History I NEED HELP ON MY HOMEWORK

The Role of Foreign Trade in the Rozvi Empire: A Reappraisal Author(s): S. I. Mudenge Source: The Journal of African History, Vol. 15, No. 3 (1974), pp. 373-391 Published by: Cambridge University Press Stable URL: http://www.jstor.org/stable/180666 . Accessed: 19/09/2011 19:36

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp

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

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

http://www.jstor.org

journal of African History, xv, 3 (1974), pp. 373-39I 373 Printed in Great Britain

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE: A REAPPRAISAL'

BY S. I. MUDENGE

THE question of the relationship between external trade and the growth of political power in pre-colonial Africa has recently attracted an increasing number of scholars.2 Among students in this field there has been a growing tendency to assume, either implicitly or explicitly, that external trade always led to the formation or enlargement of states in pre-colonial Africa. Shorn of qualifications, the 'trade stimulus hypothesis' may be boldly summarized thus: trade allowed for a high degree of concentration of wealth in the hands of one man, usually the chief; this was possible because the chief always monopolized all external trade among his people; and this highly concentrated wealth allowed for the development and continued existence of more elaborate state systems than was previously possible. Often this 'trade stimulus hypothesis' has been applied with some justifi- cation. But lately some writers have tended to assume uncritically that wherever external trade was present, it must have been a major source of power. Indeed, in certain cases scholars have come dangerously close to implying that external trade represented all the economic life in an African society. Once the presence of external trade has been established, its consequences are then assumed; analysis of its actual impact on the political system and its relationship to internal production and consumption in the given polity becomes unnecessary, and all effort is expended in explain- ing the organization of trade routes and marketing and the types of goods involved. The hypothesis itself becomes the explanation of the role of external trade within the given polity. This runs the risk of turning the theory into historical juju, in which the problems are supposed to vanish when the faithful recite the necessary incantations.

In this paper, I will try to show how such a wholesale and uncritical application of the above hypothesis to the study of one polity, the Rozvi empire (c. I684-I833/4),3 has led to a serious distortion of the true picture

1 I wish to express my gratitude to Richard Gray and John Peterson for their valuable comments on an earlier draft of this article. This paper was first read at the Institute of African Studies, Fourah Bay College. Additional research for the paper was made possible through a generous grant provided by the Humanities Research and Equipment Fund of Fourah Bay College. The map reproduced here was made by A. A. J. Elba, carto- graphic unit, Fourah Bay College.

2 Among the many works published in this field may be cited K. 0. Dike, Trade and Politics in the Niger Delta (Oxford, 1956); G. I. Jones, The Trading States of the Oil Rivers (London, I963); D. Birmingham, Trade and Conflict in Angola (Oxford, I966); more recently R. Gray and D. Birmingham (eds.), Pre-Colonial African Trade: Essays on Trade in Central and Eastern Africa before I900 (London, I970).

3 The Rozvi empire discussed here was founded by a man known in oral tradition as Dombo, but referred to simply as Changamire in contemporary Portuguese documents.

374 S. I. MUDENGE

of that state. In their description of the basis of the power wielded by the Rozvi Mambos (emperors) previous writers have tended to over-emphasize the importance of external trade at the expense of the contribution of internal factors. We have been presented with a picture of the 'unwarlike Rozwi',4 presiding over such a loosely connected 'tribal confederacy' that the Rozvi emperors could not even exact tribute from their subjects except as a form of trade.5 Regarding the 'religious and military bases of the Mambo's power', one writer refers to 'the slenderness of that power politically.'6 Thus the picture is that of an empire where the coercive powers of the emperors are judged to have been rather slender. The only aspect of this empire that, thanks to the survival of contemporary Portu- guese documents, most of these writers accept is the existence of a vigorous external trade, mainly in gold. External trade was so vital to the Rozvi Mambo's position and influence that it was 'strictly controlled and confined by the Mambo or Changamire, who had a monopoly of gold, the main export'.7 It was this control of trade which provided the wealth with which the Rozvi Mambos could secure whatever influence they wielded over their so-called subjects. Yet one writer even casts doubt on the possibility o such influence by dismissing the very concept of a Rozvi 'empire' and its 'power and glory' during the eighteenth century as a 'myth'.8

Recent research allows us to question some aspects of this picture of the Rozvi empire. In particular, the claim that external trade was the major source of the influence wielded by the Rozvi Mambos now emerges as being somewhat exaggerated. On the other hand, the roles of such internal factors as military, religious and administrative organization are seen to have been much more significant than we had hitherto been led to believe.

I

THE POLITICAL AND ADMINISTRATIVE ORGANIZATION OF THE ROZVI EMPIRE

(i) Rozvi kingship The Rozvi empire was ruled by a hereditary ruler with the title of

Mambo. Succession to the Rozvi throne was collateral, or perhaps more precisely patrilineal adelphic, i.e. the kingship devolved from brother

For the argument that the name Rozvi and the Rozvi nation did not exist before Dombo's wars (I684-95) see S. I. Mudenge, 'The Rozvi Empire and the Feira of Zumbo' (un- published Ph.D. thesis, London, 1972), 35-43.

4 R. Summers, Zimbabwe: a Rhodesian Mystery (Cape Town, I963), 96. 6 A. K. H. Weinrich, 'Karanga History and the Mwari Cult', Cultures et developpement,

II, 2 (I969-70), 394-6; N. Sutherland-Harris, 'Trade and the Rozwi Mambo', in Gray and Birmingham, Pre-Colonial African Trade, 244, 246.

Sutherland-Harris, 'Trade and the Rozwi Mambo', 245. Ibid. 243; see also A. Isaacman, Mozambique: the Africanization of a European

Institution: the Zambezi Prazos, I750-I902 (Madison, Wisconsin, 1972), 82. 8 H. H. Bhila, 'The Manyika and the Portuguese 1575-I863' (unpublished Ph.D.

thesis, London, 1I97I), 71.

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 375

to brother, eventually reverting first to a son of the eldest brother and then to a son of the next brother and so on. But this was only the principle; in practice other factors played their part. Support at court or in the kingdom was equally important in determining who was to succeed. In I695-69 and again in I76810 we have contemporary Portuguese written sources showing that sons could and did succeed their fathers. On the other hand, as is to be expected, most of the traditional sources emphasize how faith- fully the patrilineal adelphic principle was adhered to. This form of succession can be most complex and often causes succession problems." This was certainly the case within the Munhumutapa empire in the seventeenth as well as the eighteenth centuries; compared to the Mutapa empire in the eighteenth century, which was plagued by succession diffi- culties, the Rozvi empire appears to have been almost a paragon of stability and orderliness.12 The only politically significant succession dispute in the Rozvi empire known to have taken place was in I768.13

It would appear that the Rozvi were able to minimize the dangers of succession wars to such an extent in part because of an institutional device they created to reduce the known harmful effects of succession rivalries. According to some Rozvi traditions, when a Mambo died and no successor was immediately appointed, the head of one of the royal Rozvi clans with the dynastic name of Tumbare would take over as regent until a Mambo was elected.'4 The house of Tumbare, as we shall see below, used to provide the highest general of all the Rozvi armies, who was also the Chief Collector of tribute, an office approximating that of Chancellor of the Exchequer. This post would appear to have been hereditary to the house of Tumbare and in effect made Tumbare the second most powerful personage in the empire after the Mambo himself. This constitutional mechanism allowed the princes to haggle, plot and manoeuvre for the succession without necessarily throwing the empire into a blood-bath. Above all, it meant that the claimant who could secure Tumbare's support had a very good chance of succeeding to the throne.

9 Biblioteca da Ajuda Lisboa (B.A.L.), 51-VII-34, fl.sI, "Proposta que fizerao os mora- dores dos Rios de Cuama ao Senhor V. Rey", Senna, 27 June I698, Manuel Rebello.

10 Arquivo Hist6rico Ultramarino, avulsos de Mocambique, Caixa (A.H.V.AV. de Moq. Cx.) 13 Tette, 20 July 1768, Inficio de Mello Alvim to Governor-General (G.G.).

11 See G. Kingsley Garbett, 'Religious aspects of political succession among the valley Korekore (N. Shona)', in E. Stokes and R. Brown (eds.), The Zambesian Past (Manchester, I966), 152-6. Garbett shows just how confusing the situation could become. See also J. F. Holleman, 'Some "Shona" Tribes of Southern Rhodesia' in E. Colson and M. Gluckman (eds.), Seven Tribes of British Central Africa (Manchester, 1951), 380-2.

12 E.g. between 1759 and 1766 no less than six Mutapas ruled. Each was in turn deposed. From 1766 to i8o6 there were sporadic wars between the ruling Mutapas and pretenders to their throne. It could also be argued that it was the seventeenth-century succession wars within the Mutapa empire which, to a large extent, enabled the Portuguese prazo holders to gain such a strangle-hold over the valley lands during that period.

13 A.H.U. AV. de Moc. Cx. I3, Tette, 20 July I768, Inacio de Mello Alvim to G.G. 14 E.g. S. Nenguwo, 'Oral work among the Rozvi: a few notes', in Conference of the

History of the Central African Peoples (Rhodes-Livingstone Institute for Social Research, Lusaka, I963), 6.

376 S. I. MUDENGE

300E

Zumbo

Tete.

Sena' I . -. . Bindura

MANGWENDE>L

MANYIKA1M'kt'

200S -<_

.

K.a MAHUNYJ (Feira)

Manyanga. (I *Dhlodhlo 'I Sofala 20S *,0Bulawayo S

z

Modern Boundary of Rhodesia --- 30 The Heart of Rozvi Empire

Miles Boundary of Rozvi Empire appr --

50 0 50 100 IS0

50o 0 S0 100 150 20

Kilometres

Although some scholars have suggested that the Rozvi rulers may have had attributes of divine kingship, I have myself found little evidence to support this.15 Constitutionally the Rozvi Mambo embodied the executive and legislative as well as the judiciary powers within the empire. But in practice he was assisted, at his court,16 by a Dare (Council) of magota (Councillors).17 Represented in the Dare were such branches of the Rozvi polity as the priesthood, the military leaders, and provincial governors as well as some of the leading Rozvi imperial houses. Although the Mambo would no doubt have had the last say on most issues, he nonetheless had to try to carry his Dare with him. Some of the magota, e.g. the priesthood, claim to have had the right and duty to advise, warn or reprimand a

15 Mudenge, 'Rozvi Empire', io6-8. 16 The Rozvi capital, according to tradition, was variously at Khami, Dhlo-dhlo and

Manyanga. 17 F. W. T. Posselt, Fact and Fiction (Bulawayo, I934), I57; Marodzi, 'The Barozwi',

Native Affairs Department Annual (NADA) (I924), 90.

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 377

Mambo who did not rule well. Many of the magota were hereditary. This meant that their position did not depend solely on the whim of the Mambo. All this points to the fact that although the Dare was primarily there to advise the Mambo it was also, in part, a constitutional check on possible excesses by the Mambo himself.

(ii) Rozvi military power and organization

Previous writers have greatly underestimated the military capacity of the Rozvi.18 A study of contemporary sources suggests a very different picture. According to one Portuguese document of I785, 'The Brozes [Rozvi] are feared even by the monhaes19 [vanyai], who after these Rozvi are more daring than anybody else.... Because of their great prowess and daring they [Rozvi] fight fiercely and furiously in battles.'20 And here is another somewhat biased description of the Rozvi, coming from I789:

The immediate subjects of this chief [Changamire] are much devoted to him. . . but they pass their whole lives in indolence of sensuality and the activities of spoliation, hold agriculture and commerce in contempt, and, thinking themselves a race superior to the rest of mankind, they consider work a degradation ... six or seven of these desperadoes, will ... intimidate two or three hundred Africans of other tribes.21 According to Abraham, the quotation ends thus: 'Such are the freebooters with whose aid Changamera has succeeded in making all his conquests and compelled the entire population of several districts to quit their habitations and fly to the northern side of the Zambezi.'22

Many more quotations could be given from the eighteenth-century Portuguese sources, all of which corroborate the image of the Rozvi as a fierce, powerful and warlike people. But before we can agree either with the Portuguese account of Rozvi fighting power or with the more recent notion of the 'unwarlike' Rozvi, we should look briefly at what is known of the Rozvi military machine. From contemporary Portuguese sources we learn that the Rozvi armies were organized into centuries like those of the Romans.23 A century or company of soldiers was known as a missoca (regiment).24 These missocas were under the command of cabos25 or field

18 Sutherland-Harris, 'Trade and the Rozwi Mambo', 245; R. Summers, Zimbabwe: A Rhodesian Mystery (Cape Town, i965), 96; Bhila, 'Manyika', 7I.

19 These were the soldiers of the Mutapas. 20 A.H.U. AV. de Mo9. Cx. zz, Tette, ii June I785, Antonio Manoel de Mello e Castro

to Martinho de Mello e Castro, Secretary of State for Overseas. 21 J. J. N. de Andrade, 'Descripcao de Estado em que ficavao os negocios da Capitania

de Mossambique nos fins de Novembro do anno de I789', in Arquivo das coldnias (Lisbon, I9I7), I, 95. Translation by D. P. Abraham, 'The Monomotapa dynasty,' NADA (I959), 75.

22 I have personally been unable to trace where this part of the quotation comes from. 23 J. J. N. de Andrade, 'Descripcao', 95. 24 A.H.U. AV. de Mo9. Cx. zz, Tette, i i June I785, Ant6nio Manoel de Mello e Castro

to Martinho de Mello e Castro. 25 F. de M. de Castro, Descripfpo dos Rios de Senna anno de I750 (Nova Goa, Imprensa

nacional, i86I), 32, para. 66.

378 S. I. MUDENGE

commanders. Above the cabos were the inhabezes26 (same rank as generals). The inhabezes were themselves below Tumbare, the highest general. And at the top was the Mambo himself who was the Commander-in-Chief. Although it seems to have been customary for the Mambos to lead their armies in person,27 we have oral and documentary evidence that this was not always done.28

The Rozvi army had an impressive variety of weapons: bows and arrows, daggers, assegais, shields, battle axes and cudgels.29 According to one Portuguese source, the Rozvi fighters used the cudgel 'with such deadly accuracy that they can throw it from a long distance away and the blow is almost always fatal'.30 They also possessed guns, and in the early nine- teenth century there were four small cannons at the Mambo's palace.3' Although by the later nineteenth century the reputation of the Rozvi as a people possessing guns reached as far south as the modern Transvaal,32 we have little evidence that guns were ever important weapons in the Rozvi armies.33

On the battlefield, the Rozvi used a fighting formation which closely resembled that of Shaka, the Zulu king. According to a Portuguese source in I758, the Rozvi used to deploy their men as follows: a main body of fighting men known as viatte would be in the middle, flanked by two horns called mulomo acumba.34 In the nineteenth century Shaka was to use this fighting formation with devastating effect. When the Rozvi squadrons were in battle they are said to have been arranged in such a way that, while one was engaging the enemy, another would be at its rear taking care of the wounded, replenishing and animating the vanguard, as well as making certain that none of those in front could retreat.35 It is also said that the Rozvi favoured close combat, face to face in an open field.36 But it should

26 Ibid. 27 Posselt, Fact and Fiction, I47. 28 F. de M. de Castro, Descripp7o, 32-3, paras. 66-7; G. Fortune, 'A Rozvi Text with

translation and notes', NADA (I956), 83-4; D. M. Colley, 'The Fate of the last Bashankwe chief', NADA (1927), 65-6.

29 A.H.U. AV. de Moc. Cx. 17, Anon., 'DescripfAo Corografica do Reino de Manica seus Custumes e Leis'; J. J. N. de Andrade, 'Descripcao', 95; A.H.U.AV. de Moc. Cx. 22, Tette, i i June 1785, A. M. de Mello e Castro to M. de Mello e Castro.

30 A.H.U. AV. de Mo. Cx. 22, Tette, i i June 1785, A. M. de Mello e Castro to M. de Mello e Castro.

31 A.H.U. AV. de Moc. Maco 2I, I June I83I, Anon., 'Acressentamento' (annexe) to the 'Memoria sobre a doafAo do Territorio Bandire'.

32 T. Wangemann, Geschichte der Berliner Missionsgesellschaft, Bd. IV: Die Berliner Mission im Bassutolande (Berlin, I877), 70.

33 See R. Gray, 'Portuguese Musketeers on the Zambezi'; 37. Afr. Hist., XII, 4 (1971),

531-3; A. Atmore, J. M. Chirenje and S. I. Mudenge, 'Firearms in South Central Africa', J. Afr. Hist. XII, 4 (1971), 55I-2.

3 I. C. Xavier, 'Noticias dos Dominios Portugueses na Costa de Africa Oriental', in A. A. de Andrade (ed.), Relafres de Mofambique Setecentista (Agencia Geral do Ultramar, Lisbon, I955), I45.

35 A. A. de Andrade, Rela_fes, I45; also A.H.U. AV. de Mop. Cx. 22, Tette, ii June 1785, A. M. de Mello e Castro to M. de Mello e Castro.

36 'Corografica', as cited in n. 29; also A.H.U. AV. de Moc. Cx. 22, Tette, i June I785, A. M. de Mello e Castro to M. de Mello e Castro.

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 379

not be thought that the Rozvi mode of fighting was inflexible: far from it. Elsewhere it has been shown that flexibility and resourcefulness were an integral part of Rozvi warfare.37 In I684 it was Dombo's unconventional tactics that enabled him to win the critical battle for Mahungwe. At any rate that is how Fr. Conceigdo saw it when he wrote of Dombo's victory thus: 'He was so wily and cunning that after being defeated by our arms, he vanquished us with his strategems.'38

The Rozvi seem to have had a formalized mode of mobilizing their forces. It is said that when the Mambo wished to summon his men of war he would sound his gun (or was it one of his four cannons?) and then Tumbare would beat his war drum known as Dittiwe.39 At the sound of Dittiwe, criers would go out shouting 'Chisadza Mhomwe [Muhomwe]' ('fill your bags with food').40 The mapfumo, i.e. squadrons, would then assemble at the court. When it comes to the training of the Rozvi armies we know very little. The Portuguese sources are not very helpful on this subject. For although they tell us of 'military drill' or 'exercises' as well as 'archery training', they never really explain what these involved. All we are told is that the Rozvi military exercises were called pemberafoens and these are said to have entailed 'many leapings and multi-articulated move- ments that they perform with their heads during battle or military exer- cises.'4' But pemberafoens-which comes from the Shona (pembera, kupembera, to ululate)-entails singing, chanting, ululating and brandishing of spears, and is not used for military training but as a morale rouser to build self-confidence and intimidate the enemy. It may well be that fight- ing skills were acquired largely through hunting.42

Before going into battle the Rozvi armies were always doctored. Tradi- tions abound claiming that the Rozvi used supernatural powers against their enemies. We are told that the Rozvi could change the colour of cattle, summon bees to fight for them if need be, and send their enemies to sleep by magic. They could make their warriors brave and/or bullet-proof or impossible to pierce with an arrow or spear.43 Even Portuguese sources remark on this reputation of the Rozvi. In I698 the Viceroy of India wrote that the Portuguese soldiers-black and white-in the Rivers of Sena believed that the then Rozvi Mambo had magic oil with which he could kill anyone simply by touching the person with it. The Viceroy implored the Portuguese king to send a new lot of soldiers from

37 Mudenge, 'Rozvi Empire', 74-90, II6. 38 A. da Conceicao, 'Tratado dos Rios de Cuama (i696)' in J. H. Cunha Rivara (ed.),

0 Chronista de Tissuary (Goa, I867), I05, para. io6. 39 Fortune, 'Rozvi Text', 78. 40 Posselt, Fact and Fiction, I56. 41 A.H.U. AV. de Moc. Cx. 22, Tette, II June I785, A. M. de Mello e Castro to M.

de Mello e Castro. 42 Mudenge, 'Rozvi Empire', I22-3. 43 Fortune, 'Rozvi Text', 8i, 83, 84; Colley, 'Last Bashankwe Chief', NADA (I927),

66; Conceicao, 'Tratado dos Rios de Cuama', io6, para. I IO; S. Muhlanga, 'In the Early Days', NADA (I926), IO9.

380 S. I. MUDENGE

metropolitan Portugal who would not believe in such superstition.44 The Portuguese in the Rivers of Sena had cause to fear Rozvi magic, because in I693 after the great Rozvi Mambo, Dombo I, had slaughtered all the Portuguese at Dambarare, he had two Dominican priests flayed and their heads cut off and carried in front of his army. On that occasion, it is re- ported, Dombo also disinterred the bones of some of the Portuguese and had them crushed in order to prepare a powerful medicine for his soldiers.45 Fr. A. Conceigao insists that Dombo was 'a most skilled wizard'.46 This association of the Rozvi with the supernatural clearly gave their armies a vast psychological advantage over their potential opponents.

Finally, we must consider the question of numbers. This is hard to determine. But from Portuguese sources we can deduce that the Rozvi armies were fairly large by the standards of the time. We find that in I684 immediately after Dombo had sustained heavy losses in his pyrrhic victory against the Portuguese at Mahungwe, he still had enough men to face an invasion from the emperor Mutapa Mukombwe. If we are to believe con- temporary Portuguese evidence, on the latter occasion Dombo killed no less than 5,000 men of Mutapa Mukombwe, as well as some of the latter's grandees and Portuguese allies.47 Again, in I743, when the Portuguese, now reconciled to the domination of the Rozvi, were in trouble around Tete and requested help from the Rozvi emperor, he sent, it is said, about z,ooo fighters to help them. But when the 2,000 reached Zumbo, the Portuguese authorities seem to have changed their minds, and so they asked the Rozvi men of war to return to their home. Before the army reached home it met some trading agents of the Portuguese traders of Zumbo and decided to rob them of 500 pastas48 of gold. When this was reported to the Mambo, the latter sent another army which is reported to have annihilated the first 2,000 Rozvi who had originally been sent to help the Portuguese.49 In I772 the Feira of Zumbo was blockaded by Ganiam- baze, a Mutapa prince fighting for the Mutapa throne against prince Changara. Again the Rozvi sent an army which relieved the Feira of Zumbo.50 In I78o-I the Rozvi are said to have sent another army of about 3,000 men to help the Feira of Zumbo which was under pressure from one prince Casiresire.51 Also in the I780s the Rozvi sent an army to the king- dom of Manyika to ensure that the king of Manyika did not harass the

44 B. A. L., 51-VII-34, fl. 35, Goa, I2 December I698, Ant6nio Luis da Camara Coutinho to king.

45 Concei-ao, 'Tratado dos Rios de Cuama', io6, paras. I09-IO. 46 Ibid. 105, para. I03. 47 B.A.L. 51-VII-34, fl. 50, Senna, 27 June I698, 'Proposta que fizerao os Moradores

dos Rios de Cuama ao Senhor V. Rey', Manoel Rebello; A.H.U. AV. de Mo. Cx. 2,

Senna, 24 July I685, Caetano de Mello de Castro to king. 48 A pasta was a weight of about I6j ounces. 49 F. de M. de Castro, Descripfao dos Rios de Senna, 32-3, paras. 66-7. 50 A.H.U. AV. de Mo. Cx. I4, Mo9ambique, I5 Aug. I773, Balthazar Manoel Pereira

do Lago. 51 Mudenge, 'Rozvi Empire', 275-8.

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 38I

Portuguese traders while the latter were building a fortress at Feira of Manyika.52 Traditional sources also furnish other instances of the use of the Rozvi forces.

Here we see not the 'myth of power'53 or the 'unwarlike Rozvi',54 but a well-organized and well-armed military force which during the eighteenth century permitted the Rozvi Mambos to protect the Feira of Zumbo and its trade route to Butwa; allowed the Rozvi rulers at least on one occasion to keep peace between the Portuguese and the king of Manyika; and as we shall presently see, enabled the Rozvi Mambos to levy tribute on their subjects. In short, the army provided the Mambos with important diplo- matic and political leverage which they used at times in order to indulge in extra-territorial interventionist policies both north and south of the Zambezi. In the words of one Governor-General of Mozambique it made the Rozvi Mambo 'the most powerful [ruler] of those interiors',55 or, as one secretary to the government of Mozambique says, it made the Mambo 'the terror of the hinterland'.56 To assert that the military base of the Rozvi Mambo's power was politically slender, is to run against the available evidence and to make the words lose all meaning.

(iii) Religion: Its Political Significance

That the functionaries of traditional Shona religion have historically played, and continue to play, a politically significant role in Shona society is a point few scholars would now challenge.57 Professor T. 0. Ranger has convincingly shown that in the crisis created by the imposition of colonial rule on the Africans of Southern Rhodesia in the late nineteenth century it was the officers of the Mwari cult and the Mhondoro (spirit mediums) who provided the grapevine-like network for the coordination of the revolt of I896-7.58 And Richard Brown has pointed out that in the nineteenth century Ndebele kingdom the Mwari cult co-operated with the Ndebele rulers and acted as a valuable secret service in that kingdom.59 Going back to the eighteenth-century Rozvi empire, some indications in oral tradition allow us to postulate a somewhat similar politically active role for the religious arm within the Rozvi empire.

52 'Corografica', as cited in note 29; also Bhila, 'Manyika', 6o. 53 Bhila, 'Manyika', 7I. 54 Summers, Zimbabwe: a Rhodesian Mystery, 96. 55 A.H.U. AV. de Moq. Cx. I4, Mo9ambique, I5 Aug. I773, B.M.P. do Lago. 56 L. A. de Figueiredo, 'Noticia do Continente de Mogambique e abreviada relagdo do

seu comercio', in L. F. de Carvalho Dias (ed.), Fontes para a Historia, Geografia e Come'rcio de Mofambique (Sec. XVIII), Anais, vol. ix, Tomo i (Junta de Investigacaes do Ultramar, Lisbon, I954).

57 In the present guerrilla insurgence in north-eastern Rhodesia, for example, the Mhondoro (spirit mediums) seem to be playing an important role on both sides in the battle to win the support of the rural Africans.

58 T. 0. Ranger, Revolt in Southern Rhodesia I896-7 (London, I968). 69 R. Brown, 'External Relations of the Ndebele Kingdom', in L. Thompson (ed.),

African Societies in Southern Africa (London, I969), 268.

382 S. I. MUDENGE

First of all we are told that the priesthood was represented on the Mambo's council.60 This enabled the Mambo and his council to exploit the religious authorities' more discreet sources of information about outlying provinces. Furthermore the priesthood is said to have been in- volved in the investiture of some of the vassal rulers.6' In this way the priesthood may have helped to sweeten the political act of vassalage in- volved in the investiture ceremony by hiding it behind the mystique and sanctity of their office. And by being so closely associated with the Mambo, religion became one more integrative factor within the empire. Yet we would be in error if we were to conceive of the Mwari cult as having been no more than a willing tool in the hands of the Rozvi rulers. Conflict between the religious officers and the Rozvi Mambos appears to have been a perennial feature of their relationship. This may have been so partly because the priesthood claimed to have had the right and duty to advise and admonish those Mambos who did not rule well.62 Thus the religious authorities may well have been a constitutional watch-dog on a Mambo's theoretically absolute power. In a negative way, the importance of the religious authori- ties as supporters of the Rozvi rulers is further revealed in the fact tha virtually all the known Rozvi traditions account for the fall of their empire as being a consequence of a rupture between the religious and political leaders in the empire.63

(iv) Provincial Administration

The Rozvi rulers accepted the principle of heredity in the administration of their provinces. This meant that many of the provincial chiefs were the hereditary rulers of their regions. The only Rozvi Mambo who does not seem to have cared about the principle of heredity was Dombo I, himself a usurper. But on the whole his successors respected it. The question that requires to be answered is, therefore, how the Rozvi were able to control these hereditary local rulers.

The first way in which the Rozvi controlled these rulers was through the investiture ceremony known as kugadza ishe (to invest a chief). According to one source, 'If the people chose a new chief without consulting their overlord [the Rozvi Mambo], it was said in common parlance that they had no chief, but the one selected or approved by the Warozvi was held to be the true chief'.64 This view is supported by many other sources.65 The process of choosing a new chief began at the local level where the practices of that particular chiefdom would be observed. Once a candidate had been locally chosen he either went in person to the Rozvi court to seek approval

60 E.g. Marodzi, 'The Barozwi', NADA (I92z4), 9O; Fortune, 'Rozvi Text', 8i-2. 61 Fortune, 'Rozvi Text', 8I-2. 62 Ibid. 63 E.g. Posselt, Fact and Fiction, I 58; A. M. Sebina, 'The Makalaka', African Studies,

VI, 2 (I947), 85-9I- 64 E. M. Lloyd, 'Mbava'. NADA (I925), 62.

65 E.g. Fortune, 'Rozvi Text', 82; Posselt, Fact and Fiction, I55-6; Weinrich, 'Karanga History', 395-6.

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 383

or sent a deputy to obtain the approval on his behalf. The Rozvi could and did at times withhold approval. They could even bypass a duly elected candidate and impose their own choice.66 But instances of this practice seem to have been rare. The investiture of a new chief took place in his own kingdom. It was performed by a representative of the Rozvi Mambo, who handed the regalia, which included black and white calico as well as a sheep- skin, to the new chief.67 Through the investiture ceremony the Rozvi Mambo conferred legitimacy and authority on the vassal chiefs. But, most important of all, by being invested by the Rozvi, a vassal chief proclaimed his subservience to the latter. There is virtually no evidence to show that before the I790S any vassal chief successfully challenged the Rozvi right to invest subject rulers.

The other way in which the Rozvi Mambos controlled the loyalty of their subjects was by the imposition and collection of an annual tribute (Mupeta wamambo).68 Tribute collection had important political and economic implications. Politically, the payment of tribute implied continued allegiance of a vassal ruler to the Rozvi Mambos and refusal was taken as indicating rebellion.

The chief tribute collector of the Rozvi empire was none other than the Tumbare himself.69 That a man of Tumbare's standing-the highest general of the Rozvi armed forces and possible regent to the Rozvi empire- should have been made responsible for tribute collection is a sign of the importance attached to the function. According to one source, if Chikanga, the king of Manyika, ever dared to stop paying tribute to the Rozvi Mambo, he 'without doubt, would suffer severe retribution' and the Rozvi Mambo would 'bring Chikanga to his senses ... because the said Changa- mire and his soldiers are of a temperament so vain that they cannot put up with defiance without seeking revenge and complete satisfaction'.70 Oral sources confirm that any vassal chief who did not pay his tribute was visited by Rozvi soldiers.71 Tribute was paid in cloth, beads, hoes, axes, gold, ivory, skins, cattle, tobacco, foodstuffs and whatever else the various regions could produce.72 Through this tribute the Rozvi Mambo acquired vast quantities of wealth from his empire.

Finally, another way in which the Rozvi used to control their empire was by the appointment of regional governors or representatives. This is of such significance that it is most unfortunate that at present we know so

66 D. P. Abraham, 'The Principality of Maungwe: Its History and Traditions', NADA (I95 i), 63.

67 Posselt, Fact and Fiction, 140; Lloyd, 'Mbava', 62-3; S. X. Botelho, Memoria Estatistica sobre os Dominios Portuguezes na Africa Oriental (Lisbon, I833), I68.

68 Hist. Mss. Misc. IRU4/1 /i, Rukara account, National Archives, Salisbury. 69 'Corografica', as cited in note 29. 70 Ibid. 71 Posselt, Fact and Fiction, I45; K. Robinson 'A History of the Bikita District', NADA

('957), 77. 72 Posselt, Fact and Fiction, I56; Robinson, 'History of Bikita', 8o; Nenguwo, 'The

Rozvi: a few notes', 6; Hist. Mss. Misc. /RU4/I/i, Rukara account; 'Corografica' as cited in note 29; Weinrich, 'Karanga History', 396.

25 AH XV

384 S. I. MUDENGE

little about it. But we have enough evidence to establish the existence of such officers. What we lack is precise knowledge of their functions. Dr D. Beach has found oral sources claiming that a Rozvi house by the dynastic name of Gwangwava used to settle quarrels among the Rozvi subject chiefs in the present Charter District.73 In Mahungwe at the court of Makoni the house of Tandi was permanently stationed as a representative of the Rozvi Mambo.74 A similar role for the priestly house of Nerwande may be postulated in Mangwende kingdom.75 Dr Weinrich claims that her researches have shown that the Rozvi empire 'was divided into pro- vinces, and each province was administered by a royal agnate whom the king appointed provincial governor'.76 Perhaps when Dr Weinrich publishes her researches more fully we will be in a better position to assess the true significance of these provincial governors. But even from the little we know it is already clear that these governors were an important link in the control of the Rozvi empire.77

II

EXTERNAL TRADE AND THE ECONOMIC BASIS OF

THE ROZVI EMPIRE

As we have already seen, previous writers have emphasized the importance of external trade as the chief source of the economic power and political influence wielded by the Rozvi rulers. They have also asserted that the Rozvi Mambos exercised a strict monopoly on all external trade. But on this point, too, the results of recent research have led to some modifications. While it is true that the Rozvi Mambos issued decrees78 which governed

73 D. N. Beach, 'The Rising in south-west Mashonaland I896-7' (unpublished Ph.D. thesis, University of London, I97I), 6i-2. 74 Ranger, Revolt, 289.

76 W. Edwards, 'The Wanoe: a short historical sketch', NADA (I926), 28; Abraham, 'Chaminuka', 43. 76 Weinrich, 'Karanga History', 396.

77 Mudenge, 'Rozvi Empire', I53-9. 78 One such decree forbade all non-Africans, Portuguese and Goanese, to enter Butwa

for the purposes of trade. Trade between the Rozvi and the Portuguese had to be carried on through African intermediaries known as vashambadzi. The only place where non- Africans were allowed to stay for trade in the Rozvi empire was at the Feira of Manyika, where the Chikangas, the Rozvi vassal rulers of Manyika, kept a vigilant eye on their activities. So strict were the rulers of Manyika with the Portuguese that on at least one occasion the Rozvi had to intervene to restrain the ruling Chikanga's harsh treatment of the Portuguese at the Manyika Feira. But the reason why the Rozvi rulers forbade the Portuguese to wander freely in their empire was not so much economic as political. As one writer has so aptly remarked, the lesson of the seventeenth century in Zambezia was that long spoons were needed to sup with the Portuguese. The Rozvi Mambos understood the lesson only too well. As if to underline the fact that the ban on non-Africans was motivated by political considerations and not by a desire to enforce a trade monopoly, any Rozvi subject was free to trade at the Feira of Manyika whenever he chose to. Although trade between the Portuguese at Zumbo and the Rozvi empire was carried on largely through the vashambadzi, there is no evidence to suggest that the Rozvi rulers forbade their own subjects to visit Zumbo for trade. The distance as well as the need to traverse foreign lands and the inhospitable country of the Zambezi valley, rather than any decree of the Rozvi rulers, probably discouraged visits by Rozvi subjects to Zumbo. And, in any case, the regular visits by the vashambadzi made the journey to Zumbo unnecessary.

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 385

some aspects of external trade in their empire, they do not appear to have restricted the whole trade to themselves. The whole case of the trade monopoly of the Mambos has rested on two or three Portuguese documents. One of these documents is a well-known general report in I750 by Francisco de Mello de Castro.79 According to this report, the Rozvi rulers forbade their subjects to trade in pure gold. They could only trade in the less valuable (i.e. impure) gold. Elsewhere I have shown that the gold sent by the Rozvi Mambos to the Feira of Zumbo during the eighteenth century was invariably 'ouro bruto', i.e. impure gold.80 Thus the monopoly reported by de Mello de Castro was not in fact about pure versus less valuable (presumably meaning impure) gold: the monopoly, if that was what it was, was about different grades of impure gold. Both the Mambos and their subjects traded in impure gold. Even with the most up-to-date means of state control, such a monopoly would have been meaningless, as it would have been virtually impossible to enforce. The second document which has been used to argue that the Rozvi Mambos had a monopoly of trade is another general report, written in I758 by Ignacio Caetano Xavier, which alleges that one chief Caroa, a vassal of the Rozvi Mambo, had found a big gold nugget but was unwilling to give it to the Mambo.81 Caroa threw the nugget, tied to a buoy, into a river and from time to time traded chipped pieces of the nugget to the traders of Zumbo (presumably on the pretext that it was worthless gold!).

The third document used to show that the Rozvi Mambos had a mono- poly of trade is a letter written by a captain of Zumbo in I769, on the basis of evidence he received from the envoys of a Rozvi Mambo.82 Clearly this is an important piece of evidence, and it is therefore necessary to quote it exactly: 'Furthermore the said Changamire had instructed them [his envoys] to say that he had ordered three messocas of Munhais to go and punish some of his bareiros for failing to report to him the discovery of new bares.' What the document establishes is that the Mambo was clearly interested in knowing what new bares his bareiros had discovered. But who were the Mambo's bareiros? The word 'bareiros' comes from 'bar', which in eighteenth-century Mozambique meant, among other things, a gold mine or a gold mining settlement or region. North of the Zambezi the bareiros were gold miners (African and Portuguese). We might thus sup- pose that the bareiros of the Rozvi Mambo were his gold miners. But here we are told that bareiros was an alternative term for 'chefes', i.e. chiefs.83 This implies that the bareiros were special chiefs who may well have been specially connected with the Mambo's gold mining. Beyond suggesting that the Mambos may have had their own royal mines, perhaps in the

19 F. de M. de Castro, Descrippfo dos Rios de Senna, 33, para. 69. 80 Mudenge, 'Rozvi Empire', 206, 229, 385-9. 81 Xavier, 'Noticias', in A. A. de Andrade, Relaf5es, 171-2. 82 'Inventario do Fundo do Seculo XVIII' in Mofambique Documentario Trimestral,

no. 84, 1955, doc. I65, pp. 90-7 Zumbo, I3 Mar. 1769, Manuel da Costa. 83 The document reads 'regulos e bareiros ou chefes'.

386 S. I. MUDENGE

districts immediately surrounding their capitals, worked under the control of some royal agents (bareiros), there is not much more we can safely deduce from the above document. Certainly we would be stretching the evidence if we postulated the existence of a strictly enforced monopoly of external trade by the Rozvi Mambos on the basis of the above document alone. Yet it is on the basis of two of the above documents that the Rozvi Mambo's strict 'state control' of gold trade has been largely based.

We can get a clearer picture of the problem if we look briefly at the trade system between Zumbo and the Rozvi empire as it emerges from Portuguese documents about Zumbo. African traders, known as vasham- badzi, traded throughout the Rozvi empire. They were itinerant traders who moved from village to village and chiefdom to chiefdom selling their goods. There were no special market days or market places where trade was conducted. There is certainly no evidence to support a view that all trading may have been held at the court of the Rozvi Mambo.84 The nearest we come to this possibility was in I803 when an attempt was made by the Portuguese of Zumbo to introduce a new method by which the trade of Butwa could conceivably have been controlled by the Rozvi Mambo's court. On this occasion, the Portuguese requested the then Mambo to send one of his grandees to Zumbo, so that he could accompany the Zumbo vashambadzi to and from Zumbo. The chief function of the grandee was to ensure that the vashambadzi did not trade in the bares of Butwa before they had presented themselves at the Mambo's court:

where the said Changamira, in his presence, ought to entrust the said Mussam- bazes85 with their pangana9oens86 to his respective bareiros, with the express command that as soon as the bartering was over, a report about the fazendas87 and the Mussambazes should be made to him [Changamira] and together with the same grandee they [the Mussambazes] should be sent back annually to this villa so that they could give the accounts of the panganagoens in the presence of the said grandee.88

The suggestion was never put into practice because in I804 Zumbo (Mucariva) was destroyed. The above document has sometimes been used to argue that the Rozvi Mambo had a monopoly of trade. But from other Zumbo documents coming before i803 it is clear that trading operations did not take place only at the Mambo's court but all over the bares of Butwa. These bares of Butwa were found virtually all over what is now Southern Rhodesia, stretching from Bulawayo to Bindura and from the Sanyati to the Save river. And the Zumbo vashambadzi traversed all that region, bartering for gold from village to village without necessarily ever visiting the Mambo's court.89

84 Sutherland-Harris, 'Trade and the Rozwi Mambo', 258. 85 Same as Vashambadzi. 86 Term meaning trade goods or loads. 87 Word used by the Portuguese in the Zambezi valley for cloth or goods. 88 A.H.U. AV. de Moc. Cx. 42, Mucariva, 29 Jan. I803, Manoel Francisco do Rozario. 89 A.H.U. AV. de Moc. Cx. 23, Mossambique, 3 Dec. I786, Manoel de Mello e Castro

to M. de M. e Castro; A.H.U. AV. de Mo9. Cx. 37, 9 Oct. i8oo. Pedro Ant6nio Jose da

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 387

Perhaps one of the most revealing pieces of evidence against the view that the Mambo had a monopoly of all trading operations, and especially the gold trade, is provided by the fact that among the items in which vassal chiefs paid tribute were gold, ivory, beads and cloth.90 To pay one's tribute in gold and ivory implies the possession of these items before pay- ment. And payment in beads and cloth indicates an ability to engage in foreign trade, since these were the chief returns from the external trade in gold and ivory. A policy of total monopoly would have been difficult and expensive to enforce, and was probably an inefficient means of getting the gold. In a monopoly situation, the subject chiefs would have had little incentive to dig for much gold. On the other hand, getting the gold through tribute was less difficult and served a dual purpose-economic and political. The Mambos had a reasonably good system of enforcing payment of tribute by their vassals, but they would have needed a much bigger bureaucratic machine than the one we have described above to have had an enforceable and workable monopoly in external trade. Through this system, the profits of external trade filtered through the Rozvi political machinery from the bottom to the top. By allowing his subjects to trade, the Mambo avoided rousing undue resentment, but through tribute collection he benefited from that trade.

Now that we have cleared some of the impediments to a proper assess- ment of the role of foreign trade in the Rozvi empire, we should try to evaluate the contribution this made to the Mambo's power base. There is no doubt that Rozvi rulers greatly valued their trading links with the Portuguese, especially those at Zumbo. Evidence for this has already been seen in the fact that in I743, I772, and I78I the Rozvi sent powerful armies to protect the Feira of Zumbo from its enemies.91 The Rozvi Mambos also protected the vashambadzi against robbery while trading in Butwa. This policy of the Rozvi prompted one governor-general of Mozam- bique to write, 'Changamira, the most powerful [chief] of those interiors, is a man without many defects, except for his colour and paganism; because [if he had many defects] it would not be easy to understand his trading policy and the reason why he is able to make others obey him and love the Portuguese' (my emphasis).92 It is doubtful whether the Changa- mire's 'saintliness' was the sole reason for his conduct. Enlightened self- interest seems to have been more important. From the Portuguese the Rozvi received luxury goods, e.g. different varieties of beads, sombreiros, sea shells, rosaries of fake coral, handkerchiefs, chinaware, brass bells, scissors, candles, cloth of various colours and aguadente ('fire water', a

Cunha to D. Rodrigo de Sousa Coutinho; A.H.U. AV. de Moc. Cx. 21, Zumbo, 20 June I784, C. de Sousa to Governor of the Rivers; See also Annais do Conselho Ultramarino, parte nao official, Serie II, I86-7, 'Mappa das minas conhecidas no Districto de Senna', Senna, 30 July I857, Izidoro Correia Pereira.

90 See note 72 above. 91 See notes 49-51. '2 A.H.U. AV. de Moq. Cx. I4, Mocambique, 15 Aug. 1773, B.M.P. do Lago.

388 S. I. MUDENGE

kind of brandy).93 These exotic goods no doubt made life at court more colourful for their imperial majesties. But these were luxury goods which, as the documents show, came in small quantities. At the time of the investi- ture of a new chief, the Rozvi Mambos used to send black and white calico as part of the regalia of the new chief. Clearly therefore imported cloth played an important part in the Rozvi political system. But it can hardly be said to have been vitally important for the survival of the Rozvi empire. Besides the question of investiture, there is little else we have seen to warrant even the suggestion that the wealth of the Rozvi kingdom was based on external trade.94

It is also apparent that not all the luxury items consumed in the Rozvi empire were obtained from trade with the Portuguese: some were internally produced. From the few items archaeologists have been able to recover since the destruction and pillaging of the stone ruins of Rhodesia around the end of the nineteenth century, it has been deduced that the Karanga craftsmen worked iron, copper, bronze and gold. They could draw wire, make bangles, manufacture knives, razors, rings, bracelets, chains, earrings and so on, using these metals. In addition, they carved various objects in ivory and soapstone, and made beads from ostrich eggs and other shells. Since well before the Portuguese ever came to eastern Africa, the Shona have woven a highly prized cotton cloth known as machira.95 While it is true that the Rozvi economy was a largely agricultural one, it would be an error to assume that externally acquired luxury trade goods always have the same effect on all forms of subsistence agricultural economies, irrespec- tive of the precise nature of the mode of economic production attained within the state in question. In the Rozvi kingdom, some of the external luxury goods, like beads and cloth, had equally valuable internal substi- tutes.

We may in fact conclude that agriculture and pastoralism, especially the latter, appear to have been the real economic basis of the Rozvi empire. The Shona people, as one seventeenth-century Portuguese writer observed,

will not exert themselves to seek gold unless they are constrained by necessity for want of clothes or provisions, which are not wanting in the land, for it abounds with them, namely millet, some rice, many vegetables, large and small cattle and many hens. The land abounds with rivers of good water, and the greater number of the Africans are inclined to agricultural and pastoral pursuits in which their riches consist [my emphasis].96

9 For the detailed examples of the goods the Rozvi used to receive see Mudenge, 'The Rozvi Empire', 385-9.

94 Sutherland-Harris, 'Trade and the Rozwi Mambo', 264. 95 D. Randall-MacIver, Mediaeval Rhodesia (London, I906, reprinted I971), 45, 49,

58; K. R. Robinson, Khami Ruins (Cambridge, I959), I44-58; P. S. Garlake, Great Zimbabwe (London, I 973), III -3 I.

96 A. Bocarro, 'Decada', in G. M. Theal (ed.), Records of South Eastern Africa (R.S.E.A.) (Cape Town; reprinted I964), III, 355.

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 389

Comments on the economic importance as well as the number and size of the cattle found in the kingdom of Butwa are scattered in Portuguese documents from the sixteenth century to the nineteenth century. For example, in I506 Diogo de Alcaqova wrote that in about I490 a ruler of Butwa offered about 4,ooo hornless cows to the Munhumutapa97; and in I569 Fr Monclaro informs us that the cattle of Butwa were 'as big as the large oxen of France'.98 But according to Fr Gomes, so big were these Butwa cattle in the mid-seventeenth century that one had to stand when milking them.99 The importance of cattle to the economy of Butwa was underlined by Fr Joao dos Santos who, writing at the end of the sixteenth century, claimed that the people of Butwa were less interested in digging for gold because 'they are much occupied with the breeding of cattle of which there are great numbers in these lands'.100 A further indication of the numbers of Butwa cattle is provided by the case of one mid-eighteenth century mushambadzi from Zumbo who collected as many as 8oo head of cattle from his trading activities in Butwa.101 Even in the early nineteenth century Butwa cattle were still being sold at the Feira of Manyika to the Portuguese.102 The Rozvi used their cattle at that period both as baggage animals and for riding,103 in addition to using them for meat and as a form of storing wealth. Cattle were useful to pay the bride present (rovora). Above all, a man with many cattle could use them to increase his influence by the system of kuronzera, in which a man gives some of his cattle for safe keeping to another man. The person to whom the cattle have been ronzerwa can use the cattle in virtually any way he wishes, except that he cannot dispose of them without the owner's consent. Because the owner of the cattle can, if he wishes, take back his cattle, this gives him influence over his 'client'. It would appear that the first Rozvi Mambo, Dombo I, himself remembered in Portuguese sources as a notable cattle baron,104 may have risen to power and founded the Rozvi empire on the strength of his cattle-keeping activities and his ability to redistribute the cattle through the kuronzera system.105

97 Diogo de Alcacova to King, Cochin, 20 Nov. 1506, in Theal, R.S.E.A. i, 64. 98 Fr. Monclaro, 'Viagem q'fizerao os p.es da Companhia de Jesus com Franco Barretto

na conquista de Monomotapa no anno de 1569', in Theal, R.S.E.A., III, 237 and 227. "' E. Axelson (ed.), 'Viagem que fez o Padre Ant.0 Gomes, da Companhia de Jesus ao

Imperio de de [sic] Manomotapa; e assistencia que fez nas ditas terras de. Alg'us annos' [addressed to Padre Joao Marachi, S.J., Verc&-Salcete, z January 1648], Studia, III (I959),

'97. 100 J. Dos Santos 'Ethiopia Oriental', in Theal, R.S.E.A., VII, 274. 101 'Fundo do Sec. XVIII', in Mopambique, no. 84 (I955), 90-7, doc. i65, Zumbo,

I3 Mar. I769, Manuel da Costa. 102 G. J. Liesegang, 'Beitriige zur Geschichte des Reiches der Gaza Nguni im sudlichen

Mo:ambique, I 820-I 895' (Ph.D. thesis, K6ln, I 967), 36. 103 'Acressentamento', as cited in note 3I. 104 'Corografica', as cited in note 29, Conceicao, 'Tratados dos Rios de Cuama', I05,

para. I03. 105 In this respect cattle may have played not a dissimilar role for Dombo I as they

were to do for Moshoeshoe I, who according to P. B. Sanders, gained his supremacy over the other Sotho chiefs mainly because he possessed more cattle than they, and used the

390 S. I. MUDENGE

In conclusion I would like to stress that this article is not meant to be a general attack on the importance of external trade (the so-called 'trade stimulus hypothesis') in the rise and development of some pre-colonial African states. It is a corrective of its uncritical application. Some of its exponents do not always investigate the true nature and development of the modes of internal economic production attained in each state under investigation. They fail to appreciate the fact that the role and impact of external trade are to a large extent determined and defined by the socio- political and economic factors obtaining in each state. Instead they tend to concentrate on the description of the types of goods traded, their marketing and such other aspects of the organization of external trade. They rarely analyse the actual impact of external trade on a particular state under consideration. Such impact is often assumed a priori.

In the case of the Rozvi empire this approach has led to a distorted picture of that polity. A closer examination of the Rozvi empire reveals that most of the externally acquired luxury goods had equally valuable internal substitutes. And, in any case, the socio-political significance of such wealth does not appear to have been nearly as great as that of the legendary cattle of Butwa. These cattle were the true savings banks and insurance policies of the ordinary Rozvi, which the Rozvi rulers used above all to distribute for safe-keeping to their subjects, who were then entitled to the usufruct. In this way, cattle provided visible social, economic and political bonds between the Rozvi rulers and their subjects.

The Rozvi empire was able to benefit from and enjoy the fruits of external trade as well as to ensure the safe flow of that trade largely because of its power, much of which was internally generated. Thus the Rozvi example underlines the fact that if one is to make a meaningful assessment of the impact of outside forces on any society, be it African or otherwise, one must first of all look at the level of social, political and economic development attained by that society before coming to any conclusion.

SUMMARY

Recent studies by the historians of pre-colonial Africa have tended to assume that external trade has always led to the formation or enlargement of states, and was crucial for the continued existence of these states. An example where an uncritical application of the above 'trade-stimulus hypothesis' has led to some distortion of reality has been in the study of the Rozvi empire in Southern Rhodesia in the eighteenth century. Previous students of the Rozvi empire have claimed that the latter was such a loosely connected tribal confederacy that its internal power bases-given as military and religious-were politically so slender that on their own they could not have sustained whatever power the Rozvi ruler wielded. Instead, it is said that the main source of the power exercised by the

cattle through the Mafisa system (the Sotho equivalent of the Shona kuronzera) to increase his influence. See P. B. Sanders, 'Sekonyela and Moshweshwe; failure and success in the aftermath of the Difaqane', in Y. Afr. Hist. x, 3 (I969), 439-55.

THE ROLE OF FOREIGN TRADE IN THE ROZVI EMPIRE 39I

Rozvi Mambo came mainly from the latter's ability to redistribute the profits of external trade, especially that of the gold trade of which he is said to have had a strictly enforced monopoly.

This paper tries to show that the claim that the Rozvi empire was based mainly on foreign trade is not supportable on the basis of the evidence so far advanced. To do this the paper challenges the view that the Mambo had a monopoly of trade. It shows that this conclusion is arrived at through a mis- understanding of the sources. In reality, the Rozvi rulers took part in external trade and influenced it, but they did not possess a controlling monopoly. Rozvi subjects were also free to take part in external trade, provided they obeyed the Mambo's decrees for regulating and restricting possible subversive activities by foreign traders in the Rozvi empire. The picture of the Rozvi empire that emerges from contemporary Portuguese records and oral evidence is that of a relatively stable, virile and powerful polity which from time to time used to despatch expeditionary forces to regions well beyond its territorial boundaries in order to maintain law and order to protect Rozvi trade and political interests. The strength of the Rozvi empire is seen to have come from internal rather than external fac- tors. Cattle rather than external trade appear to have been the most important economic power base of the Rozvi rulers.