The African City: A History / Edition 1 available in Paperback
This book is comprehensive both in terms of time coverage, from before the Pharaohs to the present moment and in that it tries to consider cities from the entire continent, not just Sub-Saharan Africa. Apart from factual information and rich description material culled from many sources, it looks at many issues from why urban life emerged in the first place to how present-day African cities cope in difficult times. Instead of seeing towns and cities as somehow extraneous to the real Africa, it views them as an inherent part of developing Africa, indigenous, colonial, and post-colonial and emphasizes the extent to which the future of African society and African culture will likely be played out mostly in cities. The book is written to appeal to students of history but equally to geographers, planners, sociologists and development specialists interested in urban problems.
About the Author
Bill Freund is Professor of Economic History and Development Studies at the University of KwaZulu-Natal. He is the author of many books, including The African Worker (Cambridge University Press, 1995).
Read an Excerpt
Cambridge University Press
978-0-521-82109-4 - The African city : a history - by Bill Freund
Urban Life Emerges in Africa
In the preface and, by implication, in the title of this chapter, I have suggested that cities evolve; I accept an essentially evolutionary model in my analysis of what follows. It is possible to argue rather that there are or were uniquely African kinds of cities or towns, before the incorporation of Africa into world-systems with wide economic networks and defining urban cultural structures, as a cultural statement about Africanness. However, the assumption here is that there are several reasons why urban life emerges anywhere: environmental, ritual, political, and economic, all of which will be examined in more detail. This would be true for any major area in the world and up to a point may come together in very different and quite unique combinations. This volume will emphasise that the evolutionary model needs to be modified to an important extent by the incorporation of earlier elements into later urban development, just as forms of rural settlement may be carried into urban ways of living. Old cities are inevitably accretions with layers that survive from their past, cultural if not physical. The division of this African urban history into chapters that look at the impact of incorporation into the beginnings of a world-economy, at colonialism and at the post-colonial situation, tries to give character to evolutionary change, but it does notmean to suggest that there are no continuities from one phase to another.
Africa is an ideal setting for studying the beginnings of urbanisation. In many regions of the African continent, the rise of new towns belongs to the relatively recent historical record, while a considerable number of archaeological excavations have been aimed at trying to discern the nature of urban life in the past, at just how, when, and why it developed. This chapter considers the information available on the character of early urbanisation and highlights what we are able to surmise about towns and cities at different phases of their development. It will involve making some mighty leaps in terms of distance between times and places.
This chapter will not proceed chronologically; its first pages will move northwards in space and focus on different early types of urban settlements, settlements with little evidence of influence from outside Africa. Such types are far from being mutually exclusive, but the examples given are meant to highlight particular aspects more clearly. On a time frame, we will go back as far as five thousand years ago to the Old Kingdom of Egypt – but, where such settlements show little evidence of influence from the increasingly commercialised global networks in touch with the West, some of my examples are from a relatively recent date, including the first example I will be giving.
As one of the authors cited below, anthropologist John Peel, has written about the large Yoruba urban settlements of south-western Nigeria, particularly as they may have been before the nineteenth century, they often seem to defy simple categorisation on a rural-to-urban continuum. The link between urban places discussed has to be understood as conceptual rather than linear in order to do justice to the structure of ideas that follow in the next pages. Thus we will often interrogate why and to what extent these places are urban. We shall move through a number of descriptive sequences before taking stock by looking at the types of urban structures as a whole. It is important to stress that if they fail to meet certain contemporary criteria of what a city should be like, such settlements should not be dismissed but rather embraced with interest for their unique configurations and contribution to the cultural development of mankind.
In the second half of the chapter, the narrative will come closer to following a conventional order in time, and major outside influences assimilated into African experience – Greek, Punic, Roman, and Islamic, as well as earlier sub-global world-systems – will be taken on board. Here the urban becomes incontestable: the economy became more varied and involved intensified specialisation. Urban life had to be sustained by systematic agricultural surplus from outside in part determining urban-rural relationships. A definite and distinct urban culture within the system emerged. Whether by incorporation, conquest, or other means of change, this did represent an evolutionary shift in North Africa in the first three cases and far more widely in the fourth. The characteristic urban features of these systems will be highlighted in the descriptions of the second half of the chapter.
MAP 1. Old African towns and cities.
|Image not available in HTML version|
We shall in fact begin at a fairly late point on the time frame. In southern Africa – at the western edge of where Bantu-speaking agriculturalists settled in what is today Botswana – there have for several centuries been surprisingly large human agglomerations. Contemporary social scientists who specialise in the study of this area have called them “agro-towns.” These agro-towns may have contained ten to twenty thousand people before the coming of colonial rule, although the evidence also suggests that they expanded very substantially in the context of the insecure and unstable conditions of the nineteenth century. Shoshong, capital of the Ngwato state in the nineteenth century, may have attracted thirty thousand people, although they were rarely all present in the town at once. Kanye, Serowe, and Molepolole in present-day Botswana are surviving examples of this phenomenon.
Europeans were astonished at the size of Dithakong, the most southerly such town. When they encountered it at the start of the nineteenth century, it was as large as the colonial capital of Cape Town. Although the agro-towns were invariably the core of important Tswana chiefships, it is striking to note that further to the east, in somewhat wetter country, closely related Sotho speakers showed little sign of taking to such large settlements. Nor did they exist in the time equivalent of the European Middle Ages when Tswana speakers apparently first settled in Botswana. It was only somewhat later, after a period of desiccation and retreat eastwards followed by resettlement after the year 1500, that this kind of unusual settlement pattern emerged.
There is no straightforward explanation for why this happened. The agro-towns are certainly emblematic of the power of chiefs gathering together a variety of people under their sway. In fact, the structure of the towns resembled a series of villages based on descent and affiliation to a chief or elder; a distinct feature was the space for the kgotla, a communal and ceremonial meeting ground that virtually defines what community means to the Tswana. But Tswana chiefs were not immeasurably powerful, nor was this the only way a chief in Africa could gain submission. Up to a point, defense may have been a factor in the gathering of large numbers of people. The sheer size of the agglomerated population in open country could represent a formidable deterrent to an invading band of some sort. Similarly, concentration was certainly related to ecological choices. Good water supplies, the presence of a remarkable hill, were typically features of the large settlement. In no way, however, could the Tswana economy be said to be so rich as to have supported urbanisation in the sense of a surplus that could sustain many non–food producers. Family members, and notably the women, had to scatter widely to farm and to gather foodstuffs. Young men spent most of their time living at cattle posts which were maintained at a long distance from the town, often on land which was too dry to support agriculture.
Concentration created problems and, as a result, until the changes of the nineteenth and twentieth centuries, which lent commercial and administrative purposes to towns that they had never before enjoyed, they died after a time. In the words of Neal Parsons, the size of Tswana towns accelerated the course of a cycle, exactly like that known to villages, where the town had to be moved and the population shifted, often with considerable frequency. Tswana urbanisation promoted a “cycle of depletion of local grasslands, cultivable soils, wood and water supplies.”1 In particular, Parsons believes that the efflorescence and subsequent disappearance of the Tswana town closely correlated to the depletion of wood resources in its vicinity. Thus Shoshong died to be replaced by Phalatswe – Old Palapye – at the close of the 1880s. Dithakong had long disappeared by then. The agro-town structure did not lend itself without outside stimulus to the emergence of economic activities that were specifically urban in character. However, we need to respect as one facet of human evolution this rare but not unique kind of preference for a herding and farming people to choose to live and develop their sense of community in settlements the size of large towns.
If we look for analogies in southern Africa, for a pattern with which to configure Tswana urbanisation, there is one possibility that stretches back much further in time. Over a period of some centuries, stone construction of an impressive nature took place at relatively permanent settlements in south-central Africa, mainly in modern Zimbabwe – which derives its modern name from what seems to be a Shona word for such settlements (madzimbahwe, chiefs’ residences) – but extending into the territory of Botswana. Less architecturally impressive stone construction in fact was undertaken extensively all over the South African Highveld, continuing into recent times. Most of this construction is accompanied by the traces of relatively small settlements, villages, but not all. Some clearly mark fairly large communities. The oldest site associated with this pattern is at Mapungubwe, in the Limpopo Province of South Africa near to the river of that name. Mapungubwe is a hill site with some impressive graves and beautiful art objects that can be associated with the beginnings of a gold trade to the coast more than a thousand years ago. The residence of a royal family or clan atop the hill is a remarkable feature. But as an urban community it seems to have been quite small and limited.
Further north are the ruins of what we call Great Zimbabwe, not far from the modern town of Masvingo, Zimbabwe. Here there are esthetically awesome remains – a beautiful circular tower, high walls sometimes shaped to allow for step construction, ornate walling patterns representing impressive man-hours of labour, construction above the valley site on a hill that earlier archaeologists dubbed the Acropolis, where daily requirements would have to have been brought laboriously by porters. There is much that we shall never know about these ruins, but there are some points concerning Zimbabwe on which scholars seem to agree. One is that the valley contained a dense community of mud- and wood-constructed homes – many plastered in their heyday – where people once lived. Up to fifteen thousand of them may have been resident at once on a site of seven hundred hectares.2 David Beach has imagined this as “a great mass of packed huts that spread across the valley in between the marshes and up the hillsides on terraces...basically a mid thirteenth to fourteenth century construction.”3 The stone remains were certainly not houses. The walls, of little use for defense, may at most have served to block off activities, perhaps sacred activities, from the mass of the population.
We will never know exactly what purpose the different structures served. The archaeologist Tom Huffman has made a series of creative suggestions based on his study of the sacred iconography of Venda-speaking people who live south of the frontier in South Africa and who may maintain more of the older culture of Great Zimbabwe than the people who live closer to the ruins today. At the least, they suggest an added element we need to consider, the importance of the sacral: urban sites can serve as ideal placements for ceremonies and activities that link men to ancestors and to gods in a way that knits together a “community.” The sacred element has clearly sometimes been of real significance in explaining the roots of urban agglomeration and may have been the main element here. If Huffman is right, sacred activities associated with a powerful chiefly dynasty, or dynasties, were more important than any economic impulse in bringing together this large community, which was comparable to the big Tswana agro-towns in size.
Historians feel confident that Great Zimbabwe was the centre of a state that traded gold with the Indian Ocean coast, and that it was a city in effect linked by an umbilical cord to Kilwa, which controlled this trade at its height, and which was the most impressive urban community that developed on the coast of East Africa before the coming of Europeans. We shall look at Kilwa later in the chapter. But though there are traces of long-distance trade in the ruins of Great Zimbabwe (Indian beads, a Persian bowl, Chinese porcelain), they have an extraneous quality – shards of pottery and coins rather than the unveiling of a real merchant quarter or any site that indicates intensive commercial activity. The decline of this remarkable settlement may have to do with shifts in the gold trade, although this is not a settled point amongst scholars. Nor is it really clear how the gold trade impacted on the strengthening of political power in the region.
Scholars have generally concluded that the large population concentration at Great Zimbabwe, despite all the human effort that went into the moulding of its stones, was not sustainable after a time, just as with the Tswana agro-towns. The valley in which the ruins are situated seems to have become agriculturally barren through intensive cultivation and/or climate change. It was abandoned by around 1450, perhaps after two hundred years of settlement, and no community of any size was thereafter reestablished there. Whatever environmental reason there may have been for settling this site initially – if one existed – it was an evanescent one.
There are numerous other similar ruins in the region, notably further west in increasingly dry parts of Matabeleland and into northern Botswana, but the stonework there was not so extensive or impressive and the scale of settlement smaller. Beach suggests that they were cultural outliers, perhaps established by fragmenting branches of chiefly families, with less and less wealth or link to overseas trade. This seems even truer of the Shona chieftaincy capitals described by Portuguese visitors from the sixteenth century, generally further north. The Mutapa dynasty frequently built stockaded towns with little or no stoneworks and more emphasis on defense; this was the ruling power that profited from the gold trade in its later centuries of existence. Great Zimbabwe was perhaps a kind of urban experiment that failed rather than evolving along a path of greater complexity and sophistication. Perhaps the Tswana agro-town idea was influenced by this type of larger settlement in some way and represents its only later successor.
São Salvador and Gondar
Within a century or so of the end of Great Zimbabwe, contact with Europeans was a factor in the making of other impressive urban sites in the African interior. However, the urban character of these sites remained incomplete. One example was Mbanza Kongo, the capital of a large and powerful state south of the Congo River within modern Angola which traded – especially for slaves – with the Portuguese from the fifteenth century. The Portuguese were very interested in the Kongo kingdom as an ally and made determined efforts to assimilate it to a European model, in particular to Christianise it. The royal family patronised a literate Christian culture for generations, and willingly took on some forms of European statehood in the eyes of their trading partners. The capital, impressively sited on a mountainous plateau, attracted mainly retainers of the royal house but also became the site for a set of Christian buildings, constructed from stone under the direction of Europeans at first – but for some time with cadres of Africans who had mastered relevant building techniques – and housing monks and priests. It probably reached its apogee in the middle of the seventeenth century.
To Europeans, it was the passably noble city of São Salvador. Facing the plaza, accompanying a cathedral and palace, was a small walled town inhabited by the Portuguese. To Africans, Mbanza Kongo remained a place where narrow paths ran between walled compounds with enough space for livestock and garden farming but with room for agriculture as town and countryside merged into one another. Even the palace was not inhabited for very long; it was probably inconvenient and dirty compared to the typical large compound structure. Royal patronage remained so much the basis for economic activity that private property in particular locations seems never to have developed. There was little space for even rudimentary urban civil society, yet the aura of sacred power associated with this site long outlived the secular decline of the kingdom after the end of the seventeenth century, and the physical locale retained a cultural importance as a remnant of its urban glory long after the political significance of Mbanza Kongo had faded. “Mbanza” has long had, perhaps even before the creation of this town, an association with urbanity and civility in the Kikongo language, distinguishing it from village life. Yet beside the remarkable stone facades that spelt civilisation to Europeans, an African community with different associations had actually breathed life into São Salvador and made it a vital place shared by people. Thus the eighteenth-century city, which remained of considerable regional sacred and even political importance, with population concentrations at times as large as in the past, continued to be identified with the old stone buildings, even whilst they fell increasingly into ruins.
Thousands of kilometres from Mbanza Kongo, another Portuguese-influenced stone city was rising on the Ethiopian plateau – Gondar. The emperor Fasilidas, who reigned in the second quarter of the seventeenth century, despite the expulsion of the Jesuits and the restoration of Coptic Orthodoxy as the church of Ethiopia, was responsible for authorising the construction of palaces and churches at this town one hundred kilometres or so north of Lake Tana. By contrast with Mbanza Kongo, this construction process continued over several generations. This was an apparently flagrant departure from Ethiopian history, where no permanently constructed capital had existed for a thousand years.
The Ethiopian highlands have been the home of an indigenous class society of lords and peasants for many centuries; however, although trading settlements have undoubtedly existed for a very long time, urbanisation was a weak force. What happens if one does look back a further one thousand years from the time of Fasilidas? Parallel to the existence of the late Roman Empire, a kingdom which converted to Christianity and where a Semitic language ancestral to present linguistic uses was dominant, focussed on the city of Aksum. Aksum remained of great importance for many centuries and reemerged as a Christian centre and market town in medieval and later times. However, the archaeologist David Phillipson has recently concluded that whereas “Aksum was of a size and importance to merit the term ‘city ’...there is no evidence that Aksum was a city as that term is sometimes understood. Its structures, as presently known, comprised large buildings of unknown purpose, but clear elite associations, as well as funerary and other monuments, and religious buildings.”4 So far as we know, ordinary folk lived at some distance from this sacred core near to cultivable fields. Aksum lacked defensive walls. The commercial life of the Aksumite state must have been considerable (it minted coinage), but perhaps not much of it went on within this early African city; its political and sacred role doubtless reflected yet older Ethiopian traditions that have been discerned only vaguely so far by archaeologists.
Thereafter, later Ethiopian rulers travelled constantly with their courts in order to keep control of their subjects; they did not try to contain unruly subordinates within a town’s wall. Ethiopian Christianity focussed particularly on monasteries rather than urban cathedrals. Before Gondar was constructed, the court would often consist of hundreds of tents housing camp followers. Not only was this understood as a means for the court to impress its authority on key regions, it had as well an environmental logic. The weight of this kind of crude natural exploitation of foodstuffs, timber, and other products fell only seasonally on particular localities if the court moved about, and in absentia, the region was allowed to recover. Less impressive buildings by far marked the existence of regional market centres and towns which have attracted relatively little attention from historians of Ethiopia.
Gondar itself was important for its association with royalty and, through royalty, with the church, rather than for particularly significant commercial activities. A sacred iconography detailed this importance. Donald Crummey’s recent study records no less than eleven important royal churches in the vicinity. He has also recently insisted that the change initiated by permanent construction in Gondar was less dramatic than meets the European eye. Seventeenth-century “Gondar was above all a winter residence, a place where the court, and its ever-widening circles of minions and dependents, passed the season of the rains.”5 The ambulant nature of the state continued after the rise of permanent construction in Gondar into the nineteenth century through a period of royal weakness. By the middle of the eighteenth century, the commercial pace of life was picking up and Gondar was acquiring more economic importance (unlike Mbanza Kongo). Surviving documentation indicates the growing frequency of house sales and purchases; wealthy people began to have a stake in the permanent prosperity of Gondar independent of the fate of the enfeebled royal dynasty. This pattern of commercialisation, which began to transform older population agglomerations everywhere in nineteenth-century Africa, if they survived its convulsions, will emerge more clearly in the following chapter. Gondar and Mbanza Kongo are in some respects another type of early town compared to south-central African agro-towns and the walled remains of Zimbabwe, although their respective evolutions diverged.
© Cambridge University Press
Table of Contents
Introduction; 1. Urban life emerges in Africa; 2. African cities and the emergence of a world trading economy; 3. Colonialism and urbanisation; 4. Cities in revolt: the long term crisis of South African urbanism; 5. The post-colonial African city; 6. Globalisation and the African city: Touba, Abidjan, Durban.