Sâo Francisco do Conde in 1836 which suggests that it was indeed possible
even for slaves in Salvador and other parts o f Bahia to purchase newly en
slaved human commodities in Africa. A crioula slave, Luciana Maria da C on
ceiçào, wanted to purchase a slave as a dowry for her granddaughter. For
that purpose, she sent her money to a friend, who went to Africa and ac
quired a Nagô woman who was to be named Jeronima at baptism. Jeronima
was subsequently delivered to Luciana as her slave at Engenho Cahipe. De
spite her initial motivation, Luciana changed her mind; she kept Jeronima
as her own personal slave and sent her to the city for wage earning, while
Luciana herself continued to w ork as a slave on the plantation.28 We do
not have the ending to this story. It is noteworthy that while working on a
plantation Luciana could manage to save enough money to purchase an
other slave. One should not be amazed at the fact that this slave woman,
who remained enslaved herself, wanted her granddaughter to own a slave
wom an as a dowry. We wonder if Luciana was finally freed, possibly by
the accumulated earnings o f her own slave Jeronima or by substituting
Jeronima for herself. We do not even know if Luciana made the effort to gain
her own freedom. But this example does serve to illustrate how agents, pos
sibly African-born ex-slaves, traveled to Africa on a regular basis to trade in
slaves and to make them available at reasonably inexpensive prices to buyers
such as Luciana in Salvador.29
Newly arrived Africans were usually chosen as substitute slaves because
their prices were less than those for skilled Brazilian-born slaves and also
acculturated African-born slaves.30 This explains why the practice o f sub
stitution virtually disappeared with the termination o f the transatlantic
slave trade (1851). Twenty-five letters declare that substitute slaves were
African-born and/or new arrivals; the other ten letters do not provide infor

mation as to the slaves’ place o f birth or “nation.” W ith the consent o f his
or her own owner, a slave purchased the substitute, acculturated and trained
that slave in special occupational skills, and finally '‘traded in” the slave in
exchange for freedom.
As the case o f Francisco illustrates, some slaves purchased as their sub
stitute slaves o f the same “ nation.” O f nineteen cases in which both the
slave and the substitute were identified, the replacements for six Nag6 slaves
were also Nagd. This could be attributed to the law o f probability because
individuals defined as Nag6 comprised one-third o f African-born ex-slaves.
But studies on manumission in colonial and imperial Brazil by Schwartz,
Karasch, and Higgins also note the coincidence o f “ nation” between the
slave and the substitute, thereby confirming that this coincidence was not
confined to any specific ethnicity.31
We conclude that substitution was an expedience attractive to both owner
and slave. Karasch is correct in stating that “ (ojwners w illingly accepted
these ‘trade-ins,’ for they did not have to acculturate them, and they solved
the problem o f capital depreciation. In place o f an aging African woman,
they might receive a teenage boy with years o f service in the future.”32 For
the slave, the financial advantage o f purchasing a new African-born slave for
much less than he or she would need to accumulate for self-purchase was
offset by the consideration that the slave had to take time to acculturate the
new slave before the latter was a viable replacement. Some slaves, in their
capacity as slave owners, may have worked side by side with their own slaves,
or sent them out onto the street as wage-earning slaves to earn money for
several years, before finally offering them as substitutes for themselves. In
this context, purchasing a substitute who shared the same ethnicity made
sense. This eased the problem o f instructing the new arrival, who did not
understand the Portuguese language or Luso-Brazilian culture. The sharing
o f the same ethnicity between the ex-slave and the substitute was no mere
coincidence, as Schwartz and Karasch imply. Rather it reflected the creation
o f ethnic identity in New World slavery, in which enslaved individuals o f
African birth struggled to survive on a daily basis. Cognition and behavior
are different phenomena; sharing the same ethnic identity does not mean
that such individuals would autom atically create ethnic group solidarity
and/or behave as members o f the same ethnic group.
The practice o f substitution was not exclusive to Salvador but, from what
we have learned from preceding studies on manumission, seems to have
been m ainly used in an urban environment, such as the port cities o f Sal
vador and Rio de Janeiro and the mining towns o f Minas Gerais.33 Substi

tution was less prevalent in agricultural regions, such as Paraty, Paraiba, or
even the Bahian Rec6ncavo.34 This should not be taken to imply that sub
stitution was an exclusively urban phenomenon, but substitution rarely oc
curred in rural agricultural areas because o f the plantation slave system and
the relative lack o f socioeconomic m obility o f slaves who were field hands.
In one exception at the beginning o f the nineteenth century, Henry Koster
observed a similar practice on a sugar plantation o f the Benedictines in
Jaguaribe in Pernambuco. More than one hundred slaves, all Brazilian-born,
belonged to this plantation, whose overseer was a male mulatto slave. He
first purchased the freedom o f his mulatto wife, who had been owned by
the Order, and then their children’s freedom. Subsequently, he offered the
friars, in exchange for his own freedom, tw o African-born slaves whom
he owned. The friars refused to accept his proposal on the grounds that the
estate could not be properly managed without him .35
Substitution as a form o f self-purchase was unique to Brazil, as far as
New World slavery is concerned. The origin o f substitution as self-purchase
goes back to Roman times. In imperial Rome, which constituted an urban
slave society, there was a special feature o f some slaves' condition, called the
peculium, which was a fund o f money controlled by slaves themselves. The
slave could use it for investments or, ultimately, for self-purchase. Further
more, slaves could use the peculium to purchase other slaves to stand in for
them (vicariuslvicaria). The peculium was usually reserved for urban skilled
slaves, not for rural workers.36 Among many studies on manumission out
side o f Brazil, the only example I have found o f substitution is o f a Man-
dingo slave in Port-of-Spain, Trinidad, in 1830, very close to the date o f
abolition in 1834. The Muslim preacher named John Mohammed Bath pur
chased a slave as a substitute for him self as a “colonial negro" o f the British
government, since the government had declined to accept Bath’s money for
his self-purchase. The urban Mandingo community, whose members were
Muslims, used part o f its considerable economic assets to function as an
emancipation society. As in the case o f nineteenth-century Salvador, some
Mandingos in Trinidad became slave owners. Carl Campbell interprets the
retention o f African names as evidence o f their determination to “return”
to Africa, but does not address the process o f substitution in terms o f eth
nicity and ethnic identity in New World slavery. Campbell's implicit as
sumption is that the Mandingos, who shared a strong ethnic identity, owned
non-Mandingo slaves, and some o f them “traded in” non-Mandingos for
their own freedom.37
The African-born population’s re-creation o f identity as ex-slaves started

to take place while they were still enslaved but at the very moment when
they became determined to recover their lost freedom. Such highly m oti
vated individuals o f African birth found some interesting ways to manipu
late their ethnic identities for self-purchase, as the case o f substitution
clearly illustrates.
To Be an Ex-Slave o f A frican Birth:

Neither Enslaved nor Free
The acquisition o f a manumission letter by self-purchase did not to

tally end the African-born individual's desperate struggles for freedom. The
letter o f liberty was a legal document for transfer o f property (the slave)
from the former owner to the new owner, namely the slave him self or her
self, and this had to be formally registered and publicly notarized before a
notary (tabaliao). Once this form ality was completed, the ex-slave was le
gally free. It was the ex-slave’s awesome responsibility to preserve the origi
nal letter o f liberty; loss or inability to produce it on demand would make
the ex-slave vulnerable to re-enslavement.
Freedom, o f course, did not alter much o f the external appearance o f
African-born ex-slaves; they could not be easily distinguished from their
enslaved counterparts, who constituted the majority o f the African-born
population. W hether enslaved or freed, most had been born free in Africa,
and their shared cultural otherness distinguished them from the Brazilian-
born population o f African descent. There were only a few visible signs o f
their newly acquired free status. First, ex-slave street laborers who worked
in gangs, as porters, transporters, and artisans were not chained at the ankle
or neck. Second, ex-slaves were entitled to wear shoes; the British lady Maria
Graham describes shoes as the “mark o f freedom” in her travel journal.38
Perhaps with their shoes on, African-born slaves o f both sexes continued
to work with their co-workers o f African birth, both slaves and ex-slaves,
side by side, and were engaged in the same occupations as when they had
been enslaved. Their jobs could have been stigmatized by association with
slavery, and the free-born population may not have wished to take them
up. But it was African-born people’s unique occupational skills that en
abled them to earn extra money as slaves and to purchase their freedom
in the end. African-born ex-slaves usually continued to be engaged in the
same occupations, in which they could fully utilize their ethnic back
grounds and/or take full advantage o f the enslaved population o f African
birth.

Gender determined the African-born individuals experiences to a con
siderable degree. As discussed in chapter 2, African-born women were most
prominent in the inform al market o f the city and gained a substantial
amount o f financial power; as examined earlier in this chapter, their mar
keting skills enabled them to purchase freedom with cash, usually with the
payment o f more than their commercial value as slaves. Among the African-
born ex-slave population, women continued to dominate the urban infor
mal market. For instance, in 1846, in Conceiçào da Praia parish, nearly 90
percent o f market-stall keepers among 331 African-born ex-slaves were fe
males. Likewise, among 207 African-born ex-slaves in Santana parish in
1849,92.7 percent o f market persons and 87.5 percent o f market-stall keep
ers were women.39
As for men o f African birth, skilled artisans made up an important por
tion o f the male ex-slave population. Am ong the African-born freedmen
who lived in Conceiçào da Praia parish in 1846,15 out o f 96 men were arti
sans.40In Santana parish in 1849, African-born ex-slave artisans (who num
bered 25 out o f the total number o f 133) earned an average daily wage o f 962
réis, which is significantly higher than that o f chair carriers (794 réis), as
well as o f those who were classified as wage-earners (793 réis).41 Because o f
the better econom ic opportunity enjoyed by slave and ex-slave artisans,
the number o f free artisans o f African descent gradually grew in colonial
and early nineteenth-century Salvador, and their increasing financial power
eventually enabled the establishment, at the mid-nineteenth century, o f an
unusual voluntary association called the Society for the Protection o f the
Needy, which we shall discuss extensively in chapter 7.
Among male skilled artisans, barber-surgeons occupied a unique position
and constituted a significant part o f the African-born male ex-slave popu
lation. Despite the fact that full-tim e slave barber-surgeons were only about
one percent o f the whole male slave population, barber-surgeons consti
tuted a significantly larger part o f the ex-slave population.42 Six barber-
surgeons represented 6.3 percent o f the African-born freedmen in Con-
ceiçâo da Praia parish (1843), whereas five ex-slave barber-surgeons (3.8
percent o f the total African-born ex-slave population) earned the highest
daily wage, 18384 réis, o f all wages in Santana parish (i849).43 Am ong the
wills of ex-slaves, tw o ex-slave testators o f African birth identified their pro
fessions as barber-surgeon.44 One o f them, Bernardino da Sena, made his
will in the 1820s despite the fact that he was in perfect health. He intended
to “return” to Africa to live there until his death; he chose to take passage
by being employed as a barber-surgeon on the ship Gema Marina, owned

by Alves da C ruz Rios.45 Another ex-slave originally from Angola, Antônio
José da Silva, earned the wage o f i$ooo réis a month as a sailor on a smack
owned by Francisco José Vieira, who resided both in Angola and in Rio de
Janeiro, and whose ship's master was José Gomes da Rocha. The smack was
engaged in the slave trade from Angola to Pernambuco.46
Some African-born freedmen, in fact, became slavers themselves. Such
was the case o f the Hausa freedman named Francisco da Rocha (see chap
ter 3). Francisco made no mention o f his ex-owner's occupation or his own
previous occupation as a slave, but it is possible that Francisco had worked
as a sailor or hired out as a wage-earning slave on a ship to West Africa.
Francisco’s Hausa origin, together with his bilingual skills, may have facili
tated his entry into the slave business in Salvador, where there were many
slave traders who were themselves o f African descent. Later his ethnic ori
gins, coupled with his linguistic abilities, may have helped Francisco to es
tablish good business relationships w ith local slavers and merchants in the
ports o f West Africa. Francisco had a business contract with Manoel José de
Almeida, resident o f Salvador, for the supply o f slaves; Francisco ordered
Manoel to arrange for the transportation o f slaves from Costa da Mina, and
Francisco was to receive 6:500$000 réis for having sold four slaves. Luis
Marques, overseer o f the schooner Esperança, owed Francisco 85$ooo réis,
the remainder to be paid on the sale o f one newly arrived slave boy, one
female slave, three barrels o f olive oil, and one sack o f pepper. From this
voyage Francisco brought back as cargo slaves for various persons. He also
did business with a prèto, Bernardino da Sena, in Salvador, and arranged two
assignments for José da Costa de Miranda, which resulted in three slaves.
Francisco owed to the African-born ex-slave Angelo i3$ooo réis, which
Francisco had received from Angelo to employ Angelo in Costa da Mina,
although such employment did not materialize.47 Thus Francisco da Rocha
had developed extensive business networks in Salvador with business part
ners, employees on ships, and his customers, among whom were the free-
born population o f African descent and ex-slaves. The port city o f Salvador
gave Francisco the opportunity to decide for him self to engage in the slave
trade with Africa.
The British traveler Robert Walsh in Rio de Janeiro had noted the in
volvement o f former slaves o f African birth in the transatlantic slave trade
in 1828. “ There has been such a range for acquiring this sort o f property,”
he wrote, “that negroes themselves who had obtained their freedom, fre
quently sent ventures to Africa to purchase their countrymen, who were
brought back to them in exchange for the beads and looking-glasses which

they sent out There were many African-born “ Brazilian” merchants and
slavers in the late eighteenth and early nineteenth centuries who returned
to reside in West African ports, such as Agoué, Porto-Novo, Badagry, Popo,
Lagos, or Abéokuta, at least for a part o f the year.49
As a result o f transatlantic trade in which African-born male slaves and
ex-slaves played an im portant role, com m unication networks developed
between Salvador and various parts o f Africa from which not only men
but also women benefited greatly. Some maintained contact with their kins
folk in Africa, sent their Brazilian-born children to them, or themselves
emigrated to Africa after having been freed.50 Maria Antônia, an Angolan
woman, was m anum itted unconditionally by her owner, Bernardo Silva
Mello, solely because o f her “good services ” She gave birth to a pardo son
named Raphael Fortunato da Silva, who embarked for Costa da M ina in
1830. She did not even know whether he was alive or not, although she
named him as her primary heir; as such he was to inherit two-thirds o f her
possessions.51 The free-born crioula woman Jeronima Maria da Conceiçâo
was the child o f the Gêge woman Anna Joaquina and Bernardino José Bar
bosa, both o f whom were dead. Jeronima had been raised in the house o f a
businessman who lived in the lower city, José Alves C ruz dos Rios, and his
wife, and later married an African-born man named André da Conceiçâo
da França. They had a son, Manoel, who died at age nine months. Then
André went alone to the coast o f Africa, jeronima named her former owner
as executor o f her w ill and her sister, Joanna Francisca, as her heiress.52
T h e Re-creation o f Identity as “ Free”
Upon gaining freedom against all the odds, African-born ex-slaves

immediately adopted fam ily names. For an ex-slave, having a last name,
which symbolized the status o f being legally free, was very important to re
store human dignity.53 In nineteenth-century Salvador, as in many other
parts o f the Americas, ex-slaves usually adopted the last names, either in full
or in part, o f their ex-owners.54 This is also the case o f the Bahian Recôn-
cavo. For instance, at the Jesuit-owned Engenho Santana o f Ilhéus in 1752,
all slaves were listed with family names, since the Jesuit administration had
assigned family names or asked the slaves to select them in order to regu
larize family life among slaves. Am ong 108 slaves whose family names could
be easily established, more than 70 percent bore the names o f administra
tors o f the estate.55 At least in the city o f Salvador, the naming custom ap
pears to have been practiced on the initiative o f the ex-slaves and does not

seem to have been used by ex-owners to identify or control slaves. So com
mon were some first names, such as José or Maria, among different own
ers* slaves, some o f whom often worked together as wage-earners, that it
was necessary to differentiate between them. Maria, for example, became
known as Sr. Santana’s Maria. Therefore, it was convenient for ex-slaves to
adopt their owners’ surnames, by which they had been commonly known
for many years, as their own formal surnames upon being freed.
Furthermore, African-born ex-slaves were very eager to be perceived as
free by the larger society. Therefore they mimicked the white elite and took
up their distinctive behaviors and attitudes, most notably slaveholding and
multi-membership in lay sodalities.
The majority o f African-born ex-slaves could not afford to own slaves or
any other forms o f property. They had made every effort to purchase free
dom, but that does not mean that all o f these people found it possible to
escape from poverty and successfully become property owners. For instance,
79.7 percent o f African-born ex-slaves (263 out o f 330) who resided in San
tana parish (1849) and 89.9 percent o f them (186 out o f 207) in Conceiâo
da Praia parish (1846) did not own a single slave. But when some o f the
African-born ex-slave population— naturally a very small percentage— did
become successful enough to own property, they usually chose to invest
their money in slaveholding; after all, Salvador in the early nineteenth cen
tury was a slave society, in which only slaveholding could make a person
a member o f the society. Some ex-slaves (36 for Santana and 17 for C on
ceiâo da Praia), both men and women, actually owned more than one slave,
and one African-born ex-slave in Conceiâo da Praia parish owned nineteen
slaves.56The wills o f African-born ex-slaves support this point very clearly;
usually only property owners registered their wills for the sake o f inheri
tance. For instance, among the 261 African-born ex-slaves who registered
their wills between 1790 and 1850, slave owners constituted the majority:
72.3 percent for men (81 out o f 112), and 82.9 percent for women (122 out
o f 147).57 Slaveholding was certainly a norm among African-born ex-slave
property owners who could afford it.
M any such ex-slave property owners o f African birth simultaneously
owned both African-born slaves, including those who shared the "nations”
with them, and crias, namely Brazilian-born slaves raised in their owners’
households. Interestingly, those ex-slave property owners o f African birth
treated each group differently; they favored the Brazilian-born slaves over
their African-born counterparts.58This “double standard” is readily appar
ent from those owners’ wills. Whereas many Brazilian-born slaves were lib

erated unconditionally on their owner’s death and even were occasionally
named as heirs to an owner's estate, many African-born slaves— including
those o f the same “nations" as their owners'— were usually coartado, that is,
obliged to pay a fixed price to the heirs for their freedom within a limited
period o f time after the death o f the owner.59 A Nagô ex-slave might have
instructed his or her Nagô slaves in their common language and demon
strated their shared cultural behaviors, but this did not necessarily mean
that the owner favored the slave who shared the owner’s ethnicity.
The property-owning African-born ex-slave population also held mul
tiple memberships in black lay sodalities; they adopted an ostentatious at
titude they had observed in the white laity’s participation in lay sodalities,
not only in their own white sodalities, which limited memberships only to
whites, but also in mulatto ones, which did not allow African-born people
and crioulos to join them, and even black sodalities, as discussed in chapter
3. In the late colonial period and the early nineteenth century, holding the
multi-membership o f sodalities was a common practice among well-to-do
African-born ex-slaves, many o f whom were slave owners themselves. For
instance, among the 159 wills o f ex-slaves (63 men and 96 women) registered
in Salvador for the early nineteenth century, nearly half were members o f
one or more sodalities, and a third were affiliated w ith more than one so
dality.60
A good example is Ana Maria do Carmo, an African-born woman origi
nally from Costa da Mina, who registered her w ill on July 4,1811. Ana Maria
do Carmo was an ex-slave o f the Reverend Luis de Sousa, who had liberated
her on payment o f 84$ooo réis. She herself owned three African-born fe
male slaves (Antônia, Maria, and Rosa), all also from Costa da Mina, and
had limited assets in the form o f gold, furniture, and used clothes. She had
no children by her late husband Manoel Martins de Miranda da Cruz, or by
anybody else, so she ordered that her property was to be inherited by José
Maria do Carmo, who had been her slave but had been liberated. Ana Maria
do Carmo declared in her w ill that the habit o f Saint Francis should be her
shroud, that the vicar o f her parish should officiate at the funeral service,
and that her cortège was to be accompanied by sixteen priests. Her burial
was to be in the church o f her sodality o f O ur Lady o f the Rosary in the
Pelourinho, where she had held the office o f judge. Ana Maria was also a
member o f the sodality o f O ur Lady o f the Rosary in the church o f Santo
Antônio parish. On the day o f her death, the executor o f her will was to say
twelve masses for her soul, and another twelve masses for the saint o f her
name, each at the fee o f 320 réis. She also left an additional 3$ooo réis for

masses to be said for her former owner, for twelve masses for the soul o f her
late husband, and for another twelve masses for her soul, each for the fee o f
320 réis. Her funeral was to be accompanied by sixteen members o f the so
dalities to which she had belonged, and each o f them would be given alms
o f 40 réis for such participation.61
An extreme example was the case o f the African-born freedman M axi-
miano de Freitas Henrique. He had been a judge o f the black sodality o f O ur
Lady o f the Rosary in the Pelourinho and a treasurer o f the black sodality
o f G ood Jesus o f Redemption (Sr. Bom Jesus da Redempçào), and served on
the advisory boards o f the black sodality o f Saint Ephigenia in the Car
melite monastery and the black sodality o f Saint Benedict in the same con
vent. He was also a member o f the black sodality o f Our Lady o f the Rosary
on Rua de Joâo Pereira and o f the black sodality o f Jesus, Mary, and Joseph
(Jesus Maria José), whose sect was the Carmelite monastery. In his w ill, he
stipulated burial in the Franciscan monastery, accompanied by all members
o f the black sodalities o f which he was a member.62
We have to emphasize the cruel reality that freedom did not change the
larger society’s perceptions o f African-born ex-slaves; no matter how afflu
ent African-born ex-slaves became, their wealth never enabled them to hold
memberships in either white or mulatto lay sodalities; they were entitled to
belong to black sodalities only. Their multi-membership in black sodalities
carried prestige and signified their superior position as well-to-do ex-slaves
only among the population o f African descent; it did not make a significant
appeal to the larger society as long as such ex-slaves’ extended participation
was confined to black sodalities. Freedom could never “whiten” African-
born ex-slaves.
Conclusion
Enslaved Africans, most o f whom had been born free in the Old
World and remembered what freedom had meant, made all possible at
tempts to regain freedom from New World slavery. Their desperate efforts
resulted in the prevailing practice o f substitution as a form o f self-purchase,
through which slaves themselves became slave owners, rather than helping
other slaves gain freedom; their shared experiences o f enslavement (and ex
ploitation) did not necessarily help the enslaved (and the exploited) develop
a collective identity as enslaved Africans in the New World, with which they
would have acted as a group against the institution o f slavery. Realizing that
self-purchase was virtually their only means to gain freedom, African-born

slaves had to use and manipulate anything available that could work to their
advantage, such as their shared ethnicity with new arrivals.
After many years o f hard work, some African-born ex-slaves surely gained
what they had yearned for: the recovery o f their lost freedom, but it was not
the same freedom that most o f them had experienced as free-born in the
O ld World. They immediately recognized the limitations o f their regained
freedom and struggled to become “free” in a slave society. Sadly, they always
ended up identifying themselves as ex-slaves only by distinguishing them
selves from the enslaved population.

5 The Convergence of Identity,
1831-1880
Upon regaining freedom by self-purchase, African-born women and men,

who had learned how to utilize their ethnicities to their advantage, expected
to be treated as equals with the white elite, but they immediately learned
that their regained freedom had lethal limitations: their freedom had been
fatally damaged and permanently crippled. African-born ex-slaves learned
the white elite’s distinctive values and attitudes, imitated their behaviors,
but nothing could elevate their ex-slave status to that o f the free-born. As
ex-slaves, they continued to wonder how to deal with their own otherness.
They did not see it in terms o f race; instead they tended to ascribe the lim i
tations o f their freedom to their African birth, which was associated most
notably with their imperfect acquisition o f the Portuguese language and
their insufficient understanding o f Luso-Brazilian culture. It made their
naturalization impossible and therefore kept them from gaining equal legal
rights as Brazilian citizens.
This chapter discusses the processes by which African-born ex-slaves, as
early as the 1830s, came to develop a newly convergent ethnic identity as
African-born. It fostered the formation o f a distinctive ethnic com m unity
on both sides o f the Atlantic Ocean after the mid-nineteenth century: as
“African-born” in Salvador and as “ Brazilian” back in Africa.
The Creation o f A frican -B orn (Africano ) Identity
The African-born ex-slave population started to converge their di

verse ethnic identities into a broader one during the 1830s, in response to
the larger society's changing perceptions o f them. The Bahian authorities
had become more and more alert to and suspicious o f the activities o f the
African-born population after a series o f slave uprisings took place between
1807 and 1830. The law o f 1830 prohibited people o f African birth, both
slaves and ex-slaves, from traveling freely. They were obliged to be in pos
session o f passports (pasaportes) whenever they went beyond “the city, vil-
lages, settlement, or plantation and region” where they resided, whether for
reason of business or for any other purpose. Whereas the passport for a slave
had to be signed by the owner, an ex-slave had to obtain the passport from
the criminal judge residing in his or her locality. Passports indicated the
name o f the person, destinations, and permitted period o f travel. W ithout
it, any African-born ex-slave was liable to imprisonment for eight days for
transgression o f the law.1
Furthermore, the decree o f 1831 changed an important part o f the land
scape o f the port city o f Salvador, in whose busy streets newly arrived slaves
had been most commonly traded because o f the absence o f a central slave
market. W ith the enactm ent o f the decree, African-born slaves were no
longer taken directly into the port o f Salvador from slave ships. Yet, the il
legal slave trade continued to take place on a large scale from Africa to Bra
zil, including the province o f Bahia. Although no statistical data are avail
able to trace the numbers o f illegally imported African-born slaves who
were brought into the city o f Salvador, we may assume that Salvador might
have received fewer and fewer African-born slaves after 1831, due to the con
tinuing demands for slaves as plantation labor in the Recôncavo. As dis
cussed in chapter 1, slaves still constituted 42.0 percent o f the whole popu
lation o f Salvador in 1835, but the proportion declined to 26.1 percent by
1855. Yet African-born slaves continued to outnumber the Brazilian-born
slave population until mid-century. For instance, 56.2 percent o f slaves pur
chased and sold in Sé parish for 1838-1848 were African-born, while among
the slaves (1,170 individuals) registered in 154 inventories o f their owners for
1808-1849, 67 percent were African-born.2 Joào José Reis estimates that in
1835 African-born slaves constituted 63 percent o f the enslaved population
o f Salvador.3
During the 1830s Salvador remained an urban slave society, while among
the African-born population diverse ethnic identities could no longer be re
vitalized on a daily basis in the absence o f infusions o f newly arrived African
slaves o f diverse ethnic origins. Regardless o f the freedom they had ulti
mately regained, the urban African-born ex-slave population began to iden
tify themselves more and more w ith African-born slaves as African-born,
beyond legal distinctions, whereas their distinctive gender identities contin
ued to differentiate individual experiences, as a well-organized uprising in
Salvador on January 24-25,1835 clearly showed.
This armed uprising has been long known as the Malê (African Muslim)
revolt, because o f the allegedly prominent role o f Muslims among the par
ticipants.4 It was originally planned for dawn on the day o f Nossa Senhora
The Convergence o f Identity; 2831-1880 93

da Guia (O ur Lady o f Guidance), Sunday, January 25, which came after the
festival o f Senhor do Bonfim (the Lord o f the G ood Ending). O n this day,
it was the custom o f many residents o f Salvador to leave the city to attend
the church o f Bonfim, located at the northern edge o f the city, a church
noted for its miracles and, as such, a place o f pilgrimage, and later to spend
the night on Itapagipe peninsula.5 And dawn was the time o f the day when
slaves left their owners’ houses to fetch water, so many o f them could join
the uprising without arousing suspicion.6The rebels, all o f whom were wage-
earners in the city, were to set fire to several places in the city simultane
ously to distract the police and the troops. They were then to be joined by
slaves coming by water from Santo Amaro, a town o f the Recôncavo. O n the
evening o f the 24th o f January, however, the revolt was denounced to the
authorities.
Unlike the previous aborted slave uprisings between 1807 and 1830 (as
discussed in chapter 3), in the case o f the Malê revolt there are police records
and court testimonies detailing the circumstances and the personalities in
volved in this denunciation, and the story is worth telling at length here.
Prominent was the role o f two African-born fireedwomen, both Nagô.
One was named Guilhermina Roza de Souza, resident o f Vitôria parish in
the house o f her former owner, Firmiano Joaquim de Souza Velho. Guilher-
mina heard o f the uprising from the father o f her children, Domingos Fur-
tunato, a Nagô freedman. Domingos was a former slave o f Furtunato Jozé
da Cunha, who had lived in the lower city. O n the 24th, Domingos over
heard some negro boatmen (negros de saveiros) talking o f the impending
arrival o f some negros in Salvador on the 25th. They were to join with the
leader Aruna, or Uahuna, who had already arrived some days before, and to
take over the city with other negros on the following day. Their intent was
“to kill whites, cabras, and crioubs, as well as those negros o f another faction
or group (banda) who did not wish to join the uprising; the pardos would
be spared for use as lackeys and slaves.” Domingos wanted to alert his ex
owner, who was living in Pilar parish, to this impending danger, by asking
somebody to write a note for him to this effect. W hile Guilhermina was
standing at the window o f her house digesting this news, she heard a couple
o f Nagôs saying that those negros who were to rebel with the help o f those
from Santo Amaro should go to the public fountains in the lower city to
fetch water at dawn. Guilhermina went to tell her own ex-owner and, on her
way home, encountered her comadre (comother), Sabina da Cruz, also a
Nagô freedwoman o f José Manoel Gonçalves, resident o f Itapagipe. Sabina
was searching for a certain white man to warn her ex-owner that many ne-

gros had been gathering on Rua da Guadalupe and that firearms were being
distributed.7
Around four o’clock in the m orning on the 24th, Sabina had quarreled
with the father o f her children, Victoriano, also Nagô, who was known by
his African name o f Sule. A fter spending the day vending in the lower city,
she returned to her house to find everything inside a mess. Convinced that
Victoriano was the culprit because o f their quarrel in the morning, she be
gan to search for him by going from house to house o f his acquaintances.
In one such house on Rua da Guadalupe, negros from Santo Amaro were
gathered. In the corridor o f the house, Sabina saw many Nagôs assembling
and heard them talking in the language o f Nagô. She was scared and did not
go further to enter the room where the meeting was taking place. But she
did catch a glimpse o f a Nagô-Bâ (Yoruba from the kingdom o f Egba)
woman, whose African name was Edum, leaving the room. Sabina did not
know her name in Portuguese, the language o f “the white’s land” ( terra de
branco). Edum carried her baby on her back in a shawl, and Sabina recog
nized her immediately because Edum had bought some yams from Sabina
earlier on the same day. Sabina learned from Edum that Sule was in the
meeting room but would not come out until the following morning, when
the men were scheduled to “take over the land ( terra).” The yams Edum had
bought from Sabina were for Aruna, the leader, who was now inside the
room, armed and well-prepared, together w ith a lot o f people. In response
to Sabina’s statement that all o f them would be the possessors not o f the
land (terra) but o f a whip-lashing (senhor de surra) on the following day,
Edum told her to wait and see until the next day. Am ong others in the room,
Sabina recognized only a male slave whose African name was Honomonin,
and whose owner was the baker, Pedro, resident o f Pilar parish. The slave
Honomonin was rented out to Sabina’s negro neighbor, Belchior de Silva
Cunha. Terrified by what she had seen and heard, Sabina fled and ran to
report the momentous events to her comother (comadre) Guilhermina. This
was the information that Guilhermina took to the authorities.8
This short story, which I have reconstructed m ainly based on the few
African-born ex-slave women’s court testimonies, raises some interesting
points. First, the participants o f the Malê revolt regarded as their enemies
not only the whites but also all o f the Brazilian-born population o f African
descent regardless o f color, and even other ethnic groups among African-
born. As discussed in chapter 2, the African-born population in Salvador
identified themselves and were referred to by others by their original African
names. Gossip and rumor circulated quickly, often through the instrument
The Convergence o f Identity, 1831-1880 95

of African languages and thus w ithin ethnic boundaries. Second, in the
court records o f the Malê revolt, Salvador and/or Bahia was characterized
as “the white's land” in distinction to Africa as “our land .”9 This we-they
dichotomy indicated that the African-born population began to develop a
convergent ethnic identity as African-born in opposition to the white elite
in Salvador after having spent a number o f years in the New World. Third,
this story reveals the existence o f associations among women o f African
birth. The tw o Nagô freedwomen, Guilhermina and Sabina» were related to
each other by fictive kinship as comothers, and seem to have developed
communication networks and commercial associations with other women
o f African birth who were engaged in the informal market o f the city. The
market was the place where symbolically important commodities, such as
yams, were brought by water from the Recôncavo, and were used for can-
domblé rituals and traded by such women o f African birth. It is significant
that the Nagô-Bâ woman Edum bought the yams from Sabina for the leader
Aruna. These yams may have been intended for use in a ritual in celebration
o f the prospective uprising.
W ith the denunciation by the tw o African-born freedwomen, the police
immediately began to investigate and increased patrols in the city. Around
one o’clock in the morning on the 25th, a group o f patrols searched a two-
storied house located on the Ladeira da Praça. The first floor was occupied
by the pardo tailor Domingos Martin de Sâ; his parda companion Joaquina
da Santa Anna; their infant child; and a Nagô slave, Ignâcio, whom Dom in
gos owned. Domingos rented the basement to tw o African-born freedmen,
Manoel the caulker, who was Nagô, and Aprigio the bread peddler and chair
carrier, who was Nagô-Oyo. W ith the information from a neighbor that a
number o f negros had gathered in this basement, the troops broke into the
place, from which some sixty rebels came out shouting and shooting. They
easily overcame the small group o f police in front o f the house and pro
ceeded to the municipal jail. Here they hoped to rescue the African-born
Muslim slave Pacifico, who had been held prisoner because o f his owner’s
debt to the Carmelites. A fter failing to invade the jail, they attacked the
guards o f the governor’s palace and injured some o f them. In this, they were
joined by another group o f rebels. T he rebels proceeded southward, and
more people joined them on their way to Rua da Vitôria. A fter an unsuc
cessful attack on the fortress o f Sào Pedro, they headed back to Vitôria,
where they were united with another group who had been waiting at the
convent o f Mercês. They returned downtown, appeared on Rua da Ajuda,
and proceeded to Largo do Palâcio, where they failed once again to breach

the jail. The final battle between the rebels and the Municipal Cavalry took
place in Agua de Meninos, which was one and a half miles away from the
Bonfim Church. After about fifteen minutes o f fighting, the rebels were de
feated. Unaware that the uprising was already over, between six and seven
o’clock in the m orning o f the 23th, six armed slaves o f Joao Francisco Ratis
set fire to their owner’s house and attempted to reach Agua de Meninos, but
all were killed.10
It is difficult to ascertain exactly how many people participated in this
revolt. Howard Prince estimates the number at between 400 and 500.11 A f
ter an immediate and rigorous search and investigation o f the houses o f
African-born slaves and ex-slaves, the police arrested a total o f 326 individu
als. O f these, 284 were brought to trial.12 These participants in the revolt
seemingly included many slaves owned by British and other foreign mer
chants who lived in the wealthiest residential district o f Vitdria parish. O f
the 160 slaves who were incriminated, 50 were owned by foreigners. The
owners o f 45 slaves were British. The police were particularly suspicious
about a potential British anti-slavery agenda behind the revolt, but there
was no basis for their suspicion.13 British merchants, who were Protestants,
were presumably less concerned with the commitment o f their charges to
the Catholic faith, and on many occasions British owners, in fact, did claim
“ British privilege” for runaways.14 The local white population o f Salvador
resented the British privilege and charged the British with “ inciting their
own slaves to insurrection and preparing them to emulate the horrors o f
Saint Domingue.” 15
On the one hand, there are remarkable similarities between the Maid re
volt and the earlier slave uprisings (chapter 3). First, very few o f the Brazilian-
born population o f African descent participated. Second, the Maid revolt
was a predominantly male phenomenon; the proportion o f women who
participated in the revolt was very small (less than 10 percent). In contrast
with their male counterparts, African-born slave women o f diverse ethnic
backgrounds, m any o f whom worked together as market women on the
street, associated with one another and therefore developed a distinctive
gender identity beyond ethnicity (see chapter 2).16 The prominent absence
o f women from the revolt may also be attributable to the prevalent matri-
local family among the urban African-born population in the early nine
teenth century. We should remember that the denunciation o f the Maid
revolt in 1835 to the police was by tw o Nagd ex-slave women who were co
mothers, both o f whose common-law spouses o f African birth participated
in the revolt. Third, as in the slave uprisings in the 1820s and 1830, Nagds

predominated (68.1 percent o f all defendants).17 The court records o f the
revolt show that m any non-Nagô individuals o f African birth defended
themselves by referring to their respective “nations” and use o f different
languages. For instance, the Hausa freedman Narcizo Pinheiro, whose occu
pation was peddler o f cloth on the street, was arrested w ith tw o other Hausa
freedmen who shared the same house w ith him on Rua da Oraçâo. In court,
Narcizo declared that since he did not comprehend the Nagô language he
could not have known that the Nagôs had been planning to revolt, nor had
he him self been one o f the rebels. He continued: “ (E]very day at the time
o f Ave Maria I was at home, and did not go out after that. Some days in the
morning I saw various negros enter the house o f Belchior and Gaspar da
Silva, but I did not know why. I just ignored them.” 18
On the other hand, the Malê revolt was different from the uprisings o f
the African-born population in the earlier period in the following three as
pects. First, the revolt was not really a slave uprising; ex-slaves were promi
nent and comprised 41.4 percent o f the defendants. The Malê revolt mani
fests the creation o f a collective identity, beyond distinction o f legal status,
as people o f African birth.19 Second, the Malê revolt involved African-born
ex-slaves o f diverse “nations ” African-born slaves and ex-slaves, who had
already spent many years in Brazil and had gained a relatively good sense o f
Luso-Brazilian culture, joined forces to rebel against the larger society. The
majority o f slave participants (81.8 percent) were identified as Nagô, which
is a representative composite o f the enslaved population o f African birth, as
had been the case o f the slave uprisings from 1807 to 1830, but it suggests
that, as in the preceding cases, sharing the same ethnic identity continued
to be a determinant factor for the enslaved African-born population in tak
ing collective actions against power holders. By contrast, ex-slaves were
composed o f diverse West African “nations” : Nagô (48.6 percent); Hausa
(19.5 percent); Tapa (6.4 percent); and Gêge (5.5 percent).20 This supports
the point made in chapter 4 that African-born ex-slaves had created a col
lective ex-slave identity beyond diverse ethnic identities, but we should take
special note that despite their regained freedom those African-born ex
slaves joined African-born slaves for this uprising in 1835. Third, Islam
played important roles in this specific uprising. This was apparent from the
rebels* desperate attempts to rescue the highly respected African-born Mus
lim slave Pacifico (Lucian by his African name or Biali by his Arabic name)
from jail. His presence itself seems to have been very important among the
population o f African birth. Despite the fact that Pacifico him self had not
been involved in the uprising, he was later sentenced to six hundred lashes.21

In the Malê revolt, the rebels wore Muslim-style clothes with protective
good-Iuck charms and papers.22 O ther Muslim teachers who were found
guilty during the court hearings on this revolt included Aluna, the Nagô
slave who was a water seller and whose owner had recently gone to live on
a plantation in Santo Amaro; the Nagô freedman Victoriano (or Sule), an
itinerant vendor o f cloth; and the Tapa slave Luis.23 During the intensive
investigation that followed, the police seized papers and books written in
Arabic, as well as ritual garments, banners, and rings. They also found evi
dence o f sheep having been sacrificed in the houses o f many rebels, includ
ing in the basement sublet to Manoel and Aprigio. “Father” Manoel was also
a Muslim preacher.24
The Malê revolt in Salvador, however, should not be interpreted as an Is
lamic holy war (jihad) against Christianity, as the French sociologist Roger
Bastide has maintained in his The African Religions in Brazil, and most re
cently Paul E. Lovejoy has suggested in his diasporic interpretation o f New
World slavery.25 Nor was it a reflection o f any political or religious situation
in West Africa, nor a "nativistic” movement with religious millenarian over
tones, as scholars including Nina Rodrigues and Howard Prince have sug
gested.26Then, what were the roles o f Islam in the Malê revolt? M ay we in
terpret Islam not as a religion but as an ethnic identity that was created in
New World slavery and that bound the rebels o f West African birth together,
as Joâo José Keis has suggested?27
Some o f the papers in Arabic seized by the police had been written by
highly literate Muslims, probably educated in their homelands o f Africa,
and contained instructions to the insurgents coming from Vitôria parish to
“seize the country, kill all the whites, and go to a meeting place where they
would be joined by others from the Recôncavo,” but many participants had
only copies o f single verses from the Koran, perhaps written by barely liter
ate people who became adherents o f Islam after being transported to Bra
zil.28 Probably, for some o f those who were not even necessarily believers in
the religion o f Islam, there was the belief that by carrying the sacred pas
sages from the Koran in Arabic they would be protected from the bullets o f
police and soldiers. According to the anthropologist Mariane Ferme, who
studies gender among the present-day Mende o f West Africa, many West
Africans who are not Muslims by religion, nor literate in Arabic, believe
documents in Arabic to be imbued with strong sacred powers.29T he leaders
planned the revolt and committed their plans to paper in Arabic, whereas
many o f the participants, who were Nagôs, communicated verbally with
one another in Nagô. That this would be their usual practice was clearly

illustrated by the story o f the denunciation by the two African-born ex
slave women. This raises the question o f why the insurgents chose to use
the Arabic language, which had been closely identified w ith the religion o f
Islam, in their written planning o f the revolt. Was the motivation solely that
Arabic could be useful as a secret code for communication among those
African-born men?
Jack G oody’s theoretical monograph on the Mate revolt, albeit largely
based on secondary sources, discusses the significance o f the written religion
o f Islam in this particular revolt. First, the fact that writing was employed
in the uprising to make secret arrangements by means o f letters suggests
that the superior planning was related to literacy. Sharing the common w rit
ten language o f Arabic enabled a careful planning o f a revolt that possibly
recruited several hundred people. Secondly, the words from the Koran were
not simply a matter o f magical power or material protection but o f religious
faith. Thirdly, the written religion o f Islam seems to have functioned as an
ideological foundation for the rebels. The notion o f “death to the whites,”
which appeared not only in the Mate revolt but also in some o f the earlier
uprisings, was also characteristic o f many recent uprisings in West Africa,
where the word nasala for “white” is derived from the Arabic form o f Naza-
rene, Christian. Therefore, according to Goody, ethnic and even racial
definitions were dominated by religious overtones.30Thus the Mate revolt
took on anti-Christian forms and characters. Furthermore, Goody, citing
Eugene D. Genovese’s seminal work entitled From Rebellion to Revolution
(i 979)»31 discusses the importance o f literacy in major New World slave up
risings during the Age o f Revolution in general and maintains that the re
sults o f Islam’s com m itm ent to literacy were particularly marked in the
Mate revolt.32 In contrast to an oral religion, which specifies a population
boundary, a written religion is defined in relation to a text to which refer
ence is made. Consequently, the written religion is by definition capable o f
crossing ethnic boundaries and o f recruiting adherents, as in the case o f the
Mate revolt, on a non-ethnic basis and o f overcoming the ethnic divisions
among the African-born population in Salvador.33Thus, the written religion
o f Islam could have provided a solid religious foundation that enabled some
o f the African-born population in the New World to join together in the
name o f their own god, Allah, beyond ethnic divisions, to rebel against the
larger society whose full members believed in another god, namely Jesus
Christ; people o f African birth and their owners or ex-owners in Salvador
did not belong to the same religious community. Such religious separation
o f slaves from slave owners must have been the key factor in the relative

success o f the Malê revolt, because o f which, according to Genovese, “ Bahia
had come close to becoming another Haiti.”54
In the Malê revolt o f 1835, in contrast to the preceding slave uprisings
from 1807 to 1830, we do not find among the enslaved male African-born
population ethnic identities developed strongly enough to transcend geo
graphical urban/rural borders in Bahia. Whereas the majority o f the slave
rebels were Nagôs, the revolt also involved African-born freedmen o f di
verse West African “nations.” The participants in the revolt were in almost
equal numbers slaves and ex-slaves, many o f whom had already spent many
years in Brazil and had gained a relatively good sense o f Luso-Brazilian cul
ture. The aim o f the participants in the Malê revolt had been to rebel against
the larger society, which was composed o f the white elite and Brazilian-born
people o f African descent. To this end, they employed the Arabic language
as the means o f communication for this very special event and, more im
portantly, used the written religion o f Islam as a shared symbol o f their col
lective (if temporary) African-born male identity.
The Malê revolt had an enormous impact on the larger society's percep
tions o f the African-born population in general; they came to be viewed as
the most decisive obstacle to social order in slave society. In the first three
decades o f the nineteenth century, slave owners had been most apprehensive
o f African-born slaves, whom they classified by specific “nations,” such as
Hausa for the earlier periods and, later, Nagô, who were regarded as most
prone to rebellion. A fter the Malê revolt, in which both slaves and ex-slaves
o f diverse African “ nations” participated, the larger society began to rapidly
develop fear and suspicion o f all the African-born population. People o f A f
rican birth, both enslaved and freed, thus came to be viewed as the most
dangerous threats to public security.35
As a result, the Legislative Assembly o f the province o f Bahia, being con
fused and terrified, wrote to the General Legislative Assembly in Rio de
Janeiro suggesting establishment o f a Brazilian colony, to which African-
born ex-slaves could be repatriated in any port on the coast o f Africa. They
cited the case o f the British colony o f Sierra Leone, to which maroon slaves
had been sent from Jamaica. The imperial government o f Brazil sent a consul-
general, Lieutenant E. A. da Veiga, to Luanda to pursue this possibility, but
his residency was denied by the Portuguese authorities.36
The provincial assembly o f Bahia also issued a new law (Law no. 8) on
M ay 5,1835 to control the African-born population by severely restricting
their legal rights.37 African-bom ex-slaves became obliged to pay a head tax
o f io$ooo réis annually.38They were no longer allowed to acquire any form
The Convergence o f Identity, 1831-1880 to i

o f property.39O f course, many ignored these regulations and acquired prop
erty such as real estate and slaves, but registered it in the names o f their
Brazilian-born children, kinsfolk, or even slaves or ex-slaves they had raised
in the household as crias. For example, the African-born freedman Joaquim
de Sâo José, who purchased his second wife, the African-born slave named
Maria do Bonfim and her four children, registered ownership o f his wife
and children in the name o f his legitimate daughter Verlssima Maria José,
born to the late Rosa Barbosa, an African-born freedwoman who had been
his first wife.40 Furthermore, Article 18 o f the same law prohibited any real
estate owner, renter, sublessee, agent, or administrator from renting or leas
ing houses to African-born ex-slaves, as well as to slaves. Penalty for in
fringement was ioo$ooo réis.41
O n M ay 27,1835, within a month following the passage o f Law no. 8, the
provincial assembly passed another law to establish the office o f foreman in
charge o f wage-earners (ganhadores) in the city o f Salvador. The law defined
wage-earners as those engaged in the practice o f wage-earning, both by land
and sea, whether they were slaves or ex-slaves. Each foreman was paid his
salary by the wage-earners in his work gang. All wage-earners were obliged
to register their homes, the districts in which their homes were located, the
names o f owners in the case o f slaves, their qualification, and the type o f
service in which they were gainfully employed. This registration was to be
renewed monthly. Any individual who was employed in the service o f wage
earning without registration was liable to a fine o f io$ooo réis.*2 Although
this law does not specify sex o f wage-earners, apparently it refers exclusively
to men who worked in gangs as wage-earners, such as porters, stevedores,
and sailors. Furthermore, throughout most o f the early nineteenth century,
virtually all male wage-earners were o f African birth, whether enslaved or
freed. For instance, the 370 inventories o f slave owners I consulted do not
include a single slave wage-earner o f Brazilian birth for 1808-1849.43 Well
aware o f this fact, the municipal council established this law to strengthen
their legal control o f the African-born population.
The provincial law o f 1842 also obliged African-born ex-slaves to claim
annually to reside in Bahia. Each had to register his or her name, “ nation,”
marital status, age, and present address, along with the year o f manumission.
The richness o f such sources can be gauged by the following example: “On
September 12,1842, in this Secretariat o f Police, in the presence o f a judge
o f the court o f appeals and Antônio Simoes da Silva, police chief o f the
province, Antônio Xavier de Jesus, an African-born freedman, Nagô, thirty
years o f age, married to Felicidade Francisca Fernandes, an African-born

freedwoman, also Nag6, and who lived from his business dealings, declared
his residence to be in Cais das Ananas, and that he had been liberated in the
year o f 1835. He presented the document certifying that he had paid the head
tax as stipulated in Article 8 o f the provincial law for 1842-1843. He obtained
a residency permit for one year.”44
O n November 2,1850, the president o f the province o f Bahia issued an
other new decree applicable to the African-born population. It prohibited
boatmen o f African birth from plying at the quayside and within the harbor
o f Salvador, as this service was henceforth the exclusive prerogative o f the
Brazilian-born population. The British consul, Edward Porter, reported that
about 750 African-born ex-slaves were thus thrown out o f employment as
boatmen and that the decree— ostensibly issued for the purpose o f “encour
aging national labor”— in fact was intended to “gradually induce the white
population to engage in out-door labour.”45
W ith the establishment o f a series o f new laws restricting the legal status
and upward social m obility o f the African-born population, particularly
men, the Bahian authorities undertook the reduction o f a dangerous com
ponent o f the African-born ex-slave population by deporting them to A f
rica. Already in August and September o f the year o f the Maid revolt (1835),
rumors o f a possible insurrection were circulated in Salvador. Panicked and
distracted, the Bahian police soon imprisoned 300 to 400 African-born ex
slaves. O f these, 148 were expelled in a Brazilian schooner on November 12,
1835, to the port o f W hydah in West Africa. O n the 5th o f the same month,
there were further deportations to Africa.46 Interestingly, this forced depor
tation o f African-born ex-slaves to Africa stimulated voluntary departure
for their homelands in Africa by many other African-bom ex-slaves, who
were under severe daily supervision by a police force that continued to ha
rass them without cause. Individually, these African-born ex-slaves took
passage on Brazilian or Portuguese ships which sailed regularly between
Salvador and the Bight o f Benin. Am ong such emigrants were Antdnio da
Costa and Joao Monteiro, who had become wealthy slave traders. O n Janu
ary 25,1836, these tw o well-to-do freedmen chartered the English schooner
Nimrod so that they and 160 other African-bom ex-slaves could emigrate to
Africa. Under the British flag, these African-born passengers were well pro
tected from patrols, which might have otherwise suspected that the Nimrod
was engaged in the clandestine slave trade. The Nimrod dropped passengers
at the ports o f Arriba Athuna, Agoud, Aunim (Lagos), and Minas Pequenas.
Other African-born ex-slave men and women soon followed this success
ful example o f group emigration from Salvador.47 Passports were issued to

422 African-born in 1835 and to 329 in 1836. T he percentage o f African-born
among the total number o f people to whom passports were issued was 33
percent and 59 percent, respectively.48
The Bahian authorities welcomed and encouraged African-born ex-
slaves’ repatriation movements to Africa as much as possible. Law no. 420
o f June 7,1851 exempted from the aforementioned head tax o f io$ooo réis
those African-born ex-slaves who volunteered to leave Brazil forever w ithin
three months o f the enactment o f the law, on the condition that they would
never return.49 A t the same time the elite continued to impose more and
more econom ic pressure on the African-born ex-slave population. That
same year o f 1851 witnessed the establishment o f new annual taxes on vari
ous licenses issued to African-born ex-slaves, or “free Africans”: 20$000 réis
to buy and sell merchandise; 6$ooo réis to carry chairs (cadeiras); 308000
réis on sailing vessels known as saveiros; and 2o$ooo réis on boats used for
disembarkation o f people and cargo.50 No doubt, these legal regulations at
mid-century that discriminated against the African-born population be
cause o f their foreign status motivated people o f African birth living in Sal
vador to consider the option o f emigrating to Africa.
This large-scale emigration o f African-born ex-slaves to Africa seems to
have been largely limited to Salvador and Bahia.51 One o f the few cases out
side Salvador is reported by the Americans the Reverend James C. Fletcher
and the Reverend Daniel R Kidder in the city o f Rio de Janeiro. In 1851, sixty
Mina ex-slaves paid four thousand dollars as passage money to leave for
Badagry on the Bight o f Benin, and they arrived there safely. The following
year, a deputation o f eight or ten M ina ex-slaves waited for a delegation
from the English Society o f Friends. They had money for their passages, but
wanted to be assured that the coast was free o f slavers.52 Probably the large-
scale migration o f African-born ex-slaves and their families from Salvador
to Africa was partly attributable to Bahia’s depressed agro-export economy.
This contrasted with the coffee-boom ing Southeast (Rio de Janeiro and Sâo
Paulo), where even those o f African birth could find better economic op
portunities. In 1865, the French traveler Joseph Arthur de Gobineau, the
Count o f Gobineau, while observing the return o f Mina ex-slaves to Africa,
noted that some ex-slaves left Salvador for Rio de Janeiro, where oftentimes
they would make a small fortune.53
By mid-century, African-born emigrants from Brazil, mainly from Salva
dor, together with Luso-African slave traders, their former servants, and
slave ship captains, had settled in the major coastal towns o f the Bight o f
Benin such as Agoué, W hydah, Porto Novo, and Lagos.54 The British ex

military officer John Duncan, visiting W hydah in 1845, observed that the
residents o f this coastal town included many who had obtained freedom by
purchase, and others who had arrived as servants or slaves o f slave dealers
who had come from Brazil. The owners in their wills had manumitted these
slaves at the owners' deaths.55 Duncan also observed that in the countryside
ten or twelve miles distant from W hydah, many places were well cultivated
by “people returned from the Brazils.” These people had originated from
Foolah and Eya, had been enslaved, and had been transported at the age o f
twenty to twenty-four years to the port o f Badagry, from where they had
been shipped to Brazil.56
These coastal towns accepted “ returnees” not only from Brazil but also
from other parts o f the world. African-born ex-slaves who emigrated from
Cuba were later to join “ Brazilians.” 57 They were also to be joined by A fri
cans (mostly Yorubas) who had been captured on slave ships and emanci
pated in Sierra Leone by the action o f the Royal Navy, but who returned to
the Bight o f Benin and chose to settle in these towns. These people who
moved from Sierra Leone were com m only called Saros.58 African-born ex
slaves who emigrated from Salvador came into contact with the ex-slave re
turnees from Cuba and w ith the Saros, as well as with the indigenous popu
lation o f West Africa, and they created their own ethnic identity, which
separated them from others.
The Developm ent o f C ollective A frican-Born

Identity, ca. 1851-1880
The African-born population in Salvador quickly developed their

new ethnic identity as African-born after the mid-nineteenth century. The
year 1851 marked the official termination o f the transatlantic slave trade, and
after that date notarial records no longer indicate specific African “nations”
for the African-born population, as had hitherto been the case. In manu
mission letters, the percentage o f African-born ex-slaves who were identi
fied by “ nations” declined from 83 percent for 1808-1842 to less than half
(48.8 percent) for 1851-1884.59 This change is also apparent not only in le
gal registers o f slaves in SI parish and inventories o f slave owners, but also
in wills o f ex-slaves and parish records o f baptism and marriage.60Specifi
cation o f African “ nations” probably mattered less after the mid-nineteenth
century than it had previously, either for official use in records or for the
African-born population themselves; it had already been two decades since
the port o f Salvador stopped importing new arrivals directly, and the African-

born ex-slave population had gone through a substantial amount o f accul
turation. As we have discussed, their newly converged “African-bom ” iden
tity, at least among men, played an important role in their choice o f partici
pation in the Malê revolt.
Legal freedom was a key factor in African-born ex-slaves’ choice o f holy
matrimony. One example is the African-born ex-slave Gonçalo de Sousa,
who purchased his freedom from his owner, the priest Luis de Sousa, for the
price o f i2o$ooo réis. By June 25,1842, when Gonçalo made his w ill, he had
married Ana Maria do Sacramento, by whom he had no children. However,
as a bachelor, Gonçalo had fathered three natural children. The first two,
Gonçalo de Sousa and Maria de Sousa, were by a cabra woman named Maria
das Neves, whose legal status was free-born or ex-slave. T he mother o f
the third and last, Martiniana, was an African-born woman (prêta) named
Mariana, who was probably Gonçalo’s own slave. Martiniana was baptized
as an ex-slave (forra)y and her father registered her manumission letter at
the notarial office in 1834. Before his death in 1845, Gonçalo named as ex
ecutors o f his w ill, in order o f preference, first his wife’s son Francisco José
Moreira, secondly his w ife, Ana Maria, and thirdly his son, Gonçalo de
Sousa. Whereas he named Ana Maria as his “universal heiress,” he appointed
her son, his stepson Francisco, as guardian o f his second daughter, M artini
ana, who was in her early teens.61
In the case o f the union between an ex-slave and a slave, they usually
waited to contract marriage until the latter attained the status o f an ex-slave,
so that their future children would be born free. This was usually the case
o f a freedman and a slave woman, and the former often contributed money
for the latter’s self-purchase. The wills o f ex-slaves include six such cases.62
The manumission letters do not describe the relationship between the slave
woman and the non-kinsman who paid for her freedom.63 Some ex-slave
husbands liberated their wives’ slave children fathered by other men be
fore their marriage. Such was the case o f Sebastiâo Alves da Rocha, an ex
slave originally from Angola. Sebastiâo purchased the freedom o f his fian
cée, Maria Pedro, who was a slave o f José Joaquim Pires, also from Angola,
in order to m arry her. Each had a child by other partners before marriage.
After they married, Sebastiâo also purchased the freedom o f Maria’s daugh
ter, Ana, from the same owner. But neither Sebastiâo nor Maria had enough
money to pay the price for Ana, namely 4008000 réis. Sebastiâo borrowed
300$000 réis from Manoel Fernandes Suenpira, and a legal document to this
effect was notarized. A fter the death o f his wife Maria, Sebastiâo lived with
his own son, o f the same name, a bricklayer in Santana parish. Sebastiâo the

son had been bom o f a prêta, Laurença, then a slave, and had been baptized
by Sebastiâo the father as his son. Whereas Sebastiâo the elder instituted his
son as his heir, he not only made it clear that Ana had no right to inherit
any o f his property, but also insisted that she still owed him 303$200 réis,
namely the total amount o f the loan o f 300$000 réis, interest o f 200 réist
and notarial fee o f 3$ooo réis, since Ana had never returned any amount o f
money to him.64
Since a child born o f a slave woman was automatically the property o f
the mother’s owner, mother and child frequently shared the same surname
as ex-slaves o f the same owner. Despite the fact that they never shared the
same family names with their children born to slave women, many fathers
recognized the consanguinity o f their natural children and left money in
their wills for such children to purchase their freedom i f they were still en
slaved, and legitimized them as their heirs in those cases where they were
already legally free. Recognition o f paternity was most important to patri
lineal inheritance.
There are tw o interesting wills made by freedmen in which the testator
officially rejected his paternity o f a child who was recognized by others as
his natural offspring. The first case is Bento Dias Coelho, a bachelor ex-slave
from Costa da Mina. In his will registered on May 2,1816, Bento left i6$ooo
réis to the crioulo Bernardino, since Bernardino claimed that he was the son
o f Bento. However, Bento insisted that he did not recognize Bernardino as
his son, but that he left the money to him because o f his “conscience”65 As
we shall discuss later, until 1847 Brazilian law offered ample means in a situa
tion such as that o f Bernardino to prove patrilineal succession. Bernardino
or his m other’s fam ily could easily have established Bento's paternity in
court. Perhaps it was recognition o f such popular awareness o f the facts o f
the case, and not his conscience, which led Bento to leave the money to
Bernardino.
The second case is the African-born freedman Ronvaldo de Cerqueira.
Ronvaldo refused to recognize Clemente de Cerqueira Pinto as his son, al
though Clemente was born o f his first wife, Maria Rosa de Conceiçâo, after
their marriage. In his w ill o f August 16,1873, Ronvaldo confidently asserted
his opinion that Maria had been already pregnant by another man at the
time o f their marriage. A fter Maria’s death, Ronvaldo remarried, choosing
Felicidade Maria da Conceiçâo as his second wife. He named his second wife
as the “universal heiress” o f the property that he had bought with her money
after their marriage.66Since this occurred after the enactment o f the succes
sion law o f 1847, Clemente was denied the right o f patrilineal inheritance.

In the case o f comm on-law conjugal unions between ex-slaves and/or
free-born people o f African descent, natural children always took the fa
ther’s surname, as would legitimate children. Some unmarried parents surely
made their natural children share their fathers’ surnames, by which pater
nity could be widely recognized and which would become useful in justify
ing any subsequent claim to patrilineal inheritance. This custom o f taking
natural fathers’ surnames is exemplified by the case o f Narcesa Francisca de
Oliveira, a crioula ex-slave woman. She had three legitimate children by
Anselmo Lopes: Theodoro, Joáo Feliciano, and Francisca Maria. A fter her
husband’s death, she bore four natural sons: José Gonsalves Ramos, Gonsalo
Ramos, Dom ingos Gonsalves Ramos, and Vicente Ferreira Ramos. They
shared their common surname, Ramos, with their natural father, Narcesa’s
lover or common-law spouse.67
Fathers often shared their first names with sons, whereas mothers and
daughters rarely shared the same first name. Naming a son after his father is
a common practice in any patrilineal society, but in the context o f nineteenth-
century Salvador, where common-law marriages among slaves did not al
ways imply co-residence o f partners and where children were usually raised
by mothers, this naming practice helped children born as slaves to maintain
father-son relationships in those cases in which the father and mother ended
their relationship. Despite strong matrilocality among the population o f A f
rican descent, their intergenerational kinship relationships extended bilat
erally, while fathers’ recognition o f consanguinity was evidenced in their
naming practices.
Both grandmothers and grandfathers recognized their ties to their slave
grandchildren, paid cash for their purchase o f freedom, left property or
money to them, and even named them as their heirs. For instance, the African-
born ex-slave Valentino Paschoal left half o f his property to his wife, Maria
Espirito Santo Pires, also an African-born ex-slave, and the rest to his still
enslaved granddaughter, Belimira, for her freedom. Belimira was one o f
the four children o f his natural daughter, whom Valentino had had by a
slave woman.68Although parent-child relationships seem to have extended
bilaterally, it is likely that matrilineal ties survived in the long run with
much greater frequency than patrilineal ones; many more grandmothers
than grandfathers referred to their grandchildren in their wills. Maria de
Araujo Ribeiro, an ex-slave woman from Costa da Mina who died on O c
tober 3,1815, named her granddaughter, Anna Florencia de Almeida, as the
heiress o f two-thirds o f her property, and left the rest to her little great-
granddaughter, named Anna, daughter o f Anna Florencia and her husband,

Sergeant-Major Luis Gonsaga de Barros. Maria appointed her son-in-law as
executor o f her will.69
In nineteenth-century Salvador, ex-slaves* accounts o f how they viewed
holy matrimony and the institution o f marriage have not been documented.
The few exceptions are some ex-slaves* brief references to their own mar
riages in their wills, but only in terms o f inheritance. The Gêge ex-slave
woman Joana Maria da Conceiçâo had been separated from her husband,
the African-bom freedman Antônio da Costa, resident o f Ladeira de Santa
Tereza. W hen she made her will on July 28,1842, she clarified that she and
her spouse each had total control over their respective property. Therefore,
she did not leave any o f her property to her husband.70This is the only will
I have found in which a married couple were separated from each other.71
Another example is the African-born ex-slave José Alves Matheos, who was
married to Justina Maria do Nascimento, whose legal status, place o f birth,
and color were not stated. In his w ill, dated December 16,1867, he left his
house to his wife to live in with her sister Martinha, her son Eugério, her
family, and her goddaughter (afilhada) M artinha, on the condition that
his wife lived “as a widow and in a chaste manner.”72 Thus, by controlling
her rights o f inheritance, knowing that she would need the house to take
care o f her own family, José Alves Matheos tried to prevent his wife from
remarrying or even having a relationship w ith another man.
Most common-law conjugal unions among ex-slaves were monogamous
and lasted for many years. A fter the mid-nineteenth century, many wills o f
African-bom ex-slaves mentioned the names o f their common-law spouses
(companheiros or companheiras)> also African-born ex-slaves, who often
served as executors o f their wills. Through their longtime monogamous
unions, some common-law couples had several natural children between
them.73
Polygamous unions were exceptional. One example is afforded by the
African-bom freedman Antônio de Almeida, who fathered natural children
by different women. A bachelor, he fathered three children by the African-
born freedwoman Esperança Bages: Cosm a, Martiano, and Paulina. He also
had one daughter, named Maria Laurença, by Justina Baptista do Reis, an
African-born freedwoman. W hen he registered his w ill in 1869, he was living
with the African-born freedwoman Maria Antônia Guimarâes.74
After mid-century, African-born ex-slaves started to contract marriages
very late in life, sometimes even on the deathbed o f one o f the spouses. The
763 listings in the marriage register o f Conceiçâo da Praia parish over forty-
six years (1843-1888) include eighteen male and fif teen female African-born

ex-slaves. Fifteen marriages were registered between African-born ex-slave
spouses, but often with partners o f both sexes aged anywhere from their
forties through their eighties, and only after many years o f common-law
conjugal union.75 The wills o f ex-slaves include four cases o f deathbed mar
riages. One example is the old African-born ex-slave Rafaina Rosa Friandes,
who married Joao de Oliveira Sampaio, also an African-bom ex-slave, on
June 25,1863, since she had no expectation o f surviving a serious illness.
They had lived in an "illicit” union for many years. She died, as a married
woman, on August 13,1863. A fter her death, her husband married another
African-bom ex-slave woman, Marcolina Porfina Vianna, and passed away
on M ay 19 ,1867.76This raises the question o f why such deathbed marriages
became more prevalent after the 1850s and whether such marriages were
solely among African-born ex-slaves.
Historian Linda Lewin’s study o f traditional Brazilian succession law
reveals that such a change in marriage patterns did not take place exclu
sively among ex-slaves who made wills leaving bequests, but was widespread
among all property owners who had natural offspring. W ith Law no. 463 o f
September 2,1847, Brazilian legislators rewrote the rules o f patrilineal suc
cession for natural offspring. Paternal “recognition” conferred succession
rights on natural offspring, and equal rights continued to be presumed to
exist for all appropriately recognized natural children until 1917, but the legal
definition o f paternal recognition did change over time. Until 1847, civil law
construed proper recognition o f paternity to include not only notarized
declarations and those inserted in wills but also common knowledge. The
law o f 1847 drastically restricted the procedural opportunities for fathers
to declare paternity o f natural offspring. Furthermore, it ended a natural
offspring's opportunity to establish kinship in court for the purpose o f pa
ternal succession. O n the other hand, by the same law, natural offspring
autom atically gained succession rights by virtue o f their parents1 subse
quent marriage.77 This explains why, in the later part o f the nineteenth cen
tury, property-owning ex-slaves contracted marriages very late in their lives.
In the late nineteenth century, when specific “nations” o f African-bom
were no longer mentioned in baptismal records, mothers and their coparents
were often African-born.78 In the manumission letters, one male and six
female Brazilian-born slaves (1808-1842) and tw o male and seven female
Brazilian-born slaves (1851-1884) were manumitted w ith payments from
their godparents.79 Unfortunately, the manumission letters usually m en
tioned only the names o f godparents, and only one o f the sixteen godparents

for Brazilian-born slaves at baptism was identified as a Nagô freedman.80Yet
these data at least suggest that slave mothers chose as godparents those who
could later afford to pay for a slave child’s freedom.
In some wills o f ex-slaves, their godparents (padrinhos and madrinhas)
or coparents ( compadres and comadres) were named as executors, whereas
testators left money or property to their godchildren.81 By the terms afil-
hados and afilhadas, they were usually referring to godchildren, but some
times they included younger people to whom they were emotionally tied as
quasi-kinsfolk, but without the relationship, such as their coparents' chil
dren or ex-owners’ children, being recognized by formal Catholic rituals.
For instance, Pedro Pires Monçâo, a crioulo ex-slave, named as executors
o f his w ill his godmother, Maria Francisca Rodrigues Seixas, resident o f
Q uinze Mistérios and, secondly, his crioula w ife, Bernardina Rodrigues
Seixas (a daughter o f the late Gêge woman Maria Thereza) together with
his godfather, José Pereira da Silva. Pedro was determined to grant a letter
o f liberty to the prêta Thereza, one o f his slaves and his afilhada, a year after
his death.82 Another example is the w ill o f Matheos da Silva Guimarâes»
originally from Costa da M ina, who had married the late Joaquina da Silva
Vasconcellos, also from Costa da Mina, but had no children by her. As ex
ecutors o f his will he named in descending order o f preference his compad-
resy Joaquim de Santa Ana Lima and Joào Luis Ferreira, and his comadre,
Bernardina Joaquina das Neves. His compadre Joaquim de Santa Ana Lima
was also the godfather o f his Gêge slave woman Antônia, who was to be
freed because o f her “good services” upon the death o f her owner and was
to be his heiress.83
Unfortunately, the legal status or color o f a godparent, coparent, or god
child was not often mentioned in wills o f ex-slaves. Researching o f these
wills in chronological sequence does suggest an interesting evolution in the
changing patterns o f fictive kinship networking among people o f African
descent. Just after mid-century, African-born ex-slaves, who were often co-
parents, began to establish relationships with one another as debtors and
creditors. This practice, clearly revealed through their wills, suggests that
coparenthood came to function as a solid base for establishing economic
relationships among African-born ex-slaves, who began to develop a com
mon collective identity after the termination o f the Atlantic slave trade.84
Occasionally, a person o f financial means had several creditors and debt
ors concurrently, some o f whom were his or her coparents. It is not clear
whether fictive kinship networks as coparents were developed on the basis

o f a prior debt-credit relationship or vice versa, but these tw o kinds o f re
lationships overlapped each other to a great degree and did strengthen in
dividual ties among African-born ex-slaves.
Furthermore, in the late nineteenth century, African-born ex-slaves started
to establish a pattern o f naming as the executors o f their wills those who
were identified as African-born ex-slaves, including their kinsfolk and fic
tive kin sfolk Some executors o f wills were also heirs. Am ong the 325 wills
o f ex-slaves (159 for 1808-1849 and 166 for 1850-1888), 15 fall into this cate
gory, and all were registered after mid-century.85One example was an African-
born freedwoman, Maria Roza Pereira, who registered her w ill on the 18th
o f June, 1867, and died less than three months later. She had been baptized
in Santo Antônio parish and was an ex-slave o f Severo José Pereira, who had
liberated her unconditionally for her “good services.” Maria Rosa was the
widow o f Francisco de Santana. She had neither legitimate nor natural ch il
dren. As executors o f her w ill she named first Henriqueta Joaquina do Bom-
fin, the widow o f Benedicto Fernandes Galiga; secondly, Lucrecia Luisina;
and thirdly, her nephew, Manoel Luisino. Each was African-born and an ex
slave. Maria Rosa owned a house at Rua do Genipapeiro. She had loaned
500$000 réis to her nephew Manoel for his self-purchase. In her w ill, she
waived repayment o f 200$000 réisy but requested Manoel to return the re
mainder to the executor o f her w ill, since these funds were to cover the
cost o f her bequests. She left to Oliva, a daughter o f Cora, an African-
born woman, 50$000 réis; to her goddaughter Joana, an African-bom slave
owned by a baker on Rua de Säo M iguel, 50$000 réis; and to her niece
Damiana, daughter o f Rosa, 25$ooo réis. Maria Rosa also left 50$000 réis to
Lucrecia Geisina, an African-born freedwoman, who had taken care o f her
during her illness. Finally, she named as her heiress Henriqueta, an African-
born freedwoman and the first executor o f her w ill.86
Another example is the A frican-bom barber-surgeon José Domingos
dos Santos, whose “ nation” was Nagô. José was an ex-slave o f Lourenço
Domingos dos Santos, also a barber-surgeon. His ex-owner had been a resi
dent o f Rua do Cais Dourado (Pilar parish), where José still resided. In his
will o f April 14,1859, José named as executors o f his will three African-bom
ex-slaves, all o f them barber-surgeons and barbershop owners: Simäo Bento
de Cabalho, resident o f Rua do Cais Dourado o f Pilar parish; Manoel A n
tônio Guimarâes, resident o f Rua de Dom José o f Sé parish; and Antônio
Marcelino d ’Almeida, resident o f Rua de Dom José o f Sé parish.87 This is
one o f those rare cases in which the occupations o f the testator and o f all
the executors were indicated in the will.

The evidence overwhelm ingly supports the hypothesis that in the late
nineteenth century, the African-born ex-slave population, which included
longtime common-law ex-slave spouses, started to develop actual and fictive
kinship networks among themselves, and that these came to function as a
base for eco n o m ic relation sh ip s.98
The Creation o f “ Brazilian” Identity Back “ Home”
After the mid-nineteenth century, African-born ex-slaves who re

turned to Africa started to create Brazilian identity in Lagos, West A f
rica. Throughout the nineteenth century, the port o f Lagos maintained the
strongest commercial relations with Salvador, particularly for the trade o f
slaves in exchange for Bahian tobacco. A fter mid-century, the slave trade
declined but was replaced by trading in palm oil (used to manufacture
candles and soup, and as a lubricant) and oil made from the kernel o f the
palm nut (a raw material for margarine). Salvador’s African-born popula
tion continued to im port great amounts o f cotton cloth, called pano da
costa, and kola nuts from Lagos, as well as diverse ritual objects.89 A fter the
establishment o f the British colony o f Lagos in 1850, many African-born
ex-slaves from Salvador began to settle in this colony, rather than leaving
for their own specific homelands, as had formerly been the case. In Lagos,
the Pax Britannica, instituted in 1851, protected their legally free status so
that they were in no danger o f re-enslavement.90 Lagos also gave these im
migrants good economic opportunities.
Lagos, as a common destination, was likely to be gradually identified with
their shared image o f “Africa” in the minds o f the African-born population
in Salvador. In the 1850s and 1860s, African-born ex-slaves often took great
pride in publicizing their emigration to Africa in the classified advertise
ments o f local newspapers, such as Diàrio da Bahia and Jomal da Bahia,
before embarkation in Salvador. Examples are “Pedro Antônio Felicimo de
Santana, an African-born ex-slave, his Brazilian-born son Eleodorio, and
Felicidade, an African-born freedwoman, will go back to the coast o f Africa
(Costa da Africa)”; “Salvador Ramos das Neves w ill go back to the coast o f
Africa, taking his family, Maria Luisa da Conceiçào, his daughter Teda Maria
das Mercez, their former slaves o f Brazilian birth (crias forros), Cam ila and
Guimarâes, and Dona Maria Bareto, Joâo Maria Porciana and Maria Izabel” ;
and “Tito and Antônia, African-born ex-slaves, will leave for the coast o f
Africa with their crioulo sons, Luiz and Acurcio.”91 The destination o f such
emigrants was never more specific than the “coast o f Africa (Costa da A f

rica),” perhaps because the specific name of the town to which they were
headed, such as Lagos, did not matter to them. To such emigrants, of para
mount importance was to M go back” to the symbolically reconstructed Africa
as an image far removed from reality. Ironically, because of such advertise
ments, some prospective emigrants who had not yet fully met the payments
for their self-purchase found that their owners refused to release them from
bondage, and thus they were not able to take passage for Africa.92
Emigration to Africa usually took place as a family unit, which included
Brazilian-born children. Emigration kept the immediate fam ily together
but sometimes separated kinsfolk on both sides o f the Atlantic Ocean. This
was the case o f the African-born freedwoman Luiza Francisca Gonçalves,
who had married Antônio Gonçalo, by whom she had six legitimate chil
dren (Joaquim, José, Raimundo, Manoel, Severiano, and Carolina). W hen
the time came for her to make her w ill on September 4,1864, her husband
and all o f their children were already dead, so Luisa named as her “common
heirs” the children o f her three sisters and her tw o Brazilian-born former
slaves. Her nephews and nieces included the African-born children o f Atipé,
named Arm ado and Margarida, who were already back on the coast o f A f
rica.93
Once they settled in the colony o f Lagos, these African-born immigrants
from Salvador, both Catholics and Muslims, separated themselves from the
indigenous population and immigrants from other regions, such as the Saro,
and contracted church marriages among themselves.
A male Muslim teacher, resident o f the British colony o f Lagos in what is
now Nigeria in West Africa, registered his w ill in the city o f Salvador with
the date o f March 9,1877. His name was Victoriano Maria, and he was a
native o f Iguinem on the coast o f Africa and the son o f the African woman
Omalia. Probably in the early nineteenth century, Victoriano Maria was im
ported to Salvador as a slave. This resulted in separation from his mother,
whom he knew only by her African name. Somehow he had managed to
keep in contact with her, so that he learned o f her death. In Salvador, he
fathered a natural son, Theodoro, who later fathered a boy named Manoel
Theodoro de Aleluia, thereby making Victoriano a grandfather. After gain
ing his manumission letter, Victoriano returned to Africa, leaving behind in
Salvador his son and grandson, with whom he continued to communicate
across the Atlantic. In Lagos, Victoriano married three wives under Islamic
law. His wives all bore Portuguese names: Constança Joaquina do Sacra
mento, Lourença Catarina Chaves, and Catarina Maria da Conceiçâo. These

three marriages in Lagos did not result in any children. Victoriano became
a man o f means in Lagos. In his w ill, he left all his possessions in Lagos,
including tw o houses, to his three wives in Lagos and 200$000 réis to his
grandson in Salvador.94 His w ill does not provide any more information on
his wives, but it is assumed that they were also African-born immigrants
from Salvador. Despite his baptism as a Roman Catholic at the tim e o f his
enslavement and a superficial forced conversion to Christianity, Victoriano
emphasized in his w ill that he “had been living and would die in the faith
o f the law o f Muhammed.” He was not the only African-born person in
Salvador whose religious faith was Islam. This makes all the more impressive
the fact that, among the 325 wills o f ex-slaves I consulted, no other person
publicly declared his or her Islamic faith in an official document.
Victoriano Maria and other “ Brazilians” who settled down in Lagos
preserved Portuguese language and Brazilian customs, built schools and
Catholic churches in the Luso-Brazilian style, and observed religious holi
days and festivals, such as that o f O ur Lady o f the G ood Ending (Nossa
Senhora do Bonfim), the counterpart o f O ur Lord o f the G ood Ending,
which was most popular in Salvador. They also built a Brazilian cemetery
so that they could be buried together in a common place, in contrast to the
local custom o f burials within the homes. In “ Brazilian” schools, instruction
was in Portuguese as their first language, and in English as their second. O n
the other hand, their African identity encouraged them to abandon Portu
guese family names, which they had adopted from their former owners
upon being freed in Brazil, and to take Yoruba family names.95
In December 1853, there were already some 130 families o f these im m i
grants in Lagos.96 The population o f Brazilians in Lagos rose to 4,000 in
1868, but this immigration movement o f African-born Brazilians gradually
declined after 1870.97 This coincides with the period during which there was
a decrease in the number o f references to specific places o f birth in manu
mission letters granted to ex-slaves in Salvador. A fter 1870, there was also a
rapid increase in the percentage o f ex-slaves whose identifications in the
manumission letters did not include places o f birth.98T he decrease in num
bers o f African-born ex-slaves and their families emigrating to Africa after
1870 was related to the decrease in the proportion o f African-born people in
the overall population o f African descent. This led to the transformation o f
self-identities among peoples o f African descent, most o f whom were by
now Brazilian-born. By 1881, the population o f Brazilians in Lagos decreased
to 2,723 among a total population o f 37,452, but emigration o f African-born

ex-slaves from Salvador to Lagos continued until the end o f the century. The
anthropologist Manuela Carneiro da Cunha estimates the total number o f
these emigrants at nearly 8,000."
A fter the abolition o f slavery in Brazil (May 13,1888), some Brazilian im
migrants in Lagos went back to Salvador or sent back their African-born
children.100 Thus, fam ily networks developed on both continents. Brazil
ians o f African descent on either side o f the Atlantic were in constant con
tact. In June 1972, historian J. Michael Turner interviewed the president and
some members o f the Lagos Brazilian Descendants Society at their monthly
meeting held at the Casa de Agua, home o f the president. They talked to
Turner about their memories o f parents and relatives who had returned to
West Africa.10'
Among people o f African descent in Salvador, “Africa,” both as a symbol
and as reality, has remained important until the present day and is con
stantly evoked in candomblé, carnival, and other cultural-religious activities
throughout the year. Even today on the streets o f Salvador a passerby will
hear people o f African descent talking about their kinsfolk in Africa. The
memories o f Africa have been constantly reinterpreted and revitalized in the
context o f their daily activities.
Conclusion
D uring the 1830s the African-born population began to create a

convergent African-born identity beyond diverse ethnic identities. The re
creation o f ethnic identity was manifested in the occurrence o f the Malê
revolt in 1835. Islam emerged as a common symbol shared by the partici
pants o f African birth, both enslaved and freed, and as such transcended
ethnic divisions; the written religion o f Islam was closely associated and
identified with literacy in Arabic, which enabled cross-ethnic communica
tion and the creation o f an African-bom identity.
A fter mid-century, when identifications for the population o f African
birth in notarial and legal records gradually converged from specific A fri
can "nations” to African-born, a significant change was observed in African-
born ex-slave families. African-born ex-slaves, who constituted a minority
among the population o f African descent, began to adopt Luso-Brazilian
patterns o f conjugal unions characterized by co-residence o f partners, yet
without contracting holy matrimony. Unlike the previous situation, where
slaves formed conjugal unions but such common-law couples did not live in
the same households, now African-born ex-slave partners coresided. As a

result, their Brazilian-born children were raised in a double-headed house
hold. They had their newborn natural children baptized with godparents
who were also African-born ex-slaves. This led to greater stability o f urban
families and stronger emotional bonds among the ex-slave population o f
African birth.
Yet, it should be emphasized that these changes in family and kinship
among African-born ex-slaves were not merely the result o f their legally free
status. These changes certainly involved a change in their self-identities,
which occurred when the African-born population, both slave and freed,
started to merge their identification by specific “ nations” into a common
identity as African-born. Property-owning African-born ex-slaves started
to name as executors o f their wills those who were identified also as Afri
can-born ex-slaves, including their actual and Active kinsfolk. African-born
ex-slaves, who were often coparents, also began to deal with one another as
debtors and creditors. These financial relations were frequent enough to be
regarded not merely as sporadic circumstances, but to form a pattern o f be
havior Coparenthood had come to function as a solid base for African-born
ex-slaves. Thus, the African-born population in this period became bound
together through kinship and fictive kinship, and also through the sharing
o f a broader and convergent ethnic identity as African-born.
The development o f family and kinship networks among African-born
ex-slaves coincided with their new “return” movement to Africa. The family
unit often provided the context for such decision making. A fter the Malfi
revolt o f 183s» the Bahian authorities continued to impose more legal re
strictions on the African-born population in general, which spurred many
African-born ex-slaves to immigrate to West African towns. Once the Brit
ish colony was established in Lagos, African-born common-law spouses o f
diverse ethnic origins, with their Brazilian-born children, started to return,
often in a family unit, to this common destination, where their legally free
status would be protected by the Pax Britannica. For such emigrants, Lagos
came to symbolize Africa. It was as African-born, not as Nagds or Gdges,
that they returned, and as such no longer did they feel the need to return to
their original homelands. In Lagos, African-born immigrants from Salvador
separated themselves from other inhabitants to develop their Brazilian iden
tity. It should be noted that their decisions at this stage were predominantly
family oriented. This contrasted with the individuality o f decision making
and participation by individual males in the preceding slave uprisings in
Bahia in the first half o f the nineteenth century. In Lagos, these immigrants
from Salvador geographically separated themselves from the local popula

tion and other inhabitants, including returnees from other parts o f the New
World, and identified themselves as Brazilians. Thus a Brazilian com m unity
came into being in Lagos. The Brazilians in Lagos contracted church mar
riages among themselves and preserved the Portuguese language and Luso-
Brazilian culture. O n the other hand, their new African-bom identity en
couraged them to abandon Portuguese family names and replace them with
Yoruba names.
After 1870, emigration o f African-born ex-slaves to Africa gradually de
clined. This was related to the quick decrease in numbers o f the African-
born population in Salvador, but also attributable to their own changing
identities. The rapid reduction in statements as to birthplace in manumis
sion letters after 1870 probably means that the birthplace no longer mattered
for individual self-purchase. During the last decade o f slavery, African-born
male ex-slaves outnumbered their female counterparts by the ratio o f 3:2,
whereas the percentage o f women among Brazilian-born ex-slaves reached
new heights because o f the high demand for young male workers.102Owners
were more w illing to manumit slave men o f African birth, who had worked
as field hands and transporters throughout their lives and were now ill,
handicapped, and aging, whereas owners could still squeeze labor as domes
tics from African-bom slave women around the same age.
A fter the mid-nineteenth century, “Africa” was no longer a trans-ethnic
label or category simply imposed by the larger society for the purpose o f
slave control; it became a common destination shared by African-bom ex
slaves o f diverse ethnic origins. After having spent many years in Salvador,
their differences in ethnicity and ethnic identity started to be overcome by
their commonness as African-born in contrast to Brazilian-born, especially
once the slave trade was banned in 1831. Individual memories o f their re
spective homelands in A frica had been reconstructed and changed over
time and were to be replaced by the singular image o f Africa as their com
mon homeland, which was largely an imaginary creation among the African-
bom population through their experiences o f New W orld slavery. Thus
African-bom ex-slaves created a new collective identity as African-bom in
Salvador and as Brazilians in Lagos. They formed a com m unity with the
common symbol o f Africa or African birth, which led them to seek “true”
belonging in Africa, where they were once again to be classified as “outsid
ers” by indigenous populations. It may be one o f life’s ironies that in a sense,
they remained outsiders despite their voluntary relocation after they gained
legal freedoms. This unchanged outsider-ness was the hallmark o f African
slavery in the New World.

In Part I (chapters 1 and 2) and Part II (chapters 3 and 4), we have so far
discussed the creation o f identity by the African-bom population in slav
ery and freedom: the creation o f new ethnic and gender identities; collec
tive representations o f such identities in urban slavery; the emergence o f
ex-slave identity among the African-born population; and the creation o f
African-born identity in Salvador and o f Brazilian identity in Lagos o f West
Africa after the termination o f the transatlantic slave trade. O ur next ques
tion is: In contrast to the African-born population, most o f whom were
born free in the Old World and all o f whom were brought to the New World
as slaves, how differently did the Brazilian-born population o f African de
scent create their identities in slavery and freedom? This will be the subject
o f Part III.

Part Three
To Be Brazilian-Borriy circa 1808-1888
Yet this vacuum is not new to her. It has an

edge; somewhere in the bottom lid is the
distance. She has seen it lurking in the eyes
o f all white people. So. The distance must be
for her, her blackness. All things in her are
flux and anticipation. But her blackness that
accounts for, that creates, the vacuum edged
with distance in white eyes.
— Toni Morrison, The Bluest Eye
6 The Creation of Disparate
Identity, 1808-1851
The Brazilian-born slave population in Salvador created their identities in a

quite different way than did their African-born counterparts. They had
never been legally free; they were born into slavery in Brazil. Naturally they
created a different notion o f freedom from that o f the African-born popu
lation. As Sidney W. M intz elucidates: “The slaves who were born into slav
ery may have known about ‘being free’ as their parents told them o f it. But
they also knew o f ‘freedom* as defined by the colonial societies in which they
lived.” 1 Brazilian-born slaves did not identify themselves with the African-
born population; the larger society always favored the former over the latter.
As a result, for instance, very few Brazilian-born slaves participated in the
frequent slave uprisings o f the first decades o f the nineteenth century (see
chapter 3).
This ch apter, by focu sin g on the period from 1808 to 1851, w hen B razilian -
born were outnumbered by African-born among the slave population in Sal
vador, examines historical processes by which the Brazilian-born popula
tion o f African descent created their disparate identities in accordance with
color distinctions in slavery and freedom. We shall also discuss how the free-
born crioulo population started to express a newly emergent collective racial
identity by establishing new lay sodalities o f their own.
B o m as Slaves in Brazil: K inship and Fictive K inship
A high illegitim acy rate among Brazilian-born slaves resulted from

the rare incidence o f church marriage among the urban slave population
(see chapter 3). Records o f births for Penha parish (1844) in Salvador show
that only 4 out o f 26 slave children were legitimate.2 Among 679 Brazilian-
born slaves (354 men and 325 women) baptized in Santo Antônio parish,
who were registered between 1808 and 1869, only 8 males and 10 females
were registered as legitimate.3
Even those Brazilian-born slave children whose parents were unmarried
benefited from the presence o f strong kinship ties, especially with their slave
mothers, unlike the case o f the African-born population. Since a child bom
to a slave woman was automatically the property o f the mother’s owner, re
gardless o f the legal status o f the father, a slave child was raised by his or
her mother in the household o f their owner without the constant presence
o f the father, who was usually under a different ownership. Strong mother-
child ties among the Brazilian-born slave population were fostered and
strengthened by the owners* practice o f selling, purchasing, and manumit
ting the slave mother and her child together.
Mother and child(ren) were especially often sold and purchased together
in the case o f a child still being in infancy or early childhood. Am ong 2,608
listings in the legal register o f slaves purchased and sold in Sé parish dur
ing the period 1838-1888,131 (5 percent) fall into the category o f mother-
child(ren) sales. By contrast, in only four documents were fathers and sons
sold and purchased together.4 To our surprise, only one whole slave family,
all o f whose members were Brazilian-born, was transferred without separa
tion. This family was composed o f parents and their tw o children: the 30-
year-old father (crioulo)t the 22-year-old mother (cabra); their 8-year-old
son (crioulo), and 2-year-old daughter (cabra),* Needless to say, this com
mon practice o f purchase and sale o f child and mother together did not
derive from any humanitarian consideration on the part o f owners about
not breaking up mother-child ties. Raising an infant slave with a very low
sale price was a costly business, especially in the absence o f the mother.
Poor nutrition and unsanitary conditions contributed to a high infant mor
tality rate. For these reasons, people were reluctant to sell or purchase infant
slaves. For the same reasons, mother and Brazilian-born child(ren) were oc
casionally manumitted together, often unconditionally, and in “apprecia
tion” by the owner o f the mother’s “good services.” Sixty-three out o f the
3,516 manumission letters for the years 1808-1884 describe mother-child
manumission in appreciation o f the mother’s good services.6 Two-thirds o f
these cases were unconditional and gratis manumissions, because o f the
mother’s good services. One exceptional case is the unconditional liberation
o f a whole Brazilian-born slave family: Gormano, pardo, the father; Maria
Magarida, the mother; and their children Rodrigo (crioulo)9Ferigrino, Fel-
sarda, and Maria (all cabras).7 O nly tw o slave children were officially re
corded as their owner’s illegitimate children. Both mulattos, born to the
same owner by his slave woman Marcelina, were manumitted.8We suspect

there were numerous cases in which their biological fathers, who owned
their mothers, manumitted slave children.
Their mothers manumitted Brazilian-born children with the payment
o f money with greater frequency than their fathers. Among 3,516 manumis
sion letters for the period 1808-1884» fifty-five cases o f Brazilian-born chil
dren were paid for by mothers (thirteen slaves, nineteen ex-slaves, and
twenty-three with no indication o f legal status). By contrast, only eleven
fathers (one ex-slave and ten without any identification o f legal distinction)
paid money for their children's freedom.9 Some o f the mothers themselves
were still enslaved when their children were manumitted. The mother and
child(ren) were usually owned by the same owner, so that the slave mother
was in a good position to negotiate her infant child's price for freedom and
schedule o f payments with an owner who was concerned about the cost o f
raising the child. Some Brazilian-born slave children were manumitted with
the assistance o f their mothers' spouses. An example is the case o f Pedro,
son o f the African-born ex-slave woman Luduvina Rebelo, resident o f Rua
da Saldanha. She had had this son by another man before her marriage to
Manuel Alves da Silva. There was no offspring o f her marriage to Manuel.
Luduvina, together with Manuel, got her son freed at age ten. A fter Pedro
obtained his letter o f liberty, Luduvina asked Manuel to teach his stepson
Pedro his skills as a carpenter, and both gave every assistance to Pedro until
he reached adulthood. In her w ill Luduvina insisted that her son had no
right to inherit any o f the couple's property.10 One possible explanation is
that her husband had made most o f their common property and had some
body else, probably his own child b y another woman, in mind to inherit it.
Brazilian-born slaves' intergene rational kinship relationships developed
bilaterally only with fathers' recognition o f consanguinity. This has been
already illustrated by the manumission o f child slaves paid by their fathers.11
Such was the case o f Bernardino da Sena, an African-born freedman origi
nally from Costa da Mina, who had been a slave o f the Franciscans for more
than twenty-five years, until he substituted another slave o f a value equal
to that stipulated for purchase o f his own freedom. He never married, but
as a bachelor fathered a son named José Maria Vilela by Luduvina Proza, an
African-born woman from Agomê, who was a slave o f Colonel Manoel José
Vilela de Carvalho. Bernardino paid 200$000 réis to the same owner to get
his slave son freed.12 One extraordinary case is that o f the Nagô freedman
Tomé de França, resident o f Pilar parish. He had four different companions,
perhaps at different times, all African-born (Esperança Pinto, Ignez María
The Creation o f Disparate Identity, 1808-1851 125

do Bomfim, Maria do Pilar, and Catharina), all o f whom were slaves at the
time o f the relationship and each o f whom bore Tomé a child: Maria da
Piedade (19 years o f age at manumission), Ventura da Franca (11 years o f age
at manumission), Juliana da Franca (7 years o f age at manumission), and
losé da Franca (less than 8 months at manumission). Tomé de Franca not
only liberated these African-born mothers and his Brazilian-born crioulo
children but also, in his will o f December 26,1868, declared all o f these chil
dren legitimate and named them as his heirs.13 Having been manumitted,
his children obtained the legal right to succeed as his testamentary heirs. No
slave had the legal capacity to accept or reject an inheritance before the Free
Womb Law o f 1871.14
In urban slavery in Salvador, African-born enslaved fathers often shared
their first names with sons o f Brazilian birth, whereas mothers and daugh
ters rarely shared first names. Naming a son after his father is a common
practice in any patrilineal society, but in the context o f nineteenth-century
Salvador, where common-law unions among slaves did not always accom
pany the coresidence o f partners and children were usually raised by m oth
ers, this naming practice helped children born as slaves to maintain father-
son relationships in those cases in which the father and mother were to end
their relationship. Bilateral kinship relationships developed with the recog
nition o f consanguinity by fathers, and this was evidenced in their naming
practices.
Brazilian-born slaves could also receive financial support from their grand
parents. Some grandparents paid cash to purchase their grandchildren's free
dom, some left property or money to them, and some named their grand
children as their heirs. The African-born freedman Victoriano Maria, who
lived in the colony o f Lagos as a professional Muslim teacher, left some
money to his sons son in his w ill registered in Salvador (see chapter 5).15
Another African-born ex-slave named Valentino Paschoal left half o f his
property to his wife, Maria Espirito Santo Pires, also an African-born ex
slave, and the rest to his still enslaved granddaughter, Belimira, for her free
dom. Belimira was one o f the four children o f his daughter, whom Valen
tino had had by a slave wom an.16
Although parent-child relationships seem to have extended bilaterally, it
is likely that matrilineal ties survived in the long run with much greater
frequency than patrilineal ones; many more grandmothers referred to their
grandchildren in their wills than did grandfathers. Maria de Araujo Ribeiro,
an ex-slave woman from Costa da Mina, who died on October 3,1815, named
her granddaughter, Anna Florencia de Almeida, as the heiress o f two-thirds

o f her property and left the rest to her little great-granddaughter named
Anna, daughter o f Anna Florencia and her husband, Sergeant-Major Luis
Gonsaga de Barros. M aria appointed her son-in-law as executor o f her
w ill.17
Brazilian-born slaves benefited from their fictive kinship as another
source o f financial support. During the early nineteenth century, nearly
90 percent o f Brazilian-born slaves o f both sexes were baptized only with
godfathers, most o f whom were free-born, as in the case o f African-born
slaves for the same period.18 One naturally suspects that some o f the god
fathers may very well have been the fathers o f the Brazilian-born slave chil
dren, but unfortunately their relationships were not mentioned in baptismal
records. In the early nineteenth century, Brazilian-born slave children were
sometimes baptized w ith African-born godfathers, who shared the same
“nation” with their slave mothers, who were also o f African birth.19Brazilian-
born slave children could have benefited from their mothers’ ethnic net
works. Some godfathers indeed paid for the manumission o f Brazilian-born
slaves; sixteen godparents (seven for the years 1808-1842 and nine for 1851-
1884) paid money to purchase the freedom o f Brazilian-born slaves.20 By
contrast, only three godfathers contributed to the manumission o f three
African-born slaves, all women, for the years i8o8-i884.21 This suggests that
slave parents, mothers in particular, could have chosen as their children's
godparents those who might be able to afford to pay for the freedom o f their
godchildren, or in cases where they had no right to choose their children's
godparents, they could have at least successfully developed good relation
ships with their coparents for their children’s sake.
T h e Pursuit o f Freedom in M anum ission
The practice o f manumission greatly favored Brazilian-born over

African-born: Brazilian-born slaves were manumitted more frequently than
their African-born counterparts throughout the nineteenth century (see
chapter 3).22 Even during the first three decades o f the nineteenth century,
when the African-born slave population largely outnumbered their Brazilian-
born counterparts by tw o to one, the larger society always favored the latter
over the former in the practice o f manumission.23 Brazilian-born slaves
benefited from the practice o f both unconditionally and conditionally un
paid manumission. The individual slave’s birthplace determined to a great
degree the experiences with manumission; Brazilian-born had a much bet
ter chance to gain freedom than African-born.
The Creation o f Disparate Identity; 1808-1851 127

In many cases, Brazilian-born slaves were granted freedom uncondition
ally for such reasons as “good services,” "fidelity and obedience,” “love o f
God,” and “love o f raising the slave.” As Stuart B. Schwartz points out, these
were not real motives for manumission but usually a kind o f necessary “pre
condition” or minimum requirement.24 In most cases o f conditional unpaid
manumission, the owners obliged their ex-slaves to “accompany and serve”
them (or sometimes their spouses, children, or siblings) until the death o f
the latter. Those thus manumitted conditionally, who were now named ag-
regados (dependents in households), continued to work for their former
owners in the same way as formerly when they had been slaves.25 Despite
their legally free status, these Brazilian-born ex-slaves remained totally sub
ject to other people’s power. Owner-slave relations were thus maintained
with little modification after the slave gained freedom.
The most common path to freedom for Brazilian-born slaves was unpaid
manumission, either unconditional or conditional, and not manumission by
self-purchase as in the case o f their African-born counterparts. But this does
not mean that their individual experiences in manumission were monolithic
or homogeneous; all Brazilian-born slaves were not treated equally in the
practice o f manumission. Most importantly, gender, age, and skin color de
termined the individual’s chance in manumission. Interestingly, two-thirds
o f Brazilian-born ex-slaves were women, most o f whom had developed
strong and/or intimate interpersonal relationships with their owners and
their families as domestics, some o f whom were forced to become their
owner’s mistresses. As for age, children were favored over adult slaves in the
practice o f manumission.26 T h irty percent o f Brazilian-born ex-slaves in
Salvador for 1808-1884 were classified as children.27 This is a much higher
percentage than among the slave population o f Brazilian birth, as illustrated
by the adult-child ratio o f 85:15 am ong slaves purchased and sold in Sé
parish for the period 1838-1888.28 In combination o f gender and age in the
' practice o f manumission, women comprised nearly 60 percent o f Brazilian-
born adult ex-slaves and 70 percent o f Brazilian-born child ex-slaves. Fur
thermore, some newly born and child slaves were baptized as forros on the
grounds that the owner was grateful to the slave’s mother, also a slave under
the same ownership.29 In only one o f these cases did the owner become his
slave’s godfather in order to “benefit him [the slave].”30As long as “cheaper”
human com m odities were continuously supplied from Africa, owners in
Brazil usually preferred to manumit Brazilian-born slaves at birth and dur
ing childhood unconditionally and without payment. Because o f the high

risk o f infant and child mortality, raising them was much more costly than
purchasing newly arrived African-born adult slaves,31
Skin color also determined the Brazilian-born slave’s experiences in the
practice o f manumission. Mulatto slaves had a better chance o f manumis
sion than their black counterparts; lighter skin color gave the slave a great
advantage; the criow/o-mulatto ratios for Brazilian-born ex-slaves were 64:36
for 1808-1842 and 63:37 for 1851-1884.32 Brazilian-born slaves purchased and
sold in SI parish suggest higher percentages o f crioulos, namely 75.2 percent
for 1838-1848 and 67.4 percent for 1852-1888.33 Slaves registered in the 370
inventories from Salvador show an even higher percentage o f crioulos: 89.3
percent for 1808-1849 and 78.2 percent for 1850-1888.34 Mulatto slaves, re
gardless o f gender and age, evidently had easier access to freedom than did
their crioulo counterparts, but this tendency is most clearly observed in the
case o f female child slaves.35 There is no doubt that some o f these mulatto
slave girls were their owners’ illegitimate children by slave women; their bio
logical fathers might have manumitted them out o f affection, although they
would still have been raised by their mothers in the same household, as
quasi-slave domestic servants.
A further factor weighed by owners was economic. Male slave children
were valued as capital more highly than their female counterparts and thus
were more worthy o f an owner’s investment o f time and money; for those
with good occupational skills, slave men were usually more highly valued
in monetary terms than were slave women.
The manumission o f the Brazilian-born slave population constantly pro
duced a specific group o f the newly manumitted, who were characterized
as predominantly female, child/young, and mulatto, and many o f whom
were to be easily and sm oothly incorporated into domestic servitude, in
which they were legally free but in a situation that was not very different
from slavery. The rest, namely male, adult, and black, who were supposedly
much more suitable for plantation slave labor and who constituted the ma
jority o f the Brazilian-born population, were forced to remain in slavery
for use in all sorts o f manual labor, and later as human commodities for the
interprovincial slave trade with the coffee-boom ing Southeast. Thus the
racial/gender hierarchy among the Brazilian-born slave population was pre
served and strengthened with a sufficient supply o f slave labor force for the
maintenance o f the urban slave society.
From the early colonial period, the free Brazilian-born population o f A f
rican descent had expanded through recruitment, through manumission,

and through reproduction, with a lower child mortality rate than that o f the
slave population. Free people o f African descent created and re-created dis
parate identities over the years, most importantly in accordance w ith color
and gender.
T h e Creation o f Identity in Freedom
The Brazilian-born population o f African descent, either enslaved,

freed, or free-born, had always been divided into various categories o f color
since the early colonial period. In nineteenth-century Salvador, color func
tioned as a clear index o f the individual’s social status, particularly in the
case o f holy matrimony. Marriage never had been a norm for the general
populace in Salvador, but marriage rates am ong the legally free popula
tion were certainly much higher than among slaves, who rarely contracted
church marriages. Slave status was the most crucial obstacle to church mat
rimony, and ex-slaves certainly enjoyed m arrying with much greater fre
quency than the enslaved population. This is well exemplified by the case
o f property-owning African-born ex-slaves, who married their longtim e
common-law spouses after gaining freedom but late in their lives, often on
their deathbed, obviously for the sake o f inheritance (see chapter 5). Mar
riage registers in Penha parish and C o n cei^ o da Praia parish clearly suggest
the correlation between marriage and color. Being lighter-skinned meant
being better off, socioeconomically; one could afford more easily the cost o f
a church marriage. Also, when marriage was contracted, it was largely en-
dogamous: marriage partners were chosen w ithin the same categories o f le
gal distinction (free-born, freed, or slave) and place o f birth (Brazilian-born
or African-born), as well as o f color (white, crioub, pardo, or cabra)?6 These
data for marriage in nineteenth-century Salvador concur with demographic
studies for the overall population o f Brazil conducted by Thom as W. Mer
rick and Douglas H. Graham: marriage rates certainly reflected “a distinct
racial hierarchy.”37 This suggests that marriage functioned as an important
institution to strengthen pre-existing group solidarity, not to develop new
networks across group boundaries.38 Such endogamous tendencies are a
clear reflection o f the ways in which the population o f African descent cre
ated their identities in relation to the larger society's perceptions o f race and
color.
Interracial marriage rarely took place, at least officially; color determined
one's choice of marriage partner. When interracial marriages did occur,

they were most frequently between white men and Brazilian-born mulatto
(pardo or cabra) women o f legally free status. For instance, the complete
census data o f 6,194 residents for Sào Pedro Velho parish in 1835 include no
case o f interracial marriage.39 Interracial unions usually occurred in the
form o f concubinage. Such is the case o f the African-born freedwoman
Faustina Maria dos Prazeres, who never married. She purchased her free
dom for ioo$ooo réis from her owner, Rita Maria dos Prazeres, and then as
a concubine o f Captain Cristóváo da Rocha Pinto, Faustina M aria bore
three daughters, respectively named Julia, Claudiana, and Maria da Penha.
In his w ill Captain Rocha Pinto recognized their daughters as well as his
other children as his heirs. A fter his death, Julia also passed away, so that
three crioulo slaves o f minor age (Raimunda, Ardeza, and Calistato), whom
her father had given to Julia during his life, passed into the possession o f
Faustina Maria, who liberated all three slaves in her w ill dated July 7,1840.
Faustina Maria left her property to her two surviving daughters as her legal
heiresses, and she appointed as executor o f her w ill the grandfather o f her
daughters, Captain-M ajor Cristóváo da Rocha Pinto, Sr.40
For the free Brazilian-born population, color was a “symbol o f social
status,” in the words o f the anthropologist Verena M artinez-Alier (a.k.a.
Stolcke) in her study o f nineteenth-century Cuba.41 Those who were cate
gorized by others and/or identified themselves as the same color formed an
endogamous group for the sake o f marriage as a social institution, at least
temporarily.42 O n the one hand, marriage across color boundaries m ay have
brought dishonor to the family o f a spouse who married down, as in the
case o f nineteenth-century Cuba.43 O n the other hand, in Brazil, which
never established “ legal colors” for the prevention o f interracial marriage,
the individual could have more freedom to officially change his or her color
to match the socioeconomic status o f the spouse, particularly in the case
where he or she would move up on the social ladder by means o f mar
riage. Therefore, statistical evidence for the prevalence o f marriage within
the same color category does not necessarily suggest that color was an ab
solute determinant for the official choice o f marriage partners. The signifi
cance o f one’s individual color may have been regarded (and overlooked or
dismissed) as a personal preference among the free population o f African
descent. But the lighter skin o f the pardo or cabra definitely gave that popu
lation a great advantage in manumission, and furthermore various awesome
benefits for upward social m obility in a racially stratified society, benefits
which the dark-skinned population could not possibly expect. Naturally the
The Creation o f Disparate Identity; ¡808-1851 131

mulatto population preferred to separate itself from the black population.
It tended to comprise a rather exclusive, homogeneous social group or stra
tum and created a distinctive identity in freedom.
From the early colonial period, mulattoes (pardos)>w ho were often bom
free or had a much better chance for freedom in manumission, had refused
to identify themselves with “blacks,” namely the African-born and crioulo
populations. Mulatto women and men established and developed their own
lay sodalities, which came into being later than black sodalities. Mulatto so
dalities accepted and welcomed white members, as their black counterparts
did, but did not allow either slaves (both African-born and Brazilian-born),
African-born ex-slaves, or crioulos, either free or slaves, to be members. M u
latto sodalities were often dedicated to O ur Lady o f the Liberation (Nossa
Senhora do Livramento), O ur Lady o f Succor (Nossa Senhora do Amparo),
Our Lady o f Conception (Nossa Senhora do Conceiçâo), and O ur Lady o f
Guadalupe (Nossa Senhora do Guadalupe). During the colonial period m u
lattoes had established the sodality o f O ur Jesus o f the Cross (irmandade de
Nosso Senhor Jesus da Cruz) in the Church o f Palma; the sodality o f O ur
Lord the Good Jesus o f Patience (irmandade do Nosso Senhor Bom Jesus da
Paciêntia); and the sodality o f O ur Lady o f Conception o f Boqueirào (ir
mandade da Nossa Senhora da Conceiçâo do Boqueirào), both in the church
o f Sào Pedro parish.44
The free mulatto population’s creation o f identity was reflected in cer
tain collective actions that they took during the Age o f Revolution. They
had become receptive to the ideas o f the Enlightenment and initiated an
insurrectionary plot generally known as the Tailors’ Revolt (Conspiraçâo
dos Alfaiates).45 This event started on August 12,1798, when the urban resi
dents in Salvador awoke to a proclamation posted on a church door, calling
for liberty, freedom, and independence. The police arrested forty-six men
from Salvador and the Recôncavo. Among thirty-tw o participants, all men,
brought to trial, twenty-three were pardos (twelve free men and nine slaves),
and ten were whites. O nly one man was African-born: a newly arrived slave
from Costa da Mina. Seven free men and four slaves were tailors by occu
pation, which is the reason why this conspiracy has become known as the
Tailors’ Revolt, or the Conspiracy o f the Tailors.46 The Bahian authorities
accused four free pardos as chief conspiratory plotters: Luis Gonzaga dos
Birgens, soldier; Joâo de Deus Nascimento, tailor; Lucas Dantas d’Amorim
Torres, soldier; and Manuel Faustino dos Antos Lira, tailor, all o f whom were
executed. The authorities prosecuted four white upper-class intellectuals,
three o f whom were members o f a secret Masonic society, the Knights o f the

Light, which had been established on July 14 ,1797.47 Contemporary docu
mentation shows that the idea o f “ liberty,” derived from the Age o f Revo
lution, was critical in spurring participants to collective rebellion, but that
their motivations varied from one individual to the next.48
Most o f the participants in the Tailors' Revolt were “elite'1Brazilian-born
people o f African descent, as generally observed in slave uprisings in other
parts o f the New World.49 But unlike the other cases, many were free-born;
the Tailors’ Revolt was allegedly plotted mainly by free pardo artisans. The
ideas o f the revolutionary era prevailed only among a specific group o f free
light-skinned men who were skilled artisans, and these people utilized the
new concept o f “liberty” to take collective political actions.50We should em
phasize that no women participated in the Tailors' Revolt, either. It was an
entirely male phenomenon, as in the case o f all the slave uprisings in Bahia.
There existed a limited amount o f upward social m obility for the free popu
lation o f African descent, albeit only for men, regardless o f color; small
numbers o f free men o f African descent, mostly those o f Brazilian birth,
found an opportunity in a m ilitary career to move up on a social ladder, only
w ithin the well-defined parameters. In colonial and imperial Brazil, military
service depended on the free poor, who were predominantly those o f A fri
can descent, particularly in the case o f Salvador; the army was perceived as
an assembly o f the poor, the ignorant, and the lazy. The social status o f sol
diers, who were poorly and irregularly paid, was only slightly higher than
that o f slaves.51 Most soldiers remained poor but, as Joan £. Menzar dis
cusses in the case o f the Northeast for 1830-1875, m ilitary service in colonial
and imperial Brazil responded to social differences among the free poor, and
the army accentuated such divisions as those between the honorable and the
dishonorable, the worker and the vagrant, and the free and the slave.52 The
rank and file were predominantly poor, since a large number o f men were
exempt from m ilitary service by virtue o f their occupation and status.53
In colonial Brazil, colored militia regiments had come into existence in
all major Brazilian cities by the eighteenth century. As in the case o f mulatto
sodalities, urban mulatto regiments were usually formed at a later date than
those o f blacks.54 Free soldiers o f African descent in Salvador were further
more divided into separate militia regiments by color distinction. By the late
eighteenth century, Salvador had established four m ilitia regiments orga
nized by color and occupation: tw o white regiments (one for leading mer
chants and another for less wealthy whites); the “henriques” (named after
Henrique Dias, the black hero o f the war against the Dutch from 1645 to
1654), enlisting free crioulos; and the free mulatto regiment. In the first de

cades o f the nineteenth century, the militia had been reorganized into six
regiments but with the same division by color. The six regiments were the
following: one company o f mounted gentlemen, forming the government
o f honor; one squadron o f flying artillery; tw o regiments o f whites, almost
all tradespeople; one regiment o f mulattoes; one o f free blacks, whose num
ber totaled 4,000 in the early 1820s.55 Thus, as in the case o f black and lay
sodalities, free blacks and mulattoes had to belong to separate regiments by
color.
Furthermore, there was an irreconcilable cleavage o f identity among the
“free black” men in the same regiment. Schwartz documents a very interest
ing petition made by crioulos in Bahia in 1775: The black militia regiment o f
Bahia complained that African-born ex-slave men were appointed to posts
o f commands, against the usual practice in Bahia and Pernambuco. The
crioulo supplicants claimed that African-bom people were “suspect in their
faith and orthodoxy and were capital enemies o f the whites against whom
they often plotted rebellion,” while identifying themselves as “loyal servants
o f the Crown who had proven their value on many occasions.” Interestingly,
the governor o f Bahia responded to this petition by pointing out that the
exclusion o f African-born was not due to any royal order but rather to
the practice o f the regiment.56 We have no way o f knowing to what degree
African-born ex-slaves were recruited for the free black regiment, but this
particular petition made by the crioulo men clearly suggests that there had
been clear divisions in identity among the free “black” men who belonged
to the same regiment because o f the same color. Crioulos did not identify
themselves with African-born ex-slaves, although, as “blacks,” people o f A f
rican birth and the crioulo population had established and developed the
prominent form o f voluntary associations together: black lay sodalities.
The presence of the free black regiments functioned by design for the
maintenance of the existing social structure based on color hierarchy by in
corporating free black men into the power systems with only a very limited
amount of upward social mobility. The free black men were forced to locate
themselves close to the bottom o f the larger society, albeit at least not at the
very bottom of it, which was occupied solely by the slave population.
T h e Em ergence o f a N ew Free-Born Black

( Crioulo Livre) Identity
Demographic changes among the population o f African descent in

Salvador progressed gradually once the slave trade was officially banned in

Brazil (1831). The proportion o f the free Brazilian-born population o f Afri
can descent began to expand. W ith their increasing financial power, the free-
born male crioulo population created a collective identity and established
and developed new sodalities o f their own around the same time the laity’s
participation in sodalities began to decline rapidly.
The "Disappearance” o f Black Lay Sodalities
Black sodalities virtually “disappeared” from the sociocultural scen

ery in Brazil during the nineteenth century, as several historians o f Brazil
have already noted. The disappearance took place when the general laity’s
active participation in lay sodalities declined rapidly because o f the secu
larization process o f Luso-Brazilian culture; during the nineteenth century
the Church itself became a less decisive factor, with the diminishing impor
tance o f lay sodalities whose origins lie in medieval Europe.57
In fact, in the case o f Salvador, regardless o f race, gender, or color, the
genera] free laity’s participation in lay sodalities declined sharply before the
mid-nineteenth century. At the beginning o f the nineteenth century, for in
stance, nearly a hundred lay sodalities, either white or black, existed and
functioned in Salvador, and more than 85 percent o f the free-born residents
o f both sexes who registered their wills in Salvador held memberships in
lay sodalities.58 But this percentage decreased rapidly during the late nine
teenth century: 20 percent o f men and 24.2 percent o f women were mem
bers o f lay sodalities for 1851-1854; and only 19.3 percent men and 8.4 percent
women for i878-i885.59The wills o f African-born ex-slaves demonstrate the
same disposition in their memberships o f black sodalities; among the 166
wills o f ex-slaves I consulted, very few ex-slaves o f either sex (94 males and
72 females) who registered their wills between 1850 and 1888 were members
o f even a single sodality.60 The number o f black sodalities mentioned in
their wills also decreased from twenty (1808-1849) to eleven (1850-1888).6!
As a result, for instance, in the year 1862, the black sodality o f the Rosary in
the Pelourinho, while still holding 1,412 members (825 males and 587 fe
males), accepted only 8 new members (3 males and 5 females), and lost 4
members through death 62
The major reason for the aforementioned change in the laity’s participa
tion in sodalities is that long-prevailing Baroque religious values were losing
ground in nineteenth-century Brazil. Indeed, around 1850, the form o f reg
istered wills in Salvador became more abbreviated, and their statements
became simplified and more straightforward, which is attributable to the

ongoing secularization process o f Luso-Brazilian culture. N o longer did
Luso-Brazilian culture place importance on formal participation in religious
festivals and other ceremonial activities under the name o f a specific patron
saint. Neither did ostentatious display at a funeral, which required atten
dance o f fellow sodality members, serve as an indication o f the deceased’s
social status, as had hitherto been the case for the preceding centuries. No
matter how important church burials might still be for sodality members,
after the middle o f the nineteenth century no lay sodality in Salvador was
entitled to bury any o f its members w ithin its church. A fter the prevalence
o f yellow fever in Salvador in the year o f 1849, burials w ithin churches were
strictly prohibited; people had to be buried in one o f three cemeteries at a
short distance from the city.63
Interestingly, despite the above-mentioned secularization o f Luso-Brazilian
culture, increasing religiosity was noted in Salvador at around the same
time. It was clearly manifested in the rapid penetration o f lay Christianity
in the form o f Marian devotion among the Brazilian-born slave population.
The cult o f the Virgin M ary and its miraculous manifestations have always
been extremely popular over the centuries in the Iberian Peninsula and in
Portuguese and Spanish overseas colonies, according to C . R. Boxer. This
cult was not limited to the poor and lowly, but was common to all ranks
and conditions.64 In Brazil, Marianism, which defined women’s roles in the
family, and veneration o f the Virgin M ary were o f unparalleled importance
and far exceeded devotion to any single saint.65 In 1848, for instance, James
Wetherell noted that there were niches in the walls at many street corners
o f Salvador, and in these were offerings by the faithful under statues to the
Virgin Mary.66
In Salvador, during the first four decades o f the nineteenth century, the
Virgin M ary (Nossa Senhora) was not recorded as godmother in baptis
mal records for either African-born adult slaves or Brazilian-born child
slaves. But in the late 1840s, the custom o f invoking the Virgin M ary as god
mother started to prevail for baptism o f the Brazilian-born slave population
in Salvador. For the period 1848-1849, 470 baptismal records in Santo An-
tdnio parish reveal that forty Brazilian-born infant slaves (nineteen boys
and twenty-one girls), as well as two African-born slave women, were bap
tized with the Virgin M ary as their godmother. For the years 1858-1859 and
1868-1869, there were thirty-eight Brazilian-born slaves (twenty boys and
eighteen girls) for whom the Virgin M ary was named godmother. Many o f
these godchildren were baptized w ith human godfathers, but others did not

have any godfathers.67 Being baptized and having the Virgin M ary as god
mother has possessed spiritual meaning for the population o f African de
scent, perhaps as well as for whites. There is a contemporary dimension to
this practice, in that the custom prevails today among people o f color and
whites alike o f baptizing new-born babies and choosing the Virgin M ary as
their godmother.68
Invoking the Virgin M ary as godmother for baptism has been a com
mon Catholic religious practice that derives from early medieval Europe.69
Through the Virgin M ary as a mediator, the faithful could reach God/
Christ.70God and Christ were more feared than loved; the former was a re
mote and brooding eminence, whereas the latter was represented either as a
child or sacrificed on the cross— images that referred to plague and judg
ment. By contrast, M ary was a beloved intercessor who worked to deflect or
soften the harsh judgment o f a stern God.71 Thus, by haying the Virgin
M ary as godmother as well as having the name o f Mary, a lay person’s soul
was connected with God/Christ, and that soul might eventually be saved. In
medieval Europe lay Christian practices came into being such as selecting
the Virgin M ary or a local female patron saint as godmother, choosing mul
tiple sets o f godparents, and having the officiating priest serve as sponsor at
baptism.72 In the mid-sixteenth century the Council o f Trent finally prohib
ited these lay Catholic practices, which had occurred in the Iberian world
over centuries. Yet these practices were slow to disappear and continued to
exist in the Iberian Peninsula and in Brazil, especially in the baptism o f
slaves or among the lower class.73 Secularization o f Luso-Brazilian culture
led to the reemergence o f such lay Catholic customs among the general
populace in Salvador during the early years o f the nineteenth century. Prob
ably many slave owners forced their slaves to practice the same Christian
customs, but there is no doubt that in the 1840s slaves and the free popula
tion o f African descent, many o f whom were Brazilian-bom, began to ac
cept them in their own search for a Christian faith.74
In Salvador, lay Christianity penetrated the population o f African de
scent after the 1830s, when the influx o f new arrivals from Africa started to
drop drastically. The Brazilian-born population o f African descent, both
slave and legally free, more easily accepted Luso-Brazilian culture and social
norms. The population o f African descent, including slaves, began to find
more and more symbolic meanings in lay Christianity, which had reemerged
among the general laity in Salvador, in relation to the secularization o f Luso-
Brazilian culture. It is no coincidence that the lay Christian practice o f in-
The Creation o f Disparate Identity,; 1808-1851 137

yoking the Virgin M ary as godmother became prevalent among the popu
lation o f African descent when the laity's participation in sodalities began
to disappear rapidly.
The Establishment o f New Black Sodalities
The black laity's rapidly declining participation in black sodalities,

however, is by no means the end o f the story for black sodalities in Salvador.
In the late colonial period, crioulos had established sodalities dedicated to
Good Jesus o f Martyrs (Senhor Bom Jesus dos Martirios), and venerated
Saint Benedict in all churches in Salvador.75 But, o f course, most o f those
crioulos were slaves, and their sodalities were inclusive in terms o f race, eth
nicity, gender, and birthplace, as in the case o f “African” black sodalities.
The crioulo population did not just give up this special form o f voluntary
associations, which their enslaved ancestors had sought after and skillfully
elaborated for their specific needs and priorities; free-born blacks o f Brazil
ian birth (crioulos livres) established and developed two new black sodalities
with a much more exclusive character than had hitherto been the case.
A new crioulo sodality came into being at a meeting in the chapel o f O ur
Lady o f the Rosary in the Church o f the Q uinze Mistérios o f Santo Antônio
parish on September 16,1832. It was named the sodality o f O u r Lady o f
Solitude and Support o f the Needy (irmandade de Nossa Senhora da Sole-
dade Am paro dos Desvalidos). The founder and judge was Manoel V ic
tor Serra, a free-born crioulo man whose occupation w§s wage earning. The
other founding members, all o f whom were free-bom crioulo men, included
officers Manoel da Conceiçâo (cabinet maker), as treasurer; Luis Teixeira
Gomes (bricklayer), as scribe; José de Nascimento, as procurator; José Maria
Vitela, as a member o f the administrative committee. Other founding mem
bers were Gregório M. Bahia (cabinet maker), Ignácio de Jesus, Bernabé Al-
vano dos Santos, Bernardino S. de Souza (bricklayer), Pedro Fortunato de
Farias (bricklayer), Gregório de Nascimento (carter), Balthazar dos Réis
(cabinet maker), Manoel Sacramento Conceiçâo Roza (cabinet maker),
Teotonio de Souza (vinegar maker), Francisco José Repino (caulker), Daniel
Correia (wage-earner o f the canto o f Pilar parish), Roberto Tavares (water
carrier), José Fernandes do Ô (bacon seller), and Manoel M artins (worker
at the Lenha wharf). There was no woman among all the officeholders. The
sodality placed its safe box in the keeping o f the vicar o f Santo Antônio
parish, the Reverend Joaquim José de Sant’Anna. The three different keys to
this safe box were kept in the hands o f the above-mentioned judge, trea

surer, and scribe.76 In August 1832, the founder o f the sodality, Manoel V ic
tor Sena, proposed the creation o f a lottery, with printed tickets o f 320 riis,
as a potential financial source for the sodality. His proposal was approved,
and the lottery was named “ Loteria Nossa Senhora da Soledade.”77 Unfor
tunately, we have no way o f knowing the composition o f the membership
o f the crioulo sodality o f O ur Lady o f Solitude and Support o f the Needy;
the statutes and the other prim ary documents o f the sodality, which were
housed in the Church o f Quinze M istlrios, have been lost, and failure to
maintain the church has eliminated any hope o f finding them.
The crioulo sodality o f O ur Lady o f Solitude and Support o f the Needy
did not just “disappear” during the 1840s; instead it survived this critical
decade and even thrived with the financial power o f the members. This par
ticular sodality continued to function at the chapel in Quinze M istlrios
Church during the 1830s and 1840s; then on December 17,1848, its head
quarters were transferred to the black church o f O ur Lady o f the Rosary in
the Pelourinho. This new free-born crioulo sodality was to transform itself
into a new type o f mutual-aid “black” association named the Society for the
Protection o f the Needy (Sociedade Protetora dos Desvalidos), as we shall
discuss in chapter 7.
Another new sodality o f free-born crioulos came into being in December
1843 in Agua de Meninos o f Pilar parish. This was named the sodality o f
Saint Francis o f Paula o f Free-Born Crioulos Born in the Empire o f Brazil
(irmandade do Glorioso S£o Francisco de Paula de crioulos livres nascidos
no Imperio do Brasil).
The prologue to its statutes tells us an interesting “history” o f this so
dality. The Reverend Father Ant6nio Borges Monteiro was in the habit o f
taking his walks to visit the house o f an elderly friend to converse w ith him
for a couple o f hours every afternoon. One day, Father Monteiro found a tin
veronica bearing the image o f Saint Francis o f Paula. He picked it up and
locked it in a drawer at home, but three days later he again found the same
veronica in the same place. Recognizing a miracle, he bought the house o f
his friend and ordered an area cleared where there was a fugitive slave com
munity (quilombo) and bush and weeds. Father Monteiro made use o f all
o f his financial resources in building a chapel. He stipulated that the chapel
be dedicated to Saint Francis o f Paula and placed the image o f St. Francis,
which he ordered to be made, in the capela-mor (primary chapel). At his
death, he left all o f the possessions that he had inherited from his parents
to establish the patrimony for the chapel and to provide for the upkeep o f
the veneration o f the saint. He named as his executor Theotonio de Amorim

Falçâo, and in one o f the clauses o f his w ill, Father Monteiro declared him
self as the Father Founder. He ordered his executor to make his godchild,
named Francisco de Paula Borges Monteiro» responsible for the chapel, to
apply his endowment to its administration, and to let him be the chaplain.
In another clause o f his w ill the Father Founder ordered that if, some day,
one or more priests o f the order o f Saint Francis o f Paula were to appear
in Salvador, his executor or the aforesaid godchild was to let the priest or
priests maintain the chapel. This would nullify the provision charging his
godchild with this responsibility. However, the executor died in 1819 w ith
out seeing Father M onteiros wishes brought to fruition. The chapel was
abandoned until, in 1843, some crioulos appeared. W ishing to venerate Saint
Francis o f Paula and support the chapel, they requested that the archbishop
grant permission for them to adopt the chapel as the seat o f their new so
dality.78
We have noted that Agua de Meninos had formerly been a place o f refuge
for fugitive slaves. It was also there that African-born slaves and ex-slaves
had engaged in their final battle with the municipal troops in the Malê re
volt o f 1835 (chapter 5). This story or oral tradition may be symbolically
important in considering the establishment o f a free-born crioulo sodality,
since they chose the chapel built in the place where a quilombo had once
been formed and which was to become the battlefield for the last stage o f
the Malê revolt.79As for the governing body o f this crioulo sodality, whereas
no limitation was placed on slave status, officeholders had to be crioulos. The
charitable activities o f the sodality were directed toward assisting the indi
gent in general by creating a mutual insurance society (monte pio).80 The
mutual insurance society was governed by a board composed o f a director
and six brothers, which acted independently o f the governing body (mesa)
o f the sodality. O nly male members who were in good standing and whom
the scribe o f the sodality recommended were allowed to participate in the
mutual insurance society. Dues were 25$ooo réis, which could be paid in
monthly installments o f 500 réis over four years. The first month’s contri
bution was deposited by the board in an account in the Caixa Econômica
da Provincia. M inim um capital o f 400$000 réis was to be established and
placed on loan, principally to the elderly and the sick who could not provide
for themselves. W ives and children were entitled to receive financial assis
tance in those cases where the death o f a husband would result in destitution
for the survivors.81
Although the membership was not exclusive in terms o f legal status,
color, or gender, this new sodality clearly represented the interests o f the

free-born crioulo population. The establishment o f a mutual insurance so
ciety within the sodality represented husbands’ concern for their wives and
children. Among the free crioulo population, family had become an impor
tant institution, in which gender roles were redefined and gender identi
ties were re-created to a degree. Yet many such couples were not united by
church matrimony; they were merely common-law spouses.82
Conclusion
In the eyes o f the larger society, the “otherness” o f Brazilian-born

slaves was not determined culturally but merely in terms o f race. By being
born into slavery in Brazil they naturally shared w ith their owners the same
language (Portuguese) and the same Luso-Brazilian culture. Whereas they
had never been free and therefore did not have a clear notion what freedom
meant, Brazilian-born slaves had a much easier access to freedom than their
African-born counterparts. First o f all, Brazilian-born slaves had their par
ents, other kinsfolk, and sometimes godparents who could pay cash for
the purchase o f their freedom or negotiate with their owners for freedom.
Furthermore, rather than struggling for the opportunity o f self-purchase,
Brazilian-born slaves usually chose to maneuver in their interpersonal rela
tionships with their owners so that they could be manumitted without pay
ment, although it might well mean that they would have to remain in quasi-
slavery servitude under the absolute domination o f their former owners
despite their legally free status. The practice o f manumission greatly favored
women, mulattoes, and children among the Brazilian-born population.
The Brazilian-born population o f African descent by no means shared a
single, homogeneous identity among themselves; they had been always di
vided by color, and the mulatto population had separated themselves from
the crioulo population, most o f whom remained enslaved. During the late
1830s and 1840s, when the urban Bahian society was quickly transforming
itself from slave society to slave-owning society, the crioulo population be
came more and more receptive to Luso-Braziiian culture, which was rapidly
secularizing itself with the la itys declining participation in sodalities and
the penetration o f lay Christianity. This is when the free-born crioulo popu
lation established tw o new sodalities that covered their new needs. W ith
their new financial resources, they started to create a collective identity
based on their memories o f the glorious history o f their enslaved ances
tors, while (re)defining gender identities in the new framework o f double
headed households.
The Creation o f Disparate Identity, 1808-Ì851 141

7 The Labyrinth of Identity,
1851-1888
W ith the official termination o f the transatlantic slave trade in 1851, the
institution o f slavery started to decline gradually nationwide, while the
booming interprovincial slave trade continued to move the still active yet
predominantly Brazilian-born slave population from the Northeast to the
coffee-producing Southeast. By the mid-nineteenth century the urban slave
population in Salvador had already diminished to no more than one-third o f
the whole population o f color.1 During the late nineteenth century, Brazilian-
born slaves substantially outnumbered their African-born counterparts: by
a ratio o f 64:36 among those purchased and sold in Sé parish for the years
1851-1884 and by a ratio o f 58:42 among slaves registered in the 370 invento
ries o f slave owners.2
The free Brazilian-born population had come to dominate among the ur
ban population o f African descent in Salvador by the turn o f the 1870s. In
1872, when the first national census was taken for Brazil, the free population
o f color constituted 57.3 percent o f the whole population o f Salvador. By
contrast, the slave population had decreased to only 11.6 percent o f the
whole population, and they constituted less than a quarter o f the whole
population o f color in the city.3 T he hitherto prominent figures o f male
street slaves as chair carriers had already disappeared from the streets o f Sal
vador. In the last tw o decades o f the slavery regime, urban slaves were em
ployed mainly as artisans if males and as domestics if females.
This chapter will discuss the creation o f racial identity by focusing on the
expanding Brazilian-born population of African descent from the mid
nineteenth century until the year o f abolition. In order to do so, we shall
first discuss how the notion o f freedom changed after the mid-nineteenth
century, and how and why the terminology of color changed in the last few
decades of the nineteenth century. Our emphasis will be placed on a unique
mutual-aid association named the Society for the Protection of the Needy
and on its members' creation of racial and gender identities during the last
few decades o f the slavery regime.
The Changing Notion o f Freedom
The notion o f freedom in Brazil changed rapidly during the late

nineteenth century. First o f all, being under the strong influence o f Great
Britain, Brazilian abolitionist movements advanced quickly and intensified
during the last few decades o f the slavery regime.4 Naturally, abolitionism
had much more success in the urban centers, where free labor had already
made progress.5 In Salvador, abolitionist societies were established in the
1850s and 1860s, and they contributed money for manumission, but only to
Brazilian-born slaves and m ainly to infant and child slaves. In my sample
o f manumission letters, for instance, twelve men and twelve women, all
Brazilian-born, were manumitted w ith payments o f money for the years
from 1851 to 1884, from such abolitionist societies as the Society o f June the
2nd (Sociedade de Dois de Junho), September the 7th Liberating Society
(Sociedade Libertadora Sete de Setembro), Comm ercial Association (Asso-
ciaâo Comercial), Comm ercial Abolitionist Society (Sociedade Abolition-
ista Comercial), Masonic Felicity Beneficiary Society (Sociedade Masónica
Felicidade Beneficiencia), Fraternal Society o f Sergipe (Sociedade Fraterni-
dade Sergipana), Rocini Philharmonic Society (Sociedade Filarmónica Ro-
cini), Terpriano Philharmonic Society (Sociedade Filarmónica Terpriano),
and Bahian Liberating Society (Sociedade Libertadora Bahiana).6 These
abolitionist societies not only contributed money to the slaves’ self-purchase
but also filed criminal lawsuits against slave owners who mistreated their
slaves. For instance, the Bahian Liberating Society filed legal papers w ith the
date o f January 3,1887 for the case o f the 30-year-old Brazilian-born slave
Silvestre, resident o f the engenho Agua Comprida, against the slave named
Ana d’Argolo.7
Brazil’s involvement with international warfare resulted in the emanci
pation o f small numbers o f male slaves but did not improve the slave popu
lation’s official opportunities o f freedom. D uring the Paraguayan War
(1865-1870), the Brazilian government was w illing to buy the freedom o f
slaves who agreed to serve in the m ilitary; their owners had to agree to
emancipate such slaves. In Bahia, the owner was paid the price o f 1:400(000
réis per slave but only 400(000 réis in cash and the rest in public bonds. Also,
Brazilian citizens who were called for m ilitary service were allowed to free
slaves and enlist them as substitutes. The province o f Bahia officially sent a
total o f some 1,800 ex-slave men to the war. According to Barbara Trosko,
this number probably represents only slaves who were bought and freed by
The Labyrinth o f Identity; 1851-1888 143

the government, and does not include slaves who went to war as replace
ments for the upper classes or as volunteers or runaways.8
Most importantly, the Free W omb Law (Lei do Ventre Livre) o f September
20,1871, widely known as the Rio Branco Law, declared legally free the new
born children o f slave women, but only on the condition that the owners
took care o f the children until the age o f eight. Although the “free” children,
or ingênuos, were obliged to work in a state o f quasi-slavery until the age o f
twenty-one, this law resulted in a large ingênuos population; the number o f
ingênuos in Brazil expanded quickly and amounted to 439,831, according to
the report by the Imperial Government as o f June 30,1888. The Free Womb
Law also established an official emancipation fund to be used for the annual
manumission o f slaves in all provinces. For the first time in Brazilian his
tory, slaves were granted the legal right, with the consent o f owners, to have
savings (peculio). Slaves were entitled to purchase freedom i f they could pre
sent an amount equal to their “value ” The law also freed slaves owned by
the state, including those o f the imperial family, and slaves in unclaimed
inheritances or abandoned by their owners.9
W ith the enactment o f the Free W omb Law, slaves in Brazil became le
gally eligible for self-purchase. Now that more and more slave owners were
willing to manumit their slaves for monetary payment, some slaves in Sal
vador managed to negotiate with their owners for monthly payments for
self-purchase. Such were three Brazilian-born slaves who were freed in 1871
on condition they make monthly payments: Romana, 29-year-old crioulo,
daughter o f Virginia, Angola, w ith 6$ooo réis a month; M artinho, crioulo,
with io$ooo réis a month; Marcelino, 26-year-old crioulo, son o f the Ango
lan slave woman, io$ooo réis a m onth.10Other slaves made an effective use
o f public deposits for their savings. For instance, the slave woman Eduarda
paid 5oo$ooo réis, which she had saved in the public deposit, to eventually
obtain her letter o f liberty with the date o f March 20,1883.11
T he urban slave population in Salvador soon became keenly aware o f
their newly granted legal rights and indeed fought for them in court. In 1872,
a parda slave woman, Augusta, represented by her owner, Domingos Joâo
José de Almeida Costa, a resident o f Santana parish, hied a criminal law suit
against a free-born crioulo man, Carlos Hermes da Purificaçào, charging
that he had deceived her with intent to steal 6oo$ooo réis. Augusta had saved
this large amount o f money in her account in the public deposit o f the Caixa
Econòmica (Public Saving Bank) to purchase her own freedom. Carlos had
been having an affair w ith Augustas sister, Ubaldina Anna de Conceiçâo,
slave o f Major Joaquim Domingos Lopes. One day, Carlos heard from Ubal-

dina o f Augusta’s large nest egg and began to scheme to take it from her.
First, Carlos proposed marriage to Ubaldina, and they became engaged. The
new fiancé made up the fake story that the September the 7th Liberating
Society (Sociedade Libertadora Sete de Setembro), a Bahian abolitionist so
ciety, would pay some money toward U baldinas freedom, but that amount
o f money would not be enough. But i f Augusta were to withdraw her sav
ings and loan that money to Carlos, this sum together with the money from
the abolitionist society would be enough for Carlos to purchase Ubaldina's
freedom. Carlos begged Augusta to do so for her sister's sake. Augusta was
finally persuaded to do as he asked for her sister's happiness. Ten months
passed after Carlos had taken the money and disappeared. Augusta reported
the matter to the police, and they quickly arrested Carlos. O f course, the
money would never be regained by the poor slave woman Augusta.12
The Free Womb Law completely broke the social ideology o f urban slav
ery in Brazil. As Brazilian historian Sidney Chalhoub discusses in his study
o f the manumission o f slaves in the city o f Rio de Janeiro during the nine
teenth century, especially in urban areas o f Brazil the ideology o f manumis
sion was an “essential aspect o f work-discipline under bondage” For, before
the Free Womb Law, owners and slaves wdid abide and live according to the
assumption that each owner's exclusive power to decide his slave's access to
freedom was a key organizing principle o f daily social relations.” 13W ith the
enactment o f the law that “ humanized” the human chattel, the owner be
came unable to discipline a slave for work and to exercise absolute power
over the slave. The U.S. historian Sandra Lauderdale Graham, utilizing court
records and focusing on female slave prostitution in the city o f Rio de Ja
neiro, also demonstrates that, with the enactment o f the Free Womb Law, Bra
zilian officials prosecuted the owners o f slave prostitutes and removed their
slave property in the city's campaign to eliminate forced prostitution in gen
eral.14 Lauderdale Graham reports a case in which a slave woman named
Honorata found allies among her clients for her charges against her widowed
mistress, M aria Elenteira. Three o f Honorata's clients, a retired sergeant
and two casual laborers, all Brazilian citizens, collaborated in court and tes
tified that Honorata had been forced to engage in prostitution.15
The Free Womb Law made another important change among the urban
slave population o f Salvador. Now that any child born to a slave mother was
legally declared free-bom , some common-law slave spouses chose to con
tract church matrimony. In the national census o f 1872, those who had been
married (registered as married as well as widowed) constitute 14.8 percent
o f men and 12.5 percent o f women among slaves; these figures are much
The Labyrinth o f Identity 1851-1888 145

lower than the 35.3 per cent o f men and 17.0 per cent o f women among the
legally free population (both free-born and freed).16 However, considering
that there were virtually no slave marriages reported for the previous years
(zero, at least until 1855), this is still a remarkable increase. The percentage
o f married slaves in Salvador was even higher than that for the overall Bra
zilian slave population (8 percent),17 and the percentage o f married slave
women (12.5 percent) had grown close to those in Bahia (14.9 percent).18The
percentages o f the married among the population o f African descent in Bra
zil furthermore rose significantly after the abolition o f slavery: from 14 per
cent for men (slaves, 8 percent; free, 20 percent) and 15 percent for women
(slaves, 8 percent; free, 21 percent) in 1872, to 24 percent for men and 25
percent for women in 1890. These increased percentages suggest that slave
status was indeed a critical obstacle for church marriage and that legally free
status created more opportunity for ex-slaves to contract matrimony. How
ever, the change in the percentages is also attributable to a tendency in the
1890 census to classify coresidence o f ex-slave common-law spouses as mar
riages.19 It was in the very year o f 1890 that civil marriage was finally recog
nized as legal in Brazil so that consensual unions without holy matrimony
might be counted as marriages.
Changes in the T erm inology o f C olor
The Brazilian-born population o f African decent had been always

labeled/categorized and divided by color distinctions, but in the early 1870s
the official usage o f multiple color distinctions started to change. Unlike the
preceding local censuses o f Salvador, and despite the com m on distinctions
o f color in colonial and imperial Brazil, the first national census o f 1872 di
vided the legally free Brazilian-born population o f African descent into only
two dichotomous categories o f color: prito (black) and pardo (brown). At
least in this new classification for census taking, the pardo population in the
city o f Salvador outnumbered pritos by 2:1 among the entire population o f
African descent. This is attributable to the rapid growth o f the free pardo
population, which constituted more than 70 percent o f the free population
o f African descent. The pardo population had grown also among the slave
population, but nearly two-thirds o f slaves were still classified by the color
o f prito.20 O f course, we do not know exactly how the census taker identi
fied an individual’s color, but most likely census taking relied largely on self
declaration, at least by legally free individuals. It is worth emphasizing that
the majority o f the Brazilian-born free population o f African descent iden

tified themselves as pardos, declining to be regarded as prêtos; being identi
fied and/or classified as prêto generally indicated being socioeconomically
close to slave status. But it does not mean that all o f the self-declared “mu
latto” population was regarded as such, and not as blacks, by the larger so
ciety. Since there existed no subjective criterion for the identification o f the
individual’s color, people tended to self-identify w ith the color the larger
society preferred. Therefore, based on their legally free status, free people o f
African descent in nineteenth-century Salvador were likely to choose pardo
over prêto as their color, beyond visible skin colors and other ostensible
phenotypes.21
By the beginning o f the 1870s, the folk term inology o f color in Salvador
also had begun to change to a considerable degree. The term crioulo, which
specified Brazilian birth and the color black, had disappeared.22 O f course,
now that the majority o f the population o f African descent in Salvador was
Brazilian-born, there was not really a point in specifying the individual’s
birthplace. For the same reason, the term prêto, which no longer referred
specifically to African-born, either enslaved or freed, began to refer merely
to the skin color o f black.23 At the same time, the term negro, which had
referred to the enslaved African-born and had mainly been used in official
correspondence and documents (but not in any notarial records), gradually
came to identify the general black population, regardless o f birthplace or
legal status. O f course, the word negro would have been taken as offensive
when it was used to address people w ho did not identify themselves as
blacks. For instance, on April 8,1872, a self-identified white man, Aristides
Ricardo de Santana, a typographer, native o f Salvador and resident in Sâo
Pedro parish, filed criminal charges against Dr. Eloy Martins de Souza and
his son Braulio Soares Martins de Souza, residents o f Vitôria parish, and
against Joâo Tavares da Silva Godinho, native o f Portugal, as their accom
plice, for their having publicly shouted at Santana and called him a negro.24
One may now wonder how free Brazilian-born "blacks” came to identify
themselves in a changing context during the last decades o f the slavery re
gime. The Society for the Protection o f the Needy provides an exemplary
case for our discussions.
T h e Creation o f Racial Identity:

T h e Society for the Protection o f the N eedy
The black sodality o f O ur Lady o f Solitude and Support o f the

Needy, which had been established in 1832 as one o f the two new free-born

crioulo sodalities (see chapter 6), transformed itself in the year o f 1851 into
a new type o f mutual-aid association named the Society for the Protection
o f the Needy (Sociedade Protectora dos Desvalidos). Interestingly, this his
torically significant transformation o f the crioulo sodality took place in the
very same year that the transatlantic slave trade was officially terminated,
although we w ill never know whether the officeholders o f the Society in
deed chose the tim ing or it was merely coincidental.
The Society for the Protection o f the Needy has maintained strong ties
with the black church o f O ur Lady o f the Rosary in the Pelourinho from
the transfer o f the headquarters o f its predecessor, the black sodality o f O ur
Lady o f Solitude and Support o f the Needy, to the church on December 17,
1848 until the present day. O n M ay 10,1868 the Society eventually moved
out o f the church, because o f disagreements with the black sodality o f O ur
Lady o f the Rosary over use o f the church, and rented a place o f its own on
Rua do Bispo. The Society has been located since 1877 at Cruzeiro de Sâo
Francisco, no. 17, near the Church o f Saint Francis, but has continued to
have strong ties with the black church o f O ur Lady o f the Rosary in the
Pelourinho.25 Until today the majority o f the members o f the Society for
the Protection o f the Needy have also been members o f or associated with
the black sodality o f O ur Lady o f the Rosary in the Pelourinho. For instance,
Sr. Aloisio Conceiçâo Rocha, the former treasurer and longtim e member o f
the Society, has a cousin (Padre Rocha) who is the priest o f the black church
o f O ur Lady o f the Rosary in the Pelourinho.26
It was on August 26,1873 that the statutes o f the Society for the Protection
o f the Needy were finally approved by the governor o f the state o f Bahia.
W hile there was no restriction on the number o f members, the Society lim
ited its membership to “all Brazilian citizens whose color is black (prêto),
who have sufficient income for subsistence, and who are neither under the
age o f eighteen nor over fifty.” A member was to pay 158000 réis as an en
trance fee, and i$ooo réis in monthly dues.27 Small wonder the statutes o f
the Society did not employ the term crioulo, which had already disappeared
from the folk term inology o f color by the turn o f the 1870s. It is especially
noteworthy that the term negro, which used to refer to both African birth
and the enslaved status, was not used, either; instead the term prêto, which
had come to refer to color in the official terminology, as in the case o f cen
suses, was chosen to separate themselves from the majority o f free people
o f African descent, who preferred to identify themselves as pardos. Further
more, unlike the case o f all black lay sodalities in Salvador, the Society for
the Protection o f the Needy surely did not depend on well-to-do whites fi

nancially, nor even allow them to become members. Obviously exclusion o f
well-to-do whites who could have made additional contributions from the
membership cost every member much higher entrance and monthly fees,
but it seems to have been worth creating a racially exclusive voluntary as
sociation that represented their own specific interests.
Whites were not the only group that the Society for the Protection o f the
Needy excluded from its membership; it did not allow either mulattoes or
people o f African birth to join. W hat do these seemingly strict exclusions
mean? W hy did the members o f the Society for the Protection o f the Needy
choose to identify themselves as those whose color was “black” (prito)7. As
we discussed earlier in this chapter, being perceived as “black” was almost
synonymous w ith being a slave in nineteenth-century Salvador. Therefore
many o f the free Brazilian-born population o f African descent, particularly
those who attained a higher socioeconomic standing with specific occupa
tional skills, naturally wanted to identify themselves not as “blacks” but
“browns” in the case o f census taking. True, ostensible skin color indeed
mattered for one's social standing, and the individual choice o f color was
largely limited by a person’s phenotype, although one’s self-identification
did not always correspond with the larger society’s perception o f the per
son’s color. Having a lighter skin color was a great advantage in Brazilian
society, and preference for it was widely prevalent among the Brazilian-born
population o f African descent, either slave or free. In 1855, for instance,
James Wetherell reported in Salvador that am ong the “Creole blacks,” a
child who was whiter than his or her mother was looked upon with pride.28
It is very likely that the special emphasis o f the Society for the Protection o f
the Needy on the color “black” in its membership represented some free-
born blacks’ collective racial frustrations, caused by the larger society’s
unchanged perceptions o f blackness, regardless o f their newly acquired
socioeconomic positions (and financial resources). Such individuals o f A f
rican descent would be tolerated individually by the larger society as self
identified “ browns,” in accordance with their class positions, but the mean
ings o f collective blackness would never change; despite their individual
social color o f “brown,” they would continued to be treated as “blacks” on
many occasions because o f their race. Forming their own “black” associa
tion, whose membership was exclusive by race and class, they thought to be
their effective means to (re)define their blackness in their own terms, not in
the view o f the white dominant society, thereby creating a new racial iden
tity at a collective level.
The establishment and successful development o f the Society for the Pro

tection o f the Needy signifies the emergence o f a collective racial conscious
ness among some o f the Brazilian-born population o f African descent after
the mid-nineteenth century. Represented among the membership were a
variety o f artisans and other skilled urban workers such as musicians and
barber-surgeons, but there were no domestic servants, who were very poorly
paid.29 Furthermore, the Society restricted the age o f members from eigh
teen to fifty, when male workers could be professionally most productive.
The Society for the Protection o f the Needy was a new type o f mutual-aid
association exclusively for men w ho were socioeconomically stable, with
specified occupational skills. Unlike the case o f lay sodalities, the Society
did not provide any funeral services or any other charitable activities for
members.
The exclusive character o f the Society for the Protection o f the Needy was
also very noticeable in its treatment o f women and children. The statutes
obliged every member to “accompany the entrance o f his wife, mother, and
children; otherwise, he would be fined i$ooo rixs.”30 Here we see the same
kind o f concern about family as in the mutual insurance society established
within the new free-born crioulo sodality o f Saint Francis o f Paula (estab
lished in 1842), as discussed in chapter 6. Women were not entitled to hold
memberships in the Society for the Protection o f the Needy and were in
cluded only marginally, in the same way children were, and they were re
garded as an “appendix,” either as the wife or mother o f a full member, who
had to be a “black” male. It m ay suggest reconstruction o f gender relations
among the free-born population o f African descent in the late nineteenth
century. Unlike slave women, especially those o f African birth, who engaged
in the informal market in the city, most free Brazilian-born women o f A f
rican descent worked as full-tim e domestic servants in their employers*
houses, while living with their common-law husbands and children. The
Society for the Protection o f the Needy represented the interests o f the free
“black” household that was headed by a man who held a stable employment
with his artisanal skills. This may suggest that, after mid-century, not only
among African-born ex-slaves, as seen in their “return” movement to Africa
(chapter 5), but also among other groups o f people o f African descent, the
family had become a much more important institution in individual deci
sion making. This contrasts with slave uprisings and the M ali revolt in the
early nineteenth century, in which participation was largely limited to men
(chapters 3 and 4).
W ith rare exceptions, the Society for the Protection o f the Needy did not
accept women as full members until around 1950.31 Present-day officehold

ers o f the Society, who are longtim e members, insist that the Society did not
discriminate against women as regular members, and ascribe the absence o f
women’s formal participation to their “ historical lack o f socio-political con
sciousness since “most o f them were domestic servants.” O f course, being
domestic servants does not necessarily indicate women's lack o f any collec
tive sociopolitical consciousness. The present-day officeholders’ statement
on gender roles in the late nineteenth century does reflect much o f patriar-
chalism prevalent in Brazilian culture. Yet it may enable us to consider re
construction o f gender relations among the population o f African descent
after about the mid-nineteenth century. Whereas young male slaves contin
ued to be sold into the interprovincial slave trade, the city constantly de
manded female domestic servants. Unlike slave women, especially those o f
African birth, maids were not hired out to the street for wage earning. Free
Brazilian-born women o f African descent were full-tim e domestic servants
in their employers’ households, and they had to devote their spare time to
taking care o f their spouses and/or children at home. Women became more
limited in their daily mobility, and no longer moved beyond the boundary
between the “house” and the “street” as freely as before, as discussed in the
case o f African-born slave women in chapter 2. Am ong the free-born popu
lation o f African descent, the urban space may have come to be divided
largely by gender into tw o distinctive spheres: the “street” for men, and the
“ house” for women.
The Society for the Protection o f the Needy has attempted to participate
in local politics, with mixed results. For instance, a longtime member be
came a perennial candidate for election to the city council. Com m itted to
developing its collective racial consciousness, the Society has been very elo
quent on the strong presence o f racism in present Bahian society, which is
still often denied by many Bahian intellectuals. Unlike the situation during
the late nineteenth century, when it excluded the African-born population
from its membership and was proud to be a voluntary association exclu
sively for “black” Brazilian citizens, the present-day Society for the Protec
tion o f the Needy has been promoting Africanism among the present popu
lation o f African descent in Salvador by welcoming ambassadors and other
visitors from African countries. One older member o f the Society traveled
to Angola in the early 1980$. Granted that Afro-Brazil in general has been
developing an extrem ely complicated and ambiguous relationship with
continental A frica, the Society’s new “ Pro-Africa” agenda has been per
ceived by a younger generation o f Afro-Brazilian activists as a “redoubt for
an older generation not on the frontline o f protest.” But at the same time,

African ambassadors and other international visitors are “more likely to be
‘taken* to the Society’s premises”32 In fact, the president and a few other
longtime members o f the Society, based on the “ international” nature o f
their association, asked the author, a Japanese national, to make connections
with the Japanese government and major foundations for the financial sup
port o f the Society.
The Society for the Protection o f the Needy has remained a predomi
nantly black male association, although its current membership is officially
open to any Brazilian citizen, regardless o f race, color, or gender.33 Despite
the continuing success o f their “black” association, the members o f the
present-day Society for the Protection o f the Needy continue to struggle to
identify themselves as blacks {negros). W hile they are very proud o f their
own unique association, they cannot help but reveal their feelings o f being
caught between their collective self-assertion as “blacks” and the larger so
ciety’s perception o f them as “blacks’*based on their phenotypes. One day
during my frequent visits to the headquarters o f the Society for the Protec
tion o f the Needy in 1989, 1was told by several longtime members, including
the president, that the criterion for membership during the late nineteenth
century had been the “color o f the gums.” They had to be “black,” because
those with “pink-colored” gums were regarded as mixed, not blacks. The
elderly officeholders eagerly showed me their “pink-colored” gums and told
me that they would not have been qualified for membership back in the
nineteenth century; they would have been mestizoes (mixed).34 Needless to
say, this “racial” criterion is by no means correct, but this story reveals the
members’ ambiguous “black” identity. They virtually insisted that at least
during the nineteenth century they would not have been naturally classified
as “blacks,” who were not favored in the larger society and constituted a
lower social stratum than the mulatto population; and even today they may
well change their social color to “mulatto” because o f their education and
higher class standing than the masses. O n a separate occasion for an infor
mal chat, one o f these longtime members, in front o f a few other members,
insisted on the presence o f strong racism in present-day Salvador, and told
me that whenever he made friends with a white man, he was always ad
dressed as a negro by his friend, despite the fact that they shared the same
socioeconomic standing. In other words, the members o f the Society insist
that they have become “blacks” by choice; but they are by no means natu
rally “black” in accordance with the larger society's pejorative perception o f
race. We should not fail to note that those who choose to identify themselves
as “blacks” generally distance themselves from the poor majority, who are

by all means labeled as “blacks” by the larger society, not only for their
phenotypes, but also for their poverty, and who do not always identify
themselves as such.
Conclusion
The term inology o f color in Salvador changed in relation to the

rapid decline o f the institution o f slavery, opportunities for freedom, dis
appearance o f the aging population o f African birth, and expansion o f the
free Brazilian-born population o f African descent. In the last tw o decades
o f the slavery regime in Brazil, the meanings o f color were redefined in the
new context o f free labor.
All in all, color is indeed a legacy o f African/black slavery in present-day
Salvador, as well as in all o f Brazil, where color distinctions indicate not only
visible differences in physical appearance derived from non-European an
cestry, but also individual socioeconomic standings. In short, color in Brazil
is a labeling/categorization o f the individual by others (i.e., Brazilian so
ciety’s view), m ainly based on phenotype or ostensible “genetic” character
istics, such as skin color, hair, lips, and nose, which do not im ply any exact
genealogy or ancestry. In order to clarify this point, Daphne Patai, in her
book Brazilian Women Speak: Contemporary Life Stories (1988), transcribes
her conversation w ith tw o sisters in their early thirties, both o f whom are
highly educated feminists. The sisters were born to the same father and
mother in a racially mixed fam ily in the city o f Rio de Janeiro. Because
o f the fam ily’s long history o f miscegenation, the sisters happened to in
herit different sets o f physical features: the older sister Alma is completely
“white” in appearance, while the younger sister Julia’s phenotype is regarded
as “black.” Alma and Jtilia have been treated accordingly by the larger so
ciety and even within the family, which has created racial cleavages between
the two sisters.35These terms o f color are, o f course, a representation o f Bra
zilian culture and are ambiguous even w ithin Brazilian culture, as Patai
points out.36
Interestingly, this color label attached to the individual is changeable to
a considerable degree according to the person’s socioeconomic status. In his
Pulitzer Prize-winning book, Neither Black nor White: Slavery and Race Re
lations in Brazil and the United States (1971), Carl N. Degler observes that in
Brazil, where there is no definite line between blacks and whites because o f
the existence o f the large mulatto population, it is possible for the individual
o f color to raise his or her socioeconomic status and therefore eventually to
The Labyrinth o f Identity, 1851-1888 153

change (whiten) his or her social color.37 In short, the color by which one is
labeled by others is the indication o f one's socioeconomic position w ithin
the racially stratified society and in relation to others with different physical
characteristics. Individuals o f color, whose self-perceptions are inevitably
limited by the larger society’s view, maneuver to manipulate social classi
fication for the creation o f racial identity. This sociocultural mechanism
to determine one’s racial identity with public affirmation is well exempli
fied by the anthropologist Virginia R. Dom inguez in her provocative book,
White by Definition: Social Classification in Creole Louisiana (1986).3*
Generally speaking, being perceived as “black” in Brazil means being lo
cated at the lowest edge o f the socioeconomic ladder, with relatively little
formal schooling. C olor continues to carry social stigmas resulting from the
long history o f African/black slavery in Brazil.39 People o f African descent,
both individually and collectively, continue to suffer from the dishonor o f
having ancestors w ho had been enslaved, and are forced to fight various
forms o f prejudice and discrimination. Two U.S. journalists published in
1994 an illum inating episode in a Dallas newspaper article focusing on the
issue o f race in Brazil. In 1993 Daniel Sacramento ran for city council o f Sal
vador; his platform was to improve jobs, housing, and security for “blacks.”
His sister, who is a professor o f sociology at the Federal University o f Bahia
and campaigned for him, remembers a typical remark from a woman after
entreaties to vote for him: “I am not voting for him. He is black” despite
the fact she herself is phenotypically black, at least in the eyes o f those two
U S. journalists.40We, o f course, do not know if this woman, as well as other
prospective voters who gave a similar response to the candidacy o f Mr. Sac
ramento, would identify herself as black. She is very unlikely to do so since
she is literate and aware o f her voting power, which means she is well above
the poor majority. In Brazil, to be black means to be poor; it carries a con
notation that goes beyond indicating merely that this individual has differ
ent phenotypical characteristics derived from African ancestry.41 In Salva
dor, as well as in the entire nation o f Brazil, it may well be that blackness is
still largely “dissembled and hidden”42 and is perceivable only to those who
seek to see it.
The creation o f a collective racial identity by the members o f the Society
for the Protection o f the Needy in Salvador during the late nineteenth cen
tury exemplifies the omnipresent nature o f the short-lived movimento ne
g r o or black movement, which modern Brazilian history has repeatedly w it
nessed. The movimento negro, which began with the Frente Negra Brasileira
(Brazilian Black Front) first established in the city o f Sao Paulo (1931-1937),

has evolved over the years through more prominent incarnations, such as
the Teatro Experimental do Negro (TEN; Black Experimental Theater) o f
the 1940s and 1950s, “ Black Soul” o f the 1970s, and the Movimento Negro
Unificado (United Black Movement) o f the 1980s and 1990s.43 The movi
mento negro is an umbrella label that includes various kinds o f political
empowerment movements by various groups o f Afro-Brazilians that have
taken place in different locations at different times over the years.44 In this
long process o f social mobilization, Afro-Brazilian activists have had to con
struct the negro as a new political subject; the negro has come to be defined
as a socially constructed category that encompasses different colors and em
braces the shared African ethnicity for the sake o f political empowerment.45
Although the term negro has been thus employed in all o f these movements
as the “signified/signifier o f a race-conscious subject (equivalent to African-
American),” there has been no clear consensus on the definition o f the ne
gro even among the Afro-Brazilian activists.46 At the same time the leader
ship o f the movimento negro has been predominantly middle-class, which
George Reid Andrews, in his monograph titled Blacks and Whites in S&o
Paulo, Brazil, 1888-1988 (1991), notes as “class exclusivity” in the case o f the
city o f S3o Paulo during the twentieth century. Andrews has documented
middle-class Afro-Brazilians1 various political attempts to set themselves
apart from poor and working-class Afro-Brazilians, many o f whom have
refused to identity themselves as negros (blacks).47
The same kind o f alleged “class exclusivity” has been observed in the
commercial success o f the new biweekly Brazilian magazine Ra$a Brasil
(Brazil Race). In 1994 the first issue o f Ra$a Brasil sold out immediately at
every newsstand in southern Brazilian cities. This made a sensational head
line, and the magazine soon received international attention. Ra$a Brasil has
been modeled after the U.S. magazine Ebony; and it advocates the beauty o f
“ blackness,” which is represented in their unique hair, body, dress codes,
sexuality, and so on. We should not fail to note that Ra$a Brasil has been
“criticized for its advertisements on skin lightness and light-skin models,”
as in the case o f their U.S. counterpart Ebony,l48 But, o f course, Ra$a Brasil
cannot escape Brazilian realities and customs, which persistently negate
blackness and encourage individuals o f color to change their “colors” in ac
cordance w ith their better socioeconomic standings. The magazine has sur
vived financially to this day, w ith the support o f urban middle-class Afro-
Brazilians, who have desired to give positive affirmations to their collective
blackness, while at an individual level being strongly tempted to shed their
race through “whitening ” In fact, many o f these white-collar Afro-Brazilian

readers are lighter-skinned, and most o f them would rather pass as mulat-
toes in the larger society.49 O n the other hand, the poor majority, all o f
whom are naturally darker-skinned, largely lack a collective racial con
sciousness and identity, while overlooking/denying the presence o f preva
lent racial prejudice and discrimination against peoples o f color in Brazilian
society.50O nly with sufficient financial resources have Afro-Brazilians been
able to make attempts to create a positive black identity, albeit by inevitably
excluding the poor majority. In other words, class has been used by such
Afro-Brazilian activists as a shared springboard for the creation o f a collec
tive racial identity. Ironically this is exactly the same mechanism, albeit with
the opposite effect, as in an old, yet very popular, Brazilian saying: “Money
whitens [color] ” M oney gives an individual the power to choose his or her
color, even as black.
But, sadly, the story does not end here. As long as whiteness signifies
power and blackness is identified w ith poverty, many o f the self-declared
Brazilian “blacks” find themselves being caught between their politically
constructed racial identity and the individual social “color.” Such internal
dilemmas have been m aking every movimento negro extremely fragile and
vulnerable, with the critical lack o f a long-lasting group racial solidarity.
In Brazil, class and color have been closely intertwined with each other;
middle-class Afro-Brazilians have been allowed to pass individually as mu-
lattoes or even whites in Brazilian society, thereby making blackness asso
ciated forever with poverty only.
Afro-Brazilians have created and re-created their distinctive identities
over the years from the beginning o f the slavery regime until the pres
ent day. And identity, either individual or collective, is indeed a choice in
relation to others. But, whichever color (black, mulatto, or white) Afro-
Brazilians collectively choose as their common racial identity, as long as the
mechanism remains the same for the exclusion o f the lower-class and un
derclass Afro-Brazilian masses, they continue to be caught w ithin their own
labyrinth o f identity. They may utilize their financial resources to create a
collective racial identity only by distinguishing and isolating themselves
from the poor Afro-Brazilian majority, who deny their blackness and lack
collective racial consciousness. Sadly, Afro-Brazilians1 movements for col
lective political empowerment have been demonstrating their fatal limita
tions, which derive from the am biguity o f race, fluidity o f color categories,
and negative connotations o f blackness, all o f which have been prevalent in
Brazilian culture over the years.

Conclusion
Slavery surely changed individuals as well as communities. But this study

o f slavery in the major Atlantic port city o f Salvador, Brazil, during the
nineteenth century reveals complex patterns o f change in identity. Slavery
profoundly changed perceptions o f ethnicity, gender, and race.
To this day people o f African descent have always predominantly popu
lated Salvador. This is attributable to the long history o f large-scale ur
ban slavery in combination with the continuing massive influx o f African-
born slaves to Salvador that lasted from the late sixteenth century to the
mid-nineteenth century. During the nineteenth century, many international
travelers from Europe and North America, undoubtedly all whites, took spe
cial note o f the visible prominence o f the large black population in the city
and described them as a people who naturally shared the same collective
identity. But such an ostensible homogeneity in the eyes o f outsiders was
oftentimes quite misleading. All those who were categorized by others un
der a single label/category did not always share the same identity; individual
actors o f African decent in nineteenth-century Salvador created and re
created distinctive ethnic, gender, and racial identities in relation to politi
cal, socioeconomic, and cultural changes in the larger society.
T h e Creation o f Ethnic Identity
Enslaved Africans created their ethnic identities in the context o f

New World urban slavery. For the first several decades o f the nineteenth
century, two-thirds o f the slave population in Salvador was African-born,
with diverse ethnic backgrounds. Such enslaved peoples o f African birth
were labeled/categorized by the larger society for the purpose o f slave con
trol; they were reduced into a certain number o f African “nations,” which
did not always correspond to their original ethnic identifications back in
continental Africa. T he enslaved African-born population was quick to
learn how to manipulate these labels/categories o f African “nations” for the
creation o f ethnic identity. Although they were known in the slave society
by their newly given Portuguese/Christian names, among themselves the
African-born population continued to speak their native African languages,
while identifying themselves and referring to one another by their original
African names.
Sharing o f the same ethnicity became a solid foundation for the African-
born population’s daily gatherings and groupings on the streets. Ethnic
identity functioned as a determinant factor in the African-born slave’s
choice o f a partner or in parents’ choice o f their Brazilian-born child’s god
parents) for baptism. It was also well represented in the establishment and
formation o f voluntary associations, as well as in the frequent occurrence
o f slave flights and uprisings.
Already by the mid-i830s, the African-born enslaved population’s diverse
ethnic identities began to converge into a broader ethnic identity, based on
their common place o f birth, Africa. This significant transformation o f d i
verse ethnic identities manifested itself in the course o f the M a li revolt
(1835), in which several hundred African-born men, both slaves and ex
slaves, were united under the common symbols o f Islam and their Yoruba
ethnicity. The larger society started to perceive the population o f African
birth as a most serious obstacle to social order. In the late 1830$, as a result
o f societal pressures and in the face o f new legislation that limited the status
o f the African-born ex-slave population, more and more o f that population
voluntarily emigrated, typically in a family unit (common-law spouses o f
African birth and their Brazilian-born children), to Africa, sometimes to
their original hometowns. At the mid-nineteenth century, the very same
year that the transatlantic slave trade was officially terminated, African-
born ex-slaves’ repatriation movements from the New World (i.e., Salvador)
to the Old World (i.e., Africa) converged into their collective settlement in
the newly established British colony o f Lagos, in what is the present-day N i
geria. In Lagos, such African-born “ returnees” (and their Brazilian-born
children) from Salvador separated themselves from the local, indigenous
populations as well as from those o f African birth who had returned from
other parts o f the New World, such as Cuba, and identified themselves as
“Brazilians.” Thus the creation o f ethnic identity also took place back in the
Old World.
A ll in all, in slavery and freedom, enslaved Africans did not identify
themselves exclusively in terms o f race, and their creation o f identity in the
New World took place primarily within the parameters o f ethnicity as well
as gender. By re-creating strong ethnic identities and maintaining various
ethnic continuities w ith their original African cultures, the African-bom

population in Salvador demonstrated remarkable resilience against the odds
o f New World urban slavery.
T h e Creation o f G ender Identity
Enslaved Africans' creation o f distinctive gender identities under

mined the further development o f their collective ethnic identities. W hile
the urban African-bom slave population was balanced by sex, w ith women
even slightly outnumbering men, such distinctive features o f urban slavery
in Salvador as wage earning and the small scale o f slaveholding determined
important ways by which the African-bom population created their gender
identities. African-bom slave men under different ownerships were usually
hired out to the streets to work in gangs as porters and transporters. Those
who shared the same ethnicity were arranged to w ork in the same gang
mainly for their owners* convenience, but that this had the unintended con
sequence o f m aking a significant contribution to the strong combination o f
gender w ith ethnic identity among the male slave population o f African
birth, which developed beyond the geographical urban/rural boundaries,
cannot be denied. By contrast, A frican-bom slave women were widely en
gaged in the informal market o f the city, associated with one another across
ethnic boundaries, and developed their collective gender identity in common
without sharing the same ethnic identity. Small wonder that few African-
born slave women participated in ethnic slave uprisings, which were a pre
dominantly male phenomenon. We also take note that African-bom slave
women, particularly in the case o f market women, gained much more finan
cial resources than many A frican-bom slave men who worked as manual
laborers on the streets. Their financial resources, for instance, enabled them
to hold membership in black lay sodalities, although it still did not allow
them to become officeholders. W ithin the context o f New World slavery and
in the specific case o f Salvador, Bahia, the common pattern w ithin which
African-born partners formed temporary or long-lasting bonds w ith mates
but did not coreside further strengthened African-born slave women as they
(re)created their distinctive gender identity.
This creation o f gender identity took a completely different form in the
Brazilian-born slave population, bom and raised exclusively w ithin Luso-
Brazilian culture. They created their gender identities often in their close,
interpersonal relationships w ith their owners. Brazilian-born slaves were
manumitted most often without payment, either conditionally or uncondi
Cottclusion 159
tionally, and Brazilian-born women were freed with much greater frequency
than their male counterparts. Yet most o f these Brazilian-born ex-slave
women remained in their former owners’ households, in which they contin
ued to work as quasi-slave domestic servants. Unlike the case o f African-
born market women, Brazilian-bom ex-slave women were unable to move
much on the socioeconomic ladder on account o f their financial resources.
W ith the expansion o f the free-born crioulo (Brazilian-born black) popu
lation in accordance with the rapid destruction o f urban slavery after the
mid-nineteenth century, gender relations were redefined and gender identi
ties were re-created w ithin the new framework o f a two-headed household
now that common-law spouses coresided, with their children. Men became
the household heads and breadwinners for their families, while women, who
worked as full-tim e domestic servants in their employers’ households, had
their gender roles re-created in relation to their domestic roles, largely as
wives and mothers.
T h e Creation o f Racial Identity
The Brazilian-bom population o f African descent separated them

selves from their African-born counterparts and identified themselves pri
marily in terms o f race, but this racialization does not mean that they all
shared a single racial identity. For, either slave or legally free, they had al
ways been divided into various color categories according to the individual’s
skin color and other phenotypes. As for the legally free population, color
functioned as an index o f one’s socioeconomic status, for instance, for the
choice o f a marriage partner. Slave owners always preferred pardos (browns;
mulattoes) over crioulos (blacks). Naturally the former had always had a
much better chance in manumission and separated themselves firom the lat
ter, the majority o f whom had always been slaves, and established their own
mulatto sodalities that excluded blacks, namely the African-born and crioulo
populations. Over time the free mulatto population came to constitute a dis
tinct social stratum, and their racial identity was represented in the collec
tive actions they took in the Tailors’ Revolt in 1798. They had been strongly
influenced by Enlightenment ideas in the conceptualization and the execu
tion o f the Revolt, which was an all-men event.
By contrast, the free crioulo population had been allowed to enjoy only a
very limited amount o f upward social mobility, most notably through mili
tary recruitment into the black regiments. As free black men, they were not
located at the very bottom o f the slave society, but as black men they were

very close to it. The presence o f the black militia regiments in the slave so
ciety was designed to maintain the social order and hierarchy based on racial
slavery.
The creation o f racial identity is exemplified by the development o f the
Society for the Protection o f the Needy during the last decades o f the slavery
regime. The Society had been established in 1832 as one o f the tw o new free-
born crioub sodalities and subsequently transformed itself into a new type
o f voluntary association in 1851. Its statutes were approved by the govern
ment o f Bahia in 1873. The Society for the Protection o f the Needy turned
out to be the first exclusively "black” voluntary association in Brazil that did
not extend membership to whites and mulattoes, nor was it open to the
African-born population, which had been disappearing in any case. The
statutes limited membership o f the Society to male Brazilian citizens, aged
eighteen to fifty, whose skin color was prêto (black) and who possessed
the requisite means and, therefore, could afford to pay the relatively high
registration and membership fees. Interestingly, for self-identification they
avoided using the term negro, which referred to the general Brazilian-born
black population from the 1870s. Most o f the members o f the Society were
skilled artisans who could earn stable salaries; they did not include domestic
servants, who were poorly paid. W ith their newly emerging financial power,
these men chose not to change their color from “black” to “mulatto”; in
stead they (re)created their collective racial identity as “blacks" by separat
ing themselves from the poor black majority. Only improved socioeconomic
standing made one’s choice o f color possible. As this creation o f a “new”
identity shows, class exclusivity was a noticeable characteristic o f Brazilian
racial identity then and now.
* * *
Individuals who appear to be homogeneous and undifferentiable to oth

ers, either as a group or community, or who are labeled by the larger so
ciety under a single category because o f their ostensible shared sameness in
phenotype and/or color do not necessarily share a common identity. Indi
viduals create and re-create their identities in relation to other people’s per
ceptions and categorization o f them, and the creation o f identity invariably
takes place within the context o f power relations.
In the New World, particularly in Brazil and in the antebellum American
South, African/black slavery developed and prevailed as a major socioeco
nomic institution, and, as a result, race and class have been intricately inter
Conclusion 161
twined w ith each other in the establishment o f power relations and struc
tures and their maintenance. T he face o f power has been white because o f
the legacy o f racial slavery, although the definition o f whiteness varies from
one culture to another. Nonetheless, all the non-white populations in the
New World have been regarded as a social and political minority, and their
minority status has been marked by their visible otherness.
In Brazil, African/black slavery prevailed and flourished as a national in
stitution for the country's agro-export economy for more than three centu
ries. The non-white population had always been a majority in number from
the early colonial period throughout the nineteenth century. In the case o f
nineteenth-century Salvador, which this book treats extensively, some 70
percent o f the whole population was classified as people o f color or non
white. The white elite had to establish some sophisticated mechanisms by
which only they could recruit a select, elite group o f the population o f A f
rican descent as individuals, through the political practice o f clientage and
patronage, into the power systems and permit them, as mulattoes, not as
blacks, to move up on a social ladder.1 At the same time, the larger society
divided the majority o f the non-white population into various categories,
such as "nations” (for African-bom ) and colors (for Brazilian-born), which
effectively hampered them from creating a common collective ethnic/racial
identity. T his book has demonstrated that the creolization o f the Afro-
Brazilian population did not automatically result in the historical creation
o f a single, homogeneous racial/ethnic identity. Afro-Brazilians have ma
nipulated various labels/categories in relation to their skin colors and other
phenotypes, as well as their class positions, and have individually played
politics o f identity within their socio-racial/ethnic group designations.
By contrast, in the United States, slavery did not become a national insti
tution; it evolved only regionally w ith the development o f the plantation
economy in the antebellum South, with a much smaller number o f slave
imports from Africa, approximately one-ninth as many as entered Brazil.2
African Americans have always constituted a numerical minority, whose
presence as a collective could not be perceived as a critical numerical threat
to white supremacy in the colonial society. For the sake o f social control, the
larger society has lumped together all people o f African ancestry and clas
sified them as blacks by disregarding other ancestries they have also inher
ited, such as European, Native American, and Asian, although, as Melissa
Nobles's study demonstrates, the official category o f "mulatto” was not re
moved from the U.S. censuses until 1930.3 In the United States, historically,
having “one drop o f blood” o f an African ancestry autom atically deter

mined one’s race as black, no matter how white one appeared to be. Indi
vidual variations in color and other phenotypes among African Americans
have been treated mainly as a personal preference. In the late 1980s and early
1990s the U.S. society observed a crucial change in its racial categorization
w ith the advancement o f the multiracial movement; people increasingly
questioned the long-standing, rigid racial categories solely based on single
ancestors and raised many issues concerning multiracial identity.4 W hile it
became common for us to see in the United States the category o f “others”
for those who inherit more than one ancestry (and therefore are unable to
identify themselves w ith one o f the traditional rigid racial/ethnic catego
ries), the rule o f thumb has not changed at all. In the United States, African
ancestry continues to play the central role in determining one's (either sin
gular or multiple) racial identity.
* * *
Since the late fifteenth century, which was marked by the emergence o f
the Portuguese empire, the rapid expansion o f world capitalism has moved,
relocated, and marginalized a great number o f people as a forced (enslaved
and indentured) or voluntary (immigrant) labor force. As a result, world
history has repeatedly witnessed the formation o f various diaspora com m u
nities, these geographically diverged and culturally displaced peoples' crea
tion o f new identities as minorities in the larger societies, and their repa
triation movements to their homelands in search o f their “true” identities.5
An example is the “emancipated Africans” in Sierra Leone, who migrated to
Jamaica as indentured servants, and ultimately “ returned” to Africa, as stud
ied by historian Monica Schuler for the period 1846-1865.6Another example
is reported by anthropologist George Gmelch in his collection o f narratives
by thirteen Barbadians who have “returned” from various places in England
and North America.7
Around the mid-1980s, from Brazil, as well as from Peru, Argentina, and
other Latin American countries, albeit on a much smaller scale, people o f
Japanese descent began to migrate on a large scale to Japan, whose boom
ing economy demanded a massive manual labor force. Such labor move
ments o f Japanese Brazilians to Japan have been called dekassegui in Brazil,
and the term has been incorporated into the Brazilian Portuguese vocabu
lary. Many Japanese Brazilian migrant workers were white-collar workers in
Brazil, often with college degrees, but chose to work as unskilled laborers
in Japan.8 Needless to say, their decision on this transnational labor migra
Conclusiort 163
tion from Brazil to Japan was made first and foremost for financial reasons;
they could easily earn wages several times higher than when working in
Brazil. But we should not fail to point out that at the same tim e Japanese
Brazilians' labor migration to Japan demonstrates a striking resemblance
to African-born ex-slaves* emigration from Salvador to West Africa. Japa
nese Brazilians, who have been categorized and identified as “ Japanese”
(japonisJjaponisa)t with pejorative racial/ethnic biases and stereotypes de
spite their middle-class position in the larger society, had long romanticized
Japan as their “homeland,” regardless o f their Brazilian birth and citizen
ship, and tended to identify themselves in ethnic terms as Japanese, not as
Brazilians. In Japan, in June 1990, new laws were enacted to grant permanent
residency to all non-Japanese citizens whose parents or grandparents were
o f Japanese nationality so that Japanese industry might attract and more
readily obtain cheap labor from overseas. This change in labor laws contrib
uted to the sudden great increase in the numbers o f Japanese Brazilian m i
grant workers who left Brazil. Ironically, upon arrival in Japan, Japanese
Brazilians, many o f whom look exactly like Japanese, have been regarded as
the “other” regardless o f the shared Japanese ancestry because o f their cul
tural differences and lack o f language skills. Faced by ethnic prejudice and
discrimination, Japanese Brazilian workers and their families have sepa
rated themselves from the Japanese and formed “ Brazilian” communities.9
Many have moved back to Brazil after having saved a substantial amount
o f money, which has been used typically for the purchase o f real estate
and business, but others continue to move back and forth between Brazil
and Japan, still being unsure o f their ambiguous Japanese Brazilian iden
tity.10Clearly these Japanese Brazilians' “repatriation” to Japan constitutes a
present-day, global-age counterpart to the historical experiences o f African-
born ex-slaves and their Brazilian-born children who took a passage from
the city o f Salvador to Africa and re-created their ethnic identity as Brazil
ians in Lagos during the late nineteenth century.11
This book has demonstrated that the creation o f identity by enslaved A f
ricans and their descendants in nineteenth-century Salvador, Brazil was, in
fact, not unique to its specific time, space, and historical context. In our
global age, w ith the expansion o f global markets, people and com m odi
ties are circulating beyond national boundaries, and as a result we have
been observing the rapid formation o f seemingly homogeneous com m uni
ties in the United States. T his is exemplified by the rapid emergence o f
non-English-speaking “ blacks” in Washington, D.C.; “Hispanics” in Miami,

Florida; and “Asians” in Atlanta, Georgia; all o f whom, in reality, have been
heterogeneous groups whose ethnic compositions continue to change as a
result o f the ongoing massive transnational migrations. It is no longer pos
sible to study all the important issues o f identity by confining our discus
sions to a specific national/racial/eth nic t e r r i t o r y . This case study on the
population o f African descent in a specific nineteenth-century Atlantic port
city in Brazil clearly suggests intriguing leads for our understanding o f iden
tity in this global world, despite the fact that the verb “globalize” was not
coined until 1944.13 Thus m y historical study is testimony to the fact that
studies o f the past can indeed help and promote a fuller understanding o f
the present. We are able to continue to learn from history how to perceive
and work on many sociocultural “problems,” including the important issues
o f identity in our global age.
* * *
We cannot deny that the Middle Passage and New World slavery de
humanized all enslaved Africans and their descendants to the most damag
ing degree. But all o f their terrifying experiences did not necessarily result
in “social death.” As this book has illustrated at many points, peoples o f
African descent in nineteenth-century Salvador, Brazil demonstrated a re
markable degree o f resilience, self-identity, and self-consciousness during
the time o f slavery. Their actions, hopes, and memories all indicate that
Afro-Brazilians, individually or collectively, never lost a strong sense o f who
they were. That is one important dimension o f the power o f identity. O f
course, Afro-Brazilians’ struggle for freedom did not end with the abolition
o f slavery in 1888. T he institution o f racial slavery surely ended in legal
terms, but racism and racial discrimination, which derived from the long
history o f slavery, have hampered Afro-Brazilians as a group from rising in
the society. At the same time Brazilian society has always encouraged and
allowed them to move up individually on a social ladder, albeit by changing
their social color from “black” to “mulatto” or even “white ” This resulted
in the continuing critical divisions among the Afro-Brazilian population,
which was reflected in a series o f various short-lived black movements dur
ing the twentieth century. To be black has remained largely an individual
choice for racial politics in Brazil. Afro-Brazilians, who chose to identify
themselves as “black,” have continued to fight rigorously for the creation and
(re)affirmation o f racial identity for their collective political empowerment.
Conclusion 165
Glossary
afilhadofafilhada godson/goddaughter
agregados dependants in households
alforria manumission
amas de leite wet nurses
armando warehouse and processing plant for whale products
arróba unit o f weight; one arrdba is approximately 52 pounds
batuque dance o f African origin practiced by persons o f Afri
can descent on the street after work and on weekends
and holidays
candomblé Afro-Brazilian religion resulted from West African
religions’ syncretism with Catholicism during the
slavery regime
cabra mulatto
caboclo person o f Euro-Indian ancestry
caixa unit o f weight; one caixa is around 40 arrdba
canto street corner; in nineteenth-century Salvador, the
street corner where African-born slaves and ex-slaves
o f the same nation gathered on the work day
companheiroSy companheiras common-law spouses
conto Brazilian currency; one conto was equivalent to 1,000
mil-riis, and written i:ooo$ooo riis
copadreJcomadre coparent
cria Brazilian-born slave raised in his or her owner’s house
hold
crioulo Brazilian-born black
emancipados the emancipated
engenho sugar mill; sugar plantation by extension
farinha manioc flour
fazenda plantation
forroiforra ex-slave
ganho wage earning
ganhadorlganhadora wage-eamer
irmatidade lay sodality
juntas unions
liberto! liberta ex-slave
livre free-born
madrinha godmother
malungos shipmates
mercadejas market vendors
mocambo fugitive slave community
mulato/mulata mulatto, often with pejorative connotations
negros novos newly arrived African-born slaves
pardo/parda mulatto
padrinho godfather
pataca 320 rHs
peculio savings
pessoas de cór persons o f color
prétolpréta African-born black
quitanda market stall
quitandeiroi quitandeira market-stall keeper
quilombo fugitive slave community
Recóncavo hinterland o f the city o f Salvador (Bahian RecAn-
cavo); the word's original meaning is cave, hole, and
land surrounding a port or a city
real former Brazilian currency, plural, rtis; 1,000 rtis is
written 1S000 riis, and called mil-rtis
terrenos temple grounds or sacred meeting places
venda market stall
168 Glossary
Notes
Introduction
On the notion o f identity, see, for instance» A. L. Epstein, Ethos and Identity:
Three Studies in Ethnicity (London: Tavistock Publications, 1978), pp. 100-101;
Edmund Leach, A Runaway World? (London: British Broadcasting Corpora
tion, 1968 [1967]), p. 34; Anita Jacobson-Widding, “ Introduction,” in Anita
facobson-Widding, ed., Identity: Personal and Socio-cultural (Uppsala, Swe
den: Acta Universitatis Upsaliensis, 1983), p. 13; Dorinne K. Kondo, Crafting
Selves: Power►Gender, and Discourses of Identity in a Japanese Workplace (Chi
cago: University o f Chicago Press, 1990), p. 48; and Richard Handler, “Is 'Iden
tity* a Useful Cross-Cultural Concept?” in John R. Gills, ed., Commemora
tions: The Politics o f National Identity (Princeton, N.J.: Princeton University
Press, 1994).
Historians have studied human behaviors and actions in the colonial setting,
based on archival sources and other historical documents. Such is the case o f
historian Rhys Isaac’s ethnographic method, named “action approach.” By
“translating* his historical data o f colonial Virginian plantation society as
"doings o f particular people in particular circumstances," Isaac, as the "field
ethnographer,** leads us to a full understanding o f society not as “primarily a
material entity” but as “ a dynamic product o f the activities of its members— a
product profoundly shaped by the images the participants have o f their own
and others* performances” (italics in original). See Rhys Isaac, “ Ethnographi-
cal Method in History: An Action Approach,” Historical Methods 13, no. 1
(1980): 43. See also Rhys Isaac, The Transformation o f Virginia 1740-1790
(Chapel Hill: University o f North Carolina Press, 1982). Another excellent
monograph on human behaviors is William B. Taylor, Drinking, Homicide,
and Rebellion in Colonial Mexican Villages (Stanford, Calif.: Stanford Univer
sity Press, 1979). O f course, we are well aware o f the limitations, both quanti
tative and qualitative, o f historical evidence on human behavior. As the histo
rian Robert F. Berkhofer, Jr. states: wSince the historian can never see the real
actions o f his subjects, he must derive behavioral manifestations from the
subjective documents he reads according to some theoretical scheme that
postulates that relationship” See Robert F. Berkhofer, Jr., A Behavioral
Approach to Historical Analysis (New York: Free Press; and London: Collier-
Macmillan, 1969)1 p* 18.
Dauril Alden, “ Late Colonial Brazil, 1750-1808,” in Leslie Bethell, ed., Colonial
Brazil (Cambridge: Cambridge University Press, 1987), p. 289.
4- The population o f the city o f Rio de Janeiro in 1808 was 54,245, which
doubled by 1822 and rapidly doubled again (222,313) by 1872. See Mary C.
Karasch, Slave Life in Rio de Janeiro 1808-1850 (Princeton, N.J.: Princeton
University Press, 1987). p* 61; and Alden, “Late Colonial Brazil,” p. 289.
5. Howard M. Prince, “Slave Rebellion in Bahia 1807-1835" (Ph.D. diss., Colum
bia University, 1972). pp. 15-16. Conto was a colonial and imperial Brazilian
currency measurement. One conto was 1,000,000 riis. The real (plural riis)
was the Brazilian currency o f account. One caixa o f sugar was around forty
arrdbas (an arrdba is approximately 32 pounds).
6. Epstein, Ethos and Identity, p. 5.
7. J. Clyde Mitchell, “TYibe and Social Change in South Central Africa: A Situa
tional Approach," Journal of Asian and African Studies 5 (1970): 85. On the con
cept o f “situation” in social anthropology, see, for instance, Max Gluck man,
“Analysis o f the Social Situation in Modern Zululand,” Bantu Studies 14
(1940): 1-30,147-174.
8. Mitchell, “Tribe and Social Change," p. 85.
9. J. Clyde Mitchell, The Kalela Dance: Aspects of Social Relationships among
Urban Africans in Northern Rhodesia (Manchester: Manchester University
Press, 1956); idem, Tribalism and the Plural Society (London: Oxford Univer
sity Press, i960); idem, “Tribe and Social Change." Elizabeth Bott, A. L.
Epstein, and J. C. Mitchell devised “network analysis” to examine the con
crete networks developed by migrant workers and their families living in
urban areas. Elizabeth Bott, Family and Social Network: Roles, Norms, and
External Relationships in Ordinary Urban Families (London: Tavistock Publica
tions, 1959); A. L. Epstein, “The Network and Urban Social Organization,**
Rhodes-Livingstone Journal 29 (1961): 29-62; J. C. Mitchell, ed., Social Networks
in Urban Situations (Manchester: Manchester University Press, 1969). The
Manchester School did not use or develop the term “ethnic group," or the con
cept o f “ethnicity” in this earlier stage o f their studies. On the school’s studies
o f ethnicity, see Abner Cohen, Custom and Politics in Urban Africa: A Study of
Hausa Migrants in Yoruba Towns (Berkeley and Los Angeles: University of
California Press, 1969); idem, Two-Dimensional Man: An Essay on the Anthro
pology of Power and Symbolism in Complex Society (London: Routledge 8c
Kegan Paul, 1974); Abner Cohen, ed., Urban Ethnicity (London: Tavistock Pub
lications. 1974); and Epstein. Ethos and Identity. On the historical creation o f
ethnicity, see Eric Hobsbawm and Terence Ranger, eds., The Invention o f Tra
dition (Cambridge: Cambridge University Press, 1983); Terence Ranger, The
Invention of Tribalism in Zimbabwe (Gweru, Zimbabwe: Mambo Press, 1985);
and Leroy Vail, ed., The Creation o f Tribalism in Southern Africa (London:
James Currey, 1989).
10. Mitchell, “Tribe and Social Change," pp. 88-89.

11. Ibid., p. 89.
170 Notes to pages 2-3

E. P. Thompson, The Making of the English Working Class (New York: Vintage
Books, 1966 (1963)).
Ibid., p. 9.
Ibid. My italics.
Herbert G. Gutman, The Black Family in Slavery and Freedom, 1750-1925
(New York: Vintage Books, 1977 [1976]); Herbert G. Gutman, “ Interview with
Herbert Gutman,” in Middle Atlantic Radical Historians' Association, Henry
Abelove, et al.» eds., Visions of History (New York: Pantheon Books, 1983),
p. in . Gutman states: “ In other words, there had to be connectedness. The im
portant discovery in the book, I think, is not that slaves lived in families that
were frequently broken, but that the family could serve the slaves as a way of
creating social and class connections far more important than the family. Call
these connections what you will! Once you know such connections exist, then
you can also find out when, why, and how they came together" Ibid., p. 211.
Gutman, Black Family p. 3. He also states his book is “about a special aspect
o f working class history.” See Gutman, “ Interview with Herbert Gutman,”
p. 206.
Gutman, “Interview with Herbert Gutman,” p. 205. See also Gutman, Black
Family, p. 3. Gutman states: . . understanding nineteenth-century slave
behavior and belief (the developed class) required a reexamination o f the
process by which Africans first became a social class in North America (the
new class)” (my italics). See Gutman, “Interview with Herbert Gutman,” p. 207.
On Gutman's usage o f the term “class,” Ira Berlin states: “He [Gutman] used
the term ‘class’ gingerly, preferring ‘workers' and other synonyms precisely
because Class’ analysis had been identified with a belief that economic struc
ture determined behavior and belief.. . . Thompson's understanding o f class
as the precipitate o f common experiences within a system o f productive rela
tions, and o f class consciousness as the articulation o f those experiences, was
also Gutman's.” See Ira Berlin, “Herbert G. Gutman and the American Work
ing C la s s in Herbert G. Gutman, Power and Culture: Essays on the American
Working Class, ed. Ira Berlin (New York: Pantheon Books, 1987).
Sidney W. Mintz and Richard Price, An Anthropological Approach to the Afro-
American Past: A Caribbean Perspective (Philadelphia: Institute for the Study
o f Human Issues, 1976). This book has been republished with a new preface
as The Birth of African-American Culture: An Anthropological Perspective
(Boston: Beacon Press, 1992).
Mintz and Price state: “An African heritage, widely shaped by the people
imported into any new colony, will have to be defined in less concrete terms,
by focusing more on values, and less on sociocultural forms, and even by
attempting to identify unconscious ‘grammatical’ principles, which may
underlie and shape behavioral response. To begin with, we would call for an
examination o f what Foster [George M. Foster] has called 'cognitive orienta
tions,' on the one hand, basic assumptions about social relations (what values
Notes to page 4 171

motivate individuals, how one deals with others in social situations, and mat
ters o f interpersonal style), and, on the other, basic assumptions and expecta
tions about the way the world functions phenomenologically (for instance»
beliefs about causality, and how particular cases are revealed).” See Mintz
and Price, Birth of African-American Culture, pp. 9-10; idem, Anthropological
Approach, p. 5.
20. For the discussion o f creolization in colonial British America, see T. H. Breen,
“Creative Adaptations: Peoples and Cultures,” in Jack P. Greene and J. R. Pole,
eds.» Colonial British America: Essays in the New History of the Early Modem
Era (Baltimore: Johns Hopkins University Press» 1984); and Jack P. Greene,
Pursuits of Happiness: The Social Development of Early Modern British Colonies
and the Formation o f American Culture (Chapel Hill: University o f North
Carolina Press, 1988)» especially chapters 7 and 8.
21. Mintz and Price, Birth o f African-American Culture, p. 14; Mintz and Price,
Anthropological Approach, p. 7. Their italics. See Stephan Palmié’s excellent
critiques on this cultural creolization model: Stephan Palmié, “Slave Culture
and the Culture o f Slavery in North America: Some Remarks on Historiogra
phy and the Present State o f Research,” and “ Ethnographic Processes and Cul
tural Transfer in Afro-American Slave Populations,” both in Wolfgang Binder,
ed., Slavery in the Americas (Würzburg: Kdningshausen 8c Newman, 1993).
22. Stephan Palmié, “Spies or Spades? Racial Classification and Ethnic Conflict in
Miami,” AmerikastudienJAmerican Studies 34 (1989): 211-221.
23. Anthony P. Cohen, The Symbolic Construction of Community (New York: Tavi
stock Publications, 1985), p. 12.
24. Suzanne Oboler, Ethnic Labels, Latino Lives: Identity and the Politics of
(Re)Presentation in the United States (Minneapolis: University o f Minne
sota Press, 1995).
25. See, for instance, Thomas Skidmore, “ Bi-rarial U.S.A. vs. Multi-racial Brazil:
Is the Contrast Still Valid?” Journal of Latin American Studies 25, no. 2 (1993):
373 - 3 8 6 .
26. Sidney W. Mintz, “ Defining Black and White,” Johns Hopkins Magazine 46,
no. $ (1994): 10. His italics.
27. George Reid Andrews, Emilia Viotti da Costa, and Franklin W. Knight,
“George Reid Andrews, Blacks and Whites in Sáo Paulo, Brazil, 1888-1988
(Madison: University of Wisconsin Press, 1991). Introduction, Two Com
ments, and a Rejoinder by the Author,” Luso-Brazilian Review 29, no. 2
(1992): 141-158,146.
28. Sidney W. Mintz, “A Melting Pot— Sort of,” Johns Hopkins Magazine 46, no. 4
(1994): 25. His italics.
29. David W. Cohen and Jack P. Greene, “ Introduction,” in David W. Cohen and
Jack P. Greene, eds., Neither Slave nor Free: The Freedmen of African Descent
in the Slave Societies of the New World (Baltimore: Johns Hopkins University
Press, 1972), p. 7. Their assertion that everywhere else these color categories

“ became increasingly vague and confused within a century o f the beginnings
o f the slave societies” (ibid., p. 7) is obviously not correct.
30. C. R. Boxer» Race Relations in the Portuguese Colonial Empire ¡415-182$
(Oxford: Clarendon Press, 1963), p. 120.
31. A.J. R. Russell-Wood, “Colonial Brazil»” in Cohen and Greene, eds., Neither
Slave nor Free* pp. 84-35.
32. George Reid Andrews, Blacks and Whites in S80 Paulo, Brazil 1888-1988 (Madi
son: University o f Wisconsin Press, 1991), p. 249.
33. Joseph C. Miller, Way o f Death: Merchant Capitalism and the Angolan Slave
Trade 1730-1830 (Madison: University o f Wisconsin Press, 1988); Orlando
Patterson, Slavery and Social Death: A Comparative Study (Cambridge» Mass.:
Harvard University Press, 1982).
1. A “C apital o f A frica” in Brazil
1. The Jesuits were expelled from Brazil in 1759.

2. James Henderson, A History of the Brazil: Comprising its Geography, Com
merce, Colonization, Aboriginal Inhabitants, & C . & C . & C . (London: Long
man, Hurt, Rees, Orme, and Brown, 1821), pp. 336-337; Robert Dundas,
Sketches o f Brazil: Including New Views on Tropical and European Fever, With
Remarks on a Premature Decay of the System Incident to Europeans on their
Return from Hot Climates (London: John Churchill, 1852)» pp. 200-201; and
Charles Frederick Hartt, Geology and Physical Geography of Brazil (Boston:
Fields, Osgood, 1870)» p. 334.
3. Hartt» Geology and Physical Geography, p. 335. Hartt also stated that in 1867
a street to connect the two parts o f the city was in the process o f construc
tion. See also George Gardner, Travels in the Interior of Brazil Principally
through the Northern Provinces and the Gold and Diamond Districts during
the Years 1836-1841 (London: Milford House, 1987 [1846]), pp. 74-75; and L. F.
de Tollenare, Notas dominiais (Recife: Governo de Estado de Pernambuco,
1979 )* P- 212. Nineteenth-century Salvador was composed o f ten parishes: SI,
Sao Pedro Velho, Santana, Conceiô da Praia, Vit6ria, Rua da Passo, Pilar,
Santo Ant6nio, Brotas, and Penha, until a new parish, Marl, was created in
1870 by subdividing Penha parish. See Anna Amelia Vieira Nascimento, Dez
freguesias da cidade do Salvador: aspectos sociais e urbanos do siculo XIX (Sal
vador: Fundaô Cultural do Estado da Bahia, 1986), pp. 28-51.
4. Robert Abl-Lallemant, Reise durch Nord=Brasilien im Jahre 1859,2 vols.
(Leipzig: F. D. Bradhaus, i860), vol. 1, p. 10.1 am indebted to Mrs. Hannelore
Russell-Wood for the translation from the original text in German. The Portu
guese translation is defective. See Robert Abl-Lallemant, Viagem pelo norte do
Brasil no ano de 1859, trans. Eduardo de Lima Castro, 2 vols. (Rio de Janeiro:
Instituto Nacional do Livio, 1961).
5. See, for instance, Robert Walsh, Notices of Brazil in 1818 and 1829, 2 vols. (Lon-
Notes to pages 6-12 173

don: Fredrick Westley and A. H. Davis, 1830), vol. 2, p. 182; Gardner, Travels in
the Interior of Brani, p. 78; Thomas Ewbank, Life in Brazil: A Journal c f a Visit
to the Land of the Cocoa and the Palm (New York: Harper & Brothers, 1856),
p. 439; Robert Elwes, A Sketcher’s Tour round the World: Illustrations from
Original Drawings by the Author (London: Hurst and Blackett, 1854), p. 94;
and Wilson Burgess, Narrative o f a Recent Visit to Brazil to Present an Address
on the Slave-Trade and SlaveryIssued by the Religious Society of Friends (Lon
don: Edmund Marsh, 1853), p. 49.
6. Thales de Azevedo, Povoamento da cidade do Salvador (Säo Paulo: Companhia
Editora Nacional, 1955)* pp. i 9 i->9 3 ; Brasil, Diretoria Geral de Estatistica,
Recenseamento da populafào de Imperio do Brasil a que procedeu no dia 1 de
agosto de 1872 (Rio de Janeiro: Leuzinger, 1873-1876), pp. 508,510.
7. See Stuart B. Schwartz, Sovereignty and Society in Colonial Brazil The High
Court of Bahia and Its Judges, 1609-1751 (Berkeley and Los Angeles: University
o f California Press, 1973)*
8. Stuart B. Schwartz, Sugar Plantations in the Formation o f Brazilian Society,
Bahia, ¡550-1835 (Cambridge: Cambridge University Press, 1985), p. 78.
9. On sugar planters in Bahia, see Schwartz, Sugar Plantations, pp. 264-294.
On the urban life o f sugar planters and their families in nineteenth-century
northern Brazil, see Gilberto Freyre, The Mansions and the Shanties: The
Making of Modem Brazil, trans, and ed. Harriet de Onis (New York: Alfred A.
Knopf, 1963).
10. James Lockhart and Stuart B. Schwartz, Early Latin America: A History of
Colonial Spanish America and Brazil (Cambridge: Cambridge University
Press, 1983), p. 198; Schwartz, Sugar Plantations, p. 52; and Russell Menard
and Stuart B. Schwartz, “W hy African Slavery? Labor Force Transitions in
Brazil, Mexico, and the Carolina Lowcountry,” in Binder, ed. Slavery in the
Americas, p. 95. See also Màrio Maestri, “C on sid eralo sobre o cativeiro do
'negro da terra* no Brasil quinhentista,” Estudos Ibero-Americanos 16, no. 1/2
(1990): 197-210.
11. Alexander Marchant, From Barter to Slavery: The Economic Relations o f Portu
guese and Indians in the Settlement o f Brazil, 1500-1580 (Baltimore: Johns Hop
kins University Press, 1942); and Schwartz, Sugar Plantations, pp. 34-35.
12. Schwartz, Sugar Plantations, pp. 35-37.
13. I b id ., p. 16.
14. On the sugar plantation society in Bahia, see Schwartz, Sugar Plantations.
Also, on Bahian planters and merchants, see John W. Kennedy,wBahian Elites,
1750-1822,” Hispanic American Historical Review (hereafter HAHR) 53, no. 39
(i 9 7 3 ): 415-439; Rae Jean Dell Flory, “Bahian Society in the Mid-Colonial Pe
riod: The Sugar Planters, Tobacco Growers, Merchants, and Artisans o f Salva
dor and the Recòncavo” (Ph.D. diss., University o f Texas at Austin, 1978); Rae
Jean Dell Flory and David Grant Smith, uBahian Merchants and Planters in
the Seventeenth and Early Eighteenth Centuries,” HAHR 58, no. 4 (1978): 571-

594 ; Catherine Lugar, “The Merchant Community o f Salvador, Bahia,
1780-1830” (Ph.D. diss., State University o f New York at Stony Brook» 1980).
15. On Indians and Indian slavery in Brazil» see Marchant, From Barter to Slav
ery; Stuart B. Schwartz, uIndian Labor and New World Plantations: European
Demands and Indian Responses in Northern Brazil,'“ American Historical Re
view (hereafter AHR) 83, no. 3 (1978): 43-78; and John Hemming, Red Gold:
The Conquest o f the Brazilian Indians (Cambridge, Mass: Harvard University
Press, 1978).
16. Schwartz, Sugar Plantations, chapter 3; and Robert Edgar Conrad, World of
Sorrow: The African Slave Trade to Brazil (Baton Rouge: Louisiana State Uni
versity Press, 1986), pp. 15-19.
17. Schwartz, Sugar Plantations, p. 422. One arrôba is approximately 32 pounds.
18. Alden, “ Late Colonial Brazil," pp. 314-318; and David Eltis, Economic Growth
and the Ending of the Transatlantic Slave Trade (New York: Oxford University
Press. 1987), pp. 48-49.
19. AJden, MLate Colonial Brazil* p. 311.
21. AJden, “ Late Colonial Brazil," p. 312.
22. Ibid., pp. 286-287» 289.
23. Joseph C. Miller, “Slave Prices in the Portuguese Southern Atlantic, 1600-
1830,win Paul E. Lovejoy, ed.» Africans in Bondage: Studies in Slavery and Slave
Trade (Madison: University of Wisconsin Press, 1986), pp. 43-44. Various
scholars, with mixed results, have attempted to periodize the slave trade. An
early proponent was Luis Viana Filho, who identified four “cycles," namely the
Guinea cycle in the sixteenth century; the Angola cycle in the seventeenth cen
tury; the Mina Coast and Bight o f Benin cycle in the eighteenth century until
1815; the fourth and last phase, namely the period o f clandestine slave trade,
1816-1851. See Luís Viana Filho, O negro na Bahia, 3rd ed. (Rio de Janeiro:
Nova Fronteira, 1988), p. 38. The French scholar Pierre Verger also viewed the
trade in cyclical terms, but developed these as follows: the Guinea cycle, dur
ing the second half o f the sixteenth century, the Angola and Congo cycle in
the seventeenth century; the Mina Coast cycle, during the first three quarters
o f the eighteenth century; the Bight o f Benin cycle between 1770 and 1850,
including the period o f clandestine slave trade. See Pierre Verger, Trade Rela
tions between the Bight of Benin and Bahia from the 17th to 19th Century, trans.
Evelyn Crawford (Ibadan, Nigeria: Ibadan University Press, 1976), p. 1. This
book is an English translation o f Pierre Verger, Flux e Reflux de la Traite des
Negres entre le Golfe De Bénin et Bahia de Todos os Santos du Xvii au Xix Siècle,
3 vols. (Paris: Mouton, 1968). A Portuguese translation was later published as
Pierre Verger, Fluxo e refluxo do tráfico de escravos entre o Golfo do Benin e a
Bahia de Todos os Santos, dosséculos XVII a XIX, trans. Tasso Gadzanis, 2nd
ed. (Sào Paulo: Editora Corrupio, 1987). On Portuguese slave trade from
Angola to Brazil, see Miller, Way of Death. On the transatlantic slave trade

in general, see Philip D. Curtin, The Atlantic Slave Trade: A Census (Madison:
University o f Wisconsin Press» 1969); and Joseph E. Inikori and Stanley L.
Engerman, eds., The Atlantic Slave Trade: Effects on Economics, Societies* and
Peoples in Africa, the Americas, and Europe (Durham, N.C.: Duke University
Press, 1992). For the African side o f the Atlantic slave trade, see a collection o f
narratives by West Africans in the eighteenth and nineteenth centuries edited
by Philip D. Curtin, Africa Remembered: Narratives by West Africans from the
Era of the Slave Trade (Madison: University o f Wisconsin Press, 1967).
24. Paul E. Lovejoy, “Background to Rebellion: The Origins o f Muslim Slaves in
Bahia,” Slavery and Abolition 15, no. 2 (1994): 154-155.
25. Leslie Bethell, The Abolition of the Brazilian Slave Trade: Britain, Brazil and
the Slave Trade Question 1807-1869 (Cambridge: Cambridge University Press,
1970), especially p. 137, note 2; Robert Edgar Conrad, “Neither Slave nor Free:
The Emancipados o f Brazil, 1818-1868," HAHR 53» no. 1 (1973): 50-70; and
Conrad, World of Sorrow; chapter 3.
26. Arquivo Público do Estado da Bahia» Salvador» Brazil (hereafter APB), Se^ào
Histórica, maôs 1404-1406,2896. On “free Africans,” see James Bandinel»
Some Accounts o f the Trade in Slaves from Africa as Connected with Europe and
Africa (London: Longman, Brown, 1862); W. D. Christie, Notes on Brazilian
Questions (London: Macmillan, 1865); Conrad, World of Sorrow, chapter 7.
On the employment of “free Africans” in Bahia, see, for example, "Fala de
Francisco Gonsalves Martins” (1852)» in Fala receitada da Assembléia Legisla
tiva (falos e relatorios dos presidentes em exercio da Provincia) (Bahia, 1839-
1889), pp. 27-28; APB, maô 2885, R e la jo dos africanos livres illicitamente
i m p o r t a d o s n o I m p e r io , a p p r e h e n d id o s e d e s tr ib u d o s n a C a p ita l d a P r o v in c ia
da Bahia. On their employment at the Santa Casa da Misericòrdia o f Salvador,
see Arquivo da Santa Casa da Misericòrdia, Livros do assentamento dos afri
canos livres à servilo da Santa Casa, nos. 52-62, B2-200.
27. Eltis, Economic Growth, p. 244.
28. Ibid., p. 199.
29. Ibid., pp. 243-244-
30. I consulted baptismal records registered in Santo Antònio parish for two out
o f every ten years between 1809 and 1869. My data include 211 male and 202
female African-born adult slaves, and distribution is as follows: 44 males and
55 females (1809-1810); 80 males and 61 females (1818-1819); 80 males and 80
females (1828-1829); 2 males and 2 females (1838-1839); 4 females (1848-1849);
and 5 males and 1 female (1858). After 1851, when the transatlantic slave trade
to Brazil was officially terminated, there were only 6 African-born slaves bap
tized in Santo António parish, and they appear to have gone unbaptized for at
least eight years after arrival in Brazil. The number o f baptisms o f African-
born slaves dropped drastically in the 1830s, a direct result of the virtual ter
mination o f the influx o f slaves from Africa to Salvador in 1831. See Arquivo
da Curia Metropolitana de Sào Salvador da Bahia (hereafter ACMS), Livros
de batizados, freguesia de Santo António. These years are 1809-1810,1828-

1829,1838-1839» 1848-1849» 1858-1859» and 1868-1869. The irregularity in this
sample, namely for 1809-1810, is because the data in 1808 were totally inade
quate. Church archives o f Salvador seem to have lost most o f the registers for
that year. It should be also emphasized regarding baptismal records that Santo
Antônio parish is one o f the few whose data were nearly complete among
major parishes o f Salvador in the nineteenth century. For instance, those of
Sé are not acceptable, because o f the physical deterioration o f the registers.
31. Bethell, Abolition of the Brazilian Slave Trade, chapter 12. The number o f
slaves imported from Africa to Brazil for 1851-1855 amounted to 6,100 (3,300
for the South o f Bahia; 1,900 for Bahia; and 900 for the North o f Bahia). No
more staves were imported to either Bahia or the North o f Bahia, but 300
slaves were still imported to the South o f Bahia in 1856. Eltis, Economic
Growth, p. 244. On the abolition o f the transatlantic slave trade in general,
see David Eltis and James Walvin, eds., The Abolition o f the Atlantic Slave
Trade: Origins and Effects in Europe, Africa, and the Americas (Madison: Uni
versity o f Wisconsin Press, 1981).
32. M. I. Finley, “Slavery,” in International Encyclopedia of the Sodai Sciences,
vol. 14 (New York: Macmillan and Free Press, 1968), p. 308.
33. Kàtia de M. Queirós Mattoso and her former graduate students at the Fed
eral University o f Bahia have produced substantial research on urban slavery
and urban society in Salvador during the nineteenth century. See Kátia de
M. Queirós Mattoso, Entre esclave au Brésil: XVIe-XIXe (Paris: Hachette,
197 9 )» which has been translated by Arthur Goldhammer as To Be a Slave in
Brazil, 1550-1888 (New Brunswick, N.J.: Rutgers University Press, 1986), and
Bahia, siculo XIX: urna provincia no impèrio (Rio de Janeiro: Editora Nova
Fronteira, 1992); Maria Inés Còrtes de Oliveria, O liberto: o seu mundo e os
outros (Sào Paulo: Corrupio, 1988); Joào José Reis, Rebeliào escrava no Brasil
a história do levante dos malès (1835), 2nd ed. (Sào Paulo: Editora Brasiliense,
1987), which has been translated by Arthur Brakel as Slave Rebellion in Brazil
The Muslim Uprising of 1835 in Bahia (Baltimore: Johns Hopkins University
Press, 1993); and Maria José de Souza Andrade, A mào de obra escrava em Sal
vador 1811-1860 (Sào Paulo: Corrupio, 1988).
34. Curtin, Atlantic Slave Trade, pp. 19,41-47; and Herbert S. Klein, “African
Women in the Atlantic Slave Trade,” in Claire C. Robertson and Martin A.
Klein, eds., Women and Slavery in Africa (Madison: University o f Wisconsin
Press, 1983).
35. Claire C. Robertson and Martin A. Klein, “Women’s Importance in African
Slave Systems,” in Robertson and Klein, eds., Women and Slavery in Africa,
especially pp. 3-11; and William D. Phillips, Jr., Slavery from Roman Times to
the Early Transatlantic Trade (Minneapolis: University o f Minnesota Press,
1985)» pp. 122-123.
36. E lt i s , Economic Growth, p . 257*
37. Arquivo Municipal da Cidade do Salvador, Salvador, Brazil (hereafter AMCS),

Livros de escrituras da compra e venda de escravos, freguesia da Sé, nos. 82.1-

82.20. Sé parish, which was founded in 1552 as the second oldest parish in
Bahia, had the most important role as the administrative and ecclesiastical
center, not only for the city, but also for the captaincy o f Bahia. In about 1724,
one quarter o f the population o f the city was concentrated in Sé, and in 1807,
21.8 percent o f the population were residents o f this parish located at the cen
ter of the city. See Schwartz, Sugar Plantations, p. 88; and Lois dos Santos Vil-
hena, A Bahia no sécula XVIII, vol. 2 (Bahia: Editora Itapuà, 1969), p. 460.
38. APB, Se^áo Histórica, maô 1549, cademo 2 (Sào Pedro Velho parish, quar
teiráo 15 [1855]); ma$o 1602 (Sé parish, quarteiroes 20,21, 22,23 [1855]); Santo
Antònio parish, 1st district, quarteiráo 18 [1855]; Sáo Pedro Velho parish, quar-
teirìo 1 [1855]; Sào Passo parish, quarteiráo 6; ma(o 1605 (Santana parish,
quarteirao 3 [1855]); San Antònio parish, 1st district, quarteiráo 15 and 2nd dis
trict, quarteiráo 11 [1855]; and Brasil, Recenseamento da populando, p. 510.
40. Andrade, A màe de obra escrava, pp. 129,130.
41. APB, Se^áo Judiciária» Inventários da capital (1808-1888). The 370 inventories
1 consulted comprise 14 (registered for the period 1808-9); 30 (1810-19); 16
(1820-29); 37 (1830-39); 57 (1840-49); 88 (1850-59); 58 (1860-69); 4 2 (1860-69);
and 28 (1880-88). Male-female ratios in these data are 2:1 among the African-
born and 49:51 among the Brazilian-born for the period 1808-1849; and 59:41
among the African-bom and 49:51 among the Brazilian-born for the period
1850-1888. Among the slaves o f African birth, men largely outnumbered
women. This is because the 370 inventories o f slave owners include some
owners o f large-scale plantations o f suburban Salvador, where the majority
o f the registered slaves were male field workers. For instance, the inventory o f
Pedro de Espirito Santo, registered in 1850, showed that he owned 126 slaves,
o f whom only 45 were women (APB, Inventàrio da capital, 04/1624/2093/01).
Therefore I do not interpret these data derived from the inventories o f slave
owners as negating my discussion on balance by sex in urban slavery.
42. Maria Luiza Marcilio, “The Prices o f Slaves in XlXth Century Brazil: A Quan
titative Analysis o f the Registration o f Slave Sales in Bahia,” Studi in memoria
di Federigo Melts, vol. 5 (Paris: Giannini Editore, 1978), pp. 90,92. In writing
amounts o f réis, the symbol used in the U.S. as a dollar sign ($) separates the
thousands from the hundreds, as a comma does in writing U.S. dollar amounts.
43. AMCS, Livros de escrituras de compra e venda de escravos, freguesia da Sé.
44. Manuel Moreno Fraginals, Herbert S. Klein, and Stanley L. Engerman,
“The Level and Structure o f Slave Prices on Cuban Plantations in the Mid-
Nineteenth Century: Some Comparative Perspectives,” AHR 88, no. 5 (1983):
1201-1218, especially 1219-1223.
45. APB, Inventários da capital. Forty-two individuals (11.4%) owned 11-20 slaves.
Nine owned 20-29 slaves, while the rest owned 39,54,57, and 126 slaves.
46. Reis, Rebeliào escrava, p. 25.
47. APB, Se^áo Histórica, maô 1070-4, artigo 3 (December 7,1832).

4&. O n slave wet-nurses in the city, see Robert Edgar Conrad» Children o f God's
Fire: A Documentary History of Black Slavery in Brazil (Princeton» N.J.: Prince
ton University Press, 1983), pp. 133-140.
49. Maria Graham briefly mentioned the loja. See Maria Dundas Graham, Journal
o f a Voyage to Brazil and Residence There, during Part of the Years 1821, 1822,
1823 (New York: Frederick A. Praeger, 1969 [1824]), p. 149. The surviving cen
sus records o f 1855, for instance, reveal the residence pattern among live-out
urban slaves. See APB, maôs 1549» 1602 and 1605.
50. James Wetherell, Brazil Stray Notes from Bahia: Being Extracts from Letters, &
c.y during a Residence of Fifteen Years, ed. William Hadfield (Liverpool: Webb
and Hunt, i860), pp. 16-17.

51. Johann B. von Spix and Carl Friedrich P. von Martius, Viagem pelo Brasil,
vol. 2, trans. Lucia Furquim Lahmeyer, 2nd ed. (Rio de Janeiro: Impresa
Nacional, 1938), p. 301.
51. Edison Cameiro, Ladinos e crioulos (Rio de Janeiro: Editòra Civilizadlo
Brasileira, 1964), p. 9.
53. APB, maô 2898, R e la jo dos africanos libertos existentes nesta freguesia,
Santana, February 11,1849.
54. A le x a n d e r M a r jo r ib a n k s , Travels in South and North America, 5th e d . (L o n
d o n : S i m p k i n , M a r s h a l l , 1854)» P P - 6 7 - 6 8 .
55. On “hiring-out” and “hiring one's time out,* see Richard C. Wade, Slavery
in the Cities: The South 1820-1860 (New York: Oxford University Press, 1964),
pp. 38-40,48-49-
56. APB, Inventários da capital; and Reis, Rebeliào escrava, pp. 208*209. Karasch
analyzes these diverse “functions” o f urban slaves in early nineteenth-century
Rio de Janeiro. See Mary C. Karasch, “ From Porterage to Proprietorship: Afri
can Occupations in Rio de Janeiro, 1808-1850” in Stanley L. Engerman and
Eugene D. Genovese, eds., Race and Slavery in the Western Hemisphere: Quan
titative Studies (Princeton, N.J.: Princeton University Press, 1975); idem, “Sup
pliers, Sellers, Servants, and Slaves,” in Louisa Schell Hoberman and Susan
Migden Socolow, eds., Cities and Society in Colonial Latin America (Albuquer
que: University o f New Mexico Press, 1986); idem, Slave Life, chapter 7. See
also Luiz Carlos Soares, “Os escravos de ganho no Rio de Janeiro do século
XIX,” Revista Brasileira de História 8, no. 16 (1988): 107-142*
57. Bandinel, Some Accounts, p. 288.
58. On the coffee plantations and labor in the Southeast, see Stanley J. Stein, Vas-
souras, a Brazilian Coffee County, 1830-1900: The Role of Planter and Slave in a
Plantation Society (Princeton, N.J.: Princeton University Press, 1985 (1957)).
59. On the internal slave trade, see Herbert S. Klein, The Middle Passage: Com
parative Studies in the Atlantic Slave Trade (Princeton, N.J.: Princeton Univer
sity Press, 1978), chapter 5; and Conrad, World o f Sorrow, chapter 8.
60. Thomas W. Merrick and Douglas H. Graham, Population and Economic Devel

opment in Brazit 1800 to the Present (Baltimore: Johns Hopkins University
Press» 1979), pp. 50-52-
61. Klein» Middle Passage, pp. 102-105» n 4-

62. Finley, “Slavery,” p. 308. According to Finley, genuine slave societies include
classical Greece (except Sparta) and Rome» the American South» and the Ca
ribbean, whereas slave-owning societies are found in the ancient Near East (in
cluding Egypt), India, and China. See ibid.» p. 306. His italics.
63. Finley emphasizes that it is not the absolute number or proportion o f slaves,
but rather their location and function that characterize a society that holds
and uses slaves. He maintains: “ It does not matter, in such situations» whether
as many as three-fourths were not slaveowners» or whether slavery was not
fairly widespread outside the elite in domestic or other nonproductive roles.”
See ibid., p. 308.
64. Philip D. Morgan, in his study on varying race relations in the first British
Empire» employs the same definition o f “slave societies” and “slaveowning
societies” : “In the former, some slaves exist; in the latter, slavery is the deter
minative institution. A slaveowning society may become a slave society» but
only when a significant proportion o f its population is enslaved (say, for
argument’s sake, more than 20 percent) and, more important, when slavery
becomes central to the economic functioning o f that society” Morgan does
not tell how he has calculated 20 percent for the distinction. See Philip D.
Morgan, “ British Encounters with Africans and African-Americans, circa
1600-1780»" in Bernard Bailyn and Philip D. Morgan, eds., Strangers within
the Realm: Cultural Margins of the First British Empire (Chapel Hill: Univer
sity o f North Carolina Press, 1991), p. 163. His italics.
65. Among the 3,156 letters o f liberty I consulted for this study, there were only 16
in which ex-owners declare that they themselves were African-born ex-slaves.
See APB, LJvros de notas and maô 135. This may be a reflection o f the fact
that a very small percentage o f ex-slaves could become slave owners. For
instance, 186 out o f 207 (89.9%) African-born ex-slaves who resided in Con-
cei<ô da Praia parish (1846) did not own a single slave. Likewise, 263 out of
33 i (7 9 *5% ) African-born ex-slaves who resided in Santana parish (1849) were
not slave owners. See APB» maô 6472, Arollomento dos africanos libertos. . .
Conceiâo da Praia, 1846; ibid, maô 2898, Relaô dos africanos libertos. . .
Santana. On non-slave owners in colonial and imperial Brazil, see Iraci del
Nea> da Costa, Arruiu-miudu: um estudo sobre os nHo-proprietdrios de escravos
no Brasil (S3 o Paulo: MGSP Edit6res Ltda., 1992).
66. AMCS, Livros de escrituras de compra e venda de escravos, freguesia da Si,
nos. 82.1-82.20. The Municipal Archives o f Salvador (AMCS) house numbers
o f legal documents concerning slaves registered in the parishes o f the city.
Among them, those for Si parish have been preserved almost in their entirety.
Maria Luiza Marcilio consulted the same type o f slave sale registers (the total
o f 2,527) in Pilar parish o f Salvador for 1838-1882. However, she analyzes only
544 cases o f slaves who were either domestics or field hands, and without any

physical defects, registered for 1850-1882» and does not make any breakdown
o f the period. See Marcflio, “ Prices o f Slaves,” pp. 83-97.
APB, Livros de notas and maô 2898.
Moreno Fraginals, Klein, and Engerman, “Level and Structure o f Slave
Prices,” pp. 1207-1209.
APB, Livros de notas and ma<;o 2898. In writing amounts o f rtis, the colon
separates the millions from the thousands, as a comma does in writing US.
dollar amounts.
AMCS, Livros de escrituras de compra e venda de escravos, freguesia da Si.
Klein, Middle Passage, p. 101.
APB, Livros de notas and maô 135* On slave prices recorded in the inventories
o f slave owners in Salvador for 1811-1888, see Andrade, A mdo de obra escrava,
pp. 207-211.
Merrick and Graham maintain: “Approximately one out o f every two freed
slaves gained their freedom through self-purchase. This proportion grew
(and gratis manumissions declined) during periods o f rising slave prices.”
See Merrick and Graham, Population and Economic Development, p. 53.
AMCS, Livros de escrituras da compra e venda de escravos, freguesia da Si.
Occupations o f male slaves: domestics, 326 (66.3%); fieldwork, 126 (25.6%);
wage earning, 16 (3.2%); artisans, 15 (3.0%); maritime, 7 (1.4%); others, 3
(0.5%). Occupations o f female slaves: domestics, 510 (81.5%); fieldwork, 109
(17.4%); wage earning, 5 (0.8%); maritime, 2 (0.3%). Occupations of slaves
were not listed in these legal registers until 1861.
For figures on slaves in nineteenth-century newspaper advertisements, see
Gilberto Freyre, O escravo nos anuncios dejomais brasileiros do stculo XIX,
2nd ed. (S§o Paulo: Companhia Editora Nacional, 1979). The first monograph
on urban domestic slaves in Brazil is Gilberto FreyreTs Sobrados e mucambos
(1936), whose English translation is The Mansions and the Shanties. The Mak
ing o f Modem Brazil
Schwartz, Sugar Plantations, pp. 52-53.
A. J. R. Russell-Wood, The Black Man in Slavery and Freedom in Colonial
Brazil (New York: St. M artini Press, 1982), and personal communication.
My research does not support Barbara Trosko*s assertion on the time of
change in terminology: “With the arrival o f the Paraguayan War in 18(¡rj,forro
ceded to the more genetic term liberto.n See Barbara Rose Trosko, “The Liberto
in Bahia before Abolition” (M.A. thesis, Columbia University, 1967), p. 4.
2. T h e Creation o f N ew Identity, 1808-1831

Paul E. Lovejoy, Transformations in Slavery: A History of Slavery in Africa
(Cambridge: Cambridge University Press, 1983), p. 143* See also Patterson,
Slavery and Social Death, chapter 4.

2. A. C. de C. M. Saunders, A Social History o f Black Slaves and Freedmert in
Portugal, 1441-1555 (Cambridge: Cambridge University Press, 1982), chaptcr ¿.
3. Henry Koster, Travels in Brazil 1809-1815, vol. 2 (London: Longman, Hurst,
Rees, Ormes, and Brown, 1817), pp. 238-239.
4. Frederick P. Bowser, The African Slave in Colonial Peru 1524-1650 (Stanford,
Calif.: Stanford University Press, 1974), p. 47.
5. Jack Goody, “Writing, Religion, and Revolt in Bahia,” Visible Language 20,
no. 3 (1986): 319-320.
6. Ibid., p. 320.
7. Koster, Travels in Brazil, vol. 2, p. 252. My italics.
8. Curtin, Atlantic Slave Trade, pp. 184-190.
9. On slave stereotypes and prices in accordance with their “nationalities” in the
case o f South Carolina, see Daniel C. Littlefield, Rice and Slaves: Ethnicity and
the Slave Trade in Colonial South Carolina (Baton Rouge: Louisiana State Uni
versity Press, 1981), chapter 1.
10. Thomas Lindley, Narrative of a Voyage to Brazil (London: Printed for J. John
son, 1805), p. 269.
11. Koster, Travels in Brazil, vol. 2, pp. 252-258. See also Lindley, Narrative of a
Voyage to Brazil, p. 269.
12. Spix and Martius, Viagem pelo Brasil, vol. 2, p. 300.
13. See Karasch's largely static treatment o f “nations” in early ninetcenth-century
Rio de Janeiro in her Slave Life, chapter 1.
14. Koster, Travels in Brazil, vol. 2, pp. 251, 293.
15. “ Devassa do levante de escravos occorido em Salvador em 1835,” Anais do Ar-
quivo Publico do Estado da Bahia (hereafter AAEB) 38 (1968): 1-142,32-38. On
malungos, see Karasch, Slave Life, p. 298.
16. Lovejoy, “ Background to Rebellion,” p. 155.
17. Mintz and Price, Birth of African-American Culture, pp. 43,44: Mintz and
Price, Anthropological Approach, pp. 22,23. The “shipmate” relationship has
been cited as malungue in Trinidad, mati in Suriname, and batiment in Haiti.
See Mintz and Price, Birth of African American Culture, p. 44; Anthropological
Approach, p. 23.
18. Walsh, Notices of Brazil, pp. 323-328; Maria Graham, Journal of a Voyage to Bra
zil, pp. 227-228. On the slave market o f Rio de Janeiro, see also Karasch, Slave
Life, chapter 2. Once the slave trade was banned in Brazil, the Valongo itself
was declared illegal on November 7,1831, but much o f the business in slaves
continued as before, even after 1850. See Karasch, Slave Life, p. 36.
19. Lindley, Narrative of a Voyage to Brazil, p. 176.
20. A building that formerly housed the slave market on Baixa dos Sapateiros is
now owned by the municipal council o f Salvador, and has been transformed
182 Notes to pages30-33

into a museum o f West African cultures and named Casa de Benin. Founders
included the late French anthropologist Pierre Verger.
21. See, for instance» Proposta (December 2» 1828)» articles 2 and 8, Presidencia da
Provincia Conselho Geral da Provincia (Correspondencia recebida), 1829-1854*
APB, Se^áo Histórica, maô 1070-4.
22. The Pelourinho provided slave owners with an ideal place for whipping since
whipping was, as the historian Ira Berlin clarifies, “an instrument of social
discipline intended to impress not only the immediate victim but also all who
witnessed it” See Berlin, “Herbert G. Gutman,” p. 45.
23. AMCS, Livro de posturas, no. 119.6, fs. 71,107 (April 25,1843).
24. APB, Se^áo Histórica, maô 2681.
25. Ibid., maô 2951.
26. Marcus J. M. de Carvalho, Liberdade: rotinas e ruptures do escravismo no
Recife, 1822-1850 (Recife: Editora Universitaria de UFPE, 1998), chapter 13.
I thank Professor Joseph C. Love for the citation.
27. APB, Se^áo Histórica, maô 2951.
28. Ibid., mafo 2680, caixa 1047.
29. Ibid., mato 2681.
30. APB, Inventàrios da capital.
31. On the streets o f Salvador, see, for instance, Wetherell, Brazil: Stray Notes from
Bahia, p. 7; and Andrew Grant, History of Brazil (London: Printed for Henry
Colburn, 1809), p. 208.
32. On bicho da pi* set Lindley, Narratives o f a Voyage, p. 153; and Wetherell,
Brazil: Stray Notes from Bahia, p. 19.
33. The archives o f Santa Casa da Misericòrdia da Bahia house a book o f regis
ters o f slaves and “ free Africans.” Many o f these registers recorded the ethnic
mark of the individual African-born. See Arquivo da Santa Casa da Miser-
córdia da Bahia, Livro de assentamento dos escravos de Santa Casa (B.3.220);
Livro do assentamento dos africanos Livres à servilo da Santa Casa, 1852-1862
(B.2.200); and Livro de matricula dos africanos livres à servilo da Santa Casa,
1862-1864 (B.2.201).
34. Wetherell, Brazil: Stray Notes from Bahia, p. 134. His italics.
35. Ibid., p. 5.
36. Ibid., p. 5.
37. Ibid., p. 54.
38. Henderson, History of Brazil p. 340.
39. Ibid.
40. Wetherell, Brazil Stray Notes fromBahia, p. 5.
41. Biblioteca Nacional do Rio de Janeiro,Rio de Janeiro, Brazil (hereafter BNRJ),
II-34.6.57.

42. Spix and Martius, Viagem pelo Brasil, vol. 2, p. 300.
43. Nina Rodrigues, Os africanos no Brasil, 3rd ed. (S3 o Paulo: Companhia
Editdra Nack>nal> 194$), chapter 6.
44. Donald Pierson, Negroes in Brazil A Study o f Race Contact at Bahia (Chicago:
University o f Chicago Press, 1941), pp. 72-73.
45- For example, many African-born participants, both slave and ex-slave, in the
M ali Revolt o f 1835 were identified by other people o f African birth with their
African names only. See the trial and judicial records o f the Revolt, many of
which were published by the State Archives o f Bahia as“Devassa do levante
de escravos occorido cm Salvador em 1835,” AAEB 38 (1968):1-142; and “ 1835
Insurreiîo de escravos,’* AAEB 40 (1970): 9-170. See also Prince, “Slave Rebel
lions,” p. 186; and Reis, Rebeli&o escrava, p. 197 (Table 7).
46. F ra n k lin W. K n ig h t a n d M a rg a re t E. C r a h a n , “T h e A fric a n M ig ra tio n a n d
t h e O r ig in o f a n A f ro - A m e r ic a n S o c ie ty a n d C u ltu r e ” i n C r a h a n a n d K n ig h t,
e d s ., Africa and the Caribbean: The Legacies of the Link ( B a l t i m o r e : Jo h n s H o p
k i n s U n i v e r s i t y P r e s s , 1979)1 p* 12.
47. Epstein, Ethos and Identity, p. 40.

48. “Devassa do levante,” p. 7.
49. Reis, Rebeli&o escrava, p. 190.
50. J. C. Mitchell, “Perceptions o f Ethnicity and Ethnic Behaviour: An Empirical
Exploration,” in Abner Cohen, ed., Urban Ethnicity (London: Tavistock Publi
cations, 1974), p. 1.
51. Jacqueline Jones, Labor of Love, Labor o f Sorrow: Black Women, Work, and the
Family from Slavery to the Present (New York, Basic Books, 1985), p. 38. Recent
works on slave women in the Caribbean include Hilary McD. Beckles, Natural
Rebels: A Social History of Enslaved Black Women in Barbados (New Bruns
wick, N.J.: Rutgers University Press, 1989); Barbara Bush, Slave Women in
Caribbean Society 1650-1833 (Bloomington: Indiana University Press, 1990);
and Marietta Morrissey, Slave Women in the New World: Gender Stratification
in the Caribbean (Lawrence: University Press o f Kansas, 1989).
52. Maria Graham, journal o f a Voyage to Brazil, p. 133. See also Grant, History
of Brazil, pp. 235-236; Marjoribanks, Travels in South and North America,
pp. 9 4 - 9 5 .
53. Wetherell, Brazil' Stray Notes from Bahia, p. 53; Maijoribanks, Travels in South
and North America, p. 46.
54. James C. Fletcher and Daniel P. Kidder, Brazil and the Brazilians: Portrayed in
Historical and Descriptive Sketches, 9th ed. (Boston: Little, Brown, 1879),
PP- 47 5 - 4 7 6 .
55. Wetherell, Brazil Stray Notes from Bahia, p. 53. The British traveler Alexander
Marjoribanks made a similar observation; “The pole rests upon their bare
shoulders, and in order to mark time they have a singular sort o f cry as they

proceed along the streets« sung in a kind o f melancholy cadence, that proceed
ing from those in front being on a higher key than from those behind.” See
Marjoribanks, Travels in South and North America, p. 46.
56. AMCS, Livro de posturas» no. 119.1» f. 132.
57. Wetherell, Brazil: Stray Notes from Bahia, p. 106. It is only in the late nine
teenth century that communication facilities were substantially improved. In
1853» a major steamboat line named the Companhia de Navegado a Vapor
Bahiana started service around the Bay o f All Saints. Steamboat service along
the inland S3 o Francisco River began in 1865. In 1863 the £. F. Bahia-Sáo Fran
cisco rail line reached Alagoinhas» 123 kilometers north o f the capital, but it
was not until 1896 that the railroad finally connected the capital to the Sao
Francisco River. See Dain Borges, The Family in Bahia, Brazil 1870-1945 (Stan
ford, Calif.: Stanford University Press, 1992), pp. 21-22.
5$. Wetherell, Brazil Stray Notes from Bahia, p. 26.
59. Borges, Family in Bahia, p. 21.
60. Maria Graham, Journal o f a Voyage to Brazil, p. 145.
61. Spix and Martius, Viagem pelo Brasil, p. 290.
62. Wetherell, Brazil Stray Notes from Bahia, pp. 26-27.
63. Ibid., pp. 31,107-108; Spix and Martius, Viagem pelo Brasil, vol. 2, p. 290.
64. Wetherell, Brazil: Stray Notes from Bahia, p. 87. Wetherell continued to state:
M[I]t is their custom to mix in small quantities o f a crystallized substance
called cong.’ in a powdered state, and which is imported from the coast of
Africa** (ibid., p. 87). James Prior mentioned the prevailing practice o f smok
ing among the population o f the city, particularly the lower class. See Prior,
Voyage along the East Coast of Africa to Mosambique (London: Printed for
St. Richard Phillips, 1819), p. 104.
6>. Wetherell, Brazil: Stray Notes from Bahia, p. 26.
66. Karasch, “From Porterage to Proprietorship,” p. 379. George Gardner stated
in 1836 that the small boats and canoes in the bay o f Rio de Janeiro were
“all manned with African blacks.” See Gardner, Travels in the Interior of
Brazil, p. 4.
67. Grant, History of Brazil, p. 29; Wetherell, Brazil Stray Notes from Bahia, p. 139.
68. Lindley, Narrative of a Voyage to Brazil, pp. 152-153.
69. Gardner» Travels in the Interior of Brazil, p. 78; Tollenare» Notas dominiais,
pp. 218-222; and Elwes, Sketcher’s Tour, pp. 97-98.
70. On African American seamen in British America and the theme of African
Atlantic communication, see Julius Sherrard Scott III, “The Common Wind:
Currents o f Afro-American Communication in the Era o f the Haitian Revo
lution” (Ph.D. diss, Duke University, 1986); and W. Jeffrey Bolster, “ T o Feel
Like a Man’: Black Seamen in the Northern States, 1800-1860,” Journal of

American History 76, no. 4 (1990): 1173-1199; idem, Black Jacks: African Ameri
can Stamen in the Age of Sail (Cambridge, Mass.: Harvard University Press»
199 7 )*
71. Andrade, A m&o de obra, p. 129.
72. Lyman L. Johnson, “Artisans,” in Hoberman and Socolow, eds.. Cities and
Society, p. 240. According to Johnson, analysis o f the average prices paid for
young unskilled African-born slaves and for skilled slave artisans reveals that
profits o f 200 or 300 percent were not uncommon, even after the costs o f the
slave’s maintenance were subtracted. See ibid., p. 240.
73. Ibid., p. 240.
74. APB, Inventdrios da capital.
75. Ewbank, Life in Brazil p. 195. See also Wetherell, Brazil: Stray Notes from
Bahia, pp. 77-78,151.
76. I have not found any laws in Salvador forbidding slaves and/or free people of
African descent from selling specific commodities. However, in the British
Caribbean there were laws stipulating what slaves could selL These laws were
designed to protect poor whites who were trying to make a living in the same
way slaves did. See Michael Mullin, Africa in America: Slave Acculturation and
Resistance in the American South and the British Caribbean, 1736-1831 (Urbana:
University o f Illinois Press, 1992), p. 155. On the slaves’ economic activities in
the informal market of the Caribbean and mainland North America, see Ira
Berlin and Philip D. Morgan, eds., The Slaves' Economy: Independent Produc
tion by Slaves in the Americas (London: Frank Cass, 1991).
77. Spix and Martius, Viagem pelo Brasil p. 299.
78. Wetherell, Brazil: Stray Notes from Bahia, pp. 89-90. According to Wetherell,
by 1855 a company had been established to supply the principal portion o f the
city from pipes through the streets. In 1857, Wetherell mentioned some beauti
ful public fountains, which had been erected by the new company for supply-
ing the city with water. See ibid., pp. 152-153.
79- AMCS, Livro de posturas, 119.1, fs. 3,31.
80. Karasch, “Suppliers, Sellers, Servants, and Slaves,” p. 271.
81. John Clarke, Memorials of Baptist Missionaries in Jamaica (London, 1869),
pp. 9-n> cited by Mullin, Africa in America» p. 155; Mullin, Africa in America,
p. 131.
82. Sandra Lauderdale Graham discusses the social landscape of “house” and
“street” for female domestic servants, both slave and free, in the city o f Rio
de Janeiro for the period 1860-1910. In her model, “ house” and “street” are
dichotomous and function as reciprocal social spheres. See Sandra Lauderdale
Graham, House and Street: The Domestic World of Servants and Masters in
Nineteenth-Century Rio de Janeiro (Cambridge: Cambridge University Press,
1988). For the house/street dichotomy in Brazilian culture, see Roberto

DaMatta» A casa & a rua: espaço, cidadania, mulher e morte no Brasil (Sâo
Paulo: Brasiliense» 198s).
83. Fox-Genovese discusses the slave woman’s dual membership in the plantation
household and the slave community in the antebellum South: “Slave women
lived between the two worlds o f the plantation household and the slave com
munity. . . . Their experiences unfolded two different realities: the domination
o f their white masters and their relations within the black slave community.
Their lives and their identities as women combined their strands into a com
plex and distinctive pattern. From birth» the slave girls dual membership in
the plantation household and the slave community shaped her identity.” See
Elizabeth Fox-Ge nóvese, Within the Plantation Household: Black and White
Women of the South (Chapel Hill: University o f North Carolina Press» 1988),
p. 146.
84. Women’s monopoly o f small trade is still widely observed in many Third
World cities. See Sidney W. Mintz, “ Men, Women» and Trade,” Comparative
Studies in Society and History 13» no. 3 (1971): 247-269; Ximena Bunster and
Elsa M. Chaney» Sellers and Servants: Working Women in Lima, Peru (New
York: Praeger, 1985); and Nancy Rose Hunt, “Placing African Women’s His
tory and Locating Gender,” Social History 14» no. 3 (1989): 363.
85. Marisol de la Cadena,u‘Women Are More Indians*: Ethnicity and Gender in
a Community near Cuzco,” in Brooke Larson and Olivia Harris, eds.» with
Enrique Tandeter, Ethnicity, Markets, and Migration in the Andes: At the Cross
roads o f History and Anthropology (Durham, N.C.: Duke University Press,
1995 )» PP* 3 4 >~3 4 2*
86. Grant, History of Brazil pp. 239-240.
87. Vilhena, Bahia, vol. 1, p. 93.
88. Wetherell, Brazil: Stray Notes from Bahia, pp. 29-30.
89. For a discussion o f food, the concept o f pollution, and ethnic identity in the
Indian diaspora in Trinidad, see Aisha Khan, Ujuthaa in Trinidad: Food, Pollu
tion, and Hierarchy in a Caribbean Diaspora Community,” American Ethnolo
gist 21, no. 2 (1994): 245-269.
90. APB, Seçâo Histórica, maço 1400» Postura o f August 18,1847.
91. Wetherell, Brazil: Stray Notes from Bahia, p. 27.
92. Daniel D. Kidder, Sketches of Residence and Travels in Brazil, vol. 2 (Philadel
phia: Sorin and Ball, 1845), pp. 24-25.
93. Wetherell, Brazil Stray Notes from Bahia, p. 21.
94- AMCS, Livro de Posturas, no. 119.1, f. 122B (November 15,1785); and Vilhena,
Bahia, vol. 1, pp. 126-127.
95. Wetherell, Brazil Stray Notes from Bahia, pp. 83-84. The British Reverend
Hamlet Clark» visiting Salvador in December 1856» came across a lonely spot
in the suburbs, where eighty or more women and girls o f color were washing

clothes in a pool o f water, and were drying them in the hot sun. He described
the scene as “the cruel way in which they murdered the linen, beating it with
rough sticks, pointing with stones and amidst a chorus o f screams o f laughter
and noise that was very far removed from English ways!” See Hamlet Clark,
Letters from Spain, Algeria, and Brazil during Past Entomological Rambles (Lon
don: John van Woorst, Paternoster Row, 1867), pp. 105-106. On laundresses in
the city o f Rio, see Ewbank, Life in Brazil, pp. 113-114.
In his small, but still interesting, article entitled “Gossip and Scandal," social
anthropologist Max Gluckman stresses the important function o f gossip in
maintaining social groups. See Max Gluckman, “Gossip and Scandal: Papers
in Honour o f Melville J. Herskovits,” Current Anthropology 4, no. 3 (1963): 307-
316. Gluckman maintains that gossips and scandals “are enjoyed by people
about others with whom they are in a close social relationship." He continues:
“The right to gossip about certain people is a privilege which is only extended
to a person when he or she is accepted as a member o f a group or set. It is a
hallmark o f membership" (ibid., p. 313). On gossip networks, see also A. L.
Epstein, “Gossip, Norms and Social Network," in Mitchell, ed.. Social Net
works. Ruth Behar and Sonya Lipsett-Rivera respectively discuss the impor
tant function o f gossip in controlling women's behaviors for the sake o f patri-
archalism in the case of colonial Mexico. See Ruth Behar, “Sexual Witchcraft,
Colonialism, and Women’s Powers: Views from the Mexican Inquisition," in
Asunción Lavrin, ed., Sexuality and Marriage in Colonial Latin America (Lin
coln: University o f Nebraska Press, 1989); Sonya Lipsett-Rivera, “A Slap in the
Face o f Honor: Social Transgression and Women in Late Colonial Mexico," in
Lyman L. Johnson and Sonya Lipsett-Rivera. eds.. The Faces o f Honor: Sex.
Shame, and Violence in Colonial Latin America (Albuquerque: University of
New Mexico Press, 1998).
Epstein, Ethos and Identity; p. 14.
3. T h e Representation o f Identity, 1808-1831

J. da Silva Campos, “ Ligeiras notas sobre a vida intima; costumes e religiáo
dos africanos na Bahia," AAEB 29 (1943): 291-309.
Rodrigues, Os africanos, pp. 174-175. In the mid-i870s such cantos were princi
pally located on Rúa do Alvo, or Rúa dos Nagós as it was popularly known,
and, to a lesser degree, on other streets o f Santana parish. A large number o f
the African-born population congregated in public streets of Santana, Sé, and
Pilar parishes. See Campos, “ Ligeiras notas," pp. 292-293.
Rodrigues, Os africanos, p. 175.
Campos, “ Ligeiras notas," p. 291. In the 1870s, they cut hair for half o f a
pataca (160 réis) for every four persons. See ibid., p. 293.
Wetherell, Brazil Stray Notes from Bahia, p. 33. Wetherell also stated in 1843:
“The blacks have doctors amongst themselves; no matter what ails them,
they bind a handkerchief round their heads and consider that a cure for all

disorders” See ibid.» p. 43. His italics. O n the same functions o f African-born
barber-surgeons in nineteenth-century Rio de Janeiro» see Ewbank> Life in
Brazil, pp. 282,293. On barber shops in Rio» see Jean Baptiste Debret, Viagem
pitoresca e histörica ao Brasil, trans. Sergio Millet, vol. 1 (Säo Paulo: Martins
and Editora da Universidade de Säo Paulo, 1972)* pp-151-154.
Lindley> Narratives of a Voyage, pp. 123-124.
Ibid., pp. 71-72.
C. R. Boxer, “ Some Remarks on the Social and Professional Status o f Physi
cians and Surgeons in the Iberian World, i6th-i8th Centuries,” Revista de
histöria de Säo Paulo 50, no. 1 (1974): 209. In the years 1741-1749, all thirty-
eight licensed barber-surgeons in Salvador were people o f African descent:
seventeen slaves and twenty-one free blacks or mulattoes. In the years 1810-
1822, the thirty-three licensed barber-surgeons in Salvador were composed of
twenty slaves and thirteen free men o f African descent. See AMCS, vols. 191
and 193, cited by Russell-Wood, Black Man in Slavery and Freedom, p. 57. In
every city o f the U.S. South, barbering was the most important and lucrative
work for black men, so that this trade had been closely identified with free
blacks by the eve o f the Civil War. See Ira Berlin, Slaves without Masters: The
Free Negro in the Antebellum South (New York: Oxford University Press, 1974)»
pp. 235-236.
Russell-Wood, Black Man in Slavery and Freedom, p. 57.
See, for instance, Wetherell, BrazilStray Notes from Bahia, p. 54.
Presidlncia da Provincia Conselho da Provincia (correspondencia recebida),
December 2,1828, article 7, APB, ma$o 1070-4.
Albert J. Raboteau, Slave Religion: The invisible Institution* in the Antebellum
South (New York: Oxford University Press, 1978), p. 15.
Vilhena, Bahia, v o l 1, p. 134*
Arquivo Nacional do Rio de Janeiro, Rio de Janeiro, Brazil (hereafter ANRJ),
IJJ9-323» f* *>.
On Salvador and Rio de Janeiro as the two major port cities o f colonial Brazil,
see A. J. R. Russell-Wood, “Ports o f Colonial Brazil,” in Franklin W. Knight
and Peggy K. Liss, eds.» Atlantic Port Cities: Economy, Culture, and Society in
the Atlantic World, 1650-1850 (Knoxville: University o f Tennessee Press, 1991),
pp. 219-229.
See, for instance, O Oculo Mdgico, October 11,1866, which reported batuques
in Säo Miguel, Brotas, and Cabula, where the population o f African descent
was concentrated. The same article also referred to the funeral feast or celebra
tion which took place among “negros da Costa” in January 1857.
Religion o f African origins, syncretized with Christianity in the New World
through the history o f slavery, is called santeria in Cuba, shango in Trinidad,
vaudou in Haiti. See Raboteau, Slave Religion, p. 16. See, for an excellent an
thropological study o f santeria in the present-day United States, David Hilary

Brown» “Garden in the Machine: Afro-Cuban Sacred Act and Performance in
Urban New Jersey and Mew York” (Ph.D. diss., Yale University, 1989).
18. Studies on candomblé in Bahia include Pierson, Negroes in Brazil chapter io;
Julio Santana Braga, Ojogo de búzios: um estudo da advinaçào no candomblé
(Sâo Paulo: Editora Brasiliense, 1988); Joáo José Reis, “ Magia jeje na Bahia: a
invasáo do calundo do Pasto de Cachoeira, 1785” Revista Brasileira de História
8, no. 16 (1988): 57-81; Fayette Darcell Wimbery, “The African Liberto and the
Bahian Lower Class: Social Integration in Nineteenth-Century Bahia, Brazil
1870-1900” (Ph.D. diss., University o f California at Berkeley, 1989), chapters 4-
6; and Rachel E. Harding» A Refuge in Thunder: CamdombU and Alternative
Spaces o f Blackness (Bloomington: Indiana University Press, 2000).
19. For views on slave cultural independence expressed by colonial authorities
and slave owners in eighteenth-century Brazil, see Schwartz, Sugar Planta
tions, p. 484. See also George M. Fredrickson, The Black Image in the White
Mind: The Debate on Afro-American Character and Destiny, 2817-1914 (New
York: Harper and Row, 1971); and Celia Maria Marinho de Azevedo, Onda
negra, meào branco: 0 negro no imaginário das elites no sécub XIX (Rio de
Janeiro: Paz e Terra, 1987).
20. APB, Livro de registro de testamentos, no. 42, fs. 34-36 (October 13,1855).
21. Ibid., no. 27, fs. 128-130B (December 18,1839).
22. Ibid., no. 29, fs. 82B-85. The daughter was her heiress, and also served as ex
ecutrix o f her will. On reunions o f African-born kinsfolk in Salvador, see also
ibid., no. 13, fs. 61B-68; no. 39, is. 27-28; no. 40, fs. 76-78; and no. 40, fs. 169B-
171B.
23. For the studies o f slave families in nineteenth-century Salvador, see, for
instance, Kátia M. de Queirós Mattoso, Familia e sociedade na Bahia do
século XIX, trans. James Amado (Sâo Paulo: Corrupio, 1988); idem, “Slave,
Free, and Freed Family Structures in Nineteenth-Century Salvador, Bahia,”
Luso-Brazilian Review 25, no. 1 (1988): 69-83.
24. A. J. R. Russell-Wood, personal communication.
25. APB, Seçâo Judiciária, Livro de registro de testamentos, no. 19, fs. 276B-280B.
26. The 3,516 manumission letters consulted for the years 1808 to 1884 include
only four letters in which slaves were recorded as being married. Both in
the complete census records o f Sâo Pedro Velho parish taken in 1835 and
in the surviving census records o f Salvador in 1855, no slave was officially
recorded as married. The 470 listings in the marriage register o f Penha parish
for the half century (1808-1857) include only fourteen male slaves and seven
teen female slaves who had contracted marriages. Among them there are
twelve instances in which both partners were slaves. The marriage register
o f Conceiçâo da Praia parish for the period 1843-1888 has 763 listings, which
do not reveal any marriage in which a slave was one o f the partners. See APB,
Livros de notas and maço 2880; APB, Seçâo Histórica, maço 5685, Registro
dos fogos e habitantes da freguesia de Sâo Pedro; APB, maços 1549,1602 and

i6o$; ACMS, Livros de casamentas, freguesia da Penha; and ACMS, Livros de
casamentes, freguesia da C o n ce d o da Praia. It has long been discussed and
emphasized in historical literature on Brazilian slavery that in colonial and
imperial Brazil, slave marriage was sanctioned by the Church and legally rec
ognized by the State; some travelers from Europe and the U.S. observed and
recorded marriage ceremonies for slave couples, which their wealthy owners
sponsored. Examples are: Professor and Mrs. Louis Agassiz, A Journey in Bra
zil (Boston: Ticker and Fields, 1868), pp. 129-130; Debret, Viagem pitoresca,
pp. 174-175. 1 am grateful to Professor William B. Taylor and Mrs. Barbara
Taylor for their generous gift of a rare copy o f A Journey in Brazil
Schwartz, Sugar Plantations, pp. 387-388; Karasch, Slave Life, pp. 287-288.
In view o f inheritance, non-property owners, particularly the enslaved who
themselves were defined as property and had no legal right to possess any
form o f property, had no reason to contract matrimony for the succession
o f property. For instance, few Indian slaves were married in New Mexico for
the years 1694-1846 despite the fact that civil and ecclesiastical laws in New
Mexico guaranteed them the right to marry freely. See Ramón A. Gutiérrez,
When Jesus Came, the Com Mothers Went Away: Marriage, Sexuality, and
Power in New Mexico, ¡500-1846 (Stanford, Calif.: Stanford University Press,
1991)» p. 293.
APB, Livros de registro de testamentos. Karasch finds the same preference
for the same nation among the enslaved African-born population in early
nineteenth-century Rio de Janeiro. See Karasch, Slave Life, p. 293.
In Brazil, throughout the colonial and imperial periods, marriage, at least in
the official record, referred to church marriage only; no civil marriage was rec
ognized as legal by the State until 1890, when Brazil adopted civil marriage.
ACMS, Livros de batizados, freguesia de Santo António.
Ibid., September 24,1909; December 10,1809; December 26,1809; March 4,
1810; and October 21,1810. Nineteenth-century Salvador contrasts with the
data derived from the four Recóncavo parishes in the eighteenth and early
nineteenth centuries. Stephan Gudeman and Stuart B. Schwartz state: “No
pattern o f godparents from the same African nation can be seen in the docu
ments, and we believe that masters appointed or ‘invited’ more acculturated
slaves to serve as sponsors because o f their ability to assist in the baptized s
integration into the work force, their defining role.” See Stephen Gudeman
and Stuart B. Schwartz, “Cleaning Original Sin: Godparenthood and the Bap
tism o f the Slaves in Eighteenth-Century Bahia,” in Raymond T. Smith, ed.,
Kinship Ideology and Practice m Latin America (Chapel Hill: University o f
North Carolina Press, 1984), pp. 51-52.
The only exception in the baptismal records o f Santo António parish is
ACMS, Livros de batizados, freguesia de Santo António, October 7,1810.
Manoel Querino, Costumes africanos no Brasil (Rio de Janeiro: Civilizado
Brasileira, 1938), p. 155*

35 - Ibid., pp. 154-156.
36. APB, maço 3566/8, cited by TYosko, “The Liberto in Bahia,1*p. 47.
37. Georges Raeders, O inimigo cordial do Brasil: 0 con dì de Gobineau no Brasil,
trans. Rosa Freire d’Aguiar (Sào Paulo: Paz e Terra, 1988), p. 122.
38. In Brazil, throughout the colonial period and the early nineteenth century,
black lay sodalities proliferated in areas where a myriad o f Africans were
intensively imported as slave labor, especially in Bahia, Pernambuco, and
Minas Gerais. For African American beneficiary associations during the slav
ery regime, see Betty M. Kurk, “The African Deviation o f Black Fraternal
Orders in the United States,” Comparative Studies in Society and History 25,
no. 4 (1983): 559-591.
39. To my knowledge, in the state o f Bahia, there is only one existent black sodality
whose membership was limited to women: the sodality o f Our Lady o f the
Good Death (irmandade da Nossa Senhora da Boa Morte), which was estab
lished by slave women in 1832 in the Recôncavo town o f Cachoeira. This black
sodality— sisterhood— has never accepted any male members. According to
anthropologist Sheila S. Walker, the black sodality o f Our Lady o f Good
Death was established in the Barroquinha Church in Salvador in 1821 and ex
isted in major Bahian towns, but all disappeared except the one in Cachoeira.
See Sheila S. Walker, “The Feast o f Good Death: An Afro-Catholic Emancipa
tion Celebration in Brazil,” Sage 3, no. 2 (1986): 27-31,29. See also Wimbery,
“The African Liberto,* pp. 176-188.
40. Primary criterion for membership o f the Santa Casa da Misercórdia o f Bahia
(founded in Salvador in 1549), was “purity o f blood, without any trait of
Moorish or Jewish origin, both in the applicant and his wife.” Likewise, the
Third Order o f St. Dominic (founded in 1723), which was established by
very successful immigrants from Porto, Viana do Minho, and Lisbon, did
not accept Indians, people o f African descent, Jews, or even poor whites as
its members, and maintained the requirement o f “purity of blood” ( limpeza
de sangue). Neither did the Third Order o f St. Francis (founded in 1635). See
A. J. R. Russell-Wood, Ftdalgos and Philanthropists: T h e Santa Casa da M is
ericòrdia o f Bahia, 1550-1755 (Berkeley and Los Angeles: University o f Califor
nia Press, 1968), p. 124; and Joâo José Reis, A morte é urna festa: ritos funebres e
revolta popular no Brasil do siculo X IX (Sào Paulo: Companhia das Letras,
1991)» p. 53 .
41. A. J. R. Russell-Wood, “ Black and Mulatto Brotherhoods in Colonial Brazil: A
Study in Collective Behavior,” H A H R 54, no. 4 (1974): 576.
42. Ibid., p. 595; idem, “Examination o f Selected Statutes o f Three African Broth
erhoods,” in Albert Meyers and Diane Elizabeth Hopkins, eds., M anipulating
the Saints: Religious Brotherhoods and Social Integration in Postconquest Latin
America (Hamburg: WAYASBAH, 1988), pp. 224-245; and Manoel Cardozo,
“The Lay Brotherhoods o f Colonial Bahia,” Catholic Historical Review 33, no. 1
(1947): 24-30.

43- Stuart B. Schwartz's study o f manumission in colonial Bahia does not report a
single case in which a slave purchased his or her freedom with a loan from a
sodality. See Stuart B. Schwartz, “The Manumission o f Slaves in Colonial
Brazil: Bahia, 1684-1745»" H A H R 54, no. 4 (1974): 603-635. Manumission let
ters registered in Salvador during the nineteenth century do not include such
a case either. Wills o f ex-slaves included only one case, in which an African-
born slave man from Costa da Mina purchased his freedom with a manumis
sion fund borrowed from a black sodality o f the Rosary in Lisbon. See Mieko
Nishida, “ Manumission and Ethnicity in Urban Slavery: Salvador, Brazil, 1808-
1888,” H A H R 73, no. 3 (1993): 385; APB, Seçâo Judiciária, Livres de registro de
testamentos, no. 23, fis. 186-189.
44. Arquivo Histórico Ultramarino (Lisbon), Bahia, papéis avulsos, caixa 48 (July
8,1733). cited by Stuart B. Schwartr, “ Plantations and Peripheries, c. 1580-^.
1750,” in Bethell, ed., Colonial Brazil pp. 138-139.
45. This black church was often called the Church o f Our Lady of the Rosary of
Black People on Shoemakers Street (Igreja de Nossa Senhora do Rosàrio dos
Homens Prétos da baixa dos Sapateiros) o r . . . at the gates o f the Carmelite
monastery ( . . . nas Portas do Carmo) because o f its geographical location.
46. “Compromisso da Nossa Senhora do Rosario dos pretos da praya, ano de

1686,” chapters 6 and 19, published in A. J. R. Russell-Wood, Society and
Government in Colonial Brazil 1500-1822 (Brookfield, Vt.: Variorum, 1992),
pp. 250-253; Cardoso, “ Lay Brotherhoods,” pp. 24-30. Chapter 2 of the 1699
statutes o f the black sodality o f Saint Anthony o f Catagerona in Sâo Pedro
parish stipulates that a judge, a scribe, a treasurer, a procurator, and an
unspecified number o f majordomos should be elected by crioulo members,
while a judge, a scribe, a treasurer, a procurator, and an unspecified number
o f majordomos also were to be elected by Angolan members. Brazilian-born
black women (crioulas) and Angolan women should elect their respective
officers in the same manner. Manuscript sources on this sodality, including its
statutes dated 1699 and 1764, are located in the historical manuscript collec
tion o f the Oliveira Lima Library at the Catholic University o f America. See
Cardozo, “ Lay Brotherhoods.”
47. Arquivo da Igreja da Nossa Senhora do Rosàrio dos Homens Prétos, Salvador,
Brazil (hereafter AINS), Livro de entradas dos irmâos (1722-1786). Among
those who were admitted to the sodality between 1722 and 1786, only 1,959
members (720 men and 1,239 women) were listed with some indication o f
legal status and/or origin. They comprised 1,319 slaves (67.3%)» 259 ex-slaves
(13.2%), and 36 whites (1.8%]. African-born members with specific ethnic
identifications amounted to 238, while 264 Brazilian-born members were
counted. New members o f African birth (total 198) were composed o f 115
West Africans (103 Géges, 40 Minas, 10 Nagós, and 2 Calabars), 69 Angolans,
14 others (13 Benguelas and 1 Mozambique), while Brazilian-born members
(total 264) were divided into 142 crioulos and 122 mulattoes.

48. For the years from 1798 to 1865 the black sodality o f Our Lady o f the Rosary
in the Pclourinho accepted 1,50$ new members. See Jcferwn Afonso Bacelar
and Maria Conceiçâo Barbosa de Sousa, O Rosàrio dos Prêtos do Pelourinho
(Salvador: Fundaçâo do Patrimònio Artistico e Cultura da Bahia, 1974),
pp. 15-19* The African-born members were composed o f 126 West Africans
(97 Gegês, 20 Minas, and 9 Nagô), 35 Angolans, and 11 others (9 Benguelas
and 1 Mozambique). The Church has lost the Livro de entradas de irmâos
(1798-1865), which Bacelar and Sousa used for their research. This black
sodality still keeps a register o f admission o f new members, listed in alpha
betical order, for part of the nineteenth century. This booklet listed 282 mem
bers (31 males and 251 females): 176 slaves, 55 ex-slaves, and 51 whites and free-
born people o f color. They included 33 African-born (18 Gíges, 4 Angolans, 3
Nagôs, 1 Mina, 1 Congo, 6 unspecified), 36 crioulos, 23 mulattoes, and 9 whites.
See AINS, Livro de lançamento de irmàos, século XIX. The exact years which
this register o f admission covered are not mentioned.
49. Arquivo Histórico Ultramarino (Lisbon), Documentos da Bahia, no. 12235,
cited by Verger, Trade Relations, pp. 464-65; also cited by Russell-Wood,
“ Black and Mulatto Brotherhoods,” p. 573, note 12.
50. Verger, Trade Relations, p. 465. According to Pierre Verger, these sodalities
were exclusive to specific African nations in their membership (ibid.), but
there is no evidence available to support his assertion.
51. Nishida, "Manumission and Ethnicity,” pp. 283-284.
52. “Compromisso da Nossa Senhora do Rosario dos pretos da praya, ano de
1686,” chapters 6 and 19, published in Russell-Wood, Society and Government,
pp. 250-253; and AINS, Compromisso da Irmandade da Nossa Senhora dos
homens prétos no ano de 1820. In order to solve the problems derived from
interna] power struggle and discord, the statutes o f the black sodality were
modified in 1769, and further chapters were added in 1820. See Russell-Wood,
“ Examination o f Selected Statutes,” p. 224.
53. Patricia A. Mulvey, “ Black Brothers and Sisters: Membership in the Black Lay
Brotherhoods o f Colonial Brazil,” Luso-Brazilian Review 17, no. 2 (1980): 258.
In the new statutes o f the black sodalities o f the Our Lady of the Rosary in
the Pelourinho (1820), the scribe and treasurer had to be able to read, write,
and count, and this qualification meant that no longer were such elective
posts to be exclusively held by whites. This regulation also applied to caretak
ers. There were some changes in the new statutes. Provision o f funds for the
purchase o f freedom by slave members had been eliminated, but charitable
activities themselves had been expanded to reach out to the sick, poor, and
imprisoned. Interestingly, they had extended its religious celebrations, with
the new annual service held in honor o f the patron saint, in addition to the
mass and sermon on the third o f October, which had been stipulated in the
old statutes o f 1686. See AINS, Compromisso da Irmandade da Nossa Senhora
do homens prétos no ano de 1820. On the dependent nature o f African Brazil
ian Catholicism in the slavery regime in general, see Roger Bastide, The Afri-

can Religions of Brazil: Toward a Sociology o f the Interpenetration of Civiliza
tions, trans. Helen Sebba (Baltimore: Johns Hopkins University Press, 1978)»
pp. 109-125.
54. Russell-Wood, “ Black and Mulatto Brotherhoods,” p. 587.
55. Wetherell, Brazil Stray Notesfrom Bahia, pp. 113-114.
56. See Reis, A morte i uma festa, p. 54.
57. Bastide, African Religions, p. 113. See also Mulvey, “ Black Brothers and Sisters,”
p. 256; Raboteau, Slave Religion, p. 24. Out o f the 165 black sodalities o f colo
nial Brazil Mulvey studied, 86 were dedicated to Our Lady o f the Rosary. See
Mulvey, “ Black Brothers and Sisters,” p. 256.
58. Koster, Travels in Brazil, vol. z, p. 240.
59. Ibid., vol. 2, p. 241. See also ibid., vol. 1, pp. 273-275.
60. Bastide, African Religions, p. 113.
61. Ibid., pp. 113,116.
62. Pierre Verger, Noticias da Bahia— 1850 (Sao Paulo: Corrupio, 1981), p. 66.
63. Raboteau, Slave Religion, pp. 19,24.
64. The edifice o f the church still stands, empty and in a state o f disrepair, but
the statues o f the saints and documents have been stolen. According to Sheila
Walker, the first candombU temple in Salvador was established by women
from the Barroquinha Church. See Walker, “ Feast o f Good Death,” p. 29.
65. AMCS, Livro de posturas, fs. 119.1, fs. 63,152 (February 12,1710). In her sample
o f 100 wills o f ex-slaves (69 men and 31 women) for the period 1863-1890,
K£tia Mattoso found two wills o f freedmen who stipulated that they be bur
ied in accordance with African customs. See K4 tia M. de Queir6s Mattoso,
Testamentos de escravos libertos rta Bahia no ticulo XIX: uma fontepara 0
estudo de mentalidades (Salvador: Centro de Estudos Bahianos, 1979), p. 25
(Table 4).
66. Raboteau, Slave Religion, pp. 12-13.
67. In the city o f Rio de Janeiro, the American Reverend D. P. Kidder identified
the funeral he had witnessed o f an African-born person in Engenho Velho in
1839 with the custom held on the Gabon River o f Africa. See Fletcher and
Kidder, Brazil and the Brazilians, p. 136. Also see Conrad, Children of God*s
Fire, pp. 147-149.
68. Prince, “Slave Rebellion in Bahia,” pp. 86-151; Verger, Trade Relations, pp. 250-
294; Schwartz, Sugar Plantations, pp. 472-488; Reis, Rebeli&o, pp. 64-83; idem,
Slave Rebellion, chapter 3; idem, “Slave Resistance in Brazil: Bahia, 1807-1835,”
Luso-Brazilian Review, 25, no. 1 (1988): 111-144, especially 118-124.
69. Gardner, Travels in the Interior of Brazil, p. 20. Another example is Koster,
Travels in Brazil, vol. 2, p. 258.
70. Schwartz, Sugar Plantations, p. 475; Lovejoy, Transformations in Slavery,
pp. 140,142.

71. Goody, “Writing, Religion, and Revolt,” especially p. 333.
72. Schwartz, Sugar Plantations, p. 470. Mu-Kambo (mocambo) is an Ambundu

word meaning “hideout.” See Stuart B. Schwartz, “The Mocambo: Slave Resis
tance in Colonial Brazil,” Journal o f Social History s, no. 4 (1970): 316, note 12,
Quilombo derives from the Angolan term kilombo. See Joseph C. Miller, Kings
and Kinsmen: Early Mbundu States in Angola (Oxford: Oxford University
Press, 1976), p. 162 and chapter 8. According to Zoé Strother, who has con-
ducted her field research on art history in Zaire, kilombo means the place
for boys* initiation ceremonies for the Pende, who originally migrated from
Angola and live in rural Zaire near the border with Angola. Strother, personal
communication. Studies on fugitive slave communities in colonial Brazil in
clude: R. K. Kent, “ Palmares: An African State in Brazil,” Journal of African
History 6, no. 2 (1965): 161-175; Schwartz» **Mocamboni idem, Slaves, Peasants,
and Rebels: Reconsidering Brazilian Slavery (Urbana: University o f Illinois
Press, 1992), chapter 4*
73. Schwartz, Sugar Plantations, pp. 470-472; idem, Slaves, Peasants, and Rebels,
pp. 104-109; idem, "Mocambo” For instance, in 1789, fifty of the three hun
dred slaves o f Engenho Santana in Ilhéus (Bahia) fled to the forest to establish
a mocambo near the engenho. Schwartz documents a peace treaty proposal,
written by the slaves themselves, stating the conditions under which they
would return to slavery. See Stuart B. Schwartz, “ Resistance and Accommoda
tion in Eighteenth-Century Brazil: The Slaves* View o f Slavery," HAHR 57,
no. 1 (1979): 69-81.
74. The habit o f running away was, o f course, not uncommon, as the Ameri
can medical doctor James McFadden Gaston observed in Salvador at mid
nineteenth century. Iron shackles were sometimes put on slaves* ankles to
prevent them from running away. Despite their owners’ efforts, slaves took
flight, at times on a daily basis. See James MacFadden Gaston, Hunting a
Home in Brazil: The Agricultural Resources and Other Characteristics of the
Country— Also, the Manners and Customs o f the Inhabitants (Philadelphia:
King, Baird, Partners, 1867), p. 123. Local newspapers advertised such occur
rences and the rewards offered by their owners. Slaves “disappeared” while
going on errands, or did not bring to owners their daily or weekly wages as
wage-eamers at the agreed-upon time. It would have been intriguing to study
maroon communities in the suburbs o f Salvador in relation to the creation of
ethnic and gender identities among the fugitive slaves, since the majority of
them were African-born men. However, given the lack o f documentation, I
found it very difficult to research the subject. I thank Professor Joseph C. Love
for drawing my attention to identity formation among the maroon slaves.
75. AMCS, Livro de posturas, no. 119.4, fs. 32-32B (November 28,1733). The
municipal council determined a bush captain's daily wage according to the
distance from the city to where a fugitive slave was located. For instance, in
1733 he was supposed to be paid 280 réis in Itapoam; 320 réis at the limits of
the city and as far as Soledade, Forte de Sào Pedro, and Agua dos Meninos;
196 Notes to page 63

48o réis in Barra, Rio Vermelho, and Brotas; 640 réis one league distant from
Salvador; i $28o réis three leagues from the owner’s house; 2$000 réis in Rio
Joanes. See ibid.
76. Schwartz, uMocambo,np. 322. For instance, in the case o f the quibmbo known
as the Buraco de Tatu (“Armadillo’s Hole” ) located just east-northeast o f Sal
vador and destroyed by a Portuguese-led military expedition in 1763 after
twenty years o f existence, urban residents o f African descent in Salvador
helped the quibmbo inhabitants enter the city at night to steal powder and
shot. Ibid., p. 329.
77. Schwartz, Sugar Plantations, p. 479.
78. “Officio do Govemador Conde da Ponte para o Visconde de Anadia, no qual
o informa das provincias que tomara para destruir os Quibmbos, formados
pelos escravos fugidos dos seus senhores. Bahia, 7 de abril de 1807," Anais de
Biblioteca National do Rio de Janeiro (hereafter ABNRJ) 37 (1915): 450-451.
79. Ibid., p. 450.
80. Ibid., p. 450.
81. Schwartz states that the slave who denounced the uprising to the police
was “presumably not a Hausa,” but does not present any data to support his
assumption. See Schwartz, Sugar Plantations, p. 480.
82. ANRJ, IJJ9.319, fs. 1,11-13; “Officio do Governador Conde da Ponte para o
Visconde de Anadia, em que Ihe da parte das provincias que adoptara para
evitar um lavantamento dos escravos contra os brancos, de que tivera denun-
cia. Bahia, 16 de junho de 1807” ABNRJ 37 (1915): 460-461. See also Prince,
“Slave Rebellion,” pp. 86-96; Vferger, Trade Relations, p. 288; and Schwartz,
Sugar Plantations, p. 480.
83. “ Dom da Terra chegou, /Cento e cinquenta acabou” See Henderson, History
o f Brazil, p. 339.
84. “Dom da Terra abalou, /Cinco e cinquenta ficou” See ibid., p. 339.
85. Prince, “Slave Rebellion,” p. 17; Reis, Rebeliào escrava, pp. 26-33; idem, Slave
Rebellion, pp. 13-20; Schwartz, Sugar Plantations, pp. 434-436. Bert Barickman
discusses the export-oriented agriculture and the slave economy in nineteenth-
century Bahia by examining the prices o f manioc flour (farinha). See B. J.
Barickman, A Bahian Counterpart: Sugar, Tobacco, Cassava, and Slavery in the
RecÔncavo, 1/80-1860 (Stanford, Calif.: Stanford University Press, 1998).
86. ANRJ, IJJ9.318, fis. 9,12-14. Nina Rodrigues emphasized the important role of
the Obgoni or Ohogbo, a powerful secret society governed by Yorubas in this
1809 uprising, but there is no written evidence extant to support his synthesis.
See Rodrigues, Os africanos, p. 87.
87- See, for example, APB, Série Ordens Régias, vol. 109, doc. 73A (September 4,
1810).
88. Gardner, Travels in the Interior of Brazil, p. 78.
89. ANRJ, IJJ9.322, f. 111.

90. Dent, Year in Brazil, p. 246.
91. ANRJ, IJJ9.323, fs. 196-198.
92. Ibid., fs. 280-281.
93. BNRJ, II-34.6.57. The sound o f Ave Maria meant eight p . m . See Marjoribanks,
Travels in South and North America, p. 42. The British merchant Thomas
Lindley stated in Salvador (1803): “After sun-set each evening the bells o f the
churches in Catholic countries slowly toll several times, for all Christians to
repeat their Ave-Maria (or prayer to the Virgin), and to return thanks for the
benefit o f the day. This is so universal, that at the sound of the bell, all passen
gers stop, uncover their heads, and comply with the ceremony" See Lindley,
Narrative of a Voyage to Brazil, p. 192.
94. ANRJ, IJJ9.323, fs. 17-25. See also ANRJ, IJJ9.233, fs. 55-59.
95- BNRJ, 11-34.6.57-

96. Ibid. It was a custom in Scotland, Wales, Ireland, and some parts o f the North
o f England, as well as in Salvador, that St. John's Eve was celebrated by bon
fires. James Wetherell described it in 1848 as follows: “An immense amount is
annually spent in fireworks, which, for some time before the day, are heard
exploding in various directions; on the eve the noise becomes universal. Some
times accidents occur, one description o f fireworks used being very danger
ous: a charge tightly rammed into a case, which, after gyrating through the
air with great force, explodes with a loud report, shattering the tube into
pieces. It is called ‘buscap^,’ from ‘buscar’— to seek, to search *pe,* the foot—
and has been known for centuries under this name.” See Wetherell, Brazil
Stray Notes from Bahia, pp. 23-24. St. John’s Day, strategically placcd six
months before Christmas, celebrates the birth o f St. John Baptist and the
summer solstice. St. John was transplanted to the New World, and in each
country throughout Latin America, it was adapted to the particular charac
ter developed there. In Venezuela, for instance, St. John was adopted by
the large black population in coastal plantations and has become com
monly known as “the saint o f the blacks” in the area called Barvovento. See
David M. Guss, “The Selling o f San Juan: The Performance o f History in an
Afro-Venezuelan Community,” American Ethnologist 20, no. 3 (1993): 452-453.
In a village named Curiepe, the original St. John, called San Juan Congo,
which had been replaced by the pale-skinned San Juan Bautista at least by
1870, was said to be black but was in fact light-skinned with Caucasian fea
tures, including blond hair. The anthropologist David M. Guss discusses that
it was the absence o f color that made San Juan such a powerful symbol of
it; the blackness he represented was that o f poverty and oppression since the
arrival o f the first slaves in the early 1500s. San Juan might not have appeared
black but he was indeed poor, with his broken fingers, lack o f toes, and irregu
lar skin (ibid., p. 465).
97. Schwartz, Slaves, Peasants, and Rebels, p. 139 and p. 157, note 4-
98. Henderson, A History of Brazil, p. 344; and Patricia Ann Aufderheide, “Order

and Violence: Social Deviance and Social Control in Brazil 1780-1840” (Ph.D.
diss., University o f Minnesota, 1976), p. 76,
99. On the theme o f communication among peoples o f African descent in colo
nial British America and the Caribbean at the revolution, see Julius Scott,
“Common Wind.” In Julius Scott’s thesis, sailors o f African descent played a
significant role as the agent o f news on the Haitian Revolution in the forma
tion o f “Afro-America” in the revolutionary era. One may wonder, however,
how the geographical boundary o f “Afro-America” could be defined in the
revolutionary era. For instance, Genovese discusses New World slave rebel
lions in the Age o f Revolution, including those in non-Anglo America. See
Genovese, From Rebellion to Revolution.
100. Prince, “Slave Rebellion,” pp. 117-125.
101. Ibid., p. 130.
102. Verger, Trade Relations, p. 293.
103. Ibid., p. 310, note 24.
104. On slave uprisings from 1822 to 1831, see APB, Se^áo Histórica, maô 2845
(Insurreifáo de escravos); Verger, Trade Relation^ pp. 292-293; Prince, “Slave
Rebellion,” pp. 126-145.
105. ANRJ, IJ1.705; APB, maô 2845, fs. 11-12B. See Prince, “Slave Rebellion”
pp. 146-151; Verger, Slave Relations, p. 294.
106. Schwartz, Sugar Plantations, p. 475. See chapter 1 for a lengthier discussion.
4. T h e Re-creation o f Identity, 1808-1831

1. Sidney W. Mintz, “Tasting Food, Tasting Freedom,” in Wolfgang Binder, ed..
Slavery in the Americas (Wurzburg: Kónighausen and Neumann, 1993), p. 257.
2. Studies on manumission in colonial and imperial Brazil have covered regions
(Bahia and Paraíba), and specific towns and cities (the city o f Rio de Janeiro,
Paraty, Sabará, Campinas, and Cachoeira and Sáo Felix o f the Bahian Recón
cavo). See Schwartz, “Manumission,” pp. 603-635; Kátia M. de Queirós Mat-
toso, “A propósito de cartas de alforria na Bahia, 1779-1850,” Anais de História
4 (1972): 23-52; Trosko, “The Liberto in Bahia”; Nishida, “Manumission and
Ethnicity”; Diana Soares de Galliza, O dedínio da escravidáo na Paraíba i8$o-
1888 (Joáo Pessoa; Editora Universitária, 1979)» chapter 4; Karasch, Slave Life,
chapter 11; Sidney Chalhoub, “ Slaves, Freedmen and the Politics o f Freedom
in Brazil: The Experiences o f Blacks in the City o f Rio,” Slavery and Abolition
10,no. 3 (1989): 64-84; idem, Visoes da liberdade: urna história das ultimas
décadas da escravidáo na corte (Sáo Paulo: Companhia das Letras, 1990);
J a m e s P a t r i c k K i e m a n , “ T h e M a n u m i s s i o n o f S la v e s i n C o l o n i a l B ra z il:
Paraty, 1789-1822” (Ph.D. diss., New York University, 1976); Kathleen Joan
Higgins, “The Slave Society in Eighteenth-Century Sabará: A Community
Study in Colonial Brazil” (Ph.D. diss., Yale University, 1987), chapter 4; Robert
Wayne Sienes, “The Demography and Economics o f Brazilian Slavery: 1850-

i888” (Ph.D. diss., Stanford University, 1975)) chapter 10; Peter L. Eisenberg,
“Ficando livre: as alforrias em Campinas no s&ulo XIX,” Estudos Ecdnomicos
17, no. 2 (1987): 175-216; and Wimbery, “The African Liberto,” chapter 2. For
studies on manumission in other parts o f Latin America, see Bowser, African
Slave in Colonial Peru, chapter 10; idem, “The Free Person of Color in Mexico
City and Uma: Manumission and Opportunity, 1650-1850,'* in Stanley L.
Engerman and Eugene D. Genovese, eds., Race and Slavery in the Western
Hemisphere: Quantitative Studies (Princeton, N.J.: Princeton University Press,
1975); Lyman L. Johnson, “Manumission in Colonial Buenos Aires, 1776-1810,**
HAHR 59, no. 2 (1979): 258-279.
Koster, Travels in Brazil, vol. 2, p. 230. See also ibid., vol. 1, p. 404.
Lei N. 2040 o f September 28,1871, Art. IV, Colle$&o das leis do Imperio do Bra
zil de 1871 (Rio de Janeiro: Tipografico Nacional, 1871), part I, p. 149; Robert
Edgar Conrad, The Destruction of Brazilian Slavery 1850-1888 (Berkeley and
Los Angeles: University o f California Press, 1972), p. 191; idem, “Nineteenth-
Century Brazilian Slavery," in Robert Brent Toplin, ed., Slavery and Race Rela
tions in Latin America (Westport, Conn.: Greenwood Press, 1974), pp. 154-155;
Manuela Carneiro da Cunha, “Silences o f the Law: Customary Law and Posi
tive Law on the Manumission o f Slaves in 19th Century Brazil," History and
Anthropology 1, pt. 2 (1985): 427-443, especially pp. 428-429.
Koster, Travels in Brazil, vol. 2, p. 231.
Ibid., pp. 231-232, note.
Manuela Carneiro da Cunha, “Silences o f the Law," p. 434. My italics.
Ibid., pp. 4 3 4 - 4 3 8 .
APB, Livros de notas and maô 2880. Concerning the advantage o f Brazilian-
bom slaves in manumission, see Schwartz, “ Manumission,1* p. 612; Karasch,
Slave Life, p. 352; Kieman, “ Manumission o f Slaves," p. 92; Higgins, “Slave
Society," p. 213; and Eisenberg, “Ficando livre," pp. 189-191.
APB, Livros de notas and maô 2880. On the constant 1:2 male-female ratio
among the manumitted in Bahia for 1684-1745» 1779-1850, and 1813-1850, see
Schwartz, “ Manumission," p. 611; Mattoso, “A prop6sito," p. 41; Arnold
Kessler, “ Bahian Manumission Practices in the Early Nineteenth Century"
(paper presented to the American Historical Association, San Francisco, 1973).
cited in Schwartz, “ Manumission," p. 611. The same sex ratio is found in Rk>
de Janeiro (1807-1831), Parat y (1789-1822), and Lima (1580-1650). See Karasch,
Slave Life, p. 345; Kieman, “Manumission o f Slaves," p. 86; and Bowser, “Free
Person," p. 350. Higher proportions o f women among the manumitted are
found in other cities and towns, such as Sabard, Minas Gerais (1710-1809),
Campinas, Silo Paulo (1799-1887), Mexico City (1580-1650), and Buenos Aires
(1776-1810). Sec Higgins, “Slave Society,” pp. 205-207; Eisenberg, “ Ficando
livre," pp. 184-185; Bowser, “ Free Person,” p. 350; and Johnson, “Manumission
in Colonial Buenos Aires,” p. 262.

11. APB* Livros de notas and maço 2880; AMCS, Livros de escrituras da compra
e venda de escravos, freguesia da Sé; APB, Inventários da capital (1808-1888).
On the nearly equal distribution o f geographic and ethnic origins of the
African-born slave population among the manumitted, see Schwartz, “ Manu
mission” pp. 612-614; Karasch, Slave Life, p. 345; and Kiernan, “ Manumission
o f Slaves," p. 93*
12. In the mining town o f Sabará, Minas Gerais, for the period 1710-1809» more
than twice as many Costa da Mina slaves were freed as from the Congo-
Angola region, despite the fact that the former outnumbered the latter only
by approximately 9 percent in the inventories dating from 1725 to 1808. See
Higgins, “Slave Society,” p. 216. See, also Russell-Wood, Black Man in Slavery
and Freedom, pp. 122-123.
13. APB, Livros de notas and maço 2880. Payment for freedom was made in the
Brazilian currency o f real (réis). The manumission letters I consulted for this
study do not include a case in which the price for freedom was paid in silver
dobras, or any other type o f currency.
14. On the prices o f manumitted slaves in nineteenth-century Salvador, see Kátia
M. de Queirós Mattoso, Herbert S. Klein, and Stanley L. Engerman, “Trends
and Patterns in the Prices o f Manumitted Slaves: Bahia, 1819-1888,1* Slavery
and Abolition 7, no. 1 (1986): 59-67. On the difference between estimated value
and sales price in Paraty, see Kiernan, “ Manumission o f Slaves,” p. 137.
15. AMCS, Livros de escrituras de compara e venda de escravos, freguesia da Sé.
16. APB, Livros de notas and maço 2880.
17. APB, maço 2898» Relaçào dos africanos . . . Santana, 1849. Unfortunately
there is no way o f making connections between the price o f freedom and
daily wages by occupations based on the data mentioned in the manumission
letters, which rarely state the occupation or skills o f the slave to be freed.
Among 3,516 manumission letters for the entire period 1808-1884, only 69
state the occupations o f slaves. The letters represent 46 men and 23 women:
29 African-born, 29 Brazilian-born, and 11 whose birthplaces were not stated.
The men's occupations include 13 artisans, 9 domestics, 8 transporters, 8 field
hands, 4 barber-surgeons, and 4 wage-eamers; the women's are 11 domestics,
10 field hands, and 2 wage-eamers. Letters o f liberty (manumission letters)
rarely mention the occupation or skills o f the slave to be freed. See APB,
Livros de notas and maço 2880. Few ex-slaves mentioned their occupations
in their wills, either. See APB, Livros de registro de testamentos.
18. APB, Livre de registro de testamentos, no. 13, fs. 8B-15 (January 28,1825).
19. Galliza, Declfnio da escravidâo, p. 150.
20. The term “trade-in” is borrowed from Karasch, Slave Life, pp. 343 and 358.
21. APB, Seçâo Judiciária, Livro de notas da capital, no. 306, May 14,1852.
22. Among the wills o f ex-slaves, seven ex-slaves, all African-born, declare that
they had provided their owners with substitute slaves, but only one was identi-

ñed in terms o f “nation” APB, Livros de registro de testamentos, no. 4, fe. 5B-
9B; no. 9, fs. ioiB-106; no. 16, fs. 156-161; no. 17, fs. 145B-149B; no. 2«, fs. 98B-
101; no. 28, fe. 122-125B; and no. 47» fe* 28-30B.
23. Maria and Thereza, both Nagó, in place o f Maria, Nagó (APB, Livro de notas,
no. 206, May 31,1822); and two newly arrived African-born slave women, with
no specific names nor places o f origin, in place o f Anna, Nagò (ibid., no. 206,
June 8,1822).
24. Bricklayer in APB, Livro de notas, no. 160, July 15,1809; chair carrier in
ibid., no. 242, April 10, 1832; field worker in ibid., no. 295, February 3,1851;
and stevedore in ibid., no. 306, May 14,1852.
25. APB, Livro de notas, no. 160, July 15,1809. The same settlement: ibid., no. 242,
April 10,1832; ibid., no. 304, May 7,1852.
26. Joaquim was replaced by his namesake Joaquim (APB, Livro de notas, no. 272,
March 9 >1841); and Caetana, crioula> replaced by another Caetana, Nagò, in
the same occupation o f fieldwork (ibid., no. 295, February 3,1851).
27. APB, Livro de notas, no. 293, May 22,1821; ibid., no. 238, February 16,1832.
28. Instituto Geográfico e Histórica da Bahia, Salvador, pasta 28, doc. 11, cited by
Schwartz, “ Manumission,” pp. 626-627 (note 36).
29. African-born ex-slaves who were engaged in the transatlantic slave trade:
APB, Livro de registro de testamentos, no. 19, fs. 276B-280B; ibid., no. 28,
fe. 98B-101B. See also chapter 3.
30. Miller, “Slave Prices,” pp. 45,47; and Mattoso, Klein, and Engerman, “Trends
and Patterns” p. 66 (note 9).
31. Schwartz, “ Manumission,” p. 626; Karasch, Slave Life, p. 358; Higgins, “Slave
Society,” p. 247.
32. Karasch, Slave Life, p. 358.
33. The practice o f substitution took place in Bahia (1684-1745), the city of
Rio de Janeiro (1808-1850), and Sabará o f Minas Gerais (1789-1822). See
Schwartz, “ Manumission,” p. 626; Karasch, Slave Life, p. 358; and Higgins,
“Slave Society,” pp. 247-248. Kátia Mattoso refers to substitution but does
not provide any interpretation: “Slaves had to pay for their freedom in hard
cash, metal coin or paper money, either in a lump sum or by installments.
Some slaves redeemed themselves by giving another slave to masters.” See
Mattoso, To B e a Slave in B razil p. 158.
34. Kieman, “Manumission o f Slaves,” especially p. 148; Galliza, O declinio da
escravidáo, chapter 4. Judith Lee Allen, who conducted her archival research
in 1987-1989 in Salvador and Cachoeira, found only four cases o f substitution
among approximately 1,700 manumission letters registered in the towns o f the
Bahian Recòncavo in the nineteenth century. Allen, personal communication.
35. Koster, Travels in B razil vol. 2, pp. 266-267.
36. William D. Phillips, Slavery from Roman Times, p. 28. On the peculium, see
also Patterson, Slavery and Social Death , pp. 182-186.

37. Carl Campbell, “John Mohammed Bath and the Free Mandingos in Trinidad:
The Question o f Their Repatriation to Africa 1831-5«,” Journal o f African Stud
ies 12, no. 4 (1975 ): 4 7 2 .
38. Maria Graham, Journal o f a Voyage to Brazil, p. 108.
39. APB, maço 647, Arollomento dos africanos libertos. . . Conceiçâo da Praia,
1846; APB, maço 2898, Relaçâo dos africanos libertos. . . Santana, 1849.
40. APB, maço 6472, Arollomento dos africanos libertos que rezedem nesta
freguésia da Conceiçâo da Praia, January 31,1846.
41. APB, maço 2898, Relaçâo dos africanos libertos existentes nesta freguesia, San
tana, February 11,1849. The total number o f African-born freedmcn is 133.
42. Andrade, A mào de obra escrava, p. 129; and APB, Inventários da capital.
43. APB, maço 6472, Arollomento dos africanos libertos. . . Conceiçâo da Praia;
ibid., maço 2898, Relaçâo dos africanos libertos. . . Santana.
44. APB, Livro de registro de testamentos, no. 40, fe. 19-21 (April 14,1849); ibid.,
no. 48, fs. 116-118B (August 21,1873).
4>. APB, Livro de registro de testamentos, no. 28, fis. 98B- io i B (August 20,1827).
46. Ibid., no. 11 (September 24,1824), fs. 32B-35B.
47. Ibid., no. 19, fs. 276B-280B.
48. Walsh, Notices o f Brazil, vol. 2, p. 362.
49. See, for instance, J. Michael Turner, “Les Brésiliens— The Impact of
Former Brazilian Slaves upon Dahomey” (Ph.D. diss., Boston University,
*9 7 5 )>PP-102-125; Pierre Verger, Formation d*une société brésilienne au Golfe
du Bénin au XIXème siècle (Dakar: Centre de Hautes Etudes Afro-Ibéro-
Américaines de L'Université de Dakar, 1969). On the commercial role of
Luso-Africans in Angola, see Miller, Way o f Death, chapter 8.
50. See, for example, the recollection o f one elderly Bahian-born man o f African
descent, in Pierson, Negroes in Brazil, pp. 240-242.
51. APB, Livro de registro de testamentos, no. 23, fs. 146B-149 (October 13,1833).
5z. Ibid., no. 25, fe. 83-85B (March 30,1837).
53. Some slaves, especially those of Brazilian birth, had been given family names
before manumission. See Schwartz, Sugar Plantations, p. 402.
54. APB, Livros de registro de testamentos.
55. Schwartz, Sugar Plantations, p. 402.
56. APB, maço 2898, Relaçâo dos africanos libertos existentes nesta freguesia,
Santana, February 11,1849; APB, Seçâo Histórica, maço 6472, Alloroment
dos africanos libertos que rezedem nesta freguezia da Conceiào da Praia,
January 31,1846.
57. Oliveira, O liberto, p. 41 (Table 2). Due to the rapid transformation o f the
larger society from slave society to slave-owning society, which started to take
place in Salvador during the 1840s (see my discussions in chapter 1) and was

spurred by the termination o f the transatlantic slave trade (1851), the percent
ages o f slave owners among the ex-slave property owners declined consider
ably after the mid-nineteenth century. For the period o f 1850-1890, only 35.2
percent o f men (45 out o f 128) and 57.9 percent o f women [45 out o f 95) were
slave owners among the ex-slave population who registered their wills. See
ibid.
My interpretation deviates from Karasch on slaveholding as practiced by
African-born ex-slaves. She states: “One other factor that appears to have
encouraged ex-slaves to own slaves is that while the African trade continued,
they had the opportunity to rescue their own people from being enslaved to
Brazilians. This pattern is suggested in the negros de ganho records in which
Yorubas owned other Yorubas as slaves.” See Karasch, Slave Life, p. 211.
APB, Livros de registro de testamentos. On crias as heirs, see, for example,
ibid., no. 43, fs. 50-52. On the legal arrangement o f coartado, see Schwartz,
"Manumission,” pp. 627-628. Cf. On the custom o f coartaáón in Spanish
America, see Hubert H. S. Aimes, “Coartación: A Spanish Institution for
the Advancement o f Slaves into Freedom,” Yale Review 17 (1909): 412-431;
Herbert S. Klein, Slavery in the Americas: A Comparative Study o f Virginia
and Cuba (Chicago: University o f Chicago Press, 1967), pp. 196-200.
APB, Livros de registro de testamentos, nos. 3-34.
APB, Livro de registro de testamentos, no. 5, fe. 202B-20Ó. Ana Maria died on
October 15,1815.
Ibid., no. 31, fs. 154-157B (February 6,1844).
5. T h e Convergence o f Identity, 1831-1880
Decreto de 14 de dezembro de 1830, Collecfóo de leis do Imperio do Brazil de

1830 (Rio de Janeiro: Tipográfico Nacional, 1876), part I, pp. 96-98.
APB, livros de escrituras de compra e venda de escravos, freguesia da Sé; APB,
Inventários da capital.
Reis, Rebeliáo escrava, p. 16.
Studies on the Malé revolt o f 1835 include Etinne Ignace, "A revolta dos Malés
(24 para 25 de Janeiro de 1835),” Revista de Instituto Geográfico e Histórico da
Bahia 14, no. 33 (1907): 129-149; Rodrigues, Os africanos, chapter 2 (Os negros
maometanos no Brasil; originally published in the Jornal do Comércio do Rio
de Janeiro, dated November 2,1900); R. K. Kent, “African Revolt in Bahia:
24-25 January 1835,” Journal o f Social History 3, no. 4 (1970): 334-356; Prince,
Slave Rebellion, pp. 152-224; Reis, Rebeliáo escrava; idem, Slave Rebellion; idem,
"Slave Resistance,” pp. 124-132; and Lovejoy, “ Background to Rebellion.”
On the miraculous Church o f Bonñm, James Wetherell observed in 1856:
“Numbers o f devotees, particularly blacks and mulattoes, dressed in their
best and with bare feet, are seen wending their way each to present a candle as

an offering.. . . Wax models of arms, legs, breasts, 8c c., with different diseases
simulated thereon, are suspended round the room__ It was formerly the cus
tom o f the sailors belonging to the slave vessels to bring one o f their sails to
be blessed previous to departure; or having vowed a sail when in distress, they
would escort with bare feet the sail, tied up with garlands o f flowers, to the
Church, and then, after offering the same, redeem i t Upon the ceiling is a
large allegorical painting representing, amongst other things, sailors in the act
o f presenting a sail to Christ" See Wetherell, Brazil: Stray Notes from Bahia*
pp. 121-122. See also Lindley, Narrative o f a Voyage to Brazil, pp. 190-192. Ac
cording to the British medical doctor Robert Dundas, the “ higher classes" of
the inhabitants o f Bahia resorted to the suburb o f Bonñm for sea-bathing,
during the hottest months o f the year, namely December, January, February,
and March. See Dundas, Sketches o f Brazil* p. 240.
Verger, Trade Relations* p. 294.
“ Devassa do levante,” pp. 61-62.
Ibid., p. 63. According to Nina Rodrigues, the name o f “Aruna” was a reten
tion o f the Arabic term “Alufa,” which means teacher or priest. Rodrigues,
Os africanos* p. 9 4 *
The expression “white man's land” seems to have been commonly used
among the population o f African descent in nineteenth-century Salvador.
One old returnee from Bahia in Lagos, named Francisco, mentioned to Father
Baudin during his stay for 1874-1875 that he had been born in Bahia, “in the
terra dos brancos*” and continued to say, “I was happy then, in that bounteous
land o f Brazil! What beautiful churches, what lovely houses!” See Manuela
Carneiro da Cunha, “ Introduction" in Marianno Cameiro da Cunha, Da sen-
zala ao sobrado: arquitetura brasileira na Nigèria e na República Popular do
Benin (Sáo Paulo: Nobel, 1985), p. 42.
“ 1835 Insurreiq&o de escravos," pp. 9-170.
Prince, “ Slave Rebellion,” p. 169.
Ibid., p. 177, based on APB, ma$o 2847 (Insurrei^ào de escravos, 1835), fs. 1-18.
Verger, Trade Relations* p. 303.
Goody, “Writing, Religion, and Revolt,” p. 324.
Parkinson to the Duke o f Wellington, F O 13/21 (Bahia, January 29,1835)» cited
by Goody, “Writing, Religion, and Revolt," p. 324. On anti-British feelings pre
vailing among Brazilians because of Britain's interference with the slave trade,
see Marjoribanks, Travels in South and North America* p. 47.
On the severe shortage o f women in New World maroon societies, see
Richard Price, “Introduction: Maroons and their Communities," in idem,
ed., Maroon Societies: Rebel Slave Communities in the Americas* 2nd ed. (Balti
more: Johns Hopkins University Press, 1979), pp. 18-19.
Reis, Rebeliào escrava* p. 172.

18. "Devassa da levante” p. 135. Slaves were obliged to return home after the
sound o f Ave Maria, namely eight o'clock at night, and were prohibited from
going out after nine o’clock.
19. Reis» Rebeliào escrava, p. 172.
20. Ibid.
21. See the testimony by Pacífico in the court, in "Devassa da levante” pp. 84-85.
22. “ 1835 Insurreiçâo,” p. 40.
23. “Devassa da levante," pp. 72-73. See Verger, Trade Relations, pp. 300-301.
24. "1835 Insurreiçâo,” p. 13.
25. Bastide, African Religions, p. 106; Lovejoy, “Background to Rebellion ” Love
joy *s synthesis is largely based on a sample o f 108 slaves from the Central
Sudan who were sold into the transatlantic slave trade. See ibid., pp. 176-180.
26. Rodrigues, Os africanos, pp. 108-119; Prince, “Slave Rebellion,” p. 234.
27. Reis, Rebeliào escrava, parte 2; idem, Slave Rebellion, part 2.
28. Vincent Monteil, “Analyse de 25 documents arabes des Matés de Bahia (1835)”
Bulletin de l'Institut Fondanetake d'Afrique Noir, tomo 29, série B, no. 1-2
(1967): 88-98; and Rolf Reichert, Os documentos Arabes do Arquivo do Estado
da Bahia (Salvador: Universidade Federal da Bahía, 1970).
29. Mariane Ferme, personal communication.
30. Goody, “Writing, Religion, and Revolt,” pp. 328-329.
31. Eugene D. Genovese, From Rebellion to Revolution: Afro-American Slave
Revolts in the Making of the Modem World (Baton Rouge: Louisiana State
University Press, 1979), p. 44.
32. Goody, "Writing, Religion, and Revolt,” p. 332.
33- Ibid., p. 333.
34. Genovese, From Rebellion to Revolution, p. 32. Genovese continues: “ In the
hands o f a skillful anti-Christian leader the religious cry could be made to
separate the slaves from the white community and thus transform every ris
ing into a holy war against the infidel. When master and slave appealed to
the same God, the same book, the same teachings, the task of Nat Turners
became much more difficult__ The difference came not with the abstract
character o f the Christian tradition but with the reduction of revolutionary
potential inherent in the deeper separation o f religion from class and espe
cially ethnicity” (ibid.).
35. ANRJ, IJ1.707» no. 9.
36. Verger, Trade Relations, pp. 314-316.
37. APB, Seçâo Legislativa, Assembleia legislativa provincial, Registros de leis
approvadas pela assembleia, no. 1, Lei 8, fs. 9-16.
38. Ibid., fs. 11-11B (Article 8).

39- Ibid., fs. 14B-15 (Article 17).
40. APB, Livro de registro de testamentos, no. 40, fs. 29B-32 (August 31,1859). See
also ibid., no. 40, h . 186-188B (June 5* i860); ibid., no. 43» fs. 50-52 (August n,
1863); ibid., no. 46, fs. 172B-175 (March 25,1868); ibid., no. 47, fs. 28-30 (Au
gust 13,1872).
41. APB, Seçâo Legistlativa, Assembleia legislativa provincial, Registros de leis
approvadas pela assembleia, no. 1. f. 15.
42. Ibid., Lei 14, fs. 28B-30.
43. APB, Inventários da capital.
44. Títulos de residencia à africanos libertos, APB, Seçâo Histórica, maço 5664» f. 1.
45. British and Foreign State Papers, vol. 40 (London: James Ridway & Sons,
PicadiUy, 1850-1851), pp. 428-429.
46. ANRJ, IJ1.707, no. 32; Verger, Trade Relations, p. 316.
47 * Verger, Trade Relations, pp. 317-322.
48. Ibid., p. 563, note 3.
49. Manuela Carneiro da Cunha, Negros, estrangeiros: os escravos libertos e sua
volta à Africa (Sào Paulo: Editora Brasiliense, 1985), p. 80.
50. APB, Projecto de Lei no. 19 à Assembléia Legist rat iva da Provincia, 1851»
Art. 15, cited by Trosko, ‘‘The Liberto in Bahia,” p. 37.
51. On back-to-Africa movements from Salvador, see Lorenzo D. Turner, “Some
Contacts o f Brazilian Ex-Slaves with Nigeria, West Africa,” Journal o f Negro
History 27» no. 1 (1942): 55-67; Pierson, Negroes in Brazil pp. 238-239; Richard
D. Ralston, “The Return o f Brazilian Freedmen to West Africa in the 18th and
19th Centuries " Canadian Journal o f African Studies 3, no. 3 (1969): 577-592;
Verger, Formation d*une société Brésilienne; idem, Trade Relations, chapter 16;
J. Michael Turner, uLes Brésiliens”; Manuela Carneiro da Cunha, Negros,
estrangeiros; idem, “Introduction”; and Lisa A. Lindsay, “ ‘To Return to the
Bosom o f their Fatherland': Brazilian Immigrants in Nineteenth-Century
Lagos,” Slavery and Abolition 15, no. 1 (1994): 22-50.
$2. Fletcher and Kidder, Brazil and the Brazilians, p. 136.
53. Raeders, O inimigo cordial do Brasil, p. 122.
54. Mariano Carneiro da Cunha studies the Luso-Brazilian architecture o f these
West African coastal towns in the nineteenth century. See Mariano Carneiro
da Cunha, Da senzala ao sobrado.
55. John Duncan, Travels in Western Africa, in 1845 & 1846, Comprising a Journey
from Whydah, through the Kingdom of Dahomey, to Adofoodia, in the interior,
vol. 1 (New York: Johnson Reprint Corporation, 1967 [1847]), p. 138.
56. Ibid., vol. 1, p. 185.
57. On the “return” movement from Cuba, see, for instance, Rodolf Saracino, Los
que volvieron a Africa (Havana: Editorial de Ciencias, 1988).

58. Duncan, Travels in Western Africa, vol. 1, p. 186; and lean Herskovits Kopy toff,
A Preface to M odem Nigeria: The uSierra Leoneans” in Yoruba, 1830-1890
(Madison: University of Wisconsin Press, 1965).
59. APB, Livros de notas and maô 2880.
60. APB, Livros de registro de testamentos; ACMS, Livros de balizados, freguesia
de Santo António; and ACMS, Livros de casamentos, freguesia de Conceiô
da Praia, freguesia de Penha.
61. APB, Livro de registro de testamentos, no. 31, fs. 132B-135B.
62. Ibid., no. 10, fs. 91-96B (June 12,1823); ibid., no. 29, fs. 103-107 (March 3,1838);
ibid., no. 31, fs. 87-89B (April 6,1846); ibid., no. 40, ft. 29B-32 (February 13,
1857); ibid., no. 52, fs. 145B-149B (February 21,1873); ibid., no. 61, fs. 168B-170
(August 15,1887).
63. APB, Livros de notas and ma<;o 2880.
64. APB, Livro de registro de testamentos, no. 3, fs. 142B-145B (November i, 1841).
65. Ibid., no. 6, fs. 231-235 (May 2,1816).
66. Ibid., no. 51, fs. 141B-144B.
67. She named all o f her legitimate and natural children and her six grand
children (children o f her legitimate children) as her heirs. Ibid., no. 22, fs.
271-273B.
68. Ibid., no. 41, h . 64-65B (November 26,1861). For grandfathers, see also ibid.,
no. 30, fs. 63 (October 19,1840); ibid., no. 41, fe. 64-65B (November 26,1861);
ibid., no. 52, fs. 39-41 (March 9,1877).
69. 5 ,206B-210. O n g r a n d m o t h e r f i g u r e s , s e e a l s o i b i d , n o . 5, fs. 25B-
Ib ic L , n o .
28B ( M a r c h 17,1815); i b id ., n o . 22, fs . 4B-8 ( J u l y 29,1832); i b i d ., n o . 22, fs . 52-
55B ( O c t o b e r 9,1832); i b id ., n o . 31» fs. 102B-105 ( F e b r u a r y 27,1839); i b i d . ,
n o . 34, fs. 8-11B ( J a n u a r y 7,1837); a n d i b i d . , n o . 51, fs . 101-104 ( J u l y 27,1867).
70. Ibid., no. 30, fs. 57-59B.

71. Sandra Lauderdale Graham reports, albeit based on two civil court records,
on a similar case o f church marriage and separation o f an African-born ex
slave couple in nineteenth-century Rio de Janeiro, and discusses the meanings
o f honor for women among the urban poor. See Sandra Lauderdale Graham,
“ Honor among Slaves,” in Johnson and Lipsett-Rivera, eds., Faces of Honor.
72. APB, Livro de registro de testamentos, no. 46, fs. 43-45B.
73. Legitimate children bom to African-born ex-slaves: ibid., no. 42, fs.i6oB-
163B; ibid., no. 44, fs. 161-163B; ibid., no. 4 7 »(s. 107-109; ibid., no. 40, fs. 188B-
190B. Among the 325 wills o f ex-slaves, 214 (94 men and 120 women) never
had any children.
74. APB, Livro de registro de testamentos, no. 4 9 »fs* 165B-168.
75. ACMS, Livros de casamentos, freguesia da Conceúpáo da Praia.
76. APB, Livros de registro de testamentos, no. 43, fs. 16-18; ibid., no. 44. fs. 106B-
109. Other examples o f matrimony o f longtime common-law spouses before
208 Notes to pages ¡05-110

death include: ibid., no. 14, fs. 192-96 (June 8,1847); ibid., no. 60, fs. 55-58
(December 20,1884); ibid., no. 61, fs. 71B-77 (July 21,1886). The expression
“ illicit” union was commonly used in wills to refer to common-law marriage
unsanctified by holy matrimony.
77. Linda Lewin, “Natural and Spurious Children in Brazilian Inheritance Law
from Colony to Empire: A Methodological Essay,” Americas 48, no. 3 (1992):
363-368.
78. ACMS, Livros de batizados, freguesia de Santo Antônio.
80. The crioula Antônia, seventeen years o f age, was manumitted on payment of
900$000 réis by her godfather José de Sào Joào, Nagô (APB, maço 2880, Febru
ary 6,1879).
82. Ibid., no. 19, fs. 214-217B (January 7,1830).
83. Ibid., no. 23, fs. 270B-274 (September 23,1832).
85. Ibid.
86. Ibid., no. 44, fs. 195-198.
87. Ibid., no. 40, fs. 19-21.
88. Criminal records discussing kinship and other networks exclusively among
African-bom ex-slaves include: APB, Autos Crimes, 3552.4 (1865); and 3559.10
(1872).
89. Manuela Cameiro da Cunha, “ Introduction," p. 18. On the importance of
kola nuts as a trade commodity in Africa, see Paul E. Lovejoy, Caravans of
Kola: The Hausa Kola Trade 1700-1900 (Zaria, Nigeria: Ahmadu Bello Univer
sity Press, 1980).
90. Some African-bom ex-slaves still went back to their hometowns in the inte
rior, such as Abeokuta, Ilesha, and Oyo, despite ongoing wars and high risk of
re-enslavement. Manuela Cameiro da Cunha reports an oral tradition that Pa
Callisto and his parents were manumitted in Bahia and went back to Ilesha,
their hometown, where they were re-enslaved. Callistos father is said to have
committed suicide to escape from captivity. See Manuela Carneiro da Cunha,
“ Introduction,** p. 64, note 4.
91. Didrio da Bahia, May 7,1868; Jornal da Bahia, February 23,1857; and Didrio
da Bahia, April 29,1863. See also J. Michael Turner, “ Les Brésiliens,” pp. 55-63.
92. One example is Antônio de Silva, whose owner was Antônio Vieira da Silva
{Jornal da Bahia, March 5,1857).
93. APB, Livro de registro de testamentos, no. 46, fs. 188-191B.
94. Ibid., no. 52, fs. 39-41.
95. Manuela Carneiro da Cunha, Negros, estrangeiros, pp. 152-204.

96. Verger, Trade Relations, p. 544.
97. Manuela Carneiro da Cunha, Negros, estrangeiros, p. 214.
98. APB, Uvros de notas and maô 2880.
99. Manuela Carneiro da Cunha, Negros, estrangeiros, pp. 210-216.
100. Lorenzo D. Turner, “Some Contacts,” which was based on his interviews con
ducted in Salvador, 1941-1942.
101. J. Michael Turner, “Les Brésiliens,” p. 83.
6. T h e C reation o f Disparate Identity, 1808-1851

1. Mintz, “Tasting Food, Tasting Freedom," p. 257. 1 concur with Mintz: "The
vast differences between the enslaved and those who were bom into slavery
are too often ignored" See ibid., p. 270, note 1.
2. APB, Quadro dos nascimentos da freguesia da Penha (1844), maô 1549.
3. ACMS, Livros de bat izados, freguesia de Santo António, from which 1 con
sulted data for two out o f every ten years (1809-1810,1828-1829,1838-1839,
1848-1849,1858-1859, and 1868-1869).
4. AMCS, Livros de escrituras de compra e venda de escravos, freguesia da Sé.
All four cases o f father-child transaction took place in the 1860s.
5. Ibid., no. 82.6, December 7,1867.
6. A P B , L iv ro s d e n otas and m aô 2880.
7. APB, Livro de notas, no. 116, January 25,1811.

8. Ibid., no. 272, Nov. 17,1841; and ibid., no. 275, October 30,1841.
10. APB, Livro de registro de testamentos, no. 38, ft. 177B-179B (September 21,
1848). See also ibid., no. 35, fs. 16-18 (May 18,1850).
12. APB, Livro de registro de testamentos, no. 28, fs. 98B-101B (August 20,1827).
See also ibid., no. 30, fs. 142B-145B (November 1,1841); ibid., no. 4 4 * ft* 135-137
(November 10,1863); and ibid., no. 58, ft. 65B-68 (April 18,1882).
13. Ibid., no. 51, fs. 185B-189B.
14. Lewin, “Natural and Spurious Children," p. 389.
15. APB, Livro de registro de testamentos, no. 52, fs. 39-41.
16. Ibid., no. 41, fs. 64-65B (November 26,1861). For grandfathers, see also ibid.,
no. 30, fs. 63 (October 19,1840); and ibid., no. 41, ft. 64-65B (November 26,
1861).
17. Ibid., no. 5 ,206B-210. On grandmother figures, see also ibid., no. 5, ft. 25B-
28B (March 17,1815); ibid., no. 22, ft. 4B-8 (July 29,1832); ibid., no. 22, fs. 52-
no Notes to pages 115-127

55B (October 9,1832); ibid., no. 31, fs. 102B-105 (February 27,1839); ibid.,
no. 34, fs. 8-11B (January 7,1837); and ibid., 110. 51, fs. 101-104 (July 27,1867).
ACMS, Livros de batizados, freguesia de Santo Antônio. After mid-century,
a higher percentage o f Brazilian-born slaves were baptized with both god
fathers and godmothers; women seem to have begun to take a more active
role in fictive kinship.
ACMS, Livros de batizados, freguesia de Santo Antônio, February 2,1810;
May 14,1810; October 7,1810; December 13,1813.
APB, Livros de notas and maço 2880. Grandmothers: Anacleta, crioula, daugh
ter o f Thomara (already dead), crioula slave o f the same owner, with iso$ooo
réis from her grandmother, Joaquina, Nagô; and Thedolina, 11-year-old parda
with money from her grandmother, Mana, o f African birth. APB, Livro de
notas, no. 297, July 1,1851; ibid., no. 416, March 27,1872. Godparents: only one
o f the godparents was identified as African-born; José de Sâo Joào, Nagô, paid
the price o f Antônia, 17-year-old crioula (900(000 réis). See APB, maço 2880,
February 6,1879- In some cases, relations o f purchasers o f freedom to the
slaves were not mentioned in the manumission letters. For instance, two
Brazilian-born sisters, Euporoma (6 years old) and Justina (9 years old),
daughters o f their owner’s former slave, Verdina, a crioula, were liberated
with 1:200$000 réis donated by Manuel Joaquim Gomes Vilças, because he
“appreciated the services” not only o f their mother but also o f their grand
mother. In 1880, Guirina, a 23-year-old crioula, received her letter o f liberty
on payment o f 8oo$ooo réis: 40o$ooo réis from herself and another 400$000
réis from Margarida dos Sassos, who had rented out Guirina for fieldwork. In
the case o f Luis, crioulo, George Harris Duder paid 950(000 réis for his free
dom, but on condition that Luis had to serve him for seven years. See APB,
Livro de notas, no. 416, April 17,1872; APB, maço 2880, February 16,1880;
ibid., November 13,1879. On the purchasers o f freedom for Brazilian-born
child slaves in Rio de Janeiro, see Karasch, Slave Life, pp. 347-350.
APB, Livros de notas and maço 2880.
Nishida, “ Manumission and Ethnicity,” p. 374. In Bahia for the earlier period
1684-1745, Brazilian-born ex-slaves made up 69 percent o f the total freed,
while African-born ex-slaves were only 31 percent. See Schwartz, “Manumis
sion,” p. 612. On the advantage o f Brazilian-born slaves in manumission in
other parts o f Brazil, see Karasch, Slave Life, p. 352; Kieman, “Manumission
o f Slaves,” p. 92; Higgins, “ Slave Society," p. 213; and Eisenberg, “ Ficando
livre,” pp. 189-191.
APB, Livros de notas and maço 2880. Concerning the advantage of Brazilian-
born slaves in manumission, see Schwartz, “Manumission,” p. 612; Karasch,
Slave Lifet p. 352; Kieman, “Manumission o f Slaves,” p. 92; Higgins, “Slave
Society,” p. 213; and Eisenberg, “ Ficando livre,” pp. 189-191.
Schwartz, “ Manumission,” p. 619.
APB, Livros de notas and maço 2880.

26. In this study, I classify those o f more than 15 years o f age as falling into the
c a te g o r y o f “ a d u lt ” b ased o n th e a g e fo r w o r k in g a n d m ilita r y se rv ic e fo r
males and reproduction and domestic responsibility for females. I regard
as “children” those who are identified by diminutives such as crioulinhol
crioulinhoy pardinhoipcrdinha, and mulatinhofmulatinha, but whose ages
are not specified.
27. APB, Livros de notas and maço 2880. Brazilian-born child slaves rarely
purchased their freedom. Among those o f Brazilian birth who purchased
freedom, only 8.1 percent o f males and 7 percent o f females were children.
See APB, Livros de notas and maço 2880.
28. AMCS, Livros de escritura de compra e venda de escravos, freguesia da Sé.
On the advantage o f children, see Schwartz, “ Manumission,” pp. 615-616;
Kiernan, “Manumission o f Slaves,” p. 102; Eisenberg, “ Ficando livre,” p. 192;
Bowser, “ Free Persons,” pp. 350-351.
29. APB, Livros de notas and maço 2880. Eleven boys and 16 girls belong to this
category for 1808-1884.
30. Manoel, crioub, son o f the female slave. See APB, Livro de notas, no. 164, Janu
ary 25,1811.
31. On high child slave mortality in Brazil, see, for instance, Merrick and Gra
ham, Population and Economic Development, p. 61; Conrad, World of Sorrow,
pp. 14-15.
32. APB, Livros de notas and maço 2880. 1 regard all o f those who were classified
as pardos, cabras, mulatos, and mestiços together as “mulattoes” in contrast to
crioulos.
33. AMCS, Livros de escritura de compra e venda de escravos, freguesia da Sé.
34. APB, Inventârios da capital.
35. APB, Livros de notas and maço 2880. Concerning the advantage o f mulatto
slaves in manumission, see Schwartz, “Manumission,” p. 618; Kiernan, “Manu
mission o f Slaves,” p. 92; Higgins, “Slave Society,” p. 215; Eisenberg, “ Ficando
livre,” p. 187; and Johnson, "Manumission in Colonial Buenos Aires,” pp. 264-
265.
36. See, for instance, ACMS, Livros de casamentos, freguesia da Penha 1808-1857;
and freguesia da Conceiçâo da Praia (1843-1888). The marriage records in
early nineteenth-century Rio de Janeiro show the same tendency: Marriages
that trussed color and civil status barriers were uncommon. See Karasch,
Slave Life, p. 291.
37. In the first national census o f 1872,30 percent o f the white population in Bra
zil was registered as married, whereas only 8 percent o f the slave population
was married. The percentage o f the married among the legally free (freed and
free-born) population of color was significantly higher than among slaves,
whereas within the broad category o f free people o f color, free “mulattoes”
enjoyed a slight advantage over free “blacks." See Merrick and Graham, Popu
lation and Economic Development, pp. 58-60.

38. Dain Borges discusses Bahian endogamy among elite families for the period
1600*1870, as a means o f not dividing inheritance and abo as a patriar
chal strategy to preserve family continuity. See Borges, Family in Bahia,
pp. 240-241.
39. APB, maô 5685, Registro dos fogos. . . Sáo Pedro (1835).
40. APB, Livro de registro de testamentos, no. 28, pp. 148-150B.
41. Verena Martinez-Alier, Marriage, Class and Color in Nineteenth-Century
Cuba: A Study o f Racial Attitudes and Sexual Values in a Slave Society (Lon
don and New York: Cambridge University Press, 1974), chapter 6.
42. In the colonial Mexican city o f Antequerra (Oaxaca), which was a slave-
owning society but never a slave society, such group boundaries were situa
tional and flexible, and individual cross-group mobility always existed. Chil
dren were classified with regard to “social race” at baptism, and frequently
separate registers were kept for Spaniards, Indians, and castas (a generic term
for all mixed bloods). Yet, the same individuals were categorized or identified
themselves in different ways in different situations. Marriage was often con
tracted across socioracial lines. See John K. Chance and William B. Taylor,
"Estate and Class in a Colonial City: Oaxaca in 1792,” Comparative Studies in
Society and History 19, no. 4 (1977): 454-487.
43. Martinez-Alier, Marriage, Class and Colour, chapter 7. On marriage and
family honor in colonial Mexico, see Patricia Seed, To Love, Honor, and Obey
in Colonial Mexico: Conflicts and Marriage Choice, ¡574-1821 (Stanford, Calif.:
Stanford University Press, 1988); Ann Twinan, Public Lives, Private Secrets:
Gender, Honor, Sexuality, and Illegitimacy in Colonial Spanish America (Stan
ford, Calif.: Stanford University Press, 1999).
44. Verger, Trade Relations, p. 465; and idem, Noticias da Bahia, p. 65. On the
establishment o f the mulatto sodality o f Our Lord o f Good Jesus o f the Cross
(irmandade de Nosso Senhor Bom Jesus da Cruz) in 1751 and its oral tradi
tion, see Ignácio Accioli de Cerqueira e Silva, Memórias históricas e políticas
da Bahia, vol. 5 (Salvador: Imprensa Oficial do Estado, 1931), p. 241.
45. The Tailors’ Revolt has been identified by some historians as the first slave
uprising in Brazil, but does not seem to have represented the interests o f the
general enslaved population. Studies on the Tailors* Revolt include: Afonso
Ruy, A primeira revoluto social brasileira (1798), 2nd ed. (Salvador: Tipo
gráfico Beneditina Ltda, 1951); Kátia M. de Qucirós Mattoso, Presenta francesa
no movimento democrático bahiano de ¡798 (Salvador: Editora Itapui, 1969);
Donald Ramos, “Social Revolution Frustrated: The Conspiracy of the Tailors
in Bahia, 1798,” Luso-Brazilian Review 13, no. 1 (1976): 74-90; Schwartz, Sugar
Plantations, pp. 476-478; Judith Lee Allen, “Tailors, Soldiers, and Slaves: The
Social Anatomy o f a Conspiracy” (M.A. thesis, University o f Wisconsin-
Madison, 1987).
46. Allen, “Tailors, Soldiers, and Slaves,” p. 8. Allen did not distinguish between
free-born and ex-slaves among the free pardos.
Notes to pages ¡30-132 213

47- Ramos» “Social Revolution Frustrated,” p. 78.
48. Anne Plrotin-Dumon, in her study o f the Lesser Antilles for the years 1789-
»794, successfully discusses the emergence o f politics among the free and
enslaved male population o f African descent. In revolutionary Guadeloupe,
different groups o f people adopted the key concepts, such as “liberty” and
“freedom,” but interpreted and used them for different purposes. See Anne
P£rotin-Dumon, “The Emergence o f Politics among Free-Cdoreds and Slaves
in Revolutionary Guadeloupe," Journal o f Caribbean History 25, nos. 1-2
(1991): 100-135.
49. Schwartz points out that the Tailors' Revolt seems to fit into the model that
historians Michael Craton and Eugene D. Genovese respectively draw on to
document an important change in the type or nature o f slave rebellions dur
ing the Age o f Revolution. Craton maintains that in the British Caribbean,
maroon-style rebellions led by “unassimilated” African-born slaves in the
earlier period were replaced by movements led by elite creole slaves. Genovese,
in his comparative study o f slave rebellions in the New World, maintains: “Un
til the Age o f Revolution the slaves did not challenge the world capitalist sys
tem within which slavery itself embedded. Rather, they sought escape and
autonomy— a local, precapitalist social restoration. When they did become
revolutionary and raise the banner o f abolition, they did so within the con
text o f the bourgeois-democratic revolutionary wave.” See Schwartz, Sugar
Plantations, pp. 472-473; Michael Craton, Testing the Chains: Resistance to
Slavery in the British West Indies (Ithaca, N.Y.: Cornell University Press, 1982);
Genovese, From Rebellion to Revolution, p. xxi. Most recently, Michael Mullin
attempts to establish a typology to explain changes in major slave rebellions
in Anglo America (namely the American South and the West Indies) between
1736 and 1831 by setting up three historical stages o f “acculturation** for the
slave population. These are: “the unseasoned** (from the 1730s to the 1760s);
“plantation slaves” (from the late 1760s to the early 1800s); and “the assimi
lated” (from the late 1760s to the second quarter o f the nineteenth century).
See Mullin, Africa in America, especially pp. 268-273. On American slavery in
the revolutionary era (1770-1823), see David Brion Davis, The Problem o f Slav
ery in the Age o f Revolution (Ithaca, N.Y.: Cornell University Press, 1975). On
slave resistance in the American South during the same period, see Sytvia R.
Frey, Water from the Rock: Black Resistance in a Revolutionary Age (Princeton,
N.J.: Princeton University Press, 1991).
50. Schwartz discusses the distinction between African-born and Brazilian-born,
as well as between crioulos and mulattoes, resulting in different political ac
tions. See Schwartz, Sugar Plantations, p. 47 3 -
51. Michael C. McBeth, “The Brazilian Recruit during the First Empire: Slave or
Soldier?” in Dauril Alden and Warren Dean, eds., Essays Concerning the Socio
economic History o f Brazil and Portuguese India (Gainesville: University Press
o f Florida, 1977), p. 71.
52. Joan E. Menzar, “The Ranks o f the Poor: Military Service and Social Differen-
214 Notes to page 133

tiation in Northeast Brazil, 1830-1875” H A H R 72, no. 3 (1992): 335-351. Men-
zar argues that the Quefora Quilos revolt, which took place in northeast Brazil
in November and December o f 1874, was a manifestation o f the desire o f the
poor “honorable” working people to maintain their status separate from
slaves and "undesirables” See ibid., pp. 347-351.
53. F. W. O. Morton, “The Military in Bahia, 1800-1821,” Journal o f Latin American
Studies 7, no. 2 (1975): 257-258. A royal decree o f 1822 stipulated that all Brazil
ian single males between the ages o f 18 and 35 were subject to service in the
army, whereas the so-called nonrecruitment law exempted the following cate
gories: married men and only sons; overseers and administrators of cattle
ranches, plantations, and brick factories; seamen; merchants; students; and
cattle herders, bricklayers, carpenters, and fishermen as long as they “actually
exercised their craft and were well behaved ” When the national guard was
created on August 14,1831, all Brazilian men between 18 and 60, but only from
families whose annual incomes qualified them to vote, became subject to na
tional guard service. This financial qualification distinguished the national
guard from the army. See Menzar, “ Ranks o f the Poor,” pp. 337-340.
54. Russell-Wood, Black Man in Slavery and Freedom, p. 86.
55. Maria Graham, Journal o f a Voyage to Brazil, p. 141. Maria Graham praised
the black regiment in Salvador as “unquestionably the best trained, and most
serviceable, as a light infantry” See ibid., p. 141.
56. APB, Ordens régias 54 (January 30,1754); Arquivo Histórico Ultramarino (Lis
bon), Bahia, papéis avulsos, caixa 66 (December 8,1756) and maço 18 (May 11,
1756), cited by Stuart B. Schwartz, “The Formation of a Colonial Identity in
Brazil,” in Nicolas Canny and Anthony Pagden, eds., Colonial Identity in the
Atlantic World, 1500-1800 (Princeton, N.J.: Princeton University Press, 1987),
p. 48.
57. Kátia M. de Queirós Mattoso, Bahia, siculo XIX: urna provincia no impèrio
(Rio de Janeiro: Editora Nova Fronteira, 1992), p. 404; Julita Scarano, “Black
Brotherhoods: Integration or Contradiction?” Luso-Brazilian Review 16, no. 1
(1979): 11; and Maria Inés Còrtes da Oliveria, O liberto, pp. 84-85.
58. Mattoso, Bahia, século XIX, pp. 401,400.
59. Oliveira, O liberto, p. 84.
60. APB, Livros de registro de testamentos, nos. 35-61. This confirms the findings
o f Kátia Mattoso in her study o f 200 wills o f ex-slaves in Salvador. Accord
ing to her analysis, for the years 1790-1826 those who did not belong to any
sodality numbered only 14.9 percent o f men and 17 percent o f women. These
relatively low percentages escalated to 98.5 percent for men and 87 percent for
women for the period 1863-1890. See Mattoso, Testamentos de escravos libertos,
p. 23 (Table 2). Mattoso consulted 100 wills o f ex-slaves for each period: 53
men and 47 women for 1790-1826 and 31 men and 69 women for 1863-1890.
Oliveira’s study o f 482 wills of ex-slaves for 1790-1890 strongly supports this
point and furthermore indicates a gradual process o f change in ex-slaves’ atti-

tudes toward lay sodalities. The percentages o f those who did not mention
membership in any sodality are 21.6 percent (men) and 18.5 percent (women)
for 1790-1830; 50.0 percent (men) and 42.4 percent (women) for 1830-1850;
and 96.1 percent (men) and 84.2 percent (women) for 1850-1890. See Oliveira,
Liberto, p. 84.
62. AINS, Livro de termo e resoluâo, f. 104. These data do not have any break
down by legal distinction, birthplace, or color. Among the ex-slaves who regis
tered their wills, there were 13 men and 20 women who held memberships in
this specific black sodality for 1808-1849, but for 1850-1888 only one was a
member o f this sodality. See APB, Livros de registro de testamentos.
64. C. R. Boxer, Women in Iberian Expansion Overseas 1415-1815Some Facts, Fan
cies and Personalities (New York; Oxford University Press, 1975), chapter 6, re
lates the popularity o f the cult o f the Virgin Mary to the notion o f the basic
inferiority o f women in colonial Iberian Society. William B. Taylor discusses
the historical process in which the Virgin o f Guadalupe, a representation o f
the Virgin Mary, became the symbol o f future Mexico before independence.
See William B. Taylor, “The Virgin o f Guadalupe in New Spain: An Inquiry
into the Social History o f Marian Devotion,” American Ethnologist 14, no. 1
(1987): 9-33. See also William B. Taylor, Magistrates o f the Sacred: Priests
and Parishioners in Eighteenth-Century Mexico (Stanford, Calif.: Stanford
University Press, 1996), chapter 11. On the indigenous population's response
to Christianity in the case o f late colonial Mexico, see Stephanie Wood,
“Adopted Saints: Christian Images in Nahua Testaments o f Late Colonial
Toluca,” Americas 47, no. 3 (1991): 259-294; and James Lockhart, The Nahuas
after the Conquest: A Social and Cultural History o f the Indians o f Central
Mexico, Sixteenth through Eighteenth Centuries (Stanford: Stanford Univer
sity Press, 1992), chapter 6. 1 am grateful to Professor William B. Taylor for
allowing me to read two chapters o f his book before publication. On the cult
and veneration o f the Virgin Mary in general, see Hilda Graef, Mary: A His
tory o f Doctrine and Devotion (Westminster: Christian Classics and London:
Sheed & Ward, 1985 [1963 (Part 1) and 1965 (Part 2)]); and Marina Warner,
Alone o f All Her Sex: The Myth and the Cult o f the Virgin Mary (New York:
Alfred A. Knopf, 1976).
65. Borges, Family in Bahia, pp. 58 and 63.
66. Wetherell, Brazil: Stray Notes from Bahia, p. 24.
67. ACMS, Livros de batizados, freguesia de Santo Antdnio. In Brazil, church
practice was codified in the synod in 1707 and published in the Constituifoes
primeiras do Arcebispado na Bahia in 1720. According to this code, baptism
was to be given to newborn infants by a parish priest within eight days of
birth, and each child was to have only one godmother (over the age o f 12) and
one godfather (over the age o f 14). The baptism o f newly arrived and unaccul-
turated slaves called for special religious instruction to ensure that they under-

stood their obligations as members o f the church. See Schwartz, Slaves, Peas
ants* and Rebels, p. 139.
68. The staff o f the Arquivo da Curia Metropolitana de Sâo Salvador da Bahia
(ACMS), personal communication.
69. From the twelfth century on» increased importance was given to the Virgin
Mary in accordance with the ongoing feminization process o f religious lan
guage. Women continued to be perceived as physiologically and spiritually
weaker, defective in body and moral fortitude, but as being equally worthy
o f salvation. See Caroline Walker Bynum, Jesus as Mother: Studies in the Spiri
tuality o f the High Middle Ages (Berkeley, Los Angeles, and London: University
o f California Press, 1982), p. 135.
70. Taylor, Magistrates o f the Sacred, chapter 11 and personal communication.
71. Taylor, Magistrates o f the Sacred, chapter 11.
72. Schwartz, Slaves* Rebels* and Peasants* p. 139.
73. Ibid , pp. 139 and 150. The practice o f baptizing children with the Virgin
Mary as godmother took place in other parts o f Brazil during the nineteenth
century. See Gudeman and Schwartz, “Cleaning Original Sin,” p. 52; Renata
Pinto Venâncio, “Nos limites da sagrada familia: illegitimidade e casamento
no Brasil colonial,” in Ronaldo Vanifes, éd., Histôria e sexualidade no Brasil
(Rio de Janeiro: Graal, 1986), pp. 113-121; Elizabeth Anne Kuznesof, “Sexual
Politics, Race and Bastard-Bearing in Nineteenth-Century Brazil: A Question
o f Culture or Power,” Journal o f Family History 16, no. 3 (1991): 256; Schwartz,
Slaves* Peasants* and Rebels, p. 150; Robert M. Levine, “ 4Mud-Hut Jerusalem*:
Canudos Revised,” HAHR 68, no. 3 (1988): 525-576. According to Marcia Elisa
de Campos Graf this custom was common in nineteenth-century southern
Brazil. Maria Elisa de Campos Graf, personal communication.
74. The historian Suzanne Desan discusses the re-emergence o f lay Christianity
in the case o f revolutionary France. See Suzanne Desan, Reclaiming the Sa
cred: Lay Religion and Popular Politics in Revolutionary France (Ithaca, N.Y.:
Cornell University Press, 1990).
75. Verger, Trade Relations* p. 465.
76. Arquivo da Sociedade Protetora dos Desvalidos, Salvador, Brazil (hereafter
ASPD), Livro de termos do ano de 1832; Verger, Trade Relations* pp. 457 - 4 5 9 ;
and Julio Santana Braga, Sociedade protetora dos desvalidos: uma irmandade de
cdr (Salvador: Ianamâ, 1987). Not only did both Verger and Braga misidentify
Manoel Victor Serra and the founding member o f the sodality as free (or
freed) Africans, but they misinterpreted the primary purpose o f the Protec
tive Society as an emancipation pool (junta de alforria). Their argument was
grounded on the presence o f the safe box, which still exists in the Society, but
sodalities often had safe boxes, such as the case o f the black sodality o f Our
Lady o f the Rosary in the Pelourinho. For example, see chapter 18 o f its stat
utes (1820): AINS, Compromisso da Irmandade da Nossa Senhora do homens
prétos no ano de 1820.

77- ASPD, Livro de termos e acordâos, f. 7B (August 4,1833).
78. Compromisso da lrmandade do Glorioso Säo Francisco de Paula Filial à
Matriz de Nossa Senhora do Pilar» de crioulos livres nascidos no Imperio do
Brazil, ano de 1844, Arquivo da Venerével Ordern Terceira de Nossa Senhora
do Monte do Carmo (Salvador, Brazil). I am grateful to Professor A. J, R.
Russell-Wood for the copy o f this compromisso.
79. The popularity o f the cult o f the Virgin o f Guadalupe, patron saint o f
Mexico, especially among the indigenous population, is attributable to the
alleged miraculous appearance o f the Virgin in December 1531. Her church
was subsequently built at a historically special place, which had been sacred
to the Aztec maize goddess. See George M. Foster, Culture and Conquest:
America’s Spanish Heritage (New York: Wenner Gren Foundation for Anthro
pological Research, 1960), p. 207.
80. Russell-Wood, “ Examination," pp. 247-248; idem, "Black and Mulatto Broth
erhoods” pp. 577-578.
81. Compromisso da lrmandade do Glorioso Säo Francisco de Paula (1844), chap
ters 21-32, Arquivo da Venerável Ordern Terceira de Nossa Senhora do Monte
de Carmo; and Russell-Wood, Black Mart in Slavery and Freedom, p. 152.
82. In the remaining census records o f 1855, among the free population, 52.2 per
cent o f the couples who shared the same households were not married. See
Mattoso, Familia e sociedade, p. 82.
7. T h e Labyrinth o f Identity, 1851-1888

1. Brasil, Recenseamento da populaçâo, pp. 508 and $10.
2. AMCS, Livros de escrituras da compra e venda de escravos, freguesia da Sé,
nos. 82.1-82.20; APB, Seçâo Judiciária, Inventários da capital (1808-1888).
3. Brasil, Recenseamento da populaçâo, pp. 308,310.
4. Conrad, Destruction o f Brazilian Slavery, chapters 9-12.
5. Emilia Viotti da Costa, The Brazilian Empire: Myths and Histories (Chicago:
Dorsey Press, 1988), p. 147.
7. APB, Seçâo Judiciária, Autos Crimes, 3531.8. O n Bahian abolitionism and abo
litionist societies, see Pierson, Negroes in Brazil, pp. 54-59. On abolitionism in
Brazil, see Conrad, Destruction o f Brazilian Slavery, pp. 121-277; Robert Brent
Toplin, The Abolition of Slavery in Brazil (New York: Atheneum, 1972).
8. Compra e venda de escravo para a Guerra do Paraguay 1867, APB, Seçâo
Histórica, maço 2970; Trosko, “ Liberto in Bahia," p. 30 and footnote 48.
9. Conrad, Destruction o f Brazilian Slavery, pp. 90-117; Toplin, Abolition of
Slavery, pp. 20-21.
10. APB, Livro de notas, no. 411, August 8,1871; ibid., no. 411, August 12,1871; ibid.,
no. 416, August 2i, 1817.

11. APB, Livro de notas, no. 711, March 20,1883.
12. APB, Sc<ô Judiciiria, Autos Crimes, no. 2360.5.
13. Chalhoub, “Slaves, Freedmen and the Politics o f Freedom,” p. 78. See also
idem, Visdes da liberdade, chapter 2.
14. Sandra Lauderdale Graham, ‘‘Slavery’s Impasse: Slave Prostitutes, Small-Time
Mistresses, and the Brazilian Law o f 1871,” Comparative Studies in History
and Society 33, no. 4 (1991): 669-694. On slave prostitution in nineteenth-
century Rio de Janeiro, see, for instance, Luiz Carlos Soares, Rameiras, ilhoas,
poiacas. . . . A prostitui&o no Rio de Janeiro do stculo X IX (S3 o Paulo: Editora
Atica, 1992), chapter 5.
15. Lauderdale Graham, "Slavery’s Impasse,*1 p. 676.
16. Brasil, Reccnscamcnto da populaf&o, pp. 510, 512.
17. Merrick and Graham, Population and Economic Development, p. 59.
18. Slenes, “ Demography and Economics," p. 420, Table 9.2.
19. Merrick and Graham, Population and Economic Development, p. 59, particu
larly Table IV-3 and note 5* Robert Conrad cites the same data as proof that
ex-slaves rushed to register their marital status once they obtained their free
dom, but does not pay any attention to the problem in the Brazilian census
data. See Conrad, World o f Sorrow; p. 13.
20. Brasil, Recenseamento da popula&o, pp. 508,510.
21. The percentage o f self-identified mulattoes continued to grow in post-
emancipation Salvador: from 35.1% (1890) to 38.4% (1944), while that o f the
prtto population did not change (26.3% for both 1890 and 1940). See Kim D.
Butler, Freedoms Given, Freedoms Won: Afro-Brazilians in Post-Abclition SAo
Paulo and Salvador (New Brunswick, N.J.: Rutgers University Press, 1998),
p. 134, Table 7. Needless to say, these census data do not suggest a scale o f fur
ther racial mixture but demonstrate Afro-Brazilians* continuing denial of
blackness and eagerness to pass as mulattoes.
22. Dain Borges states that the usage o f crioulo became impolite in Bahia after
the turn o f the twentieth century. See Borges, Family in Bahia, p. 292.
23. In present-day Salvador, the term prito, which is regarded as extremely pejora
tive, is rarely used to refer to a dark-skinned person, unless it is intended to
offend the person.
24. APB, Autos Crimes, 2594.2. Conducting his research in the late 1930s, the U.S.
sociologist Donald Pierson stated that the term negro was “seldom heard in
Bahia,” since it was regarded as “ a palavra pesada, a harsh, even offensive
term ” See Pierson, Negroes in Brazil, p. 138. Pierson recorded terms and popu
lar expressions for various phenotypes among the mulatto population. Ibid.,
PP* 135- 138.
25. Manoel Francisco dos Santos, Discurso Proferido na Sociedade Protetora dos
Desvalidos (1876), APB, maô 5306.

26. 1 was very fortunate to learn interesting oral history in person from longtime
members o f the Society for the Protection o f the Needy on many occasions,
although I had not planned to conduct interviews. Our conversations usually
took place in the main meeting room o f the Society for the Protection o f the
Needy in the afternoons over cool maté tea with much sugar, which I was
generously offered whenever I visited there for consultation o f their historical
documents. Those old members, all Afro-Brazilian men, gathered around one
by one and started to speak to me about their histories, which must have been
constructed and reconstructed since the days o f slavery, together with their
activities as members of this association and their individual experiences as
“ black” Bahians. My gratitude goes to Sr. Aloisío C o n ce d o Rocha and his
fellow members o f the Society for the Protection o f the Needy as well as his
cousin Padre Rocha. Sr. Rocha, who is a devoted Catholic, has been also ac
tively engaged in candomblé throughout his life, although he does not hold
the title o f pai de santo. Every one o f those members o f the Society for the
Protection o f the Needy conveyed to me win an individual way the collective
experience o f a conquered people” and revealed uhis experiences as they are
embodied in, and embody, the history o f his society,” as Mintz describes his
Puerto Rican informant named Tazo in his monograph entitled Worker in
the Cane. See Sidney W. Mintz, “The Sensation o f Moving, while Standing
S till” American Ethnologist 16, no. 4 (1989): 792. Also, as Mintz perceives his
relationship with Tazo (ibid., pp. 794-795), I feel the people o f the Society for
the Protection o f the Needy had "chosen" me; 1 happened to “come along”
when those individuals were ready to tell their stories. See Sidney W. Mintz,
Worker in the Cane: A Puerto Rican Life History (New York: W. W. Norton,
1974 [i960]).
27. Estatutos da Sociedade Protetora dos Desvalidos (1873), titulo 1 and art. 9, pub
lished by Braga, in Sociedade protetora, pp. 79-89. 1 was not able to consult the
original copy o f the Estatutos (1873), since Sr. Rocha, who is in charge o f their
Archives, was unable to locate it during my research period.
29. ASPD, Livro de matrículas (1870-1950); Documentos dos anos (1870-1874).
30. Estatutos da Sociedade Protetora dos Desvalidos (1873), art. 11.
31. The first female member o f the Society for the Protection of the Needy, who
was an elderly lady but in excellent health in June 1989, came to the headquar
ters relatively often, but just briefly each time, during my research, but she
never joined the table where the male members and I gathered for talks.
32. Anani Dzidzienyo, “Africa-Brazil: Ex Africa Semper Aliquid Novi,” in Larry
Crook and Randall Johnson, eds., Black Brazil: Culture, Identity, and Social
Mobilization (Los Angeles: UCLA Latin American Studies Center, University
o f California, 1999) and personal communication.
33. Estatutos da Soaedade Protetora dos Desvalidos (Salvador: Sociedade Protetora
dos Desvalidos, 1974).

34- The term mestiço was employed in the Brazilian national censuses of 1890 but
was again replaced by pardo in 1940. See Butler, Freedoms Given, Freedoms
Won, p. 134. Today mestiço is not used commonly.
35. Daphne Patai, Brazilian Women Speak: Contemporary Life Stories (New Bruns
wick, N.J.: Rutgers University Press, 1988), pp. 12-15. Patai capitalizes the
terms as White and Black, while I put them in quotation marks.
36. Patai, Brazilian Women Speak»p. 11. For an excellent discussion o f color and
gender identity among Afro-Brazilian women, see John Burdick, Blessed
Anastâcia: Women, Race, and Popular Christianity in Brazil (New York: Rout-
ledge, 1998), chapter 1.
37. Carl N. Degler, Neither Black nor White: Slavery and Race Relations in Brazil
and the United States (Madison: University o f Wisconsin Press, 1986 [1971]),
chapter V.
38. Virginia R. Dominguez, White by Definition: Social Classification in Creole
Louisiana (New Brunswick, N.J.: Rutgers University Press, 1986).
39. The Brazilian Institute o f Geography and Statistics (IBGE) decided in 1980 to
begin analyzing and publishing racial data in dichotomous form by dividing
the population into whites (brancos) and blacks (prètos). According to George
Reid Andrews, in Sâo Paulo, Rio de Janeiro, and the states o f southern Brazil,
such a North American dichotomous vision o f race hierarchy as blacks versus
whites “corresponds not just to ‘objective* statistical indicators but to subjec
tive Brazilian perceptions o f race as well ” Andrews, Blacks and Whites in
Sâo Paulo, pp. 250-251. Since the turn o f the 1970s, many demographers, both
Brazilians and non-Brazilians, have discussed "racial inequalities” in Brazil
based on quantitative analysis o f census data. The Brazilian sociologist Nelson
do Valle Silva, in particular, maintains that the complex web o f color termi
nology can be reduced into three categories: blacks, pardos, and whites, and
concludes that the race-related differences between blacks and pardos are far
less significant than the difference between “whites” and “non-whites” See
Nelson do Valle Silva, “Updating the Cost o f Not Being White in Brazil,** in
Pierre-Michel Fontaine, ed., Race, Class, and Power in Brazil (Los Angeles:
Center for Afro-American Studies, University o f California, Los Angeles,
1980). See also Carlos Hasenbalg, Nelson do Valle Silva, and Mârcia Lima,
Cor e estratifiçào social (Rio de Janeiro: Contra Capa Libraria, 1999).
40. David L. Marcus and Lennox Samuels, “Melting Pot Coming to a Boil,” Dal
las Morning News, January 16,1994, p. 31A. My italics. I am grateful to Profes
sor William B. Taylor for a copy o f the article. This newspaper article does
not specify the original Portuguese word that has been translated as “black.”
41. Recent studies on blackness and racial identity in Latin America include
Guss, “Selling o f San Juan”; Winthrop R. Wright, Café con leche: Race, Class,
and National Image in Venezuela (Austin: University o f Texas Press, 1990);
Peter Wade, Blackness and Race Mixture: The Dynamics o f Racial Identity
in Colombia (Baltimore: Johns Hopkins University Press, 1993); idem, “The

Cultural Politics o f Blackness in Colom bia" American Ethnologist 22, no. 2
(1995 ): 341 - 357 ; Aline Helg, Our Rightful Share: The Afro-Cuban Struggle for
Equality, 1886-1912 (Chapel Hill: University o f North Carolina Press, 1995);
and Alejandro de la Fuente, A Nation for A lt Race, Inequality, and Politics in
Twentieth-Century Cuba (Chapel Hill: University o f North Carolina Press,
2001).
42. Guss, “Selling o f San Juan,n p. 466.
43. For recent studies on the movimento negro, see Andrews, Blacks and Whites in
Sáo Paulo, pp. 146-207; Michael James Mitchell, “ Racial Consciousness and
the Political Attitudes and Behavior o f Blacks in Sáo Paulo" (Ph.D. diss., Indi
ana University, 1977); Michael George Hanchard, Orpheus and Power: The
Movimento Negro o f Rio de Janeiro and Sáo Paulo, Brazil, 1945-1988 (Princeton,
N.J.: Princeton University Press, 1994); and Butler, Freedoms Given, Freedoms
Won. See also Larry Crook and Randall Johnson, eds., Black Brazil: Culture,
Identity, and Social Mobilization (Berkeley and Los Angeles: University o f Cali
fornia Press, 1999); Michael Hanchard, ed., Racial Politics in Contemporary
Brazil (Durham, N.C.: Duke University Press, 1999); Antonio Sérgio Alfredo
Guimaráes and Lynn Huntley, eds., Tirando a máscara: ensaios sobre o racismo
no Brasil (Sáo Paulo: Paz e Tera, 2000).
44- Melissa Nobles maintains: “The umbrella term movimento negro subsumes a
variety o f activities and organizations that are held together only by the term
negro, which glosses over the regional, class, political, and ideological divisions
between and among them. More important, the use o f ‘black movement’ con
ceals the absence o f a broadly shared understanding o f ‘black’ and to whom
‘black* should be applied.” See Melissa Nobles, Shades o f Citizenship: Race and
Censuses in Modem Politics (Stanford, Calif.: Stanford University Press, 2000),
p. 146. See also Rebecca Reichmann, “ Introduction," in Rebecca Reichmann,
ed., Race in Contemporary Brazil From Indifference to Inequality (University
Park: Pennsylvania State University Press, 1999), pp. 14-20; Abdias do Nas-
cimento, Brazil Mixture or Massacret Essays in the Genocide o f a Black People,
trans. Elisa Larkin Nascimento, 2nd ed. (Dover, Mass.: Majority Press, 1989);
Abdias do Nascimento and Elisa Larkin Nascimento, Africans in Brazil A Pan-
African Perspective (Trenton, N.J.: Africa World Press, 1992); idem, “ Reflexócs
sobre o movimento negro no Brasil, 1938-1997,** in Guimaráes and Huntley,
eds., Tirando a máscara.
45. Reichmann, “Introduction," p. 11. Reichmann states: “ In accord with the
official version, the negro acknowledges and welcomes African identity,
embracing Africa’s cultural contributions to Brazilian life. In a departure
from the received ideology, however, the negro’s consciousness is heightened
about the significance o f ethnicity (rather than color) and, regardless o f his
or her own statues, identifies with the subjugation o f black people in Brazil."
See ibid, p. 11.
46. Ibid., p. 2, note 3.
47. Andrews, Blacks and Whites in Sáo Paulo, chapters 6 and 7. Andrews discusses

a tendency among light-skinned Afro-Brazilians, who identify themselves as
mulattoes, to emphasize their social superiority over darker-skinned Afro-
Brazilians. See ibid., pp. 178-179.
48. Anani Dzidzienyo, personal communication.
49. For psychological studies on mulatto identity in present-day Brazil, see, for
instance, Eneida de Almeida Reis, “Mulato: negro-náo-negro e/ou branco-náo-
branco: un estudo psicossocial sobre identidate” (M.A. thesis, Ponticia Univer-
sidade Católica, 1997).
50. Based on substantial in-depth interviews with Afro-Brazilian women I have
conducted in the city o f Sáo Paulo (1997-2001), I have discussed the remark
able lack o f racial consciousness among the urban poor, especially among the
migrant population from the Northeast, most o f whom identify themselves
as morenos (dark-complexioned), not negros. See Mieko Nishida, “Voices of
Otherness: ‘Black* and ‘Japanese* Women in Sáo Paulo, Brazil** (paper pre
sented at the Latin American Studies Association Congress, Washington, D.C.,
September 6-8, 2001); “Gender, Race, and Ethnicity in Urban Brazil: ‘Black*
and Japanese* Women in Sáo Paulo” (paper presented at the American His
torical Association Annual Meeting, San Francisco, Calif., January 5-6,2002).
Conclusion
1. See, for instance, Viotti da Costa, Brazilian Empire, pp. 239-243.
2. Ira Berlin, “Time, Space, and the Evolution o f Afro-American Society on Brit
ish Mainland North America,*’ AMR 85, no. 1 (1980): 44-78; and Curtin, Atlan
tic Slave Trade, p. 268 (Table 77).
3. Nobles, Shades o f Citizenship, p. 68. Although socially, as African Americans,
they have not been regarded as a separate racial group from blacks, mulattoes
have always occupied an ambiguous yet very important position in American
culture. See, for instance, Judith R. Berzon, Neither Black nor White: The M u
latto in American Fiction (New York: New York University Press, 1978). Berzon
maintains that “ the key elements in distinguishing the mulatto from the full-
blooded blacks are sociological and psychological rather than biological**
and deñnes the mulatto as “an individual who reaps certain advantages and
disadvantages in his interaction with both blacks and whites, advantages
and disadvantages which are a direct result o f his mixed racial heritage.” See
ibid., p . 8.
4. On the multiracial movement in the United States, see Nobles, Shades o f Citi
zenship, pp. 130-145. The movement no longer exits today.
5. For discussions on diasporas, see, for instance, Khaching Tólólyan, “ Rethink
ing Diaspora(s): Stateless Power in the Transnational Movement,** Diaspora 5,
no. 1 (1996): 3-36; James Clifford, Routes: Travel and Translation in the Late
Twentieth Century (Cambridge, Mass.: Harvard University Press, 1997),
pp. 244-277; Smadar Lavie and Ted Swedenburg, eds., Displacement, Dias
pora, and Geographies of Identity (Durham, N.C.: Duke University Press,

1996); and Robin Cohen, Global Diaspora: An Introduction (Seattle: University
o f Washington Press, 1997).
6. Monica Schuler, “Alas, Alas, Kongo**: A Social History o f Indentured African

Immigration into Jamaica, 1841-1865 (Baltimore: Johns Hopkins University
Press, 1980).
7. George Gmelch, Double Passage: The Lives o f Caribbean Migrants Abroad and
Back Home (Ann Arbor: University o f Michigan Press, 1992).
8. The original Japanese term dekasegi refers to a seasonal labor migration or

migrant to big cities within Japan. By 1991, the population of Japanese Brazil
ians numbered around 1.2 million, out o f whom some 100,000 had migrated
to Japan. For major legal issues concerning dekassegui (the Brazilian term) see
Masato Ninomiya, org., “D ekasseguiGensho n 't kansuru Simpojium Hokoku-
sho (Sáo Paulo: Sociedade Brasileira de Cultura Japonesa, 1993). As o f the year
2001,250,000 Japanese Brazilians are said to be working in Japan.
9. See, for instance, MIn Japan, Bias is an Obstacle Even for the Ethnic Japanese,”
New York Times, November 13,1991; and “Sons and Daughters o f Japan, Back
from Brazil,” New York Times, November 27,2001. Japanese journalists have
published numerous small accounts on Japanese Brazilian migrant workers in
Japan, such as MKyuzo-suru Nikkei Brajiru-jin: ‘Sokoku* Nippon de Hataraku”
(“The increasing number o f Japanese Brazilians, working in their ‘home
land’ ” ), Asahi Shimbun Weekly AERA, March 19,1991, pp. 46-48. Japanese
American novelist Karen Tei Yamashita, author o f the highly acclaimed Brazil-
Maru (1993) on Japanese immigration to Brazil, has just published a book on
Japanese Brazilian dekassegui workers based on her observations during her
six-month stay in Nagoya, Japan in 1997. See Karen Tei Yamashita, Circle K
Cycles (Minneapolis: Coffee House Press, 2001). Recent scholarly monographs
on dekassegui workers in Japan include Takayuki Tsuda, “ Strangers in the Eth
nic Homeland: The Migration, Ethnic Identity, and Psychological Adaptation
o f Japan's New Immigrant Minorities” (Ph.D. diss.. University o f California at
Berkeley, 1996); Joshua H. Roth, “ Defining Communities: The Nation, the
Firm, the Neighborhood, and Japanese Brazilian Migrants in Japan” (Ph.D.
diss., Cornell University, 1999). Other studies on dekassegui include Keiko
Yamanaka, “Return Migration o f Japanese-Brazilians to Japan: The Nikkeijin
as Ethnic Minority and Political Construct ” Diaspora 5, no. 1 (1996), pp. 65-
98; idem, “ *I will go home, but when?’ Labor Migration and Circular Dias
pora Formation by Japanese Brazilians in Japan,” in Mike Douglas and
Glenda S. Roberts, eds.> Japan and Global Migration: Foreign Workers and
the Advent o f a Multicultural Society (New York: Routledge, 2000); Yoko Sellek,
“Nikkeijin: The Phenomenon o f Return Migration,” in Michael Weiner, ed.,
Japans Minorities: The Illusion o f Homogeneity (New York: Routledge, 1997);
Reimei Yoshioka, Por que migramos do e para o Japáo: os exemplos dos barrios
das Alianzas dos atuais dekasseguis (Sáo Paulo: Massao Ohno Editor, 1995);
Renato Mikio Moriya, Fenómeno dekassegui: um oihar sobre 0$ adlescentes que
ficaram, with a preface by Paul Negri (Róndorina: Edi^óes CEFEL, 2000); and

Daniel Touro Linger» No One Home: Brazilian Selves Remade in Japan (Stan
ford, Calif.: Stanford University Press, 2001).
I am currently working on a book-Iength manuscript on Afro-Brazilian and
Japanese Brazilian women in twentieth-century Sao Paulo, which is tenta
tively entitled: “The (Re) Making o f Gender, Race, and Ethnicity: ‘Black* and
‘Japanese’ Women in S3 o Paulo, Brazil, 1888-2000” Part o f my research has
been published as a Working Paper in Latin American Studies Series by the
University o f Maryland at College Park as Japanese Brazilian Women and
Their Ambiguous Identities: Gender, Ethnicity, and Class in Sdo Paulo (2000).
I am grateful to the Latin American Studies Center o f the University o f
Maryland at College Park and the Center's director, Saul Sosnowski, for a
postdoctoral fellowship I was awarded for the 1997-1998 academic year.
See, fo r m y d eta iled d iscu ssion s o n th e su b ject, M ie k o N ish id a , “ D iasp o ric
Identities in Transformation: Repatriation Movements from Brazil and to
West Africa and Japan” (paper presented at the Latin American Studies
Association Congress, Miami, Florida, March 16-18,2000).
One such example was published in a national newspaper. Rudy E. Carlino,
author o f this very short essay, wrote o f her experience with a taxi driver in
Turkey where she once lived: “I could see his look o f perplexity: His passen
ger was a Chinese-looking woman, with an Italian name, born in the Philip
pines, and she was an American?” See her essay in the section “ Life Is Shot:
Autobiography as Haiku,” Washington Post, February 3,2002, p. Fi.
Jan Arat Scholte, “The World o f Collective Identities in a Globalizing World,”
Review o f International Political Economy 3 (Winter 1996): 565-607,575. 1 am
grateful to David J. Bachner for the citations, a copy o f his draft pages from
his bode manuscript on global education, and his continuing interest in my
study.

Bibliography
M anuscript Sources
Salvador, Brazil
Arquivo da Cúria Metropolitana de Sáo Salvador da Bahía

Arquivo da Igreja de Nossa Senhora do Rosário dos Homens Prétos
A r q u iv o M u n icip a l d a C id a d e d o Salvad or
Arquivo Público do Estado da Bahía
Arquivo da Santa Casa da Misercórdia da Bahía
Arquivo da Sociedade Protetora dos Desvalidos
Rio de Janeiro, Brazil
Arquivo Nacional do Rio de Janeiro

Biblioteca Nacional do Rio de Janeiro
Newspapers
Diário da Bahia (Salvador, Brazil)
Jornal da Bahia (Salvador, Brazil)
O Oculo Mágico (Salvador, Brazil)
Published P rim ary Sources

Abé-Lallemant, Robert. Reise durch Nord =Brasilien im Jahre 1859. 2 vols. Leipzig: F. D.
Bradhaus, i860.
------ . Viagem pelo norte do Brasil no anno de 1S59. TVans. Eduard de Lima Castro.
2 vols. Rio de Janeiro: Instituto Nacional do Livro, 1961.
Agassiz, Louis, Professor and M r s . A Journey in Brazil Boston: Ticker and Fields, 1868.
Almeida, Candino Mendes de. Codigo philippino ou ordena$des e lets do Reino de Portu
gal 14th ed. Rio de Janeiro: Typographia do Instituto Philamathico, 1870.
Bandinel, James. Some Accounts o f the Trade in Slaves from Africa as Connected with
Europe and Africa. London: Longman, Brown, 1862.
Brasil, Diretoría Geral de Estatística. Recenseamento da populado de Imperio do Brazil
a que se procedeu n o d ia i de agosto de 1872. Rio de Janeiro: Leuzinger, 1873-
1876.
British and Foreign State Papers, vol. 40 (London: James Ridway & Sons, Picadilly, 1850-
1851), pp. 428-429.
Burgess, Wilson. Narrative o f a Recent Visit to Brazil to Present an Address on the Slave-
Trade and Slavery, Issued by the Religious Society of Friends. London: Edmund
Marsh, 1853.
Candler, John, and Wilson Burgess. Narrative o f a Recent Visit to Brazil to Present
an Address on the Slave-Trade and Slavery, Issued by the Religious Society of
Friends. London: Edmund Marsh, 1853.
Christie, W. D. Notes on Brazilian Questions. London: Macmillan, 1S65.
Clark, Edwin. A Visit to South America: With Notes and Observations on the Moral and
Physical Features o f the Country, and the Incidents o f the Voyage. London: Dean
and Son, 1978*
Clark, Hamlet. Letters from Spain, Algeria, and Brazil during Past Entomological
Rambles. London: John van Woorst, Paternoster Row, 1867.
Collec^&o deleisdo Imperio do BraziL Rio de Janeiro: Tipográfico Nacional, 1831-1888.
Debret, Jean Baptiste. Viagem pitoresca e história ao Brasil Trans. Slrgio Millet. 3 vols.
in 2 tomes. Sáo Paulo: Martins and Editora da Universidade de Sáo Paulo,
1972.
Dent, Hastings Charles. A Year in Brazil with the Notes on the Abolition o f Slavery, the
Finances o f the Empire» Religion, Meteorology, Natural History,: London: Kegan
Paul, Trench, 1886.
“ Devassa do levante de escravos occorido em Salvador em 1835” Anais do Arquivo
Público do Estado da Bahia 38 (1968): 1-142.
Duncan, John. Travels in Western Africa, in 1845 and 1846: Comprising a Journey from
Whydah, through the Kingdom o f Dahomey, to Adofoodia, in the Interior. 2 vols.
New York: Johnson Reprint Corporation, 1967 [1847].
Dundas, Robert. Sketches o f Brazil including New Views on Tropical and European
Fever, with Remarks on a Premature Decay o f the System Incident to Europe
ans on Their Return from Hot Climates. London: John Churchill, 1852.
“ 1835 In su m id o de escravos.” Anais do Arquivo Público do Estado da Bahia 40 (1970):
9-170.
Elwes, Robert. A Sketcher's Tour round the World: Illustrations from Original Drawings
by the Author. London: Hurst and Blackett, 1854.
Ewbank, Thomas. Life in Brazil; A Journal o f a Visit to the Land o f the Cocoa and the
Palm. New York: Harper and Brother, 1856.
Fala receitada da Assembléia Legislativa (falos e relatorios dos presidentes em exercio da
Provincia). Bahia, 1839-1889.
Fletcher, James C., and Daniel P. Kidder. Brazil and the Brazilians: Portrayed in Histori
cal and Descriptive Sketches. 9th ed. Boston: Little, Brown, 1879.
Gardner, George. Travels in the Interior o f Brazil Prinapally through the Northern Prov
inces and the Gold and Diamond Districts during the Years 1836-1841. London:
Milford House, 1987 [1846].
Gaston, James McFadden. Hunting a Home in Brazil The Agricultural Resources and
Other Characteristics of the Country— Also, the Manners and Customs o f the
Inhabitants. Philadelphia: King, Baird, Partners, 1867.
Graham, Maria Dundas (Lady Maria Calcott). Journal o f a Voyage to Brazil and Resi
dence There, during Part o f the Years ¡821, 1822, 1823. New York: Frederick A.
Praeger, 1969 (1824].
Grant, Andrew. History o f Brazil Comprising a Geographical Account o f That Country,
228 Bibliography
Together with a Narrative o f the Most Remarkable Events Which Have Occurred
There since the Discovery; A Description o f the Manners, Customs, Religion, &
C. o f the Natives and Colonists; Interspersed with Remarks on the Nature o f
Its Soul, Climate, Production, and Foreign and Internal Commerce. London:
Printed for Henry Colburn» 1809.
Hartt, Charles Frederick Geology and Physical Geography o f Brazil Boston: Fields,
Osgood, 1870.
Henderson, James. A History o f the Brazil: Comprising Its Geography, Commerce, Coloni
zation, Aboriginal Inhabitants, & C .& C .& C . London: Longman, Hurt, Rees,
Orme, and Brown, 1821.
Kidder, Daniel P. Sketches o f Residence and Travels in Brazil 2 vols. Philadelphia: Sorin
and Ball, 1845.
Koster, Henry. Travels in Brazil 1809-1815. 2 vols. London: Longman, Hurst, Rees,
Orme, and Brown, 1817.
Undley, Thomas. Narrative o f a Voyage to Brazil. London: Printed for J. Johnson, 1805.
Marjoribanks, Alexander. Travels in South and North America. 5th ed. London: Simp-
kin, Marshal], 1854.
“Officio do Govemador Conde da Ponte para o Visconde de Anadia, no qual o in
forma das provincias que tomara para destruir os Quilombos, formados pelos
escravos fúgidos dos seus senhores. Bahía, 7 de abril, 1807” Anais da Biblioteca
National do Rio de Janeiro 37 (1915): 450-451.
“Officio do Govemador Conde da Ponte para o Visconde de Anadia, em que Ihe da
parte das provincias que adoptara para evitar um lavantamento dos escravos
contra os brancos, de que tivera denuncia. Bahía, 16 de junho de 1807” Anais
da Biblioteca National do Rio de Janeiro 37 (1915): 460-461.
Prior, James. Voyage along the Eastern Coast o f Africa to Mosambique, Johanna, and
Quiloa to St. Helena; to Rio de Janeiro, Bahia and Pernambuco in Brazil in the
Nisus Frigate. London: Printed for St. Richard Phillips, 1819.
Rugendas, Joáo Mauricio. Viagem pitoresca através do Brasil Trans. Sérgio Millet. Säo
Paulo: Livraria Martins Editora, 1972.
Spix, Johann B. von, and Carl Friedrich P. von Martius. Viagem pelo Brasil Trans.
Lucia Furquim Lahmeyer. 2nd ed. 4 vols. Rio de Janeiro: Impresa Nacional,
1938.
Tollenare, L. F. de. Notas dominiais. Recife: Governo do Estado de Pernambuco, 1979.
Walsh, Robert. Notices o f Brazil in 1818 and 1829. 2 vols. London: Fredrick Westley and
A. H. Davis, 1830.
Wetherell, James. Brazil Stray Notes from Bahia: Being Extracts from Letters, &.c, dur
ing a Residence o f Fifteen Years. Ed. William Hadfield. Liverpool: Webb and
Hunt, i860.
Secondary Sources
Accioli de Cerqueira e Silva, Ignácio. Memórias históricas e políticas da Bahia. 6 vols.
Salvador: Imprensa Oficial do Estado, 1931.
Aimes, Hubert H. S. “Coartación: A Spanish Institution for the Advancement of
Slaves into Freedom.” Yale Review 17 (1909): 412-431.
Bibliography 229
Alden, Dauril. “ Late Colonial Brazil, 1750-1808” In Leslie Bethell, éd., Colonial Brazil
Cambridge: Cambridge University Press, 1987.
Algranti, Leila Mezan. O feitor ausente; estudo sobre a escravidâo urbana no Rio de
Janeiro de 1808 a 1821. Petrópolis: Vozes, 1988.
Allen, Judith Lee. “Tailors, Soldiers, and Slaves: The Social Anatomy o f a Conspiracy.”
M.A. thesis, University o f Wisconsin-Madison, 1987.
Amaral, José Alvares do. “Crónicas dos Acontecimentos da Bahia (1809-1828)." Anais
do Arquivo Publico do Estado da Bahia 16 (1938): 50-95.
Andrade, Maria José de Souza. A m&o de obra escrava em Salvador 1811-1860. Sào Paulo:
Corrupio, 1988.
Andrews, George Reid. The Afro-Argentines o f Buenos Aires, 1800-1900. Madison: Uni*
versity o f Wisconsin Press, 1980.
-------. Blacks and Whites in Sào Paulo, Brazil 1888-1988. Madison: University o f Wis
consin Press, 1991.
Andrews, George Reid, Emilia Viotti da Costa, and Franklin W. Knight. “George Reid
Andrews, Blacks and Whites in Sào Paulo, Brazil, 1888-1988. Madison, WI:
University o f Wisconsin Press, 1991. Introduction, Two Comments, and a
Rejoinder by the Author.” Lttso-Brazilian Review 29, no. 2 (1992): 141-158.
Aufderheide, Patricia Ann. “Order and Violence: Social Deviance and Social Control in
Brazil 1780-1840” Ph.D. diss., University o f Minnesota, 1976.
Azevedo, Celia Maria Marinho de. Onda negra, medo branco: 0 negro no imaginàrio
das elites no siculo XIX. Rio de Janeiro: Paz e Terra, 1987.
Azevedo, Thales de. Povoamento da cidade do Salvador. 2nd ed. Sio Paulo: Companhia
Editora Nacional, 1955.
Bacelar, Jeferson Afonso, and Maria Conceiçào Barbosa de Sousa. O Rosàrio dos
Prétos do Pelourinho. Salvador: Fundaçâo do Patiomònio Artistico e Cultura
da Bahia, 1974.
Barickman, B. J. A Bahian Counterpart: Sugar, Tobacco, Cassava, and Slavery in the
Recôncavo, 1780-1860. Stanford, Calif.: Stanford University Press, 1998.
Barman, Roderick J. Brazil': The Forging o f a Nation, 1798-1852. Stanford, Calif.: Stan
ford University Press, 1988.
Barth, Fredrik. “ Introduction.” In Fredrik Barth, ed., Ethnic Groups and Boundaries:
The Social Organization o f Cultural Difference. Bergen-Oslo: Universitets
Forlaget, and London: George Allen and Unwin, 1969.
Bastide, Roger. The African Rdigions o f Brazil: Toward a Sociology of the Interpenetra
tion o f Civilizations. Trans. Helen Sebba. Baltimore: Johns Hopkins University
Press, 1978.
Bedeles, Hilary McL). Natural Rebels: A Social History o f Enslaved Black Women in Bar
bados. New Brunswick, N.J.: Rutgers University Press, 1989.
Behar, Ruth. “Sexual Witchcraft, Colonialism, and Women's Powers: Views from the
Mexican Inquisition.* In Asunción Lavrin, ed. Sexuality and Marriage in Colo
nial Latin America. Lincoln: University o f Nebraska Press, 1989.
Berkhofer, Robert F., Jr. A Behavioral Approach to Historical Analysis. New York: Free
Press, and London: Collier-Macmiilan, 1969.
Berlin, Ira. “ Herbert G. Gutman and the American Working Class.” In Herbert G.
230 Bibliography
Gutman» Power and Culture: Essays on the American Working Class, ed. Ira
Berlin. New York: Pantheon Books, 1987.
------ . Slaves without Masters: The Free Negro in the Antebellum South. New York:
Oxford University Press, 1974.
-------. “Time, Space, and the Evolution o f Afro-American Society on British Main
land North America.” American Historical Review 85, no. 1 (1980): 44-78.
Berlin, Ira, and Philip D. Morgan, eds. The Slaves* Economy: Independent Production by
Slaves in the Americas. London* Frank Cass, 1991.
Berzon, Judith R. Neither Black nor White: The Mulatto Character in American Fiction.
New York; New York University Press, 1978.
Bethell, Leslie. The Abolition o f the Brazilian Slave Trade: Britain, Brazil and the Slave
Trade Question 1807-1869. Cambridge: Cambridge University Press, 1970.
-------. “The Independence o f Brazil” In Leslie Bethell, ed., Brazil Empire and the
Republic, 1822-1930. Cambridge: Cambridge University Press, 1989.
Bethell, Leslie, and José Murilo de Carvalho. “ 1822-1850” In Leslie Bethell, ed., Brazil
Empire and the Republic, 1822-1930. Cambridge: Cambridge University Press,
1989.
Blassingame, John W. The Slave Community: Plantation Life in the Antebellum South.
Revised and enlarged ed. New York and Oxford: Oxford University Press, 1979.
Boles, John B. Black Southerners 1619-1869. Lexington: University Press o f Kentucky,
1983 -
Bolster, W. Jeffrey. Black lacks: African American Seamen in the Age o f Sail Cambridge,
Mass.: Harvard University Press, 1997.
-------. “ "To Feel Like a Man*: Black Seamen in the Northern States, 1800-1860." Jour
nal o f American History 76, no. 4 (1990): 1173-1199.
Borges, Dain. The Family in Bahia, Brazil ¡870-1945. Stanford, Calif.: Stanford Univer
sity Press, 1992.
Boon, James A. Other Tribes, Other Scribes: Symbolic Anthropology in the Comparative
Study o f Cultures, Histories, Religions, and Texts. Cambridge: Cambridge Uni
versity Press, 1982.
Boschi, Caio César. Os leigos e o poder [Irmandades leigas e política colonizadora em
Minas Gerais). Sâo Paulo: Editora Atica, 1986.
Bott, Elizabeth. Family and Soáal Network: Roles, Norms, and External Relationships in
Ordinary Urban Families. London: Tavistock Publications, 1959.
Bowser, Frederick P. The African Slave in Colonial Peru 1524-1650. Stanford, Calif.: Stan
ford University Press, 1974.
------ . “The Free Person o f Color in Mexico C ity and Lima: Manumission and
Opportunity, i650-i8$0.” In Stanley L. Engerman and Eugene D. Genovese,
eds., Race and Slavery in the Western Hemisphere: Quantitative Studies. Prince
ton, N.J.: Princeton University Press, 1975.
Boxer, C R. The Golden Age o f Brazil 1695-1750: Growing Pains o f a Colonial Society.
Berkeley and Los Angeles: University o f California Press, 1962.
-------. Race Relations in the Portuguese Colonial Empire 1415-1825. Oxford: Clarendon
Press, 1963.
------ . “Some Remarks on the Social and Professional Status o f Physicians and Sur-
Bibliography 231
geons in the Iberian World, i6th-i8th Centuries.” Revista de histôria de Säo
Paub 50, no. 1 (1974): 197-219.
-------. Women in Iberian Expansion Overseas 2415-1813: Some Facts, Fancies and Per
sonalities. New York; Oxford University Press, 1975.
Braga, Julio Santana. Ojogo de büzios: un estudo da adivinaçâo no candomblé. Säo
Paulo: Editora Brasiliens, 1988.
-------. Sociedade protetora dos desvalidos: uma irmandade de cor. Salvador, Bahia:
Ianamä, 1987.
Breen, T. H. "Creative Adaptations: Peoples and Cultures.” In Jack P. Greene and J. R.
Pole, eds., Cobnial British America: Essays in the New History o f the Early Mod
em Era, Baltimore: Johns Hopkins University Press, 1984.
Breen, T. H., and Stephen Innes. “Myne Owne G rou n dR a ce and Freedom on Vir
ginia’s Eastern Shore, 1640-1676. New York: Oxford University Press, 1980.
Britto, Eduard A. de Calda. "Levantes de prétos na Bahia.” Revista do Instituto
Geogräfico e Histörico da Bahia 10, no. 20 (1903): 69-94.
Brown, David Hilary. “Garden in the Machine: Afro-Cuban Sacred Act and Perfor
mance in Urban New Jersey and New York.” Ph.D. diss., Yale University, 1989.
Bunster, Ximena, and Elsa M. Chaney. Sellers and Servants: Working Women in Lima,
Peru. New York: Praeger, 1985.
Burdick, John. Blessed Anastdcia: Women, Race, and Popular Christianity in Brazil New
York: Routledge, 1998.
Burke, Peter. History and Social Theory. Ithaca, N.Y.: Cornell University Press, 1993.
Bush, Barbara. Slave Women in Caribbean Society 1650-1835. Bloomington: Indiana
University Press, 1990.
Butler, Kim D. Freedoms Given, Freedoms Won: Afro-Brazilians in Post-Abolition, Säo
Paub and Salvador. New Brunswick, N.J.: Rutgers University Press, 1998.
Bynum, Caroline Walker. Jesus as Mother: Studies in the Spirituality of the High Middle
Ages. Berkeley, Los Angeles, and London: University o f California Press, 1982.
Cadena, Marisol de la. “ ‘Women Are More Indians': Ethnicity and Gender in a Com
munity near Cuzco.” In Brooke Larson and Olivia Harris, eds., with Enrique
Tandeter, Ethnicity, Markets, and Migration in the Andes: At the Crossroads o f
History and Anthropology. Durham, N.C.: Duke University Press, 1995.
Campbell, Carl. “John Mohammed Bath and the Free Mandingos in Trinidad: The
Question o f Their Repatriation to Africa 1831-38.” Journal of African Studies
12, no. 4 (1975 ): 4 6 7 - 4 9 5 .
Campos, J. da Silva. “Ligeiras notas sobre a vida intima; costumes e religiäo dos
africanos na Bahia.” Anais do Arquivo Publico do Estado da Bahia 29 (1943):
291- 309.
Canny, Nicholas, and Anthony Pagden, eds. Cobnial Identity in the Atlantic World,
1500-1800. Princeton, N.J.: Princeton University Press, 1987.
Cardoso, Fernando Henrique. Capitalismo e escravidäo no Brasil meridional 0 negro na
sociedade escravocrata do Rio Grande do S u l Säo Paulo: Difusäo Européia do
Livro, 1962.
Cardozo, Manoel S. “The Lay Brotherhoods o f Colonial Bahia " The Catholic Histori
cal Review 33, no. 1 (1947): 12-30.
Carneiro, Edison. Ladinos e crioulos. Rio de Janeiro: Editôra Civilizaçâo Brasileira, 1964.
232 Bibliography
Carroll» Patrick J. Blacks in Colonial Veracruz: Race, Ethniáty, and Regional Develop
ment. Austin: University o f Texas Press, 1991.
Carvalho, Marcus J. M. de. Uberdade: rotinas e ruptures do escravismo no Recife, 1822-
i8$o. Recife: Editora Universitaria de UFPE, 1998.
Chalhoub, Sidney. “Slaves, Freedmen and the Politics o f Freedom in Brazil: The Expe
riences o f Blacks in the C ity of Rio.” Slavery and Abolition 10, no. 3 (1989):
64-84«
------ . Visdes da liberdade: urna história das últimas décadas da escravidáo na corte.
Sáo Paulo: Companhia das Letras, 1990.
Chance, John K., and William B. Taylor. “Cofradías and Cargos: An Historical Perspec
tive on the Mesoamerican Civil-Religious Hierarchy.” American Ethnologist 12,
no. 1 (1985): 1-26.
------ . “ Estate and Class in a Colonial City: Oaxaca in 1792." Comparative Studies in
Society and History 19, no. 4 (1977): 454-487.
Clifford, James. Routes: Travel and Translation in the Late Twentieth Century. Cam
bridge, Mass.: Harvard University Press, 1997.
Cohen, Abner. Custom and Politics in Urban Africa: A Study o f Hausa Migrants in
Yoruba Towns. Berkeley and Los Angeles: University o f California Press, 1969.
------ . Two Dimensional Man: An Essay on the Anthropology o f Power and Symbolism
in Complex Society. London: Routledge and Kegan Paul, 1974.
Cohen, Abner, ed. Urban Ethnicity. London: Tavistock Publications, 1974.
Cohen, Anthony P. “O f Symbols and Boundaries; or. Does Erti*s Greatcoat Hold the
Key?” in Anthony P. Cohen, ed., Symbolising Boundaries: Identity and Diversity
in British Cultures. Manchester: Manchester University Press, 1986.
------ . Self Consciousness: An Alternative Anthropology o f Identity. New York: Rout
ledge, 1994.
------ . The Symbolic Construction o f Community. London: Tavistock Publications, 1985.
Cohen, David W., and Jack P. Greene, eds. Neither Slave nor Free: The Freedmen o f Afri
can Descent in the Slave Societies o f the New World. Baltimore: Johns Hopkins
Cohen, Robin. Global Diaspora: An Introduction. Seattle: University o f Washington
Press, 1997.
Collier, Jane Fishburne, and Sylvia Junko Yanagisako, eds. Gender and Kinship: Essays
toward a Unified Analysis. Stanford, Calif.: Stanford University Press, 1987.
Conrad, Robert Edgar. Children o f God's Fire: A Documentary History o f Black Slavery
in Brazil. Princeton, N.J.: Princeton University Press, 1983.
------ . The Destruction o f Brazilian Slavery 1850-1888. Berkeley and Los Angeles: Uni
versity o f California Press, 1972.
------ . “Neither Slave nor Free: The Emancipados o f Brazil, 1818-1868 ” Hispanic
American Historical Review 53, no. 1 (1973): 50-70.
-------. “Nineteenth-Century Brazilian Slavery.” In Robert Brent Toplin, ed., Slavery
and Race Relations in Latin America. Westport, Conn.: Greenwood Press, 1974.
-------. World o f Sorrow: The African Slave Trade to Brazil Baton Rouge: Louisiana
State University Press, 1986.
Cope, R. Douglas. The Limits o f Racial Domination: Plebeian Soaety in Colonial Mex
ico City; ¡660-1720. Madison: University o f Wisconsin Press, 1994.
Bibliography 233
Costa, traci del Nero da. Arraia-miùda: um estudo sobre os náo-proprietários de ecravos
no Brasit Sào Paulo: MGSP Editores Ltda., 1992.
Coxt Edward L. Free Coloreds in the Slave Societies o f St. Kitts and Grenada, 1763-1833.
Knoxville: University of Tennessee Press, 1984.
Crahan, Margaret E., and Franklin W. Knight, eds. Africa and the Caribbean: The Lega
cies o f a Link. Baltimore: Johns Hopkins University Press, 1979.
Craton, Michael. Testing the Chains: Resistance to Slavery in the British West Indies.
Ithaca, N.Y.: Cornell University Press, 1982.
Crook, Larry, and Randall Johnson, eds. Black Brazil• Culture, Identity, and Social Mobi
lization. Los Angeles: UCLA Latin American Studies Center, University of
California, 1999.
Cunha, Manuela Cameiro da. Antropologia do Brasil mito, história, etnicidade. Sáo
Paulo: Brasiliense, 1986.
------ . “Introduction” In Marianno Cameiro da Cunha, Da senzala ao sobrado:
arquitetura brasileira na Nigèria e na República Popular do Benin. S9 o Paulo:
Nobel, 1985.
-------. Negros, estrangeiros: os escravos libertos e sua volta à Africa, Sáo Paulo: Editora
Brasiliense, 1985.
-------. “Silences o f the Law: Customary Law and Positive Law on the Manumission
o f Slaves in the 19th Century BraziL” History and Anthropology 1, p t 2 (1985):
417-443.
Cunha, Mariano Cameiro da Cunha. Da senzala ao sobrado: arquitetura brasileira na
Nigèria e na República Popular do Benin. Sáo Paulo: Nobel, 1985.
Curtin, Philip D. The Atlantic Slave Trade: A Census. Madison: University o f Wiscon
sin Press, 1969.
Curtin, Philip D., ed. Africa Remembered; Narratives by West Africans from the Era of
the Slave Trade. Madison: University o f Wisconsin Press, 1967.
DaMatta, Roberto. A casa & a rua: espaço, cidadania, mulher e morte no Brasil Sáo
Paulo: Brasiliense, 1985.
Davis, David Brion. The Problem o f Slavery in the Age o f Revolution 1770-1823. Ithaca,
N.Y.: Cornell University Press, 1975.
Degler, Carl N. Neither Black nor White: Slavery and Race Relations in Brazil and the
United States. Madison: University o f Wisconsin Press, 1986 [1971].
Desan, Suzanne. Reclaiming the Sacred: Lay Religion and Popular Politics in Revolution
ary France Ithaca, N.Y.: Cornell University Press, 1990.
Dzidzienyo, Anani. “Africa-BraziL Ex Africa Semper Aliquki Novi.” In Larry Crook and
Randall Johnson, eds., Black Brazil Culture, Identity, and Social Mobilization. Los
Angeles: UCLA Latin American Studies Center, University of California, 1999-
-------. “ Brazil.” In Jay A. Sigler, ed., International Handbook on Race and Race Rela
tions. New York: Greenwood Press, 1987.
-------. “ Brazilian Race Relations Studies: Old Problems, New Ideas?” Humboldt Jour
nal o f Social Relations 19, no. 2 (1993): 109-129.
Dominguez, Virginia R. White by Definition: Social Classification in Creole Louisiana.
New Brunswick, N.J.: Rutgers University Press, 1986.
Eisenberg, Peter L. “ Ficando livre: as alforrias em Campinas no siculo XIX.” Estudos
Ecónomicos 17, no. 2 (1987): 175-216.
234 Bibliography
Elkins, Stanley M. Slavery. 3rd ed. Chicago: University o f Chicago Press, 1978 [1959].
Ellison, Ralph. Invisible Man. New York: Vintage, 1982 (1952).
Eltis, David. Economic Growth and the Ending o f the Transatlantic Slave Trade. New
York: Oxford University Press, 1987.
------ . The Rise o f African Slavery in the Americas. Cambridge: Cambridge University
Press, 2000.
Eltis, David, and James Walvin, eds. The Abolition o f the Atlantic Slave Trade: Origins
and Effects in Europe, Africa, and the Americas. Madison: University o f Wis
consin Press, 1981.
Epstein, A. L. Ethos and Identity: Three Studies in Ethniaty. London: Tavistock Publica
tions, 1978.
------ . “Gossip, Norms and Social Network.” In J. Clyde Mitchell, ed.> Social Net-
works in Urban Situations: Analyses o f Personal Relationships in Central Afri
can Towns. Manchester: Manchester University Press, 1969.
------ . “The Network and Urban Social Organization" Rhodes-Livingstone Journal29
(1961): 29-62.
Fernandes, Florestan. A integrado do negro na sociedade de classes. 2 vols. Sáo Paulo:
Dominus Editóra, 1965.
Fields, Barbara. Slavery and Freedom on the Middle Ground: Maryland during the Nine
teenth Century. New Haven, Conn.: Yale University Press, 198$.
Filho, Luís Viana. O negro na Bahia. 3rd ed. Rio de Janeiro: Nova Fronteira, 1988.
Finley, M. I. “Slavery.” In International Encyclopedia o f the Social Sciences, vol. 14,307-
314. New York: Macmillan and Free Press, 1968.
Flexor, Maria Helena. Oficiáis mecánicos na cidade do Salvador. Salvador: Prefeitura
Municipal do Salvador, 1974.
Flory, Rae Jean Dell. “Bahian Society in the Mid-Colonial Period: The Sugar Planters,
Tobacco Growers, Merchants, and Artisans o f Salvador and the Recóncavo *
Ph.D. diss., University o f Texas at Austin, 1978.
Flory, Rae Jean Dell, and David Grant Smith. “ Bahian Merchants and Planters in the
Seventeenth and Early Eighteenth Centuries.” Hispanic American Historical
Review 58, no. 4 (1978): 571- 5 9 4 *
Flory, Thomas. Judge and Jury in Imperial Brazil, 1808-1871: Social Control and Political
Stability in the New State. Austin: University o f Texas Press, 1981.
Foster, George M. Culture and Conquest: America’s Spanish Heritage. New York:
Wenner-Gren Foundation for Anthropological Research, i960.
Fox-Genovese, Elizabeth. “ Placing Women’s History in History.” New Left Review 133
(1982): 5-29.
------ . Within the Plantation Household: Black and White Women o f the South. Chapel
Hill: University o f North Carolina Press, 1988.
Fragoso, Joáo Luis R. and Manolo G. Florentino, “Melcelino, filho de Inocéncia cri-
oula, neto de Joana Cabina: um estudo sobre familias escravas em Parafba do
Sul.” Estudos Económicos 17, no. 2 (1987): 151-173.
Fredrickson, George M. The Black Image in the White Mind: The Debate on Afro-
American Character and Destiny, 1817-1914. New York: Harper and Row, 1971.
Frey, Sylvia R. Water from the Rock: Black Resistance in a Revolutionary Age. Princeton,
N.J.: Princeton University Press, 1991.
Bibliography 235
Freyre, Gilberto. The Mansions and the Shanties: The Making o f Modem Brazil Trans,
a n d ed . H a rriet d e O n is. N e w York: A lfre d A . K n o p f, 1963.
-------. The Masters and the Slaves: A Study in the Development o f Brazilian Civiliza
tion. Trans. Samuel Putman. New York: Alfred A. Knopf> 1946.
-------. O escravo nos anuncios dejom ais brasileiros do siculo XIX. 2nd ed. S3 o Paulo:
Companhia Editora Nacional, 1979.
Fuente, Alejandro de la. A Nation for All- Race, Inequality, and Politics in Twentieth-
Century Cuba. Chapel Hill: University o f North Carolina Press, 2001.
Galliza, Diana Soares de. O declfnio da escravidUo na Parafba ¡850-18S8. Jo3 o Pessoa;
Editora Universit£ria, 1979*
Genovese, Eugene D. From Rebellion to Revolution: Afro-American Slave Revolts in the
Making o f the Modem World. Baton Rouge: Louisiana State University Press,
1979 .
-------. Roll, fordan, Roll: The World the Slaves Made. New York: Vintage Books, 1976.
Gluckman, Max. “Analysis o f a Social Situation in Modem Zululand.” Bantu Studies
14 (1940): 1-30,147-174.
-------. “Gossip and Scandal: Papers in Honour o f Melville J. Herskovits.” Current
Anthropology 4, no. 3 (1963): 307-316.
Gmelch, George. Double Passage: The Lives o f Caribbean Migrants Abroad and Back
Home. Ann Arbor: University o f Michigan Press, 1992.
Golden, Claudia Dale. Urban Slavery in the American South, 1820-1860. Chicago: Uni
versity o f Chicago Press, 1976.
Goody, Jack. “Writing, Religion, and Revolt in Bahia ” Visible Language 20, no. 3
(1986): 318-343-
Gorender, Jacob. O escravismo colonial 4th ed. S io Paulo: Atica, 1978.
Goulart, Mauricio. A escravid&o African# no Brasil (Das origins £ extinf&o do trafico).
S5 o Paulo: Alfa-Omega, 1975.
Graef, Hilda. Mary: A History of Doctrine and Devotion. Westminster: Christian Clas
sics, and London: Sheed and Ward, 1985 [1963 (Part 1) and 1965 (Part 2)].
Graham, Richard. “Brazilian Slavery Re-examined: A Review Article” Journal o f Social
History 3, no. 4 (1970): 431- 453 *
-------. Patronage and Politics in Nineteenth-Century Brazil Stanford, Calif.: Stanford
-------. “Slave Families on a Rural Estate in Colonial BraziL” Journal o f Social History
9» no. 3 (1976): 382-402.
Graham, Richard, ed. The Idea of Race in Latin America, 1870-1940. Austin: University
o f Texas Press, 1990.
Greene, Jack P. Imperatives, Behaviors, and Identities: Essays in Early American Cultural
History. Charlottesville: University Press o f Virginia, 1992.
-------. The Intellectual Construction o f America: Exceptionalism and Identity from 1492
to 1800. Chapel Hill: University o f North Carolina Press, 1993.
-------. Pursuits o f Happiness: The Social Development o f Early Modem British Colonies
and the Formation o f American Culture. Chapel Hill: University o f North
Carolina Press, 1988.
Gudeman, Stephen, and Stuart B. Schwartz. “Cleaning Original Sin: Godparenthood
and the Baptism o f the Slaves in Eighteenth-Century Bahia.* In Raymond T.
236 Bibliography
Smith, ed., Kinship Ideology and Practice in Latin America, Chapel Hill: Uni
versity o f North Carolina Press, 1984.
Guimaráes, Antonio Sérgio Alfredo, and Lynn Huntley, eds. Tirando a máscara: ensaios
sobre o racismo no Brasil Sáo Paulo: Paz e Terra, 2000.
Guss, David M. “The Selling o f San Juan: The Performance o f History in an Afro-
Venezuelan Community.” American Ethnologist 20, no. 3 (1993): 451- 473 *
Gutiérrez, Rámon A. The Political Legacies o f Columbus: Ethnic Identities in the United
States. Working Papers no. 16. College Park, Md.: Department o f Spanish and
Portuguese, University o f Maryland at College Park, 1995.
------ . When Jesus Came, the Com Mothers Went Away: Marriage, Sexuality, and
Power in New Mexico, 1500-1846. Stanford, Calif.: Stanford University Press,
1991-
Gutman, Herbert G. The Black Family in Slavery and Freedom, 1750-1925. New York:
Vintage Books, 1977 [1976].
------ . “ Interview with Herbert Gutman.” In MARHO, Henry Abelove, et al., eds.,
Visions o f History. New York: Pantheon Books, 1983.
-------. Power and Culture: Essays on the American Working Class. Ed. Ira Berlin. New
York: Pantheon Books, 1987*
Hahner, June E. Poverty and Politics: The Urban Poor in Brazil, 1870-1920. Albuquer
que: University o f New Mew Mexico Press, 1986.
Hall, Gwendolyn Mildo. Africans in Colonial Louisiana: The Development ofAfro-
Creole Culture in the Eighteenth Century. Baton Rouge: Louisiana State Uni
versity Press, 1992.
H a n c h a r d , M i c h a e l G e o r g e . Orpheus and Power: The Movimento Negro o f Rio de Ja
neiro and Sáo Paulo, Brazil 1945-1988. P r i n c e t o n , N.J.: P r i n c e t o n U n i v e r s i t y
P re ss, 1994.
Hanchard, Michael George, ed. Racial Politics in Contemporary Brazil Durham, N.C.:
Duke University Press, 1999.
Handler, Jerome S. The Unappropriated People: Freedmen in the Slave Society o f Bar
bados. Baltimore: Johns Hopkins University Press, 1974.
Handler, Richard. “Is ‘Identity* a Useful Cross-Cultural Concept?” In John R. Gills,
ed., Commemorations: The Politics o f National Identity: Princeton, N.J.: Prince
ton University Press, 1994.
Handlin, Oscar. Boston’s Immigrants: A Study in Acculturation. Revised and enlarged
ed. Cambridge, Mass.: Harvard University Press, 1981.
Harding, Rachel E. A Refuge in Thunder: Camdomblé and Alternative Spaces o f Black
ness. Bloomington: Indiana University Press, 2000.
Hasenbalg, Carlos, Nelson do Valle Silva, and Márcia Lima. Cor e estratifico social
Rio de Janeiro: Contra Capa Libraría, 1999.
Helg, Aline. Our Rightful Share: The Afro-Cuban Struggle for Equality, 1886-1912.
Chapel Hill: University o f North Carolina Press, 1995.
Hemming, John. Red Gold: The Conquest of the Brazilian Indians. Cambridge, Mass:
Harvard University Press, 1978.
Henretta, James A. “Social History as Lived and Written.” American Historical Review
84, no. 5 (1979): 1293-1322.
Higgins, Kathleen Joan. “Licentious Liberty” in a Brazilian Gold-Mining Region: Slav
Bibliography 237
ery> Gender>and Social Control in Eighteenth-Century Sabará, Minas Gerais.
University Park: Pennsylvania State University Press, 1999.
-------. “The Slave Society in Eighteenth-Century Sabará: A Community Study in
Colonial Brazil" Ph.D. diss., YaJe University, 1987.
Higman, B. H. Slave Populations o f the British Caribbean 1807-1834. Baltimore: Johns
Hopkins University Press, 1984.
Hobsbawm, Eric J. Nations and Nationalism since 1780: Programme, Myth, Reality. 2nd
ed. Cambridge: Cambridge University Press, 1992.
Hobsbawn, Eric J., and Terence Ranger, eds. The Invention o f Tradition. Cambridge:
Cambridge University Press, 1983.
Horton, James Oliver. Free People o f Color: Inside the African American Community.
Washington, D.C.: Smithsonian Institute Press, 1993.
Hünefeldt, Christine. Paying the Price o f Freedom: Family and Labor among Lima*$
Slaves, i8oo- i 8$4. Berkeley and Los Angeles: University o f California Press,
1994.
Hunt, Nancy Rose. “ Placing African Women’s History and Locating Gender.” Social
History 14, no. 13 (1989): 359 - 37 9 -
Ianni, Octavio. As metamorfoses do escravo. 2nd ed. Sáo Paulo: Editora Hucitec, 1988.
-------. Escravidáo e racismo. 2nd ed. Sáo Paulo: Editora Hucitec, 198$.
Ignace, Etinne. “A revolta dos Malés (24 para 25 de Janeiro de 1835)"Revista de Insti
tuto Geográfico e Histórico da Bahia 14, no. 33 (1907): 129-149.
Inikori, Joseph E., and Stanley L Engerman, eds. The Atlantic Slave Trade: Effects on
Economics, Societies, and Peoples in Africa, the Americas, and Europe. Durham,
N.C.: Duke University Press, 1992.
Isaac, Rhys. “Ethnographic Methods in History: An Action Approach." Historical
Methods 13, no. 1 (1980): 43-61.
-------. The Transformation o f Virginia 1740-1790. Chapel Hill: University o f North
Carolina Press, 1982.
Jacobson-Widding, Anita. “ Introduction." In Anita Jacobson-Widding, ed., Identity:
Personal and Socio-culturaL Uppsala, Sweden: Acta Universitatis Upsaliensis,
1983.
Johnson, Lyman L. “Artisans." In Louisa Shell Hoberman and Susan Migden Socolow,
eds., Cities and Society in Colonial Latin America, Albuquerque: University of
New Mexico, 1986.
-------. “Manumission in Colonial Buenos Aires, 1776-1810.” Hispanic American His
torical Review 59, no. 2 (1979): 258-279.
Jones, Jacqueline. Labor o f Love, Labor o f Sorrow: Black Women, Work, and the Family
from Slavery to the Present. New York: Basic Books, 1985.
Joyner, Charles. Down by the Riverside: A South Carolina Slave Community. Urbana
and Chicago: University o f Illinois Press, 1984.
Karasch, Mary C. “ From Porterage to Proprietorship: African Occupations in Rio de
Janeiro, 1808-1850." In Stanley L. Engerman and Eugene D. Genovese, eds.,
Race and Slavery in the Western Hemisphere: Quantitative Studies. Princeton,
-------, Slave Life in Rio de Janeiro 1805-1850. Princeton, N.J.: Princeton University
Press, 1987.
238 Bibliography
------ . “Suppliers, Sellers, Servants, and Slaves.” In Louisa Schell Hoberman and
Susan Migden Socolow, eds., Cities and Society in Colonial Latin America.
Albuquerque: University o f New Mexico Press, 1986.
Kennedy, John. “ Bahian Elites, 1750-1822.” Hispanic American Historical Review 53,
no. 39 (1973 ): 415 - 4 3 9 .
Kent, R. K. “African Revolt in Bahia: 24-25 January 1835.” Journal o f Social History 3,
no. 4 (1970): 334 - 356 .
------ . “ Palmares: An African State in Brazil” Journal o f African History 6, no. 2
(1965): 161-175.
Kessler, Arnold. “ Bahian Manumission Practices in the Early Nineteenth Century.”
Paper presented to the American Historical Association, San Francisco, 1973.
Khan, Aisha. “Juthaa in Trinidad: Food, Pollution, and Hierarchy in a Caribbean Dias-
pora Community” American Ethnologist 21, no. 2 (1994): 245-269.
Kiernan, James Patrick. “The Manumission o f Slaves in Colonial Brazil: Paraty, 1789-
1822.” Ph.D. diss., New York University, 1976.
Kilson, Martin L., and Robert I. Rotberg, eds. The African Diaspora: Interpretative
Essays. Cambridge, Mass.: Harvard University Press, 1976.
Klein, Herbert S. African Slavery in Latin America and the Caribbean. New York:
------ . “African Women in the Atlantic Slave Trade.” In Claire C. Robertson and
Martin A. Klein, eds., Women and Slavery in Africa, Madison: University of
Wisconsin Press, 1983.
------ . The Middle Passage: Comparative Studies in the Atlantic Slave Trade, Princeton,
------ . Slavery in the Americas: A Comparative Study o f Virginia and Cuba. Chicago:
University o f Chicago Press, 1967.
Knight, Franklin W. Slave Society in Cuba during the Nineteenth Century Madison:
University o f Wisconsin Press, 1970.
Knight, Franklin W., and Margaret E. Crahan. “The African Migration and the Ori
gin o f an Afro-American Society and Culture.” In Margaret E. Crahan and
Franklin W. Knight, eds., Africa and the Caribbean: The Legacies of a Link.
Baltimore: Johns Hopkins University Press, 1979.
Knight, Franklin W., and Peggy K. Liss, eds. Atlantic Port Cities: Economy, Culture, and
Society in the Atlantic World, 1650-1850. Knoxville: University o f Tennessee
Press, 1991.
Kondo, Dorinne K. Crafting Selves: Power, Gender, and Discourses o f Identity in a Japa
nese Workplace. Chicago: University o f Chicago Press, 1990.
Kopytoff, Jean Herskovits. A Preface to Modem Nigeria: The “Sierra Leonians” in
Yoruba, 2830-2890. Madison: University o f Wisconsin Press, 1965.
Kraay, Hendrik. “ ‘As Terrifying as Unexpected1: The Bahian Sabinada, 1837-1838.”
Hispanic American Historical Review 42, no. 4 (1992): 501-527.
Kurk, Betty M. “The African Derivation o f Black Fraternal Orders in the United
States.” Comparative Studies in Society and History 25, no. 4 (1983): 559-591.
Kuznesof, Elizabeth Anne. “ Sexual Politics, Race and Bastard-Bearing in Nineteenth-
Century Brazil: A Question of Culture or Power.” Journal o f Family History 16,
no. 3 (1991): 241-260.
Bibliography 239
-------. “Sexuality, Gender and the Family in Colonial Brazil.” Luso-Brazilian Review
30, n o. 1 (1993): 119-132.
Lauderdale Graham, Sandra. “ Honor among Slaves" In Lyman L. Johnson and Sonya
Lipsett-Rivera, eds., The Faces o f Honor: Sex, Shame, and Violence in Colonial
Latin America. Albuquerque: University o f New Mexico Press, 1998.
-------. House and Street: The Domestic World o f Servants and Masters in Nineteenth-
Century Rio de Janeiro. Cambridge: Cambridge University Press, 1988.
-------. “Slavery's Impasse: Slave Prostitutes, Small-Time Mistresses, and the Brazil
ian Law o f 1871.** Comparative Studies in History and Slavery 33, no. 4 (1991):
669-694.
Lavie, Smadar, and Ted Swedenburg, eds. Displacement, Diaspora, and Geographies of
Identity. Durham, N.C.: Duke University Press, 1996.
Leach, Edmund. A Runaway World? London: British Broadcasting Corporation, 1968
(1967).
Lebsock, Susanne. The Free Women o f Petersburg: Petersburg, Virginia 1784-1820. New
York: Norton, 1984.
Levine, Lawrence W. Black Culture and Consciousness: Afro-American Folk Thought
from Slavery to Freedom. Oxford: Oxford University Press, 1978.
Levine, Robert M. “ ‘Mud-Hut Jerusalem’: Canudos Revised" Hispanic American His
torical Review 68, no. 3 (1988): 525-576.
Lewin, Linda. “ Natural and Spurious Children in Brazilian Inheritance Law from
Colony to Empire: A Methodological Essay." Americas 48, n a 3 (1992): 351-396.
-------. Politics and Parentela in Paranaiba: A Case Study o f Family-Based Oligarchy in
BraziL Princeton, N.J.: Princeton University Press, 1987.
Lindsay, Lisa A. “ 'To Return to the Bosom o f Their Fatherland': Brazilian Immigrants
in Nineteenth-Century Lagos.” Slavery and Abolition 15, no. 1 (1994): 22-50.
Linger, Daniel Touro. No One Home: Brazilian Selves Remade in Japan. Stanford, Calif.:
Stanford University Press, 2001.
Lipsett-Rivera, Sonya. “A Slap in the Face o f Honor: Social Transgression and Women
in Late Colonial Mexico." In Lyman L. Johnson and Sonya Lipsett-Rivera, eds.
The Faces o f Honor: Sex, Shame, and Violence in Colonial Latin America. Albu
querque: University o f New Mexico Press, 1998.
Littlefield, Daniel C. Rice and Slaves: Ethnicity and the Slave Trade in Colonial South
Carolina. Baton Rouge: Louisiana State University Press, 1981.
Lockhart, James. The Nahuas after the Conquest: A Social and Cultural History o f the
Indians o f Central Mexico, Sixteenth through Eighteenth Centuries. Stanford:
Stanford University Press, 1992.
Lockhart, James, and Stuart B. Schwartz. Early Latin America: A History o f Colonial
Spanish America and BraziL Cambridge: Cambridge University Press, 1983.
Lovejoy, Paul E. “ Background to Rebellion: The Origins o f Muslim Slaves in Bahia."
Slavery and Abolition 15, no. 2 (1994): 151-180.
-------. Caravans o f Kola: The Hausa Kola Trade 1700-1900. Zaria, Nigeria: Ahmadu
Bello University Press, 1980.
-------. Transformations in Slavery: A History o f Slavery in Africa. Cambridge: Cam
bridge University Press, 1983.
240 Bibliography
Lugar, Catherine. “The Merchant Community o f Salvador, Bahia 1780-1830" Ph.D.
diss., State University o f New York at Stony Brook, 1980.
Machado, Maria Helena Pereira Toledo. Crime e escravidáo: trabalho, Iuta e resistència
rtas paulistas ¡830-1888, Sào Paulo: Editora Brasiliense, 1987.
Maestri, Mário. “Considerado sobre o cativeiro do 'negro da terra* no Brasil quinhen-
tista.” Estudos Ibero-Americanos 16, no. 1-2 (1990): 197-210.
Marchant, Alexander. From Barter to Slavery: The Economic Relations o f Portuguese
and Indians in the Settlement o f Brazil 1500-1580. Baltimore: Johns Hopkins
Marcilio, Maria Luiza. “The Price o f Slaves in XlXth Century Brazil: A Quantitative
Analysis o f the Registration o f Slave Sales in Bahia ” In Studi in memoria di
Federigo Melis, 5 vols., vol. 5»83-97. Paris: Giannini Editore, 1978.
Martinez-Alier, Verena. Marriage, Class and Color in Nineteenth-Century Cuba: A
Study o f Racial Attitudes and Sexual Values in a Slave Society. London: Cam
bridge University Press, 1974 -
Mattoso, Kátia M. de Queirós. Bahia: a cidade do Salvador e seu mercado no siculo
XIX. Sào Paulo: Hucitec, 1978.
------ . Bahia, siculo XIX: urna provincia no impèrio. Rio de Janeiro: Editora Nova
Fronteira, 1992.
-------. Entre esclave au BrésiL XVIe-XIXe. Paris: Hachette, 1979.
-------. Familia e sociedade na Bahia do siculo XIX. Trans. James Amado. Sáo Paulo:
Corrupto, 1988.
------ . Presenta francesa no movimento democràtico bahiano de 1798. Salvador: Editora
Itapuà, 1969.
------ . “A propósito de cartas de alforrìa na Bahia, 1779-1850." Anais de História 4
(1972): 23-52.
------ . “Slave, Free, and Freed Family Structures in Nineteenth-Century Salvador,
Bahia." Luso-Brazilian Review 25, no. 1 (1988): 69-83.
-------. Testamentos de escravos libertos na Bahia no siculo XIX: urna fonte para o
estudo de mentalidade. Salvador: Centro de Estudos Bahianos, 1979.
------ . To Be a Slave in Brazil 1550-1888. Trans. Arthur Goldhammer. New Bruns
wick, N.J.: Rutgers University Press, 1986.
Mattoso, Kátia M. de Queirós, Herbert S. Klein, and Stanley L. Engerman. “Trends
and Patterns in the Prices of Manumitted Slaves: Bahia, 1819-188#.” Slavery
and Abolition 7, no. 1 (1986): 59-67.
McBeth, Michael C. “The Brazilian Recruit during the First Empire: Slave or Soldier?"
In Dauril Alden and Warren Dean, eds., Essays Concerning the Socioeconomic
History o f Brazil and Portuguese India. Gainesville: University Press o f Florida,
1977.
Mello, Pedro C. de. “Expectation o f Abolition and Sanguinity o f Coffee Planters in
Brazil, 1871-1881." In Robert William Fogel and Stanley L. Engerman, eds.,
Without Consent or Contract: The Rise and Fall o f American Slavery. Condi
tions o f Slave Life and the Transition to Freedom: Technical Papers. 2 vols.
New York: W. W. Norton, 1992.
Menard, Russell, and Stuart B. Schwartz. “W hy African Slavery? Labor Force Transi
Bibliography 241
tions in Brazil, Mexico, and the Carolina Lowcounty” In Wolfgang Binder,
ed., Slavery in the Americas. Würzburg: Königshausen and Newman, 1993.
Menzar, Joan E. “The Ranks o f the Poor: Military Service and Social Differentiation
in Northeast Brazil, 1830-1875.” Hispanic American Historical Review 72, no. 3
(1992): 335 - 351-
Merrick, Thomas W., and Douglas H. Graham. Population and Economic Development
in Brazil 1800 to the Present. Baltimore: Johns Hopkins University Press, 1979.
Metcalf, Alida C. Family and Frontier in Colonial Brazil Santana de PamaQja* ¡580-
¡822. Berkeley, Los Angeles, and Oxford: University o f California Press, 1992.
Meyers, Albert, and Diane Elizabeth Hopkins, eds. Manipulating the Saints: Religious
Brotherhoods and Social Integration in Postconquest Latin America. Hamburg:
WAYASBAH, 1988.
Miers, Suzanne, and Igor Kopytoff, eds. Slavery in Africa: Historical and Anthropologi
cal Perspectives. Madison: University o f Wisconsin Press, 1977.
Miller, Joseph C. Kings and Kinsmen: Early Mhundu States in Angola. Oxford: Oxford
-------. “Slave Prices in the Portuguese Southern Atlantic, 1600-1830." In Paul E. Love-
joy, ed., Africans in Bondage: Studies in Slavery and Slave Trade. Madison: Uni-
versity o f Wisconsin Press, 1986.
-------. Way o f Death: Merchant Capitalism and the Angolan Slave Trade ¡730-1830.
Madison: University o f Wisconsin Press, 1988.
Mintz, Sidney W. “ Defining Black and White.” Johns Hopkins Magazine 46, no. 5
(1 9 9 4 ): 8, to.
-------. “A Melting Pot— Sort o f” Johns Hopkins Magazine 46, no. 4 (1994): 24-25.
-------. “ Men, Women, and Trade.” Comparative Studies in Society and History 13, no. 3
(1971): 247-269.
-------. “The Sensation o f Moving, while Standing Still.” American Ethnologist 16,
no. 4 (1989): 786-796.
-------. “Tasting Food, Tasting Freedom.” In Wolfgang Binder, ed., Slavery in the
Americas. Wtlnzburg: Königshausen and Neumann, 1993.
-------. Worker in the Cane: A Puerto Rican Life History. New York; W. W. Norton, 1974
11960].
Mintz, Sidney W„ and Richard Price. An Anthropological Approach to the Afro-American
Past: A Caribbean Perspective. Philadelphia: Institute for the Study o f Human
Issues, 1976.
-------. The Birth o f African-American Culture: An Anthropological Perspective. Boston:
Beacon Press, 1992.
Mintz, Sidney W., and Eric R. Wolf. “An Analysis o f Ritual Co-Parenthood (Com
padrazgo).” Southwestern Journal o f Anthropology 6, no. 4 (1950): 341-368.
Mitchell, J. Clyde. The Kalela Dance: Aspects o f Social Relationships among Urban Afri
cans in Northern Rhodesia. Manchester: Manchester University Press, 1956.
-------. “Perceptions o f Ethnicity and Ethnic Behaviour: An Empirical Exploration."
In Abner Cohen, ed., Urban Ethnicity. London: Tavistock Publications, 1974.
-------. Tribalism and the Plural Society. London: Oxford University Press, i960.
-------. “Tribe and Social Change in South Central Africa: A Situational Approach ”
Journal o f Asian and African Studies 5 (1970): 83-101.
242 Bibliography
Mitchell, J. Clyde, ed. Social Networks in Urban Situations. Manchester: Manchester
Mitchell, Michael James. “ Racial Consciousness and the Political Attitudes and Behav
ior o f Blacks in Säo Paulo.” Ph.D. diss., Indiana University, 1977.
Monteil, Vincent. “Analyse de 25 documents arabes des Malés de Bahia (1835).** Bulletin
de ¡’Institut Fondanctake d'Afrique Noir>tomo 29, sèrie B, no. 1-2 (1967)- 88-98.
Moreno Fraginals, Manuel, Herbert S. Klein, and Stanley L. Engerman. “The Level
and Structure o f Slave Prices on Cuban Plantations in the Mid-Nineteenth
Century: Some Comparative Perspectives.” American Historical Review 88,
no. s (1983): 1201-1218.
Morgan, Philip D. “British Encounter with Africans and African-Americans, circa
1600-1780.” In Bernard Bailyn and Philip D. Morgan, eds., Strangers within
the Realm: Cultural Margins of the First British Empire. Chapel Hill: University
o f North Carolina Press, 199L
Moriya, Renato Mikio. Fenómeno dekassegui: um olhar sobre os adlescentes que ficaram,
with a preface by Paul Negri Róndorina: Ediôes CEFEL, 2000.
Morrissey, Marietta. Slave Women in the New World: Gender Stratification in the Carib
bean. Lawrence: University Press o f Kansas, 1989.
Morrison, TonL The Bluest Eye. New York: Plum, 1994 (1970).
Morton, F. W. O. “The Military in Bahia, 1800-1821” Journal o f Latin American Studies
7, no. 2 (1975): 249-269.
Mullin, Michael. Africa in America: Skive Acculturation and Resistance in the Ameri
can South and the British Caribbean, 1736-1831. Urbana: University o f Illinois
Press, 1992.
Mulvey, Patricia A. “Black Brothers and Sisters: Membership in the Black Lay Brother
hoods o f Colonial Brazil” Luso-Brazilian Review 17, no. 2 (1980): 253-279.
------ . “The Black Lay Brotherhoods o f Colonial Brazil.” Ph.D. diss., City College o f
New York, 1976.
------ . “ Slave Confraternities in Brazil: Their Role in Colonial Society.” Americas 39,
no. 1 (1982): 39-68.
Mura, David. Turning Japanese: Memories o f a Sansei. New York: Doubleday, 1991.
Nascimento, Abdias do. Brazil Mixture or MassacreÌ Essays in the Genocide o f a Black
People. Trans. Elisa Larkin Nascimento. 2nd ed. Dover, Mass.: Majority Press,
1989.
Nascimento, Abdias do, and Elisa Larkin Nascimento. Africans in Brazil A Pan-African
Perspective. Trenton, N.J.: Africa World Press, 1992.
------ . “ Reflexöes sobre o movimento negro no Brasil, 1938- 1997” In Antonio Sérgio
Alfredo Guimaràes and Lynn Huntley, eds., Tirando a máscara: ensaios sobre 0
racismo no Brasil Sáo Paulo: Paz e Terra, 2000.
Nascimento, Anna Amélia Vieira. Dezfreguesias da cidade do Salvador: aspectos sociais
e urbanos do sáculo XIX. Salvador: Funda^äo Cultural do Estado da Bahia,
1986.
Nash, Gary B. Forging Freedom: The Formation o f Philadelphia’s Black Community,
1720-1840. Cambridge, Mass.: Harvard University Press, 1988.
Nazzari, Muriel. Disappearance o f the Dowry: Women, Families, and Social Change in
Säo Paulo, Brazil (1600-1900). Stanford, Calif.: Stanford University Press, 1991.
Bibliography 243
Ninomiya, Masato, org. “DekasseguiGertsho ni kansuru Simpojium Hokoku-sho. Sáo
Paulo: Sociedade Brasileira de Cultura Japonesa» 1993.
Nishida, Mieko. “Diasporic Identities in Transformation: Repatriation Movements
from Brazil and to West Africa and Japan” Paper presented at the Latin
American Studies Association Congress, Miami, Florida, March 16-18» 2000.
-------. “From Ethnicity to Race and Gender: Transformations o f Black Lay Sodalities
in Salvador, Brazil.” Journal o f Social History 32, no. 2 (Winter 1998): 329-348.
-------. “Gender, Race, and Ethnicity in Urban Brazil: ‘Black* and ‘Japanese* Women
in Sáo Paulo.” Paper presented at the American Historical Association Annual
Meeting, San Francisco, CA, January 3-6,2002.
-------. Japanese Brazilian Women and Their Ambiguous Identities: Gender, Ethnicity,
and Class in Sáo Paub. Latin American Studies Working Paper no. 5. Latin
American Studies Center, University o f Maryland at College Park, 2000.
-------. “Manumission and Ethnicity in Urban Slavery: Salvador, Brazil, 1808-1888.”
-------. “Voices o f Otherness: 'Black* and ‘Japanese' Women in Sáo Paulo, Brazil.”
Paper presented at the Latin American Studies Association Congress, Wash
ington, D.C., September 6-8,2001.
Nobles, Melissa. Shades o f Citizenship: Race and Censuses in Modem Politics, Stanford,
Calif.: Stanford University Press, 2000.
Oboler, Suzanne. Ethnic Labels, Latino Lives: Identity and the Politics o f (Re)Presenta-
tion in the United States. Minneapolis: University o f Minnesota Press, 1995.
Oliveira, Maria Inés Córtes de. O liberto: o seu mundo e os outros. Sáo Paulo: Corrupio,
1988.
Palmié, Stephan. “Ethnogenetic Processes and Cultural IVansfer in Afro-American
Slave PopulationsIn Wolfgang Binder, ed., Slavery in the Americas. Würz
burg: KOnigshausen 8c Newman, 1993.
-------. “Slave Culture and the Culture o f Slavery in North America: Some Remarks
on Historiography and the Present State o f Research.” In Wolfgang Binder,
ed., Slavery in the Americas. Würzburg: Kdnigshausen 8c Newman, 1993.
------ . “Spies or Spades? Racial Classification and Ethnic Conflict in Miami.”
Amerikastudien/American Studies (1989): 211-221.
Patai, Daphne. Brazilian Women Speak: Contemporary Life Stories. New Brunswick,
N.J.: Rutgers University Press, 1988.
Patterson, Orlando. Slavery and Social Death: A Comparative Study. Cambridge, Mass.:
Harvard University Press, 1982.
Pérotin-Dumon, Anne. “The Emergence o f Politics among Free-Coloreds and Slaves
in Revolutionary Guadeloupe.” Journal of Caribbean History 25, nos. 1-2
(1991): 100-135.
Phillips, Caryl. Crossing the River. New York: Vintage, 1993.
Phillips, William D., Jr. Slavery from Roman Times to the Early Transatlantic Trade.
Minneapolis: University o f Minnesota Press, 1985.
Pierson, Donald. Negroes in Brazil A Study o f Race Contact at Bahia. Chicago: Univer
sity o f Chicago Press, 1941.
Preston, James J., ed. Mother Worship: Theme and Variations. Chapel Hill: University
o f North Carolina Press, 1982.
244 Bibliography
Price, Richard, ed. Maroon Societies: Rebel Slave Communities in the Americas. 2nd ed.
Prince, Howard M. “Slave Rebellion in Bahia 1807-1835.” Ph.D. diss., Columbia Univer
sity, 1972.
Querino, Manoel. Costumes africanos no BrasiL Rio de Janeiro: Civilizaqto Brasileira, 1938.
Raboteau, Albert J. Slave Religion: The “Invisible Institution” in the Antebellum South.
New York: Oxford University Press, 1978.
Raeders, Georges. O inimigo cordial do BrasiL 0 conde de Gobineau no Brasil. Trans.
Rosa Freire d'Aguiar. Sao Paulo: Paz e Terra, 1988.
Ralston, Richard D. “The Return o f Brazilian Freedmen to West Africa in the 18th and
19th Centuries.” Canadian Journal o f African Studies 3, no. 3 (1969): 577-592.
Ramos, Donald. “Social Revolution Frustrated: The Conspiracy o f the Tailors in
Bahia, 1798.” Luso-Brazilian Review 13, no. 1 (1976): 74-90.
Ranger, Terence. The Invention o f Tribalism in Zimbabwe. Gweru, Zimbabwe: Mambo
Press, 1985.
Reichert, Rolf. Os documentos ardbes do Arquivo do Estado da Bahia, Salvador: Univer-
sidade Federal da Bahia, 1970.
Reichmann, Rebecca. “ Introduction.” In Rebecca Reichmann, ed., Race in Contempo
rary Brazil From Indifference to Inequality. University Park: Pennsylvania
State University Press, 1999.
Reis, Eneida de Almeida. “Mulato: negro-nào-negro e/ou branco-nào-branco: un
estudo psicossocial sobre identidate.” M.A. thesis, Pontfcia Universidade
Católica, 1997.
Reis, Joào José. “A elite baiana face os movimentos sociais, Bahia: 1824-1840.” Revista
de História 54, no. 108 (1976): 342-384-
------ . “ Magia jeje na Bahia: a invasto do calundo do Pasto de Cachoeira, 1785.”
Revista Brasileira de História 8, no. 16 (1988): 57-81.
------ . A morte é urna festa: ritos funebres e revolta popular no Brasil do século XIX.
Sào Paulo: Companhia das Letras, 1991.
------ . Rebeliäo escrava no BrasiL a história do levante dos malés (1835). 2nd ed. Sào
Paulo: Editora Brasiliense, 1987.
------ . Slave Rebellion in Brazil The Muslim Uprising o f 183$ in Bahia. Trans. Arthur
Brakel. Baltimore: Johns Hopkins University Press, 1993.
------ . “Slave Resistance in Brazil: Bahia, 1807-1835.” Luso-Brazilian Review 25, no. 1
(1988): 111-144.
Reis, Joào José, and Eduard Silva. Negociaflo e conflito: a resisténcia negra no Brasil
escravavista. Sào Paulo: Companhia das Letras, 1989.
Reiter, Rayna Rapp, ed. Toward an Anthropology of Women. New York: Monthly
Review, 1974.
Robertson, Claire C., and Martin A. Klein. “Women’s Importance in African Slave
Systems.” In Claire C. Robertson and Martin A. Klein, eds., Women and Slav
ery in Africa. Madison: University o f Wisconsin Press, 1983.
Robertson, Claire C., and Martin A. Klein, eds. Women and Slavery in Africa. Madi
son: University o f Wisconsin Press, 1983.
Rodrigues, Nina. Os africanos no BrasiL 3rd ed. Sào Paulo: Companhia Editòra
Nacional, 1945.
Bibliography 245
Rosaldo, M. Z., and L. Lampherc, eds. Women, Culture, and Society. Stanford, Calif.:
S ta n fo rd U n iv e rsity Press, 1974.
Roth, Joshua H. “ Defining Communities; The Nation, the Firm, the Neighborhood,
and Japanese Brazilian Migrants in Japan." Ph.D. diss., Cornell University,
1999.
Ruiz, Vicki. Cannery Women, Cannery Lives: Mexican Women, Urbanization, and the
California Food Processing Industry, 1930-1950. Albuquerque: University of
New Mexico Press, 1987.
Russell-Wood, A. J. R. “Aspectos da vida social das irmandades leigas da Bahia no
siculo XVIII ” Universitas, no. 6/7 (1970): 189-204.
-------. “ Black and Mulatto Brotherhoods in Colonial Brazil: A Study in Collective
Behavior” Hispanic American Historical Review 54, no. 4 (1974): 567-602.
-------. The Black Man in Slavery and Freedom in Colonial BraziL New York; S t Mar
tin’s Press, 1982.
-------. “Colonial Brazil” In David W. Cohen and Jack P. Greene, eds., Neither Slave
nor Free: The Freedmen o f African Descent in the Slave Societies o f the New
World. Baltimore: Johns Hopkins University Press, 1972.
-------. “Examination o f Selected Statues o f Three African Brotherhoods.” In Albert
Meyers and Diane Elizabeth Hopkins, eds., Manipulating the Saints: Religious
Brotherhoods and Social Integration in Postconquest Latin America. Hamburg:
WAYASBAH, 1988.
-------. Fidalgos and Philanthropists: The Santa Casa da Misericòrdia o f Bahia, 1550-
1735. Berkeley and Los Angeles: University o f California Press, 1968.
-------. “ Ports o f Colonial Brazil” In Franklin W. Knight and Peggy K. Liss, eds., The
Atlantic Port Cities: Economy, Culture, and Society in the Atlantic World, 1650-
1850. Knoxville: University o f Tennessee Press, 1991.
-------. “ Prestige, Power, and Piety in Colonial Brazil: The Third Orders o f Salvador."
-------. Society and Government in Colonial Brazil, 1500-1822. Brookfield, Vt.: Vari
orum, 1992.
-------. “Women and Society in Colonial Brazil.” Journal o f Latin American Studies 9,
no. 1 (1977): 1-34.
Russell-Wood, A. J. R., ed. From Colony to Nation: Essays on the Independence o f BraziL
Ruy, Afonso. A primeira revoluto social brasileira (1798). 2nd ed. Salvador: Tipografo
Beneditina Ltda. 1951.
Saffioti, Heleieth I. B. Women in Class Society; Trans. Michael Vale. New York: Monthly
Review Press, 1978.
Santos, Juanita Elbein dos. Os nàgó e a morte: pàde, àsèsè e o culto egun na Bahia. 4th
ed. Petrópolis: Vozs, 1986.
Saracino, Rodolfo. Los que volvieron a Africa Havana: Editorial de Ciencias, 1988.
Saunders, A. C de C. M. A Social History o f Black Slaves and Freedmen in Portugal,
1441-1555. Cambridge: Cambridge University Press, 1982.
Scarano, Julita. “ Black Brotherhoods: Integration or Contradiction?” Luso-Brazilian
Review 16, no. 1 (1979): 1-17.
246 Bibliography
------ . Devoçào e escravidào: a irmardade de nossa senhora do Rosârio dos Prétos do
distrito diamarttino no sécub XVIII. 2nd ed. Sâo Paulo: Ed. Nadonal, 1978.
Scholte, )an Arat. “The World o f Collective Identities in a Globalizing World.** Review
o f International Political Economy 3 (Winter 1996): 565-607.
Schuler, Monica. “Alas, Alas, Kongo*: A Social History o f Indentured African Immigra
tion into Jamaica, 1841-1865. Baltimore: Johns Hopkins University Press, 1980.
------ . “ Ethnic Slave Rebellion in the Caribbean and the Guianas.” Journal o f Social
History 3 (1970): 374-385.
Schwartz, Stuart B. “The Formation of a Colonial Identity in Brazil.** In Nicolas
Canny and Anthony Pagden, eds., Colonial Identity in the Atlantic World,
1500-1800. Princeton, N.J.: Princeton University Press, 1987.
------ . “Indian Labor and New World Plantations: European Demands and Indian
Responses in Northern Brazil.” American Historical Review 83» no. 3 (1978):
43-78.
------ . “The Manumission o f Slaves in Colonial Brazil: Bahia, 1684-1745.** Hispanic
American Historical Review 54, no. 4 (1974): 603-635.
------ . “The Mocambo: Slave Resistance in Colonial Brazil.” Journal o f Social History
3, no. 4 (1970): 313-333-
------ . “Plantations and Peripheries, c. 1580-c. 1750.” In Leslie Bethell, e d , Colonial
Brazil Cambridge: Cambridge University Press, 1987.
------ . “Recent Trends in the Study of Slavery in Brazil.” Luso-Brazilian Review 25,
no. 1 (1988): 1-25.
-------. “Resistance and Accommodation in Eighteenth-Century Brazil: The Slave's
View on Slavery." Hispanic American Historical Review 57, no. 1 (1979): 69-81.
------ . Slaves, Peasants, and Rebels: Reconsidering Brazilian Slavery. Urbana: Univer
sity o f Illinois Press, 1992.
------ . Sovereignty and Society in Cobnial Brazil The High Court o f Bahia and Its
Judges, 1609-1751. Berkeley and Los Angeles: University o f California Press,
1973-
------ . Sugar Plantations in the Formation o f Brazilian Society, Bahia, 1550-1835. Cam
bridge: Cambridge University Press, 1985.
Scott, Joan. “Gender: A Useful Category o f Historical Analysis.” American Historical
Review 91, no. 5 (1986): 1053-1075.
Scott, Julius Sherrard, III. “The Common Wind: Currents o f Afro-American Commu
nication in the Era o f the Haitian Revolution.” Ph.D. diss., Duke University,
1986.
Scott, Rebecca J. “ Exploring the Meaning o f Freedom: Postemancipation Societies
in Comparative Perspective." Hispanic American Historical Review 68, no. 3
(1988): 407-428.
------ . Slave Emancipation in Cuba: The Transition to Free Labor 1860-1899. Princeton,
Seed, Patricia. To Love, Honor, and Obey in Cobnial Mexico: Conflicts and Marriage
Choice, ¡574-1821. Stanford, Calif.: Stanford University Press, 1988.
Sellek, Yoko. “ Nikkeijn: The Phenomenon o f Return Migration ” In Michael Weiner,
ed., Japan’s Minorities: The Illusion o f Homogeneity. New York: Routledge, 1997.
Bibliography 247
Silva, Maliene Rosa Nogeira da. Negro na rúa: a nova face de escravidáo. Sáo Paulo:
Hucitec, 1988.
Silva, María Beatriz Nizza da. Sistema de casamento no Brasil colonial Sáo Paulo:
EDUSP, 1984.
Silva, Nelson do Valle. “ Updating the Cost o f Not Being White in Brazil.” In Pierre-
Michel Fontaine, ed„ Race, Class, and Power in Brazil. Los Angeles: Center for
Afro-American Studies, University o f California, Los Angeles, 1980.
Sio, Arnold A. “Marginality and Free Coloured Identity in Caribbean Slave Society”
Slavery and Abolition 8, no. 2 (1987): 166-182.
Skidmore, Thomas. MBi-racial U S A . vs. Multi-racial Brazil: Is the Contrast Still
Valid?” Journal o f Latin American Studies 25, no. 2 (1993): 373-386.
-------. Black into White: Race and Nationality in Brazilian Thought. New York:
-------- . "R a cia l Ideas a n d S o c ia l P o lic y in B ra zil, 1870-1940.” In R ic h a rd G r a h a m , ed .,
The Idea o f Race in Latin America. Austin: University o f Texas Press, 1990.
Sienes, Robert Wayne. “The Demography and Economics o f Brazilian Slavery: 1850-
1888.” Ph.D. diss., Stanford University, 1975.
-------. “ Escravidáo e familia: padrees de casamento e estabilidade familiar numa
comunidade escrava.” Estudos Económicos 17, no. 2 (1987): 217-227.
-------. “ Lares negros, olhos brancos: história da familia escrava no siculo XIX ”
Revista Brasileira de História 8, no. 16 (1988): 189-203.
Smith, Raymond T., ed. Kinship Ideology and Practice in Latin America. Chapel Hill
and London: University o f North Carolina Press, 1984.
Soares, Luiz Carlos. “Os escravos de ganho no Rio de Janeiro do siculo XIX.” Revista
Brasileira de História 8, no. 16 (1988): 107-142.
-------. Rameiras, ilhoas> polacas. . . : a prostituito no Rio de Janeiro do século XIX.
Sáo Paulo: Editora Atica, 1992.
Sousa, Paulo Cesar. A Sabirtada: a revolta separatista da Bahia (1837). Sáo Paulo:
Editora Brasiliense, 1987.
Souza, Laura de Mello e. O diabo e a terra de Santa Cruz: feitiâria e religiosidade popu
lar no Brasil colonial Sáo Paulo: Companhia das Letras, 1986.
Stein, Stanley J. Vassouras, a Brazilian Coffee County, 1850-1900: The Roles o f Planter
and Slave in a Plantation Society. Princeton, N.J.: Princeton University Press,
1985 [1957].
Stuckey, Sterling. Slave Culture: Nationalist Theory and the Foundations o f Black
America. New York: Oxford University Press, 1985.
Tannenbaum, Frank. Slave and Citizen: The Negro in the Americas. New York: Alfred A.
Knopf, 1947.
Tavares, Luis Henrique Dias. Comércio proibido de escravos. Sáo Paulo: Editora Atica,
1988.
Taylor, William B. “ Between Global Process and Local Knofwledge: An Inquiry into Early
Latin American Social History” In Olivier Zunz, ed., Reliving the Past. The
Worlds o f Social History. Chapel Hill: University o f North Carolina Press, 1985.
-------. Drinking, Homicide, and Rebellion in Colonial Mexican Village. Stanford,
Calif.: Stanford University Press, 1979.
248 Bibliography
------ . Magistrates o f the Sacred: Priests and Parishioners in Eighteenth-Century
Mexico. Stanford, Calif.: Stanford University Press, 1996.
------ . “The Virgin o f Guadalupe in New Spain: An Inquiry into the Social History
o f Marian Devotion.” American Ethnologist 14, no. 1 (1987): 9 - 33 *
Thompson, E. P. The Making o f the English Working Class. New York: Vintage Books,
1966 (1963).
Thornton, John. Africa and Africans in the Making o f the Atlantic World, 1400-1680.
Cambridge: Cambridge University Press, 1992.
Iblölyan, Khaching. “ Rethinking Diaspora(s): Stateless Power in the Transnational
Movement.” Diaspora 5, no. 1 (1996): 3-36.
Toplin, Robert Brent. The Abolition of Slavery in Brazil New York: Atheneum, 1972.
Trosko, Barbara Rose. “The Liberto in Bahia before Abolition.” M.A. thesis, Columbia
University, 1967.
Tsuda, Takayuki. “Strangers in the Ethnic Homeland: The Migration, Ethnic Identity,
and Psychological Adaptation o f Japan's New Immigrant Minorities.” Ph.D.
diss., University o f California at Berkeley, 1996.
Turner, J. Michael. “Les Brésiliens—-The Impact o f Former Brazilian Slaves upon
Dahomey.” Ph.D. diss., Boston University, 1975 *
Turner, Lorenzo D. “Some Contacts of Brazilian Ex-Slaves with Nigeria, West Africa."
Journal o f Negro History 27, no. 1 (1942): 55-67.
IWinan, Ann. Public Lives, Private Secrets: Gender; Honor, Sexuality, and Illegitimacy in
Colonial Spanish America. Stanford, Calif.: Stanford University Press, 1999.
Vail, Leroy, ed. The Creation o f Tribalism in Southern Africa. London: James Currey,
1989.
Venàncio, Renata Pinto. “Nos limites da sagrada familia: illegitimidade e casamento
no Brasil colonial.” In Ronaldo Vanifes, ed., Histôria e sexualidade no Brasil.
Rio de Janeiro: Graal, 1986.
Verger, Pierre. Flux e Reflux de la Traite des Negres entre le Golfe De Bénin et Bahia de
Todos os Santos du Xvii au Xix Siècle. 3 vols. Paris: Mouton, 1968.
------ . Fluxo e refluxo do tráfico de escravos entre o Golfo do Benin e a Bahia de Todos
os Santos, dos séculos XVII a XIX. Trans. Tasso Gadzanis. 2nd ed. Sâo Paulo:
Editora Corrupio, 1987.
------ . Formation d'une société Brésilienne au Golfe du Bénin au XIXème Siècle. Dakar:
Centre de Hautes Etudes Afro-Ibéro-Amé ricanes de L'Université de Dakar,
1969.
------ . Noticias da Bahia— 1850. Sâo Paulo: Corrupio, 1981.
------ . Os libertos: sete caminhos na Hberdade de escravos da Bahia no sécuio XIX. Sâo
Paulo: Corrupio, 1992.
------ . Trade Relations between the Bight o f Benin and Bahia from the 17th to 19th Cen
tury. Trans. Evelyn Crawford. Ibadan, Nigeria: Ibadan University Press, 1976.
Vilhena, Lufs dos Santos. A Bahia no século XVIII. 3 vols. Bahia: Editôra Itapuâ, 1969.
Viotti da Costa, Emilia. The Brazilian Empire: Myths and Histories. Chicago: Dorsey
Press, 1988.
------ . Crowns o f Glory,, Tears o f Blood: The Demerara Slave Rebellion o f 1823. New
York and Oxford: Oxford University Press, 1994.
Bibliography 249
-------. Da senzala à colònia. 2nd ed. S io Paulo: Livraria Editora Ciéncias Humanas
L td a., 1967.
Wade, Peter. Blackness and Race Mixture: The Dynamics o f Racial Identity in Colombia.
-------. “The Cultural Politics of Blackness in Colombia ” American Ethnologist 22,
no. 2 (1995): 341- 357.
Wade, Richard C. Slavery in the Cities: The South 1820-1860. New York: Oxford Univer
sity Press, 1964.
Walker, Sheila S. “The Feast o f Good Death: An Afro-Catholic Emancipation Celebra
tion in Brazil.” Sage 3, no. 2 (1986): 27-31.
Warner, Marina. Alone o f A ll Her Sex: The Myth and the Cult o f the Virgin Mary. New
York: Alfred A. Knopf, 1976.
Wimbery, Fayette Darceli. “The African Liberto and the Bahian Lower Class: Social
Integration in Nineteenth-Century Bahia, Brazil 1870-1900” Ph.D. diss., Uni
versity o f California at Berkeley, 1989.
Wood, Peter H. Black Majority: Negroes in Colonial South Carolina from 1670 through
the Stono Rebellion. New York: Alfred A. Knopf, 1974.
Wood, Stephanie. “Adopted Saints: Christian Images in Nahua Testaments o f Late
Colonial Toluca.” Americas 47, no. 3 (1991): 259-294.
Wright, Winthrop R. Café con leche: Race, Class, and National Image in Venezuela.
Austin: University o f Texas Press, 1990.
Yamanaka, Keiko. “ ‘I will go home, but when?* Labor Migration and Circular Dias
pora Formation by Japanese Brazilians in Japan ” In Mike Douglas and
Glenda S. Roberts, eds., Japan and Global Migration: Foreign Workers and
the Advent o f a Multicultural Society. New York: Routledge, 2000.
-------. “ Return Migration o f Japanese-Brazilians to Japan: The Nikkeijin as Ethnic
Minority and Political Construct” Diaspora 5, no 1 (1996): 65-98.
Yamashita, Karen Tei. Brazil-Maru. Minneapolis: Coffee House Press, 1993.
-------. Circle K Cycles. Minneapolis: Coffee House Press, 2001.
Yanagisako, Sylvia Junko. "Family and Household: The Analysis o f Domestic Groups.”
Annual Review of Anthropology 8 (1979): 161-205.
-------. Transforming the Past: Tradition and Kinship among Japanese Americans. Stan
ford, Calif.: Stanford University Press, 1985.
Yoshioka, Reimei. Por que migramos do e para 0 Japào: os exemplos dos barrios das
Alianzas dos atuais dekasseguis. Sao Paulo: Massao Ohno Editor, 1995.
250 Bibliography
Index
Abé-Lallemant, Robert, u Armando, 66

Abolition, 2,116 Arrdba, 15
Abolitionist societies, 143
Acaraji, 45 Bahia, captaincy/providence/state, 12,13
Afilhada, 107, 111 Bahian Recftncavo, 13,40,83,93,94; sugar
African “ nation” (nafdo),30,31,32,38,39,50, economy of, 14-15
98,101,102,105,116,157,162; stereotypes of, Bandinel, James, 21
31-32 Baptism, African slaves of, 30,54
African-bom ex-slaves: common-law unions, Barber-surgeons, 49,85; also employed as
107,109; coparenthood, uo-111; expulsion musicians, 49-50
of, 103; family and kinship, 117; fam ily name, Bastide, Roger, 60,99
87-88,107; fictive kinship, $4; gender and Batuque, 50-51; divided by African “nations,'*
gender relations, 8$; godparenthood, 111; 50-51; prohibitions of, 64,68
grandparents, 108; laws, 102-103; marriage, Bay o f All Saints (Bahia de Todos os Santos),
54,106-107,109-110; membership in black 11,16
Uy sodalities, 89-90; occupations, 85-87; pa Benguela, 31,35
ternity, 107; property-owning, 88; repatria Bicko da p^y$
tions to Africa, 87,103-10$; slavers and slave Bight o f Benin, 15,103,105
merchants, 86-87; slave-owning, 88-89; Bight o f Biafra, 15
wages, 85 Blacks, 11-12,152-153; definitions of, 6; “ polite
Africanism, 151 ness” of, 36
Africano: identity, 92; africano livre (free Afri Black Church o f O u r Lady o f the Rosary in
can), also called emancipados (the emanci the Pelourinho, 33,148
pated), 16,25,104 Black identity, am biguity of, 152-155
Afro-Brazilians, 155-156; activists, 155,156; Black sodality: burial o f the dead with non-
movements of, 156; working-class, 155 Christian rituals, 61-62; “disappearance”
Age o f Revolution, 132,133 of, 135-138; formation and development, 55-
Agoé, 103,104 62; identification o f African gods and god
Aldetra, 14 desses with Christian saints, 61; membership
Alma de leite* 19 o f whites and mulattoes, 59-60; multiple
Andrade, Maria losé de Sousa, 18 membership by African-bom ex-slaves, 89-
Andrews, George Reid, 155 90; new black sodalities, 138-141; positions o f
Angola, 14» 30,31,32,58 crioulos, 58-59,134; women and gender, 59
Anjico, 31 Black sodality: o f G ood Jesus o f the Necessity
Antebellum South, 161 and Redemption in the Corpo Santo church,
Anti-slave trade law (1831), 2,8,25; illegal slave 57; o f G ood Jesus o f Redemption, 90; o f
trade, 16-17; impact on African slavery in Sal Jesus, Mary, and Joseph in the Carmelite
vador, 16; impact on the landscape o f Salva- monastery, 90; o f O u r Lady o f the Good
dor, 93 Death in the Barroquinha church, 57; o f
Arabic, 63; literacy in, 99-100 O u r Lady o f M artyrs in the Barroquinha
Arriba Athuna, 104 church, 60; o f O ur Lady o f the Rosary (in
Conceiîo da Praia, 57; in the PeJourinho, 58- Cento, 2
59» 89» 148; in Santo AntOmo, 89; and on Kua Consensual unions among slaves, 53-54
de Jo3 o Perreira, 90); o f Saint Anthony o f Coparent, 110,111
Catagerona in the chapel o f S io Pedro, 57; Cdr, See Color
o f Saint Benedict (in Conceiîo da Praia, Count o f Arcos, D. Marcos de Noronha, 37,
60; in the Saint Francis monastery, 61; and 66,67
in the Carmelite monastery, 90); o f Saint Count o f Gobineau, Joseph Arthur de
Ephigenia in the Carmelite monastery, 90 Gobineau, 55,104
Black Soul, 155 Count o f Ponte, Jo&o de Saldanha de Gama,
Blackness, 13 63» 65
Boxer, Charles C , 6,136 Coparents o f the fig tree, 67
Brazil: “discovery" o f (1500), 13; independence C rah an, Margaret E., 38
o f (1822), 2; republic o f (1889), 2; transition Creolization, o f diverse African cultures, 4
from mercantilism to free trade (1808), 2,65 Cria, 88,102; cria fbrro, 113
Brazilian identity, 113 Criouto, 26,123,133» 134»135. 13». 141» 147.
Brazilian succession law, 110 148,160
Brazilian-born slaves, 123-130; kinship and Crioulo livre, identity formation of, 134-141
Active kinship, 124-127; manumission of, 127- Crioulo sodalities, 138-141»161
130; skin color, 129 C urtin, Philip D., 31
Brazilians: in Lagos, 115-116,118, i$i; in Japan,
163-164 Day o f the Kings, 49
Brazilwood, 13 Dekassegui, 163
British'Brazilian Mixed Commission, 16 Diaspora communities, formation of, 163
British privilege, 97 Dom inguez, Virginia R., 154
Bush captain, 63 D onatary system, 14
Duncan, John, 105
Cabray 26,131
Cabodo, 26 Egba,38
Cadena, Marisol de la, 44 El Mina, 31
Cafre (Kaffir), 6 Elkins, Stanley, 74
Caixa de empr&timo, 55 Elwes, Robert, 42
CandombU, 96; casa de candombU, 51 Emancipado, 15,25
Canto, 48-50 Engenho, 18,66
Cam eiro da Cunha, Manuela, 75 Engerman, Stanley L., 23
Carta de alforria, 73 Engomadeira, 46
Caruru, 45 Enslavement, 30
Carvalho, M arcu s). M. de, 34 Epstein, A. L., 3,47
Categorization, 3,46 Escrow, 6
Chalhoub, Sidney, 145 Ethnic: affiliation o f the individual, 36; cate
C ivil marriage, 146 gory, 47; communication networks, 52; gath
Class cxUuiivity, 155 ering and grouping, 48-51; identity, 32-38,
Cohen, Anthony P., 5 49.62,70,157-159; labels, 5; origins, 26; soli
Cohen, David W., 6 darity, 8; symbol, 36
Color, 9,26,129,131,146-147» 153- 154.162; cod Ethnicity: definition of, 38; in New World slav
ing system, 6; labels of, 7 ery, 38
Comadre, 94,95,96 Etlis, David, 18
Com m on-law marriage: among ex-slaves, 109; Ewbank, Thom as, 43
among slaves, 108
Companheiro/companheira, 109 Falua, 42
Congo, 31,32 Ferme, Mariane, 99
252 Index
Finley, Moses I., 17,20 Irmandade. See Lay sodality
Fletcher, James C ., and Daniel P. Kidder, Islam, 8,63,96,99,116
40,104
Forra, 25 Jabti (Ijebu), 38
Fraginals, Manuel Moreno, 23 Japanese Brazilian: ethnic identity, 164;
Free population o f color, emergence of, 6 repatriation movements to Japan, 164
Fite Womb Law (Lei de Ventre Livre, 1871), 2, Jesuits, 14
74.144. M5 Jihad, 62-63,9 9
Freedom: for African-born slaves, 84; for Johnson, Lyman, 43
African-bom ex-slaves, 87-91; for Brazilian- Jones, Jacqueline, 39
born slaves, 123; legal, 106; loss of, 43 Junta, 55,56; Junta dos Africanos, 5$
Frente Negra Brasileira, 154
Karasch, M ary C., 44,82
Gabon, 31 Klein, Herbert S., 22,23,24
Gardner, George, 62 Knight, Franklin W „ 38
G<ge (Ewe), 31,37,51» 59 »98 Koster, Henry, 30,31,60-61,74-75
Gender, 49,8$; gender identity, 39-46,141,
159-160 Lagos, 87,103,104,113,114,117, 159; “ Brazil
Genovese, Eugene D., 100 ians” in, 115-116,158
Gentio (gentile), 13 Lauderdale Graham, Sandra, 145
Global: age, 9,164,165; world, 165 Lay Christianity, 137-138
Gmelch, George, 163 Lay sodality: for the population o f color, 7;
Godparenthood, 54,111 divided by race, 56
GoUs, 15 Lew in, Linda, 110
Gold Coast, 31 Liberto, 26
Golden Law (Lei Aurea, 1888), 2 Lindley, Thom as, 31,33,49
Goody, Jack, 30-31,63,100 Lingua gtral, 37
Gossip and rumor, 95 Loja, 19
Graham, Maria Dundas, 84 Lovejoy, Paul L , 99
Greene, Jack P., 6
Gutman, Herbert G ., 4 Mae de santo, 51
M ali revolt (1835), 8,32,38,93-101,103,158
Haitian Revolution, 15,68 Malungo. See Shipmate
Hammock carrier, 40 Mandingo, 83
Hausa, 37,51,62,64,66,98 Manumission, 74-84; laws, 74-75; paid manu
Henderson, James, 36,64-65 mission, 24,76-77; unpaid manumission,
Htnriques, 133 128-129
Higgins, Kathleen Joan, 76 Marjoribanks, Alexander, 20
Hispanic, color and identity, 5 Market woman, 45,159
House, vs. street, 44,155 Marketing, monopolized by African-born
women, 44
Ibeji, 61 Martinez-Alier, Verena, 131
Identity: cards of, 29-31; creation of, 29; defini Massapi soil, 14
tions of, 1; formation of, 4-6; representations Mating, 52-53
of, 48-70; studies on, 3-5 M atto Grosso, 15
Indian: “good" and bad,” 14; slavery, 14 Menzar, Joan E., 133
Indio forra, 25 Mercadeja, 45,78
Ingènuo, 144 Merrick, Thom as W., and Douglas H.
Interprovincial slave trade, 21,22,24 Graham, 24,130
Inter-relatedness, 3 Mestizo, 26,152
Index 253
Middle Passage, 9,29 Pataca, 20
M ilitia regiments: black, 134; colored, 7,133-134 Patai, Daphne, 153
Mina, 31,32; Costa da M ina, 64,74,86,87 Pecuiio, 144
Minas Gerais, 14,82 Peculium, 83
Mingâo, 45 Peiourinho, 33
Mintz, Sidney W., 4,6,32,123 Plantation slavery, 34
Mitchell, J. Clyde, 3,4,38 Port-of-Spain, Trinidad, 83
Mocambo, 63 Porto Novo, 104
Movimento negro, 154-156 Portuguese Court: transfer o f from Lisbon to
Movimento Negro Unificado, 155 Rio de Janeiro (1808), 1; im pact on the trans
Mozambique, 31 atlantic slave trade to Brazil, 16
Mucama ( mucumba), 19 Prito, 6,25,146-147,148-149,161
Mulato, 26 Price, Richard, 4,32
Mulatto identity, 132-133 Prince, Howard, 97,99
Mulatto sodality, 56,132; o f O ur Jesus o f the
Cross in the Church o f Palma, 132: o f O ur Quilombo, 63,64,65,68,139,140
Lady o f Guadalupe, 56; o f O u r Lady o f C on Queirbs Law (1850), 17
ception o f Boqueirào in Sáo Pedro, 132; o f Quitanda, 44
O ur Lord o f Patience in Sáo Pedro, 132 Quitandeira, 44,78
Multiracial movement, 163
Muslim, 93,96 ,9 9 Raboteau, Albert J., 50,61
Raça Brasil 155
Nagô ( Yoruba), 31,32,37,38,51,59.69; Race: perceptions of, 6; categories in the U.S.
Nagó-Bá (Egba), 38, 95 ; Nag&-G«ge, 51; censuses, 162
Nagô-Oyo, 96 Racial identity, 160-163
Names, African, 37-38,47,80,95; Portuguese/ Racial ization, 160-161
Christian, 30-31,37,46,80,115,157; Yoruba, 115 Rebolo, 31
Nation/nationality. See African “ nation” Reis, Joflo José, 19, 93 »99
(naçào) Returnees, 105,159
Native languages, use o f among slaves, 37 Rio de Janeiro, 15,16,50,51,82,86,104
Negro, 6,13,25,148; negro da terra, 13; negro de Rodrigues, Nina, 37,99
Guiñé, 13; negro de ganho, 69; ne$ro novo, 69; Russell-Wood, A. J. R., 53,60
negro de saveiro, 94
Nobles, Melissa, 162 Saint Benedict: monastery, 4; the Moor, 61
Saint Dom ingue, 15
Oboler, Suzanne, 5 Salvador, city of, 1-2* 13; cidade alta and cidade
O gun, 61 baixa of, 11; location and descriptions of, 11;
Oshun, 61 history of, 12; population of, 11-12
O ur Lady: o f Conception, 132; o f Guadalupe, Santa Casa da Misercôrdia, 11,16,56
132; o f Guidance, 94-95; o f the Liberation, Sio Paulo, 154
15a o f Succor, 132 Saios, 105
Oyo, 38,62-63 Savings associations, 55
Schuler, Monica, 163
Pai de santo, 51 Schwartz, Stuart B., 81,82
Palmié, Stephan, 5 Sedan chair carrier, 21,40
Paraguayan War, 143 Self-purchase, by African-bom slaves, 77-84
Paraíba, 15,83 Semana santa, 66
Paraty, 83 Sergipe, 15
Pardo, 26,131,132,146,148,160 Shipmate, 32,39
Passport, 92 Shoes, as a mark o f freedom, 84
254 Index
Shopana, 61 Thom pson, E. P., 4
Sierra Leone, 105» 163 Tobacco, 41; chew ing of, 41
Slave men; artisans, 42-43; communication Trade-in, 79,82,83. See also Substitution as a
among, 40; fishermen, 42; Bights of, 62-64; form o f self-purchase
functions of, 39; porters and transporters, 41; Transatlantic slave trade to Brazil, 14-17; term i
prices, 18,23-24; seamen, 42; stevedores, 41; nation o f (1851), 2,81,105,142
uprisings, 64-70; working in gangs, 38,40 Transnational labor migration, 163-164
Slave population: sex balance in urban slavery, TVibc and tribalism, 3; tribalism as urban eth
17-18; sex imbalance among the Africans im nicity, 3
ported to Bahia, 18 Tupinambds, 13
Stave society, 17,22,23,24; transformation to Tupiniquins, 13
slave-owning society, 93 TUrner, J. Michael, 116
Slave women: as domestics, 18; functions of,
4 4 -4 6 Urban ethnicity, 3
Slave-owning society, 21,22,23,24 Urban slavery: decline of, 21; multi-functions
Slavery in the U.S., 162 o f urban slaves, 2; scale o f urban slavehold-
Slaves: daily interactions, 18; kidnapping of, ing, 19; sex balance in the slave population,
34; market for, 33; sources of, 14-1$ 17-18
Sobrado, 13 Urban studies in South-Central Africa, by
Social death, 9,16$ British social anthropologists at the Univer
Society for the Protection o f the Needy, 139, sity o f Manchester, 3
142,148-153
Sodality o f O ur Lady o f Solitude and Support
Valongo, o f Rio de Janeiro, 22
for the Needy, 138-139
V atapi, 45
Sodality o f Saint Francis o f Paula o f Free-
Venda, 44
Born Crioulos B om in the Empire o f Brazil,
Vilhena, Luis dos Santos, 44
139-141
Viotti da Costa, Emilia, 6
Sousa, T h o m l de, 12
Virgin Mary, as godmother, 136-138
Spix, Johann B. von, and Carl Friedrich P. von
Martius, 20,32,37,43
Street peddling and vending: by slaves, 43; as Wage earning, 20-21,34-35,48,102
“woman’s job," 44 Walsh, Robert, 86
Substitution as a form o f self-purchase, 79-S4 W ay o f death, 9
Sugar cultivation of, 14; economy, 15; produc West Africa, 57,58; as a slave source, 14
tion, 2 West African religions, 50
Wetherell, James, 30,36,49
Tailors’ Revolt (1798), 132-133; 160 W hite sodality, 56
Tannenbaum, Frank, 74 W hitening, 155
Tapa(N upe), 31,37,92 W hydah, 52,103,105
Teatro Experimental de Negro, 155
Terra de bronco* 95 Yam, 95,96
Terreiro de candombU, 51 Yem anji, 60
Index 255
M IE K O N ISH ID A is Assistant Professor o f H istory at H artw ick C ollege
in O neonta, N ew York. She held a Predoctoral Research Fellowship at
the C arter G . W oodson Institute o f the U niversity o f V irgin ia and a
Rockefeller Foundation Postdoctoral Fellowship at the Institute o f Latin
Am erican Studies o f the U niversity o f Texas at Austin.

Slavery and Identity

Uploaded by

Copyright:

Available Formats

Slavery and Identity

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

Slavery and Identity

Uploaded by

Copyright:

Available Formats

Slavery and Identity

Blacks in the Diaspora

Slavery and Identity

601 North Morton Street

Library o f C o n g r a i Cataloging-ln-Publication Data

Nishida, Mieko, date

1. A “ Capital o f A frica” in Brazil 11

PART T W O : T O BE A F R I C A N - B O R N A N D F R E E D , C I R C A 180 8-1880

6. T h e Creation o f Disparate Identity, 1808-1851 123

ACM S Arquivo da Cùria Metropolitana de Säo Salvador da Bahia, Salva­

AAEB Anais de Arquivo Público do Estado da Bahia

Identity is not given. It is one’s situational synthesis o f self in relation to

2 Slavery and Identity

This historical case study focuses on the creation o f individual iden­

Anthropological Studies on Ethnic Identity

M y approach to ethnic identity among the African-born popula­

M y discussions on the historical creation o f collective identity have

4 Slavery and Identity

Race and C o lo r in Brazil

Perceptions o f race and color in Brazil differ considerably from the

6 Slavery and Identity

Structure o f the Book

8 Slavery and Identity

Abé-Lallemant, who characterized the city o f Salvador as “a capital o f A f­

Salvador, Bahia, and the Transatlantic

The C ity o f the Savior (cidade do Salvador), founded by Governor

12 Slavery and Identity

A “Capital o f Africa” in Brazil 13

14 Slavery and Identity

A “Capital o f Africa” in Brazil 15

16 Slavery and Identity

Urban Slavery in N ineteenth-C entury Salvador:

Sex Ratios among the Urban Slave Population

O ne o f the major characteristics o f urban slavery in Salvador is

A uCapital o f Africa” in Brazil 17

18 Slavery and Identity

The Scale o f Urban Slaveholding

Urban slaveholding took place on a much smaller scale than on the

A “Capital o f Africa ” in Brazil 19

Urban slavery in Salvador was also characterized by its wage-earning

A Brazilian has slaves, he sends them out to work at different trades—

20 Slavery and Identity

Decline o f Urban Slavery in Salvador in the 1840s:

The above-mentioned distinctive characteristics o f urban slavery

A “Capital o f Africa” in Brazil 21

22 Slavery and Identity

A “Capital o f Africa ” in Brazil 23

24 Slavery and Identity

Categories for People o f A frican Descent

The population o f African descent in nineteenth-century Salvador

A “Capital o f Africa” in Brazil 25

26 Slavery and Identity

There are no paths in water. No signposts.

An African-born individual’s creation o f new identity began to take place

Enslavement and the Imposition o f

In Angola, on the day before embarkation, slaves were usually assembled

30 Slavery and Identity

The Creation o f New Identity 1808-1831 31

The Creation o f Ethnic Identity

ACM S Arquivo da Cùria Metropolitana de Säo Salvador da Bahia, Salva

This historical case study focuses on the creation o f individual iden

M y approach to ethnic identity among the African-born popula

Abé-Lallemant, who characterized the city o f Salvador as “a capital o f A f

African slaves occupied the urban public space o f Salvador in ac

Black lay sodalities functioned as a very special type o f voluntary as

The dances rtegros are accustomed to performing to the sound o f disso

Ethnic identities were clearly manifested in the African-born popu