Palmares: An African State in Brazil

Author(s): R. K. Kent
Source: The Journal of African History, Vol. 6, No. 2 (1965), pp. 161-175
Published by: Cambridge University Press
Stable URL: http://www.jstor.org/stable/180194 .
Accessed: 13/03/2011 01:22

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .
http://www.jstor.org/action/showPublisher?publisherCode=cup. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].

Cambridge University Press is collaborating with JSTOR to digitize, preserve and extend access to The
Journal of African History.

http://www.jstor.org
Journal of African History, VI, 2 (I965), pp. I6l-175.

PALMARES: AN AFRICAN STATE IN BRAZIL*

R. K. KENT
WITH 0 UT slaves from Africa, reported an early Portuguese source, 'it is
impossible to do anything in Brazil ' 1 Although prior arrivalsare suspected,
the first known landing of slaves from Africa on Brazilian soil took place
in I552.2 In I580, five years after the founding of Loanda and on the eve
of Brazil's sugar boom, there were no fewer than io,ooo Africans in Brazil.3
Fifty years later, Pernambuco alone imported 4,400 slaves annually from
Africa.4 It also contained i5o engenhos,or a third of the total sugar-mill
and plantation complex in Brazil.5 In I630, the Dutch West India Com-
pany (WIC) captured Pernambuco, and within a decade Portugal had
abandoned Brazil to the Dutch. It was ultimately the decision of local
settlers, the moradores, to fight the West India Company that led to
restoration of Portuguese control in I654. The Dutch retreat from Brazil,
however, was secured through a joint Afro-Portuguese effort which gave
the Black Regiment of Henrique Dias its colonial fame. If early settlement
and a sugar-based economy could not have been sustained without the
African labourer, neither could the Portuguese continue to hold Brazil
without the African soldier. The subsequent evolution of Brazil is no less
a story of Euro-African enterprise. Exploitation of gold and diamonds in
the eighteenth century, pioneering shifts of population from the coast to
the interior, dilution of monoculture, formation of mining states or advent
of an abolitionist movement in the nineteenth century were all dependent
on the same combination.6 The blend of race, language and culture in
contemporary Brazil confirms this evolution.
Africa's impact on Brazil and, more generally, the role of the Negro7
* An earlier paper on Palmares was read at University of Wisconsin's seminar in Latin-
American History, under Dr. John Phelan. I am indebted to the Editors of the JAH
and to Dr. Jan Vansina for suggestions which made the revision easier. I am also grateful
to the Foreign Area Program for its I963-4 Fellowship at UW. There is no connexion
between the Program and the views expressed herein.
1 Quoted by J. A. G. de Neto Mello in Novos Estudos Afro-Brasileiros, 1937, II, 205. The
original source is Breve Discurso Sobre o Estado das Quatras Capitanias Conquistadas (n.d.).
2 Artur Hehl Neiva, Provenencia das Primeiras Levas dos Escravos Africanos, Anais,
Instituto Historico e Geographico Brasileiro (I949), IV, 49I-2.
3 Pedro Calmon, Historia do Brasil (959), II, 347.
4 Charles R. Boxer, Salvador de Sa and the Struggle for Brazil and Angola, I602-1686

(1952), 225.
5 Frederic Mauro, Le Portugal et l'Atlantique, 1570-I670 (I960), 193 (Table).
6 The standard English-language works are: Charles R. Boxer, The Golden Age of
Brazil (I96I); Gilberto Freyre, The Masters and the Slaves: A Study in the Development
of Brazilian Civilization (1956); and Arthur Ramos, The Negro in Brazil (I939).
7'Negro', as a term used in colonial Brazil, did not apply to pretos ('Blacks') alone. It
included sometimes pardos or gente do cor, people 'of colour' not easily accepted as either
pretos or brancos ('Whites'). It also applied to crioulos or those born in Brazil of African
or mixed parentage, to ladinos or those who spoke Portuguese and usually espoused the
Catholic faith, and to the Africanos or those who were neither Portuguese-speaking nor
native to Brazil.
 i62 R. K. KENT
in Brazilian history and society are subjects of an extensive literature.8 Its
principal stress is on assimilation rather than divergence, and frequently
the early colonial society has been postulated from descriptions left by
European and North American travellers who visited Brazil much later.9
It is hence not surprising that active Negro resistance to slavery in Brazil
has not received comparable attention and is consequently less known.
According to one working definition, there were three basic forms of
active resistance, namely fugitive slave settlements called quilombos,
attempts at seizure of power, and armed insurrections which sought
neither escape nor control but amelioration.10 The latter two prevailed in
the first half of the nineteenth century, a period of political transition in
Brazil and of accelerated slave trade with Africa." They encompass, for
example, nine Bahian revolts between I807-35, which involved a number
of Hausa, Yoruba and Kwa-speaking groups, as well as the Ogboni Society,
Muslim alufas and even a back-to-Africa movement.12 The quilombos
constitute a pre-nineteenth-century phenomenon and are of considerable
interest to the African historian. They came closest to the idea of recrea-
ting African societies in a new environment and against consistently heavier
odds. Once formed, the quilomboswere regarded as a threat to the Portu-
guese plantation, an inducement for escape from the slave hut. They
were rarely, therefore, allowed to last a long time. Of the ten major
quilombos in colonial Brazil, seven were destroyed within two years of
being formed. Four fell in the state of Bahia in I632, I636, I646 and I796.
The other three met the same fate in Rio in i650, Parahyba in I73I and
Piumhy in I758. One quilombo,in Minas Gerais, lasted from I7I2 to I7I9.
Another, the 'Carlota' of Mato Grosso, was wiped out after existing for
twenty-five years, from I770 to I795.
Nothing, however, compares in the annals of Brazilian history with the
'Negro Republic' of Palmares in Pernambuco. It spanned almost the entire
seventeenth century. Between I672-94, it withstood on the average one
Portuguese expedition every fifteen months.'3 In the last entrada against
Palmares, a force of 6,ooo took part in 42 days of siege.14 The Portuguese
8 For a recent summary in the pages of this Journal, see Jose Hon6rio Rodrigues, 'The
Influence of Africa on Brazil and of Brazil on Africa', J.A.H., III (i962), 49-67.
9 Cf. J. Codman, Ten Months in Brazil (Boston, I867); L. Couty, L'Esclavage au Bresil
(Paris, i88 i); C. H. Dent, A Year in Brazil (London, i886); G. Gardner, Travels in the In-
terior of Brazil, I836-I84I (London, I 849); D. P. Kidder, Sketches of Residence and Travels
in Brazil (London, I845); H. Koster, Travels in Brazil, I809-I8I5 (Philadelphia, I817,
2 vols.); A. de Saint-Hilaire, Voyages dans l'Interieur du Bresil (Paris, i852); and J. B. von
Spix and C. F. P. von Martius, Travels in Brazil, I8I7-I820 (London, I824, 2 vols.).
10 Edison Carneiro, 0 Quilombo dos Palmares, I630-I695 (I947), 13.
11Alfredo Gomes, Achegas para a Historia do Trafico Africano no Brasil, Anais,
I.H.G.B. (1950), V, 56 (Tables I and II).
12 See: Etienne Ignace Brasil, La Secte Musulmane des Males et leur Revolte en 1835,
Anthropos (I909), IV, 99-105 and 405-15; A. E. de Caldas Britto, Levantosde Pretosna
Bahia, Revista,I.H.G.B. (Bahia)(1903), xxix, 69-90; and Nina R. Rodrigues,OsAfricanos
no Bresil (I945) for pertinent chapter.
13, 14 Sebastiao da Rocha Pitta, Historia da America Portugueza (first edition 1730),
i88o, Book VIII, 239; Robert Southey, History of Brazil, III, I8I9, 27; Ernesto Ennes,
 PALMARES: AN AFRICAN STATE IN BRAZIL I63

Crown sustained a cumulative loss of 400,000 cruzados,'5 or roughly three

times the total revenue lease of eight Brazilian Captaincies in i6I 2 .16 As
Brazil's classic quilombo,Palmares gained two more distinctions. It opened
the study of Negro history in modern Brazil. Minutes of the Brazilian
Historical Institute reveal that Palmares caused lively discussions in I840,
and that search for written materials relative to it began in I85 I .17
Important gaps in knowledge persist, but enough primary sources have
been found and published to trace the development of Palmares, to examine
it as a society and government, and to suggest its significance to both
Brazilian and African history.'8
I
Early writers attributed the birth of Palmares to Portuguese-Dutch
struggles for Pernambuco, from which slaves profited by escaping in
groups.'9 They made no reference to Palmares as a quilombo. Southey
came across the term in a Minas Gerais decree of I722.20 An official letter,
sent from Pernambuco to Lisbon in i692,21 contains the first and only
definition of Palmares as a quilombo in primary sources. The point is
worth stressing. The accepted definition of a quilomboas a fugitive slave
settlement has been continuously applied to Palmares since the turn of this
century, and the problem of interpretation has been more difficult as a
result. An early nineteenth-century historian, for example, could easily
classify Palmares as the 'unusual exception, a real government of escaped
Blacks on Brazilian soil'.22 But subsequent identification of the state
'The Palmares " Republic " of Pemambuco: Its Final Destruction', The Americas
(1Io49). V. 20o-10.
15 Emesto Ennes, As Guerras nos Palmares (1938), 113.
16 Engel Sluiter in preface to the 'Report on the State of Brazil, i6i2', Hispanic American
Historical Review (i949), XXIX/4, 52I; full Portuguese text reproduced, 52i-62. It is
generally accepted the Diogo de Campos Moreno, aide to governor Diogo de Meneses
(I608-I2), penned the 'Rezao do Estado do Brasil'.
17 Revistado InstitutoHistorico e Geographico
Brasileiro(I84I), III, I5I-4; (I85I),XIV,491.
18
First-hand descriptions (i645), 'Diario da Viagem do Capitao Joao Blaer aos Pal-
mares', actually written by Jiirgens Reijmbach (extract from the Brieven en Papieren uit
Brasilien, translated into Portuguese by Alfredo de Carvalho for the Revista do Instituto
ArqueolodgicoPernambucano (I902), x, 87-96; (i675-8) 'Relacao das Guerras feitas aos
Palmares de Pernambuco no tempo do Governador d. Pedro de Almeida', R.I.H.G.B.
(i859), XXII, 303-29. Material from the Municipality of Alagoas: (i668-80) 22 'Docu-
mentos' from the Segundo Livro de Veracoes (n.d.). Material from the Arquivo Historico
Colonial in Lisbon: (i67 1-97) 95 'Documentos' collected and published by Ernesto
Ennes (1938), 133-484. First-hand descriptions and materials from Municipality of
Alagoas have been printed together in the document appendix to Edison Carneiro's
0 Quilombo dos Palmares, 2nd edition (1958), 20I-68, and will be quoted from that
volume. Material of dubious value: 'Memoria dos feitos que se deram durante os primeiros
anos da Guerra com os Negros quilombolas, dos Palmares', R.I.H.G.B. (i876), xxxix,
293-322. This would-be 'Memoria dos feitos' has been discredited as a doctored copy of
the 'Relacao das Guerras', published in the same Review seventeen years earlier.
19 Cf. Pitta, i88o, Book VIII, 235; Southey (i8i9), III, 23; Domincos Do Loreto
Couto, Desaggravos do Brasil e glorias do Pernambuco (i757), Book VIII, chap. 4; and
F. A. de Vamhagen, Historia geral do Brasil, 1930 ed., III, 31 9.
20 History (I 8 I 9), III, 248 and 248n.
21
Montebello to Overseas Council, 5 Sept. i692, in Ennes (1938), 243.
22 Ferdinand
Denis, Resume de l'Histoire du Bresil (i825), I47.
i64 R. K. KENT
which was a major historical event with a mere colony of escaped slaves
could not provide a framework to fit the problem. Ki-lombo, according to
Cavazzi,23was a Jaga war camp, and there is no lack of sources which have
translated it correctly as 'arrayal '.24 Could a historico-linguistic link
between a Palmares in a formative stage and the Jaga ki-lombo perhaps be
assumed, faute de mieux?
Slaves who freed themselves by escaping into the bush became some-
thing of a problem several decades before the Dutch took Pernambuco. In
I597, a Jesuit Father, Pero Rodrigues, was able to write that the 'foremost
enemies of the colonizer are revolted Negroes from Guine in some moun-
tain areas, from where they raid and give much trouble, and the time may
come when they will dare to attack and destroy farms as their relatives do
on the island of Sao Thome'.25 Shortly after his arrival from Portugal,
governor Diogo Botelho (i 602-o8) learned from an Amerindian chief
named Zorobabe that there was a 'mocambo . . . of Negroes from Guine
. .. in the palmares of river Itapicuru'.26 Zorobabe was asked to destroy
the mocamboand return with slaves, but 'few were brought back since the
Indians killed many and Zorobabe sold some along the way'.27 If the
Itapicuru mocambowent almost unnoticed by Portuguese authorities, this
was not the case with a similar manifestation farther north. In the Cap-
taincy of Pernambuco, reported a high official in i6I2, 'some 30 leagues
inland, there is a site between mountains called Palmares which harbours
runaway slaves . .. whose attacks and raids force the whites into armed
pursuits which amount to little for they return to raid again. . . . This
makes it impossible to . . . end the transgressions which gave Palmares its
reputation.'28 Diogo Botelho, before he left Brazil, sent a punitive expedi-
tion to Palmares.29
Clearly, quilombodoes not appear in the vocabulary of early seventeenth-
century Brazil. Instead, the fugitive slave settlement is known as mocambo,
an appropriate description since mu-kamboin Ambundu means a hideout.30
Around I603, palmares was simply any area covered by palm trees. There
was no connexion between the Itapicuru mocambosouth of Sergipe and
the Palmares of Pernambuco. Palmares was not regarded as an ordinary
mocambo.By i6I2, it had a considerable reputation. It was an organization
23 Giovanni Antonio Cavazzi da Montecuccolo, Istorica descrittione de' tre regni Congo,

Matamba e Angola (I687), 207; (I690), I63 and I65 (Chi-Lambo Plan).
24 B. J. de Souza, Dicionario da Terra e da Gente do Brasil (I939), 267; H. Capello and

R. Ivens, De Benguella a Terras de Iacca (i88i), I30.

2 Quoted by Rend Ribeiro, 'Relations of the Negro with Christianity in Portuguese

America', The Americas (I958), XIV/4, 458.

26 Frei Vicente do Salvador, Historia do Brasil, 1560-1627 (I96I ed.), 3I5.
27 Do Salvador (I96I), 3I5.
28 'Report on the State of Brazil', H.A.H.R., XXIX/4,553; Carneiro (I958), 90; Carneiro

worked with an earlier copy.

29 'Correspondencia do Diogo Botelho, Governador do Estado do Brazil (I602-I628)',
R.I.H.G.B. (I9I0), LXXIII/I, 86 and I5I.
30 A. M. Perdigao Malheiro, A Escraviddo no Brasil (I867), part III, 2I; Renato
Mendonca, A Influencia Africana no Portugues do Brasil (1935), 220.
 PALMARES: AN AFRICAN STATE IN BRAZIL i65
with which the moradorescould not cope alone. The foundation of Palmares
thus appears to have taken place in I605 /o6, possibly earlier, but certainly
not later. As the report of i6I2 indicates, the first Portuguese expedition
against Palmares attained little by way of military victory. Nothing else,
however, is heard of Palmares until the mid-i63os. Do Salvador's history
of Brazil, written in I629, and recently published official documents for
the years I607-33 are equally silent on Palmares.31 In i634, a Pernam-
bucan morador described Palmares as a 'great calamity'.32 The Dutch
viewed it as a 'serious danger' in i640.33 Increasing palmarista militancy
after I630 can safely be associated with slaves who took advantage of the
Dutch presence to escape and who eventually found their way into
Palmares. It is also certain that Palmares antedates the Dutch in Brazil
by at least a quarter of a century. Given an earlier origin, and the absence
of quilombo from the contemporary vocabulary, it is even less probable
that Jagas were the founders of Palmares. It would be tempting to accept
a recent claim that the Jagas gave Palmares its ruling dynasty, after being
sent to Brazil in i6i6 by the Angolan governor, Luis Mendes de Vascon-
cellos,34 who assumed office in August i6I7 and fought against the
Ngola with Jaga auxiliaries.35 A large contingent of Jagas was sold into
slavery after a punitive expedition against Kasanje in i62436 and may have
reached Brazil along with other prisoners from the guerra preta. But the
account of Andrew Battell,37 who was with the Jagas until I603, shows
nothing to indicate that any of them could have landed in Pernambuco by
I605. There remains the alternative of 'Negroes from Guine'.
'Negroes from Guine' were mentioned long before I597 in connexion
with attempted rebellion.38 Rocha Pitta, a contemporary of Palmares,
held that it was founded by 'forty Negroes from Guine' who had aban-
doned plantations around Porto Calvo.9 But the 'Guine of early Portu-
guese sources is not a fruitful geographical expression. It stood for nearly
anything between a limited section of West Africa and the entire continent.
'Slaves from Guine', according to the i6I2 report, 'are bought dearly
because of the gifts and duties which must be paid for them in Angola '.40
Henrique Dias wrote a letter-in i648 most likely-which stated that the
Black Regiment was composed mainly of 'Angolas' and crioulos with a
3' Biblioteca Nacional, Livro Primeiro do Governo do Brasil, z607-1633 (1958).
32
Francisco de Brito Freire, Nova Lusitania, Historia da Guerra Brasilica, I675, Book
VII, 525n.; also R.I.H.G.B. (I84I), III, I53n.
" 'A Situacao do Negro sob o Dominio Hollandes', in Novos Estudos Afro-Brasileiros
( I937), II, 2 I7.
34 Jan Vansina, 'The Foundation of the Kingdom of Kasanje', YAH, IV/3, I963, 370.
35 Mario Martins de Freitas, Reino Negro de Palmares (1954), I, 266 and 278.
38 See E. G. Ravenstein's 'A Sketch of the History of Angola', published together with
The Strange Adventures of Andrew Battell of Leigh (I9OI), I67-8.
37 Strange Adventures, I-87.
38 Da Nobrega to Tome de Sousa (5 July, 1559), lines 83I-55, in Cartasdos Primeiros
Jesuitas do Brasil 954) III, ioi, edited by Serafin Leite.
38 Historia, Book VIII (i88o), 235.
40 'Report', H.A.H.R. (I949), 523.
 i66 R. K. KENT
sprinkling of 'Minas' and 'Ardras '.4l With Loanda as the undisputed slave
funnel from the I580S until well into the seventeenth century,42 it is quite
unlikely that more than a handful of palmaristas originated outside the
Angola-Congo perimeter. Crioulos-in Pernambuco of I605-could not
have been numerous either. All of this leads to the only plausible hypo-
thesis about the founders of Palmares. They must have been Bantu-
speaking and could not have belonged exclusively to any sub-group.
Palmares was a reaction to a slave-holding society entirely out of step with
forms of bondage familiar to Africa. As such, it had to cut across ethnic
lines and draw upon all those who managed to escape from various planta-
tions and at different times. The Palmares which emerged out of this
amalgam may be glimpsed in a little more detail during the second half
of the seventeenth century.

II
Dutch activities concerning Palmares, from I640 until the Reijmbach
expedition of I645, are known mainly through Barleus43 and Nieuhof.44
They begin with a reconnaissance mission by Bartholomeus Lintz, a
Dutch scout who brought back the first rudimentary information about
Palmares. Lintz discovered that Palmares was not a single enclave, but a
combination of many kleine and two groote units. The smaller ones were
clustered on the left bank of the Gurungumba, six leagues from its con-
fluence with the larger Paraiba and twenty leagues from Alagoas. They
contained 'about 6,ooo Negroes living in numerous huts '.4 The two large
palamars were deeper inland, thirty leagues from Santo Amaro, in the
mountain region of Barriga, and 'harboured some 5,ooo Negroes '.A In
January I643, the West India Company sent its Amerindian interpreter,
Roelox Baro, with a force of Tapuyas and several Dutch regulars to 'put
the large Palmares through "fire and sword," devastate and plunder the
small Palmares'.47 Baro seems to have returned without his men to
report that 'ioo Negroes of Palmares were killed as against one killed and
four wounded Dutchmen, our force having captured 3 I defenders, includ-
ing 7 Indians and some mulatto children'.48 The four Dutchmen and a
41 Freyre (I956), 30I; Mauro (I960), I53. Affonso de E. Taunay dates the letter to
I648 in vol. VIII (1946) of his Historia Geral das Bandeiras Paulistas, with pp. 37-I69
devoted to Palmares.
42 See Dieudonne Rinchon, La Traite et l'Esclavage des Congolais par les Europeens
(1929), 59-80; and Luiz Vianna Filho, 0 Negro na Bahia (1946), for appropriate tables.
43 Caspar Barleus, Nederlandsch Brazil onder het bewind van Graaf J'ohan Maurtiz
(1923), translated from Latin by S. P. L'Honore Naber. I am indebted to Mr. Bruce
Fetter of the University of Wisconsin for his translation of the Dutch text pertinent to
Palmares.
44 Johan Nieuhof, 'Voyage and Travels into Brazil and the East Indies', in A. & J.
Churchill (eds.), Collection of Voyages and Travels (1704), II, I-369. Taunay (I946), VIII,
55, implies heavy borrowing from Barleus' (I647) Rerum per octenium in Brasilia.
45, 46 Barleus (1I923), 315.
47, 48 Barleus (1923), 370. Baro left a subsequent Relation du Voyage... au Pays des
Tapuies (3 April-14 July I647), translated from the Dutch by P. Moreau (Paris, I65I).
 PALMARES: AN AFRICAN STATE IN BRAZIL I67
handful of Tapuyas were found two months later. There was no one with
them.
A second Dutch expedition left Selgado for Palmares on 26 February
I645. It was headed by Jiirgens Reijmbach, an army lieutenant who kept
a diary for thirty-six consecutive days. His task was to destroy the two
groote Palmares. On i8 March Reijmbach reached the first and found that
it had been abandoned months earlier. 'When we arrived the bush growth
was so thick that it took much doing to cut a path through.'49 Three days
later, his men located the second one. 'Our Brasilenses managed to kill
two or three Negroes in the bush but most of the people had vanished.'50
Their king-the few captives told Reijmbach-' knew of the expedition for
some time because he had been forewarned from Alagoas'.51 This Pal-
mares, reads the entry of 2I March,52

is equally half a mile long, its street six feet wide and running along a large
swamp, tall trees alongside.... There are 220 casas, amid them a church, four
smithies and a huge casa de conselho;all kinds of artifactsare to be seen....
(The) king rules ... with ironjustice, without permittinganyfeticeirosamongthe
inhabitants;when some Negroes attemptto flee, he sends crioulosafterthem and
once retakentheir death is swift and of the kind to instill fear, especiallyamong
the Angolan Negroes; the king also has anothercasa, some two miles away, with
its own rich fields.... We asked the Negroes how many of them live (here)
and were told some 500, and from what we saw aroundus as well we presumed
that there were 1,500 inhabitantsall told.... This is the Palmaresgrandesof
which so much is heard in Brazil, with its well-kept lands, all kinds of cereals,
beautifully irrigatedwith streamlets.

In military terms, Reijmbach fared no better than his two predecessors,

Bartolomeu Bezzerra and Roelox Baro. An undestroyed Palmares, of which
'so much is heard in Brazil', remained free of further interference by
Pernambucan authorities until I672. The ensuing two decades can best be
described as a period of sustained war which ended in the complete destruc-
tion of Palmares in i694. As is often the case, warfare and more intimate
knowledge of the enemy went together, and the growing information about
Palmares in the i670s threw light on its evolution during the twenty-seven
years of relative peace.
'Our campaigns', complained a group of Pernambucan moradores in
I68I, 'have not had the slightest effect on the Negroes of Palmares ...
who seem invincible.'53 The claim was not altogether true. Of the eight
expeditions between I672 and i68o, two did hurt Palmares. They were led
by capitdo-mor Fernao Carrilho, who had distinguished 'himself in
the destruction of mocambos in the Captaincy of Sergipe del Rey.'54
The Carrilho entradas of i676-7 produced the most extensive first-hand
49-52 Diario, Caneiro (0958), 255-8.
5 Val de Reis et al. to Overseas Council (6 May i68i), Ennes (1938), I36.
54 Val de Reis et al. to Overseas Council (6 May 068i), Ennes (1938), I35.
i68 R. K. KENT
report ever found. The Palmares of I677 encompassed over sixty
leagues :55
In the northeast,mocamboof Zambi,located i6 leaguesfrom Porto Calvo; north
of it, at 5 leagues' distance, mocambo of Arotirene;along it two others called
Tabocas;northeast of these, at I4 leagues, the one of Dombabanga;8 leagues
north another, called Subupuira;another 6 leagues north, the royal enclave of
Macoco;west of it, at 5 leagues, the mocamboof Osenga;at 9 leagues from our
Serinhaem,northwest,the enclaveof Amaro;at 25 leaguesfrom Alagoas,north-
west, the palamarof Andalaquituche,brother of Zambi;and between all these,
which arethe largestand most fortified,there are othersof lesser importanceand
with less people in them.
There was no doubt, went the report, that Palmares maintained its 'real
strength' by providing 'food as well as security' for the inhabitants-
largely tillers of land who planted 'every kind of vegetables' and knew how
to store them against 'wartime and winter'. All the inhabitants of Pal-
mares considered themselves :56
subjects of a king who is called Ganga-Zumba,which means Great Lord, and he
is recognizedas such both by those born in Palmaresand by those who join them
from outside; he has a palatialresidence,casasfor membersof his family, and is
assisted by guardsand officialswho have, by custom, casaswhich approachthose
of royalty. He is treatedwith all respect due a Monarchand all the honours due
a Lord. Those who are in his presence kneel on the ground and strike palm
leaves with their hands as sign of appreciationof His excellence. They address
him as Majesty and obey him with reverence. He lives in the royal enclave,
called Macoco,a name which was begotten from the death of an animal on the
site. This is the capitalof Palmares;it is fortifiedwith parapetsfull of caltrops,
a big danger even when detected. The enclave itself consists of some 1,500
casas. There are keepers of law (and) their office is duplicatedelsewhere. And
althoughthese barbarianshave all but forgottentheir subjugation,they have not
completelylost allegianceto the Church. There is a capela,to which they flock
whenevertime allows, and imagensto which they direct their worship.... One
of the most crafty, whom they venerate as paroco, baptizes and marries them.
Baptismalsare, however, not identicalwith the form determinedby the Church
and the marriageis singularlyclose to laws of nature.... The king has three
(women), a mulata and two crioulas. The first has given him many sons, the
other two none. All the foregoingapplies to the cidadeprincipalof Palmaresand
it is the king who rules it directly; other cidadesare in the charge of potentates
and major chiefs who govern in his name. The second cidadein importanceis
calledSubupuiraand is ruled by king's brother(Gana) Zona.... It has 8oo casas
and occupies a site one squareleague in size, right along the river Cachingi.It is
here that Negroes are trained to fight our assaults(and weaponsareforgedthere).
Nearly three decades of peace had a number of important results in the
internal evolution of Palmares.
r5Relacdo das Guerras, Carneiro (I958), 202.
"6 Relacdo das Guerras, Cameiro (I958), 203-4.
 . . . ... ... rr... : ......
... ............

4 - A~~~~~~~~~r
4~~~~~~~~~~~~~~~~~~4 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
. .. . ......
& '- .....

04 ~ ~....... ~ 4
..

. .. ... ~
For theSouthern Pernambuco
an~~~.........the inset shows artst'.cocep
tion f Pamare .Bot masaefo.Bres(67)adlaen
doubt.that the Potgeeegnoadte.fia tt fPlae
domnaedth.svetenthcetuy.e.amuc
40ei~~~~~~~~~~~~~

1**

g3;-~~~~~~~~~~~~~~~~~~~~~~~
, - - - - - - ! 6 .~~~~~~~~~~~~~~~~~
 PALMARES: AN AFRICAN STATE IN BRAZIL I69

Instead of the two major palmars of I645, there were now ten. There
was a very substantial element in the Macoco of those native to Palmares,
people unfamiliar with engenho slavery. Afro-Brazilians continued to
enjoy preferential status, but the distinction between crioulos and Angolas
does not appear to have been as sharp as it was in i645. There was a
greater degree of religious acculturation. The reference to a population
composed mainly of those born in Palmares and those who joined from
outside suggests that slaves had become less numerous than free com-
moners. According to Pitta, the only slaves in Palmares were those
captured in razzias.57 But they had the option of going out on raids to
secure freedom by returning with a substitute. This is confirmed by
Nieuhof, who wrote that the main 'business' of palmaristas 'is to rob the
Portugueses of their slaves, who remain in slavery among them, until they
have redeemed themselves by stealing another; but such slaves as run over
to them, are as free as the rest.'58
Although slim and often corrupted, the linguistic evidence leads to two
unavoidable conclusions. The king and most of the hierarchy at the head
of individual mocambos were not crioulos. Macoco/Makoko points to
Loango ;59a Tabocas/Taboka to Ambundu ;b Andalaquituche/Ndala
Kafuche to Kisama;c Osenga/Osanga/Hosanga to Kwango ;d Subupuira/
Subusupu hara vura and Zumba to Zande ;e Dombabanga/Ndombe+
banga to a Benguella-Yombe composite! Arotirene appears to be
Amerindian.9 Zambi/Nzambi and Ganga/Nganga, respectively 'divinity'
and 'lord', are too widely used in Central Africa to be traced further.
Given as 'brother', Zona may be an extreme corruption of Mona, an
equally common term. Amaro/Amargo derives from a very bitter kind of
wild-growing tea shrub, chimarrao,h which is close enough to cimarrones,
as marooned slaves were called in the West Indies.60 The principle of
cujus regio ejus religio, slightly bent to accommodate ethnic sub-groups,
cannot be deduced from this evidence. What it does affirm, however, is
that Palmares did not spring from a single social structure. It was, rather,
an African political system which came to govern a plural society and thus
give continuity to what could have been at best a group of scattered
hideouts.
The almost equally long years of peace and war between I645-94 point
to Palmares as a fluctuating 'peril'. While not necessarily unfair to the
merits of a particular event, the Portuguese took it for an article of faith
57 Historia (i88o), Book VIII, 236.
58 Collectionof Voyages(I704), ii, 8.
59 a--f. A. de 0. de Cadornega,Historia Geral das GuerrasAngolanas(1942 ed.), III, 235,
249, I72, i86 and 240-I and Furtado map of I790/I6?-I7? Lat.; L. M. J. Visconde de
Paiva Manso, Historia do Congo-Documentos (I877), I76-7 and 283-5; Jan Vansina,
Personal Communication (30 January I964); C. R. Lagae, La Langue des Azande, III (I 925),
I45-I75 (Zumba=one of the group and consensus; elected ruler; Subusupu hara vura=
forged weapon of war). g. Probably Araticum of the Arapua mountains in Pernambuco.
h. James L. Taylor, A Portuguese-English Dictionary (I958), 40.
60
J. H. Parry and P. M. Sherlock, A Short History of the West Indies (I963), I4 and 39.
170 R. K. KENT
that Palmares was an aggressor state. No written document originat-
ing within Palmares has come to light. It probably does not exist. The
late Arthur Ramos made a search for oral traditions in the 1930s. It
yielded only an annual stage play he was able to attend in the township of
Pilar:61

The sensation of security (in Palmares)diminished after the first attacksof the
colonists. The PalmaresNegroes reacted by increasing their defences . . . to
maintain their little republic, the Negroes were forced to make sorties to the
neighbouring Indian villages and the towns of nearby valleys. This brought
about (more) reprisals.... The play recalls this sequence of events as it persists
in the memory of the people.

However blurred by the passage of time, the play at least allows for
aggression on each side. There is no need to depend, in this case, on
collective memory to look for evidence with which both the specific and
broad nature of the 'peril' can be illustrated.
Pernambucan authorities did not view Palmares from the perspective of
the moradoreswho were in contact with it. They were too far removed
from the general area of Palmares. Reijmbach, for example, had to march
at a fast clip for twenty days to reach it from the coast, which the Pernam-
bucan governors-Dutch or Portuguese-seldom left. The governors did,
however, respond to morador pressure. 'Moradores of this Captaincy,
Your Majesty, are not capable of doing much by themselves in this war....
At all hours they complain to me of tyrannies they must suffer from [the
Negroes of Palmares]. 62 Among the complaints most frequently heard
were loss of field hands and domestic servants, loss of settler lives, kid-
napping and rape of white women. Two of the common grievances do not
stand up too well. Women were a rarity in Palmares and were actively
sought during razzias. But female relatives of the morador did not con-
stitute the main target, and those occasionally taken were returned un-
molested for ransom.63 Checking the 'rape of Sabines' tales, Edison
Carneiro discovered one exception to the ransom rule, reported by a
Pernambucan soldier in i682.64 Equally, close examination of documents
in the Ennes and Camara de Alagoas collections- I 7 in all-failed to
reveal a single substantiated case of a morador killed in palmarista raids.
Settler lives appear to have been lost in the numerous and forever un-
recorded 'little' entradas into Palmares. They were carried out by small
private armies of plantation owners who sought to recapture lost hands or
to acquire new ones without paying for them. Some of the moradoreshad
secret commercial compacts with Palmares, usually exchanging firearms
for gold and silver taken in the razzias.65 Evidence of this is not lacking.
A gubernatorial proclamation of 26 November I670 bitterly denounced
61 Ramos (I939), I09-I6. Pilar, in the district of Alagoas, was formerly Aloea or Cariri.
62
Sotto-Maior to Overseas Council (8 August I685), in Ennes (I938), I42.
64
63 Southey (i8i9), III, 24. 0 Quilombo (1958), 62. 65 Pitta (I88o), 237.
 PALMARES: AN AFRICAN STATE IN BRAZIL I71

'those who possess firearms' and pass them on to palmaristas 'in disregard
of God and local laws '.66 In I687, the state of Pernambuco empowered a
Paulista Colonel-of-Foot to imprison moradores merely suspected of
relations with Palmares, 'irrespective of their station '.67 Town merchants
are also known to have carried on an active trade with Palmares, bartering
utensils for agricultural produce.68 More than that, they 'were most useful
to the Negroes . . . by supplying advance information on expeditions
prepared against them (and) for which the Negroes paid dearly.'69 And
Reijmbach's entry of 2I March I645, makes it clear that this relationship
was an old one.
Loss of plantation slaves, through raids as well as escape, emerges as the
one solid reason behind the morador-palmarista conflict. The price of
slaves is known to have increased considerably by the late i66os. The very
growth of Palmares served to increase its fame among the plantation slaves.
'More and more Negroes from Angola', wrote a governor in I67I, 'have
now for some years fled on their own from the rigor de cativeiro in mills
and plantations of this Captaincy.'70 But this growth was not one-sided.
Salients in the moradorfrontier, which had protruded from the littoral by
the early i640s, contracted between i645-54, a decade of Portuguese-
Dutch struggle for Pernambuco. Contacts with Palmares were thus
minimized until new bulges began to form. In a painstaking study of
territorial expansion in Brazil, Felisbello Freire has shown that this move-
ment away from the coast began in the late I650s from Bahia, Sergipe and
Espirito Santo.71 It was retarded by no more than a few years for southern
Pernambuco. The northern section merely took a little longer. 'The
Negroes', writes Carneiro, 'had good relations with moradores,as long as
the latter kept their slave huts and plantations away from the free lands of
Palmares.'72 But what looked like free lands to the Portuguese were not
regarded in the same light by rulers of Palmares, and neither party under-
stood the problem. There were, to be sure, no 'great frontier' proportions
in the inland movement of the concluding seventeenth-century decades.
According to Basilio Magalhaes, it was an 'expansao pequena', at fifty or
so leagues inland.73 Palmares was, however, well within it. Towards the
end of the seventeenth century, its territorial domain was estimated at
about I,I00 square leagues.74 'Those who live in a state of constant
danger,' reads another proclamation, 'are people in the vicinity of the
mocambosbelonging to Palmares.'75
66 'Bando do Govemador
Fernao de Souza Coutinho acerca de armas proibidas-
Palmares', in Carneiro (1958), 227-8. 67 Condicoes e Capitulos, Ennes (1938), 24.
68 Ramos 64. 69 Freitas (I954),
(I939), 29I.
70 'Carta do Governador
Fernaode Souza Coutinho' (i June I67I), Ennes (I938), 133.
71 F. F. de Oliveira Freire, Historia Territorial do Brazil (I906), 6-io6.
72 0 Quilombo (1958), 76.
73 Expansdo Geographica do Brasil Colonial(I935), I7-I7I.
7 Ennes (I949), 2I2.
75 'Bando do Governador Fernao de Souza Coutinho (4 September I672)', Carneiro
(1958),23I.
172 R. K. KENT
The hard-hitting Carrilho entradas of I676-7 evoked at least one
response familiarto Palmaresbesides warfare. As he had done earlier,
whenever a new governorcame to Pernambuco,Ganga-Zumbasued for
peace. The terms, however,were new and rathersurprising. On i8 June
I678 :76
the junior lieutenantswhom don Pedro (de Almeida)had sent to Palmares
returnedwith three of the king'ssons and I2 more Negroeswho prostrated
themselvesat the feet of don Pedro.... They broughtthe king'srequestfor
fealty,askingforpeacewhichwasdesired,statingthatonlypeacecouldendthe
difficulties
of Palmares,
peacewhichso manygovernorsandleadershadproferred
butneverstuckto; thattheyhavecometo askforhisgoodoffices;thattheyhave
neverdesiredwar;thattheyonlyfoughtto savetheirownlives;thattheywere
beingleft withoutcidades,withoutsupplies,withoutwives. ... The kinghad
sent themto seek peacewith no otherdesiresbut to tradewith moradores, to
havea treaty,to servehis Highnessin whatevercapacity;it is only the liberty
of thosebornin Palmaresthat is now beingsoughtwhilethosewho fled from
our peoplewill be returned;Palmareswill be no moreas long as a site is pro-
videdwheretheywillbe ableto live,at hisgrace.
Three days afterthe embassy'sarrival,the new governor,Aires de Souza
-replacing Almeida-called a council of state. He proposedthat a draft
treaty be sent to Ganga-Zumbaextendingpeace, the requestedliberties,
and the releaseof palmaristawomen who seem to have constitutedby far
the largestgroupof captives. The council agreed,and a sargento-mor who
had served in the Black Regimentand knew how to read and write, was
sent to the Macoco, 'paraque lesse e declarasseao rei e aos mais o tratado
de paz'.77 Ganga-Zumba was confirmed as supreme ruler over his people.
The question of Palmares'territoriallimits was not settled in any precise
way. 'The solemnity which surrounded all these acts,' wrote Nina
Rodrigues, 'gave a real importanceto the Negro State which now the
Colonytreatedas one nationwould another,(for) this was no mere pact of
a strong party concludedwith disorganizedbands of fugitive Negroes.'78
On paper, the treaty seemed conclusive. But, there were peculiarities
in the immediatesituation. A strong detachment,which had been attack-
ing Palmaressince i677 or early i678, was not demobilizedand a group of
Alagoar moradores,led by a spokesmannamed Joao da Fonesca, made
certainthat it would remainthere.79 The ink was hardlydry when Aires
de Souza Castro began to distributesome 192 leagues of land to sixteen
individualswho had taken part in wars againstPalmares,Carrilhoalone
obtaining a twenty-league sesmaria.80By I679, a palmarista 'captain
named Zambi (whose uncle is Gana-Zona) was in revolt (with Joao
76 'Relacao das Guerras', Carneiro (1958), 2I9.
77 'Relacao das Guerras', Carneiro (I958), 22I.
78 Rodrigues (1945), 146.
79 'Joao da Fonesca pede a Camara de Alagoas mantimentos para a tropa estacionada nos
arredores dos Palmares (26 January I68o)', Carneiro (1958), 244-5.
80
'Relacao das Legoas de Terra', Documento t8, Ennes (I938), I53.
 PALMARES: AN AFRICAN STATE IN BRAZIL I73
mulato, Canhonga, Gaspar (and) Amaro, having done the person of
Ganga-Zumba to death'.81 By March I68o, Zambi was being called upon
to surrender, without success.82 The war was on once more.
Reactions to the treaty, on both sides, are revealing. Ganga-Zumba's
peace proposal contained two clauses which could not be fulfilled. To
allow a sovereign, if vassal, state to exist in Pernambuco would have meant
to reverse a 15o-year-old policy of exclusive Portuguese claim to Brazil.
The Almeida-de Souza move was, therefore, a tactical one. It was, as
Ennes stated after careful study, 'an easy way of postponing that question
which already had, without any positive accomplishment, consumed
infinite time'.83 Conversely, to hand over to the Portuguese half or more
of some i5,000-20,000 palmaristas84-a difficult logistical problem in its
own right-would have required the kind of obedience which only a
modern totalitarian state can secure.
The native-newcomer ratio was not identical in every mocambo of
Palmares. The Macoco, at forty-five leagues from Porto Calvo, must have
had a far greater number of the native-born than did the mocambosof
Zumbi, at sixteen leagues from Porto Calvo, and Amaro, at nine leagues
from Serinhaem. Socio-cultural differences, moreover, between crioulos
and recent arrivals from Africa were not sufficiently great to challenge the
unity of Palmares, which stood against the Portuguese economic and
political order. The diplomacy of Ganga-Zumba, an elected ruler, might
have worked had the promise to return those who found refuge in Palmares
been observed. It might have worked if Palmares had been contiguous to
other similar states facing an intrusive minority. Again, it might have
worked if Palmares had been a homogeneous society with hereditary rulers.
None of these conditions were present. In its time and place, Palmares had
only two choices. It could continue to hold its ground as an independent
state or suffer complete extinction. Zambi's palace revolt finally brought
the unyielding palmarista and moradorelements to full agreement.
Six expeditions went into Palmares between I 68o-6. Their total cost
must have been large. In I694, the Overseas Council in Lisbon was
advised that Palmares caused a cumulative loss of not less than i,000,000
cruzados to the 'people of Pernambuco'.85 The estimate appears exag-
gerated unless the 400,000 cruzados contributed directly by the Crown
was included. A single municipality did, however, spend 3,000 cruzados
(I09,800 reis) in the fiscal year I679/80 to cover the running cost of Pal-
mares wars,86 and a tenfold figure for the local and state treasuries would
seem modest for the six years. Casualties aside, the results did not justify
81' Bando do sargento-mor Manuel Lopes chamando a obediencia o capitao Zambi dos
Palmares (26 March i68o)', Cameiro (1958), 247.
82 Ibid. 83 Ennes (1949), 202.
84 'Relacao das Guerras', Cameiro (1958), 2o6.
85
Caetano de Melo e Castro to Overseas Council (i8 February I694), Ennes (1938), I95.
8B 'Quantia despendida pela Camara de Alagoas com a guerra nos Palmares (20 July
i68o', Cameiro (I958), 248.
I74 R. K. KENT
the cost. Palmares stood undefeated at the end of I686. It was apparent
that the state of Pernambuco could not deal with Palmares out of its own
resources. In March I687, the new governor, Sotto-Maior, informed
Lisbon that he had accepted the services of bandeirantesfrom Sao Paulo,
'at small expense to the treasury of Your Majesty'.87 The Paulistas of the
time were Portuguese-Amerindian metis and transfrontiersmen, renowned
in Brazil for special skills in jungle warfare. Their leader, Domingos
Jorge Velho, had written to Sotto-Maior in I685 asking 'for commissions
as commander-in-chief and captains in order to subdue . . . (Palmares).88
Largely because Lisbon could not be convinced that their services would
come cheap, the Paulistas did not reach Pernambuco until i692. In
crossing so great a distance, I92 lives were lost in the backlands of Brazil,
and zoo men deserted the Paulista ranks, unable to face 'hunger, thirst
and agony'.
The story of Palmares' final destruction has been told in great detail.
Two-thirds of the secondary works discuss the Paulistas and the I69os,
some sixty of the ninety-five documents in the Ennes collection refer to
little else, and Ennes has published a useful summary in English.89 The
Paulistas had to fight for two years to reduce Palmares to a single fortified
site. After twenty days of siege by the Paulistas, the state of Pernambuco
had to provide an additional 3,ooo men to keep it going for another
twenty-two days. The breakthrough occurred during the night of 5-6
February I 694. Some 200 palmaristas fell or hurled themselves-the point
has been long debated-'from a rock so high that they were broken to
pieces'. Hand-to-hand combat took another zoo palmarista lives and over
500 'of both sexes and all ages' were captured and sold outside Pernam-
buco. Zambi, taken alive and wounded, was decapitated on 2o November
I695. The head was exhibited in public 'to kill the legend of his im-
mortality'.

III
The service rendered by the destruction of Palmares, wrote one of Brazil's
early Africanists, is beyond discussion. It removed the 'greatest threat to
future evolution of the Brazilian people and civilization-a threat which
this new Haiti, if victorious, would have planted (forever) in the heart of
Brazil'.90 Indeed, Palmares came quite close to altering the subsequent
history of Brazil. Had they not experienced the threat of Palmares in the
seventeenth century, the Portuguese might well have found themselves
hugging the littoral and facing not one, but a number of independent
African states dominating the backlands of eighteenth-century Brazil. In
spite of hundreds of mocamboswhich tried to come together, Palmares was
87Quoted by Ennes (1949), 205.
887 November i685, Ennes (1949), 204; translation here is Ennes' own.
89 The Palmares 'Republic' (1949), 2oo-i6.
90 Rodrigues (I945), 137.
 PALMARES: AN AFRICAN STATE IN BRAZIL 175

never duplicated on Brazilian soil. This is ample testimony of its impact

on the Portuguese settler and official. They organized special units, under
capitaes-do-mato or bush-captains, to hunt for mocambosand nip them in
the bud. And they sought to prevent, at ports of entry, an over-concen-
tration of African slaves from the same ethnic group or ship. This policy
was abandoned in the wake of the Napoleonic wars, and the immediate
repercussion came by way of the nine Bahian revolts after 1807. The well-
established thesis that uninhibited miscegenation and the corporate nature
of the Portuguese society in Brazil produced a successful example of social
engineering must also take into account the historical role of Palmares.
Palmares was a centralized kingdom with an elected ruler. Ganga-
Zumba delegated territorial power and appointed to office. The most
important ones went to his relatives. His nephew, Zambi, was the war
chief. Ganga-Zona, the king's brother, was in charge of the arsenal.
Interregnum problems do not seem to have troubled Palmares, the history
of which spans about five generations of rulers. Zambi's palace revolt did
not displace the ruling family. Assuming that Loanda was the main
embarkation point for Pernambucan slaves, which is confirmed by the
linguistic evidence, the model for Palmares could have come from nowhere
else but central Africa. Can it be pinpointed? Internal attitude toward
slavery, prostrations before the king, site initiation with animal blood, the
placing of the casa de conselho in the 'main square', or the use of a high
rock as part of man-made fortress lead in no particular direction. The
names of mocambochiefs suggest a number of possible candidates. The
most likely answer is that the political system did not derive from a parti-
cular central African model, but from several. Only a far more detailed
study of Palmares through additional sources in the archives of Angola
and Torre do Tombo could refine the answer. None the less, the most
apparent significance of Palmares to African history is that an African
political system could be transferred to a different continent; that it
could come to govern not only individuals from a variety of ethnic groups
in Africa but also those born in Brazil, pitch black or almost white,
latinized or close to Amerindian roots; and that it could endure for almost
a full century against two European powers, Holland and Portugal. And
this is no small tribute to the vitality of traditional African art in governing
men.