OVERDUE FINES ARE 25¢ PER DAY
PER ITEM

Return to 506k drop to remove
this checkout from your record.

.31 K”'

  

   
 
   

 

 

( ., 20 '
“~49ng
' 5.1;; .
o K125 r ,
3’ {ézfl‘;435 :?§ 25 IE) UQflkagéi3ju
' I ,’ 48 KO)
, ~7
5;’ ‘- q
210

THE QUEST FOR AN AFRICAN ELDORADO:
SOFALA, SOUTHERN ZAMBEZIA, AND THE

PORTUGUESE, 1500-1865

BY

Terry Howard Elkiss

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of History

1979

ABSTRACT

THE QUEST FOR AN AFRICAN ELDORADO:
SOFALA, SOUTHERN ZAMBEZIA, AND THE
PORTUGUESE, 1500-1865

BY

Terry Howard Elkiss

This dissertation details and analyzes the history
of the Mozambican port of Sofala, in relationship to the
Portuguese and to the peoples of southeast Africa, 1500-
1865. Based upon European and African archival sources,
interviews, published documents, and secondary works, the
study emphasizes Luso-African perceptions and relations.
The conclusions reached suggest that the wealth of Sofala
was a gross exaggeration, and the settlement was preserved
by Lisbon only because of its legendary and symbolic image.
Significantly though, the myth of Sofala initially stimu-
lated Portuguese trade and conquest throughout Southern
Zambezia. The various Shona states, however, never fell
under the firm grip of Lusitanian authority and, in fact,
often utilized the Portuguese for their own political ends.

As early as the tenth century, Sofala acquired a
fabulous reputation for its gold, thanks to imaginative

Muslim writers. Some of the tales reached the court at

Terry Howard Elkiss

Lisbon and inspired royal efforts at overseas expansion.
The Portuguese, however, erroneously believed that Sofala
was a single entrepOt rather than an entire coastal region.
As a result, in 1505, Lisbon established her presence at a
port which she misjudged to be the one and only Sofala.
The settlement soon proved to be of little commercial value.
The Portuguese subsequently sought aggrandizement in the
unknown interior where they ignorantly and impetuously
became entangled in local politics. Luso-Shona compromise
and conflict were features of the region's history until
the arrival of Nguni peoples in the early nineteenth
century. Their conquests altered the complexion of south-
east Africa, and led to the abandonment of Sofala in 1865.
For 350 years of turmoil and disappointment, Lisbon
retained the unremunerative Sofala and glorified its
significance. The settlement had become a Portuguese
symbol of an earlier halcyon age and the hope for a more
radiant future. The legendary reputation of "Golden Sofala"
remained intact, although the port itself was blighted.
The present study seeks to dispel the myth and to provide
an insight into Sofala's wider historical significance for

Portugal and Southern Zambezia.

TO HARRY AND BELLE ELKISS

ii

ACKNOWLEDGMENTS

During the several years required to research and
write this dissertation I have benefited from the kind and
wise counsel of many individuals. Although it is impossible
to express my gratitude to all those who contributed, I
wish to acknowledge a few for their advice, criticism, and
support. First, I must thank Professors Harold Marcus,
Eleanor Huzar, Larry Robbins, David Robinson, and Harry Reed
for serving on my doctoral guidance committee. A special
debt is owed to my chairman, Harold Marcus, for his pene-
trating insights on my topic and his long enduring patience.

Many other scholars have also generously offered
me their time and suggestions including: Edward Alpers, Eric
Axelson, D. N. Beach, E. E. Burke, B. A. Datoo, R. W.
Dickinson, T. N. Huffman, Anténio da Silva Rego, and the
late James Hooker. I express my appreciation to all of them.

Last, I offer my sincerest thanks to my friends and
family. To my colleagues at the African Media Center, Diane
Pflugrad and Robert Cancel, I am grateful for their good
will and unselfish help. I am also deeply indebted to two
unequaled friends, James P. Johnson and Charles McClellan,

for their good natured encouragement during my ”darkest days."

iii

My most profound gratitude, however, is expressed to my

parents, Harry and Belle Elkiss, for their unfailing con-

fidence and enthusiastic interest in my work. I respectfully

offer them my most heartfelt thanks.

iv

TABLE OF CONTENTS

LIST OF FIGURES O I O O O O O O O O O O O O 0

LIST OF ABBREVIATIONS . . . . . . . . . . . .

INTRODUCTION 0 O O O O O O O O O O O O O O 0

Chapter

I.

II.

III.

IV.

VI.

THE ARABIC IMAGE OF SOFALA . . . . . .

Bilad as-Sufala . . . . . . . . . .
Mambone: A Pre-Portuguese Sofala . .

PRELUDE TO PORTUGUESE SOFALA . . . . .

TERRA INCOGNITA: SOFALA AND THE STATES
OF SOUTHERN ZAMBEZIA . . . . . . . . .

Fortress and Factory: Sofala's Early
Years, 1505-1530 a o o o o o o o o
The Shona and Early State Building .

The Rise of the Mutapa Empire: Another

Look at the Evidence . . . . . . .

- States of the Sofala Hinterland . .
A Poor and Miserable Business: Gold
Production in Southern Zambezia .

MERCHANTS, MISSIONARIES, AND MUSKETEERS

IN THE INTERIOR, 1531-C. 1600 . . . .

IN PURSUIT OF THE GOLDEN FLEECE,
1600-1800 c o o o o o o o o o o o o o

Penury and Profit in Seventeenth
Century Southeast Africa . . . . .

Misfit of Empire: Eighteenth Century
Southern Zambezia . . . . . . . .

REQUIEM FOR GOLDEN SOFALA . . . . . .

Page
vii

viii

34

50

50
59

63
73

81

88

112

112
134

150

Page
VII. SOFALA RETROSPECTIVE: FACTS AND FANTASIES . . 170

LIST OF REFERENCES 0 O O C I O O O O O C O O O O O O 180

vi

LIST OF FIGURES

Figure Page
1. Modern Mozambique . . . . . . . . . . . . . . . 4
2. The East African Coast . . . . . . . . . . . . 8
3. Southern Zambezia and Southeast Africa . . . . 15
4. States of the Sofala Hinterland . . . . . . . . 75

5. The Gaza Empire . . . . . . . . . . . . . . . . 155

vii

A.H.M.
A.H.U.
Ajuda
A.N.T.T.
A29
B.N.M.
8&4
_Bs_GL

Documentos

 

DRI
FOUCP.

IJAHS

JAH
JRGS

JSAIMM

POROO.

R.G.S.

LIST OF ABBREVIATIONS

Arquivo Histérico de Mocambique (Maputo)

Arquivo Histérico Ultramarino (Lisbon)

Bilbioteca da Ajuda (Lisbon)

Arquivo Nacional da Torre do Tombo (Lisbon)

Archivo Portugués Oriental, Nova edicao

Biblioteca Nacional de Madrid

Boletim da Sociedade de Estudos de Mocambique

Boletim da Sociedade de Geographia de Lisboa

Documentos sobre os Portugueses em Mogambigue
e na Africa Central/Documents on the
Portuguese in Mozambique and Central
Africa

Documentos Remetidos da India

Filmoteca Ultramarina Portuguesa (Lisbon)

African Historical Studies/International
Journal of African Historical Studies

Journal of African History
Journal of the Royal Geographical Society

Journal of the South African Institute of
Mining and Metallurgy

Livros das Moncoes
Public Record Office (London)

Royal Geographical Society (London)

viii

RSEA Records of South Eastern Africa

 

SAAB South African Archaeological Bulletin

 

ix

INTRODUCTI ON

In 1910, a disgruntled visitor to the Portuguese
colony of Mozambique asserted, "the early history of this
strange section of East Africa should not be, even if it
could be, written . . . the endless tale of persistent,
widespread and continuous butchery would not be good to
read."1 Although this bias has been challenged by a handful
of recent monographs, as yet no comprehensive history of

Mozambique exists.2

The void is partly the result of both
scholarly and political difficulties faced by the potential
researcher. Archives in Lisbon, for example, have been a
Kafkaesque nightmare to the historian. The numerous repos-
itories are filled with crumbling documents which remain
often uncatalogued or filed under a whimsical system not
even understood by the various administrators. Fieldwork
in Mozambique itself has been hampered by the difficulty in

reaching any sites beyond the larger towns and the obstinacy

of the former Portuguese regime in granting visas. Currently,

 

1O. W. Barrett, "The Portuguese Colony of Mozambique,"
National Geographic, XXI (1910), 807.

 

2M. D. D. Newitt, "Towards a History of Modern Moqam-
bique,” Rhodesian History, V (1974), 33.

 

military activity, supported by the independent government
at Maputo, has precluded research in certain regions.

Despite these obstacles, historical reconstruction
is being undertaken. This dissertation attempts to contri-
bute to the process by examining the history of Portugal's
first Mozambican settlement, Sofala, and, more generally,
by evaluating Lisbon's efforts at imperial expansion in
Southern Zambezia3 before the "scramble."

Only for the well documented sixteenth century has

4 The limited

Sofala received more than passing attention.
nature of these studies has left a series of questions
unanswered. Little has been written on how the Portuguese
came to View Sofala as a fabled city of gold or why they
pertinaciously maintained the settlement in light of its
insolvency. Too often, it has been glibly asserted that

Sofala was the opulent entrepot of the East African coast

which handled vast quantities of bullion.S The evidence

 

3Defined here as the region between the Zambezi and
Limpopo Rivers, stretching from the coast westward to the
present Botswana-Rhodesia (Zimbabwe) border.

4See, for example, Alexandre Lobato, A Expansao
Portuguesa em Moqambique de1498 a 1530, 3 vols. (Lisbon,
1954-1960); R. WI Dickinson, "Sofala and the Rivers of
Cuama: Crusade and Commerce in South-East Africa, 1505-
1595" (unpublished M. A. thesis, University of Cape Town,
1971); Eric Axelson, Portuguese in South-East Africa,

1488-1600 (Johannesburg, 1973).

 

 

 

5Most recently see Bailey Diffie and George Winius,
Foundations of the Portuguese Empire, 1415-1580 (Minneapolis,
1977), 343; Philip Curtin, gt 31., African History (Boston,
1978), 143.

 

presented below suggests a revision of this perspective;
Sofala was less significant to Lisbon for what it produced
than for what it represented. In addition, a case can be
made that subsequent Portuguese interest in southeast
Africa was determined, at least partly, by the myth of
Sofala.

Although this dissertation relies upon a corpus of
Portuguese documentation, attention has been given to the
relevant oral and archaeological data. This material
allows an added insight into the African response to Lisbon's
machinations. Europeans did not freely determine the
course of events anywhere in Africa. In Southern Zambezia,
sophisticated states existed long before Lusitania's
arrival. It would be ludicrous to recount the travails of
the Portuguese without examining the decisive part played
by the local population. Lisbon's plans were thwarted less
by the venality and ineptitude of her agents than by actions

of Shona and, later, Nguni peoples.

Vi lo 03an Porto A . I“
0
Hum lo
. 1
hobo“ a. Noun-b1 no I .
Zu-bo
‘l'oto . Angocho
\\
\
\
Son:
\' ' no I Mano
\
O R. ‘1‘
o
Us For,
Boil-o
3“ . Nova. Soto lo
Chinoono

Sub! R . Ion. lumbono

    

 

lbputo

' Bouruto

I Numbno

 

0 at loo loo

Figaro I. Modern lounbiquo

CHAPTER I

THE ARABIC IMAGE OF SOFALA

Bilad as-Sufala

 

The earliest references to Sofala appear in the
commentaries of medieval Muslim geographers and seafarers.
With rare exception, they regarded Sofala not as a single
port, but as a lengthy and low-lying section of the East

Africal littoral.l

This perilous region, the Bilad as-
Sufala, apparently marked the southern limit of the land of

Zanj and adjoined the territory of the Wakwak.2 According

 

1The Arabic root safala literally means "low-lying"
and in support of this etymology, al-Masudi stated that
"whenever a mountain stretches for some distance below the
sea, it is given . . . the name al-sofala"; Gabriel Ferrand,
”Sofala," in The Encyclopedia of Islam, IV, ed. M. T.
Houtsma, gt al.(Leiden, 1934), 469-470; also see H. N.
Chittick. "Observations on Pre-Portuguese Accounts of the
East African Coast," in Sociétés et Compagnies de Commerce
en Orient et dans l'Ocean Indien, ed. Michel Mollat (Paris,
1970), 131-132; R. W. Dickinson, "Angoche and the Sofala
Shoal," Rhodesiana, 34 (1976), 28-37.

 

 

2The word Zanj has no precise meaning, and though
it has been employed as a racial, cultural, and ethnic term,
its most frequent usage has been in the geographical sense,
to describe a large section of the East African coast.
Wakwak also has no exact definition, but may have been a
region south of Sofala inhabited by Khoisan-speaking peoples.
See John Wansbrough, "Africa and the Arab Geographers," in
Language and History in Africa, ed. David Dalby (London,
1970), 97-99: H. N. Chittick, "The Peopling of the East

to al-Masudi (?-956), Sofala was "the farthest goal of

Muslim travel in Africa."3

Beyond this zone, due to

currents and winds, the sea was thought to be unnavigable.
Those vessels which strayed beyond Sofala reportedly never
returned.4

Although the location of the Bilad as-Sufgla cannot

 

be precisely determined, it was regarded by early Arab and
Persian mariners as a land of considerable wealth. Both
al-Masudi and his near contemporary Buzurg Ibn Shahriyar

agreed that Sofala was a source of gold and ivory.5

BY
their time, Muslim merchants had established a vast commer-
cial network linking the distant shores of the Indian Ocean.

Testimony to this flourishing trade is provided by

the widespread distribution of similar Islamic ceramic

 

African Coast," in East Africa and the Orient, eds. H. N.
Chittick and Robert Rotberg (New York, 1975), 20-25; J.
Spencer Trimingham, "The Arab Geographers and the East
African Coast," in East Africa and the Orient, 116-122;

E. E. Burke, "Some Aspects of Arab Contact with South East
Africa," in Historians in Tropical Africa (Salisbury, 1962),
98; Raymond Mauny, wThe Wakwak and the Indonesian Invasion
in East Africa in 945 A.D.," Studia, 15 (1965), 7-16.

 

 

 

3al-Masudi quoted in S. M. Z. Alavi, Arab Geography
in the Ninth and Tenth Centuries (Aligarh, 1965), 76-77.

 

 

4Edward Sachau, trans. and ed., Alberuni‘s India
(Lahore, 1962), I, 362.

 

5Alavi, Arab Geography, 76-77: al-Masudi, "The Ivory
Trade," in The East African Coast: Select Documents, ed.
G. S. P. Freeman-Grenville (Oxford, 1962), 15-16; Buzurg Ibn
Shahriyar, The Book of the Marvels of India, trans. Marcel
Devic and Peter Quennell (London, 1928), 56, 152-153.

 

 

 

sherds from Zanj to the sub-continent. At Kilwa, for

instance, the yellow and green sgraffiato earthenware un-

 

covered is identical to the pottery found at the Pakistani
ports of Tiz and Dabul.6 Traders, of course, dispensed a
variety of other "exotica" including spices, perfumes, and
cloth which have not been so fortuitously preserved.
Muslim merchants also served as intermediaries for

7

the commerce between Sofala and the orient. Indians were

the primary purchasers of Zanj ivory which they fashioned

8 In

into ornaments and utilized in Hindu ceremonies.
exchange for the elephant tusks, artisans, located mainly
in Cambay, supplied the African coast with a steady stream

9 A plethora of these baubles has

of colored glass beads.
been discovered along the littoral and within the interior.
Besides beads, it is likely that Indian manufacturers pro-
duced various textiles for the long-distance trade.

China, in turn, was also interested in obtaining

Zanj ivory. Al-Masudi noted that fashionable Chinese

 

6H. N. Chittick, "Kilwa: A Preliminary Report,"
Azania, I (1966), 10: idem, Kilwa: An Islamic Trading City
on the East African Coast (Nairobi, 1974), II, 302-304,
pl.II; James Rirkman, "The History of the Coast of East
Africa up to 1700," in Prelude to East African History, ed.
Merrick Posnansky (London, 1966), 113.

 

 

 

 

7Sachau, Alberuni's India, II, 141-142.

 

8al-Masudi, "Ivory Trade," 16: G. S. P. Freeman-
Grenville, "Islam and Christianity in East Africa,"'
African Ecclesiastical Review, II (1960), 197.

 

9See A. J. Arkell, "Cambay and the Bead Trade,"
Antiquity, x (1936), 295-305.

 

 

 

 

Adon
GIN Gun-darn:
iogodiohu

:rovo
mu
“Blind!

onhuoo
INDIAN
Q hmthr 00““

not oo Soloom

Kilwo

  
    

Mozambiquo I.

Quounno “Manoeu-

Sofala
llnbono

 

0 niloo 300
Inhlnhuno

Dolores any

 

Figaro 2. Tho Soot African Cooot

 

officials dared not appear in public unless they were
carried in ivory palanquins. The Arab writer further
obserVed that ivory was burned in temples as a form of

10 The rich finds of early Chinese porcelain

incense.
along the East African coast confirm a commercial rela-
tionship, although it is unlikely, that before the fif-
teenth century, China actually initiated any voyages of
her own.11
In order to exploit the Indian Ocean traffic,

Muslim colonies were established at strategic points in-
cluding the East African seaboard. Although the evidence
is scanty, Islamic settlement along the coast probably

began as early as the ninth and tenth centuries.12 It has

been claimed that during this period the port of Mogadishu

 

10al-Masudi, "Ivory Trade," 16.

11See H. N. Chittick, "The Coast of East Africa,"

in The African Iron Age, ed. Peter Shinnie (Oxford, 1971),
119, 126; Teobaldo Filesi, China and Africa in the Middle
Ages, trans. David Morrison (London, 1972); James Kirkman,
"The Culture of the Kenya Coast in the Later Middle Ages,"
SAAB, XI (1956), 92, 94-95; G. S. P. Freeman-Greenville,
"Chinese Porcelain in Tanganyika," Tanganyika Notes and
Records, 41 (1955), 63-66.

 

 

 

12Chittick, "Peopling of the East African Coast,"
37-38, 41; Randall Pouwels, "Tenth Century Settlement of
the East African Coast: The Case for Qarmatian/Ismacili
Connections," Azania, IX (1974), 65-74. Professor Donald
Abraham has informed me that Persian Muslims founded an
entrepot at or near contemporary Sofala between A.D. 825
and 846. He, however, did not reveal his sources; letter
from D. P. Abraham, 25 August 1975.

10

initiated a thriving trade with Sofala for gold and ivory.13

Unfortunately, there is no material evidence to document

14 In fact, it appears that

Mogadishu's early predominance.
Benadir ports, like Mogadishu, evolved only over centuries
as they were settled by successive waves of Muslim traders,
political refugees, and exiles. The colonists integrated
themselves into the indigenous societies and then began a
slow process of expansion southward which eventually brought
Sofala under commercial domination.15
Although al-Idrisi (1100-1166) did not provide

confirmation of the Sofala—Mogadishu connection, his writings

suggested that by the twelfth century the Bilad as-Sufala

 

was well known to Muslim merchants. As resident intellec-
tual at the Sicilian court of Roger II, al—Idrisi acquired
enough information from commissioned informants, as well
as traders and adventurers, to prepare a geographical

treatise which contained a detailed description of the East

 

13Joéo de Barros, "Da Asia" (1552), in RSEA, ed.
G. M. Theal, VI (London, 1900), 234, 273; Burke, "Arab

Contact," 96.

14H. N. Chittick, "An Archaeological Reconnaissance
of the Southern Somali Coast," Azania, IV (1969), 117. Two
Persian inscriptions discovered date only to the thirteenth
century, see Enrico Cerulli, Somalia, I (Rome, 1957), 2-3, 9.

15See H. N. Chittick, "The 'Shirazi' Colonization of
East Africa," JAH, VI (1965), 275-294; B. G. Martin, "Arab
Migration to East Africa in Medieval Times," IJAHS, VII
(1974), 367-390; Pouwels, "Tenth Century Settlement," 65-74.

11

African coast.16

Al-Idrisi's Sofala was composed of various
centers, including Jantama and Dandama, which he character-
ized as collections of villages. In addition, he apparently
mentioned other regional ports (such as, Sayuna, Bukha, and
Daghouta) but, unfortunately, translations are incomplete
and confused.17 A major problem also remains as to the
precise location of the sites and whether some of them
concentrated on the commerce of particular goods. Export
specialization would explain why al-Idrisi asserted that

the inhabitants of Jantama and Dandama sold neither gold

nor ivory, but only iron.18

Ingots were then exported to
India, where they were reworked into the wggtz steel which
gave fame to the Damascus sword.

Since iron is not found along the East African

littoral, the location of the deposits is as uncertain as

the exact whereabouts of the Bilad as-Sufala. One intri-

 

guing possibility would be the Zambezi region of present

 

l6al-Idrisi, India and the Neighboring Territories,
trans. and ed. 8. M. Ahmad (Leiden, 1960), 12-18.

17Ibid., 23. Ahmad's recent translation, for
example, is deficient in many respects. See also Burke,
"Arab Contact," 100; Marcel Devic, Le Pays des Zendis
(Paris, 1883), 83; M. Guillain, Documents sur 1'Histoire,
la Géorgraphie et le Commergede 1'Afrique Orientale (Paris,
1856), I,A228; A. Matecka, :La Cate Orientale_de 1'Afrique
au Moyen Age d'apres le Kitab Ar-Rawd Al-MiCTar de Al-
Himyari (XVeS)," Folia Orientalia, IV (1962), 332.

 

 

 

 

 

18al-Idrisi, India and the Neighborinngerritories,

23.

12

Rhodesia (Zimbabwe). Writing in the sixteenth century,
the Portuguese friar Joao dos Santos indicated that "in
these lands of Mocaranga, much iron is . . . found and of
such good quality that some Portuguese took it from these
parts to India for the manufacture of guns."19 In 1881, a
scientific expedition exploring the Zambezi located
numerous iron mines and ore deposits in the Sena and Tete
regions.20 There is no reason to believe these resources
could not have been exploited at a date compatible with
the account of al-Idrisi.

Other sites located throughout Southern Zambezia,
and even much farther south, may also have provided possible
iron sources. Particularly notable is the region of Phala-
borwa in the northeastern Transvaal. Investigation has
indicated that extensive mining and smelting of both copper
and iron were practiced here from the end of the first
millennium A.D. Shafts were dug with great accuracy and
depth, though sizeable magnetite pebbles could also be

found on the surface.21 Undoubtedly, there was a flourishing

 

19Joao dos Santos, "Eastern Ethiopia" (1609), in
RSEA, VII (London, 1901), 285.

20Paul Guyot, Voyage au Zambese (Paris, 1895), 322-
324. Recent investigation confirms that Rhodesia's "largest
iron smelting deposits seem to be in the Zambezi valley and
to the north of that river"; Roger Summers, Ancient Mining
in Rhodesia (Salisbury, 1969), 212.

 

 

 

21N. J. van der Merwe and Robert Scully, "The Phala-
borwa Story: Archaeological and Ethnographic Investigation

13

trade from this Transvaal site, since local needs cannot
explain the intensive mining operations. While the nature
of the early commerce is unknown, it should be noted that
in more recent times the BaPhalaborwa people often traded
their goods as far as the coastal towns of Delagoa Bay,
Inhambane, and probably even Portuguese Sofala.22
It therefore seems quite possible that Jantama and
Dandama could have functioned as Sofala's iron entrepots,
with ore supplies arriving from numerous different sources.
To present this hypothesis is not to suggest that other
commodities were insignificant during the twelfth century.
According to several French scholars, al-Idrisi indicated
that gold of excellent quality also existed in the region

of Sofala.23 A more thorough and precise translation of

the original Arabic manuscript is required; the result may

 

of a South African Iron Age Group," World Archaeology, III
(1971), 178-196; Brian Fagan, "The Later Iron Age in South
Africa," in African Societies in Southern Africa, ed.
Leonard Thompson (London, 1969), 54, 66; C. E. More, "Some
Observations on 'Ancient' Mining at Phalaborwa," JSAIMM,
LXXIV (1974), 227-232.

 

 

22T. M. Evers, "Iron Age Trade in the Eastern
Transvaal, South Africa," SAAB, XXIX (1974), 33-37. The
historical link between the Transvaal and Sofala was also
emphasized in my conversations with present residents of
Nova Sofala; interviews with Paulo Goncalves and Mzee
Makandara, 16 May 1974, Nova Sofala, Mozambique.

23See_for example, Devic, Pays des Zendjs, 173:
Ferrand, "Sofala," 470; Guillain, Afrique Orientale, I, 226:
al-Idrisi does not omit reference to the precious mineral
as indicated in R. W. Dickinson's, "Sofala--Gateway to the
Gold of Monomotapa," Rhodesiana, 19 (1968), 36.

 

 

 

14

well be that such items as gold and ivory were associated
with specific localities.

Al-Idrisi's late contemporary, Yakut al-Rumi (1179-
1229), makes no mention of Sofala's iron trade, but clearly
notes its wealth in gold. Unlike al-Idrisi, Yakut was a
Persian trader and probably acquired his knowledge from
personal experience. It is unlikely that he actually visit-
ed Sofala in his travels, although oddly he did refer to it

as "the last known town of the Zeng."24

While this may
indicate that one of the significant Sofalan ports, such as
the gold entrepot, could have appropriated the name of the
entire region, Yakut may simply have been in error. Un-
deniably, later Arabic commentaries persistently referred
to Sofala as a vast land.

Yakut's significance is not his contribution to the
continued geographical confusion surrounding Sofala, but
his brief discussion of how trade was conducted there. He
explained that foreign merchants brought their wares to a
prearranged Sofalan site where they left them and withdrew
until the local people made an offer in gold placed nearby.25

This description of the so-called silent barter is reminis-

cent of similar tales from other African regions and may be

 

24Ferrand, "Sofala," 470.

ZSIbid.

15

 

 

C
|
0
Q
\
‘\
. ' "‘ \
.O ' \o "‘ I
’ . o . \.
v annex: a. 'o' f:
1' ,'
t C .0 \- .\ f.’
. ‘. a
. .o . \‘ U
'I' 3
€
;
|
r’
r!
“.
TOflA '.
Khumi Zimbabwe ,' Sofa in
\ ° . 1'
""2 5 Sohi R . “mm"
' INDIAN OCEAN
,' I
Limpopo R.\
PM lobom '. lnhonhune
O
‘Q
0
I
-lo¢oo Buy

. -’-..

L ..-

 

(F mi loo 250

 

Figure 3. Southern Zomba“: and Southooot Africa

 

16

of dubious historical validity.26

It is possible though
that in Yakut's time the Sofalan coast had not yet been
colonized and trading patterns were still rudimentary. In
fact, the Persian commentator while noting Mogadishu's
commercial significance, does not link it to trade with
Sofala.27

By the thirteenth century, Kilwa Kisiwani, under
an expansionist dynasty, had emerged as the most important
emporium on the East African coast. Part of her dominant
position was related to control over the commerce of Sofala.
Utilizing her more southerly location, Kilwa assumed loose
hegemony over her neighbors. Although it is unlikely that
the carrying trade with Sofala was completely monopolized,
exorbitant tariffs were charged passing vessels. Three
hundred years later, a Portuguese official noted the method
of taxation which Kilwa employed. He pointed out that
traders were required to pay in coin and in merchandise
before departing for Sofala. On their return they were
further assessed on their gold. Kilwa even had an arrange-

ment with Mombasa to catch violators.28 It is difficult to

 

26See P. F. de Moraes Farias, "Silent Trade: Myth
and Historical Evidence," History in Africa, I (1974), 9-24;
Lars Sundstrom, The Exchange Economy of Pre-Colonial Tropi-
cal Africa (London, 1974), 22-30.

 

 

 

27Interview with E. E. Burke, 9 May 1974, Salisbury,
Rhodesia; Chittick, "'Shirzai' Colonization," 275.

28Diogo de Alcacova to King, 20 November 1506, in
Documentos, I (Lisbon, 1962), 397-399.

 

l7

discern how merchants could turn a sufficient profit if
they were fleeced so mercilessly; undoubtedly, many engaged
in contraband trade.

Kilwa's ascendancy over Sofala has led most commen-
tators to believe two related points: Sofala was a specific
town, and it was located at the site of the later Portuguese
village. Neither observation is verified in the Arabic
literature or by the scanty archaeological work. The source
of this mischief, which now pervades all the historiography,
is probably Joao de Barros's tongue-in-cheek tale relating
Kilwa's discovery of Sofala. According to the sixteenth-
century account:

A man was fishing in a canoe outside the bar of Kilwa,
near an island called Miza, and caught a fish on the
hook of the line which he had cast into the sea; and
feeling from the struggles of the fish that it was very
large, in order not to lose it he unmoored his boat and
let the fish go which way it would, and sometimes the
fish dragged the boat and sometimes it was carried by
the currents, which are very strong there, so when the
fisherman wished to return to port he was so far from
it that he could not find it. Finally after suffering
hunger and thirst, he arrived more dead than alive at
the port of Sofala where he found a ship from Mogadaxo
which was trading there, in which he returned to Kilwa
and related what had passed and what he had seen of

the gold trade.29

From this report it would be logical to conclude that
Sofala was a single entity. Barros further noted that once
in the possession of the lucrative trade, the Muslim rulers

of Kilwa always sent governors to Sofala, and that one of

these officials was in charge when the Portuguese reached

 

29Barros, "Da Asia," 273-274.

18

30

the port. Barros's primary source, the Chronicle of the

 

Kings of Kilwa, is no longer extant, although a summary of

 

the work is contained in an anonymous Arabic history of
Kilwa.31 The latter version, however, includes none of
the above material on Sofala. Therefore, interpretation
rests solely upon the comments of a Portuguese historian
who could not read Arabic and has been criticized for
making statements that do not always correspond to the
documentation.32 Yet, modern scholars assume that Barros's
Sofala was synonymous with the Portuguese one. It is more
likely that the Portuguese historian had no reason to
suspect that early Sofala may once have had a broader
identity, and that it had not always existed at the same
site.

This hypothesis, although unverified, has the virtue
of not affronting the existing historical record. While
archaeological investigation of Portuguese Sofala has been
hampered by the sea's encroachment, the work undertaken thus

far indicates an almost absolute lack of pre-fifteenth-

 

3°Ibid., 274.

31See "An Arabic History of Kilwa Kisiwani c. 1520,"
in Select Documents, 34-49.

 

32C. R. Boxer, "Three Historians of Portuguese
Asia," Boletim do Instituto Portugues de Hong Kong, I (1948),
24. I. A Macgregor, "Some Aspects of Portuguese Historical
Writing of the Sixteenth and Seventeenth Centuries on South
East Asia," in Historians of South East Asia, ed. D. G. E.
Hall (London, 1961), 182.

 

 

19

century artifacts and no evidence of early residence
structures.33 Of course, sites can be scoured by tides and
more thorough examination is required. A case, however,
can be made that contemporary Sofala was only the latest in
a series of ports along a stretch of coast, with settlement
taking place only shortly before the Portuguese conquest.34
In addition, Arabic sources dating from Kilwa's
ascendancy, with only one possible exception, continued the
earlier tradition of enumerating the various towns of the

extensive Bilad as-Sufala. The Arab-Spanish geographer Ibn

 

Said (c. 1214-c. 1286) cited Sofala's capital as Sayuna and
its most southerly entrepot as Daghouta.35 He added that

Sayuna's inhabitants were non-Muslims who were ruled by a

36

local king. Since the Moorish writer made no mention of

 

33Interview with R. W. Dickinson, 3 May 1974, Salis-
bury, Rhodesia; R. W. Dickinson, "Archaeological Investiga-
tions at Sofala, Mocambique, 9-18 June 1969" (unpublished
report, Salisbury, 1969); idem, "Archaeological Investiga-
tions at Sofala, Mocambique, 29 May-ll June 1970" (unpub-
lished report, Salisbury, 1970); idem, "The Archaeology of
the Sofala Coast," SAAB, XXX (1975), 90.

34Idem, "Excavations at Sofala," Rhodesia Science
News, IV (1970), 423.

 

35In an attempt to identify these ports it has been
argued that the river upon which Sayuna was located was the
Zambezi, while Daghouta was the ancient site of Inhambane;
see W. G. L. Randles, South East Africa and the Empire of
Monomotapa on Selected Printed Maps of the 16th Century
(Lisbon, 1958), 11-14; Lereno Barradas, "Inhambane de
Outrora," Monumenta, VI (1970), 21-22.

 

 

 

36Ferrand, "Sofala," 470.

20

Kilwa's domination of this port, either his sources were
dated or Sayuna was not an important enough emporium to
require resident agents. Perhaps Kilwa controlled the

transit trade to its south, but maintained officials at

only the Sufalat adh-Dhahab: the Sofala of Gold. Abu al-

 

Fida (1273-1331), the Arab prince who dabbled in geography
and history, revealed that after passing Sofala's capital,
Seruna (Sayuna), one arrived at Leirana. He noted, "Ibn
Fathuma, who visited the town, said that it was a seaport
where ships put in and whence they set out. The inhabitants

profess Islam."37

Possibly, Leirana was Kilwa's gold entre-
pot; the information is simply too inadequate to determine
the answer.

While the Muslim writers continued to point out
other products found at Sofala, it was gold that preoccupied
their interests from the thirteenth century onwards. Ibn
al-Wardi (1290-1349) claimed that "Golden Sofala" possessed
remarkable quantities of the precious metal: "one of the
wonders of the land of Sofala is that there are found under
the soil, nuggets of gold in great numbers."38 Like his
predecessors, he did not suggest any relationship between

this region and Kilwa in his study. Only the knowledgeable

Arab traveler Ibn Battuta (1304-1377) implied such a

 

37Abu al-Fida, "Malindi, Mombasa, and Sofala," in
Select Documents, 23.

 

38Ibn al-Wardi quoted in Ferrand, "Sofala," 471.

21

connection. Ibn Battuta, visited Kilwa in c. 1331 and "was
told by a merchant that the city of Sufala lies at a dis-
tance of half a month's journey from Kfilwa . . . and from
Yufi gold dust is brought to Sufala."39 Yufi was apparently
a site in the interior, perhaps Great Zimbabwe, where gold
was collected and dispatched to the coast.40 The Sofalan
town Ibn Battuta cited must have been within Kilwa's
political grasp. Unfortunately, the Arab adventurer did
not travel further south and provided no unequivocal con-
firmation.41

For Arab geographers, even during the late four-
teenth century, Sofala remained a vague toponym designating
a large region.42 This stretch of coast contained many
entrepots, some of which gained fame for supplying gold,
ivory, and other commodities to distant lands. The fortunes

of these Sofalan collection centers probably varied in

 

_ 39H. A. R. Gibb, trans. and ed., The Travels of Ibn
Battuta, II (Cambridge, 1962), 380.

 

40T. N. Huffman, "The Rise and Fall of Zimbabwe,"
JAH, XIII (1972), 362; also see below Chapter III.

41One frustrated scholar has seriously asked whether
he actually even traveled south of Mogadishu; J. Spencer
Trimingham, "Notes on Arabic Sources of Information on East
Africa" in East Africa and the Orient, 275. See also H. N.
Chittick, "Ibn Battuta and East Africa," Journal de la
§ociété des Africanistes, XXXVIII (1968), 239-241; Burke,
"Arab Contact," 101.

 

 

 

42See, for example, Ibn Khaldun, The Mugaddimah,
trans. Franz Rosenthal (London, 1958), I, 99, 123.

 

22

relation to political and economic developments, as well as
to geological alterations of the coast. While the literary
and archaeological evidence is too fragmentary to identify
the sites with assurance, a strong case can be made for

the existence of one such port just south of the eventual

Portuguese settlement.

Mambone: A Pre-Portuguese Sofala

 

It appears likely that Mambone, a port at the mouth
of the Sabi (Save) River, once served as an emporium for
foreign merchants and a gateway to the interior. Before
the arrival of the Portuguese, Mambone may even have been

the Bilad as-Sufala's principal gold entrepot. This hypo-

 

thesis is based partly upon the recent Portuguese translation

of an ambiguous Arabic nautical poem, the al-Sofaliya by

 

Ahmed Ibn Madjid (c. 1432-1500).43 The author was largely
concerned with providing a seafaring guide to the perilous
Sofalan coast, where "if you commit an error the world will

forget your existence."44 The unique contribution, however,

 

43T. A. Chumovsky, ed., Trés Roteiros Desconhecidos
de Ahmad Ibn-Madjid, Port. Trans. M. Malkiel-Jimounsky
(Lisbon, 1960), 17-56. For criticism of the editing and
translation of the Leningrad Ms. see G. R. Tibbetts, Arab
Navigation in the Indian Ocean before the Cominggf the
Portuguese (London, 1971), 430; Lereno Barradas, "O Sul de
Mocambique no Roteiro de Sofala do Piloto Ahmad Ibn-Madjid,"
Revista da Universidade de Coimbra, XXII (1970), 159; idem,
1?Sobre o Roteiro de Sofala do Piloto Arabe Hamad Ibn-Madjid,"
Studia, 32 (1971), 14-15.

 

 

 

 

 

 

44Chumovsky, Trés Roteiros, 43.

 

23

of the al-Sofaliya was Ibn Madjid's insertion of previously

 

unmentioned toponyms and titles. Specifically, he noted

that to the south of Sofala was Mulbaiuni (or Malanbfini),

 

 

a place inhabited by Muslims, but frequented by "infidels"

ruled by a great and wealthy king called Mung-batfir.45 It

 

is conceivable that this statement was a veiled reference
to trade between Mambone and the rich Southern Zambezian
gold area of Torwa (Butua) controlled by a Shona chief. A
1512 Portuguese report noted that the ruler of Torwa "has
much gold which is mined in his land along the fresh water
rivers, and he is as great as the king of Menomotapa and is
always at war with him."46 Although the exact location of
Torwa is not known, it was probably an extensive geographi-
cal area encompassing the goldfields of southwestern Rho-
desia (Zimbabwe).47

Another intriguing verse in the al-Sofaliya provides

 

some hint of a commercial route from coast to interior.

 

45Ibid., 43- 44; There is no concurrence of opinion
on the identification of Mulbaiuni/Malanbuni, but G. R.
Tibbetts, who has examined the original manuscript, states
that the site is "almost certainly Mambone"; Tibbetts, Arab
Navigation, 431. The question of the Muna- -batur' 3 identity
i§falso controversial. It may be a Shona praise title, see
W. G. L. Randles, "La Fondation de l'Empire du Monomotapa,"
Cahiers d'Etudes Africaines, XIV (1974), 222.

 

 

46Gaspar Veloso to King, c. 1512, in Documentos, III
(Lisbon, 1964), 185.

 

47Peter Garlake identifies Torwa with the region of
Guruhuswa to the west of Great Zimbabwe; P. S. Garlake,
Great Zimbabwe (London, 1973), 16, 171, 177.

 

24

According to Ibn Madjid, south of Sofala, there was a route
to the rich mines which was utilized for more than two
months each year,48 possibly along the meandering Sabi-Lundi
River. In the late nineteenth century a series of resolute
English explorers attempted to venture from the Sabi

estuary inland. Without exception, they notified the Royal
Geographical Society that the river was largely unnavigable
due to its narrow banks, shallow depth, and occasionally
raging currents.49 More recently, the French geographer
Francois Balsan dismissed the Sabi-Lundi waterway as a
possible ancient commercial highway for similar reasons.50
Yet, this negative evidence is open to challenge. A
thorough survey of the Sabi and Lundi Rivers concluded that

for, perhaps, "three or four months" during the year craft

of good size could travel about 250 miles from the coast.51

 

48Chumovsky, Trés Roteiros, 47.

 

49R. G. S., anon. letter, 10 October 1880, in the
Journal Mss. of C. E. Foot and T. L. Phipson-Wybrants, 1881;
R. G. S., A. Vaughan Williams to the secretary of the R. G.
S., 3 October 1892, in the Papers of A. Vaughan Williams,
1892; St. Vincent Erskine, "Journey to Umzila's, South-East
Africa in 1871-1872." JRGS, XLV (1875), 36, 94; idem,
"Third and Fourth Journeys in Gaza, or Southern Mozambique,
1873-1874 and 1874 to 1875," JRGS, XLVII (1878), 28-29, 35-
36.

 

50Francois Balsan, "La Route de l'Or du Matabele-
1and," BSEM, XXXII (1963), 99, 103; idem, "Ancient Gold
Routes of the Monomotapa Kingdom," Geographical Journal,
CXXVI (1970), 240, 244-246.

 

51H. de Laessoe, "The Lundi and Sabi Rivers," Pro-
ceedings of the Rhodesia Scientific Association, VI (1906),
128.

25

Ibn Madjid's gold route, it should be reiterated, had a
similar limited period for travel.

More convincing data for the Sabi-Lundi theory are
provided by geology and archaeology. Sufficient attention
has not been given to tectonic movements that have affected
not only the Mozambique coast, but also the course and
depth of rivers. Ports and islands have sunk into the
ocean, while waterways have become shallow and unnavigable
or even silted up. The rate of this process appears to
have been very rapid. In addition, rainfall patterns have

52 These

also significantly altered over the centuries.
points make it possible to suggest that the Sabi-Lundi net-
work might have provided a more effective link to the
hinterland than is presently the case. Interestingly, a
tributary of this river system, the Mtilikwe, passes near
the ruins of the earliest notable gold collection center,
Great Zimbabwe. It is not known, however, if this stream
was ever utilized for commerce or, indeed, whether it has
undergone various changes.53

While it would be encouraging to be able to delin-

eate a chain of ancient ruins following the river network,

 

52Summers, Ancient Mining in Rhodesia, 206-208.

 

53Balsan, "Route de l'Or," 99; H. von Sicard, "The
Ancient Sabi—Zimbabwe Trade Route," NADA, XL (1963), 12.

26

no such pattern has been discovered.54 Roger Summers has
investigated the region near the confluence of the Sabi and
Lundi which, according to legend, was once the site of a
man-made dock. He has speculated that this anchorage may
have served trading vessels which could not proceed up-

55 There are also oral traditions

stream because of rapids.
which indicate that the Sabi River was regularly utilized
as a trade route with emporia located at its mouth and

56 Corroboration for these tales must

along its course.
await the work of archaeologists.

Although the Portuguese never apparently utilized
the Sabi-Lundi network (which was probably defunct by the
date of their arrival), they were well aware of the exis-

tence of Mambone.57

The European governors regarded the
port as insignificant commercially, especially in comparison

with their stronghold, Sofala. Duarte Barbosa (?- c. 1521),

 

54Interview with T. N. Huffman, 8 May 1974, Salis-
bury, Rhodesia; O. R. De Oliviera, Amuralhados da Cultura
Zimbaue-Monomotapa de Manica e Sofala (Lourenco Marques,
1963), 19.

 

55
206-208.

Summers, Ancient Mining in Rhodesia, 156-157,

 

56Most of the material was collected by J. Blake-
Thompson and exists largely in unpublished manuscripts and
extracts, see J. Blake-Thompson, "The Oriental Ancestry of
the Aulaya-‘Nyungwe," South African Journal of Science,
XXVII (1939), 464-467. His material is quoted in Summers,
Ancient Mining in Rhggesia, 206-208; von Sicard, "Sabi-
Zimbabwe Trade Route," 8-10.

 

 

57The earliest reference I have found to Mambone ~

dates from 1511, see Anténio de Saldanha to Troilos Brandao,
27 April 1511, in Documentos, III, 71.

 

27

a minor official, noted simply that in the Hucicas Pequenas
(Sabi Delta) there were "Moorish" trading settlements.58
Pearls were apparently the only valuable commodity obtained
at Mambone, and they were of questionable quality.59 It
seems that such items as ivory and gold were no longer
shipped there, but to more northern ports such as Sofala,
Angoche, and Quelimane, and, later, to the southern town of
Inhambane. The Portuguese, however, did establish a small
garrison post at Mambone, as indicated on numerous seven-
teenth century maps.60
Even from this brief description, it is clear that
Mambone hardly impressed the Portuguese as anything more
than a typical coastal village. Yet, persistent traditions
of the port's more glorious past appeared in the commentaries
of some of the local clergy. Padre Anténio Gomes, in a
1648 document, indicated that to the south of Sofala were
ruins of the Queen of Sheba's factory. They were (supposed-

ly) located on a river preserving her name, the Rio da

Saba.61 In 1723, the prelate of Mozambique reported to the

 

58M. L. Dames, trans. and ed., The Book of Duarte

Barbosa (c. 1518), I (London, 1918), 5.

 

59Eric Axelson, Portuguese in South-East Africa,
1600-1700 (Johannesburg, 1960), 186.

 

60See Armando Cortesao and A. Teixeira da Mota,
Portugaliae Monumenta Cartographica (Lisbon, 1960), V, pls.
560 B, 578 F.

61Eric Axelson, ed., Viagem que fez o Padre Ant.o
Gomes, da Comp.a de Jesus, ao Imperio de de (sic) Manomotapa;

 

28

learned Academia Real de Histéria de Lisboa that a local

 

tradition described ships at the mouth of the Sabi that
traveled upstream to procure gold.62 The vicar of Sofala,
Friar Francisco de Santa Catarina, was even more specific
in his 1744 account. He noted that near Sofala "in a land
that is called Mambone one can still see the palace where
the Queen of Sheba was, and in the walls of the said
edifice one can discern some letters that cannot be under-
stood because they appear to be Moorish." None of the
Muslims he asked, however, could interpret the inscription

either.63

If any credence can be given to these semi-
1egendary accounts, there is some support for the View that
in a previous period Mambone was a rich entrepot servicing
foreign merchants who voyaged up the Sabi. Recent oral

traditions, in fact, suggest that at one time Mambone was

called Sofala.64

 

e assistencia que fez nas ditas terras de Algfus annos
(Lisbon, 1959), 44.

 

62Frei Manuel de S. Thomas cited in H. von Sicard,
"Pero de Covilha's 'Relaction,‘ 1587," Studia, 29 (1970),
167.

63Frei Francisco de Santa Catarina's original report
is contained in C. R. Boxer's, "A Dominican Account of Zam-
bezia in 1744," BSEM, XXIX (1960), 1-14, note esp. 9-10.
Professor Boxer interprets the curious account of Mambone as
a confused reference to Zimbabwe, see 14 n. 14. I cannot
concur since the document is not geographically muddled and
the author, resident at Sofala, would have been familiar
with Mambone.

64Summers, Ancient Mining in Rhodesia, 205.

 

29

Further confirmation of this theory through archae-
ological work has been hampered largely because of the
difficult terrain. Mambone is situated in a region where

65 Little serious

the coastal shelf is rapidly sinking.
investigation has been conducted until recently. Professor
Balsan, who only briefly visited sites in the Sabi Delta,
indicated that he found no traces of Arab occupation in

the pre-Portuguese period.66 The late Lereno Barradas,
however, carefully scrutinized sites in the Mambone area
and pointed out a number of possible ancient anchorages.

He based his claims, not only upon observation, but also

on examination of old maps and interviews with local resi-
dents. The lack of any observable ruins, he noted, was
probably the result of "strong fluvial corrosion that must

"67 The destructive force of

have destroyed everything.
both the Sabi River and the Indian Ocean, in conjunction
with devastating cyclones and tectonic movement, could have

eliminated any vestiges of an early emporium.

 

65This aspect can be most readily observed at Sofala,
where the ruins of the old Portuguese fort are under water
and only the outline of the ravaged site can be seen at low
tide. When I visited Sofala on 17 May 1974, the remnants
of the fortaleza were visible for only one hour, making
archaeological work almost impossible.

 

66Balsan, "Ancient Gold Routes," 245—246.

~67Lereno Barradas, "A Primitiva Mambone e Suas
Imediacoes," Monumenta, III (1967), 24.

 

30

Recently, R. W. Dickinson began the first excava-
tions at Mambone. He initiated his work to the southwest
of the present village at a site mentioned by tradition
and where pottery previously had been discovered.68
Dickinson found evidence of an old "Swahili" trading
community including, pottery, spindle whorls, and trade
beads. As yet, he has not been able to assign any definite
dates to the material, though nothing appears to precede
the second half of the fifteenth century.69 Further re-
search is required to clarify this picture, but there
appears sufficient reason to believe that Mambone's history
is closely aligned to any consideration of pre-Portuguese
Sofala.

It seems reasonable, at this point, to present a
speculative picture of how early commercial agents may have
utilized Mambone and the Sabi-Lundi network. Beginning no
earlier than A.D. 1000, traders began to penetrate the
Southern Zambezian plateau in order to obtain the gold mined

70

there. "Not until gold mining in Rhodesia was reasonably

 

68Ibid., 27; R. W. Dickinson, "Archaeological In-
vestigations at Nova Mambone, Mocambique," Monumenta, VII
(1971), 24-26; Balsan, "Ancient Gold Routes," 245.

 

69Dickinson, "Investigations at Nova Mambone," 24-
26; interview with R. W. Dickinson, 3 May 1974, Salisbury,
Rhodesia.

70There is no evidence that gold was ever mined in
Rhodesia before this date; T. N. Huffman, "Ancient Mining
and Zimbabwe," JSAIMM, LXXIV (1974), 240; also see below
Chapter III.

31

well organized from within would it have been profitable
for Arabs to establish a permanent settlement in the Sofala
region. Both gold-mining and the coastal trade probably
developed together."71 There are numerous rivers which may
have served itinerant merchants, but the Sabi-Lundi waterway
with its various tributaries reached one of the richest gold
production areas and near the center of a powerful Shona
state, Great Zimbabwe. Navigating from the coast, according
to the season, traders could have departed from their vessels
near the Sabi-Lundi confluence and proceeded by foot. Land
routes alongside the rivers may have been followed where
the waterways themselves were impassable.72

The commerce from the coast may generally have been
conducted by African middlemen and Swahili traders although,
conceivably, foreign craft did venture up the Sabi. Gold,
admittedly, was the major stimulus to this enterprise,
although other important commodities such as copper and

ivory were also sought in exchange for beads, printed cloth,

 

71T. N. Huffman, "Cloth from the Iron Age in Rho-
desia," Arnoldia, V (1971), 16.

72Carl Peters, The Eldorado of theAncients (London,
1902), 348; Balsan, "Ancient Gold Routes," 243-245; Hugh
Tracey, "Vias Comerciais Arabes de Sofala ao Monomotapa antes
do Estabelecimento dos Portugueses," Mocambique, 26 (1941),
33-37; Gervase Mathew, "Some Reflections on African Trade
Routes," Research Review, III (1967), 69-70; Christoph von
Oidtman, TrO Porto de Sofala e o Problema de Zimbaue,"
BSEM, XXVII (1958), 5-19.

 

 

 

32

and ceramics.73 It is also likely, as hinted by Ibn Madjid,
that African people made use of the Sabi-Lundi system to
bring their goods directly to Mambone. Almadyas or log
canoes long have been vehicles of transport, at least on

74 In either case, commercial traffic must

the lower Sabi.
have followed a strict seasonal calendar, not only to maxi-
mize river travel, but also to take full advantage of the
monsoon patterns which determined the movement of cargo
ships along the littoral and across the Indian Ocean.75
As simply a collection center and landing site for
foreign mariners Mambone may never have possessed impressive
buildings but, instead, warehouses and storage facilities.
A small resident community would have sufficed to provide
the necessary amenities and organize the commerce. The lack
of conclusive evidence for Mambone's past may be related to
both its transit function as well as the vagaries of nature.

Wealth percolated from the interior and filtered through

the port to international centers.

 

73Barradas, "Primitiva Mambone," 28. It is relevant
to note that the lower Sabi was navigated to collect ivory
in the Danda region until nineteenth-century invasions dis-
rupted this traffic; Sebastiao Xavier Botelho, Memoria
Estatistica sobre os Dominios Portuguezes na Africa Orien-
tal, I (Lisbon, 1835), 173-174.

 

74
27 n.5.

Dickinson, "Investigations at Nova Mambone,"

75See Raymond Mauny, "Notes sur le Probleme Zimbabwe-
Sofala," Studia, I (1958), 181-182.

33

The future of Mambone as an Indian Ocean entrepot
was foreshortened for natural and, probably, political
reasons. Over the years the course of the Sabi-Lundi River
altered and became impassable over long stretches. It is
impossible to calculate at what date this development may
have occurred. except that it was prior to the Portuguese
arrival. Although other regional centers probably existed,
the focal point of commercial operations was the Sofala of
Gold which apparently shifted northward to the site the
Portuguese claimed in 1505.

This outline is highly conjectural, but provides
an indication of the possible significance of one pre-
Portuguese Sofala. More work must be done not only on the
port of Mambone and along the Sabi-Lundi network, but also
at numerous other sites which possibly were part of the

Bilad as-Sufala. The research will help provide answers

 

to questions concerning the changing political conditions

in the interior as well as the nature of early trade on

the western fringe of the Indian Ocean.76

 

76The recent excavations conducted by Peter Garlake
in the interior of the Inhambane province are a step in
this direction, but the work is too incomplete to draw any
conclusions, see his "An Investigation of Manekweni,
Mozambique," Azania, X (1976), 25-47.

CHAPTER II

PRELUDE TO PORTUGUESE SOFALA

Until the end of the fifteenth century, Europe knew
little about southeast Africa or, for that matter, the
mysterious world of the Indian Ocean. While a handful of
Italian and Jewish merchants long had plied the eastern
trade, their ventures were sporadic because sea routes were
unfamiliar and Muslims dominated the commerce. The Portu-
guese voyages of discovery and conquest altered this picture
by establishing new patterns of trade and political
authority throughout the Indian Ocean complex.

Lisbon's efforts at gaining domination in the orient
were prompted by mixed motivations, both sacred and profane.
The chronicler Gomes Eannes de Azurara (c. 1410-1473)
suggested that exploration was inspired by strategic
interests and the desire for rich profits, as well as
Christian zeal.1 If the hated Muslims could be outflanked
and destroyed in the east, Portugal would secure the lucra-

tive Indian commerce and, moreover, receive pontifical

 

1Gomes Eannes de Azurara, The Chronicle of the
Discovery and Conquest of Guinea (1453), trans. and eds.
C. R. Beazley and Edgar Prestage, I (London, 1896), 27-29.

 

 

34

35

gratitude. In fact, the various Papal Bulls of this period
specifically recognized and authorized the amalgam of
imperial, commercial, and religious objectives.2 According
to King Manuel (1495-1521):
in those parts, so remote from the Roman church, the
faith of His Son our Lord Jesus Christ will not only
be published by our means and received there, by which
we shall win a reward from Him and praise and glory
from men, but also kingdoms and new states and great
riches will be won by force of arms from the hands of
barbarians.

While the Portuguese may have been inspired by such
visions of grandeur, their successes were all the more re-
markable considering the limited resources and manpower
available. These shortages, however, were not detrimental
to overseas expansion. Given Portugal's position on the
Atlantic, it was quite logical for the Iberian nation to
turn to the sea for its basic sustenance. Deficient in
cereals and meat, one of Portugal's dietary staples was
fish. This dependency was precarious, since along the
coast, the narrow continental shelf prevented the develop-
ment of an abundant home fishery. In order to supplement
local sardines, Portuguese fishermen were forced to make
long voyages, thus providing not only food, but also a
contingent of trained seafarers.

It should furthermore be noted that the only avenue

open to Lusitanian ambition was the sea. Inland expansion

 

2C. R. Boxer, The Portuguese Seaborne Empire, 1415-

1825 (New York, 1969), 20-23.

 

3Barros, "Da Asia," 160.

36

was prevented by the superior power of her neighbor, Castile.
The Portuguese, though, had one distinct advantage over the
Spanish; their small state was unified and not embroiled

in the dynastic and foreign difficulties that occupied
Castile until the end of the fifteenth century.4 Portugal
was able to benefit from an effective leadership which
closely supervised the voyages of discovery and carefully
scrutinized all possible opportunities.

Even the hated infidel was utilized as a source of
information on seafaring and distant lands. For several
centruies, the "Moors" had dominated sections of Portugal,
and their defeat in the thirteenth century did not dispel
their influence. Early Portuguese sailing directories,
roteiros, for instance, owed much to Arabic sources.5 After
the conquest of the North African commercial fortress of
Ceuta in 1415, Portugal acquired additional geographical
knowledge from Muslim captives and confiscated maps.6

Equally important as the Arab sources were the re-
ports of Italian merchants residing in Lisbon. Many of

these Venetians and Genoese possessed practical navigational

 

4See J. H. Elliott, ImperialySpain, 1469-1716 (New
York, 1963), 17-51.

 

5C. R. Boxer, "Portuguese Roteiros, 1500-1700,"
Mariner's Mirror, XX (1934), 173.

 

6Edgar Prestage, "The Search for the Sea Route to
India," in Travel and Travellers of the Middle Ages, ed.
A. P. Newton (New York, 1926), 199; Boxer, Portuguese Sea-
borne Empire, 19.

 

 

 

37

experience, while others provided the capital necessary to
finance the Portuguese fleets.7 Along with other Europeans,
the Italians also supplied the latest technology, from ship-
building and textile manufacturing to the important pro-
duction of guns and artillery. Ship mobility and firepower
were the forces that assured the success of Portuguese
imperial expansion. In the case of weaponry when local
industry could not fulfill the demand for arms, Portugal
employed its foreign profits to make purchases from Flemish
and German manufacturers.8
Portuguese mariners were well prepared for their
lengthy voyages, and, in 1488, Bartolomeu Dias and his crew
rounded the Cape of Good Hope. Lisbon, though, did not
dispatch another recorded sea expedition to the east for
nearly a decade. Considering the allure of East Africa and
India, this delay is difficult to understand. Possibly,
there was a conservative reaction among royal advisors

against further exploration. They may have opposed con-

quering new lands not only because such efforts would drain

 

7Charles Verlinden, "Le Influence Italiane nella
Colonizazazione Iberica," Nuova Rivista Storica, XXXV (1952),
254-270; Leo Magnino, "AntOnio de Noli e a Colabora ao entre
os Portugueses e Genoveses nos Descobrimentos Maritimos,"
Studia, 10 (1962), 99-115; Bailey Diffie, Prelude to Empire
(Lincoln, 1960), 49-58; J. H. Parry, The Age of Reconnais-
sance (London, 1963), 48-52.

 

 

 

8Carlo Cipolla, European Culture and Overseas
Expansion (Harmondsworth, 1970), 40.

 

 

38

their modest resources, but also because an oriental empire
would be difficult to defend. The burdensome obligations
would outweigh the possible advantages.9
Some historians, however, prefer to believe that
Portuguese caravels were not idle during these years and,
supposedly, made secret voyages in the South Atlantic and
along the southeast African coast. The expeditions, it is

claimed, remain unknown because of a deliberate policy of

obfuscation, the politica de sigilio, (policy of secrecy).10

 

It seems unlikely that such ventures could remain concealed
from the numerous foreign merchants and sailors present in
Lisbon.

The most probable explanation for the (apparent)
failure of the Portuguese to pursue immediately the sea
route to the orient was the desire by the crown to await the
detailed reports of two agents dispatched overland to Aden

at the same time that Dias was setting sail for Africa.

 

9Barros, "Da Asia," 158; Douglas Wheeler, "Anti-
Imperial Traditions in Portugal: Yesterday and Today,"
Boston University Graduate Journal, XII (1964), 127-130;
Boxer, Portuguese Seaborne Empire, 36.

 

 

10There is nothing to substantiate this hypothesis
which was originated by the late Portuguese historian Jaime
Cortesao. His views are summarized in his Os Portugueses
em Africa (Lisbon, 1968), 123, 127-129; also see Prestage
IiSea Route to India," 212-214; George H. Kimble, "Portuguese
Policy and Its Influence on Fifteenth Century Cartography,”
Geographical Review, XXIII (1933), 653-659. For criticism
of the sigiIio theory see Samuel Eliot Morison, Portuguese
quages to America in the Fifteenth Century (Cambridge,
Mass., 1940), 76-86; Bailey Diffie, I'Foreigners in Portugal
and the Policy of Silence," Terrae Incogpitae, I (1969),
23-34.

 

 

 

 

 

 

39

Pero de Covilha and Afonso de Paiva were commissioned by
King Joao II (1481-1495) to explore the Indian spice trade
and to form an alliance with the legendary Prester John,
thought to reside somewhere in the east, probably Ethiopia.11
Before the crown could formulate a definite strategy,
precise information was required on the extent of Muslim
authority and commerce in the Indian Ocean.

Unfortunately, only the outline of the peregrina-
tions of Covilha and Paiva is known. When the two reached
Aden they parted company; the Arabic speaking Covilha
traveled to India while his companion embarked for Ethiopia.
Paiva apparently died in Cairo before gaining entry to

12

Abyssinia. In contrast, Covilha was successful in re-

connoitering the spice traffic, and eventually followed

the northeast monsoon to East Africa. He even claimed that

"13

he had reached the "coast of Sofala. Like his Muslim

 

llBarros, "Da Asia," 154; Fernao Lopes de Castanheda,
"History of the Discovery and Conquest of India by the
Portuguese" (1551), in RSEA, V (London, 1901), 350-351.

12It had originally been thought that Paiva reached
Ethiopia and died there, but research in the Vatican Ar-
chives has altered this view, see C. F. Beckingham, "Notes
on an Unpublished Manuscript of Francisco Alvares: Verdadera
Infqrmagam das Terras do Preste Joam das Indias," Annales
d' Ethiopia, IV (1961), 150.

 

 

 

13Franciscojklvares, Verdadeira Informagao das
Terras do Preste Joao das Indias (Lisbon, 1540): 129. Padre
Alvares ihterviewed the elderly Covilha Who was resident
in Ethiopia when the Portuguese expedition arrived in 1520.

 

 

40

predecessors, Covilha was probably referring to a long
stretch of the littoral rather than the site of the later
Portuguese settlement.l4

Yet, despite his long absence, Covilha upon arriving
in Cairo found that his travels were not over. The death
of his compatriot meant that he had to set out for Ethiopia.
Before departing, Covilha left an account of his discoveries
for King Joao with two Portuguese Jews. It is unknown,
however, whether this report was ever successfully deliv-
ered.15 It is likely that, at least, some of Covilha's
information was relayed to Lisbon by other travelers and
merchants.

While Joao was collecting such commercial and
cartographic data, he also succeeded in negotiating the
Treaty of Tordesillas with Castile in 1494. The pact
divided the largely unexplored world between the two Iberian
powers. Portugal was now free to pursue an eastern empire
without interference from her neighbor. One of her imme-

diate goals was to capture the gold trade of Sofala, which

 

14See above Chapter I; cf. C. F. Beckingham, "The
Travels of Pero da Covilha and Their Significance," in
Actas, III (Lisbon, 1961), 4.

15Boxer, Portuggese Seaborne Empire, 34. Many
Portuguese historians assume that CoVilha's repprt was
received and utilized, see, for example, Cortesao, Portu-
gueses em Africa, 107; Lobato, Expansao Portuguesa, I, 51;
M. L. de Freitas Ferraz, ed., Documentacab Historica Mogam-
bicana (Lisbon, 1973), 7-8. AIrecent e fort to discovei
the doeument has proved fruitless; von Sicard, "Pero de
Covilha's 'Relaction,' 1587," 155-168.

 

 

 

41

had acquired exaggerated proportions, thanks to Muslim
writers. Yet, it is widely presumed that Sofala was ignored
during da Gama's first voyage to India.
After rounding the Cape in 1497, the Portuguese

captain did avoid much of the Mozambican seaboard, but only:

because the coast curves inward so far . . . that

Vasco da Gama feared the waters would draw him in

and that it was some deep bay from which he would

not be able to get out. This fear made him so

cautious in keeping far off the shore, that he

passed without seeing it, the settlement of Sofala,

so celebrated in those parts.
Da Gama was not only wary of the uncharted coast, but also
was probably convinced that the fabled land was located
farther to the north, as it appeared in Fra Mauro's Venetian
map (c. 1460).17 Apparently, Portuguese geographical infor-
mation was outdated since, as has been suggested above, the
port of Sofala, which Lisbon eventually settled, was only
the latest in a series of such regional entrepéts. Da Gama
seemingly attempted to correct this error by later gleaning
information on the gold emporium from a "Moorish" pilot
captured near Mozambique Island. While India was his ulti-
mate destination, Sofala was not disregarded.

Although da Gama failed to survey the famous south-

east African site, King Manuel, Joao's successor, was not

 

16Barros, "Da Asia," 169. On the coastal configura-
tion see figure l.

17G. R. Crone, "Fra Mauro's Representation of the
Indian Ocean and Eastern Islands," in Studi Colombiani,
III, (Genoa, 1951), 57.

 

42

deterred from prematurely announcing to his Castilian
rivals that the land of gold mines had been discovered.18
This observation was evidently one of his more modest
comments when it came to the subject of Sofala's wealth.
While information on the port remained limited, royal
hyperbole abounded.

It was not until the early sixteenth century that
the first documented Portuguese adventurer sighted Sofala.
In 1500, under the leadership of Pedro Alvares Cabral,
another flotilla was dispatched with the intention of not
only subduing India, but also establishing Portuguese
economic relations along the East African littoral. A
specific objective was to initiate commerce between Lisbon
and Sofala. For this reason, among Cabral's captains was
Bartolomeu Dias, commissioned to raise the first Portuguese
settlement at Sofala. The ill-fated Dias and his crew,
however, drowned on the outward passage and failed to reach

the Indian Ocean. Therefore, although some reports suggest

that Cabral arrived at Sofala before touching India,19 it

 

18King Manuel to Kings of Castile, 1499, in Docu-
mentos, I, 31.

lgBritish Museum, Add. Ms. 20,902, fol. 4 "Relacao
das Naos e Armadas da India," n.d.; "Extractos da Navegacao
de Pedro Alvares Cabral," in RSEA, I (London, 1898), 48.’
Barros only suggests that Cabral reached the "parcel de
Cofala" or the Sofala shoal, a considerable length of the
coast, see Barros, "Da Asia," 49; cf. Theal's faulty trans-
lation of Barros, 198. Manuel Faria e Sousa (1590-1649)
confirms that Cabral reached the "coast of Sofala," but not
the port; see his "Asia Portuguesa" (1666), in RSEA, I, 7.

43

is more likely that any information he acquired on the site
was derived from captured Muslim seamen or through his
visits to other coastal entrepots. According to a member
of Cabral's crew, the captain was informed by a "Moorish"
captive that golden Sofala was ruled by the King of Kilwa.
He further learned that Kilwa "was on the route we were to
take, and that Zaffalle [Sofala] was behind us."20 Again
the Portuguese had missed the port on their passage to India,
providing another indication of how imprecise geographical
information remained.

Only upon his return from the orient in 1501, did
Cabral actively attempt to seek Sofala. From Mozambique
Island, the commander discharged a caravel under Sancho de
Toar to investigate the emporium. The lack of Portuguese
knowledge of Sofala's commerce is evident from the items
which Toar intended to present to the local ruler. Among
the goods were little bells, woolen caps, and even "trap-
pings for hawks."21 None of these trifles indicates any
comprehension of the specific articles necessary to nego-

tiate the sophisticated Sofalan commerce; Sancho de Toar

 

’ 20Anonymous narrative, 1507, in The Voyage of Pedro
Alvares Cabral to Brazil and India, trans. and ed., William
B. Greenlee (London, 1938), 63; also see Damiao de Goes,
"Chronicle of the Most Fortunate King Dom Emanuel of Glori-
ous Memory" (1566), in RSEA, III (London, 1899), 92-93.

 

 

21Gaspar Correa, "Lendas da India" (1858), in RSEA,
II (London, 1898), 26. (Although written in the mid-
sixteenth century, Correa's work was not published for three
centuries.)

44

was ill-prepared to commence serious bartering. Actually,
it is unlikely that the Portuguese agent even set foot at
Sofala. On his arrival, he sighted several Arab vessels
and, following the method of his predecessors, he took a
Muslim hostage while sending his envoy ashore. It was the
usual procedure for Portuguese captains to refrain from
leaving their ships until their secrutiy was guaranteed;

instead, banished convicts, degredados, were carried aboard

 

for such purposes as surveying dangerous or unknown re-

gions.22

In Toar's case, his representative never returned,
and, therefore, no attempt was made to explore the site
further. The report provided by Toar to the king indicated
that Sofala was:

a small island in the mouth of a river, and the gold

which comes here is from a mountain where the mine

is situated, and it is inhabited by Moors and hsgthens

who barter the said gold for other merchandise.

Even though inadequate, the description provided

by this expedition further stimulated the European interest
in the distant Mozambican entrepot. King Manuel boastfully

wrote to Ferdinand and Isabella in 1501 that a treaty had

been arranged with Kilwa over Sofala's commerce. He,

 

22On the neglected subject of the role played by
degredado explorers see T. H. Elkiss, "On Service to the
Crown, Journal of the American Portuguese Society, X
(1976), 44-53.

 

23"Navegacao de Pedro Alvares Cabral," in RSEA, I,
49. Correa's tale of Toar's visit to the Sofalan King seems
to be a total fabrication, note his description in "Lendas
da India," 27-28.

 

45

moreover, attempted to impress upon them the frightful
demeanor of the local Africans, as well as the amount of
gold available, noting that even "the cows of the king
wear collars of heavy gold."24

Manuel was not alone in his exaggerations; Italians
resident in Lisbon related even more grandiose tales. One
Venetian claimed that there was more gold at Sofala than
anywhere else in the world,25 and Americo Vespucci wrote
that Sofala had the reputation of being as great a city as

Cairo.26

These views were still based on imagination,

since little new information actually had been gathered,

and no Portuguese settlement had been constructed at Sofala.
When Vasco da Gama set sail on his second voyage to

India in 1502, the Portuguese had not even established

commercial contact with the fabled port. The captain deter—

mined not only to complete the conquest of the spice traffic,

but also to usurp Sofala's gold trade. These efforts would

strike a devastating blow against the Muslims, Portugal's

only commercial and religious competitors in the Indian

Ocean. In June 1502, the Portuguese explorer, with part of

his fleet, arrived at Sofala in order "to inspect the port

 

24Manuel to Ferdinand and Isabella, c. 1501, in
Pedro Alvares Cabral 51.

 

2éLetter of Giovanni Matteo Cretico, 27 June 1501,
in Pedro Alvares Cabral 156.

 

26Amerigo Vespucci to Lorenzo de Medici, 4 June
1501, in Pedro Alvares Cabral 156.

 

46

and obtain information concerning the inhabitants of the
country."27 Although well received, the Europeans found
their efforts were hampered in bartering for gold. Da Gama
learned that he had arrived eight or nine days after Muslim
ships had taken most of the available supply and that a
struggle between coast and interior was delaying additional
gold from reaching the port.28 Although little was actually
accomplished by this frustrating venture, a new myth was
born. According to Tomé Lopes, one of da Gama crew, cer-
tain Sofalan Muslims possessed information proving that the
emporium once had been King Solomon's Ophir.29 In the Old
Testament, Ophir was the font of gold and spices which the
Israelite ruler sought on his triennial voyages. Lisbon's
officials enthusiastically accepted the Sofala-Ophir
connection and prepared to reap the golden dividends.
Despite the inherent dangers and continual dis-

appointments encountered by Portuguese mariners, Sofala as

 

27Goes, "Most Fortunate King," 99.

28Tome Lopes, "Navegacao da Armada de Estevam da
Gama" (1550), in App, ed. A. B. de Braganca Pereira, I-I-I-
(Bastora; 1936), 230-233. An anonymous Dutch sailor, a
member of da Gama's crew, confirmed this picture and noted
that the local people were reluctant to allow the Portuguese
to discover the actual gold sources; J. Berjeau, trans. and
ed., Calcoen (London, 1874), no pagination.

29Lopes, "Navegacao da Armada," 232-233. Biblical
figures often appear in the Koran, and Muslim traders along
the East African coast adapted these tales to their own en-
vironment, see T. N. Huffman, "Solomon, Sheba and Zimbabwe,"
Rhodesian Prehistory, VII (1975), 21.

47

Ophir became an integral part of the plan for the east.
Gold would be needed to pay for Indian pepper, and the
desire for this glittering treasure prompted da Gama's
subsequent conquest of Kilwa. This city-state not only
was important to Lisbon as a coastal center, but also was
coveted for its control of the (seemingly) lucrative
Sofalan trade. The gold tribute collected at Kilwa and
carried to Lisbon only served to enhance the reputation of
Sofala itself.30

During the following years, other Portuguese armadas
were dispatched to the orient. In 1504, Pero Afonso de
Aguiar may have initiated the first successful trading
expedition to Sofala. Unfortunately, the only extant
source for this event is the unreliable Gaspar Correa
(c. l495-c. 1563) who wrongly assigned the date of 1502 to
the venture.31 According to the chronicler, Aguiar's nego-
tiations provided him with abundant gold, as well as the
promise of loyal friendship. The bargain struck was un-

likely to have been as simple as Correa indicates.

 

30The gold was crafted into the Belem Monstrance, a
classic example of Manueline art, which can still be seen
in the Museu do Arte Antiga in Lisbon; see Robert Smith, 2A3
Art of Portugal, 1500-1800 (New York, 1968), pl. XV.

31Correa, "Lendas da India," 28-30. Aguiar was one
of Lopo Soares's fleet captains during the 1504 expedition;
Goes, "Most Fortunate King," 104. The chronological question
has been thoroughly analyzed by Alexandre Lobato in his Novos
Subsidios para a Histéria daFundagéo de Sofala (Lisbon,
1950), 136-137, and Expansao Portuguesa, I, 72, 204.

 

48

Furthermore, his suggestion that the local Sofalans were
astounded by the sight of writing32 seems unlikely consid-
ering the number of Arab merchants who frequented the port.
To reject, however, the entire Aguiar episode presents "an
unacceptable pause" of three years between da Gama's re-
connaissance and eventual Portuguese settlement.33 Al-
though India was the primary objective, it is difficult to
believe that no effort was made during this period to gain
control of Sofala and its highly valued gold sources. These
mines had been praised by King Manuel as "the richest which
have ever been reported in those parts, and of any others
which are known in the world."34
Yet, only in 1505, under the supreme command of

Francisco de Almeida, was Pero de Anhaia assigned to under-
take the establishment of a fort and factory at Sofala.
The Portuguese were about to learn the truth of Camoes's
prophetic words:

Gold will reduce the strongest fortress, it will turn

friends into traitors and deceivers, constrain the noble

to acts of infamy, make men betray their leaders to the

foe . . . Gold causes authorities to be interpreted with

something more than subtlety, makes and unmakes laws,
turns ordinary men into perjurers, and kings, times

 

32Correa, "Lendas da India," 30.

33Lobato, Novos Subsidios, 137-138.

 

34Quoted in C. R. Boxer, An African Eldorado: Mono-
motapa and Mogambique, 1498-1752 (Salisbury, 1960), l.

 

49

without number, into tyrants. Even among those who
are wholly dedicated to the service of God Omnipotent
you will find countless examples where this enchanter

has corrgpted and misled under cover, withal, of
virtue.3

 

35Luis Vaz de Camées, The Lusiads (1572), trans.
W. C. Atkinson (Harmondsworth, 1952), 198.

CHAPTER III

TERRA INCOGNITA: SOFALA AND THE

STATES OF SOUTHERN ZAMBEZIA

Fortress and Factory: Sofala's
Early Years, 1505-1530

 

 

Lisbon's strategy for subjugating the orient was to
establish a chain of coastal factories and fortresses
throughout the Indian Ocean basin. The crown considered
Sofala one of its essential East African links and, there-
fore, instructed the first viceroy, Francisco de Almeida,
to secure it from Muslim domination.

According to Almeida's 1505 regimento (royal in-

 

structions), he was to enter the port, intern the "Moorish"
traders, and confiscate their merchandise. The local
African population, in contrast, was to be well-treated:

the natives of the land you shall not hurt either their
persons or their property, for it is our wish that they
be protected and you shall tell them that in ordering
those Moors to be made captive and their property taken,
we do so by reason of their being enemies of our holy
catholic faith . . . whilst to themselves it will ever
be our pleasure to bestow every bounty and grace to
treat them well and help them as though they were our
own and that we shall ever hold them in esteem.

 

1Instructions to Francisco de Almeida, 5 March 1505,
in Documentos, I, 181.

 

50

51

The king apparently hoped that this plan would
permit the replacement of Muslim merchants with his own
commercial agents. Once the port was secure, Almeida was
to begin the construction of a fortified trading post to
monopolize the gold traffic. The monarch's scheme faltered,
however, when Pero de Anhaia, the designated captain of
Sofala, overloaded his flagship which slowly sank in
Lisbon's harbor before a startled assemblage of dignitaries.
The Viceroy was thereupon instructed to proceed with the
remainder of his fleet directly to Kilwa while Anhaia pre-
pared a new vessel.2 Sofala was, after all, only one of
several strategic and economic enclaves in Portugal's

proposed Estado da India (eastern empire). Grandiose

 

imperial plans could not be postponed to accommodate the
conquest of the fabled gold emporium.

Two months after Almeida's departure, Pero de Anhaia
with his newly out-fitted flotilla set sail for southeast
Africa and reached Sofala in September 1505. The specifics

of the captain's regimento are not known, but (presumably)

 

besides geographical information, it contained new orders
on how to deal with the merchant community. The crown may
have decided on a more subtle approach to the Muslim

traders.3 Anhaia made no attack, but instead peacefully

 

2Ibid., 257.

31n 1974, I discovered a badly mutilated document in
Lisbon which may once have been part of Pero de Anhaia's

52

ventured ashore with a few sailors.4 Yusuf, the elderly
shaikh of Sofala, and his retainers received the Europeans
and granted them permission to build a fortress. The obse-
quious behavior of the Muslims suggests that they had
learned the fate of their coastal neighbors who resisted
the Portuguese. Only a few months earlier, both Kilwa and
Mombasa had been stormed and looted by Almeida's troops.S
Yusuf had no desire to see Sofala similarly victimized.
Pero de Anhaia, nevertheless, doubted Muslim inten-
tions and therefore carefully selected a defensive location
for the stockade. The site, reportedly, was encircled by
a small river and commanded a good view of the bay.6 The
troops quickly constructed a makeshift wood and earthen

palisade to serve until a more permanent structure could be

erected.

 

regimento. The extant section noted the necessity of em-
ploying diplomacy at Sofala to insure peaceful relations and
subsequent trade and tribute; A.N.T.T., Fragmentos, Cx. 3,
Mago 3, Doc. 7, n.d.

 

 

4According to one of Anhaia's crew, the landing
party was small so that it could more easily escape a possi-
ble Muslim assault; see Martin Fernandez de Figueroa, A
Spaniard_in the Portuguese Indies (1512), trans. and ed.
James McKenna (Cambridge, Mass., 1967), 41.

 

5Account of the Voyage of Francisco de Almeida, 22
May 1506, in Documentos, I, 523-537; Barros, "Da Asia," 274.

 

6The drastic alteration of the coast over the centu-
ries precludes an exact geographical description of the
original site. For a thorough study of the early maps on
Sofala see A. Teixeira da Mota, "Cartografia Antiga de
Sofala," Monumenta, IX (1973), 5-18.

 

53

The fledgling settlement, in fact, scarcely sur-
vived its first years. The land was soon found to be
insalubrious and several of Anhaia's troops died, apparently
from malaria. Meanwhile, trade had dwindled to a trickle
because of Portuguese restrictions on Muslim commerce and
their related attempts to replace Indian textiles and
beads (which Africans demanded) with European manufactures.
The Lusitanians awaited the arrival of quantities of gold
bullion, but found that their offers of woolen caps, table
cloths and brass chamber pots attracted few Shona traders.7
Captain Anhaia, reputedly, even notified the Viceroy that
it was futile to build a permanent fortress at Sofala since
it would needlessly sacrifice men and provide Lisbon with
dominion over nothing.8

In 1506, the settlement's prospects further shrivel-
ed when the frustrated Muslim community, assisted by local
Africans, assaulted the fort. According to an eye-witness
account, Yusuf's forces might have succeeded if the weakened
defenders had not been forewarned by a Swahili defector and
consequently employed their superior firepower. While such
weaponry was often inefficient in the field, it could prove

effective in a struggle waged from a fortified encampment

 

7Pero de Anhaia to King's Treasurers, 27 October
1505, in Documentos, I, 297-299; Pero de Anhaia to Manuel
Fernandes, 29 December 1505, in Documentos, I, 341; Barros,
"Da Asia," 277.

 

 

8Correa, "Lendas da India," 26.

54

such as Sofala; even so, the Muslim siege lasted three
days.9 As a warning to others who dared to contest Lisbon's
sovereignty, Yusuf was executed and his head was displayed
on a pike outside the stockade.

Although Anhaia designated a new, and seemingly
tractable Shaikh, he took little pleasure in the event. A
caravel reaching the port a short time later found the
settlement blighted and without supplies. Anhaia, like
many of his men, was dead.10 His post had been assumed by
the factor, Manuel Fernandes, who had acquired local
notoriety when he personally decapitated Yusuf.

Fernandes's brief tenure as captain did nothing to
diminish his reputation. Just before the arrival of Nuno
Vaz Pereira, the new commandant, in 1507, Fernandes was
caught pilfering the royal coffers. Pereira's inquiry
discovered that his predecessor was also guilty of embezzle-
ment.ll Despite this villainous record, Fernandes was never

convicted and, in fact, was knighted by the king and given

 

9Figueroa, Portuguese Indies, 49-51. On the question
of firearms see Richard Gray, “Portuguese Musketeers on the
Zambezi," JAH, XII (1971), 531-533; E. A. Alpers, "Review of
Eric AxelsSHTs, Portuguese in South-East Africa, 1488-1600,"
IJAHS, VIII (1975T, 93.

 

 

10Pero Quarema to King, 31 August 1506, in Documentos,

 

I, 627.

11Enquiry Instigated by Nuno Vaz Pereira, 25 Febru-
ary 1507, in Documentos, II (Lisbon, 1963), 170-177; cf.
Axelson, Portuguese . . . 1488-1600, 59-60.

 

 

55

a coat of arms.12 It appears that the captaincy of Sofala
could be lucrative for the unscrupulous official.
Ironically, the circumspect Pereira was dismissed
not for any misdeed, but because the great distances
separating Portuguese centers made communications difficult.
After learning of Anhaia's death, the king had dispatched
a replacement, Vasco Gomes de Abreu, unaware that the
Viceroy already had sent Pereira from Goa. After arriving
at Sofala in 1507, Abreu found that Pereira had constructed

a stone fortress at the site,13

a surprising achievement
since the bulk of building material had to be imported
from Kilwa and even Portugal.

Besides assuming command of Sofala, Abreu had been
commissioned to establish a way station and trading post on
Mozambique Island. The island, in contrast to Sofala,
possessed a shoal free anchorage and was strategically
located to intercept Muslim interlopers. Abreu's men soon
began work on the new settlement.

Meanwhile, the captain was occupied with the plight

of Sofala. Official memoranda claimed that while there was

 

12Castanheda, "Discovery and Conquest," 394.

13The square bastion, later named Sao Caetano,
enclosed a small community which included not only barracks,
but also a parish church, a factory, and an observation
tower; see dos Santos, "Eastern Ethiopia," 185.

56

much gold in the land, little ever reached the fortress.14

To redirect commerce to Sofala, Muslim contraband traffic
had to be suppressed. Before Abreu could provide any
remedy, he mysteriously disappeared along the coast in 1508.
In frustration, Viceroy Afonso de Albuquerque (1453-
1515) proposed to the king that the fortress of $50 Caetano
be leased to the "Moors" of Malindi for a few years "so
that the affairs of Sofala may be better understood."15
Anténio de Saldanha, Sofala's new captain, decided upon a
more direct approach. In 1511, he launched a surprise
attack on the center of clandestine Muslim commerce, Angoche
Island. Although Saldanha looted and burned the settlement,
by the following year merchants seeking to avoid the Portu-
guese monOpoly were once more prospering there.16
At Sofala itself, the captain discovered that,
during his absence, the fortress had faced another crisis.

Yusuf's successor, Maulide, had not proved as pliable as

the Portuguese hoped. Without warning, the shaikh quit

 

14Afonso de Albuquerque to King, 6 February 1507,
in Documentos, II, 127; Duarte de Lemos to King, 30 Septem-
ber 1508, in Documentos, II, 293; Lobato, Expansao Portu-
guesa, III, 96-97.

 

 

15Summary by Ant6nio Carneiro of Letters from
Afonso de Albuquerque, 1511, in Documentos, II, 127.

 

16M. D. D. Newitt, "The Early History of the Sul-
tanate of Angoche," AAA, XII (1972), 397-406; B. A. Datoo,
"Selected Phases of the Historical Geography of the Major
Eastern African Ports" (unpublished Ph.D. dissertation,
University of London, 1968), 194, 197-200.

57

Sofala to reside in the hinterland with his African allies.
He abandoned his earlier allegiance to Lusitania and block-
aded all trade routes to the entrepot. After three months
of Maulide's harassment, a Portuguese column secretly left
the fort and proceeded inland; the shaikh was captured and
executed.17

Saldanha and his comrades were beginning to appre-
ciate that not only did they know little about the nature
of Sofala's external economy, but they were also ignorant
of the interior African states which controlled the coveted
sources of gold. As early as 1513, Pero Vaz Soares, the
factor of Sao Caetano, correctly surmised that "although
there is gold in all the land it is spread out . . . [no
communities have] it in such quantity as to allow them to
come so far [as Sofala] to trade it . . . they wait for
the merchandise to be taken to them."18 Consequently,
Sofala's expenses continued to exceed revenues. In fact,
subsidies designated for the heirs of deceased Portuguese
soldiers were spent simply to cover the fort's maintenance.
Soares added that since desired gold supplies had not

materialized, greater emphasis should be placed by Lisbon

 

17Maulide's insurrection is only known in outline
from a document provided by his executioners; see Statement
by Pero Sobrinho and Diogo Homen, 15 April 1512, in Docu-
mentos, III, 234-249; Lobato, Expansao Portuguesa, III,
102-103.

 

18Pero Vaz Soares to King, 30 June 1513, in Docu-
mentos, III, 461.

58

upon the acquisition of ivory. Elephants abounded in the
area, and their tusks could be sold profitably in India.19
In 1530, the crown officially made Sofala's ivory

a royal monopoly,20 but, by that time, the settlement was
already in a state of ruin. One observer reported:

The factory has fallen to the ground and has to be

re-timbered and the water flows through it as it

does in the street; the granary is falling down;

nobody can go into the church; the house where the

captain dwells has to be re-tiled and repaired; a

wall begun by Samcho de Thoar must be finished, so

I am told, in order that the fortress does not fall

to the ground.21

Except for officials who accumulated considerable

fortunes by fleecing the king's treasury and engaging in
private trade, Sofala had proved a financial fiasco. Yet,
Lisbon remained sanguine over $50 Caetano's prospects and
refused to consider relinquishing it. Unlike the once
opulent Kilwa, which had been quickly abandoned when it
proved unprofitable, Sofala retained an impressive biblical
reputation which even its grim reality did not diminish.
The Portuguese, however, soon realized that the acquisition
of wealth would require penetration of the interior and

the establishment of direct relations with various African

polities.

 

1912i§., 459-469; see also Dames, Book of Duarte
Barbosa, I, 8.

 

20Regulations for Sofala, 20 May 1530, in Documentos,
VI (Lisbon, 1969), 355-357, 389-391.

 

21Anténio da Silveira to King, c. 1529, in Documen-
tos, V (Lisbon, 1966), 559.

59

The Shona and Early State Building

 

When the Portuguese began to filter inland from the
coast during the sixteenth century, they encountered a
number of powerful African states. Before considering the
course of Afro-Lusitanian relations, it is necessary first
to examine the evolution of Southern Zambezia's indigenous
communities.

The fertile plateau which crowns much of the inte-
rior proved an attractive early habitat and was first
settled by hunter-gatherers.' During the initial centuries
of the first millennium A.D., however, Bantu-speakers began
entering the region and displacing (or incorporating) the

original population.22

The tall grasses, perennial rivers,
and favorable climate of the highland were suitable to
their agricultural and herding activities. These Early
Iron Age immigrants also began (to a limited extent) to
mine copper and iron both for local use and to exchange

within regional bartering networks.23

Apparently, they did
not, however, exploit the gold bearing reefs of the region,

and any gold which they obtained was probably from alluvial

 

22T. N. Huffman, "Radiocarbon Dates and Bibliography
of the Rhodesian Iron Age." Rhodesian Prehistory, 11 (1973),
4; R. C. Soper, "New Radiocarbon Dates for Eastern and
Southern Africa," JAH, XV (1974), 184-185; D. W. Phillipson,
"Early Iron-Using FEEples of Southern Africa," in African
Societies in Southern Africa, 30-35, 40-42.

 

 

230. N. Beach, Angutlinetof Shona History (forth-
coming), Chapter I; Brian Fagan, "Early Trade and Raw Mate-
rials in South Central Africa," JAH, X (1969), 2, 10.

 

60

sources. Mining and commerce were largely marginal economic
activities for these farmer-herders.24

In about the eleventh century, a second wave of
Bantu-speakers invaded Southern Zambezia and established
the related Leopard's Kopje and Zimbabwe traditions over
much of the region.25 These people, tentatively identified
as Shona,26 assimilated their culturally similar forebears
and came to dominate the interior.

It was within the economic sphere that the Shona
differed most from their predecessors by exploiting the
gold reefs of the plateau. Although cattle represented a
source of food and a mark of wealth to the Shona, it was
their export of gold, along with copper and ivory, which

placed these Bantu-speakers within the Indian Ocean complex.

 

24Beach, Outline of Shona History, Chapter I; I. R.
Phimister, "Pre-Colonial Gold Mining in Southern Zambezia:
A Reassessment," African Social Research, 21 (1976), 4. The
trading and mining pursuits of these people have been inade-
quately studied thus far.

 

 

25T. N. Huffman, "Excavations at Leopard's Kopje
Main Kraal: A Preliminary Report,“ SAAB, XXVI (1971), 85-89;
idem, "The Leopard's Kopje Tradition" (unpublished Ph.D.
d1ssertation, University of Illinois, 1974), 136-154, 180-
183, 185; Garlake, Great Zimbabwe, 155-159.

 

261t should be noted that the term Shona is, in the
strictest sense, a linguistic designation for the south-
central zone of Bantu-speakers. It is not an ethnic or
'tribal' label, although the six Shona dialect groups display
broadly similar cultural patterns from the interior to the
coast; see C. M. Doke, Report on the Unification of the Shona
Dialects (Hertford, 1931), 34-35, idem, The Southern Bantu
Languages (London, 1954), 21—22; K. G. Mkanganwi, i'The Rela-
tionship of Coastal Ndau to the Shona Dialects of the Inte-
rior," African Studies, XXXI (1972), 111-137.

 

 

 

 

61

At Leopard's Kopje sites (near modern Bulawayo) the dis-
covery of glass beads and rich funerary goods points to
increased long-distance trade, new artisan skills, and
social stratification. It is, however, by examining the
related Zimbabwe culture that a more detailed picture of
early state building emerges.

The well studied site of Great Zimbabwe provides
evidence of the evolution of Shona dominance over much of
Southern Zambezia. Located to the southeast of the plateau,
Zimbabwe provided a suitable environment for a centralized
state. Rainfall was more abundant there than in surrounding
areas, providing reliable water supplies for crops, cattle,
and the growth of a rich woodland. Lying nearby were two
other ecological zones which possessed rich red soils for
farming and open grasslands for grazing. Moreover, placer
deposits of gold around Great Zimbabwe made it additionally
attractive to the Shona.27

It was the establishment of the gold trade to the
coast which largely generated the expansion of Zimbabwe.
The income from long-distance commerce was concentrated in
the hands of Shona aristocrats who were distinguished from
their subjects by their material culture and increased
authority which included control over the redistribution of

exotic goods. The monumental architecture associated with

 

27Garlake, Great Zimbabwe, 15, 184, Huffman, "Rise
and Fall of Zimbabwe," 361; I. R. Phimister, "Ancient Mining
near Great Zimbabwe," JSAIMM, LXXIV (1974), 233-237.

 

62

Zimbabwe can be seen not only as an ostentatious display of
power and wealth, but also as public works projects to
organize the local labor force.28
Although mining and trade were significant during
Zimbabwe's height of dominance (c. 1350-1450), they were
not the primary Shona occupations, but rather part-time,
seasonal activities for a population of farmers and herders.
While commerce provided the Shona aristocracy with prestige
goods, agriculture sustained the economic system and cattle
were the mark of wealth. Even among Zimbabwe's successor
states, farming and pastoralism continued to underpin
their economies.29
By the fourteenth century Zimbabwe had extended its
political and cultural influence over Southern Zambezia, as
evidenced by the appearance of characteristic stone build-
ings throughout the region.30 Originally these satellites
may have been the seats of provincial administrators who

increasingly exerted their independence from the ruling

dynasty. Eventually, the outlying districts became capitals

 

28Huffman, "Rise and Fall of Zimbabwe," 365.

29See, for example, 8. I. Mudenge, "The Role of
Foreign Trade in the Rozvi Empire: A Reappraisal," JAH, XV
(1974) 373-391.

30Huffman, "Rise and Fall of Zimbabwe," 365; idem,
"Ancient Mining and Zimbabwe," 241; P. S. Garlake, "Rhodesian
Ruins--A Preliminary Assessment of Their Styles and Chronol-
ogy," JAH, X (1970), 505-506; idem, Great Zimbabwe, 193,
196; 1355, fInveStigation of Manekweni, Mozambique}" 25-47;
Oliveira, Amuralhados da Cultura Zimbaue-Monomotapa, 41-103.

 

 

63

of competing Shona states.31 In turn, these offshoots
often incorporated other peasant groups and enlarged their
own political sphere. Zimbabwe, therefore, was not actually
the center of any far-flung empire, but the stimulus to
state-building and economic expansion in southeastern
Africa.

It is ironic that by about the mid-fifteenth century
Great Zimbabwe had been virtually abandoned.32 The resources
in the area, including timber, grazing land, soil, and even
gold, had been depleted.33 Such overall environmental de-
terioration was common, and the Shona responded by vacating
the site. One result of the break-up of Zimbabwe was the
establishment of the Mutapa kingdom which was to obsess
Portuguese traders and officials.

The Rise of the Mutapa Empire:
Another Look at the Evidence

 

 

Until recently, the orthodox view on the origin and
development of the Mutapa state has rested upon a scatter
of Portuguese documents, some incomplete archaeological
evidence, and, most importantly, the writings of D. P.

Abraham who collected oral traditions in the 19505.

 

31Beach, Outline of Shona History, Chapter I.

 

32P. S. Garlake, "The Value of Imported Ceramics in
the Dating and Interpretation of the Rhodesian Iron Age,"
AAA, IX (1968), 29-30. Garlake's view has found wide
acceptance among historians and archaeologists.

33Phimister, "Pre-Colonial Gold Mining," 20;
Garlake, "Rhodesian Ruins," 507-508.

64

Abraham's articles, although occasionally contradictory and
poorly referenced, reveal the dominance and extent of this
empire during a brief "golden age" in the late fifteenth
century. In summary, Abraham asserted that the ruling
Zimbabwe dynasty under Mutota, in c. 1440, abandoned their
settlement for the northern Dande area along the Zambezi
River. The southern area encompassing Zimbabwe was left

in the hands of a close relative, Torwa (Togwa). Following
Mutota's death (0. 1450), his son Matope (c. l425-c. 1480)
expanded the empire to encompass most of Southern Zambezia,
installing his kinsmen as provincial governors. This unity
was short-lived as Torwa and another nearby ruler, Changamire
(Changamira), revolted after Matope's death and a subsequent
succession dispute. The Portuguese arrived to find the
ruling Munhumutapa struggling to regain his lost provinces
from his disobedient relatives, as well as attempting to

prevent any further diminution of his territory.34

 

34Only the baldest summary of Abraham's views are

possible because of the frequent alterations in his writ-
ings; see D. P. Abraham, "Early Political History of the
Kingdom of Mwene Mutopa (850-1589)," in Historians in Tro-

ical Africa, 62-66; idem, "Maramuca: An Exercise in the
Combined Use of Portuguese Records and Oral Traditions,"
JAH, II (1961), 212-214; idem, "Ethno-History of the Empire
5f_Mutapa: Problems and Metfiods," in The Histprian in Tro-

ical Africa, eds. J. Vansina, R. Mauny, and L. V.‘Thomas

London, 1964), 108-109. For excellent analyses of Abraham's
writings, see D. N. Beach, "Historiography of the People of
Zimbabwe in the 19605," Rhodesian History, IV (1973), 21-
30; idem, "The Mutapa Dynasty: A Comparison of Documentary
and Traditional Evidence," History in Africa, III (1976),
1-17.

 

 

 

 

 

 

65

Unfortunately, there is a notable flaw in this in-
terpretation: it is based exclusively upon Mutapa-biased
information.35 The sixteenth-century Portuguese tended to
accept this version of events, largely because Sofala was
probably within the Mutapa's sphere of influence when they

d36 and there were no other local sources of informa-

arrive
tion upon which to rely. Abraham compounded this imbalance
by collecting traditions from Mutapa-oriented informants
and then confirming his evidence with Portuguese records.
The result obviously has not provided a full and balanced
account of events which took place in Southern Zambezia.

In light of recent research, a very different
historical reconstruction can be proposed. First, it is
apparent that the direct successor of Zimbabwe was not the
Mutapa state, but (the site of) Khami located on the south-
west of the plateau. Migrants, tentatively identified with
the Torwa dynasty, slowly moved westward from Zimbabwe, a
process which can be traced from pottery and stonework

styles. Archaeologically, Khami was the natural outgrowth

of Zimbabwe and its immediate successor in material

 

35Beach, "Historiography," 29; Richard Gray, "Review
of Portuguese in South-East Africa, 1488-1600," JAH, XVI
(1975), 148. -—-

36See, for example, Diogo de Alcacova to the King,
20 November 1506, in Documentos, I, 391; Gaspar Veloso to
King, 1512, in Documentos, III, 183.

 

66

traditions.37 If this is the case, then Abraham's conten-
tion that Torwa was left by Mutota in the south is histor-
ically upside-down; it was Mutota who broke away from the
core state to move northward. It is, of course, not un-
usual that Abraham‘s informants would characterize a senior
and rival Torwa dynasty as junior to their own.38 Such a
maneuver for prestige was also useful in legitimizing the
Mutapa's authority to the Portuguese who were induced to
support him in his struggles against "rebellious vassals."
Assuming that the Torwa state was an independent
Shona center at Khami, the emergence of the Mutapa dynasty
can be deduced as follows. The environmental disaster at
Great Zimbabwe engendered a political crisis which caused a
segment of the population to move northward to Dande, a
more fertile region, closer to sources of gold and copper.
As Zimbabwe had done earlier, the rising Mutapa state began
to establish provincial centers, some of which paid nominal

allegiance to it but were never tightly integrated into a

 

37These ideas are associated with T. N. Huffman and
D. N. Beach with whom I conversed in Salisbury on 8 May 1974
and 10 May 1974 respectively. See Huffman, "Ancient Mining
at Zimbabwe," 241; idem, "Leopard's Kopje Tradition," 161;
idem, "Rise and Fall of Zimbabwe," 356-357, 360; Beach, Out-
line of Shona History, Chapter I; idem, "Historiography,"-_
28-30; Garlake, Rhodesian Ruins," 508; idem, Great Zimbabwe,
169-171, 180; K. R. Robinson, Khami Ruins (CamBiidge, 195977
115-117; G. Fortune, "Who Was Mwari," Rfiodesian History, IV,
(1973)] 3'40

 

 

 

38Beach, "Historiography," 29; cf. Abraham, "Ethno-
history," 108.

67

large-scale empire. The myth of the extent and power of
the Munhumutapa empire was accepted by the Portuguese who
perpetuated it throughout Europe.39

With this construction in mind, the question of
Changamire also takes on a new perspective, although little
is known about the early evolution of this important
dynasty. A 1506 Portuguese document identified the first
Changamire as a chief justice or governor within the
Mutapa empire. His power and popularity engendered the
fears of the Munhumutapa who eventually provoked him to
revolt.40 It may be suggested alternatively, that with the
break-up of Zimbabwe, Changamire simply established his own
independent state. In turn, the Mutapa state characterized
him (like Torwa) as a rebel. Unfortunately, Changamire's
affined relationship to the Mutapa dynasty is unclear and

even his original sphere of influence is unknown.41

In any
case, in c. 1494, an early alliance between Changamire and

Torwa was forged which led these dynasties to struggle with

 

39See references cited in n. 36; Filippo Pigafetta,
A Report of the Kingdom of the Congo and of the Surrounding
Countries (1591), trans. and ed., Margarite Hutchinson
TLondon, 1881), 117-119. Particularly valuable is W. G. L.
Randles, L'Ima e du Sud-Est Africain dans la Litterature
Européenne au XVI Siécle (Lisbon, 1959).

 

 

 

4oAlcagova to King, 20 November 1506, in Documentos,
I, 393-395.

 

41Abraham, "Early Political History," 64; idem,
"Ethnohistory," 108; D. N. Beach, "Historians and the Shona
Empires, Part II" (unpublished seminar paper, University of
Rhodesia, 1972), 19-27.

68

their rival for the domination of Southern Zambezia.42 The
conflict continued intermittently until the mid-sixteenth
century when, with the aid of the Portuguese, the Munhumutapa
succeeded in displacing Changamire's forces in the lands
surrounding Sofala. Thereafter, no reference to the Chang-
amire reign appears in the existing documents until its
late-seventeenth-century revival under Dombo at Danagombi
(Dhlo Dhlo). During the long hiatus, the dynasty was prob-
ably reduced to a minor power which survived along the
periphery of the Mutapa empire.43

Considering these suggested revisions, it is clear
that the Mutapa empire was never so large or powerful as it
claimed. The depiction of the region in terms of conflict
between the Munhumutapa and his rebellious underlings pro-

vides a truncated view of reality. The Torwa and early

 

42Alcacova to King, 20 November 1506, in Documentos,
I, 393-395.

 

43Abraham, however, suggests that as early as the
15005, the Changamire dynasty already had wrestled power
from Torwa and established its kingdom far to the south of
the Mutapa paramountcy (the region from which Dombo emerged
two centuries later to overrun Southern Zambezia); Abraham,
"Ethnohistory," 109. Historical records and oral traditions
though indicate that this invasion occurred much later than
Abraham speculates; see Beach, "Shona Empires, II," 19-27;
S. I. Mudenge, "An Identification of the Rozvi and Its
Implications for the History of the Karanga," Rhodesian
Histor , V (1974), 27, 31; Ant6nio Pinto de Miranda, "Monar-
chia Airicana" (17th century), in Apais da Junta de Investi-
gacoes do Ultramar, IX-I (Lisbon, 1954), 111; AntEnio da
Conceicéo,7flTratado dos Rios de Cuama" (1696), in 0 Chron-
ista de Tisguary, ed. J. H. da Cunha Rivara, II (Nova Goa,
1867), 68; E. M. Lloyd, "Mbava," NADA, 3 (1925), 62.

 

 

 

 

69

Changamire dynasties were politically underestimated by the
Portuguese who based their interpretations upon what they
were told by their Mutapa associates. Since so little is
known about either the Khami or Changamire competitors, an
examination of aspects of the more familiar Mutapa empire
is in order.

When Mutota and his retainers first migrated north-
ward to the Dande region, they were undoubtedly attracted
by mineral resources and the growing Zambezi trade. It is
rarely mentioned, however, that the Mutapa state, like
other Shona polities, was sustained by farming and herding,
with cattle one of the chief signs of wealth.44 To some
extent the Portuguese recognized these economic facts when
they complained about the insolent Africans who only mined
gold when they were coerced. Anténio Bocarro (1594-c.
1649), the royal chronicler, observed that the land of
Munhumutapa had plentiful crops and abounded in cattle
since "the greater number of the Kaffirs are inclined to
agricultural and pastoral pursuits, in which their riches

consist."45 Yet, the Mutapa state was not simply a

 

44Beach, "Shona Empires, I," 12; idem, "The Shona
Economy: Branches of Production" in The Roots of Rural Pov-
erty in Centra} and Southern Africa, eds. Robin Palmer and
Neil Parsons (Berkely, 1977), 40-47, Phimister, "Pre-
Colonial Gold Mining," 20-21.

 

 

45Anténio Bocarro, "Decade" (1876), in RSEA, III
(London, 1899), 355. The same view is reiterated in Faria
e Sousa, "Asia Portuguesa," 23. For Torwa see dos Santos
"Eastern Ethiopia," 274.

70

subsistence level peasant society. International commerce
had important effects upon the community, particularly in
the way people thought about their economy and social re-
lations.46 Although trade brought changes, the Mutapa state
was not merely a torpid puppet in the process. For the
rulers, the exchange of gold, copper, and ivory for cloth
and beads allowed a process of accumulation and expansion
to take place, analogous to Great Zimbabwe's rise. The
Portuguese, however, were unlike earlier Swahili and Muslim
merchants, their interests included political domination.
The first tentative Luso-Mutapan encounter took
place at Sofala as early as 1506, when "a Kaffir of the
interior of Menapotaque" reached the fort to initiate
trade.47 For the rising Mutapa state, an alliance with the
Portuguese had two aims: it provided the exotic goods de-
sired by the Shona rulers, for themselves and for redistri-
bution among their people, and it established a diplomatic
tie. As previously indicated, the Mutapa kingdom was not
secure; it was threatened initially by Torwa and Changamire,
and later by its provincial appointees. The Portuguese
were potential allies in such clashes. Actually, the insta-
bility characterizing the Mutapa empire was inherent in

Shona political systems. There was no clear method of

 

46E. A. Alpers, "Re-Thinking African Economic His-
tory," Ufahamu, III (1973), 120; Mudenge, "Role of Foreign
Trade," 390.

47Pero de Anhaia to King's Treasurers, 19 May 1506,
in Documentos, I, 507.

 

71

succession between dynasties or even within dynasties, and
civil war was an ever-present threat. Dos Santos observed:

This uncertainty in the succession is because the

Kaffirs say that any legitimate son of former kings

of that land can inherit the kingdom of his father,

and that he has the best right of succession who is

best fitted to govern.
Differing views on candidates opened the gate to factional
politics and conflict.

In an attempt to inhibit such fragmentation, a

three tier administrative system was devised to balance
threats to the Mutapa's authority. The core of the empire
(the Dande "metropolitan" area) was administered by the
Munhumutapa and his appointed territorial chiefs, known as
mambos. The latter were composed of patrilineal relatives,
close associates, and even senior wives, all of whom held
their posts and lands for life.49 This system theoretically
stabilized politics by surrounding the paramount ruler with
a coterie of loyal supporters who owed their positions to
him. These officials collected taxes and probably performed

ritual (or religious) obligations in exchange for their

hereditary positions and lands. To insure that the Mutapa's

 

48dos Santos, "Eastern EthiOpia," 191; also see
Beach, "Shona Empires, I," 13-14.

49Allen Isaacman, Mozambiqge: The Africanizarion of
arEuropean Institution (Madison, 1972), 7; Abraham, "Early
PoliticaIiHistory,"63; Bocarro, "Decade," 358; Beach, "Shona
Empires, I," 14.

 

 

72

perquisites were observed by the mambos, there existed an
elaborate royal bureaucracy and a powerful army.50

At the second administrative level, provincial rule,
a more difficult problem was posed. Those lineages barred
from the throne often were granted outlying lands, thereby
theoretically removing them as potential rivals. Yet, these
same regions later served as offensive bases against inef-
fectual Munhumutapas whose justification to rule was no
better than the claims of their opponents. To prevent the
provincial chiefs from attaining complete autonomy, the
Munhumutapa demanded periodic service and gifts, and even
kept the sons of his governors as hostages.51 Another
method was to recall rulers who posed a threat to his reign,
although without military intervention such an order probably
was not obeyed. As in the case of Changamire, the command
may have stimulated more friction.

Besides employing provincial appointees, the Mutapa
exerted some influence over indigenous authorities whose
lands remained under their own control. This form of indi-
rect rule formed the third administrative tier and was

based upon tax payments and allegiance. This method,

 

50Bocarro, "Decade," 356-357; Miranda, "Monarchia
Africana," 112-116; Ajuda, 51-viii-40, fol. 214v., Vasco
Fernandes Homem to Luys da Sylva, 15 February 1576.

51Bocarro, "Decade," 356-357.

73

utilized especially among the Sena and Tonga (along the
Zambezi), had only limited success.52
In providing a composite picture of political

administration it should be emphasized that there were no
clearly defined procedures, and much was dependent upon the
personality of the individual Munhumutapa. The essential
point is that, even in its halcyon days under Matope, the
empire was far more circumscribed than its representatives
cared to admit. The Munhumutapa's overlordship of his

dependencies was more de jure than de facto.53

Why the
Portuguese chose to accept the Mutapa view is related to
their initial contacts in southeast Africa as well as the
advantages they perceived would be gained from cooperation.
In practical terms, the Europeans quickly discovered that

they had to deal with several different Shona rulers,

whether they regarded them as legitimate or not.

States of the Sofala Hinterland

 

The Portuguese found that even the communities
forming the Sofala hinterland, and therefore claimed by the
Munhumutapa as part of his domain, were hardly servile.
Manyika, Teve, and Danda were really offshoots from the

empire and only nominally subject to its control.54 For

 

52Abraham, "Early Political History," 64; Isaacman,
Mozambique, 8.

 

53Beach, "Shona Empires, I," 18.

54Ibid., 17. See figure 4.

74

such distant provinces, the centrifugal forces in Shona
society tended to be exacerbated by the least instability
at the center. With little supervision over their activi-
ties, provincial governors held free rein in their lands.
Portuguese chroniclers and officials, however, perceived
the states surrounding Sofala as fragments of a once fully

55 Although the Shona ruling

integrated Mutapa empire.
dynasties were related and culturally similar, political
unity was always tenuous. Even boundaries were flexible
and subject to dispute, and when the lands around Sofala
engendered such conflicts, the Portuguese became not only
participants but also pawns.

Located to the northwest of Sofala was the important
gold producing area of Manyika.56 According to one tradi-
tion, the first Chikanga (ruler) of Manyika was the son of

one of the Mutapan provincial rulers and was himself re-

lated to the royal dynasty.57 The documents indicate that

 

55See, for example, dos Santos, "Eastern Ethiopia,"
273; Diogo De Couto, "Da Asia" (1778), in RSEA, VI, 391.

56Abraham has placed Manyika in the Inyanga and
northern Umtali districts of Rhodesia with extensions into
Mozambique; see Abraham, "Early Political History," 83, n.
44. It is difficult to be precise in light of the fluctua-
ting political situation at the time, as well as from the
documents available.

57The Chikanga was the son of the Makombe of Barwe
who was a matrilineal relation to the Munhumutapa according
to Abraham, "Early Political History," 66-67, 83 n. 44.
This view rejects the Portuguese claim that the Chikanga was
actually a son of the Munhumutapa; see n. 55 above for cita-
tions. In this case, Abraham's evidence seems stronger.

75

 

 

 

,."" \o"‘ ‘0
.o ' . . ; . ‘0
you; ‘ Zumho I ...
‘ We Tot? '..o'
uurlr1\_ %.~:
’ ' .'.'
Innmruro ‘.
BANUE ‘enn
ILUN -r-
.' NYIKA
,3 unno R.
'.‘
m i .‘ ' '
in .' Sofala
f DANUA

 

Limpopo a. \ INDIAN OCEAN

r-"

 

0 Mi lea 250

 

Figure 4. Statoo of tho Sofala Hinterland

 

76

in both domestic and foreign policy the Chikanga was
virtually an autonomous authority until the seventeenth
century. And although the Chikanga and the Munhumutapa
were old adversaries, they did cooperate when it was
mutually beneficial.58

Unlike the Mutapa state and other Shona polities,
it has recently been suggested that gold production and
trade, rather than agriculture and herding, were the main

59

pursuits of the Manyika people. Since rivers in the

region were perennial, gold washing as well as mining in
alluvial ground could have been carried on year-round.60
While proximity to the coast would have facilitated commerce
from Manyika, it still remains unlikely that such gold
working replaced farming as a full-time occupation.61 Por-
tuguese troops who reached the region late in the sixteenth
century were disappointed to discover "the difficulty and

labour of the Kaffirs, and with what risk and peril of

their lives they extracted gold from the bowels of the

 

58Francisco de Brito to King, 8 August 1519 in
Documentos, VI, 13; Ajuda, 51-viii-40, fols. 212-212 v.,
Homem to da Sylva, 15 February 1576; dos Santos, "Eastern
Ethiopia," 285-287; Bocarro, "Decade," 362-364.

 

59H. H. K. Bhila, "Trade and Survival of an African
Polity: The External Relations of Manyika from the Sixteenth
to the Early Nineteenth Century," Rhodesian History, III
(1972), ll.

 

60;§$§., Phimister, "Pre-Colonial Gold Mining," 24-25.

61Phimister, "Pre-Colonial Gold Mining," 25-26;
Beach, "Shona Economy," 52, 63 n. 124.

77

earth and from the stones."62 Lisbon's overemphasis on
gold has obscured Manyika's other economic activities. In
any case, her alluvial sources were rapidly exhausted,
and by the mid-eighteenth century gold production had
declined considerably.63
Leaving aside economic considerations, Manyika was
faced with other major concerns. Foremost was the antagon-
ism to her efforts at independence by the Munhumutapa, who
was often joined by the Portuguese. In addition, the
Chikanga was also challenged by the coastal states of Teve
and Danda. These two related communities formed the imme-
diate hinterland for Sofala and seem to have stretched from
about the Pungwe River to the Sabi River. Their boundaries
were flexible, and at different times Sofala fell under the
influence of each state. Teve's first ruler or Sachiteve
supposedly was the son of the successful competitor for the
Mutapa throne (after Matope's death).64 He, in turn, dis-

patched his own son and successor, Nyamunda, to expand the

 

62dos Santos, "Eastern Ethiopia," 218. Diogo de
Couto characterized Manyika gold working as a "poor and
miserable business" in his "Da Asia," 389. The observation
is true for Southern Zambezia in general.

63H. H. K. Bhila, "The Manyika and the Portuguese,
1575-1863" (unpublished Ph.D. dissertation, University of
London, 1971), 87. Europeans had a "bullionist fantasy"
according to Professor Curtin. "A country with gold mines
was automatically taken to be wealthy, regardless of cost
conditions"; Philip Curtin, "The Lure of Bambuk Gold," AAA,
XIV (1973), 627.

64Abraham, "Early Political History," 65-67, 84 n.

47.

78

kingdom to the south. The conquered region of Danda was
placed under the authority of Nyamunda's younger brother

65 After becoming

(or son) who assumed the title of Sedanda.
Sachiteve, Nyamunda gained Portuguese assistance in his
expansionist activities.

Nyamunda's duplicity exemplifies how Europeans
could be used as a lever in Shona factional politics. The
Sachiteve realized that the Portuguese could be enticed
with gold to support his kreach with the Mutapa state, as
well as the expansion of his own domain. For this purpose,
in 1515, he sent some of his retainers to Sofala to request
peace and friendship, and to deliver a sum of gold to the

66 In return, the Portuguese provided

fortress's captain.
men and arms which Nyamunda used to conquer the surrounding
lands including various gold mines belonging to his foe, the
Chikanga.67 Once he had subjugated the Sofalan interior,

the Sachiteve made an about-face. By 1519, the trade routes

linking Sofala to its hinterland were cut, and merchants

 

65Ibid., 67; J. K. Rennie, "Christianity, Colonialism
and the Origins of Nationalism among the Ndau of Southern
Rhodesia, 1890-1935" (unpublished Ph.D. dissertation, North-
western University, 1973), 67.

553050 Vaz de Almada to King, 8 August 1519, in Doc-
umentos, IV, 277-279.

67£§i§., 285; Brito to King, 8 August 1519, in Docu-

mentos, VI, 13; da Silveira to King, c. 1529, in Documentos,

V, 567-571.

 

79

were robbed and killed.68 Ironically, Portuguese officials
probably supported Nyamunda because they considered him to
be a weak ruler of a small state who could be manipulated
as a counterweight against the perceived predominance of
the Munhumutapa.69 The plan had dual advantages: it would
gain an African ally in the immediate region around Sofala
and would fragment the "powerful" Mutapa empire for easier
conquest. Nyamunda's intrigues, however, altered their
strategy.

From the Teve viewpoint, the economic blockade of
Sofala prevented the delivery of merchandise (primarily
cloth) that its Shona enemies used to pay their warriors.
Furthermore, rather than being dependent upon the Portuguese,
Nyamunda now set the terms of trade and established himself
as overlord of Sofala. The Sachiteve, for example, demanded
that all commerce between the two communities be conducted
by the "Moors of Sofala," which intensely angered the
officials of the fortress70 and forced the captain to admit
that Nyamunda had out-maneuvered the Portuguese with his

"ways and wiles."71

 

 

 

68Brito to King, 8 August 1519, in Documentos, VI,
11-13. .
69 .
Lobato, Expansao Portuguesa, II, 81.
70

Lopo de Almeida to King, 27 August 1527, in Docu-
mentos, VI, 277-279. Apparently, the Portuguese had been
cHarging excessive prices for their beads and cloth;»
Axelson, Portuguese . . . 1488-1600, 102.

 

71da Silveira to King, c. 1529, 569.

80

Unfortunately, after about 1529, Nyamunda vanishes
from the historical record, and it can only be guessed
that after his death the vast Teve domain followed the
pattern of other Shona polities by fragmenting. Besides
being divided into two separate states, Teve and Danda,
Nyamunda's kingdom was probably diminished by the Manyika

reconquest of its gold fields.72

For the Portuguese, the
reduced authority of Teve did not spell an end to their
difficulties. Well after Nyamunda's demise, it was reported
that interior communities often prohibited traders from

73 These states also

reaching Sofala for long periods.
frustrated the fort's establishment of communications with
the Munhumutapa. A Portuguese officer reflected that the
powerful Munhumutapa, original lord of all the region,
"wants nothing more than our friendship."74 What he de-
sired was nothing less than what the Sachiteve had earlier
attained.

During the sixteenth century the states surrounding
Sofala prevented either the Munhumutapa or the Portuguese

from threatening their autonomy. The Sachiteve was eventu-

ally able to compel the captain of Sofala to pay a yearly

 

72dos Santos, "Eastern Ethiopia," 193, 217; Couto,
"Da Asia," 390. Almost nothing is known about the Danda
state; see E. A. Alpers, "Dynasties of the Mutapa-Rozwi
Complex," AAA, XI (1970), 212.

73Joao de Sepulveda to King, 10 August 1542, in
Documentos, VII (1971), 137-139.

 

74Ibid., 139.

81

tax for safe passage through his land.75

It, however, must
be pointed out that neither Teve nor Danda was concerned
exclusively with trade, anymore than Manyika was totally
dependent on gold production for its economic base. Nya-
munda's warlike tactics or commercial inclinations did not
mean that the Teve neglected traditional economic activities.
In fact, the people were often observed herding their

76

cattle. There is also evidence that cultivation was the

predominant responsibility for many of the Teve residents,
since the Sofala hinterland was well known for its rice,

wheat, and fruit supplies.77

A Poor and Miserable Business: Gold
Production in Southern Zambezia

 

 

The proof that sixteenth-century Shona societies
did not overly emphasize mineral exploitation (particularly
gold) is evidenced from the Portuguese descriptions of such
workings. The difficulties, limitations, and dangers
inherent in Manyika mining were true throughout Southern
Zambezia. Furthermore, the combination of an increasing
internal demand for the scattered sources of gold as well

as its declining profitability in international trade (due

 

75dos Santos, "Eastern Ethiopia," 219-220.

76Brito to King, 8 August 1519, in Documentos, VI, 11.

 

77See, for example, "Memorias da Costa d'Africa
Oriental" (1762), in Relagées de Mocambique Setecentista,
ed. A. A. de Andrade (Lisbon, l955)'205; dos Santos, "Eastern
Ethiopia," 207, 209-210.

 

82

to the sixteenth-century Spanish discoveries in the Amer-
icas) encouraged draconic local restrictions on gold
extraction.78 On the whole, gold production was a high-risk,
low-return investment.

While the Shona were astute prospectors, they were
far less capable miners. Even so, they stripped the surface
mineral deposits of Southern Zambezia so completely that
most modern mines have been pegged over old workings.79 It
was in the area of deep shaft mining that the Shona lacked
the necessary technology; once easily worked deposits were
exhausted, they had to be abandoned. According to one
typical Portuguese account, the miners, consisting of men,
women, and children under a village leader, were organized
into working groups which began to dig narrow‘shafts. When
done, steps were constructed within the pit, and the workers
began to excavate by slowly passing the dirt in bowls to
the surface for washing. Gold veins were followed in every

possible direction, but sometimes the poorly constructed

 

78Roger Summers, "Notes on the Economic Bases of
Southern Rhodesia Iron Age Cultures," in Conference on the
History of the Central African Peoples (Lusaka, 1963), no
pagination; idem, Ancientggining in Rhodesia, 194; Phimister,
"Pre-Colonial Gold Mining," 22-23; Victorino Magalhaes-
Godinho, L'Economie de l'Empire Portugais aux XVe e XVIe
Siécles (Paris, 1969), 260.

 

 

79Beach, Outline of Shona History, Chapter I; Eric
Axelson, "Gold Mining in Mashonaland in the 16th and 17th
Centuries," Aptima, IX (1959), 170; Garlake, Great Zimbabwe,
69.

 

83

mines were subject to flash-floods, and rich sources had to
be abandoned for the lack of any way to pump out the water.80

Beside flooding, there were other health and safety
risks. In order to draw off foul fumes and help exfoliate
quartz reefs, fires were set at the bottom of the shafts.81
The process itself was dangerous and, on the whole, poor
ventilation must have made working conditions difficult.
There also existed the imminent threat of cave-ins because
the Shona did "not know how to sink mines or to manage
manage machinery for this work."82

Most mining took place during the winter season
(August, September, and October) when the water table had
fallen and after the harvest, thereby not interfering with
cultivation.83 In November when the rains became heavy,
excavation had to cease because of flooding. Shona pro-
duction methods never solved this problem, probably because
mining remained, much to the consternation of the Portu-
guese, a marginal activity. Besides the obvious technical

and geological limitations in Southern Zambezia, the work

 

80See Manuel Barretto, "Report upon the State and
Conquest of the Rivers of Cuama," (1667), in RSEA, III, 490.

lSummers, Ancient Mining in Rhodesia, 166.

 

82Couto, "Da Asia," 367; also see dos Santos, "East-
ern Ethiopia," 219; Francisco Monclaro, "Account of the
Journey Made by Fathers of the Company of Jesus with Fran-
cisco Barreto" (1569), in RSEA, III, 233.

83Barretto, "Report upon the State," 489-490;
Phimister, "Pre-colonial Gold Mining," 5, 12.

84

was labor intensive with a low profit return. Under these
circumstances the Shona were fully aware of their economic
priorities. An early English writer failed to understand
the Shona system when he noted:

If the negroes were couchous [conscious] of gold

themselves they might make great advantage of it,

but being extremely lazie themselves never work but

wginagrged thereto by extremity of want to oblidge
By the sixteenth century many of the easily mined deposits
had been largely worked out, and the Portuguese caused a
further contraction in operations by offering low prices.85
These features made it necessary for rulers to exert control
over dwindling gold supplies in an effort to increase the
mineral's market value.

Within the Mutapa empire, the ruler invoked the

death penalty for those of his subjects who mined gold
without his permission.86 The fear of working unauthorized

deposits was so great that accidental discoveries of ore

were covered with branches in order to warn others to flee

 

84Bodleian Library, Oxford University, Ms. Rawlinson
A. 334, fol. 63, "Description of the Gold Mines on the
Sofala Coast," c. 1693; Barros makes a similar comment in
"Da Asia," 267.

851. R. Phimister, "Alluvial Gold Mining and Trade
in Nineteenth-Century South Central Africa," AAA, XV (1974),
447 n. 9; idem, "Pre-Colonial Gold Mining," 26-27. For a
view of a comparable situation in West Africa see Curtin,
"Lure of Bambuk Gold," 623-631.

86Alcagova to King, 20 November 1506, in Documentos,

 

I, 391.

85

from the area.87 Production depended on the Munhumutapa's
consent and command, and other Shona rulers enjoyed the
same absolute authority. One Sachiteve, for instance,
executed a number of people, not for digging, but for
having provided the captain of Sofala with information

88 Such disclosures threatened the

about new gold mines.
state's integrity as well as the ruler's dominant political
and economic suzerainty.

Much of the gold traded, of course, was not from
rich veins, but from alluvial sources. Manyika, cited
above, was not alone among Southern Zambezian states in
collecting gold from streams. For some Shona the winter
season was spent panning a few grains and particles of
river gold, a process which required neither extensive
social organization nor complicated methods.89 In turn,
there were probably less restrictions placed upon the trade

of such gold by individual Shona with the Portuguese.

This irregular commerce may have satisfied, to some extent,

 

87dos Santos, "Eastern Ethiopia," 280-281. This
author often exaggerated, but there is no doubt about the
strict prohibitions in force; see also Bocarro, "Decade,"
355; Barretto, "Report upon the State," 490-491; Monclaro,
"Account of the Journey," 234; Couto, "Da Asia," 367.

88Ajuda, 51-vii-34, fol. 46v., Pedro Coelho de
Carvalho, "Papel de Noticia que Deu da Fortaleza de Sofalla,"
3 August 1698.

9Barros, "Da Asia," 266-267; Couto, "Da Asia,"
389; dos Santos, "Eastern Ethiopia," 219, 280; Barretto,
"Report upon the State," 489; Phimister, "Alluvial Gold
Mining," 445-447.

86

the African desire for exotic items without upsetting local
economic foundations. Yet, the Shona laborer's output

of gold rarely exceeded a few ounces for an entire season's
work; in return he obtained only a few pieces of cloth or
some beads.90 This unequal exchange was in itself a
natural disincentive to gold working, even without the
influence of royal prohibitions.

Eventually the Portuguese realized that gold pro-
duction in Southern Zambezia was a poor business. Disil-
1usioned by the amount of capital, equipment, and labor
necessary for development, Portuguese traders even refused
mineral concessions.91 The official mind was slower to
accept such a bleak view, and in Lisbon the vision of
African Eldorado persisted. At best, it may be said that
after 1519, the crown began to place greater emphasis upon
the ivory trade, and it became the major Mozambican export
by mid-century.

In passing, it should be noted that ivory collecting
in Southern Zambezia was certainly as dangerous as gold
mining. Little source material is available on the elephant
hunt, but the methods described by two sixteenth-century
observers reveal the high risk involved. One procedure was

for a group to attack a sleeping elephant with spears and

 

90Summers, Ancient Mining in Rhodesia, 22; Phimister,
"Pre-Colonial Gold Mining,“ 22.

 

91Monclaro, "Account of the Journey," 233, 253.

87

then retire until the animal was presumed dead. Unfortu-
nately, the hunters sometimes found that they had miscalcu-
1ated, and they instead became the victims.92 Even more
precarious was the method in which Africans employed only
small hatchets to hack at the elephant's legs until it
stumbled and was killed.93 For the Shona, ivory collection
remained another means to acquiring fine cloth held by their
rulers and imports from the Portuguese.94

It should be clear that both gold mining and ivory
hunting were adjunct activities for the Shona. Each pro-
vided highly desired items for local use and international
trade. Shona rulers not only carefully supervised produc-
tion, but also controlled long-distance commerce through
the collection of import and export duties from their own
people and, later, Portuguese traders.95 The European
arrival, however, provided a new element in the economic

and the political complexion of Southern Zambezia.

 

92dos Santos, "Eastern Ethiopia," 321.

93André Fernandes to Luis Fréis, 25 June 1560, in
Documentos, VII, 483. Undoubtedly, traps were also set to
capture big game; see E. A. Alpers, Ivory and Slaves in East
Central Africa (Berkeley, 1975), 16.

 

 

 

94Interview with Mzee Makandara, 16 May 1974, Nova
Sofala, Mozambique.

95David Chanaiwa, "Politics and Long-Distance Trade
in the Mwene Mutapa Empire during the Sixteenth Century,"
IJAHS, V (1972), 429-430; Veloso to King, 1512, in Documen-
Egg, III, 183. Professor Magalhées-Godinho contends tfiat
the Portuguese payments were customary gifts rather than
taxes, but this objection is more semantic than real; see
Magalhaes-Godinho, "L'Economie de l'Empire Portugais," 255.

 

 

CHAPTER IV

MERCHANTS, MISSIONARIES, AND MUSKETEERS

IN THE INTERIOR, 153l-C. 1600

Much to their chagrin, the Portuguese learned that
Sofala was a costly undertaking. Their solution was not
to abandon the port, but to gain control over the actual
sources of gold in the mysterious interior of Southern
Zambezia.

As early as 1511, the degredado Antonio Fernandes

 

had been commissioned to explore the vast terra incognita

 

composing Sofala's hinterland. He was instructed to estab-
lish friendly relations with the surrounding peoples and to
discover what valuable resources they possessed. Unfortu-
nately, Fernandes's peregrinations are only partially

known because he was illiterate, and his adventures were
recorded only years later by secretaries who themselves

had only the vaguest knowledge of local geography.1 For

 

1The primary documents on Fernandes's journeys are:
Veloso to King, 1512, in Documentos, III, 181-189, and
Almada to King, 26 June 1516, in Documentos, IV, 275-295.
Other scattered references appear in the literature. An
annotated version of these letters with maps is also avail-
able; see Hugh Tracey, Ant6nio Fernandes, Descobridor do

 

 

88

89

several years, Fernandes served as both explorer and roving

ambassador. Subsequently, he suggested to Portuguese

authorities that they establish permanent settlements in

the interior to guarantee their interests. Specifically,

he recommended the construction of a factory house along

the Zambezi for the acquisition of gold and ivory, and as a

restraint on illicit Muslim commerce in the region.2 It

was, however, not until about 1531 that Sena and Tete, sites

of previous Swahili fairs, became Portuguese strongholds.

Partly as a result of Sofala's failure, Lisbon had finally

decided to penetrate the hinterland of Southern Zambezia

and thereby establish her imperial presence beyond the coast.
Ironically, instead of sustaining Sofala, the

Zambezi settlements sealed its fate. Sena and Tete became

linked to the port facilities of Mozambique Island and later

Quelimane. With the lines of commerce following the Zambezi,

Sofala's continued degeneracy was guaranteed, particularly

3

because the Manyika gold commerce now had another outlet.

By trading at the new inland fairs, the Manyika people

 

Monomotapa, 1514-1515, trans. and ed., Caetano Montez
(Lourenco Marques, 1940). The English version: Hugh Tracey,
"Antonio Fernandes, Rhodesia's First Pioneer," Rhodesiana,

19 (1968), 1-26, is less useful. Recent commentators have
made extensive efforts to chart Fernandes's journeys and
identify the places he visited, note, for example, some of
the works cited in Elkiss, "On Service to the Crown," 52 n.
37, and R. W. Dickinson, "Antonio Fernandes--A Reassessment,"
Rhodesiana, 25 (1971), 45-52.

 

 

 

2Veloso to King, 1512, in Documentos, III, 187.

 

3Bhila, Manyika and the Portuguese, 24.

 

90

avoided tariffs imposed by the Sachiteve and consequent
conflicts. The Portuguese also established early commercial
posts within the Chikanga's realm, and even these factories
were probably tied to the Zambezi trade, not Sofala.4
Sofala's decline not only was a matter of competi-
tion or miscalculation, but also derived from flagrant
transgressions by 850 Caetano's officials. It is little
wonder that in 1539, the famous navigator Joao de Castro
echoed the earlier sentiments of Afonso de Albuquerque by
suggesting that the captaincy of Sofala be leased.5 It
brought little profit and no glory to the crown. The dis-
gusted Viceroy Estesvao de Gama added that it was fruitless
for the king to issue any further instructions to the
settlement, since none were ever followed. He calculated
that "were only half of them obeyed, no more would be
necessary."6 The situation was dramatically illustrated in
the late 15405, when Sofala's captain and factor ordered the
assassination of an obstructive leader of the nearby Muslim
trading community.7 They also reportedly stirred revolt in

the interior when they supported a tractable candidate for

 

4M. D. D. Newitt, Portuguese Settlement on the Zam-

besi (London, 1973), Pedro Barretto de Rezende "Of the
State of India(1634), in RSEA, II, 411-413.

 

5Axelson, Portuguese . . . 1488-1600, 133.

 

6Estevao da Gama to King, 11 November 1540, in
Documentos, VII, 125.

 

7Joao Velho to King, c. 1548, in Documentos, VII,

 

169.

91

the Changamire's throne. In retaliation, the Shona ruler
besieged the Sofalan hinterland, and, according to one
witness, if Mutapan troops had not come to the rescue, "the
fortress would have been lost and the whole land closed
and we would have starved to death."8 Despite their mis-
deeds, the two reprobate officials succeeded in accruing
massive fortunes by stealing almost all of the ivory brought
to the port. In fact, upon departure, the captain was
considered the wealthiest man in the east, with the possible
exception of the Viceroy.9
By mid-century a drastic remedy was proposed to the

crown:

I write to His Highness of the little need I think

he has of the fortress of Sofala, and of the disservice

and little service that is carried out in it, because

the captains have factors and factories there that

overshadow that of His Highness . . . if His Highness

does not think it good for his service to demolish

the fortress at Sofala . . . let him demolish forth-

with the captain's factory.10

The design was apparently rejected, in part due to Sofala's

golden mystique, as well as the efforts of various officials.

 

8Ibid., 173.

9Ibid., 173-175; also see Joao Velho to King, 4 Nov-
ember 1548, in Documentos, VII, 184-189; Afonso de Noronha
to Queen, 27 January 1552, in Documentos, VII, 257. It is
little wonder that the post of captain was highly prized by
ambitious fidal 05 for themselves and their relatives; for
example, Blblloteca Nacional de Lisboa, Cod. 886, £015. 783-
784, "Provisao que Pedio Francisco de Sa para seus Irmaos,"
24 March 1552.

 

 

10Joao de Gamarfa to Queen, 8 November 1555, in
Documentos, VII, 315.

 

92

Lisbon, however, was not unaware of the financial burden
Séo Caetano represented. The king even dispatched an
emissary to the east in an effort to discover why certain
imperial possessions continuously produced excessive and
unnecessary expenses. According to the agent, Sofala was

the most egregious example in the Estado da India, and

 

although he made recommendations, no changes resulted.11

More and more the road to riches seemed to the
Portuguese to run along the Zambezi in the direction of
their ally, the Munhumutapa. Lisbon hoped that its good
neighbor policy would permit the growth of royal commerce
and, subsequently, domination over the legendary wealth of
Southern Zambezia. In turn, the Mutapa state seemingly
viewed Portugal not only as a trading partner, but also as
a military comrade in its struggle with Shona foes.

Even before the mid-sixteenth century a Portuguese
representative had been assigned to the Mutapa kingdom and
had acquired judicial and administrative powers. According
to Joao dos Santos, the resident served "as agent in all
matters between the Portuguese and Monomotapa . . . and as
factor of Monomotapa to receive all the duties paid to him

«12

by the merchants. The existence of this office facili-

tated peaceful commerce as long as amenities and decorum

 

11Antao de Noronha to Captain and Officers of the
Fortress of Sofala, 12 April 1565, in Documentos, VIII

 

12dos Santos, "Eastern Ethiopia," 271.

93

were observed. The failure, for example, to observe the
SEEKS (the traditional payment of cloth made by each newly
arrived captain of Mozambique Island) resulted in an attack
on Portuguese merchants and the seizure of their goods.13
In no way were sixteenth-century Luso-Mutapan relations
simply a matter of European domination. Further proof is
provided by the demise of the first Portuguese missionary
enterprise and the subsequent Barreto-Homem expedition.

Shadowing the penetration of Southern Zambezia by
officials and merchants were the spiritual advisers. Their
first obligations were to tend to their own flock (within
the Portuguese settlements) and to attempt to convert
nearby African populations. The developing commercial and
political relationship with the Munhumutapa led the Portu-
guese to believe that his conversion was both timely and
appropriate. For this purpose, the Society of Jesus chose
Goncalo da Silveira (1526-1561), whose religious assignment
was also diplomatic and economic. There was, therefore, no
contradiction in the Portuguese mind between transmitting
"the light of Jesus Christ" and emphasizing the quantity of
gold available within the Mutapa state.14

It should be pointed out that da Silveira was

atypical of Portuguese priests, and, in fact, his choice

 

13Ibid., 272; Beach, "Shona Empires, I," 25 n. 2.

14Goncalo da Silveira to Society of Jesus, November
1559, in Documentos, VII, 425.

 

94

as first religious councilor to the Mutapa court was a
strategic error. Da Silveira had abandoned his aristocratic
heritage in order to devote his life to the church as a
Jesuit missionary, serving first in India before being
called to Mozambique. Even from his early career, the
father was known for his austere lifestyle and was regarded
as a man obsessed by his religious convictions.15

After reaching Mozambique Island in 1560, da Silveira
traveled south with two companions to a region inland from
the port of Inhambane. During his brief tenure there the
padre calculated that he had baptised 450 Africans.16 Da
Silveira, however, soon returned to Mozambique Island to
collect trade goods and to set sail along the Zambezi for
passage into the Mutapa empire.

The travails of the journey itself seem only to
have stimulated da Silveira's zeal. At one point, after
surviving a violent sea squall, the Jesuit reportedly said

mass in the blazing sun until his head blistered, and then

 

15On his early life, see Anténio Garcia, "Goncalo
da Silveira, Proto-Martir de Mo ambique e da Africa do Sul,
1560-1561," Monumenta, VI (1970?, 53-57.

 

16Gon alo da Silveira to Society of Jesus, 9 August
1560, in Documentos, VII, 503. The report is confirmed by
other documentary sources on da Silveira's efforts among
this mixed Tonga-Shona group. A study of these people in
the sixteenth century is provided in C. E. Fuller, "An
Ethnographic Study of Continuity and Change in Gwambe Cul-
ture" (unpublished Ph.D. dissertation, Northwestern Univer-
sity, 1955), 43-72; cf. Alan Smith, "The Peoples of Southern
Mozambique: An Historical Survey," AAA, XIV (1973), 565-566,
573.

 

95

17 Such rigid self-

refused any remedy for his misery.
discipline also was exemplified when da Silveira spent
eight days aboard ship in seclusion reading a book on the
lives of the saints while "only eating once a day a handful
of roasted grain, and refusing everything else."18 By the
time he arrived at the court of the ruling Munhumutapa
Negomo, Goncalo da Silveira was prepared for his eventual
martyrdom.

Negomo was immediately surprised by the missionary
who, unlike other Portuguese, rejected gifts of gold, ser-
vants, and cattle. Through his interpreter, the ruler
speculated that such a man was unnatural and "must have
been born of the herbs."19 To Negomo, da Silveira's other-
worldliness made him suspect. The padre meanwhile was busy
devoting himself to the massive conversion of hundreds of
the Shona elite, although it is unlikely that the effect
was more than superficial. According to a resident Portu-

guese trader, Negomo was dissuaded from the new Christian

teachings by "Moorish engangas" (wizards) who convinced him

 

l7Luis Frois, "Of the Voyage of the Father Dom
Goncalo to the Kingdom of Monomotapa and of His Happy Passing
Away" (1561), in RSEA, II, 116-117; Book Concerning the Work
of the Society of Jesus in the East, 1573, in Documentos,
VIII, 213.

 

18Frois, "Voyage of the Father Dom Goncalo," 118.

19Ibid., 120. The fact that da Silveira ate no
meat, but only fruit and millet cooked with herbs, may have
contributed to the Mutapa's observation.

96

that da Silveira represented a political threat.20 It is
unnecessary, however, to attribute the padre's undoing to

a Muslim plot; the episode appears to represent the reaction
of traditional religious authorities to Christianity.21
Furthermore, Negomo's anxieties, regarding not only Portu-
guese intervention, but also Shona competitors for his
kingdom, contributed to his mistrust Of da Silveira. The
Munhumutapa seems to have believed that the Jesuit missionary
was actually an agent of the Sachiteve Chipute, his sworn
enemy and claimant to the empire.22 The simple fact that

da Silveira had become a permanent fixture of the court also
threatened to undermine Negomo's authority.

Thus Goncalo da Silveira fell prey not only to the
encounter between competing religions, but also factional
Shona politics; the padre was perceived as an element in
the competition. His intolerant attitudes and impetuous
behavior only contributed to his doom. Da Silveira seems to
have been fascinated with his impending martyrdom: when
warned of Negomo's decision to have him assassinated, he

absolutely refused to depart, forgave his persecutors, and,

 

20Letter’of Antonio Caiado (c. 1561), in RSEA, II,
102; also see Frois, "Voyage of the Father Dom Goncalo,"
122-123.

 

21Beach, "Shona Empires, I," 20-21.

22Letter of Caiado (c. 1561), in RSEA II, 102; Frois,
"Voyage of the Father Dom Goncalo," 123; Abraham, "Early
Political History," 69.

 

97

with a smile, prepared for death.23 Luis Frois, who wrote
the dramatic account of the missionary's final days, added
that da Silveira was impatient to meet his end; he paced
back and forth "as if he wished to be already free and
reigning with Christ."24

On 16 March 1561 the Jesuit father was strangled
and his body cast into a nearby stream. Negomo reportedly
regretted the deed, but his sorrow was related to the
warning of Portuguese merchants that the padre's death
might engender a military reprisal by Lisbon.25 (During
this episode the traders were neither harmed nor expelled,
indicating the absence of general xenophobia.) The Mutapa
state, like its other Shona counterparts, sought to manipu-
late the Portuguese for its own purpose and did not seek
warfare.

The murder of a Portuguese priest, especially from
an aristocratic family, demanded immediate punitive measures
from Lisbon. Yet, the reaction was hardly swift, not being
initiated for almost a decade. It is sometimes assumed that
the misguided fiasco known as the Barreto-Homem expedition
(1569-1575) was the direct result of the execution. The

delay in response, as well as the facts of the mission, deny

 

23Letter of Caiado (c. 1561), in RSEA, II, 103.

24Frois, "Voyage of the Father Dom Goncalo," 125.

251bid., 127.

98

this contention. It may be argued that it was decided only
fater that da Silveira's death could be exploited as a
pretext for invasion and conquest.

When the fourteen-year-old Sebastiéo (1557-1578)
assumed the Lusitanian Crown in 1568, the Jesuit educated
youth was obsessed "with the idea of becoming Christ's
captain against the Infidels."26 One way of achieving such
glory was to dispatch a military expedition against the
Mutapa state. The legal justification for the king's plan
was provided by theologians and lawyers who cited the Mun-
humutapa as being guilty of encouraging Islam and brigandage
as well as da Silveira's murder. They furthermore claimed
that the Shona ruler's failure to allow commerce and the
promulgation of the gospel was punishable by "lawful"
warfare.27 Padre Monclaro, the apologist for the expedition,
repeated these indictments, but added that the anticipated
riches of the Mutapa state induced many men, including

nobles, to participate in the mission.28

 

26H. V. Livermore, A New History of Portugal (Cam-
bridge, 1966), 153.

 

27Decision of the Lawyers, 23 January 1569, in RSEA,
III, 153-156. In terms of international law, the document
was inconsistent with even the existing precepts of juris-
prudence; see J. M. Chirenje, "Portuguese Priests and
Soldiers in Zimbabwe, 1560-1572: The Interplay between
Evangelism and Trade," IJAHS, VI (1973), 42-47.

28Monclaro, "Account of the Journey," 202-204. Even
the lawyers' judgment stipulated that the Munhumutapa was to
provide tribute in gold or land, but these issues were
hedged in legal terms rather than ones of personal profit;

99

Leading the entourage was Francisco Barreto, the
former Governor of India, who commanded a company of perhaps
one thousand troops. If retribution was his motive,

Barreto certainly displayed no anxiety in postponing the
struggle. The captain had a mania for wasting time. Al-
though he departed Lisbon in April 1569, Barreto decided

to lay over in Brazil rather than proceed directly to
Mozambique, as did his subordinate Vasco Fernandes Homem.
Barreto‘s delay was both unnecessary and unusual, since

few vessels sailing to the orient stopped in Brazil. By
1565, the crown had even prohibited India bound ships from
wintering there.29

The well rested commander finally reached Mozambique
in May 1570, but complained that his provisions were insuf-
ficient for the task and therefore the expedition would
have to await supplies from Lisbon. Instead of initiating
the conquest of the Mutapa kingdom, Barreto sailed up the

East African coast as far as Pate, ostensibly to collect

overdue taxes from local potentates, but it was little more

 

Decision of the Lawyers, 23 January 1569, in RSEA, III,
156. Other chroniclers indicate that gold provided the
single inducement for the expedition; see Couto, "Da Asia,"
357-358; Faria e Sousa, "Asia Portuguesa," 21.

29On this subject, see Alexander Marchant, "Colonial
Brazil as a Way Station for the Portuguese India Fleets,"
Geographical Review, XXXI (1941), 454-465.

 

100

than a sight-seeing excursion.30 The unscheduled expedition
so exasperated the king that he censured Barreto for aban-
doning his main purpose.31 The admonition did not, however,
dissuade the captain from his procrastination. He even
considered embarking for India, where local rebellions
threatened Portuguese rule, until the Viceroy convinced him
that his own fleet was sufficient to deal with the emer-
gency.32 Barreto seems to have underestimated the difficul-
ties posed by the Mutapa campaign, and while he tarried,
more than one hundred of his crew died of disease long
before reaching the battlefield.

Further delay was caused by debate over which route
to follow into the interior. The conflict indicates how
little da Silveira's death influenced the goals of the
mission. Originally, a council had decided to depart from
Sofala and march westward into Manyika to obtain gold in
addition to an alliance with the Chikanga, the Mutapa's

33

known enemy. It was argued that the direct Zambezi

route was more dangerous; yet, beginning at Sofala certainly

 

3oBarreto, for example, wrote to the king of the
beauty and fertility of Zanzibar which greatly impressed
him; A. N. T. T., Gavetas, 2-6-30, Francisco Barreto to King,
5 August 1571. For the complete tour, see Monclaro, "Account
of the Journey," 209-216.

31Couto, "Da Asia," 360.

321bid., 360-363.

33
1600, 157.

Ibid., 361-363; Axelson, Portuguese . . . 1488-

 

101

implied a more circuitous journey. Furthermore, unless the
goal was simply to conquer the Manyika mines, the Barreto
plan made little logistic sense. Padre Monclaro, therefore,
demanded that the troops follow an undeviating course into
Mutapan territory. Barreto capitulated, according to
Diogo de Couto, because he feared the wrath of the Jesuit
hierarchy and Dom Sebastiao.34 Barreto probably would have
been intimidated by the threat of another rebuke from the
king, particularly if he were not complying with his orders.
While the monarch was interested in both God and Mammon, it
appears that the Barreto expedition was concerned only with
the latter. Monclaro, however, was condemned by Portuguese
chroniclers for his political interference, although, in
reality, the padre only was attempting to retain the
mission's original purpose.35
Finally at the end of 1571, the expedition entered
the Zambezi and proceeded to Sena, where a base camp was
established. A Portuguese trader was sent to the Munhumutapa

to announce Barreto's arrival and to deliver the message

that the Portuguese wanted only friendship and trade; the

 

34Couto, "Da Asia," 363. Monclaro's report is
strangely silent on this dispute, perhaps because of the
mission's outcome.

35Faria e Sousa, "Asia Portuguesa," 21; Couto, "Da
Asia," 363. In his analysis of the entire expedition,
Vasco Fernandes Homem observed that the Jesuit advisers had
strict orders to proceed directly to Mutapa state; Ajuda,
51-viii-40, fol. 212v., Homem to da Sylva, 15 February 1576;
also see Jeronimo de Alcéntara Guerreiro, "Inquérito em
Mocambique no Ano de 1573," Studia, 6 (1960),.9-10.

102

troops were only there to clear away underbrush along the
commercial routes.36 Meanwhile, men and horses at Sena
were dying from disease-—either malaria or cholera--attri-
buted to "Moorish" poison. With Monclaro's encouragement,
these deaths were utilized as a pretext to annihilate the
local Muslim community.37 The violence was horrific: Muslims
were impaled alive, hacked to pieces with axes, blown apart
with cannons, and torn asunder while tied to bent trees.
The Portuguese seem to have relished the massacre, less for
its anti-Muslim sentiments, than for the opportunity to
loot the "Moorish" establishment.38

From Sena the enriched Portuguese column traveled
up the river where they attacked Samungazi's Tonga. The
latter were regarded by the Munhumutapa as within his sphere
of influence, although they retained local autonomy. At
some point, Samungazi had rebelled by cutting the route
between the Shona state and Sena, and therefore, the Portu-
guese assault was welcomed by the Munhumutapa.39 The Euro-

peans inflicted a large number of casualties, but the

 

36Monclaro, "Account of the Journey," 237.

37Ibid., 223-225; Couto, "Da Asia," 370-372; Faria
e Sousa, "As1a Portuguesa," 26. Monclaro admitted that he
persuaded Barreto to believe the Muslims were responsible,
although he knew many Portuguese deaths were due to other
causes.

38C0uto, "Da Asia," 370-371.

39Ibid., 373; Faria e Sousa, "Asia Portuguesa," 26;
Monclaro, "Account of the Journey," 247; Beach, "Shona
Empires," I, 21.

103

weakened force could venture no further into the interior
and returned to Sena in October 1572 without having reached
the Mutapa state.

Greeting the troops at Sena was a representative of
Negomo who complimented the commander on his victory over
the Tonga and indicated the desire of his people to have
peaceful commerce with the Portuguese. Barreto set three
conditions: the expulsion of Muslims from the state, the
acceptance of Christianity, and the grant of Mutapan gold
mines to Lisbon.40 Portuguese envoys returned with the Shona
ambassador to receive the Munhumutapa's answer.

Not wanting to remain idle when his talents were
needed elsewhere, Francisco Barreto returned during Decem-
ber 1572 to Mozambique Island. His purpose was to collect
more supplies for the campaign and to settle a political
tangle that had developed at the fortress. While at the
coast, Barreto was unaware that in the interior Negomo had
agreed to expel the Muslims, to provide the Portuguese with
unspecified mineral concessions, and to discuss the issue of
Christianity.41 Clearly, Negomo was not simply playing a
clever diplomatic game; he was willing to sacrifice the
Muslim traders as long as Portuguese merchants could supply
the same commodities. The allocation of mines, assuming

they were in peripheral areas and of typically low yield,

 

4oMonclaro, "Account of the Journey," 247.

411bid., 251-252.

104

also posed no threat. Only on the issue of Christianity did
Negomo disarm the original demand by making it a point of
mutual consideration. In return, the Portuguese had
silenced the troublesome Tonga revolt and consequently en-
hanced Mutapan authority in Southern Zambezia.42

In May 1573, after frittering away valuable time,
Barreto finally returned to Sena. To his horror, he found
many of his men dead, and not a single one remained in
good health. A few days later the commander himself
succumbed, and the remaining force of less than two hundred,
under the authority of Vasco Fernandes Homem, returned to
Mozambique Island to convalesce.

The Portuguese had failed to reach Negomo, but they
were not yet finished. Homem fitted out a new expedition
from the crews of recently arrived ships, and in 1574, they

43 The troops first

departed for the interior via Sofala.
entered the Teve state where their reception was hostile.
The Sachiteve feared that the Portuguese penetration of the
interior would lead to the elimination of his role as

middleman between Manyika and Sofala.44 Furthermore, the

movement of a foreign army through his domain was an obvious

 

42See Beach, "Shona Empires, I," 21-22.

43Axelson, Portuguese . . . 1488-1600, 161. Many of
the soldiers fled before departure when they became aware of
what possibly lay ahead of them.

 

44Couto, "Da Asia," 387-388; dos Santos, "Eastern
Ethiopia," 217.

105

infringement on his sovereignty and evidence of conquest.
For these reasons, food was hidden and wells were polluted;
the Portuguese retaliated by burning and killing, and even
imprisoning the Sachiteve's family.45 The defeat of Teve
not only opened the route to Manyika, but also eliminated
one of the Munhumutapa's most serious competitors. While,
ironically, Goncalo da Silveira may have been killed
because he was thought to be the Sachiteve's agent, the
Portuguese removed this threat in their effort to reach the
Munhumutapa and punish him.

When Homem's company reached Manyika, they were well
received, not simply because the Chikanga was aware of what
had happened in Teve, but also for the possible leverage the
Portuguese could provide in Shona politics. Even so, it is
unlikely that the Chikanga volunteered the location of the
richest gold deposits. Although it has been noted above
that mining was a marginal activity, the Manyika people
would not have encouraged the expropriation of their mineral
wealth, if clever subterfuge could be used to mislead the
Portuguese.46 Homem was extremely disappointed in the quan—
tity of available gold, and though his mining expert found

it to be of high quality, great numbers of workers and

 

45Couto, "Da Asia," 388-389; Faria e Sousa, "Asia
Portuguesa," 29; Ajuda, 51-viii-40, fol. 214, Homen to da
Sylva, 15 February 1576.

46See the comments of H. H. K. Bhila in his "Trade
and Survival," 13.

106

machines would have been necessary for the crown to turn a
profit.47 Instead of continuing the expedition, the Euro-
peans returned to Sofala where they once again sailed for
the Zambezi. It is unclear why Homem did not continue
northward, but if, as Professor Eric Axelson suggests, he
feared a Mutapa invasion of Manyika,48 it is further proof
that Portuguese claims of regional sovereignty were only a
sham. Since the new voyage along the Zambezi was to search
for silver mines, any higher motivations for the mission's
purpose had by then dissipated.

The next assault on the Zambezi proved no more
successful. During the latter part of 1575, the fever-
plagued soldiers struggled up the river. Ambassadors of
the shrewd Munhumutapa granted the Portuguese mineral
concessions over territories they did not control, and local
rulers accordingly refused to cooperate. The exasperated
Homem returned to Mozambique Island, leaving a small contin-
gent in the interior which was subsequently ambushed and
killed.49

The first attempt at military conquest of the hinter-

land had proved a costly and ill-conceived adventure. The

 

47Ajuda, 51-viii-40, fols. 214-214v., Homen to da
Sylva, 15 February 1576; Couto, "Da Asia," 389-390; dos
Santos, "Eastern Ethiopia," 218.

48Axelson, Portuguese . . . 1488-1600, 162.

 

49Faria e Sousa, "Asia Portuguesa," 30; dos Santos,
"Eastern Ethiopia," 282-285.

107

Barreto-Homem expedition was doomed by vacilation, delays,
military blunders, and illness. Its pitiful ignorance of
the interior terrain led to inappropriate preparations--
from cumbersome wagon trains to Barreto's own unsuitable
coat of armor (which he wore during the entire campaign).
More importantly, the Portuguese underestimated their enemy
and were duped in their effort to undermine him. In reality,
they little affected the course of sixteenth-century Muta-
pan history because most of the important developments were
of an African nature.50 To the extent that they were useful,
the Portuguese had become tools in the struggles between
Shona polities.

For the Portuguese, events beyond Southern Zambezia
proved to be more crucial during the last quarter of the
sixteenth century. Young King Sebastiao involved his
country in Moroccan internal politics, which culminated in
the Spanish captivity (1580-1640). The monarch led a force
of several thousand men to Ksar el Kebir where, on 4 August
1578, a battle was fought which was so badly organized that
it ranks among the greatest military disasters in history.
The Portuguese forces, including Sebastiao, were annihi-

lated.51 The throne was assumed by the elderly Dom Henrique,

 

50Beach, "Shona Empires, I," 22.

51For a précis of these events see Livermore, New
History of Portugal, 156-158; a detailed study is provided
in E. W. Bovill, The Battle of Alcazar (London, 1952).

108

Sebastiao's uncle, who died in 1580, leaving the way open
for the related Filipe II (1580-1598) of Spain to usurp the
Portuguese crown.

The Spanish monarch assuaged his new Iberian subjects
by promising to retain their officials and respect their
laws and customs. In theory, Lusitania remained a sovereign
unit, but Castile continuously encroached upon Portuguese
local autonomy. Furthermore, Lisbon's neutrality in Euro-
pean politics was abrogated; Spain's enemies now became
hers. Portuguese overseas territory was regarded as fair
game by the Dutch, French, and English, especially since
the empire was less secure than its Spanish counterpart.

From East Africa to the orient the vulnerable Estado da

 

India was attacked. Along coastal Mozambique, the fear of
these "Inimigos da Europa" was expressed by urgent requests
to rebuild defenses, prominent among which was the de-
teriorating fortress at Sofala. Local commerce scarcely
covered Sao Caetano's expenses, and it was not unusual for
officials to seek credit in order simply to purchase trading

52

goods. In fact, although captains were instructed to

divide their time equally between the fortresses at

 

52Regimento da Fortaleza de Sofala, 27 July 1574,
in Regimentos das Fortalezas da India, ed. P. S. S. Pissur-
lencar (Bastora-Goa, 1951), 195; Book of Expediture of the
Fortresses of Sofala and Mozambique, 7 November 1574, in
Documentos, VIII, 431.

 

 

109

Mozambique Island and Sofala, they generally preferred to
reside continuously at the more profitable northerly entre-
pot.53
Within the interior, trade between the Portuguese
and the Shona states continued, although the crown arrogantly
proclaimed that gold could be obtained easily by the con-
quest of the Mutapa state. A more cautious approach was
advocated, however, due to the lack of royal capital.54 The
fiscal situation notwithstanding, in 1585, one captain
proposed yet another major campaign. He even had the
temerity to suggest that a great expedition of 2,000 men be
launched, once again, along the Zambezi.55 Meanwhile,
Shona rulers maintained control of mineral resources and
still set the terms of trade in Southern Zambezia.
During the following decade, both Portuguese traders

and the Mutapa state were intimidated by offshoots of the

so-called Zimba migrations.56 In 1592 and 1593, the Zimba

 

53B. N. M., Ms. 3217, fol. 38v., "Livro das Cidades
e Fortalezas," c. 1582.

54A. N. T. T., Cartas dos Vicereis da India, no. 58,
King to Duarte de Meneses, n.d. (c. 1585); British Museum,
Add. Ms. 25,419, fol. 183v., Francisco Roiz Silveira,.
"Governo do Estado da India Oriental," 1578.

 

55Nuno Velho Pereira to King, 29 October 1585, in
Documentos, VIII, 529. The suggestion is especially peculiar
in light of the captain's revelation that a large proportion
of the Portuguese settler population of Mozambique Island
recently had been annihilated by a Makua attack.

 

56On the ethnic identification of what were appar-
ently diverse groups see Alpers, Ivory and Slaves, 46-53.

 

110

defeated garrisons from Tete and Sena when they impetuously
crossed the Zambezi.57 Later, the Munhumutapa's own domain
was invaded by two groups from the same northern region.
The Shona ruler was powerless to suppress these enemies
without aid. The Portuguese were obligated to help, not
only due to their tributary position, but also because
trade was suffering.58 The Munhumutapa's reliance upon
them, however, was to prove costly, especially when it
became obvious to his Lusitanian supporters that he had
little authority over surrounding Shona communities. In
the ensuing period, the Portuguese attempted to salvage the
Mutapa state for their own purposes, but found that their
plans were largely thwarted by the rise of the Changamire
empire.

The sixteenth-century optimism which marked the
Portuguese entrance into Southern Zambezia had been dampened,
but not extinguished. New schemes were being constantly
considered by the royal court which both exaggerated the
mineral wealth available and the ability of the Portuguese
to determine events in the region. And although King Filipe

complained that Sofala showed no profit for the crown, but

 

57dos Santos, "Eastern Ethiopia," 293—299; Couto,
"Da Asia," 404-410.

58Bocarro, "Decade," 361-364; Beach, "Shona Empires,

I," 23.

111

only expenses,59 he was as reluctant as his predecessors

to abandon the fabled port.

 

59A. N. T. T., Corpo Cronologico, 3-21-1, King to
Mathias de Albuquerque, 31 March 1593.

 

CHAPTER V

IN PURSUIT OF THE GOLDEN FLEECE,
1600-1800
Penury and Profit in Seventeenth
Century Southeast Africa

During the seventeenth century the fiction of Portu-
gal's domain became increasingly apparent. A barren treasury
and manpower shortage made the empire vulnerable on all
fronts to European rivals. In the Indian Ocean arena inter-
lopers denied Lisbon her traditional freedom of the sea,
while on land Portuguese settlements survived, largely
through the sufferance of the indigenous populations. De-
spite these circumstances and previous experiences, the
Iberian nation remained sanguine about the prospects for
Southern Zambezia.

Although southeast Africa was hardly the focal
point of Portugal's imperial interests, it was still widely
believed that strategically and economically the region was
significant. The battered fortress at Sofala, however,
remained a persistent source of anxiety and embarrassment.
For reasons more imagined than real, it was feared that the
Dutch, French, or English would appropriate the legendary

gold emporium if innovations were not undertaken. Yet, over

112

113

the following decades Lisbon's ability to retain this
settlement was neither the result of wise administration nor
military valor. Quite simply, her competitors directed
their energies at more lucrative pieces of real estate.
The Netherlands, in particular, waged fierce campaigns
against the Portuguese which took place on three continents
and seven seas.1
As early as 1601, Dutch pinnaces were sighted off
the Sofala coast, although they apparently did not threaten
the port. Of far more interest to the Lowlanders was
Mozambique Island. Jan Huighen van Linschoten, a Dutch
merchant, sailed there during 1585, and made an optimistic
report to his countrymen. He observed that the island
provided an excellent harbor for ships engaged in the India
trade and also served as a collection center for the gold
of Munhumutapa.2 Just after the turn of the century, the
Dutch reconnoitered the isle and even managed to seize two
vessels laden with ivory. It was, however, in 1607 and 1608
that the settlement was besieged, severly damaged, and almost

captured by Holland.3

 

1See C. R. Boxer, "Portuguese and Dutch Colonial
Rivalry, 1641-1661," Studia, 2 (1958), 7-42; idem, Portuguese
Seaborne Empire, 106-127.

 

 

2Jan Huighen van Linschoten, Discours of Voyages
into ye Easte and West Indies (London, 1598), 9-10.

 

 

3The literature on the Dutch attacks is prolific and
deserves a full length study in English. See, for example,
Ajuda, 51-iv-36, Antonio Durao, "Hystorya dos Cercos que

114

King Filipe III (1598-1621) long had urged the
Viceroy of India to refurbish his East African settlements
against such assaults, but little had been done because the
means were lacking. At Sofala, in particular, the fort
was in a frightful state: it lacked ammunition and soldiers
and needed considerable repair.4 The threat of sea attack,
though, seemed remote owing to Sofala's dangerous shoal
and the shallow depth of its channel. Without the aid of
local pilots, larger ships could not reach the port or
risked running aground.5 Additionally, with the exception
of the private trade by petty officials, commerce at the
enclave was of only trifling significance. In retrospect,
the crown's continued belief that Sofala was worth retaining
and would ultimately bring glory and wealth seems absurd.

Pecuniary considerations, however, did play a domi-

nant role in the expansive Portuguese aspirations for

 

Olandes Puzerao a Fortaleza de Mo ambique o Anno de 607 e
608," 1609; Academia das Ciéncias de Lisboa, Ms. Azul 384,
fols. 1-72, untitled, n.d.; Luiz Coelho de Barbuda, "Mili-
tary Achievements of the Portuguese" (1624), in RSEA, II,
362-377; Humberto Leitao, "Um Notével Episodio do Primeiro
Cerco de Mocambique,' Ultramar, XII (1971), 5—13; Axelson,
Portuguese . . . 1600-1700, 15-29. Important sources also
exist in Dutch.

 

4A. N. T. T., 231, l, fol. 61, King to Viceroy, 27
January 1607; A. N. T. T., 231, 2, fol. 32, King to Viceroy,
12 February 1608; A. H. U., Moc., Cx. 1, Cristovao de Moura
to Joao de Forjaz Pereira, 21 March 1608.

5Rezende, "Of the State of India," 403. The growth
in ship size during the seventeenth century was an important
cause of the increase in shipwrecks; see James Duffy, Ship-
wreck and Empire (Cambridge, Mass., 1955), 50-54.

115

Southern Zambezia. The faltering and increasingly dependent
Mutapa state was central to these desires. Gatsi Rusere
(c. 1589-1623), the ruling paramount, relied heavily upon
his European comrades for military aid. Foremost among
these men was a resident of Tete, Diogo Simoés Madeira
(Inhamacoto) who assembled a private army to aid the be-
leaguered Munhumutapa. In return, on 1 August 1607, Gatsi
ceded to King Filipe all the mineral wealth of his land.6
Although the difficulties posed in exploiting these mines
and in subordinating local Shona authorities made this
concession less momentous than it appears, the transfer did
reflect a continued diminution in Mutapan authority. Fur-
thermore, it initiated a new strategy by the Iberian
monarchy for penetration of the interior.

The blueprint for subjugation of southeastern
Africa included the utilization of established forts along
the Zambezi as offensive outposts and military support for
the tractable Munhumutapa.7 Shamelessly, an advisory
council even suggested to the king that conquest of the
mines would be more easily accomplished if the "heathens"

8

were converted to Christianity. These efforts, it was

 

6Bocarro, "Decade," 366-370.

7A. H. U., Mo¢., Cx. 1, Moura to Forjaz Pereira, 21
March 1608; A. H. U., Moc., Cx. l, Alfonso da Faza, "Sobre
as Minas de Manamotapa," 24 September 1609.

8A. H. U., Moc., Cx. l, Consulta do Conselho da
Fazenda to King, 24 September 1609. Motives, in this in-
stance, were clearly temporal rather than spiritual.

116

believed, would yield not only gold, but also rich quantities
of silver from the fabled mines of Chicoa.9 There is no
evidence, however, for the existence of mines within the
region. Much to the chagrin of the Iberians, the silver of
Chicoa was as illusory as the gold of Sofala. A series of
royal appointees with grandiloquent titles failed in a
futile quest which served to exacerbate local tensions as
well as to diminish commerce.10
Meanwhile, at Sofala, the old fortress remained
unrenovated and neglected throughout the 16205. The settle-
ment reportedly lacked a captain of the garrison or indeed
troops for him to command. In fact, parts of the fort's
walls had crumbled to the ground and gates were missing or
without locks. Failure to make the necessary repairs,

officials emphasized, would permit Dutch seizure of the

port and its trade.11 King Filipe IV (1621-1665) feared

 

9For reports on the legendary wealth of Chicoa see,
for example, Bocarro, "Decade," 412- 433; Jeronimo de Alcan-
tara Guerreiro and José de Oliviera Boléo, As Minas de Prata
de Chicoa (Lourenco Marques, 1944); Axelson, Padre Ant.D
Gomes, 32; Biblioteca Nacional de Lisboa, Fundo Geral, Cod.
176, fols. 21- 23; Henrique Bravo de Moraes,fi1nformacao
sobre os Rios de Cuama e Sofala," 1726; cf. W. G. L. Randles,
L' Empire du Monomotapa du XVe Siecle (Paris, 1975), 49- 50.

 

 

 

 

 

10The details of this chapter of imperial malfeasance
are described in Bocarro, "Decade," 381-435; Axelson, Portu-
guese . . . 1600-1700, 33-77.

 

11Archivo General de Simancas (Spain), Cod. 1552,
fol. 93, letter to King, 7 March 1620; A. N. T. T., DRI,15,
fols. 231- 232, Nuno Alvares Pereira to Viceroy, 26 June 1621;
Ajuda, 50-v-32, fol. 169, "Propuesta de Francisco Vaz d' Al-
mada," c. 1622; A. H. U. Moc., Cx. 1, letter to Lu1s da

117

that the loss of Sofala or the establishment of a rival
trading station nearby would provide an opponent the
opportunity to undersell Portuguese merchandise in the
interior.12 With Mozambican commerce already diminished,
competition would destroy Portugal's frail authority in
Southern Zambezia. For this reason, definite plans were
approved to dispatch men and munitions to fortify coastal
enclaves.

Under the command of Joao da Costa, a crew which
included carpenters, bricklayers, miners, and the fortifi-
cation engineer Bartolomeu Cotao was assembled for the
passage to southeast Africa. The laborers were engaged to
restore the various fortresses and to begin mineral exploi-
tation of the hinterland. Such efforts, it was believed,
would guarantee Portuguese sovereignty in the face of not
only Dutch antagonism, but also suspicious English activi-
ties.13 When da Costa's vessels finally embarked in 1633,
their departure was so sudden that, incredible as it may

seem, Cotao and several of his colleagues were accidentally

 

Silva, 9 February 1623; B. N. M., Cod. 2362, fols. 290-291,
Pedro d'Almeida Cabral to King, 15 November 1630.

12A. N. T. T., DRI, 25, fol. 150, King to Viceroy,

1628.

13F. U. P., EM, l6-A, 21/1-2, King to Viceroy, April
1632; F. U. P. EM, 17, 31/5, King to Captain of Mozambique
and Sofala, February 1633.

118

left behind.14 The absurdity of this incident was soon
overshadowed by the expedition's extravagant claims that
Southern Zambezia was rich in gold and silver beyond any
expectations.15 The anticipated wealth of the region con-
trasted though with the actual conditions there and, in
particular, at Sofala.

In a report commissioned by the Viceroy in 1634,
Pedro Barreto de Rezende presented a bleak picture of the
port. According to the author, the fort remained in dis-
repair with neither soldiers nor ammunition. A handful of
Portuguese lived in the nearby village and, with their
slaves, could be employed to defend the palisade. Such
slaves were Sofala's only Christian converts, and they
quickly shed the faith once they gained their freedom.
Rezende added that the Sachiteve, whose kingdom surrounded
Sofala, did not even consider this foreign theology and,
moreover, treated the settlement as a tribute-paying commu-
nity to aid him in his struggle against the Munhumutapa.
The Portuguese observer concluded that since the alleged

purpose of Sao Caetano was to protect the (nominal) trade

 

14The documentation on this expedition is confusing
and is probably a reflection of the events; see King to
Viceroy, 2 April 1632, in RSEA, 230-231; F. U. P., LM, 17,
26/5, King to Viceroy 1633; A. N. T. T., DRI, 31, £51. 87,
King to Viceroy, 10 December 1633; King to Viceroy, 7 March
1634, in RSEA, IV, 239; A. N. T. T., 231, 31, fol. 434, King
to Viceroy, 17 March 1634; Margarida to Viceroy, 28 March
1636, in RSEA, IV, 272—273.

15Axelson, Portuguese . . . 1600-1700, 99-100.

 

119

in gold, ivory, and ambergris from the covetous designs of
rival nations, the fort should be refurbished.16

The captain of Mozambique Island and Sofala, Filipe
Mascarenhas, added his own warning. He pointed out that
once the Dutch occupied the port they would gain control
of the trade with Manyika and threaten Portuguese interests
in the Mutapa state.l7 Yet, even before the receipt of
these reports, the monarch proposed a new plan for Southern
Zambezia.

King Filipe's design involved the massive colonial
settlement of the region in order to promote commerce and
mining. Besides soldiers and miners, marriageable women
(from the overcrowded poor houses) would be sent to assure
the growth of the Portuguese community. Additional clergy
could aid the scheme by proselytizing Africans and assuring
not only expansion of the gospel, but also Iberian rule. Of
course, thought was also given to refortification, and
Sofala, the king noted, needed immediate attention. Besides
repairs, he suggested a garrison of two hundred men and the

construction of another nearby fortress to protect the port.

 

16Rezende, "Of the State of India," 404-406.

l7Filipe Mascarenhas to King, 4 February 1635, in
APO, IV-II-II (Bastora, 1938), 119. Mascarenhas himself
abandoned his office after less than one year because of
alleged financial loses; see Diogo de Sousa de Menezes to
King, 17 February 1635, in RSEA, IV, 280; King to Viceroy,
24 February 1635, in RSEA, IV, 263; Archivo General de
Simancas, Cod. 1469, royal letter, 24 March 1636.

120

Filipe claimed his proposals were essential to the preser-

vation and defense of an important part of the Estado da
18

 

The extravagant design for Southern Zambezia was a
misguided attempt by Spain at external reform. Early in
the seventeenth century Castile had become aware of her
diminished status as a world power, but the remedies
officials considered were frequently misconceived.19 In
this case, Filipe's notions bore little relation to the
realities which were to be faced. While royal councils
debated who would fill important posts, it became impossible
during 1635 and 1636 to find rank and file settlers for the
dangerous undertaking. This development partly derived
from the doubts expressed by mining experts about the rich-
-ness of the region's mineral deposits. Another deterent

was the hostility of Portuguese sertanejos (backwoodsmen)

 

and local Shona to any further interference from agents of
the Iberian Crown. The project was finally abandoned in
1637 because India itself was in desperate need of relief
from external invaders.

Castile's fortunes were declining elsewhere as well.

In 1640, Lisbon took advantage of a revolt in Catalonia to

 

18King to Viceroy, 24 February 1635, in RSEA, IV,
253-262; A. H. U., Moc., Cx. 1, fragment of a royal letter,
24 February 1635.

19See J. H. Elliott, "Self-Perception and Decline in
Early Seventeenth-Century Spain," Past and Present, 74
(1977), 41-61.

 

121

declare her independence with massive popular support.20

Reluctantly, the Duke of Braganca accepted the Portuguese
throne and was crowned as Joao IV (1640-1656). One of his
first priorities was the arrangement of a non-aggression
treaty with the Dutch. In truth, the pact was so poorly
observed that the king constantly warned his Viceroy to be
vigilant for no funds were presently available for the
reinforcement of the Estado da India. Joao, however, para-
doxically commanded the Viceroy to construct new fortresses
in Mozambique, including Sofala:

a new fort shall be immediately built in this port

of Sofala, in whatever way appears to you most fitting,

and in such a manner that it may defend the entrance

to the said place, convey there for this purpose all

the requisite workmen and materials, that the work may

be carried out.
There was no explanation of how this edifice was to be
funded.

During the years following the restoration of Portu-

guese sovereignty, Southern Zambezia was generally neglected
and increasingly came under the influence of ambitious

sertanejos who garnered vast land holdings, prazos.22

 

 

20Idem, Imperial Spain, 1469-1716, 337-345; Luis
d'Oliviera Guimaraes, "O Povo na Revolu So de 1640," in 993-
gresso do Mundo Portugués (Lisbon, 1940 , VII, 139-146; also
see José M. da Cunha Saraiva, "Causas do Levantamento da
Nacao em 1640," in Congresso do Mundo Portugués, VII, 59-94.

 

 

 

21King to Viceroy, 24 February 1644, in RSEA, IV,

299.

22Portugal eventually institutionalized these estates

as Prazos da Coroa. According to the legal fiction, such

122

Sisnando Dias Bayao, for example, was an influential resi-
dent of Tete by nature of the large estates he had inherited
from his father-in-law. With his African retainers, he
assisted the neighboring Munhumutapa in local struggles,
as his predecessor Diogo Simoés Madeira had done. In 1644,
Bayao expanded his operations to Teve, where he was able to
restore the dethroned ruler and received in turn new tracts
of land. Subsequently, the Portuguese adventurer penetrated
westward and introduced stockades and troops to the Torwa
region. This accomplishment proved short-lived because
after his return to Sena, Bayao mysteriously died, and his
forces were withdrawn.23
Although details are sketchy, Bayao's efforts point
out the volatile situation within Southern Zambezia during
this period. Two significant features should be noted.
First, powerful settlers with private armies could carve

out small empires, which neither the Portuguese Crown nor

Shona authorities could prevent. In the case of Bayao, the

 

lands were granted by the crown and could be held for only
three generations. In reality, the Brazos were often the
domain of strongmen who were Portuguese only in name. The
subject has recently been investigated in detail, see Newitt,
Portuguese Settlement and Isaacman, Mozambique.

 

 

23C. R. Boxer, "Sisnando Dias Bayao: Con uistador
da 'Mae da Ouro,'" in Primiero Congresso da Historia da
Expansao Portuguesa no Mundo (Lisbon, 1938), III, 101-115;
Axelson, Pagre Ant.0 Gomes, 43; Barretto, "Report upon the
State" (1667), in RSEA, III, 504-505; Newitt, Portuguese
§ettlement, 53—54; of. H. von Sicard, "A Prop6s1to de
Sisnando Dias Baiao," Studia, 16 (1965), 179-187.

 

 

 

 

 

123

political situation allowed him to exploit antagonisms for
his own benefit (and, indeed, for his descendants who also
became great landowners). Additionally important, Bayao's
conquest of a portion of the western plateau exposed the
internal weakness of the Torwa dynasty in the-16403. Its
frailty may have resulted from the rise of the nearby Rozvi
state of Changamire. These developments were to have con-
siderable influence over the region in succeeding decades.
The Portuguese government clearly could not control
events within the interior nor, for that matter, even
manage her coastal settlements. Nowhere was the situation
more pathetic than at Sofala. By mid-century the Viceroy
pointed out that it was fruitless to renovate Sao Caetano
since it had no purpose except to raise Lisbon's expendi-
tures. Joao and his overseas council, however, rejected
this advice.24 The king not only commanded its preservation,
but also magnified its virtues:
the said fortress was the cause which led to the king
of Monomotapa and others becoming tributaries to my
crown . . . it serves to hold the neighboring petty
rulers in subjection, and to protect the commerce of
my subjects and the gold and ivory trade . . . the
Moors and heathens respect the residents of the rivers

on account of the same fortress, and it is not right
to abandon the Christians there.2

 

24A. H. U., Moc., Cx. 2, Consulta do Conselho

Ultramarino to King, 8 January 1652; King to Viceroy, 6
February 1652, in RSEA, IV, 317-318.

25

King to Viceroy, 6 February 1652, in RSEA, IV,
317-318.

124

Local Mozambican officials still considered razing the
fort; since communications were so tardy, they were unaware
of royal desires. From their on-the-spot observations, they
noted that trade at Sofala was almost non-existent, and
there seemed no threat from abroad.26 The Viceroy, however,
ordered Sao Cataeno conserved amidst a flurry of court
memos on the subject.27

The volume of correspondence dictated no change in
Sofala's floundering circumstances. Nicholas Buckeridge, a
representative of the English East India Company, visited
Sofala during the 16503 and reported that its only commerce
consisted in the collection of small quantities of ivory

28 He found that the fort was in ruins and

and ambergris.
that the settlement was defended by an ill-equipped rag-tag
garrison, composed of a few mulattoes and convicts (who

were often more conversant with Shona culture than Portugu-

ese). Padre Manuel Barreto noted in 1667, that Sofala was

 

26Axelson, Portuguese . . . 1600-1700, 133; Francisco
de Lima to Governor of India, 6 January 1652, in APO, IV-
II-II, 124.

 

27A. H. U., Moc. Cx. 2, Consulta do Conselho Ultra-
marino to King, 30 January 1654; King to Viceroy, 25 Febru-
ary 1654, in RSEA, IV, 327; F. U. P., EM, 24, 12/2, Bras de

Castro to King, 4 February 1655.

28John Jenson, ed., Journal and Letter Book of
Nicholas Buckeridge, 1651-1654 (Minneapolis, 1973), 32, 43.
Ivory and ambergris are also the only commodities mentioned
in roughly contemporary Portuguese documents; see Manuel
Cesar Pereira, "Discurso sobre a Conquista das Minas de
Monomotapa,” BSGL, VIII (1888-1889), 539; Barretto, "Report
upon the State," 479-480. I wish to thank Professor E. A.
Alpers for referring me to Buckeridge's records.

 

125

almost deserted owing to the insalubrity of the land and the
redirection of trade routes elsewhere. He caustically
pointed out that no priest was to be found in the local
church, nor, for that matter, any parishioners.29 It was a
place "where none are made Christians who are not born
so."30 In short, the site had no prospects, yet Lisbon
remained adamantly convinced of Sofala's value and suspi-
cious about the intentions of other European powers.

Sofala was, of course, not Portugal's primary con-
sideration among her far-flung possessions. In Southern
Zambezia, much attention remained centered upon the Munhumu-
tapa state and how to manipulate its new ruler, Siti
Kazurukumusapa (1652-1663). As previously noted, Christian-
ity was considered a useful instrument. Although Siti
accepted the faith, conversion for him was not simply an act
of fidelity to Lisbon, but also a means of obtaining
Portuguese support for his own interests.31 Siti's
succession had never been assured, and he apparently used
the baptismal rites both to obtain Portugal's approval and

to test the political loyalty of powerful nobles who were

 

29Barretto, "Report upon the State," 479-480.

3°"Eastern Ethiopia" (c. 1631), in RSEA, II, 437.
(This document is not to be confused with the 1609 study of
the same title by Joao dos Santos.)

31"Authentic Testimony of the Baptism of the Emperor
and King Manamotapa," 14 August 1652, in RSEA, II, 445-448;
King to Viceroy, 15 March 1655, in RSEA, IV, 330.

126

reluctant to follow his example.32 Unfortunately, the
Munhumutapa did not correctly calculate the role of the
EEEEQ holders, and he opposed them. They, in turn, forged
their own pact with Siti's rivals, who assassinated him.
The new appointee Mukombwe (1663-1692) proved not to be the

sertanejos' pawn; he showed himself to be a wily ruler,

 

suggesting that, in this case, it was easier to create a
puppet than to control one.33
Despite this situation, Barretto observed that the
time was ripe for the conquest of the Mutapa state:
If this king [Mukombwe] should rise in rebellion it
would be proper to conquer the whole of the kingdom,
and divide the lands among the Portuguese, by which
there would be great favours and rich appointments
for the deserving. Then the natives would dig as
much gold as the masters, the Portuguese, required.34
More realistically, the Viceroy saw that, conquest
aside, troops were needed at the Munhumutapa's court simply
to sustain Portuguese survival in the interior. He stated
that unless men were sent quickly their interests would be
shattered "like a house without a tenant, which falls in
time without being pulled down."35 The Viceroy, however,

had no troops to offer because of the disorders elsewhere

in the east. Moreover, Portugal was in the grip of its own

 

32Beach, "Shona Empires, I," 27-28.

33$§i§., 27; Barretto, "Report upon the State," 483.

34Barretto, "Report upon the State," 484.

35Viceroy to King, 9 January 1668, in RSEA, IV, 344.

127

national crisis. King Joao had died in 1656, and his
twelve-year-old heir, Afonso VI (1656-1668), was both
retarded and partially paralyzed. As regent, his mother
Queen Luisa attempted to steer a moderate course while
threatened by both Spain and the Netherlands. Young Afonso,
in turn, showed a decided disinclination to rule. His time
was devoted to various toys, nightly attacks on citizens,
and a keen interest in the local bordellos. Afonso's
companions, though, were ambitious and prompted him to take
the throne in 1661, and to exile his mother. He served as
monarch in name only and was controlled by court favorites.
Finally, in 1668, Afonso was forced to resign, losing both
his kingdom and his wife to his brother Pedro II (1668-
1706).36

As Pedro assumed power, it was remarked that the
Portuguese empire had been reduced to a few enclaves "with
some other fortresses and places of less importance . . .
either a memorial of how much we formerly possessed in
Asia, or else as a bitter reminder of the little which we
now have there."37 One of these relics was the captaincy
of Sofala. While Mozambique Island and other East African
settlements had been attacked in the 16605 by Omani Arabs,

there was no threat to 850 Caetano from them or other

 

36Livermore, New History of Portugal, 185-196.

 

37Manuel Godinho quoted in Boxer, Portuguese Sea-
borne Empire, 128-129.

 

 

128

aggressors. Sofala's trade was so stunted that, except for
the profits of a few unscrupulous or clever individuals,
nothing was to be gained there, and money had to be borrowed
from the fortress of Sao Sebastiao.38

In an effort to revive Mozambique's entire economy,
Pedro sanctioned freedom of commerce for all of his subjects
and encouraged further emigration, although he recognized
that in the interior, settlers were among the most disrup-
tive elements.39 Pedro's advisory council also urged the
development of Southern Zambezia and suggested mounting
another large and expensive expedition which would exploit
the region's wealth and establish new towns. The unsuccess-
ful record of past efforts in this regard did not discourage
this design, and, although perhaps two thousand people
departed in 1677, little is known of the result of this
expedition. Many Portuguese undoubtedly died on the voyage
while others probably fell victim to various epidemics

40

ravaging the interior. Surviving settlers did nothing to

bring gold and glory to Lisbon. According to an official,

 

38A. H. U., India, Cx. 29, anon. report, 20 January
1672; A. H. U., Moc., Cx. 2, Consulta do Conselho Ultramarino
to Prince, 23 February 1680.

39Royal order, 13 February 1672, in RSEA, IV, 351-
352; A. H. U., Moc., Cx. 2, royal order, 9 March 1672; Prince
to Viceroy, 22 September 1672, in RSEA, IV, 357-358; Prince

to Viceroy, 14 March 1675, in RSEA, IV, 365; Prince to Vice-
roy, 14 March 1675, in RSEA, IV, 367-368.

400n the ill-fated expedition, see Axelson, Portu-

 

129

"the married men who went to those parts were only a source
of expense to the royal treasury and of discredit to the
Portuguese nation, on account of their bad conduct."41

Truculent sertanejos posed a constant problem for authori-

 

ties, but during the last decades of the seventeenth century
the foremost threat to Portuguese sovereignty arose from a
different source.

Seemingly unnoticed by officials, the Khami state's
Torwa rulers had by (at least) the 16805 been superceded by
an aggressive Rozvi dynasty under Changamire Dombo. Al-
though this process is unclear, it has already been suggest-
ed that Bayéo's earlier success in the region (in 1644) was
due possibly to the societal transformation taking place
there. The Changamire state apparently evolved over a long
period and perhaps peacefully absorbed the related Torwa
community.42 While these points remain speculative, Dombo's
activities are unequivocal.

In 1684, Caetano de Mello de Castro, capitao-geral

 

and governor of Mozambique,43 led an expeditionary force to

 

41Gaspar de Sousa de Lacerda to King, 3 July 1682,
in RSEA, IV, 423.

42The possible relationship of the Changamire and
Torwa dynasties is discussed in Chapter III; of. S. I.
Mudenge, "Eighteenth-Century Portuguese Settlements on the
Zambezi and the Dating of Rhodesian Ruins: Reflections on
the Porblems of Reference Dating," IJAHS, x (1977), 384-393.

43Pedro's dual appointment in 1682, made Mello de
Castro the highest official in the region and subordinate
only to the Viceroy; see Prince to Viceroy, 9 March 1682, in

130

the Maungwe region west of Manyika. There he encountered
Changamire's army, and a day-long battle ensued. During
that evening Dombo set numerous camp fires around his foes,
leading them to fear that new reinforcements had arrived.
Mello de Castro's forces retreated, and although Changamire's
men possessed only bows and arrows, they were victorious
with this strategy.44
This success was followed by a Rozvi rout of the
Mutapan army which had attempted to conquer Dombo's home-
land during his absence. Changamire had demonstrated his
mastery of Southern Zambezia, a phenomenon which culminated
in the 16905 with what Portuguese historian Alexandre
Lobato has called "the great disaster."4S
For almost ten years there is a hiatus in the infor-
mation on Changamire's regime. Dombo may have spent the
intervening period outside the Portuguese sphere consoli-
dating his forces and gradually gaining control of the
plateau before launching his attacks on Lisbon's settlements.

Governor Mello de Castro complained in 1687 that, since his

arrival, virtually no assistance had been sent to Mozambique

 

RSEA, IV, 414. The title capitao-geral is often defined as
captain-general, but it is more precise to translate it as
the highest ranking military officer.

 

 

44Conceicao, "Tratado dos Rios de Cuama," 105-106;
Axelson, Portuguese . . . 1600-1700, 179; Gray, "Portuguese
Musketeers on the Zambezi,fi533.

 

’ 45Alexandre Lobato, Colonizagéo Senhorial da Zam-
bezia e Outros Estudos (Lisbon, 1962), 92.

 

 

131

from the Viceroy, and that soldiers were badly needed.46

As victim of the earlier clash with the Rozvi, Mello de
Castro realized the poor condition of Portuguese defenses
against serious attack.

Despite these unstable circumstances, Changamire's
assault on the trading fair of Dambarare in 1693, caught
the settlers by surprise, and all were slain.47 Dombo had
forged an alliance with the new Munhumutapa, Nyakunembire,
who succeeded to the throne despite Portuguese opposition.
In turn, the Munhumutapa encouraged the Rozvi attack on
Dambarare, an event which produced shock waves among the
other commercial outposts, including the Zambezi stations
of Sena and Tete. Changamire, though, focused his campaigns
in Maungwe and Manyika. Subsequently, the Rozvi-Mutapa
coalition weakened when Nyakunembire was dethroned by a
Portuguese supported competitor.

In 1695, Dombo besieged the fair of Masekesa in
Manyika, when a local captain refused to pay his annual tax

and assaulted the Changamire's emissaries. More than

 

46Caetano de Mello de Castro to King, 24 June 1687,
in RSEA, IV, 430. The king commanded the Viceroy to send
troops and munitions to Mozambique; F. U. P., LM, 52, 7/4,
King to Viceroy, 23 March 1687. ——

47The tales of wholesale destruction, including
disinternment of graves and the burning of the church, des-
cribed in Conceicéo, "Tratado dos Rios de Cuama," 106, are
exaggerated, as the excavated site provides no such evi-
dence; see P. S. Garlake, "Excavations at the Seventeenth
Century Portuguese Site of Dambarare, Rhodesia," Proceedings
of the Rhodesia Scientific Association, LIV (1969), 51.

 

 

132

revenge, Dombo sought to destroy Lisbon's military and
economic authority in Manyika in order to proclaim his own
suzerainty there. He accomplished this goal and appointed
a Chikanga to his liking, possibly the deposed Nyakunem-
bire.48

Meanwhile, the traders who had escaped from Manyika
fled to Sofala and were horrified to find the old fortress
provided no defense against Changamire. Various walls were
weakened or ruined and steps leading to the bastions had
collapsed; no stone was available for repairs. In addition,
the artillery was rusted and the firearms were useless
because they lacked stocks or bolts. Since Sofala was
considered to be the next Rozvi target, thought was given
to deserting the port.49

It was therefore with a collective sigh of relief
that the Portuguese learned of the Changamire's death in
1696. Although the state he created remained an important
feature in Southern Zambezia, the conquests were over. It
can be argued that Dombo was not concerned with destroying
every vestige of Portuguese society or pushing the Europeans

into the sea; he may have sought a new modus vivendi.

 

 

48See Manuel Galvao de Silva, "Diério das Viagens
Feitas pelas Terra de Manica" (1788), in Anais, IX-I, 327;
Bhila, "Trade and Survival," 14-18; Beach, i'Shona Empires,
I," 30.

49Ajuda 51-vii-27, fol. 32, Rafael de 5. Martinho
to King, 1697; Ajuda, 51-vii-34, fols. 45v.-46v., Carvalho,
"Papal da Noticia," 3 August 1698; Ajuda, 51-vii-34, fols.
48-49, Anténio Ferreira, "Barra de Sofalla," 12 August 1698.

133

Settlers with their personal armies had long disrupted
Shona society for their own pecuniary gains, and, in turn,
the Rozvi ruler employed force in his attempt to change
relations of authority and terms of trade. Although Dombo
did not live to see such developments, his successors did
establish new relationships with the Portuguese.

By the close of the seventeenth century, Portugal
was fully aware of how vulnerable she was, both to African
and European enemies in southeast Africa. Reports about the
mineral wealth of this region now attracted Lisbon's
primary concern, for her treasury was depleted and her
empire attenuated. Only with gold and silver, it was be-
lieved, could Portugal revive her diminished stature as a
world power and complete her mission of promoting Christian-
ity and trade. Despite its dismal history, Sofala retained
the glitter, if not the gold, of a legendary past and, though
neglected, it was not abandoned. In 1698, the captain of
Sao Caetano promised to rebuild the fortress, but like his
predecessors there is no evidence that he did anything.50
Increasingly, the reality of Sofala was subordinate to a
mythical image which looked not only to the past, but also

to hOpes for the future.

 

50Ajuda, 51-vii-34, fol. 46, Carvalho, "Papel da
Noticia," 3 August 1698; cf. Axelson, Portuguese . . . 1600-
1700, 191, which claims that the captain did rebuild the
fortress, although there is nothing to cofifirm this view.
Since writing this volume, Professor Axelson has indicated
to me that he may have been in error; interview with Eric
Axelson, 27 May 1974, Cape Town, South Africa.

 

134

Misfit of Empire: Eighteenth
Century Southern Zambezia

 

By 1700, only the veneer of Portuguese dominion

survived throughout the vast Estado da India. Many settle-

 

ments had been seized by European competitors or retaken by
local authorities. Lisbon's ill-equipped troops, within
their ruined and defenseless stockades, proved no match for
these forces. Remaining possessions were often little more
than decaying monuments to past aspirations.

Although the traditional ideals of prestige, pro-
selytization, and profit were reaffirmed by the crown, it
became increasingly difficult to enlist Portuguese citizens
for eastern missions. At times, men had to be impressed
and chained aboard ship. Expeditions frequently consisted
of a few seasoned sailors and a motley band of young boys,

vagabonds, and degredados.51 Those who survived the voyage

 

were hardly ideal settlers and, when they did not desert,
they scandalized the government with their fractious be-
havior.

Local Portuguese officials and clergy also contri-
buted to the crown's embarrassment by their unauthorized
activities and breach of royal edicts. Many were tyrants

who manipulated policies for their own profit, resorting,

 

51C. R. Boxer, "The Carreiga da India, 1650-1750,"
Mariner's Mirror, XLVI (1960), 36-54, esp. 40-41; idem,
Portuguese Seaborne Empire, 313-315.

 

135

when necessary, to murder.52

The king was powerless to do
more than require cursory investigations by his Viceroy.

Despite the prevalent decadence of the Estado da

 

12912! the lands of Southern Zambezia retained their fabled
appeal for Lisbon. King Joao V (1706-1750) described the
region as "the richest that I possess in my royal domin-
ions."53 The legendary wealth of King Solomon's mines con-
tinued to hypnotize the monarch despite decades of failure
and pessimistic reports. As usual, new royal plans were
formulated to refurbish dilapidated fortresses and resus-
citate commerce.

Along the Mozambican littoral the expansion of
Dutch and British trading companies posed a growing threat
to these schemes. During the 17205, Joao was particularly
fearful of foreign intentions in Delagoa Bay. He also
expressed consternation over the safety of his emporium at
Sofala,54 although from a commercial viewpoint, the old
port with its dangerous anchorage and minimal economic

value offered little advantage to any merchant.

 

52See, for example, the account of a Mozambican
captain who executed a local ruler and replaced him with a
more tractable candidate: King to Viceroy, 15 March 1702, in
RSEA, V, 7; King to Viceroy, 19 January 1704, in RSEA, V,
13. Such incidents were not isolated, as I have previously
noted.

53King to Viceroy, 16 August 1710, in RSEA, V, 23.

54King to Viceroy, 16 April 1721, in RSEA, V, 115-
116, 118; King to Viceroy, 12 April 1723, in RSEA, V, 130;
Anténio da Silva Rego, O Ultramar Portugués no Seculo XVIII
(Lisbon, 1967), 99.

 

136

According to a 1722 census, there were only twenty-

55 It appears that few of

six Portuguese settlers at Sofala.
these residents had actually been born in Europe, and their
descendants were usually of mixed blood. The fort itself

was garrisoned by a handful of mulattoes or degredados who

 

often lived outside the ruined stockade because there were
no barracks. The fortress was, in fact, constantly threat-
ened by inundation or collapse since sections of its walls
had crumbled. By this time, the empty regimental church
was even being utilized as a cattle kraal.56

While Sofala was of little importance in the designs
of Portugal's European rivals, it also had a diminished
role in local African politics. At the beginning of the
eighteenth century, the Changamire dynasty remained the
predominant force within Southern Zambezia. Rozvi attention
was, however, focused upon the nutritive commercial and
mineral regions in Manyika and along the Zambezi. Within
the Dande area, the Changamire's influence was demonstrated
by his efforts to control Mutapan succession, which often
stimulated debilitating factional disputes.57 Additionally,

he sought to control the remunerative Zambezi trade. It was

 

551722 census report, in APO, IV-II-II, 90.

’ 56Alexandre Lobato, Evolucao Administrativa e
Economica de Moqambique (Lisbon, 1957), 37, 163.

 

 

57Biblioteca Nacional de Liaboa, Fundo Geral, Cod.
176, fols. 21-23, Moraes, "Informacao sobre os Rios;" Beach,
"The Mutapa Dynasty," 9.

 

137

under his protection that, in 1716, the Portuguese estab-
lished the fair of Zumbo at the confluence of the Zambezi
and Luangwa. Commerce was strictly supervised by the Rozvi:
from Zumbo only authorized African traders were permitted
to penetrate into the rich southwestern plateau in search
of gold and ivory. This policy, at least theoretically,

prevented the disruptive introduction of sertanejos into

 

the Rozvi state; it was not an economically hostile reaction
to Lisbon. In fact, more than once, the Changamire aided
the Portuguese in keeping Zumbo operating and the trade
routes open.58
The harmony at Zumbo was also created in Manyika,
where Portugal was permitted by Rozvi authorities to re-

59 In this

establish a trading station early in the century.
region, settlers were not only prohibited from commercial

expeditions, but also had to pay a large annual tax and were

 

58On Zumbo see Newitt, Portuguese Settlement, 75-
79; J. D. Clark, "The Portuguese Settlement at Feira,"
Northarn Rhodesia Journal, VI (1965), 275-292, esp. 275-
280; :Memorias da Costa d'Africa Oriental," 200-204; "Des-
cripcag da Capitania de Monsambique, Suas Povoacaes, e
Producoes" (1788), in Relacoes de Mogambique Setecentista,
402-403. The Changamire'sitrade monopoly was never as
thorough as implied by his restrictions, see Nicola Suther-
land-Harris, "Trade and the Rozwi Mambo," in Pre-Colonial
African Trade, eds. Richard Gray and David Birmingham
(London, 1970), 257-259; Mudenge, "Role of Foreign Trade,"
384-387.

 

 

 

 

 

59Viceroy to Antonio Cardim Troes, 21 January 1719,
in RSEA, V, 50.

138

prevented from mining.60

In return, the Portuguese sought
protection from local Manyikan authorities, who increasingly
followed an independent course. Chikangas attempted to
reassert their control over the internal economy by super-
vising all merchants and concealing mineral resources while,
at the same time, carefully avoiding actual conflict with
Rozvi agents.61 Although unsettled conditions were an
impediment, Manyikan trade was proving profitable.

This prosperity, however, only contributed to the
continued stagnation of Sofala. Under Changamire super-
vision, mineral wealth from Manyika was carried to the
Zambezi fairs, and 850 Caetano was deprived of any of this
trade. While Sofala had never been the opulent center
described in legend, the old Chikanga state had, at least,
given some credence to the tale by filtering small quantities
of gold to the fort in the past.

Sofala's real staple had been its modest ivory
trade. This commerce also disappeared after the Portuguese
established Inhambane in 1731. The new settlement was
located close to elephant country and became the primary

entrepét for the ivory traffic.62 Thus deprived of any

 

6ONewitt, Portuguese Settlement, 74; Bhila, "Trade
and Survival," 17, 19: idem, trans., and ed., "A Journal of
Manoel Galvao da Silva's Travels Through the Territory of
Manica in 1790," Monumenta, VIII (1972), 82.

 

61Bhila, "Trade and Survival," 18-28.

~ 62"Memorias da Costa d'Africa Oriental," 211-213;
Joao Baptista de Montaury," Mocambique, Ilhas Querimbas,

139

commercial purpose, Sofala no longer interested the Rozvi
and was reduced to a pawn in the turbulent relations of

a handful of aggressive sertanejos and the Teve state.

 

Although the Sachiteve seems to have regarded
Sofala as a dependency, during the first half of the
eighteenth century, he was himself the client of Portuguese
settlers who acquired vast prazos in his domain through
subterfuge or threat. In 1735, for example, the merchant
Joao Pires was killed when he ventured within Teve territory.
To retaliate, Pires's wife raised a private army and marched
into the interior. The Sachiteve, reportedly, was so
intimidated that he granted the woman a large tract of
land in compensation for her husband's death.63 Lisbon

admitted that such sertanejos had "become so powerful that

 

not only do they free themselves from the condition of
subjects, but on many occations act with the tyranny of

petty kings."64 The crown lacked the resources to thwart

 

Rios de Sena, Villa de Tete, Villa de Zumbo, Manica, Villa
de Luabo, Inhambane" (c. 1778), in Relacaes de Mogambique
Setecentista, 372; P. R. 0., Admiralty 172269, 'Cap. 0 35',
fol. 24, Jofio Juliao da Silva, "A Brief Description of the
Fort and Town of Sofalla and Adjacent Lands and Possessions
Belonging to the Captaincy," 1824; Alan Smith, "Delagoa Bay
and the Trade of South-Eastern Africa," in Pre-Colonial
African Trade, 278-279.

 

 

 

 

53p. R. 0., Admiralty 1/2269, 'Cap. 0 35', fol. 6,
da Silva, "Brief Description." On other grants see, F. H.
Ferao, "Account of the Portuguese Possessions within the
Captaincy of the Rios de Sena" (c. 1810), in RSEA, VII, 374-
375.

64Conselho Ultramarino to Viceroy, 13 February
1727, in RSEA, V, 147.

140

its countrymen from pursuing their own ambitions. At the
same time, the Sachiteve was inhibited by factional conflicts
and African foes from controlling the European intruders.

By 1740, Portuguese even resided within the Teve court
serving allegedly as scribes to aid in matters of diplo-

65 It is apparent though that their

macy and commerce.
activities did nothing to alleviate Sofala's torpor. One
official described the fort as so debased that it could
easily be subjugated by any enemy. Its defense consisted of

fifteen soldiers--mostly dagredados.66

 

Although Sofala appears an anachronism, it was only
the most egregious example of Portugal's fortunes in South-
ern Zambezia. Even the more prosperous settlements were
subject to closure by African states or the vexations of
local prazeros. As a remedy, schemes were proposed for the
transport of Indian settlers to southeast Africa because it
was believed that they could be more easily constrained
than degenerate Portuguese. Undoubtedly, though, the most
bizarre suggestion was that impoverished Irish peasants be
transferred to Southern Zambezia. It was emphasized that
such colonists would not only be more dependable and indus-

trious than Lisbon's misfits, but also that they were good

 

65A. E. X. Soares, Descripgao da Villa de Sofalla

 

66A. H. U., Moc., Maco 32, Pedro de Rego Barretto to

King, 10 November 1745.

141

Catholics who would promote the faith in the interior.67
Yet, rather than engaging in such fanciful emigration plans,
a more drastic action was undertaken by the crown.

In 1752, southeast Africa's administrative tie to
Goa was severed, and it was placed under the command of a
governor-general stationed at Mozambique Island. The idea

of dividing the two distinct regions of the Estado da India

 

had been formulated, at least, as early as 1634, but was
still being rejected by the monarchy more than a century

68 The reasons for the long standing refusal are not

later.
entirely clear, but it apparently was believed that the

Estado da India had an organic unity. This view, according

 

to one eighteenth-century official, was absurd because
Southern Zambezia and the orient had nothing in common. He
adduced:

The relationship should be seen as the same as that

of Brazil with Angola, which latter being some distance
from that great continent has always remained quite
separate resulting in an increase of its commerce and
establishment which would never have happened had
Angola been dependent on Brazil.69

 

67J. P. Monteroy Mascarenhas to Viceroy, 6 April
1746, Archivo das Colonias, I (1917), 154-156; Boxer,
African Eldorado, 11-12, 15 n. 30.

 

 

68A. H. U., Mog., Cx. 1, Consulta do Conselho de

Estado to King, 16 November 1634; Jeronimo de Alcéntara
Guerreiro, "Quadros da Histéria de Mocambique," Mocambique,
68 (1951), 12.

 

69M. D. D., trans. and ed., Account of a Joaanay
Made Overland fram Quelimane to Angoche in 1752 by Francisco
Raymundo Moraes Pereira (Salisbury, 1965), 4-5.

 

142

Under the authorization of King Jose (1750-1777) the separ-
ation was completed. Its real instigator, however, was his
influential minister, Sebastiao de Carvalho e Mello--the
Marquis de Pombal (1699-1782).7O But even the energies of
Pombal could not stem the drift of Southern Zambezia into
undisturbed lethargy.

While the Viceroy was no longer responsible for East
Africa's internal squabbles, he retained control over its
commerce with India.71 This trade, however, was increasingly
in the hands of Gujarati merchants, Banyans, not their hap-
less Portuguese colleagues. Mozambique and its hinterland
therefore remained economically subordinate to Goa and
dominated by Indian itinerant merchants. The king confirmed
this relationship when he opened East Africa to all of his
Asian subjects, and permitted them to "trade there as they
think fit, in the manner most advantageous to them." His

only stipulation was that they pay the required royal

 

70Pombal was responsible for various, extensive
domestic and external reforms during his tenure as virtual
dictator, see C. R. Boxer, "Pombal's Dictatorship and the
Great Lisbon Earthquake, 1755," History Today, V (1955),
729-736. Specific information on his reforms relevant.to
southeast Africa is provided in Lobato, Evolucao Admlnls-
trativa, and Fritz Hoppe, A Africa Oriental Portuguesa no
Tempo do Marques de Pombal, 1750-1777 (Lisbon, 1970).

 

 

 

71Boxer, African Eldorado, 12; King to Francisco de
Mello e Castro, 23 April 1752, in Relacoes de Mocambique
Setecentista, 45.

 

 

 

143

duties.72 Although Sofala was generally neglected by these
long-distance traders, the governor-general noted that an
occasional Banyan appeared at the port with the hope of
exchanging his cloth for ivory. The deplorable conditions
there and in Teve, however, precluded any extensive deal-
ings.73
Future aid to Sofala or any ultramar settlement
was inhibited by the devastating earthquake which struck
Lisbon on 11 November 1755. Consequently, during the
second half of the eighteenth century, Southern Zambezia
was dependent upon its own limited resources for survival.
Depopulation and decay had reduced several settlements and
now threatened them with the prospect of becoming ghost
towns. Governor-General Pedro de Saldanha de Albuquerque
reported that these trading centers lacked sufficient
settlers and soldiers to retain even the crumbling facade
of Portuguese sovereignty. As a minimal measure, he sug-
gested that hardy Asian women be dispatched to southeast
Africa to provide mates for Lusitania's dwindling male

population.74

 

72King to Viceroy, 5 April 1755, in RSEA, v, 224-
225; Law Relating to the Commerce of Mozambique, 10 June
1755, in RSEA, V, 228-230.

73A. H. U., Moc., Ma 0 38, Francisco de Mello de
Castro to King, 20 November 753; A. H. U., Moc., Cx. 4,
Francisco de Mello de Castro to King 22 November 1753; A. H.
U., Moc., Cx. 5, Francisco de Mello de Castro to the King,
10 November 1754.

74A. H. U., Mo ., Cx. 6, Pedro de Saldanha de

Albuquerque to King, 2 December 1758.

144

By 1762, Sofala was nearly under water, and the
garrison had been reduced in four years from thirty to
six.75 Of the few residents there, most were Africans or
mulattoes; the leading merchant was a Goan.76 The growth
of the slave trade in Mozambique during the 17605 did pro-
vide Sofala with some economic stimulus. Captives from
this area had the reputation of being stronger and more
intelligent than elsewhere, and they therefore demanded a

greater price.77

It appears though that the number exported
during this period was never great; nonetheless, the once
fabled Sofala had now become little more than a debased
enclave for the slave trade. Small quantities of ivory and
gold were also occasionally traded there, but the amounts

were so insignificant that thought, once again, turned to

abandoning the site.78

 

75Ignacio Caetano Xavier, "Noticias dos Dominios
Portuguezes na Costa de Africa Oriental," 26 December 1758,
in Relacées de Mocambique Setecentista, 155; Pedro de
Saldanha de Albuquerque to King, 30 December 1758, Archivo
das Colonias, IV (1919), 78; A. H. U., Moc., Cx. 8, Pedro
de Saldanha de Albuquerque to King, 13 August 1760; "Memor-
ias da Costa d'Africa Oriental," 205.

 

’ 76"Memorias da Costa d'Africa Oriental," 205; Hoppe,
Africa Oriental Partuguesa, 111; A. H. U., Moc., Cx. 10,
3530 Pereira da Silva Barba to Captain of Sofala, 18 April
1763; "Rellacam das Moradores Portuguezes que Assitem em
Monssabique e Seos Destrictos" (c. 1767), in Anais, IX-I,
167.

 

77"Memorias da Costa d'Africa Oriental," 208.

78A. H. U., Moc., Cx. 10, Ignacio de Mello [Alvim]
to King, 5 April 1267; "Instruc ao que o IIImo e Exmo Sr.
Governador e Capitao General Ba tazar Manuel Pereira do Lago

145

Lisbon tenaciously refused to relinquish Sao Caetano,
but continued to find it difficult to enlist settlers and
soldiers for Southern Zambezia. The Viceroy, for example,
rejected a royal edict to recruit sepoys for East Africa:
none would volunteer because "white men," who took appoint-
ments in Mozambique, returned without profit and vowed never
to go back.79 The interior was racked by inter-Shona con-
flicts, civil wars, and the imperial pretentions of local
prazeros. These disruptions were, in small part, the result
of the developing external slave trade. More importantly,
they reflected the inability of any single faction to
exert overwhelming authority for any extended period.

As early as 1730, the Viceroy claimed that the once
feared Changamire was "without power."80 Although this view
was exaggerated, it has been suggested above that Rozvi
authority was circumscribed and constantly threatened not
only by Portuguese ambitions, but also by Shona politics.
The Mutapa state, for example, was shaken by civil war from

1759 until 1763, strife which spilled over into the

 

Deo a Quem lhe Suceder Neste Governo" (1768), in Relagaes

de Mogambique Setecentista, 322-323. In 1762, it was re-
ported that the Zambezi settlements exported twelve times as
much gold as Sofala, see Hoppe, Africa Oriental Portuguesa,
244.

 

79Viceroy to Francisco Xavier de Mendonca Furtado,
28 January 1764, in RSEA, V, 240-241.

80Viceroy to King, 2 January 1730, in APO, IV-II-II,
172.

146

neighboring Tete prazos and threatened Changamire's merchants
at Zumbo.81 A few years later, fierce succession disputes
erupted in Teve, Manyika, Barue, and even within the Rozvi

82 The unstable conditions led to the abandonment

state.
of Portuguese estates or the forging of frail alliances
between prazeros and Shona communities in the interests of
self-preservation. These features were hardly an inducement
to future European settlement, and Mozambique's forts be-
came the dumping grounds for Lisbon's felons.

At Sofala, in particular, Sao Caetano appeared more

like a prison than a fortress. The governor-general finally

requested that no more degredados be sent to the settlement

 

because they lacked respect and discipline. In addition,

they did nothing for the crown and further degraded the

t.83

por These soldiers and the few mulatto settlers had

gone "native" (caffralmente) and lived in extreme poverty

 

in grass huts near the ruins of the fort. The officers were
little better off. Most could neither read nor write

Portuguese, and even the captain was regarded as an

 

81Mirada, "Monarchia Africana," 110-111; Isaacman,
Mozambique, 108-109; Beach "Mutapa Dynasty," 12.

 

82A. H. U., Diversos, Moc., Cx. l, Joao Juliao da
Silva, "Memoria sobre Sofala," 1844; P. R. 0., Admiralty
1/2269, "Cap. 0 35', fol. 25, da Silva, "Brief Description:"
"Instruccao . . . Baltazar Manuel Pereira do Lago," 323;
Isaacman, Mozambi ue, 111-112; Sutherland-Harris, "Trade and
the Rozwi Mambo," 257.

83A. H. U., Moc., Cx. l4, Baltazar Manuel Pereira
do Lago to King, 14 August 1772.

147

insignificant appointee.84 A contemporary English commenta-
tor concluded that "the inhabitants of the town and kingdom
of Sofala are a mixture of Mahometan Arabs, idolatrous
caffres and bad Portuguese Christians."85

During the last decade of the eighteenth century,
Southern Zambezia was struck by one of the periodic droughts
and consequent famines which plagued the region.86 The
abject conditions at Sofala were exacerbated by the danger
of starvation during 1793 and 1794. Settlers voiced the
fear that if food were not available soon, they would have
to leave the port. Nothing was arriving from the interior
because of crop failure and the closure of commercial

routes. Merchants who ventured outside the settlement were

 

84Biblioteca Geral da Universidade de Coimbra, M5.
3069, fol. 372 v., "Vida do Illustrissimo e Excellentissimo
Senhor Sebasti5o José de Carvalho e Mello, Marquez de Pom-
bal," 1781; J. J. Nogueira de Andrade, "Descripc5o do Estado
em que Ficav5o os Nego ios da Capitania de Mossambique" '
(1790), Archivo das Co onias, I (1917), 83- 84; A. H. U.,
Moc., Cx. 28, "Mappa da Forca da Tropa da Guarni 50 da Fort-
aleza de S50 Caetano," 20 July 1793; M. D. D. Newitt, "Igna-
cio Caetano Xavier's Account of Portuguese East Africa," in
Conference of the History of the Cenaaal African Peoples
(Lusaka, 1963), no pagination; idem, Portuguese Settlement,
207.

 

 

 

85C. B. Wadstrom, An Essay on Colonialism (London,
1794), 128.

 

86Isaacman, Mozambique, 67. Oral traditions relating
such frequent ecological disasters are common; interview with
Mzee Makandara Nova Sofala, Mozambique, 16 May 1974; also
see Franz Boas and Simango Kamba, "Tales and Proverbs of the
Vandau of Portuguese South Africa," Journal of American
Folk-Lore, XXXV (1922), 162.

 

 

 

148

robbed.87 Furthermore, in contrast to its earlier open-door
policy, Teve increasingly prohibited Portuguese penetration
into its territory. There were no further land grants to

sertanejos, and trade was so restricted that one Sachiteve

 

was assassinated by his subjects when he opened new gold
mines. The ruler's action, it was thought, would expose
Teve to further Portuguese ambition and interference.88

Despite the very real threats in southeast Africa,
Lisbon was preoccupied with imaginary fears that Napoleonic
France would seize her Mozambican ports. For unknown
reasons, it was believed that along with Inhambane and Del-
agoa Bay, Sofala would be a target.89 Ironically, even
after the Peninsula War had begun, commercial relations
along the East African coast, between French and Portuguese,
remained cordial.90

The fate of Sofala, however, was not affected by

any new European interlopers. One official described S50

 

87A. H. U., Moc., Cx, 29, Anténio Alberto Pereira to
King, 21 July 1793; A. H. U., Moc., Cx. 29, Joao de Paz e
Temes Brinha to King, 3 July 1794; A. H. U., Moc., Cx. 29,
Joao de Paz e Temes Brinha to King, 27 July 1794.

88See Xavier, "Noticias dos Dominios Portuguezes,"
155. This example, of course, indicates the extent of
internal dissention in Teve. At the end of this century the
state was threatened by new African enemies which affected
the Sofalan trade, see below Chapter VI.

89M. V. Jackson Haight, European Powers and South-
East Africa (London, 1967), 123.

 

 

9oPaul Mazery, "Mozambique and the Napoleonic Wars,"
Studia, 37 (1973), 387-399.

149

Caetano quite simply as "a small fort, much neglected."91
Although not forgotten, the decayed settlement only barely
managed to survive. The myth of Sofala, nevertheless, con-
tinued to prosper in the mind of Portuguese officials. Its
real future, however, was to be determined not in Lisbon,

but in the hinterland of Southern Zambezia.

 

91Montaury, "Mogambique, Ilhas Querimbas," 370.

CHAPTER VI

REQUIEM FOR GOLDEN SOFALA

After navigating along the Mozambican coast, Captain
James Prior observed: "notwithstanding the progress of
science and discovery, this country remains, in 1812, in
that happy state of obscurity in which it was found by Vasco
da Gama."1 His somewhat fanciful picture was subsequently
erased by events which transformed Southern Zambezian
society and increased Mozambique's economic significance
during the turbulent nineteenth century.

Long before Prior's voyages, the forces of change
had already been set in motion. As early as 1700, Tsonga
peoples had begun migrating northward from their homeland
which spread across both sides of the Limpopo. This initial
expansion led to the conquest and incorporation of the
Tonga-Shona communities located south of Inhambane and
threatened the Portuguese settlement itself until the

arrival of reinforcements forestalled the invasion. By

 

1James Prior, Voyage along the Eastern African
Coast (London, 1819), 24.

 

150

151

mid-century, however, the Tsonga had expanded throughout
southern Mozambique and even conquered portions of the Sofala
hinterland.2

Although these disruptions greatly contributed to
economic and political instability in Southern Zambezia, it
is difficult to be precise in describing the Tsonga migra-
tions. The problem is partly semantic, as well as histori-
cal. Descriptions are obscured by the tendency of Portuguese
officials of the period to refer to all unfamiliar African
groups as Landins.3 This unfortunate term lacks either
ethnographic or linguistic applicability, and even its

etymology is unclear.4 While it is safe to assume that

 

2About these events see Smith, "Peoples of Southern
Mozambique," 574-578. Literature on the Tsonga is particu-
larly confusing because of the lack of agreement on what
these people should be called. Among the various designa-
tions applied, the name Tonga has posed the greatest prob-
lems, since it is also employed in reference to a distinct
group in the region of Inhambane, as well as an unrelated
Ndau dialect of Shona; see M. Wilson, "The Sotho, Venda,
and Tsonga," in The Oxford History of South Africa, eds. M.
Wilson and L. Thompson, I (New York, 1969), 176; Doke,
Southern Bantu Languages, 22-23.

 

3For example, see Smith, "Peoples of Southern Mozam-
bique," 560; "Noticias de Moiambique e Suas Conquistas," 11
November 1754, in Lobato, Co onizacao Senhorial, 158; Xavier,
"Noticias dos Domininios Portuguezes, T145; Francisco de
Santa Tereza, "Plano e Relac5o. da Bahia Denominada de Lour-
enco Marques," 6 August 1784, in Caetano Montez, Descobri-
mento e Fundag5o de Lourengo Marques (Lourenco Marques,
1948), 164; C. J. dos Reis e Gama, Reposta das Questoens
sobre os Cafres (1796), ed. Gerhard Liesegang (Lisbon, 1966),
22.

 

 

4A spirited, if not always rational, debate has
existed over the origin of this appellation. Perhaps, the
most amusing suggestion was put forth by W. F. W. Owen who

152

these so-called Landins were actually Tsonga migrants, the
specific movements of various clans is conjectural. Despite
these questions, it may be concluded that the incursions
set the stage for the more destructive Nguni invasions to
follow.

South of the original Tsonga homeland, revolutionary
events were beginning to take shape during the last part
of the eighteenth century. In the Natal region, a combina-
tion of population growth, land pressure, and trade seem-
ingly contributed to the rise of the Zulu state which
reached its military apogee under Shaka (c. 1787-1828).5
His elaboration of earlier battle techniques and organiza-
tion and the development of a warrior society resulted in

the conquest or flight of neighboring peoples. Once again,

 

postulated that Landim (sing.) was a corruption of "L 'Indi-
an." A less improbable possibility suggests the term is
related to the Rjonga kuland a: to follow, or land a: slave;
but even these explanations are not entirely satisfactory.

A lively discussion is provided in Caetano Montez, "As

Racas Indigenas de Mocambique," Mo ambi ue, 23 (1940), 53-
66, esp. 58-64; F. Toscano, "Sobre os Indigenas Portugueses
ao Sul do Zambeze," RSEM, XLIV (1941), 144; José Luis
Quint5o, Gramatica de Xironga (Landim), (Lisbon, 1951), 8-9;
A. C. P. Cabral, Pequeno Dicion5rio de Mogambique (Lourenco
Marques, 1972), 59; W. F. W. Owen, Narrative of Voyages to
Explore the Shores of Africa, Arabia, and Madagascar (London,
18337, I, 276.

 

 

 

 

5This major event is outside the scope of this
study, but see J. D. Omer-Cooper, The Zulu Aftermath (London,
1966) 25-33; Alan Smith, "The Trade of Delagoa Bay as a
Factor in Nguni Politics 1750-1835," in African Societies in
Southern Africa, ed. L. Thompson (London, 1969), 171-189;
Max Gluckman, "The Rise of a Zulu Empire," Scientific Ameri-
ggg, CCII (1960), 159-169; idem, "The Individual in a Social
Framework: The Rise of King Shaka of Zululand," Journal of
African Studies, I (1974), 113-144.

 

 

 

 

 

 

153

Southern Zambezia was invaded from below the Limpopo. This
second shock wave was composed of related Nguni groups who
fled from Shaka's armies. Their onslaught spelled an end
to Shona domination of the interior and Portuguese safety
along the coast.

Although the information on Nguni infiltration is
far superior to our knowledge of the Tsonga, much confusion
still exists. Portuguese commentators (and later their
British counterparts) called all Nguni, Vatuas, and rarely
distinguished among them.6 They further managed to confuse
the issue by employing the terms Landim and Vatug as though
they were synonymous, despite the fact that the Tsonga and
Nguni belonged to different language groups. It is there-
fore often impossible to identify, with assurance, the
movements and actions of the numerous refugees from the
Zulu conquests.

It appears that initially four Nguni groups pene-
trated southern Mozambique in about 1821. During the next
several years these clans managed to clash not only among
themselves, but also with the local population and the

Portuguese as well. The ultimate victor in these squabbles

 

6Vatua may have been a Portuguese corruption of the
Shona vatorwa, meaning strangers or foreigners. Variations
of the term in other Bantu languages are common, such as,
the designation of so-called pygmies as "Batwa." The issue
is by no means settled; see M. Hannan, Standard Shona Dic-
tionary (London, 1961), 411; M. D. W. Jeffreys,wThe Batwa:
Who Are They," Africa, XXIII (1953), 45-54; W. H. J. Rangely,
"The Angoni," Soc1ety of Malawi Journal, XIX (1966), 70.

 

 

154

was Soshangane Nquamayo (Manicusse), whose Gaza empire
eventually covered much of Southern Zambezia.

Soshangane and his followers originally settled
near Lourenco Marques, but in 1827 they swept into the
lower Limpopo Valley, where they resided until 1835.7
During this eight-year period, the Gaza had demonstrated
their mettle by defeating both the Ngoni warriors of
Zwagendaba Jere and an invading Zulu army, and now were
prepared to expand northward. Occupying the Sofala hinter-
land, however, were the rival Nguni forces of Nxaba Msane
(Muava) who dominated local Shona communities and had
forced the Portuguese to abandon their Manyika fair.8
Nxaba, though, was no match for Soshangane's troops, who
attacked in about 1836. The vanquished Nguni either
joined the Gaza or fled toward the Zambezi to unite with
the Maseko Ngoni.9

It should not be construed that Soshangane's con-

quests were solely the result of aggressive Gaza sub-

 

7W. F. W. Owen, "The Bay of Delagoa," 1 May 1823,
in RSEA, II, 470; Gerhard Liesegang, "Nguni Migrations
between Delagoa Bay and the Zambezi, 1821-1839," IJAHS, III
(1970), 323; idem, "Aspects of Gaza Nguni Hisotry 1821-
1897," Rhode51an History, VI (1975), 2.

 

8A. H. U., Diversos, Moc., Cx. 1, da Silva, "Memoria
sobre Sofala."

9Liesegang, "Nguni Migrations," 328-333; Ian Linden,
"Some Oral Traditions from the Maseko Ngoni." Society of
Malawi Journal, XXIV (1971), 69; M. S. Alberto, Os Angones
os fitimos Povos Invasores da Angonia Portuguese," Mocambique,
27 (1947), 88.

 

 

155

 

 

 

Limpopo R.

\ - O" .Q
-.~ Tete :
o. 0 o
e. \ |
: '-
. Sen:
2
.1. :0 |‘ .o.:°':o
~'. :."'-I. u" ~°° ~-
..o'¢: ’3 .. ova... e
‘ o... .-_ .. ' .0 ' O
.-:.I.c.-'. '.: -'-‘:-‘~'o ‘w"°-".
...':'c.....‘a .... I..'OQ. 0..
u. g I _ o o . . O I
\C. o . . . I 0 e . '
. *. .I . - o .0: ‘0‘, O. o o
O.".' . -.§ ‘. a e I ..
.. , l , ‘lh‘ . I ‘ . e:
' .:s:"' "I". ' . ”I .
'. . . . a”. .. .I. . . .
". .5 .. us '. '
'7"; :I "so ' :'
.,_ ... . . o 0.. -
'.~~( ‘ ,.‘ ' 't‘, ' . of! la
‘ a O o o . . . 0' .
.'o.'. :0 , .' . .‘ t .,
‘. ... u. l ...' 1 .° u' '.
.'.~:-'..' '.‘ .'. K "a ' - l
‘ O. . ‘ ..... ' .
(‘1'... ' . o. :
.. u . . ... I- e. .
' :' . . . ..' o ’ o z.-
' '0: ' 'I '
1 $0 “ M 'R}.
’ I . O .'
\ " I. ' . ‘ . ’. .
n ' " n .. ‘-
’ f... 2 . . 30". ‘ o ‘ 1)"
a ‘ h" . - ' .r o
\' .‘e l. ' '5 ' o" ' H. ' '-.
‘ e . 0‘ ..° 0 o
u . . ' O I . ' J a. e
e e. ' . ° I '. .
Q ‘ ‘. e .. . . ‘0...- "
Q ‘.e.: . 'I . . ’ ' v...-
f . . . ‘I e
’ ..I.I O ' ‘ O
_ . Inhambane
. 'f a-
e \ I g l
: 8 .- ‘ o 0'
.‘.9\ . ‘ O
.u’ I o
.. h - —
U. '

: ourenco Marquee

|

l ”a

5-. (fifigfirtent of Gaza-Nzuni
”” Influence

 

0 miles 250

 

Figure 5. The Gaza Empire

 

156

imperialism; in reality, his own homeland was itself being
threatened by enemies. In 1833, the Zulu came uncomfortably
close when they assailed Lourenco Marques.10 The following
year the captain of Inhambane launched an expedition against
Soshangane which, though unsuccessful, may well have en-
couraged the Gaza to march northward.

Until recently, it was generally accepted that
Soshagane's arrival heralded not only Nxaba's demise, but
nearly Sofala's as well. In October 1836, the town was
stormed by large numbers of Nguni troops. Although Soshan-
gane would seem to have been the culprit, Professor Gerhard
Liesegang has suggested that Nxaba's warriors were respon-
sible.11 Perhaps, before retreating from the Gaza, Nxaba
made a final sally against the Portuguese; the documentation
is unfortunately ambiguous. There is, however, no question
about the intensity of the assault.

According to eyewitness accounts, the village was

sacked and looted by an army of 4000 Nguni. Although slaves

 

10Gerhard Liesegang, "Dingane' 3 Attack on Lourengo
Marques in 1833," JAH, X (1969), 565; A. H. U. , Moc., Maco
1-A, notebook of Antonio José Nobre, 1833.

11Liesegang, "Nguni Migrations," 326. Liesegang
bases his view on the consumate diplomatic skill displayed
by Soshangane vis- 5-vis Europeans. In an interview with a
Portuguese envoy in 1840, the Nguni leader claimed that he
was innocent of any assaults or robberies. He added that
his military reputation was so highly valued that "other
people" assumed his name in order to inspire fear. Years
later, Soshangane continued to claim that he had not attack-
ed Sofala; Caetano dos Santos Pinto, "Viagem de Inhambane as
Terras de Manicusse em 1840," Arquivo das Colonias, I (1917),
272; A. H. M., Cx, 4-75, Maco 10, Filippe da Costa Correa
to G. Oliveira Rego, 25 March 1856.

 

157

working nearby fields were slain, the bulk of the population
managed to retire to the old fortress, which scarcely
offered adequate protection. Surprisingly, no serious
attempt was made to attack the fort, as the invaders were
satisfied to depart quickly with their booty of cloth and
cattle. Since the town was not set ablaze, the Portuguese
feared that the Africans would return and catch them unpre-
pared once again. In response, several frightened residents
embarked on a small vessel for the safety of Mozambique
Island.12

Sofala appeared in imminent danger of surrender
not just to the Nguni, but also to the more destructive
forces of nature. The first decades of the nineteenth
century present an uninterrupted history of neglect and
degeneracy. The harbor could admit only the smallest ships
because it was nearly blocked by shifting sandbanks, and
even lighter vessels required local "Moorish" pilots to
guide them to a safe anchorage.l3 The appearance of any
craft, however, was a rare event because even the port's
nominal trade had been reduced to a trickle. When the

routes to the interior were not impeded, the Portuguese

 

12A. H. U., Diversos, Moc., Cx. 1, da Silva, "Memoria
sobre Sofala"; M. S. Alberto, "Carta de Sofala," BSEM,
XXVIII (1959), 130, 135-136.

 

13Prior, Eastern Coast of Africa, 25; Salt, Vo a e
to Abyssinia, 18-19; William Fisher to Albemarle Bertie, 2
September 1809, in RSEA, IX (London, 1903), 8-9; A. H. U.,
Moc., Cx. 51, Joaquim do Rosario e Monteiro to Governor-
General, 7 April 1809.

 

158

lacked both the inclination and resources to take advantage
of their opportunities. One captain, for example, complained
that despite the Sachiteve's concession of a gold producing
area, none of his compatriots could be swayed from their

lethargy to establish an official fair there.l4

Contrary
to M. D. D. Newitt's view, there are no indications of a
"vigorous revival of Portuguese activity in Sofala in the
early years of the nineteenth century."15

Only the expansion of the slave trade (briefly)
sustained Sofala, although this commerce led to an economic
boom in other Mozambican ports. The British Parliamentary
Abolition Act of 1807 encouraged Brazilian, Cuban, and
French slavers to prowl the East African coast where London's
cruisers were less active. At Quelimane, for instance,
slaves accounted for only about 17 percent of the export
revenues in 1806; fifteen years later, they represented 85
percent of the total.16 A burgeoning business was conducted
at other entrepéts, where captives were obtained as a result
of the disruptive Nguni migrations and the consequent evo-

1ution of large scale slave trading prazos.17 The British

even accused Mozambique's governor-general of abetting the

 

.14

Fer5o, "Account of the Portuguese Possessions,"

15Newitt, Portuguese Settlement, 210.

 

16Isaacman, Mozambique, 93.

 

17gggg, 85-93, 115-117, 124-153; Newitt, Portuguese

Settlement, 220-226.

 

 

159

18 Strictly speaking,

traffic while hindering their efforts.
of course, the Portuguese official was within his rights,
since Lisbon did not outlaw the commerce until 1836 (al-
though, even afterwards, the decree was flagrantly violated).
Despite the increase in the export of black cargoes
from Southern Zambezia, Sofala initially profited little.
Few slaves were traded at Sofala because of its poor
anchorage for passing ships and its distance from the
Zambezi prazos which filtered much of the human traffic.19
Captain W. F. W. Owen, an ardent abolitionist, explored
Sofala in 1823, and found it to be of "trifling importance
even to the Portuguese." Although he suspected that local
Muslims maintained nearby estates as slave depots, he found
no evidence of a flourishing commerce.20 Owen added that
the port certainly did not live up to its inflated historical

reputation:

it was with much curiosity that we looked forward to
our arrival at Sofala and with much disappointment as

 

18P. R. O., Admiralty, 1/2270, 'Cap. O 35,‘ W. F. W.
Owen to J. W. Croker, 3 June 1825.

19Prior, Eastern Coast of Africa, 25. Research in
the Portuguese archives indicates that there was little
commercial interest in Sofala during the early part of the
nineteenth century. Only an occasional small vessel sought
to make the voyage: A. H. U., Moc., Cx. 40, "Pedido para um
Passaporte," 18 May 1802; A. H. U., Moc., Cx. 48, "Pedido
para um Passaporte," 27 April 1807; A. H. U., Moc., Cx. 68,
"Relac5o de Equipagem da Escuna," 28 June 1821; A. H. U.,
Moc., Cx. 76, "Pedido para um Passaporte," 10 January 1825.

 

20F. R. O., Admiralty 1/2269, W. F. W. Owen, "The
Portuguese Settlements and Dominions of the Eastern Coast
of Africa," 15 April 1823.

160

at the total failure of our expectations. Instead of
what the fancy pictured, remains of past grandeur and
opulence, frowning in decay and falling gradually to
dust, we found but a paltry fort and a few miserable
mud-huts, the deserted abode of poverty and vice.21

Owen's bleak description was confirmed by Portuguese
commentators as well. The crumbling stockade was largely

protected by a handful of degredados or their progeny. Since

 

the troops resided outside the fortress and received their
pay in cloth, most devoted their efforts to trade, in order
to survive. It is not surprising that they were Africanized,
both racially and culturally. Soldiers took Shona wives
and accepted indigenous customs, so much SO that local
medicine men were consulted about planting and harvesting,
as well as about various illnesses.22 Sofala was such a
backwater village that, over long periods, there were no
doctors, teachers, or priests to minister to the impover-
ished population. Inevitably, though, there were always
sufficient officials prepared to serve there.

For the unscrupulous officer, Sofala provided
suitable scope to engage in private trade, which could
assure a handsome stipend. In order to turn a profit,
officials, however, had to combine larceny with evasion of

duty--an effort which never seemed to tax appointees. In

 

21Owen, Narrative of Voyages, I, 318-319. Also see
the comments of Lieutenant Boteler, one of Owen's officers;
Thomas Boteler, Narrative ofa Voyage of Discovery to Africa
and Arabia (London, 1835), I, 347-348.

 

 

 

22A. H. U., Diversos, Moc., Cx. 1, da Silva, "Memoria
sobre Sofala"; Soares, Descripgao da Villa de Sofalla, 40.

 

161

fairness, it should be pointed out that Lisbon was unable
or reluctant to furnish her representatives with adequate,
regular wages. Cynics aptly remarked that payments from
the royal coffers were made, "late, in part, or never."23
With little choice, magistrates had to become merchants.
The concept of office holding as a duty with fixed juris-
diction and compensation was alien to this period.24 Posts
of various grades were sold to the highest bidder and re-
garded almost as personal property. Only those officers
who grossly overstepped the liberal limitations of propriety
were castigated. Thus, when Luis Araijo Rosa, commandant
of Sofala, absconded with the treasury in the 18203, his
superiors agreed that he should be imprisoned because he
had not served with "appropriate decorum."25

It is unlikely, though, that the rapacious Captain
Rosa secured much capital in his unsuccessful effort to
fleece the settlement. During his time, Sofala's only sig-
nificant export was rice, which partially explains the
governor-general's refusal to approve repairs for the unre-
munerative factory. The absence of regular commerce, lack

of sufficient manpower, and obvious negligence of royal

 

23Boxer, Portuguese Seaborne Empire, 77.

 

24See H. H. Gerth and C. Wright Mills, trans. and
eds., From Max Weber: Essays in Sociology (New York, 1958),
196-244.

 

25A. H. U., Moc., Cx. 77, letter to Governor-General,
11 April 1826.

162

appointees made the re-edification of S50 Caetano an unlikely

prospect.26 Conditions continued to degenerate to the

point that in 1829/30, Sofala once again faced starvation.

A combination of drought, locust infestation, and Nguni up-

heavals in the hinterland made it impossible for the port

even to feed itself, let alone export any foodstuffs.27 An

effort to rescue Sofala initially resulted in disaster, when

two supply ships were wrecked along the perilous coast, and

only later did the necessary staples arrive.28
Despite such dreary circumstances, it was believed

that restoration of trade with the interior and revival of

agriculture would insure Sofala a brighter future.29 In

 

26A. H. U., Mo ., Maco 3, petition of Pascoal Pires
de Carvalho, 14 April 828: A. H. U., Moc., Ma 0 5, Governor
of Inhambane to Ministro do Ultramar, 15 February 1828; A
H. U., Moc., Maco 21, Francisco Miguel Rodrigues Nunes to
Governor-General 5 April 1831; "Descripc5o sobre o Estado de
Sofalla na Costa Oriental d'Africa" (c. 1829), Archivo das
Colonias, I (1917), 185-192.

 

27A. H. U., Diversos, Moc., Cx. 1, da Silva, "Memoria
sobre Sofala; A. H. U., Moc., Maco 10 letter of Jer6nimo
Francisco Ferreira, 14 September 1829; A. H. U., Moo 9., Mago’
11, petition of Pascoal Pires de Carvalho, 24 December 18
A. H. U., Moc., Maco 12, Ant6nio José da Maia to Governor-
General, 3 March 1830; A. H. U., Cod. 1424, fol. 5, Paulo
José Miguel de Brito to Francisco Miguel Rodrigues Nunes, 30
May 1830; A. H. U., Moc., Maco 12, Francisco Miguel Rodrigues
Nunes to Paulo José Miguel de Brito, 5 August 1830.

28A. H. U., Moc., Maco l4, Francisco Miguel Rodrigues
Nunes to Manuel da Silva Goncalves, 22 April 1830.

29A. H. U., Mo ., Maco 21, "Nove Documentos Relacion-
ados com a Reocupac5o o Territ6rio Bandire," 1831; A. H. U.,
Moc., Maco 25, Paulo Jose Miguel de Brito to Governor-Gener-
a1,4 September 1831; A. H. U., Cod. 1424, fols, 8- 9, Paulo
José Miguel de Brito to Francisco Miguel Rodrigues Nunes,

163

1836, any optimism was dashed by the devastating Nguni
attack which demonstrated the settlement's extreme vulnera-
bility. Ironically, in spite of the disaster, the coinci-
dental Portuguese abolition of the slave trade actually
increased Sofala's commercial importance. The port's
inaccessibility now made it an ideal haven for the illicit
traffic. Such poor harbors were difficult for British
vessels to survey because of the unpredictable currents

and shifting shoals along much of the littoral.

By the 18403, Sofala therefore had become one of
the more important sanctuaries for slavers. Meanwhile,
Mozambican officials not only disregarded Lisbon's dicta
and London's censure, but also actively assisted in in-
creasing the commerce.30 With large fortunes to be made,
these officers were reluctant to prosecute slave merchants
whose venality they shared. One governor-general audaciously
proclaimed: "Portuguese laws are for me a dead letter after
I have passed the Cape of Good Hope."31 His view was con-
firmed by a mid-century census of Sofala which revealed

968 slaves in the town--a figure representing nearly 85

 

29 December 1831; A. H. U., Cod. 1424, £015, 18-20, F.
Ant6nio José da Maia et al. to José de Amante de Lemos, 21
November 1832; A. H. U., Moc., Maco 30, "Noticia sobre
Maxanga," 4 October 1833. '

30Haight, European Powers and South—East Africa,
257-283; James Duffy, A Question of Slavery (Oxford, 1967),
41-42 0

 

 

31P. R. 0., Foreign Office 84/908, Portuguese offi-
cial quoted in Pakenham to Clarendon, 18 April 1853.

164

percent of the total population.32 This total certainly
exceeded the demands of local agriculture and documents
Portugal's failure to prohibit the trade. Sofala's new
found prosperity, however, proved ephemeral as a result of
the Gaza state.

After Soshangane's defeat of Nxaba, the Portuguese
feared another foray would be made against Sofala as soon
as the Gaza finished raiding the interior.33 The Nguni
leader, however, had more grandiose plans. By 1845,
Soshangane had subjugated all of southern Mozambique and
transformed its towns and prazos into tribute paying
dependencies.34 He never sought to drive the Portuguese
into the sea, but to reduce them to vassalage as a subordi-
nate community from whom valuable commodities could be
collected.35

From the Shona viewpoint, the growth of the Gaza
state represented the latest in a series of far-flung
empires that had engulfed their communities over the cen-

turies. Soshangane largely usurped the role of his Rozvi

 

32Soares, Descripg5o da Villa de Sofalla, 25.

 

33A. H. M., Gav. 4-63, Cod. 1742, fol. 81 v., un-
titled report, 29 August 1837; A. H. U., Mo ., Maco l-A,
Governor-General to Count of Bomfim, 16 November 1840.

34A. H. U., Diversos, Moc., Cx. 1, da Silva, "Memoria
sobre Sofala"; Randles, Empire du Monomotopa, 64; Liesegang,
"Aspects," 2-3; Isaacman, Mozambique, 122-123; J. P. R. Wal-
lis, ed., The Zambezi Expedition of David Livingstone, 1858-
1863—(London, 1956), I, 34, 73, 11,2340-341.

 

 

 

35Omer-Cooper, Zulu Aftermath, 59.

 

165

predecessors rather than establishing a new political
system. Although some Shona were incorporated into Nguni
style age-regiments and introduced to Zulu military techni-
ques, they were never totally assimilated. Fighting units
remained segregated, with fresh recruits often forced to
man the front lines. Gaza society was, in fact, so
stratified that acculturated Shona were prevented from
rising to positions of power, although those who refused to
adopt Nguni customs faced ridicule.36 For the most part,
the Gaza were satisfied with extracting periodic payments
of ivory, cattle, or cloth from the Shona and attempting

to restrict trade.37 The numerical inferiority of the
Nguni to their subjects meant that, in reality, many of

the hinterland states enjoyed a considerable degree of
autonomy.

In establishing his reign, Soshangane had mixed
diplomacy with military threat in relations with both the
Portuguese and Shona. He continued this policy by not over-
extending his empire, and thereby avoided conflict with

Voortrekkers south of the Limpopo and with the Ndebele west

 

36Ibid; Rennie, "Christianity, Colonialism and the
Origins of Nationalism," 135-164, esp. 144-147, 154; P. R.
Warhurst, "The Scramble and African Politics in Gazaland,"
in The Zambesian Past, eds. E. Stokes and R. Brown (London,
1966), 48; David Chanaiwa, "A History of Nhowe before 1900"
(unpublished Ph.D. dissertation, University of California at
Los Angeles, 1971), 133.

 

37H. H. K. Bhila, "Manyika's Relationship with the
Portuguese and the Gaza-Nguni," Rhodesian History, V11 (1976).
35-36 0

 

166

of the Sabi.38 Perhaps, even Soshangane's unsuccessful
efforts to prohibit slave dealing within his domain39 reveal
less altruism than a sagacious understanding of how this
trade could undermine his rule. Soshangane's heirs proved
less astute.

The death of the Gaza leader in 1858 precipitated
a lengthy succession struggle between two of his sons,
Mawewe and Mzila.40 The Portuguese-supported candidate,
Mzila, emerged as victor, but hardly proved obsequious. In
1862, he crossed the Sabi and settled just north of the
river. The local Shona found, to their chagrin, that Mzila
not only exacted considerable tribute but also greater
allegiance than had his father.41 He left no doubt about
his view of the Portuguese when he remarked that they were

"women" who only settled along the coast, in order to be

 

381n 1836, Afrikaners initiated the Great Trek to
escape British rule. In turn, they forced the Ndebele under
Mzilikazi to migrate across the Limpopo and establish their
kingdom (c. 1840) within the core of the Rozvi realm. Al-
though the Ndebele state represented the other major Nguni
polity in Southern Zambezia, it had little impact on Sofala
or its hinterland and has been thoroughly analyzed elsewhere.

39Gerhard Liesegang, Beitr5ge zur Geschichte des
Reiches der Gaza Nguni im sfidiichen Mogambique 1820-1895
TCologne, 1967), 244.

 

 

40Although the struggle did not actually terminate
until 1865, Mzila's victory was assured by 1862. Most of
the fighting took place in the Limpopo Valley and outside
Lourenco Marques. For an outline of events see Liesegang,
"Aspects," 4-8; J. Steven-Hamilton, The Low Veld (London,
1929), 170-176.

 

41Liesegang, "Aspects," 8; Jo5o Tavares de Almeida to
King, 7 August 1863, Archivo das Colénias, II (1918), 101.

 

167

able to flee quickly.42 After meeting the new Gaza ruler,
a British emissary noted: "The Portuguese he neither fears
nor hates, but regards . . . as necessary appendages to his
dominions to fetch and carry goods for him; and, in fact,
as being sort of enlightened Tongas [Tsonga]."43

Mzila demonstrated his scorn by robbing Sofalan
merchants and preventing the fort from obtaining needed
supplies. He negotiated with the Portuguese only when it
suited his purpose and, as officials complained, Mzila's
price for ivory was double or triple the current rate.44
Sofala's small population could not endure these exactions
and feared an imminent attack. Thus, plans were formulated
to abandon the settlement and retreat to the safer site of
Chiloane Island, fifty miles to the south.

It appears that even at this point, it was with
some reluctance that the transfer was initiated in about
1862. The sparsely populated Chiloane long had been con-

sidered a plausible alternative to the deteriorating post

 

42Warhurst, "Scramble and African Politics," 50;
St. Vincent Erskine, "Third and Fourth Journeys in Gaza, or
Southern Mozambique, 1873-1874 and 1874 to 1875," JRGS,
XLVIII (1878), 32.

43St. Vincent Erskine quoted in Liesegang, Reiches
der Gaza Nguni, 246.

 

4?A. H. M., Gav. 4-63. Cod. 1460, fol. l9v., "Actas

da Delagac5o da Junta da Fazenda, 1862-1866"; Francisco de
Paula Castro Domingues, "Relatorios," Boletim Official do
Governo Geral da Provincia de Mocambique, 12 (1871), 49.
There are no indications of further slave dealing at Sofala
in this period, testifying to Mzila's interference, in-
creased British activity, or lowered demand.

 

 

168

of Sofala. As early as 1518,45 Lisbon had utilized the
island as an anchorage and subsequently had surveyed it for
possible colonization. Yet, only the assaults of Mzila
prompted its actual settlement. After years of gradually
transporting people and goods from Sofala to Chiloane, the
change was finally completed in 1865.46 The fortress of
S50 Caetano retained a small detachment of poorly supplied

degredado troops, who were often isolated within the

 

stockade by the frequent Nguni attacks.47

Despite depressing past circumstances and its pre-
sent malaise, the mythical glory of Golden Sofala was not
forgotten. In 1871, the governor-general claimed: "The
district of Sofala, in my opinion, is the richest in all
the province."48 Fifteen years later, the former commandant
of S50 Caetano urged that the fort be immediately repaired

before this "glorious monument" of Lusitanian history was

 

45Lobato, Expans5o Portuguesa, III, 319. Even prior
to the Portuguese arrival, the island had served as a Muslim
trading station, see Dames, Book of Duarte Barbosa, I, 5;
Balsan, "Route de l'Or," 103; Owen, Narrative of Voyages, II,
411.

 

 

 

46A. H. M., Gav. 2-8, Cx. 4-76, Maco 21, letter of
8 March 1864; A. H. M., Gav. 4-63, Cod. 1460, fols. 42 v.,
53v.-54v., 60, "Actas da Delagocao."

47A. H. M., Gav. 2-8, Cx. 4-76, Ma 0 24, "Relac5o
Nominal de Todos os Degredados," January 1 65; A. H. M., Gav.
2-40, Cx. 4-82, Maco 2, report of 30 November 1876; G. L.
Sullivan, Dhow Chasing in Zanzibar Waters (London, 1873),
232-233.

 

48Paula Castro Domingues, "Relatorios," 49.

169

49 Such fanciful notions, surely, defy a

swept away.
dollars-and-cents view of Portuguese imperialism and require
an explanation as to what Sofala symbolized over the cen-

turies.

 

49A. B. Cr6 de Castro Ferreri, Apontamentos de um
Ex-Governador de Sofalla (Lisbon, 1886), 73-74.

 

 

CHAPTER VII

SOFALA RETROSPECTIVE:
FACTS AND FANTASIES
In 1506, the Portuguese Crown was informed that

"Sofala is and will always be one of the best things Your
Highness has here, albeit it might not seem so now."1
Rhapsodized for centuries as the gateway to an African
Eldorado, Sofala proved a bitter disappointment, whose
substance never approximated royal expectations. Yet,
despite its protracted record of debits, the East African
entrep6t survived. Lisbon refused to relinquish the port
and extended her meager resources in the pursuit of a
fantasy that defied economic theories of profit and loss.
For Portugal, Sofala was imbued with a legendary and symbolic
reputation which superceded its blighted reality. Moreover,
Sofala's renown influenced the disastrous course of Portu-
guese actions throughout southeast Africa. Instead of gold
and glory, the Lusitanians were rewarded with only frustra-

tion and failure.

 

lPero Ferreira Fogaca to King, 22 December 1506, in
Documentos, I, 757.

 

170

171

The fantasy surrounding Sofala, the product of
imaginative medieval Muslim writers,2 seduced a credulous
Europe of the Middle Ages. The land of Sofala was perceived
not only as a source of abundant gold, but also as an
enchanted place populated with extraordinary people and
curious beasts. As early as the tenth century, an inventive
Persian seafarer claimed that local Sofalans were so gro-
tesque that they could frighten away gigantic elephant
eating birds.3 Utilizing such misinformation, Marco Polo
elaborated upon the appearance of these coastal folk: they
"have large mouths, their noses turn up towards the fore-
head, their ears are long, and their eyes so large and
frightful, that they have the aspect of demons."4 Such
exaggerated accounts intrigued many Europeans and were
responsible, in part, for stimulating the age of explora-
tion. The pursuit of fanciful places and people was the
result of a predisposition which in the Middle Ages acknow-

ledged the possibility of even the most spurious notions.5

 

2See above Chapter I.

3Buzurg Ibn Shahriyar, Book of Marvels, 56.

 

4Manuel Komroff, trans. and ed., The Travels of
Marco Polo (New York, 1953), 309.

 

 

50n the medieval imagination and its impact on re-
naissance voyages of discovery see John K. Wright, The Geo-
graphical Lore of the Time of the Crusades (New York, 1925);
Boises Penrose, Travel and Discoverygin the Renaissance
(Cambridge, Mass., 1952); W. G. L. Randles, “Notes on the
Genesis of Discoveries," Studia, 5 (1960), 20-46; Carolly
Erickson, The Medieval V131on (New York, 1976).

 

 

 

172

Sofala was an integral fragment of this enchanted world
view--one which the Portuguese never entirely abandoned.
During the fifteenth century, Lisbon was convinced
that golden Sofala would be discovered within the realm
of the fabled Prester John. Although the legend of a rich
and sagacious monarch residing in the east was unsubstanti-
ated, it was widely credited throughout Europe.6 Prince
Henry and later King Jo5o II dispatched expeditions in
search of this possible Christian ally and his gilded
emporium. It was, however, only during the reign of Manuel
that accounts were received in Lisbon confirming that
Prester John's dominion was to be found in East Africa.
His opulent kingdom reportedly encompassed the mysterious
region of Sofala where four-eyed monsters could be seen
handling quantities of gold..7
While the Portuguese investigation of the littoral
helped to diminish the supernatural image of Sofala, a new

myth was soon inaugurated. In 1502, when Vasco da Gama

 

6For the legend of Prester John see Charles Nowell,
"The Historical Prester John," Speculum, XXVIII (1953), 435-
445; Wright, Geographical Lore, 283-286; also note above
Chapter II.

 

7Manuel to Ferdinand and Isabella, c. 1501, in Voyage
of Pedro Alvares Cabral, 51; "Italian Account of the Portu-
guese DiscoveriesW (1505), in Documentos, I, 45. Although
the Portuguese came to equate Ethiopia with the land of
Prester John, they grossly exaggerated its dimensions, see
Afonso de Albuquerque to King, 4 December 1513, in Documen-
Eog, III, 497, 499; British Museum, Cotton Tiberius Ms. D.
IX, fol. 29, "Roteiro de Dom Joham de Castro da Viagee que
os Portugueses Fizeram dela India ate Soez," 1541.

 

 

173

inspected what was called Sofala, he found neither strange
creatures nor evidence of a flourishing gold commerce. Da
Gama seemingly had happened upon a recently settled town
which had appropriated the name originally applied to much
of the coast.8 Yet, Sofala's fabled pedigree was preserved
by one of da Gama's sailors, Tome Lopes, who linked it to
the biblical grandeur of Solomon's Ophir.9 The unhesitating
acceptance of this connection by the Portuguese suggests

how potent the medieval imagination remained despite the
evidence provided by exploration.

Since, as John Milton later wrote, "Sofala [was]
thought Ophir,"lo Lisbon anticipated golden dividends by
constructing an impressive settlement there. It was not
assets, however, that the crown collected but considerable
liabilities. Nevertheless, the royal court remained san-
guine that a great mine would be found to justify its
unwavering confidence. Thus, the initial Portuguese pene-
tration into the hinterland of Southern Zambezia was a
quest for the elusive sources of King Solomon's biblical

wealth. The subsequent discovery of massive stone ruins,

 

8See above Chapter I.

9Lopes, "Navegac5o da Armada," 232-233. See above
Chapter 11.

10Milton Paradise Lost xi. 400.

 

174

mining sites, and ostensibly wealthy Shona states seemed
to provide positive confirmation for this pursuit.11

By the end of the sixteenth century, however, the
ignominious condition of Sofala and the general failure of
Portugal's efforts throughout Southern Zambezia cast a

shadow over the already diminished Estado da India. Lisbon's

 

grandiose self-image had been eroded by European competitors
who snatched away her valuable overseas territories. In
1580, Castile captured the Portuguese Crown and dominated
the small Iberian nation for sixty years. Once liberated
from Spain, Portugal found it necessary to emphasize her
unique cultural identity and even to justify her national
sovereignty to neighboring states. Lisbon defended her
claim as a world power by pointing with pride to a far-
flung empire. Dom Luis da Cunha, an eighteenth century
diplomat, for example, observed, "the conquests are what

honour us and what sustain us."12

 

llSome sixteenth century Portuguese commentators
even suggested that Ophir was actually to be found within
the Mutapa Empire, see Couto, "Da Asia," 367; Pigafetta,
Kingdom of the Congo, 118-119; dos Santos, "Eastern Ethiopia:"
276-270; also note Randles, L'Image du Sud-Est Africain.

 

12Quoted in C. R. Boxer, "Faith and Empire: The Cross
and the Crown in Portuguese Expansion, Fifteenth-Eighteenth
Centuries," Terrae Incognitae VIII (1976), 86; also see Fred
Bronner, ”A World Apart: The Role of Empire in the Problem
of Portugal's Identity," Journal of the American Portuguese
Society, IX (1975), 18-29; J. H. Parry, "Portugal Overseas:
The Search for Riches," Journal of Interdisciplinary Histopy,
II (1972), 281; R. J. Hammond, “Uneconomic Imperialism: Por-
tugal in Africa before 1910," in Colopialism in Africa, I
(Cambridge, 1969), eds., L. Gann and P. Duignan, 355.

 

175

The imperial mystique was best represented by Sofala
which symbolized, for the Portuguese, the magnitude of their
heroic age of exploration. Its significance was not in its
blighted physical reality, but in its legendary past and
exalted future. Lusitania tenaciously clung to this image,
although her perseverance was amply tested. While she
possessed the will, the means were lacking to provide more
than minimal sustenance for the settlement's survival.
Somewhat paradoxically, the crown desired the preservation
of Sofala at any cost--as long as it was not too expensive.

Portugal's frustration in holding a fabled gold
emporium which produced no revenue led some dissenting
officials to recommend Sofala's abandonment. The crown
sharply criticized such heresy, but royal solicitude did
not result in solvency for the port. In fact, its golden
image became debased as Europeans questioned whether the
miserable Mozambican town could ever have served as King
Solomon's débouché. By the early nineteenth century, Henry
Salt cautiously observed that the Old Testament did not
provide sufficient information to link Sofala with Ophir.

He particularly chastised his older contemporary James
Bruce for drawing this conclusion based only upon "visionary

principles."l3 A few years later, Captain W. F. W. Owen

 

l3Henry Salt, A Voyage to Abyssinia and Travels in
the Interior of That Country (London, 1814), 99, 101, 103;
of. James Bruce, Travels to Discover the Source of the Nile
(Edinburgh, 1790), I, 434-440, 445.

 

 

176

described Sofala as "the Ophir of Scripture," but after
surveying the site, he admitted that he was puzzled by the
absence of any opulent ruins.l4 Such scientific investi-
gation inevitably would have encouraged the debunking of
the Sofala myth, but late nineteenth-century discoveries
in Southern Zambezia granted the legendary tale renewed
credibility. 6

Although Lisbon had long claimed hegemony over the
interior of southeast Africa, crushing defeats by various
Shona states and later by the Gaza-Nguni demonstrated that
Portuguese domination was little more than a cherished
fiction. Therefore, when various European explorers
penetrated the region, there was no Lusitanian force to
impede them. In 1871, Carl Mauch, a German geologist,
reached Great Zimbabwe and noted, "I do not think I am far
wrong if I suppose that the ruin on the Kopje is a copy of
Solomon's Temple on Mount Moria and the building in the
plain a copy of that palace where the Queen of Sheba lived
during her visit to Solomon."15 Since local Shona were
thought incapable of such construction (despite the evidence
of Portuguese historical records), Mauch's explanation was

accepted. The subsequent discovery of old gold workings

 

14P. R. O. Admiralty 1/2269, Owen, "Portuguese
Settlements"; idem, Narrative of Voyages, 318-319; also see
Chapter VI.

 

15Quoted in Roger Summers, "Carl Mauch on Zimbabwe
Ruins," NADA, 29 (1952), 15.

177

seemed to confirm his conclusion. A correspondent of the

Spectator observed that, "there may well be greater wonders

 

in the heart of Africa than even the romancer dreamt
of . . . traditions are apt to have a base of reality."16
Much to the chagrin of the Portuguese, British
prospectors and settlers, in particular, were soon staking
claims over Southern Zambezia in the hope of finding another
Rand. But, like their predecessors, these new adventurers
exaggerated the land's mineral potential. Manyika, for
example, was cited as a region where "gold is found every-
where" and the source of "the richest mines the world has

n17

ever known. Such exultation is reminiscent of Lisbon's

sixteenth-century predictions and is a reminder of the
persistence of the medieval world view.18

It should not be surprising to note that even at

the turn of the twentieth century, the question of Sofala

 

16"Reality and Romance," Spectator, 28 April 1888'

 

571.

17R. G. S., East Africa Journal Mss., report by E.
de Kergarion to the British Association, 30 August 1889; also
note P. R. 0., Foreign Office 403/111, Salisbury to Petre,
31 October 1889; J. S. Galbraith, "Cecil Rhodes and His
Cosmic Dreams: A Reassessment,” Journal of Imperial and
Commonwealth History, I (1973), 173-189.

 

 

18Professor Charles van Onselen recently stated that
the myths embracing southeast Africa "cannot really be traced
to the Portuguese so much as to the nineteenth century and
the era of imperial expansion." This view skews the course
of historical continuity and suggests that, suddenly, in
the nineteenth century men began thinking in fantastic ways.
These ideas did not originate from thin air; see his
Chibaro (London, 1976), 12.

178

and Ophir was subject to scholarly debate.19 Ironically,
though, the Portuguese settlement essentially had been
abandoned by 1865. Soon thereafter, Sofala was destroyed
by the elements: the village was inundated, roads were
covered with sand, and even the crumbling fortress was
deserted. What nature did not destroy, men did. In about
1904, a shipload of convict laborers--aboard, appropriately
enough, the Qphi£--was sent to Sofala to obtain stone for
the growing port of Beira, twenty miles to the north. They
wrenched loose every available block from the battered
stockade for the edification of the new coastal center.
Thus, quite literally, modern Beira was constructed from the
foundations of Sofala's past.20

Although the old entrep6t is beyond redemption, its

legendary reputation remains largely intact,21 and provides

 

19Armand Rainaud, Le Continent Austral (Paris, 1893),
54-67; Carl Peters, King Solomon's Golden Ophir (London,
1899); idem, Eldorado of the Ancients, esp. 12, 361, 365,
385; idem, "ophir and Punt in South Africa," Journal of the
African Socie_y, I (1901-1902), 174- 183; A. H. Keane, The
Gold of ophir (London, 1901); Gustav Oppert, "Tharshisfi_find
Ophir," Zeitschrift ffir Ethnologie, XXXV (1903), 50- 72.

 

 

 

 

 

 

20"Sofala--Transferencia da Villa," BSGL, X (1891),
204; M5rio Malheiros, "Fortalezas da Mo ambique: Sofala,"
Mocambique, I (1935), 13-15; R. W. Dickinson, "Sofala,"
Rhodesiana, 10 (1964), 47, 51, 53.

 

 

 

21For the most egregious examples see Francisque
Marconnes, "The Two East African Sofalas and King Solomon's
Ophir," NADA, XIII (1935), 59, 65— 74; Juli5o Quintinha,
"Descobertas Sensacionais no Territ6rio de Manica e Sofala,"
Boletim da Sociedade Luso:Africana do Rio de Janeiro, 24
71938), 43-44; Quint5o, Gramatica de Xironga, 23; Elaine

 

 

179

a bizarre reminder of a much embroidered historical fantasy.
Few sites can claim the unwarranted grandeur that was thrust

upon Sofala, first, as part of the medieval Imago Mundi and

 

later, as the proof of biblical lore. To the Portuguese
specifically, Sofala was the symbol of a once impressive
empire and the hope for an illustrious future. Sofala
could not be viewed, therefore, in logical terms, but only
metaphysically as an abstraction for all the fabled lands
of gold and glory which were always just one day's march

away. Sofala was not merely a place, but a state of mind.

 

Sanceau, Good Hope: The Voyage of Vasco da Gama (Lisbon,
1967), 13; M. D. W. Jeffreys, Some Semitic Influences in
Hottentot Culture (Johannesburg, 1968), 13.

 

 

 

LI ST OF REFERENCES

LIST OF REFERENCES

Archival Sources

 

GREAT BRITAIN

Bodleian Librapy, Oxford University
Ms. Rawlinson A. 334

 

British Museum, London
Additional Mss. 20,902, 25,419
Cotton Tiberius Ms. D. IX

 

Public Record Office, London
Admiralty 172269, 1/2270
Foreign Office 84/908, 403/111

 

Royal Geographical Society, London
East Africa Journal Mss.
Journal Mss. of C. E. Foot and T. L. Phipson—Wybrants
Papers of A. Vaughan Williams

 

MOZAMBIQUE

Arquivo Histdrico de Mogambique, Maputo
Gavetas 2-8, 2440, 2-57, 2-58, 4-63, 5-61

 

PORTUGAL

Academia dos Ciéncias de Lisboa
Ms. Azul 384

 

Arquivo Histérico Ultramarino, Lisbon
C6dice 1424
Diversos, Mocambique, Caixa 1
India, Caixa 29
Mocambique, Caixas l, 2, 4, 5, 6, 8, 10, 14, 28, 29,
0, 48, 51, 68,76, 77
Mozambique,8 Macos l-A, 3, 5, 10, ll, 12, 14, 21, 25,

 

180

181

Arquivo National de Torre do Tombo, Lisbon
Cartas dos Vicereis da India no. 58
Corpo Cronol6gico 3-21-1
Documentos Remetidos da India 1, 2, 3, 6, 15, 25,
26, 31, 33
Fragmentos, Caixa 3, Maco 3, Documento 7
Gavetas 2-6-30

Biblioteca da Ajuda, Lisbon
50-v-32, 51-iv-36, 51-vii-27, 51-vii-34, 51—viii-40

 

Biblioteca Geral da Universidade de Coimbra
Mss. 3064-3072

 

Biblioteca Nacional de Lisboa
Fundo Geral, Codices 176, 886

 

Filmoteca Ultramarina Portuguesa, Lisbon
Livros das Moncoes l6-A (21/1-2), 17(31/5),
24(12/2), 52(7/4)

SPAIN

Archivo General de Simancas
C6dices 1469, 1552

 

Biblioteca Nacional de Madrid
C6dice 3217
Ms. 2362

 

Unpublished Papers, Theses, and
Dissertations

 

 

Beach, D. N. "Historians and the Shona Empires, Parts I and
II." Unpublished seminar paper, University of
Rhodesia, 1972.

. An Outline of Shona History. Forthcoming.

 

Bhila, H. H. K. "The Manyika and the Portuguese, 1575-1863."
Unpublished Ph.D. dissertation, University of
London, 1971.

Chanaiwa, David. "A History of Nhowe before 1900." Unpub-
lished Ph.D. dissertation, University of California
at Los Angeles, 1971.

Datoo, B. A. "Selected Phases of the Historical Geography
of the Major Eastern African Ports." Unpublished
Ph.D. dissertation, University of London, 1968.

182

Dickinson, R. W. "Archaeological Investigations at Sofala,
Mo ambique, 9-18 June 1969." Unpublished report,
Sa isbury, 1969.

. "Archaeological Investigations at Sofala, Mo am-
bique, 29 May-11 June." Unpublished report, Sa is-
bury, 1970.

. "Sofala and the Rivers of Cuama: Crusade and
Commerce in South-East Africa, 1505-1595." Unpub-
lished M.A. thesis, University of Cape Town, 1971.

Fuller, C. E. "An Ethnographic Study of Continuity and
Change in Gwambe Culture." Unpublished Ph.D.
dissertation, Northwestern University, 1955.

Huffman, T. N. "The Leopard's Kopje Tradition." Unpublish-
ed Ph.D. dissertation, University of Illinois, 1974.

Rennie, J. K. "Christianity, Colonialism and the Origins
of Nationalism among the Ndau of Southern Rhodesia,
1890-1935." Unpublished Ph.D. dissertation, North-
western University, 1973.

Oral Data

 

Interviews with:
Eric Axelson, 27 May, 1974, Cape Town
D. N. Beach, 10 May 1974, Salisbury
E. E. Burke, 9 May 1974, Salisbury
R. W. Dickinson, 3 May 1974, Salisbury
Paulo Gon alves, 16 May 1974, Nova Sofala
T. N. Huf man, 8 May 1974, Salisbury
Mzee Makandara, 16 May, Nova Sofala

Published Sources

 

Primary Source Materials

Abu al-Fida. "Malindi, Mombasa, and Sofala." The East
African Coast: Select Documents. Edited by G. S. P.
Freeman-Grenville. Oxford, 1962, 23-24.

 

Alvares, Francisco. Verdadeira Informac5o das Terras do
Preste Jo5o das Indias. Lisbon7’1540.

 

 

Andrade, A. A. de, ed. Re1a96es de Mocambique Setecentista.
Lisbon, 1955. i "—

 

183

Andrade, J. J. Nogueira de. "Descripc5o do Estado em que
Ficav5o os Negocios da Capitania de Mossambique."
Archivo das Col6nias, I (1917), 75- 96.

 

"An Arabic History of Kilwa Kisiwani c. 1520." The East
African Coast: Select Documents. Edited by G. S. P.
Freeman-Grenville. Oxford, 1962, 34-49.

 

Archivo das Col6nias. 5 vols. Lisbon, 1917-1933.

 

"Authentic Testimony of the Baptism of the Emperor and King
Manamotapa." RSEA. Edited by G. M. Theal. Vol.
II. London, 1898, 441-488.

Axelson, Eric, ed. Viagem que fez o Padre Ant.o Gomes, da
Comp.a de Jesus, ao Imperio de de (sic) Manomotapa;
e assistencia que fez nas ditas terras de Alg'us
annos. Lisbon, 1959.

 

 

 

Azurara, Gomes Eannes de. The Chronicle of the Discovepy
and Conquest of Guinea. Translated and edited by C.
R. Beazley and Edgar Prestage. 2 vols. London,
1896-1899.

 

 

Barbuda, Luiz Coelho de. "Military Achievements of the
Portuguese." RSEA. Edited by G. M. Theal. Vol.
II. London, 1898, 362-377.

Barretto, Manuel. "Report upon the State and Conquest of
the Rivers of Cuama." RSEA. Edited by G. M. Theal.
Vol. III. London, 1899, 436-495.

. "Supplement to the Report upon the State." RSEA.
Edited by G. M. Theal. Vol. III. London, 1899, 496-
508.

Barros, Jo5o de. "Da Asia." RSEA. Edited by G. M. Theal.
Vol. VI. London, 1900,1- 306.

Berjeau, J., trans. and ed. Calcoen. London, 1874.
Bhila, H. H. K. trans. and ed. "A Journal of Manoel Galvao

da Silva's Travels Through the Territory of Manica
in 1790." Monumenta, VIII (1972), 79-84.

 

Bocarro, Ant6nio. "Decade." RSEA. Edited by G. M. Theal.
Vol. III. London, 1899, 254- 435.

Boteler, Thomas. Narrative of a Voyage of Discovery to
Africa and Arabia. 2 vols. London, 1835.

 

Botelho, Sebasti5o Xavier. Memoria Estatistica sobre os
Dominios Portuguezes na Africa Oriental. 2 vols.
LiSbon, 1835-1837.

 

184

Braganca Pereira, A. B. de, ed. Archivo Portugués Oriental.
Nova edic5o. ll vols. Bastor5, 1936-1940.

 

Bruce, James. Travels to Discover the Source of the Nile.
5 vols. Edinburgh, 1790.

Buzurg Ibn Shahriyar. The Book of the Marvels of India.
Translated by Marcel Devic and Peter Quennell.
London, 1928.

 

Camées, Luis Vaz de. The Lusiads. Translated by W. C.
Atkinson. Harmondsworth, 1952.

 

Castanheda, Fern5o Lopes de. "History of the Discovery and
Conquest of India by the Portuguese." RSEA.
Edited by G. M. Theal. Vol. V. London, 1901, 298-
500.

Castro Ferreri, A. B. Cr6 de. Apontamentos de um Ex-
Governador de Sofalla. Lisbon, 1886.

 

 

Chumovsky, T. A., ed. Trés Roteiros Desconhecidos de Ahmad
Ibn-Madjid. Portuguese translation by M. Makiel-
Jimounsky. Lisbon, 1960.

 

 

Conceic5o, Ant6nio da. "Tratado dos Rios de Cuama." O
Chronista de Tissuary. Edited by J. H. da Cunha
Rivara. Vol. II. Nova Goa, 1867, 39-45, 63-69,
84-92, 105-111.

 

Correa, Gaspar. "Lendas da India." RSEA. Edited by G. M.
Theal. Vol. II. London, 1898, 1-53.

Couto, Diogo de. "Da Asia." RSEA. Edited by G. M. Theal.
Vol. VI. London, 1900, 307- 410.

Dames, M. L., trans. and ed., The Book of Duarte Barbosa.
2 vols. London, 1918.

 

"Descripc5o da Capitania de Monsambique, Suas Povoacoes, e
Producoes." Relacoes de Mocambique Setecentista.
Edited by A. A. de Andrade. Lisbon, 1955, 375- 405.

 

"Descripc5o sobre o Estado de Sofalla na Costa Oriental d'
Africa." Archivo das Col6nias, I (1917), 185- 192.

 

Documentos sobre os Portugueses em Mogambique e na Africa
Central. Documents on the Portuguese in Mozambique
and Central Africa. 8 vols. Lisbon, 1962-1975.

 

 

 

"Eastern Ethiopia," RSEA. Edited by G. M. Theal. Vol. II.

185

Erskine, St. Vincent. "Journey to Umzila's, South-East
Africa in 1871-1872." JRGS, XLV (1875), 45-128.

. Third and Fourth Journeys in Gaza, or Southern
Mozambique, 1873-1874 and 1874 to 1875." JRGS,
XLVII (1878), 25-56.

Faria e Sousa, Manuel. "Asia Portuguesa." RSEA. Edited
by G. M. Theal. Vol. I. London, 1898, 1-46.

Ferao, F. H. "Account of the Portuguese Possessions within
the Captaincy of the Rios de Sena." RSEA. Edited
by G. M. Theal. Vol. VII. London, 1901, 371-383.

Figueroa, Martin Fernandez de. A Spaniard in the Portuguese
Indies. Translated and edited by James McKenna.
Cambridge, Mass., 1967.

 

Freeman-Grenville, G. S. P., ed. The East African Coast:
Select Documents. Oxford, 1962.

 

 

Freitas Ferraz, M. L. de, ed. Documentagao Historica
Mogambicana. Lisbon, 1973.

 

 

Frois, Luis. "Of the Voyage of the Father Dom Goncalo to
the Kingdom of Monomotapa and of His Happy Passing
Away." RSEA. Edited by G. M. Theal. Vol. II.
London, 1898, 104-128.

Gibb, H. A. R., trans. and ed. The Travels of Ibn Battuta.
3 vols. Cambridge, 1958-1971.

 

Goes, Damiao de. "Chronicle of the Most Fortunate King Dom
Emanuel of Glorious Memory." RSEA. Edited by G. M.

Greenlee, William B., trans. and ed. The Voyage of Pedro
Alvares Cabral to Brazil and India. London, 1938.

 

 

Guyot, Paul. Voyage au Zambese. Paris, 1895.

 

Ibn Khaldun. The Mugaddimah. Translated by Franz Rosenthal.
3 vols. London, 1958.

 

al-Idrisi. India and the Neighboring Territories. Trans-
lated and edited by S. M. Ahmad. Leiden, 1960.

 

"Instruccao que Illmo e Exmo Sr. Governador e Capitao
General Baltazar Manuel Pereira do~Lago Deo a Quem
Lhe Suceder neste Governo." Relacoes de Mogambique
Setecentista. Edited by A. A. deiAndrade. Lisbon,
1955, 317-338.

 

 

186

"Italian Account of the Portuguese Discoveries." Documentos.
Vol. I. Lisbon, 1962, 41-75.

 

Jensen, John, ed. Journa1_and Letter Book of Nicholas
Buckeridge, 1651-1654. Minneapolis, 1973.

 

 

Komroff, Manuel, trans. and ed. The Travels of Marco Polo.
New York, 1953.

 

Linschoten, Jan Huighen van. Discours of Voyages into ye
Easte and West Indies. London, 1598.

 

Lopes, Tomé. "Navegacao da Armada de Estevam da Gama."
Archivo Portugues Oriental. Nova edicao. Edited by
A. B. de Braganca Pereira. Vol. I- I- I. Bastora,
1936, 230- 233.

 

al-Masudi. "The Ivory Trade." The East African Coast:
Select Documents. Edited by G. S. P. Freeman-

 

 

"Memorias da Costa d'Africa Oriental." Relacoes de Mocam-
bique Setecentista. Edited by A. A. de Andrade.
Lisbon, 1955, 189-224.

 

Miranda, Antonio Pinto de. "Monarchia Africana." Anais da
Junta de Investigacoes do Ultramar, IX-I (1954),
109-117.

 

Monclaro, Francisco. "Account of the Journey Made by Fathers
of the Company of Jesus with Francisco Barreto."
RSEA. Vol. III. London, 1899, 157-253.

Montaury, Joao Baptista de. "Mocambique, Ilhas Querimbas,
Rios de Sena, Villa de Sena, Villa de Tete, Villa
de Zumbo, Manica, Villa de Luabo, Inhambane."
Relacoes de Mocambique Setecentista. Edited by A. A.
de Andrade. Lisbon, 1955, 339-373.

 

Newitt, M. D. D., trans. and ed. Account of a Journey Made
Overland from Quelimane to Angoche in 1752 by Fran-
EiSco Raymundo Moraes Pereira. Salisbury, 1965.

 

 

 

Owen, W. F. W. "The Bay of Delagoa." RSEA. Edited by G.
M. Theal. Vol. II. London, 1898, 465-475.

. Narrative of Voyages to Explore the Shores of
Africa, Arabia, and Madagascar. 2 vols. London,
1833.

 

Paula Castro Domingues, Francisco de. "Relatorios." Boletim
Official do Governo Geral da Provincia de Mocam51que,
12 (1871), 49- 50.

187

Pigafetta, Filippo. A Report of the Kingdom of the Congo
and of the Surrounding Countries. Translated and
edited by Magarite Hutchinson. London, 1881.

 

 

Pissurlencar, P. S. 8., ed. Regimentos das Fortalezas da
India. Bastora-Goa, 1951.

 

Prior, James. Voyage along the Eastern African Coast.
London, 1819.

 

Reis e Gama, C. J. dos. Reposta das Questoens sobre os
Cafres. Edited by Gerhard Liesegang. Lisbon, 1966.

 

"Rellacam das Moradores Portuguezes que Assitem em Monssa-
bique e Seos Destrictos." Anais da Junta de
Investigacoes. Vol. IX-I. Lisbon, 1954, 153-170.

 

 

Rezende, Pedro Barretto de. "Of the State of India." RSEA.
Edited by G. M. Theal. Vol. II. London, 1898, 378-
426.

Sachau, Edward, trans. and ed. Alberuni's India. 2 vols.
Lahore, 1962.

 

Salt, Henry. A Voyage to Abyssinia and Travels in the
Interior of That Country. London, 1814.

 

 

Santa Tereza, Francisco de. "Plano e Rela 50 da Bahia
Denominada de Lourenco Marques." escobrimento e
Fundagao de Lourenco Marques. Edited By Caetano
Montez. Lourenco Marques, 1948.

 

 

Santos, Joao dos. "Eastern Ethiopia." RSEA. Edited by
G. M. Theal. Vol. VII. London, 1901, 1-370.

Santos Pinto, Caetano dos. "Viagem de Inhambane as Terras

de Manicusse em 1840." Archivo das Colonias, I
(1917), 269-274.

 

Silva, Manuel Galvao da. "Diario das Viagens Feitas pelas
Terra de Manica." Anais da Junta de Investigacoes
do Ultramar, IX-I (1954), 323- 332.

 

 

Soares, A. E. X. Descripqao da Villa de Sofalla. Nova Goa,
1857.

 

Sullivan, G. L. Dhow Chasing in Zanzibar Waters. London,
1873.

 

Theal, G. M., ed. Records of South Eastern Africa. 9 vols.
London, 1898-1903.

 

188

Wallis, J. P. R., ed. The Zambezi Expedition of David Liv-
ingstone, 1858-1863. 2 vols. London, 1956.

Xavier, Caetano Ignacio. "Noticias dos Dominios Portugueses
na Costa de Africa Oriental." Relagoes de Mocam-
bique Setecentista. Edited by A. A. de Andrade.
Lisbon, 1955, 139-188.

 

Secondary Materials

Abraham, D. P. "Early Political History of the Kingdom of
Mwene Mutopa (850-1589)." Historians in Tropical
Africa. Salisbury, 1962, 6i¥91.

 

. "Ethno-History of the Empire of Mutapa: Problems
and Methods." The Historian in Tropical Africa.
Edited by J. Vansina, R. Mauny, and L. V. Thomas.
London, 1964, 104-121.

"Maramuca: An Exercise in the Combined Use of
Portuguese Records and Oral Traditions." JAH, II
(1961), 211-225.

Alavi, S. M. Z. Arab Geography in the Ninth and Tenth Cen-
turies. Aligarh, 1965.

Alberto, M.’S. "Os Angones os fitimos Povos Invasores da
Angonia Portuguesa." Mocambique, 27 (1947), 79-103.

 

. "Carta de Sofala." BSEM, XXVIII (1959), 123-136.

Alcantara Guerreiro, Jeronimo de. "Inquérito em Mocambique
no Ano de 1573." Studia, 6 (1960), 7-18.

. "Quadros da Historia de Mocambique." Mocambique,
68 (1951), 6-30.

 

I I O I ’ ’
Alcantara Guerre1ro, Jer6n1mo de and Oliveira Boleo, Jose

de. As Minas de Prata de Chicoa. Lourenco Marques,
1944.

Alpers, E. A. "Dynasties of the Mutapa-Rozwi Complex."
JAH, XI (1970), 203-220.

. Ivory and Slaves in East Central Africa. Berke-
ley, 1975.

. "Re-Thinking African Economic History." Ufahamu,
III (1973), 97-129.

"Review of Eric Axelson's, Portuguese in South-
East Africa, 1488-1600." IJAHS, VIII (1975), 91-94.

 

189

Arkell, A. J. "Cambay and the Bead Trade." Antiquity, X
(1936), 292-305.

 

Axelson, Eric. "Gold Mining in Mashonaland in the 16th and
17th Centuries." Optima, IX (1959), 164-170.

. Portuguese in South-East Africa, 1600-1700.
Johannesburg, 1960.

 

. Portuguese in South-East Africa, 1488-1600.
Johannesburg, 1973.

 

Balsan, Francois. "Ancient Gold Routes of the Monomotapa
Kingdom." Geographical Journal, CXXXVI (1970),
240-246.

 

. "La Route de l'Or du Matabeland." BSEM, XXXII
(1963), 99-109.

Barradas, Lereno. "Inhambane de Outrora." Monumenta, VI

 

. ”A Primitiva Mambone e Suas Imediacoes." Monu-
menta, III (1967), 21-41.

. "Sobre o Roteiro de Sofala do Piloto Arabe Hamad
Ibn-Madjid." Studia, 32 (1971), 7-50.

. "O Sul de Mocambique no Roteiro de Sofala do
Piloto Ahmad Ibn-Madjid." Revista da Universidade
de Coinbra, XXII (1970), 157-176.

 

 

Barrett, 0. W. "The Portuguese Colony of Mozambique."
National Geographic, XXI (1910), 807-830.

 

Beach, D. N. "Historiography of the People of Zimbabwe in
the 1960's." Rhodesian History, IV (1973), 21-30.

 

. "The Mutapa Dynasty: A Comparison of Documentary
and Traditional Evidence." History in Africa, III
(1976), 1-17.

. "The Shona Economy: Branches of Production." TEE
Roots of Rural Poverty in Central and Southern
Africa. Edited by Robin Palmer and Neil Parsons.
Berkeley, 1977, 37-65.

 

Beckingham, C. F. "Notes on an Unpublished Manuscript of
Francisco Alvares: Verdadgra Informacam das Terras
do Preste Joam das Indias." Annales'd'Ethigpie, IV
11961), 139-154.

 

 

190

. "The Travels of Pero da Covilha and Their Signi-
ficance." Actas. Vol. III. Lisbon, 1961, 1-14.

Bhila, H. H. K. "Manyika's Relationship with the Portuguese
and the Gaza-Nguni." Rhodesian History, VII (1976),
31-37.

 

. "Trade and Survival of an African Polity: The
External Relations of Manyika from the Sixteenth to
the Early Nineteenth Century." Rhodesian History,
III (1972), 11-28.

Blake-Thompson, J. "The Oriental Ancestry of the Aulaya-
'Nyungwe." South African Journal of Science, XXXVI
(1939), 464-467.

 

Boas, Franz and Kamba, Simango. "Tales and Proverbs of the
Vandau of Portuguese South Africa." Journal of
American Folk-Lore, XXXV (1922), 151-204.

 

 

Bovill, E. W. The Battle of Alcazar. London, 1952.

 

Boxer, C. R. An African Eldorado: Monomotapa and Mocambique,
1498-1752. Salisbury, 1960. '

 

. "The Carreira da India, 1650-1750." Mariner's
Mirror, XLVI (1960), 36-54.

 

 

. "A Dominican Account of Zambezia in 1744." BSEM,
XXIX (1960), 1-14.

. "Faith and Empire: The Cross and the Crown in
Portuguese Expansion, Fifteenth-Eighteenth Cen-
turies." Terrae Incognitae, VIII (1976), 73-89.

 

. "Pombal's Dictatorship and the Great Lisbon Earth-
quake, 1755." History Today, V (1955), 729-736.

 

. "Portuguese and Dutch Colonial Rivalry, 1641-
1661." Studia, 2 (1956), 7-42.

. "Portuguese Roteiros, 1500-1700." Mariner's
Mirror, XX (1934), 171-186.

 

. The Portuguese Seaborne Empire, 1415-1825. New
York, 1969.

 

. "Sisnando Dias Bayao: Conquistador da Mae da ~
Ouro." Primiero Congresso da Historia da Expansao
Portuguesa no Mundo. Vol. III. Lisbon, 1938, 101-
115.

 

 

191

. "Three Historianshof Portuguese Asia." Boletim
do Instituto Portugues de Hong Kong, I, 15-44.

 

Bronner, Fred. "A World Apart: The Role of Empire in the
Problem of Portugal's Identity." Journal of the
American Portuguese Society, IX (1975), 18-29.

 

 

Burke, E. E. "Some Aspects of Arab Contact with South
East Africa." Historians in Tropical Africa.
Salisbury, 1962, 93-106.

 

Cabral, A. C. P. Pequeno Dicionario de Mogambique. Lour-
enco Marques, 1972.

 

Cerulli, Enrico. Somalia. 3 vols. Rome, 1957-1964.

Chanaiwa, David. "Politics and Long-Distance Trade in the
Mwene Mutapa Empire during the Sixteenth Century."

Chirenje, J. M. "Portuguese Priests and Soldiers in Zim-
babwe, 1560-1572: The Interplay between Evangelism
and Trade." IJAHS, VI (1973), 36-48.

Chittick, H. N. "An Archaeological Reconnaissance of the
Southern Somali Coast." Azania, IV (1969), 115-130.

 

. "The Coast of East Africa." The African Iron
Age. Edited by Peter Shinnie. Oxford, 1971, 108-
141.

. "Ibn Battuta and East Africa." Journal de la
Société des Africanistes, XXXVIII (1968), 239-241.

 

 

. Kilwa: An Islamic Trading City on the East
African Coast. 2 vols. Nairobi, 1974.

 

"Kilwa: A Preliminary Report." Azania, I, 1-36.

. "Observations on Pre-Portuguese Accounts of the
East African Coast." Societés et Compagnies de
Commerce en Orient et dans l'Ocean Indien. Edited
by Michel Mollat. Paris, 1970, 129-132.

 

 

. "The Peopleing of the East African Coast." East
Africa and the Orient. Edited by H. N. Chittick and
Robert Rotberg. New York, 1975, 16-43.

 

. "The 'Shirazi' Colonization of East Africa."
JAH, VI (1965), 275-294.

Chittick, H. N. and Rotberg, Robert, eds. East Africa and
the Orient. New York, 1975.

 

192

Cipolla, Carlo. European Culture and Overseas Expansion.
Harmondsworth, 1970.

Clark, J. D. "The Portuguese Settlement at Feira." North-
ern Rhodesia Journal, VI (1965), 275-292.

 

 

Cortesao, Armando and Mota, A. Teixeira da. Portugaliae
Monumenta Cartographica. 5 vols. Lisbon, 1960.

 

 

Cortesao, Jaime. Os Portugueses em Africa. Lisbon, 1968.

 

Crone, G. R. "Fra Mauro's Representation of the Indian
Ocean and Eastern Islands." Studi Colombiani.
Vol. III. Genoa, 1951, 57-64.

 

Cunha Saraiva, José da. "Causas do Levantamento da Nacao
em 1640." Congresso do Mundo Portugués. Vol. VII.
Lisbon, 1940, 59-94.

 

Curtin, Philip. "The Lure of Bambuk Gold." JAH, XIV
(1973), 623-631.

Curtin, Philip, Feierman, Steven, Thompson, Leonard and
Vansina, Jan. African History. Boston, 1978.

 

Devic, Marcel. Le Pays des Zendjs. Paris, 1883.

 

Dickinson, R. W. "Angoche and the Sofala Shoal." Rhodesi-
ana, 34 (1976), 28-37.

. "Antonio Fernandes--A Reassessment." Rhodesiana,
25 (1971), 45-52.

 

. "Archaeological Investigations at Nova Mambone,
Mocambique." Monumenta, VII (1971), 21-31.

 

. "The Archaeology of the Sofala Coast." SAAB,
XXX (1975), 84-104.

. "Excavations at Sofala." Rhodesia Science News,
IV (1970), 423.

 

. "Sofala." Rhodesiana, 10 (1964), 45-53.

 

. "Sofala-Gateway to the Gold of Monomotapa."
Rhodesiana, 19 (1968), 33-47.

 

Diffie, Bailey. "Foreigners in Portugal and the Policy of
Silence." Terrae Incognitae, I (1969), 23-24.

 

 

. Prelude to Empire. Lincoln, 1960.

 

193

Diffie, Bailey and Winius George. Foundations of‘the Portu-
guese Empire, 1415-1580. Minneapolis, 1977.

 

 

Doke, C. M. Report on the Unification of the Shona Dialects.
Hertford, 1931.

 

. The Southern Bantu Languages. London, 1954.

 

Duffy, James. A Question of Slavery. Oxford, 1967.

 

. Shipwreck and Empire. Cambridge, Mass., 1955.

 

Elkiss, T. H. "On Service to the Crown." Journal of the
American Portuguese Society, X (1976), 44-53.

 

 

Elliott, J. H. Imperial Spain, 1469-1716. New York, 1963.

 

 

. "Self-Perception and Decline in Early Seventeenth-
Century Spain." Past and Present, 74 (1977), 41-
61.

Erickson, Carolly. The Medieval Vision. New York, 1976.

 

Evers, T. M. "Iron Age Trade in the Eastern Transvaal,
South Africa. SAAB, XXIX (1974), 33-37.

Fagan, Brian. "Early Trade and Raw Materials in South
Central Africa." JAH, X (1969), 1-13.

. "The Later Iron Age in South Africa." African
Societies in Southern Africa. Edited by Leonard
Thompson. London, 1969, 50-70.

 

Ferrand, Gabriel. "Sofala." The Encyclopedia of Islam.
Vol. IV. Edited by M. T. Houtsma, A. J. Wensinck,
H. A. R. Gibb, W. Heffening, and E. Levi-Provencal.
Leiden, 1934.

 

Filesi, Teobaldo. China and Africa in the Middle Ages.
Translated by David Morrison. London, 1972.

 

Fortune, G. "Who Was Mwari." Rhodesian History, IV (1973),
1-20 0

 

Freeman-Grenville, G. S. P. "Chinese Porcelain in Tanganyika."
Tanganyika Notes and Records, 41 (1955), 63-66.

 

. "Islam and Christianity in East Africa." African
Ecclesiastical Review, II (1960), 193-207.

 

Galbraith, J. S. "Cecil Rhodes and His Cosmic Dreams: A
Reassessment." Journal of Imperial and Commonwealth
History I (1973), 173-189.

 

194

Garcia, Ant6nio. "Goncalo da Silveira, Proto—Martir de
Mo ambique e da Africa do Sul." Monumenta, VI
(1 70), 53-57.

 

Garlake, P. S. "Excavations at the Seventeenth Century
Portuguese Site of Dambarare, Rhodesia." Proceed-
ings of the Rhodesia Scientific Association, LIV
(1969), 23-53.

 

. Great Zimbabwe. London, 1973.

 

. "An Investigation of Manekweni, Mozambique."
Azania, X (1976), 25-47.

. "Rhodesian Ruins--A Preliminary Assessment of
Their Styles and Chronology." JAH, X (1970), 495-
513.

. "The Value of Imported Ceramics in the Dating
and Interpretation of the Rhodesian Iron Age."
JAH, IX (1968), 13-33.

Gerth, H. H. and Mills, C. Wright, trans. and eds. From
Max Weber: Essays in Sociology. New York, 1958.

 

Gluckman, Max. "The Individual in a Social Framework: The
Rise of King Shaka of Zululand." Journal of African
Studies, I (1974), 113-144.

 

. "The Rise of a Zulu Empire." Scientific American,
CCII (1960), 159-169.

Gray, Richard. "Portuguese Musketeers on the Zambezi." JAH,
XII (1971), 531-533.

. "Review of Portuguese in South-East Africa, 1488-
1600." JAH, XVI (1975), 148-149.

Gray, Richard and Birmingham, David, eds. Pre-Colonial
African Trade. London, 1970.

 

 

Guillain, M. Documents sur 1'Histoire, 1a Geographie et 1e
Commerce de 1'Afrique Orientale. 3 vols. Paris,
1856.

 

Haight, M. V. Jackson. European Powers and South-East
Africa. London, 1967.

Hammond, R. J. "Uneconomic Imperialism: Portugal in Africa
before 1910." Colonial in Africa. Edited by L.
Gann and P. Duignan. Vol. I. Cambridge, 1969,
352-382.

 

195

Hannan, M. Standard Shona Dictionary. London, 1961.

 

Historians in Tropical Africa. Salisbury, 1962.

Hoppe, Fritz. A Africa Oriental Portuguesa no Tempo do
Marques de Pombal, 1750-1777. Lisbon, 1970.

 

Huffman, T. N. "Ancient Mining and Zimbabwe." JSAIMM,
LXXIV (1974), 238-242.

. "Cloth from the Iron Age in Rhodesia." Arnoldia,

. "Excavations at Leopard's Kopje Main Kraal: A
Preliminary Report." SAAB, XXVI (1971), 85-89.

. "Radiocarbon Dates and Bibliography of the Rho-
desian Iron Age." Rhodesian Prehistory, 11 (1973),
3-8.

 

. "The Rise and Fall of Zimbabwe." JAH, XIII
(1972), 353-366.

"Solomon, Sheba and Zimbabwe." Rhodesian Pre-
history, VII (1975), 21-24.

 

Isaacman, Allen. Mozambique: The Africanization of a Euro-
pean Institution. Madison, 1972.

 

 

Jeffreys, M. D. W. "The Batwa: Who Are They." Africa,
XXIII (1953), 45-54.

. Some Semitic Influences in Hottentot Culture.
Johannesburg, 1968.

 

Keane, A. H. The Gold of Ophir. London, 1901.

 

Kimble, George H. "Portuguese Policy and Its Influence on
Fifteenth Century Cartography. Geographical Review,
XXIII (1933), 653-659.

 

Kirkman, James. "The Culture of the Kenya Coast in the
Later Middle Ages." SAAB, XI (1956), 89-99.

"The History of the Coast of East Africa up to
1700," Prelude to East African History. Edited by
Merrick Posnansky. London, 1966, 105-124.

 

Laessoe, H. de. "The Lundi and Sabi Rivers." Proceedings
of the Rhodesia Scientific Association, VI (1906i:
118-138.

 

 

 

196

Leitao, Humberto. "Um Notavel Episddio do Primiero Cerco de
Macambique." Ultramar, XII (1971), 5-13.

Liesegang, Gerhard. "Aspects of Gaza Nguni History 1821-
1897." Rhodesian History, VI (1975), 1-14.

. Beitrage zur Geschichte des Reiches der GazagNguni
im sudlichen Mocambique 1820-1895. Cologne, 1967.

. "Dingane's Attack on Lourenco Marques in 1833."
JAH, X (1969), 565-579.

. "Nguni Migrations between Delagoa Bay and the Zam-
bezi, 1821-1839." IJAHS, III (1970), 317-337.

Linden, Ian. "Some Oral Traditions from the Maskeo Ngoni."
Society of Malawi Journal, XXIV (1971), 61-73.

Livermore, H. V. A New History of Portugal. Cambridge, 1966.
Lloyd, E. M. "Mbava." NADA, 3 (1925), 62-64.

Lobato, Alexandre. Colonizacéo Senhorial da Zambezia e Outros
Estudos. Lison, 1962.

. Evolucao Administrativa e Economica de Mocambigpe,
Lisbon, 1957.

. A Expansao Portuguesa em Mocambique de 1498 a 1530.
3 vols. Lison, 1954-1960.

. Novos Subsidios para a Historia de Fundacao de
Sofala. Lisbon, 1950.

Macgregor, I. A. "Some Aspects of Portuguese Historical
Writing of the Sixteen and Seventeenth Centuries on
South East Asia." Historians of South East Asia.
Edited by D. G. E. Hall. London, 1961, 172-199.

Magalhaes-Godinho, Victorino. L'Economie de l'Empire Portu-
gais aux XVe e XVIe Siecles. Paris, 1969.

Magino, Leo, "Antonio de Noli e a Colaboracao entre os
Portuguese e Genoveses nos Descobrimentos Maritimos."
Studia, 10 (1962), 99-115.

Malheiros, Mario. "Fortalezas da Mocambique: Sofala."
Mogambique, 1 (1935), 7-15.

 

Marchant, Alexander. "Colonial Brazil as a Way Station for

the Portuguese India Fleets." Geographical Review,
XXXI (1941), 454-465.

197

Marconnes, Francisque. "The Two East African Sofalas and
King Solomon's Ophir." NADA, XIII (1935), 59-74.

Martin, B. G. "Arab Migration to East Africa in Medieval
Times." IJAHS, VII (1974), 367-390.

Matecka, A. fLa Cate Qrientale de 1'Afrique au Moyen Age
d'apres 1e Kitab Ar-Rawd Al-MiCTar de a1 Himyari
(XVeS)." Folia Orientalia, IV (1962), 331-339.

 

 

Mathew, Gervase. "Some Reflections on African Trade Routes."
Research Review, III (1967), 63-71.

 

Mauny, Raymond. "Notes sur 1e Probléme Zimbabwe-Sofala."
Studia, I (1958), 176-183.

. "The Wakwak and the Indonesian Invasion in East
Africa." Studia, 15 (1965), 7-16.

Mazery, Paul. "Mozambique and the Napoleonic Wars." Studia,
37 (1973), 387-399.

Merwe, N. J. van der and Scully, Robert. "The Phalaborwa
Story: Archaeological and Ethnographic Investigation
of a South African Iron Age Group." World Archaeology,
III (1971), 178-196.

Mkanganwi, K. G. "The Relationship of Coastal Ndau to the
Shona Dialects of the Interior." African Studies,
XXXI (1972), 111-137.

 

Montez, Caetano. "As Racas Indigenas de Mocambique. Mog m-
bigue, 23 (1940), 53-66.

Moraes, P. F. de. "Silent Trade: Myth and Historical
Evidence." History in Africa, I (1974), 9-24.

 

More, C. E. "Some Observations on 'Ancient' Mining at Phala-
borwa." JSAIMM, LXXIV (1974), 227-232.

Morison, Samuel Eliot. Portgguese Voyages to America in the
Fifteenth Century. Cambridge, Mass., 1940.

 

Mota, A. Teixeira da. "Cartografia Antiga de Sofala."
Monumenta, IX (1973), 5-18.

Mudenge, S. I. "Eighteenth Century Portuguese Settlements on
the Zambezi and the Dating of Rhodesian Ruins: Reflec-
tions on the Problems of Reference Dating." IJAHS, X
(1977), 384-393.

. "An Identification of the Rozvi and Its Implica-
tions for the History of the Karanga." Rhodesian
History,‘V (1974), 19-31.

 

198

. "The Role of Foreign Trade in the Rozvi Empire: A
Reappraisal." JAH, XV (1974), 373-391.

Newitt, M. D. D. "The Early History of the Sultanate of
Angoche." JAH, XII (1972), 397-406.

. Portuguese Settlement on the Zambesi. London,
1973.

. "Ignacio Caetano Xavier's Account of Portuguese
East Africa." Conference of the History of the
Central African Peoples. Lusaka, 1963, no pagination.

"Towards a History of Modern Mocambique."
Rhodesian History, V (1974), 33-47.

Nowell, Charles. "The Historical Prester John." Speculum,
XXVIII (1953), 435-445.

Oidtman, Christoph von. "0 Porto de Sofala e o Problema de
Zimbaue." BSEM, XXVII (1958), 5-19.

Oliveira, O. R. de. Amuralhados da Cultura Zimbaue-Monomotapa
de Manica e Sofala. Lourenco Marques, 1963.

Oliviera Guimaraes, Luis de. "0 Povo na Revolucao de 1640."
Congresso do Mundo Portugués. Vol. VII. Lison, 1940,
139-146.

Omer-Cooper, J. D. The Zulu Aftermath. London, 1966.

Onselen, Charles van. Chibaro. London, 1976.

Oppert, Gustav. "Tharshish und Ophir." Zeitschrift ffir
Ethnologie, XXXV (1903), 50-72.

 

Parry, J. H. The Age of Reconnaissance. London, 1963.
. "Portugal Overseas: The Search for Riches."
Journal of Interdisciplinagy,Histogy, II (1972),
281-286.

Penrose, Boises. Travel and Discovery in the Renaissance.
Cambridge, Mass., 1952.

Pereira, Manuel Cesar. "Discurso sobre a Conquista das Minas
de Monomotapa." BSGL, VIII (1888-1889), 539-543.

Peters, Carl. The Eldorado of the Ancients. London, 1902.

. King Solomon's Golden Ophir. London, 1899.

 

. "Ophir and Punt in South Africa." Journal of the
African Society, I (1901-1902), 174-183.

199

Phillipson, D. W. "Early Iron-Using Peoples of Southern
Africa." African Societies in Southern Africa.
Edited by Leonard Thompson. London, 1969, 24-49.

Phimister, I. R. "Alluvial Gold Mining and Trade in Nine-
teenth-Century South Central Africa." JAH, XV (1974),
445-456.

. "Ancient Mining near Great Zimbabwe." JSAIMM,
LXXIV (1974), 233—237.

. "Pre-Colonial Gold Mining in Southern Zambezia:
A Reassessment." African Social Research, 21 (1976),
1-30.

Pouwels, Randall. "Tenth Century Settlement of the East
African Coast: The Case for Qarmatian/Ismacili Con-
nections." Azania, IX (1974), 65-74.

Prestage, Edgar. "The Search for the Sea Route to India."
Travels and Travellers of the Middle Ages. Edited by
A. P. Newton. New York, 1926, 195-216.

 

Quintao, José Luis. Gramatica de Xironga (Landim). Lisbon,
1951.

 

Quintinha, Juliao. "Descobertas Sensacionais no Territ6rio
de Manica e Sofala." Boletim da Sociedade Luso-
Africana do Rio de Janeior, 24 (1938), 43-44.

 

 

Rainaud, Armada. Le Continent Austral. Parix, 1893.

 

Randles, W: G. L. L'Empire du Monomotapa du XVe au XIX-e
Siecle. Paris, 1975.

. "La Fondation de l'Empire du Monomotapa." Cahiers
d' Etudes Africaines, XIV (1974), 211-236.

 

. L'Image du Sud-Est Africain dans la Litterature
Européenne au XVIe Siécle. Lisbon, 1959.

 

 

. Notes on the Genesis of Discoveries." Studia, 5
(1960), 20-46. ‘-———-'

. South East Africa and the Empire of Monomotapa on
Selected Printed Maps of the 16th Century. Lisbon,
1958.

 

 

Rangely, W. H. J. "The Angoni." Society of Malawi Journal,
XIX (1966), 62-86.

 

"Reality and Romance." Spectator, 28 April 1888, 569-671.

 

 

200

Robinson, K. R. Khami Ruins. Cambridge, 1959.

Sanceau, Elaine. Good Hope: The Voyage of Vasco da Gama.
Lisbon, 1967.

Sicard, H. von. "The Ancient Sabi- Zimbabwe Trade Route."
NADA, XL (1963), 6-16.

. "Pero de Covilha's 'Relaction,‘ 1587." Studia,
29 (1970), 155-168.

. "A Propésito de Sisnando Dias Baiao." Studia,
16 (1965), 179-187.

Silva Rego, Ant6nio da. 0 Ultramar Portugués no Século
XVIII. Lisbon, 19671

Smith, Alan. "Delagoa Bay and the Trade of South-Eastern
Africa." Pre-Colonial African Trade. Edited by
Richard Gray and David Birmingham. London, 1970,
265-289. ‘

. "The Peoples of Southern Mozambique: An Historical
Survey." JAH, XIV (1973), 565-580.

. "The Trade of Delagoa Bay as a Factor in Nguni
Politics 1750-1835." African Societies in Southern

Africa. Edited by Leonard Thompson. London, 1969,
171-189.

Smith, Robert. The Art of Portugal, 1500-1800. New York,
1968.

“Sofala - Transferencia da Villa." BSGL, XL (1891), 203-207.

Soper, R. C. "New Radiocarbon Dates for Eastern and Southern
Africa." JAH, XV (1974), 175-192.

Steven-Hamilton, J. The Low Veld. London, 1929.

Summers, Roger. Ancient Mining in Rhodesia. Salisbury, 1969.
. "Carl Mauch on Zimbabew." NADA, 29 (1952), 9-17.
. "Notes on the Economic Bases of Southern Rhodesian
Iron Age Cultures." Conference on the History of the

Central African Peoples. Lusaka, 1963, no pagination.

Sundstrdm, Lars. The Exchapge Economy of Pre-Colonial
Tropical Africa. London, 1974.

 

201

Sutherland-Harris, Nicola. "Trade and the Rozwi Mambo."
Pre-Colonial African Trade. Edited by Richard Gray
and David Birmingham. London, 1970, 243-264.

Thompson, Leonard, ed. African Societies in Southern Africa.
London, 1969.

Tibbetts, G. R. Arab Navigation in the Indian Ocean before
the Coming of the Portuguese. London, 1971.

 

Toscano, F. "Sobre os Indigenas Portugueses ao sul do
Zambeze." BSEM, XL (1941), 65-147.

Tracey, Hugh. Ant6nio Fernandesl Descobridor do Monomotapa,
1514-1515. Translated and edited by Caetano Montez.
Lourenco Marques, 1940.

 

. "Antonio Fernandes, Rhodesia's First Pioneer."
Rhodesiana, 19 (1968), 1-26.

 

"Vias Comerciais Arabes de Sofala ao Monomotapa
antes do Estabelecimento dos Portugueses." Mogambi-
que, 26 (1941), 33-37.

Trimingham, J. Spencer. "The Arab Geographers and the East
African Coast." East Africa and the Orient. Edited
by H. N. Chittick and Robert Rotberg. New York,
1975, 115-146.

 

. "Notes on Arabic Sources of Information on East
Africa." East Africa and the Orient. Edited by
H. N. Chittick and Robert Rotberg. New York, 1975,

 

 

272-283.

Verlinden, Charles. "Le Influence Italiane nella Colonizza-
zione Iberica." Nuova Rivista Storica, XXXV (1952),
254-270.

Wadstrom, C. B. An Essay on Colonialism. London, 1794.

 

Wansbrough, John. "Africa and the Arab Geographers."
Language and History in Africa. Edited by David
Dalby. London, 1970, 89-101.

 

Warhurst, P. R. "The Scramble and African Politics in Gaza-
land." The Zambesian Past. Edited by E. Stokes and
R. Brown. Manchester, 1966, 47-62.

 

Wheeler, Douglas. "Anti-Imperial Traditions in Portugal:
Yesterday and Today." Boston Universitinraduate
Journal, XII (1964), 125-137.

 

202

Wilson, M. "The Sotho, Venda, and Tsonga." The Oxford
Historypof South Africa. Edited by M. Wilson and
L. Thompson. Vol. I. New York, 1969, 131-182.

 

Wright, John K. The Geographical Lore of the Time of the
Crusades. New York, 1925.

 

‘mmTIT“