free web hosting | free website | Business Web Hosting | Free Website Submission | shopping cart | php hosting

Wim van Binsbergen

Making sense of urban space in Francistown, Botswana

?/font>a study in urban symbolism


van Binsbergen, W.M.J., 1993, ‘Making sense of urban space in Francistown, Botswana? in: P.J.M. Nas, ed., Urban symbolism, Leiden: Brill, Studies in Human Societies, volume 8, pp. 184-228.

?1993-2002 Wim van Binsbergen[i]


1. Repertoires of meaning and the postmodern urban space[ii]

At least one third of the population of Africa now lives in urban environments, and social scientists are increasingly realizing that the zeal and competence with which earlier generations of their profession have mapped the ideological and symbolic universes of African villagers, now need to be partly diverted to the urban scene. African towns have been looked at as places where migrants from the rural areas settle, accommodate to urban living, to each other (along such major social-structural dimensions as ethnicity, class and religious affiliation), to formal and informal sectors of a predominantly capitalist economy, and to the state. Although it is more than half a century since Wirth (1938) launched his seminal distinction between urbanization (the process of becoming urbanite) and urbanism (the state of being urbanite), and since Hellman (1935) or Wilson (1942) first sought to explore African towns as social settings in their own right, our scholarly understanding of urban life-styles in Africa is still rather limited. In the 1950s and 1960s the study of African towns was in the forefront of Africanist anthropology, but this is no longer the case. In recent decades, the emphasis on ‘development?in the social-science approach to African towns has often taken for granted that which needs most to be explained: under what specific conditions and in what specific forms do towns mediate between rural-based historic cultures on the one hand, and on the other the post-colonial state (whose overt pattern of organization is largely along bureaucratic lines and whose ideology tends to be universalist), and a world-wide economy (characterized by bureaucratic formal organization, mass consumption, and proletarianization in the sense that urbanites have become largely dependent upon wage labour).

          ?          ? Against this background it is inspiring and liberating to look, as the editor of the present volume has invited us to do, at the symbolism of Third-World towns. This perspective prompts us to assess to what extent the contemporary urban environment has in fact managed to produce and nurture symbols which selectively refer to the state and the world economy, yet at the same time negotiate dilemmas of rural-derived identity and of urban-rural relations. It is here that one can begin to look for the stuff that African urbanism is made of. Is it true to say that these towns have engendered collective representations which are strikingly urban, and which offer partial and tentative yet creative solutions to such typically urban problems as incessant personnel flow, ethnic, class and religious heterogeneity, economic and political powerlessness, and the increasing irrelevance, in the urban situation, of historic, rural-derived forms of social organization (kinship, marriage, ‘traditional?politics and ritual)? Mitchell’s Kalela dance (1956) still offers a classic paradigm, stressing how at the city boundaries elements of rural society and culture (such as a rural-based ethnic identity, a minority language, expressive forms of music and dance, specific ways to organize production and reproduction in localized kin groups) may be selectively admitted onto the urban scene, yet undergo such a dramatic transformation of form, organization and function that their urban manifestations must be understood by reference to the urban situation alone. Or, in Gluckman’s (1960: 57) famous words, ‘the African townsman is a townsman?/span> [iii]

          ?          ? Statements of this nature have helped to free our perception of African urbanites from traditionalist and paternalistic projections; for according to the latter they continued to be viewed as temporarily displaced villagers whose true commitment and identity continued to lie with their rural societies of origin. The stress on the urban nature of African urbanites even amounted to a radical political challenge, in a time when the colonial (and South African) economy was largely based on the over-exploitation[iv] of rural communities through circulatory migration of male workers conveniently defined as bachelors while in town. We can therefore forgive these authors their one-sidedness, but there is no denying that they failed to address the fundamental problems of meaning which the construction of a town-based culture in the (by and large) new cities of Africa has always posed.

          ?          ? African historic societies in the present millenium have invariably displayed cleavages in terms of gender, age, class, and political power, while containing only partially integrated elements deriving from and still referring, beyond the local society, to other cultural complexes which were often remote in space and time. Yet they have offered to their members (and largely in order to accommodate those very contradictions) a fairly coherent universe, in which the human body-self, interpersonal relations, the landscape and the supernatural all featured in one composite, comprehensive world-view. In this context, the meaning of an aspect of the local society and culture consists in the fact that that element is perceived by the participants, explicitly or implicitly, as belonging to that general socio-cultural order, cognitively and emotively linked to many other aspects of that order ?a condition which produces a sense of proper placement, connectivity and coherence, recognition, identity as a person and as a group, aesthetics, bodily comfort and even healing.

          ?          ? Such meaning is for a considerable part spatially expressed and mediated, e.g. through the ground plans of dwellings, villages, capitals, national territories, through the sacralisation of the human space through shrines (including temples) and cemeteries, and through the extent to which non-domesticized parts of the landscape (the forest, hills, rivers) may take on diffuse sacred connotations.

          ?          ? Generations of anthropologists, linguists, historians and students of religion have monographically described and analyzed these rural-based universes of African meaning, assessing their variety and patterns of convergence, and tracing the ways in which these universes have partially persisted, have been eroded, eclipsed, or transformed as the case may be, under conditions of political, economic and ideological incorporation in the wider world. But what happens to meaning in town?

          ?          ? Fundamental as this question may be, we have to limit our scope considerably, and specify our analytical tools accordingly, if we are to address the specific theme of this symposium. We cannot concentrate on those aspects of meaning which are enshrined in the private, often invisible and ephemeral behaviour and interaction of urbanites in their homes, private conversations, private rituals. When we approach the topic of urban images from the point of view of meaning, we have to look for such meaning as is enshrined in publicly articulated collective representations, and then particularly those which, through their relative permanence, their anchorage in material forms of buildings, roads, spatial arrangements, are mediated through and at the same time constitute the townscape. We have to ask ourselves under what conditions such materially-expressed collective representations can at all be generated and perpetuated in the relatively new setting of an African town, whose fluid population (immensely heterogeneous when we look at geographical and ethnic origins, languages, creeds, life-styles and access to economic and political power) could rather be expected to have myriads of fragmented and mutually contesting parallel manifestations of meaning: the disconnected scraps of many different rural life-worlds the migrants left behind when becoming urban. It is particularly in the context of meaning that we see African towns as the arena where a migrant’s specific, disconnected and fragmented rural-based heritage is confronted with a limited number of ‘cosmopolitan?socio-cultural complexes, each generating its own discourse and claiming its own commitment from the people drawn into its orbit in exchange for partial solutions of their problems of meaning.

          ?          ? Before discussing these complexes, it is useful to realize that, as a source of meaning, the historic rural background culture of urban migrants is not necessarily as fragmented as the multiplicity of ethnic labels and linguistic practices in the town may suggest. Ethnic groups have a history,[v] and while the major ethnic identities featuring in and around Francistown (Tswana, Kalanga, Ndebele, Barotse, Shona, Sarwa ? the Botswana term for San ?etc.) can hardly be said to be recent, colonial creations, underlying their unmistakable differences there is in many cases a common substratum of regional cultural similarities and even identities: continuities such as a patrilineal kinship system, emphasis on cattle, similarities in the marital system, the cult of the land and of the ancestors, patterns of divination and of sacrifice, shared ideas about causation including witchcraft beliefs, converging ideas about conflict resolution and morality. The result is that even urban migrants with a different ethnic, linguistic and geographical background may yet find that they possess a cultural lingua franca that allows them to share such historic meanings as have not been mediated through the state and capitalism. Sometimes specific routinized modes of inter-ethnic discourse (such as joking relations) explicitly mediate this joint substratum. Traditional cults and independent Christian churches in town, which tend to be trans-ethnic, derive much of their appeal from the way in which they articulate this historic substratum and thus recapture meanings which no longer can be communicated with through migrants?direct identification with any specific historic rural culture. Moreover, partly on the basis or these rural continuities, urban migrants creatively develop a new common idiom not only for language communication, but also for the patterning of their everyday relationships, their notions of propriety and neighbourliness, the interpretation and settlement of their conflicts, and the evaluation of their statuses.

          ?          ? After this qualification, let us sum up the principal cosmopolitan complexes:

?The post-colonial state: a principal actor in the struggle for control of the urban space; a major agent of social control through its law-and-order institutions (the judiciary, police, immigration department); a major mediator of ‘cosmopolitan?meaning through the bureaucratically organized services it offers in such fields as education, cosmopolitan medicine, housing, the restructuration of kinship forms through statutory marriage etc.; a major context for the creation of new, politically instrumen­tal meaning in the process of nation-building and elite legitimation; and through its constitutional premisses the object (and often hub) of modern political organizations.

?The capitalist mode of production, largely structuring the urbanites?economic participation and hence their experience of time, space, causation, personhood and social relations; involving them in relations of dependence and exploitation whose ideological expression we have learned to interpret in terms of alienation (the destruction of historic meaning); but also, in the process, leading on to modern organizational forms (e.g. trade unions) meant to counter the powerlessness generated in that process; and finally producing both the manufactured products on which mass consumption as a world-wide economic and cultural expression ?in other words, as another, immensely potent form of ‘cosmopolitan? meaning ?depends, as well as the financial means to participate in mass consumption.

?World religions, which pursue organizational forms and ideological orientations rather reminiscent of the post-colonial state and the capitalist mode of production, yet tending to maintain, in time, space and ideological content, sufficient distance from either complex to have their own appeal on the urban population, offering formal socio-ritual contexts in which imported cosmopolitan symbols can be articulated and shared between urbanites, and in which ? more than in the former two complexes ?rural-based historic symbols can be mediated, particularly through independent churches.

          ?          ? The three cosmopolitan complexes each have their unmistakable manifestations in the townscape: in the form of specialized buildings, plots, quarters specifically set aside for state or industrial functions; and each offer both social organization and meaning. However, these three cosmopolitan repertoires of meaning differ considerably from the ideal-typical meaning enshrined in the rural historic universe. Although all three are historically related, they are present on the urban African scene as mutually competitive, fragmented, optional, and more or less anomic or even ?when viewed from a competitive angle ?absurd. Yet together, as more or less elite expressions, they constitute a realm of symbolic discourse that, however internally contradictory, assumes dominance over the rural-orientated, local and historic repertoires of meaning of African migrants and workers. The ways in which the cosmopolitan and the local idioms interact, are ill understood for several reasons. Those who, as social scientists, are supposed to study these patterns of interaction are, in their personal and professional lives, partisans of cosmopolitan repertoires and are likely to be identified as such by the other actors on the urban scene. Much of the interaction between repertoires is evasive and combines the assumption of rigid subordination with the practice of creative challenge and tacit symbolic resistance in private spheres of urban life where few representatives of the cosmopolitan repertoires have access. And whereas anthropology has developed great expertise in the handling of meaning in one spatio-temporal context (e.g. rural African societies) whose wholeness and integration it has tended to exaggerate, the development of a sensitive approach to fragmented and incoherent multiplicity of repertoires of meaning, each assaulted and rendered more or less meaningless by the presence of the other, had perhaps to wait till the advent of postmodernism as an attempt to revolutionarize, or to explode, anthropology.[vi] Our classic predecessors in African urban studies worked on the assumption that the African urban situation was very highly structured ?by what they called the ‘colonial-industrial complex?imposing rigid segregation and class interests, by voluntary associations, by networks.[vii] In the contemporary world, such structure is becoming more and more problematic, and the town, especially the African town, appears as the postmodern social space par excellence. Our greatest analytical problem here is that as a social space it lacks the coherent and unified structure which could produce a single, convincing repertoire of meaning ready for monographic processing; but this is not merely an analytical problem ?it appears to sum up the essence of what urban life in Africa today is about.[viii]

2. Francistown, Botswana: general features

These ideas will lead us in this paper’s attempt to make sense, i.e. to assess the problem of meaning, of urban space in Francistown, N.E. Botswana, with 60,000 the second largest town in the country, and since 1980 going through rapid expansion. Originating in a White gold-mining settlement almost a century ago (and therefore without any historical background in the Tswana ward system which continues to organize, to some extent, many other towns in the country), despite the decline of mining in the 1910s Francistown developed into a major railhead, distributive and industrial centre, and nodal point in the flow of migrant labour all over Southern Africa. A segregated town until Bo­tswana’s Independence (1966), its basic spatial organization has remained dominated by the state and capitalism, and their personnel requirements. Squatter areas and site-and-service areas (which have often replaced squatter areas or perpetu­ated them in an upgraded form), as well as fully-serviced housing estates, have meanwhile diversified the old layout, accommodating an unceasing influx of people from all over Botswana and surrounding countries, and have extended the range of spatial expression of the population’s economic and class differences.

          ?          ? Functional specialization, however striking and meaningful a feature of any townscape, is not in itself to be equated with urban symbolism. How can we distinguish between socio-economic association, and symbolism, in the context of the analysis of socially-articulated meaning in the urban space? For instance, in Francistown the toponym ‘Minestone?(‘Gem?, as the designation of a high-status residential area, carries overtones of prestige, conspicuous consumption, reference-group behaviour orientated towards a dominant White category of people, life-style and culture. In a way, the latter phrase summarizes the social symbolism of the name Minestone. Yet such symbolism appears to be more diffuse, less ‘properly symbolic? than e.g. tthe symbolism of the low, circular or semi-circular mud fence,[ix] which we find in many residential plots in Francistown as a reminiscence of the viable rural order and of the enclosed-yet-accessible patrilineally-recruited homestead as the epitome of that order. An intermediate case between these extremes would perhaps be that of the nicknaming of residential areas by reference to international toponyms, e.g. in everyday Francistownian conversation ‘Beyrouth?is understood to be the large, supposedly chaotic, violent and working-class squatment of Somerset[x] West, while ‘London?is the supposedly aspiring middle-class, orderly and bourgeois site-and-service scheme of Somerset East Extension. Much further reflection is needed on the theory of (urban) symbolism.[xi] When it comes to the sacralization of urban space I suppose some inspiration could be derived from theoretical work on shrines, regional cults and pilgrimage.[xii] Another interesting question would then be to define the conditions for the urban space and its symbolism, under the modern state and capitalism, to have room for secular or civil shrines ?in other words, monuments.

          ?          ? One of the most characteristic features of Francistown has been the relative absence of publicly articulated carriers of cosmopolitan meaning in the townscape: it was, and (despite the building boom since the late 1980s) largely still is, a town without street names, statues, monuments, striking public buildings, without spatially articulated public symbolism. The White- and Indian-dominated two-street business centre with its modern architecture of banks and shopping ‘malls? the new single-level civic centre, fast-food and video outlets, during the day is invaded by African people who only conditionally, awkwardly, for specific purposes of employment or consumption, venture out of their distant housing, preferably under elaborate protective clothing and make-up. The town’s symbolic barrenness, even in the central district and certainly outside it, conveys the message that, for many of its inhabitants, it is merely a temporary and scarcely convenient space where cos­mopolitan meaning, however implied to be dominant and inescapable, is seldom articulated, internalized, or shared, is always problematic, and often rejected; at the same time this state of affairs points, as we shall see, to fundamental contradictions in the relation between the state, capital and Christianity in the social formation of which Francistown forms part.

          ?          ? In order to do justice to the fragmented and evasive nature of urban symbolism in Francistown, we have to remain close to the descriptive data and traverse the town in all directions, keenly looking for clues. With some apology, I invite the reader on a guided tour, hoping that in the end it will prove rewarding. It is not my intention to write, in the following pages, a mere travel guide to Francistown; but neither is it my habit to jump to conclusions, even if in the anthropological study of symbolism such is not uncommon practice.

Francistown[xiii] can be reached by train (along the track which since the end of the nineteenth century has linked South Africa to what is today Zimbabwe, Zambia, Zaire and Angola), by airplane (already in the 1930s its airport played an important role in the transport of labour migrants to and from the Witwatersrand, as recruited by the WENELA[xiv] agency) and by road: the excellent tar road across Botswana from Tlokweng at the South African border bifurcates at Francistown, one fork leading to the Zimbabwe border at c. 80 kilometres north of the town, the other fork northwest-bound to Nata (180 km) and from there either to Kasane-Kazangula-Zambia or to Maun, capital of northwestern Botswana.

          ?          ? A number of striking features reveal themselves already on approach.

          ?          ? The first feature is that Francistown is an urban island, sharply cut out amidst the rural environment of Botswana’s North East district, and separated from the nearest (African-populated) villages by a broad (25 km) fringe of large-scale farms and mining concessions. There is hardly a peri-urban area with transitional patterns of land use and human settlement. This is due to the fact that as from the late nineteenth century, most of what is today Botswana’s North East district has been held as freehold land by the Tati Company (and its legal predecessors)[xv]. The mining concessions and White-owned commercial farms have extended right to the city boundary. Unmistakably urban squatter areas like the sprawling Somerset West have been encroachments on Tati Company land, and the town’s planned extension in southern direction (Phase IV etc.) in the 1980s was dependent upon and constrained by the extent to which Tati Company land could be acquired by the Francistown Town Council. In the Francistown case, therefore, the conceptual boundary between town and village is very much a spatial reality. Maintaining urban-rural ties in this situation always involves substantial travelling.

          ?          ? The absence of a peri-urban fringe has implications for the ritual articulation between town and country: while sacrifices reminiscent of the historic rural culture are being carried out at the town’s handful of mediumistic sangoma lodges (located at private residential plots)[xvi] and at a few Independent churches, there is no market for domestic animals in Francistown (one can scarcely buy a life chicken at the vegetable market, and then only on Saturdays), and as a result every urban sacrifice puts the patient or the latter’s sponsor in jeopardy: it is no mean achievement for a townsman without personal motor transport to extract a goat from the distant countryside. However, like many other Francistownians the leaders of sangoma lodges have their own footholds (in the way of cattle posts) in the countryside of northeastern Botswana, and it is not only for reasons of the availability of sacrificial animals, but also for symbolic reasons of avoiding pollution, alienness, sorcery and social disapproval, that certain rituals (e.g. initiation of a novice into the Shumba cult, which is greatly feared among urbanites)[xvii] can only be performed in this rural environment.

          ?          ? As a result of the absence of a peri-urban fringe, also, there is virtually no horticultural activity in Francistown, which further reinforces the conceptual and symbolic opposition between town and country. Very few people cultivate vegetables in their private plot; many people, however, have gardens in their village home or have a working relationship with close kinsmen who have. The town council operates a farming scheme for the destitute at the Tati river bank in Somerset East Extension, but its share in the total consumption of vegetables in Francistown appears to be negligible.

A second feature is Nyangabgwe Hill, which towers massively over the sprawling town and whose tall military communication pylon, lit with red signal lights at night, can be seen from all over town. For the larger part, the steep hill is uninhabited, although the southern slope has accommodated, since the mid-1960s, the PWD (Public Works Department) squatter area due to be demolished in 1989 but still there.

          ?          ? A third feature is the railway, which for almost a century now has been Francistown’s vital artery. The original railway line[xviii] dissects the urban area from south to north, and there is no place in town where one does not hear the characteristic railway sounds by day and especially by night.

          ?          ? A fourth feature is Tati River, whose wide bedding ondulates through the entire urban area, intersecting with main roads and the railway and creating a sandy, shrub-fringed boundary between residential areas. Like most rivers in Botswana, the Tati River is far from perennial and in fact contains water only a few days a year. Among its smaller tributaries in the urban area, the most important is Inchwe River. Monarch area, for decades a sprawling uncontrolled African mine-workers?settlement to be demarcated and upgraded around 1980, in terms of transport by motor roads constitutes a virtual peninsula bounded by the Tati and Inchwe rivers.

          ?          ? The fifth feature is the relative absence of vertical architectural accents. Francistown has the familiar Southern African ‘township?appearance only too well known from world-wide television broadcasting on the South African revolution: rows and rows of tiny, sink-covered houses at ground level, each standing in its own tiny plot, serviced by dusty, often unpaved roads. Towards dusk the smell of charcoal and wood stoves fills the air and the fading light is filtered through the blanket of smoke extending over the entire town. In the most recent years, the skyline of Francistown has developed considerably. Until the late 1980s the only high-rising architectural features were: the gigantic Dumela storage elevators in the Industrial Site seven kilometres north of the town, and the airport’s main military hangar, scarcely one kilometre west of the town centre and adjacent to the residential areas of Aerodrome, White City, Bluetown and Masemenyenga. Since, impressive complexes have been added: the Botswana Meat Corporation abattoir on the town’s southern entrance; the massive Nyangabgwe Hospital; and in the town’s business centre the colonial-style single-level stores and offices are now rapidly being replaced by imposing, brick-and-glass architectural wonders of two or more levels high, which also increasingly fill the many open spaces that used to exist between the centre’s grid of streets.

Francistown (cf. diagram 1) has a clearly defined spatial structure, which largely derives from the functional specialization of parts of the town in the colonial period. The Tati River constituted the main social and geographical boundary within the town: that between Whites, the dominant elite in control of commerce, industry (particularly mining) and the colonial state; and Africans, who only as salaried workers could enter the formal-economy sphere of Francistown, and for the rest depended for their upkeep in town.on ties of co-residence, kinship, sexual and marital relations, ethnic identification, and informal sector activities. For as long as Francistown was racially segregated, it was impossible for Africans to live east and north of the Tati River, with the exception of the Monarch peninsula adjacent to Monarch Mine. Moreover, few of those thousands of Africans who did live in their allotted space across the Tati River, held any formal title to their plots ?the overwhelming majority were de iure squatters. The main exception was White City, a significant name in itself, evoking the Whites as a standard reference group of the relatively well-to-do Africans living in that residential area on a limited number of plots in freehold or leasehold.

(insert diagram 1)

Diagram 1. The urban space of Francistown

          ?          ? Indian families were very limited in number before the recent economic boom of the town which started with the termination of the liberation war in Zimbabwe in 1980. People classified as coloured would have their allotted residential area at Coloured Stands (now Satellite South), more than two kilometres away from the ‘Central Area?of White business and residence.

          ?          ? East of the Tati and Inchwe rivers was the White reserve, the ‘real Francistown? articulated into a few constituent parts, which I shall describe one by one.

3. The old core area of Francistown

Government Camp. ‘Government Camp?lies in the northern part of the area enclosed by the Tati River, Inchwe River, Nyangabgwe Hill, and the railway line. Here were the professional and living quarters of the colonial civil servants, such as the magistrate, medical and veterinary officers, etc. It is here that the oldest permanent buildings of Francistown are to be found, such as the old court house, which later was incorporated in Francistown’s first hospital, the Jubilee Hospital[xix]; the latter was recently surpassed by the immensely impressive Nyangabgwe Hospital, situated at the other end of the town centre. In order to accommodate the labourers of mainly the Public Works Department, and in recognition of the scarcity of housing in the town, the PWD squatter area was allowed to develop at the edge of Government Camp since the mid-1960s [xx]

          ?          ? Adjacent to Government Camp there are the state prison and the BDF (Botswana Defence Force) barracks.[xxi] Due west of the barracks is the Francistown stadium, which is not only the lively scene of weekly football matches drawing a crowd of thousands, but also packed to the brim at national festivals (President’s Day, Independence Day), when local politicians (especially those of the Botswana Democratic Party ?BDP ? which has ruled the country since Independence), dance troupes and gymnastic teams from local schools, musical bands and even the town’s sangoma lodges with their ecstatic dances, offer a programme in which entertainment, political edification and traditional culture intermingle in a most significant manner.

          ?          ? Near the BDF barracks, Government Camp also contains an old White cemetery now gone into disuse. A number of current cemeteries are dispersed among the outlying residential areas of Francistown, but I must admit that I did not make a systematic study of them as aspects of the use of urban space, as rudimentary shrines or monuments.

Light Industrial Site. Adjacent to Government Camp there is the Light Industrial Site, where colonial-style stores, workshops and factories are increasingly replaced by impressive modern architecture. It is here that a large proportion of the employed population of Francistown earns a living. Here the Labour Office is located, where the labour disputes that occur in this town at an astronomic rate are arbitrated, and the jobless register for employment. The area is particularly the heart of the wholesale trade: Francistown services not only the North East district, but the whole of northern Botswana and in addition large sections of Zimbabwe and Zambia; and any Francistownian who has managed to lay his hands on a retail license or can mobilize alternative forms of access, does his shopping in the several enormous wholesale establishments situated in this area. Here we also find, spatially isolated from the rest of town, the extensive Catholic Mission, with a church building and a secondary school, with its latin name appropriately hinting at status mobility through education: Mater Spei, ‘Mother of Hope? Tucked away in a corner between Government Camp, the railway and the Light Industrial Site are the Railway Quarters, whose architecture is even more reminiscent of colonial conditions than that of Government Camp: the railway workers are still accommodated in tiny zinc imitations of African round huts, densely packed in straight rows and without the slightest comfort, services or privacy.

Area W. The southern part of the area demarcated above is that of Area W, where around the now defunct WENELA complex a site-and-service area of lower middle-class aspirations has developed in the last decade.

Central Business District. Immediately east of the railway line (with a slight extension to the west, into the Light Industrial Site, via a foot bridge across the railway) there is the Central Business District, situated along two parallel streets ?Haskins Street, which borders on the railway and halfway of which we find Francistown Railway Station, and Blue Jacket Street, one block to the east.

          ?          ? The railway station contains the only conspicuous monument Francistown can boast: on the platform bordering on Haskins Street, enclosed in a high wire fence, and at the dead end of a stretch of track apparently laid down for the sole purpose of creating this secular shrine, we see an old locomotive, ritually displayed as if to remind us that it is to the railway that Francistown has owed its continued life. Being the real thing rather than its image, it is scarcely convincing as a monument.

          ?          ? Also other transport functions are located in the Central Business District. The town’s main taxi rank is along Haskins Street immediately south of the railway station; from here one can board line taxis to most residential areas and to surrounding villages. The bus station is half a kilometre away, in the southwestern corner of the district.

          ?          ? Streets and side-streets in this old core part of town are very wide, according to van Waarden (1986) in order to allow a sixteen-oxen span to turn without difficulty. There is an intricate system of back alleys: sanitation lanes now gone obsolete, through which also the African workers could get to their places of employment. I would not be surprised if African customers, too, were expected to gain access to the shops through these alleys.

          ?          ? Haskins Street, named after a local trading dynasty nearly as old as Francistown itself,[xxii] used to be the principal shopping street, and it is here that we find the retail outlets of old patronized by Africans. The major ones are still owned by the Haskins company and the Tati Company, and still in the familiar colonial-style buildings which dominate the entire street. Along Haskins Street we also find the two now somewhat disreputable hotels whose various bars, along with a few other such establishments concentrated in this same district, play a major part in Francistown social life: the Tati Hotel and the Central Hotel.

          ?          ? The restructuring of the town road plan and the building boom of the 1980s have however caused Blue Jacket Street (named after a local gold mine) to surpass Haskins Street, and it is here that the powers of capitalism are more emphatically visualized: shopping arcades and malls, video outlets, fast-food outlets, supermarkets, and oversized buildings (completed since 1989) housing Barclays Bank, the Tati Company Office, Air Botswana, chique clothes?and sports shops. Here is also, in a building owned by Francistown mayor Mr. Ebrahim (a Motswana of Indian ancestry, born in Ramokgwebana, 80 km north of Francistown on the Zimbabwe border, and a member of the town’s other major trading family beside the Haskins’s), a superbly finished spacious cinema, as well as a fully-stocked electrical appliances shop. Along Blue Jacket Street we also find the pride of Francistown Town Council, the new (but single-storey) Town Council offices in red brick, whose H-shaped ground-plan encloses an ornamental garden from the Blue Jacket Street side, and a spacious car park from the back; the figure ?0? reproduced nearly man-high in the garden in angular, computer-type white letters, reminds us that this nice and modest civic centre, completed in 1986, was the fruit of twenty years of Independence. It was also the fruit of an opposition municipal regime which soon after the completion of the building was supplanted by a ruling-party one. Adjacent to the roundabout which gives access to Blue Jacket Street from the south, there is the luxurious Thapama Lodge Hotel, the obvious choice for the many civil servants and industrialists passing through booming Francistown.

          ?          ? However, the general revamping process now going on has not yet extended to some of the most significant buildings along Blue Jacket Street: the Magistrate’s Court and adjacent local government offices still look very much like they did thirty or forty years ago.

          ?          ? One block away from the civic centre is the vegetable market, a few of whose stalls are occupied by electronic repair shops and seamstresses. In the midst of the Central Business District’s proud insistence on White-dominated world-wide mass consumption and enterprise, the vegetable market under a few shady jacaranda trees is the most stereotypically ‘African?place of Francistown, where hardly any Whites come to shop and where the ‘colourful?display of vegetables, the preparation and consumption of African meals (a staple of maize porridge with meat relish) in the back of the market, and the loud music blasting away from radios and stereos already repaired, constitute an enclave of uncomplicated absence of inhibition amidst the intimidating display of White symbolic and economic dominance elsewhere in this part of town.

          ?          ? Throughout the Central Business District, moreover, the streets are lined with peddler’s businesses. Many vendors operate from stalls whose framework remains overnight; others squat on the pavement and display their goods on spread-out pieces of cloth. Items for sale vary from food (fat cakes and cooked mealies, particularly), via cloth and clothing, to cheap watches and jewelry, cosmetics, ladies?bags, wallets and other indispensable items of accomplished urban life. With the exception of prepared food[xxiii] all items for sale in the peddlers?circuit have been manufactured, and imported from either South Africa or the Far East. Articles of local manufacture, e.g. household implements including baskets, are rare in Francistown; if available, they are distributed not so much through peddling but by such large-scale manufacturers and sellers of curios as BGI and Bushman Products in the Light Industrial Site, and then the more likely buyers are not Africans but Whites. Sometimes these items are peddled at the vegetable market. The main other place where African artefacts are being sold in the streets is in the car park in front of Francistown’s Spar supermarket ?the sellers are Zimbabwean Africans, the buyers Whites.

          ?          ? In addition to the peddlers, there are a few cobblers and seamstresses working on the pavement, for their own account and independently from the commercial firms in front of whose shops they have found hospitality. Their customers are invariably African. Most people in the streets and in the shops are African, anyway. If Whites (and African elites) are less conspicuous on the sidewalks, it is because these are the people with motor cars, and although the town centre is small enough to park one’s car somewhere and do one’s shopping on foot, this is not what one prefers, for combined reasons of social status, the risk of theft from cars, and the insistence on social distance across class and racial lines.

          ?          ? Even in the large supermarkets whose assortment (imported lock, stock and barrel from South Africa with which Botswana entertains a customs union) specifically caters for White/elite tastes and budgets. Yet, although their shopping baskets often contain only one or a few items, more and more African customers now find their way from the ‘traditional?retail outlets along Haskins Street (with their austere shop interiors and their crude security regulations against African shop-lifting, but also with their assortment carefully attuned to African taste and experience, and by and large slightly lower prices) to the fancy, less strictly policed and occasionally cheaper supermarkets like OK, Spar and Fairways. In fact, many small peddlers of food and groceries in the outlying residential areas do their ‘wholesale?purchases at these supermarkets rather than at the wholesalers in the Light Industrial Site ?for the latter they tend to lack the time, transport facilities, license, and (since real wholesale purchases have to be in bulk) the money.

          ?          ? The African non-elite presence in the old core of Francistown and particularly the Central Business District touches on fundamental issues in the use and signification of space in this town. We shall come back to this important point after completing our tour of the town centre.

Central area. Immediately to the northeast of the Central Business District there is the old White residential area (‘Central Area?, with spacious plots containing houses ranging from decent to luxurious. With the Central Business District, this is the only part of town where streets have formal names displayed on street signs. Most of these names derive from obscure White characters in the early decades of the town, or from early Bechuanaland history in general: Guy Street, Feitelberg Street, etc. The important thing about these street names is that they are very few, and that no African Francistownian would ever use them nor know to which street or which historical personage they refer. This aspect of the urban symbolism of Francistown is only meaningful to the local Whites.

          ?          ? Interesting is the location of the University of Botswana’s local institute for adult education on the very corner of Lobengula Avenue and Khama III Avenue ?of course, a century ago Tati district itself could become a mineral concession, a focal point of peripheral-capitalist expansion in mining and agriculture, and hence the cradle of Francistown, because it constituted more or less a no-man’s land between the Ndebele king Lobengula’s and the Ngwato king Khama III’s territories.

          ?          ? In this area we also find Francistown’s John Mackenzie and Clifton[xxiv] primary schools catering for the (predominantly White) local elite.

          ?          ? As an expression of White concepts of urban space, the Central Area is also the only part of town where two small parks can be found, as well as a public swimming pool. However, their constituting formal public spaces in an elitist though post-segregation area makes these amenities pose a dilemma typical of Francistown today: since these places are open to working-class Africans they are no longer visited by Whites/elites (who have access to secluded alternatives at the Thapama and Marang hotels, the Francistown club and private houses), and while African youths massively patronize the municipal swimming pool, the parks tend to be empty as a sign of latent contestation.

Area M, Area I, Area A. South of the ‘Central Area?more recent residential areas have developed, which in terms of class association are however rather continuous with the White and elitist connotations of the Central Area. Minestone is the most prestigious residential area in Francistown, apart from the larger and older (and almost exclusively White-occupied) Tati Riverside plots on both sides of the Tati river.

4. African working-class and White-cum-African elite: Invisible boundary and the contestation of urban space

We have noted the emphatic presence of African working-class people in Francistown’s Central Business District.

          ?          ? To irreverently paraphrase Gluckman, African customers are customers, and local commerce has understood that it does pay to keep social and security thresholds sufficiently low for potential patrons from the African working class to cross them. In this respect Francistown might give one the illusion of an effectively desegregated society. All facilities in the Central Business District are of course in principle open to anyone regardless of race or social class.

          ?          ? In fact however, the standard image of Furnivall’s (1944) plural society (however discredited the concept is from a theoretical perspective, and however obsolete in a well-functioning post-colonial society like Botswana’s) would meet the social reality of Francistown better: the presence of well-defined social classes, with large differences in political and economic power, whose membership badges tend to feature somatic characteristics such as skin colour along with such acquired status symbols as style of dress, mode of transportation and language use, and who meet (and rather uneasily at that) in the reserved social space of the marketplace only. ‘Marketplace?should then not even be taken literally (Whites in Francistown shun the vegetable market, the peddling circuit and such market-like retail outlets as are found at the bus station, Jubilee hospital, and in the outlying residential areas) but be understood as ‘formal outlets of commodity distribution? And just as non-elite Africans risk to be relegated to the status of trespassers and potential burglars whenever they seek to enter the elitist residential plots of Minestone and along the Tati River, so there are still many situations in the Central Business District where the White uniform of colour, and/or the elitist uniform of smart dress, are a requirement for entry or at least for decent service; Thapama Lodge is a case in point, and so are most company offices and, somewhat further away from the town centre, the premisses of the Francistown Club;[xxv] Clifton School; or the Marang Hotel which caters almost exclusively for a White clientele directly from, or at least orientated towards, South Africa.

          ?          ? Alternatively, there is a marked sense of uneasiness (generally covered under excessive politeness, which occasionally may give way to downright hostility) when Whites try to break through the invisible boundary which lies over even the Central Business District of Francistown (and anywhere else in that town), and when they venture afoot in the streets, the bus station, the vegetable market, the sanitation alleys, the peddlers?circuit, the non-elitist drinking places of the town centre.

          ?          ? Working-class Africans on the one hand, and Whites/African elites on the other, make a fundamentally different use of Francistown’s central space and attach a fundamentally different meaning to it.

          ?          ? The Whites/African elites own it, are competent in the complex socio-economic procedures, varieties and choices of commodities offered there, and for them the architectural and planological designs, the rare street names, the manner of display and the assortment in the shops, patterns of commercial etiquette, reinforce their sense of identity and security as a dominant class; and by their means of transportation and their choice of venue they limit to a minimum the extent to which they should be required to rub shoulders with working-class Africans.

          ?          ? The subjective townscape of the White Francistownians still does not greatly differ from that of the many South African tourists passing through on their way to the distant game reserves of the Chobe and the Okavango Rivers: ‘a few decent shops (their own familiar OK, Spar and Fairways!) and filling stations along a major through-road (Blue Jacket Street), a few passable garages on the other side of the railway in the Light Industrial Site, an excellent camping and curio shop in between (BGI), a nice, White-frequented hotel and camping site at adequate distance from the shops (Marang Hotel), and as a backdrop to these essentials, barely registering onto consciousness, ‘‘the African location’’ ?(more than 50,000 people!).

          ?          ? Alternatively, African working-class people, although far more conspicuous than Whites and African elites in the streets of the Francistown town centre, know only too well that they do not own the place. They are aware of invisible thresholds of incompetence, embarrassment, suspicion and lack of spending power when they venture from the streets into the shops except for those few shops known to cater explicitly (albeit in a humiliating way) for the African clientele. When they enter the buildings of the Central Business District (and the same would apply for the Light Industrial Site and the elitist residential plots) it is far more likely that they do so as wage labourers (or as would-be wage labourers asking for a job) than as respected clients. Their apparent appropriation of the public space of the streets, as pedestrians, as eaters of fast food, as peddlers and their clients, as people on their way to and from work, school, hospital, church, shops and government offices, may subconsciously seek to express an aspiration of ownership as a reflection of political ideals in an independent and economically thriving Botswana, but in reality is more a form of ‘being cast out into the streets??the streets being the one public space which they know cannot be denied to them, not even in the centre of Francistown which for hundreds of kilometres around is the focal point of White and capitalist domination of African life.

          ?          ? The contestation of public space along lines of class and colour, and as a result the emergence of parallel, scarcely intersecting patterns of use of the urban space, is not limited to Francistown’s Central Business District. If for a moment we look at the town not so much as a residential space but as a network of formal road communications within and across material boundaries such as the railway line, the Tati River and the Inchwe river, and fenced plots, it is very revealing that Africans (who, as we have seen, are the town’s typical pedestrians) use the formal road system only selectively and negotiate the main boundaries in a manner very different from White/elite motorists. The African pedestrians insist on shortcuts across the formal road system, use the railway track and the sandy river beds[xxvi] as just another obvious passage which can be utilized at whatever convenient point rather than as a boundary which can only be negotiated at formal bridges and crossings. In fact the spontaneous footpaths quite often lead through plot fences, whose wiring is casually lifted in the process, or trampled underfoot. A case in point is the Francistown Club golf course, which is used, simultaneously and virtually without mutual recognition of each other’s presence, by two different sets of people: paid-up members using the terrain for its intended purpose, and numerous working-class pedestrians taking short-cuts across the terrain.

          ?          ? In other words, the map of urban transportation of African pedestrians is not the same as that of White/elite motorists ? it refers to a different town altogether.

          ?          ? The division of the urban space in terms of class and colour is merely one aspect of the major cleavage dividing Francistown society. Its effects are far from limited to the use and conceptualization of space, but amount to a compartmentalization of life worlds and symbolic universes. In addition to the private domain which each African family, or each White and Indian family for that matter, in Francistown negotiates in the face of other such families within their own category, there is a kind of ‘collective private?domain on either side of the class boundary, where the common knowledge on one side of the boundary becomes a carefully guarded secret on the other side. The Mwali cult,[xxvii] sangoma-hood, the significance of rivers and hills, therapeutic par­tici­pation in Independent churches, are considered private if collective African matters which are virtually impossible to articulate in conversations and interactions across the boundary. As a White, operating from a house I had rented for my family in the non-elite site-and-service area of Somerset East Extension, becoming aware of that boundary and desperately looking for means to cross it dominated my participatory research for the better part of a year (van Binsbergen 1991b).

          ?          ? Although this imperceptible boundary, as we have seen, is an interactional and a conceptual one much more than a geographical one, Francistown’s old town centre is very much perceived, by the African population, as outside their own collective private world, as inimical and alien. ‘Going to town? therefore, in other words leaving the residential space of the outlying townships and entering the Central Business District, the Light Industrial Site, or Government Camp, is strongly felt to be the crossing of a boundary, which requires significant and costly rites of passage, whose main purpose is self-protective. No matter how informal one’s attire when around the house (where many women go barefoot, with a wrapper as their principal garment, and occasionally exposing the breasts; men have matching forms of informal residential dress, featuring e.g. sleeveless undervests, patched-up trousers and shoes with holes) it is virtually inconceivable to enter ‘town?without having washed one’s entire body, having put on freshly washed and ironed clothing, preferably of a kind local Whites would consider to be in the range of ‘over-dressing? or without decent shoes, or on a bicycle instead of in a taxi or (if at all unavoidable) on foot. Perfumes, other cosmetics, and protective African medicine help to ease one across the apparently massive and intimidating boundary. It is a continuous source of amazement to see perfectly spic-and-span townsmen emerge, as a matter of course, from the most humble and dilapidated dwelling houses, and one can only admire their skill in keeping up these standards even when their journey leads across several kilometres of muddy road on a rainy day. It is possible to explain away some of these precautions as strategic in the lively local market of occupational and amorous success, and in the struggle for peer group esteem; but we are dealing with a strictly observed institution here, not with mere individual strategies. On the one hand this over-preparation is defensive reference-group behaviour, an expression of insecurity lest one be found out to be a mere working-class or peasant African in a White- and elite-dominated town; on the other hand there is, after all these preparations, a sense of self-assertion, of exorcising such humiliation as one has internalized, and of claiming a right to the central urban space by adopting its appropriate uniform ?after the rite of passage one hopes to become a legitimate if only temporary member. The striking point remains that these painstaking preparations are made not just for a full day’s work at the office or factory or for major shopping; I have known people to go through the entire routine just for the purpose of buying a few litres of paraffin at the Haskins filling station on Blue Jacket Street.

5. The outlying residential areas: Strategies and commitments

So far some basic characteristics of Francistown’s core area, dominated by the state, capitalism and White culture ?the latter particularly in the form of formal bureaucratic organization and mass consumption, and characterized by a peculiar, invisible boundary of class and (to a considerable extent still) race. Let us now turn to the outlying areas of the town.

          ?          ? We have seen how in the era of segregation the residential space reserved for the African population was located west of the Tati and Inchwe rivers, where moreover they rarely held title to the land they occupied but were squatters. In other words, prior to Independence the African population of Francistown was largely a population of squatters. Squatments eventually even developed east of the rivers: PWD, Satellite and particularly Somerset East, all of which seem to pre-date Independence by at least a few years.

          ?          ? Various strategies were adopted to improve the housing situation, reduce the insecurity and crowding in the existing African residential space, and accommodate the massive influx of urban migrants from all over the country, both as a result of commercial, industrial and governmental expansion, and as a result of the declining rural economy under conditions of drought, rural class formation and over-grazing. The first official attempts to formalize squatters?occupancy status in Francistown go back to the late 1960s.[xxviii] Subsequently, existing squatter areas (such as Monarch, Bluetown, Riverside North, Tatitown, Tati West, Riverside South, a section of Somerset West, and Block 2) were upgraded: plots were demarcated (typically at considerably larger size ?ideally 400 m2 ?than the units which squatters had appropriated spontaneously or had managed to maintain in the face of continued influx of migrants), and those persons and household units considered to be too many were given priority as candidate occupants of newly designed site-and-service areas which came to form a wide circle around Francistown’s older core: the residential areas of Area S, Aerodrome, Area W, Somerset East Extension, Donga, Block 7, and Area L, among others. With the exception of these ‘victims of thinning?[xxix] the site-and-service plots were reserved for Francistownians with a regular monetary income not exceeding P3,000 per annum.[xxx] Their exclusive use rights to the plot were defined by a formal ‘Certificate of Rights?(framed, this document constitutes a standard ornament on the walls of Francistown houses in site-and-service areas), they qualified for a low-interest building loan, and were required to pay a service levy of P8.50 per month. Despite inevitable misuse it can be said (van Binsbergen 1989a) that it was by and large the target population which benefited by the SHHA scheme, and thousands of plots have been distributed on its basis.[xxxi]

          ?          ? All demarcated plots in fully-serviced, site-and-service and upgraded residential areas in Francistown lie immediately on a formal road which is supposed to be regularly maintained by the Town Council, and which features on the maps of the department for Town and Regional Planning. Yet none of these many streets has a name (a state of affairs sometimes deplored in the local freely-distributed advertisement paper, the Northern Advertiser), and people can only describe their own address in these areas by reference to a four- or five-digit plot number (which is seldom displayed anyway) or to far from unique local landmarks, such as the area’s vending stalls, a bend in the road, etc. In this use of plot numbers and the absence of any higher-level shared designation at the street or neighbourhood level, we may see an almost caricaturish expression of the ‘atomisation?of urban life under conditions of capitalism and the modern state. At the same time it points to the absence of a shared discourse of reference and meaning between planners and their clients, the town dwellers.[xxxii]

          ?          ? SHHA houses have to be built against strict specifications (whose observance is frequently inspected and constitutes a condition for the building loan to be made available), but the occupant has a wide choice of alternatives, from the simplest two-room house to accomplished two-bathroom villas. Here the only uniformity is a demarcated plot of 400 m2, with metal markers on the corners, and a dug latrine pit covered by a slab ?on which the occupant has to erect a standard-type toilet building. The Botswana Housing Corporation, alternatively, has erected hundreds of serviced houses in each of the three categories of low, medium and high cost. Their architecture is standardized and cannot be influenced by the occupant.

          ?          ? These three planning strategies of upgrading, site-and-service schemes and fully-serviced housing, in the face of increased influx proved insufficient to prevent squatting, as an unplanned, people’s fourth strategy. The town’s residential areas with their four strategies of structuring and occupying the urban space constitute a laboratory where the central theme of this paper and of our symposium is tested out in various experiments. The results of the experiments in the four situations, however, are difficult to compare with each other since in each situation the occupants?freedom to structure the private urban space is substantially different.

          ?          ? By definition, this freedom is maximum in the case of the squatments. It would be a fallacy to expect urbanites to use this freedom exclusively to implement rural, ‘traditional? patterns in their use of the urban space. The squatter’s appropriation of land in the urban fringe does not particularly emulate a perennial rural pattern any more. State control (through the Land Boards) over rural land in Botswana has greatly altered patterns of rural occupancy in the course of the twentieth century (cf. Werbner 1970, 1982). One should not entertain too romantic an idea about the structuring of space in contemporary Botswana villages: in their outlying parts considerable room and a degree of unboundedness may still prevail, but in village centres residential plots are often directly adjacent to each other; with their standard size of 1,600 m2 (40x40 m) they are only four times larger than urban SHHA plots; they tend to have wire fences; neighbours are often non-kin; and problems of privacy, annoyance, theft, moral indignation, religious rejection, absence of effective social control and witchcraft suspicion occur between rural neighbours much in the same way now as they do in town. In one respect squatters are better off than villagers: there is no obvious, rural, ‘traditional?equivalent for the possibility of converting squatting into formal occupancy rights, such as occurs in Francistown (and other Botswana towns) whenever squatments are upgraded or removed.[xxxiii]

          ?          ? And again, when similarities emerge in people’s spatial responses in these four situations, this may be not by their own choice but as an effect of the converging planological measures and considerations of civil servants. Thus, all plots in fully-serviced, site-and-service, and upgraded residential areas are to be situated on a public access road however humble; this does not particularly reflect people’s reluctance to share their private space with their neighbours in case the latter have no direct access to the road, but mainly represents a bureaucratic conception of the ideal urban space.

          ?          ? Squatments, outside official control, provide a testing ground for people’s minimal conceptions of urban privacy, and here there is considerable variation: from the relatively spacious plots, demarcated by a low row of stones or a single-wire fence, in many squatments, to ?under conditions of extreme overcrowding, in PWD ?dwellings which (although independently occupied by occupants who are not mutually related) are back-to-back and sometimes even without proper access to the foot paths. These developments are adaptations to urban conditions. With a similar degree of optionality, however, rural continuities occur at the same time, for instance in the relative lack of insistence on specific toilet facilities; in the adornment of house walls with pictorial patterns in mud of a contrasting colour; the construction of a low mud fence within which cooking and social conversation may take place; or the attempt to surround oneself with kinsmen and more in general with people from one’s home village as urban neighbours (cf. diagram 2). The latter strategy, so marked in PWD, is only feasible in squatments (with possible traces in upgraded squatments) since individualized bureaucratic procedures determine plot allocation in site-and-service and fully-serviced areas.

(insert diagram 2)

diagram 2. Clusters of fellow-villagers in the PWD squatter area, Francistown, April?989.[xxxiv]

          ?          ? Such continuities are more than the mere nostalgic and pathetic return to familiar, rural patterns: they are an attempt to render the urban residential space meaningful by structuring it in terms of a shared, ancestral world-view. Such attempts will have to remain selective: actors are only partially, through socialization, language use and continued urban-rural ties, participating in their rural background, and ?largely surrounded by non-kin strangers anyway ?for the rest have to survive, at the same time, in an urban life-world dominated by White culture, the state and capitalism. Among African Francistownians today, there is considerable variation in rural-urban ties and continued rural investment. Francistown, in other words, does not always play the same role in the process of social reproduction of its inhabitants. A useful idea of modes-of-production analysis[xxxv] is that, to the extent to which people remain dependent upon non-capitalist, rural-based modes of production, and hence remain involved in urban-rural ties if they are townsmen, to that extent (and in those situations involving such ties) will they continue to subscribe to the ideological components of these non-capitalist modes of production. What is the range of variation of such involvement?

          ?          ? There is certainly the characteristic Botswana pattern[xxxvi] of people who, as individual members of extended families involved in long-term economic and social strategies, straddle the urban-rural divide, invest on the one hand in cattle and a rural base, ?in education, a formal-sector career, and an urban base, on the other. I am still analyzing my quantitative data on Francistown, but I estimate that this group would account for nearly 50% of adults in Francistown, both men and women. In these cases people’s dealing with the urban space could be characterized as ‘reticent use?rather than as ‘eager appropriation? In their social reproduction, the town’s role is more or less balanced against that of the village and the cattle post. Their urban house is looked upon as mainly a sensible investment and is readily let to tenants even if it offers levels of comfort and status appeal exceeding such alternative accommodation as the landlord will have to use when in town. When people in this category use the town house themselves (it is often the scene of an endless procession of rural kin and tenants, each staying only for a relatively short period), it is largely left unadorned. No time or money is wasted on an ornamental garden, and the interior contains the bare minimum and (as will be appreciated by anyone who has seen women try and prepare food in the rain and wind in a kitchen-less urban plot) often offers less comfort than a village house would. Under such conditions, the town house is appreciated for its utilitarian rather than its symbolic or status value. It is in this context that we must view squatters?stoic acceptance ?which has little to do with poverty, since others in the same income bracket do opt for modest modcomfs?of low levels of comfort (absence of a toilet, bathroom and kitchen; small rooms; absence of outdoor space). From this point of view it can also be understood why many non-squatters lack enthusiasm for a nice yard embellished with flowers, rendered more private by wire fencing, shrubs or a tall brick fence, and more impressive by fancy brickwork and a hand-carved front-door. For the people in this category, the urban space is experienced primarily in pragmatic terms: the town is an environment where one can make money, go to hospital, send one’s children to a good school ?identity, and the valourisation of the living space by historic symbols imbued with profound meaning, are concerns reserved for the village. Their town house does not look like a home, and in fact home is emphatically elsewhere.

          ?          ? In this category of town-dwellers balancing their urban and rural commitment, women, meanwhile, are in a slightly different position from men, since the burden of reproduction (childbirth and particularly bringing up the children) is more or less exclusively falling on them as female head of households and unmarried mothers;[xxxvii] and while such patterning of their reproductive roles would not in the least stigmatize them nor jeopardize their taking part in long-term extended-family strategies, they will often find that they have to fence for themselves and their children. Women have a fair (probably even advantaged) access to the Francistown job market, but mainly in the lowest paid jobs. Women also get a fair deal from the housing department, as statistical analysis has shown (van Binsbergen 1989a). Considering a woman’s difficulty of looking after her rural economic interests single-handed, and taking into account that for women (in view of their lion’s share of reproductive and productive labour under austere rural conditions) urban living is even more attractive than it is for men, we see adult women (i.e. mothers) in Francistown insist somewhat more on the acquisition and development of urban housing than men of the same age and background. But in the case of women, too, the town house means an asset in a long-term survival strategy, much more than a requisite for modern status achievement.

          ?          ? Around this prevailing balanced position minority responses exist on both sides:

?On the one hand the unsuccessful urban migrant who does not manage to adopt a viable utilitarian attitude towards the town, and spends a period in town either as a co-resident kinsman with a more committed urbanite; as an independent squatter; or as a passing tenant in a squatter area. These characteristics would particularly apply to the Sarwa squatters among the population of the most devastated and miserable squatment of Francistown, Masemenyenga. For people in this category, the town does not yet play a major role in their social reproduction, even if the viability of their rural alternatives has greatly declined.

?On the other hand those whose rural ties are rather weaker than average (because they are second-generation townsmen, or in the process of upward social mobility with a secure source of income from capitalism or the state, because of family conflicts, of the declining viability of the Botswana rural economy, etc.), for whom the town is the main locus of social reproduction, and who have developed a sense of urban identity,. The latter is usually combined with a desire to acquire the status symbols of modern urban living: wardrobe, dresser, display board, lounge suite, television set, other consumer electronics, gas stove etc. Their house does look like a modern home, and advertisements, furnishers?and mail-order catalogues are spelled to keep informed of fashionable trends and affordable offers. For people in this category the house is not just a utilitarian asset, but a central means of self-expression ?a material structure, still, but one charged with meanings derived from the ‘modern?i.e. cosmopolitan, domains of life. There is of course a minimum financial requirement here: even with widely available (but intimidating and humiliating) hire-purchase facilities such as are offered by the large furniture shops in Francistown’s Central Business District, one needs to have a regular if moderate income to qualify. Yet it is amazing to see how people manage, over the years, to accumulate consumer goods with the kind of minimum formal-sector income that still falls within the SHHA limits. For the people in this category, building itself (with all the implications of artisan’s skills, knowledge of building materials and their prices, the necessity of gaining access to transport facilities and of organizing labour) becomes a self-expression and a passion, by which one asserts one’s aspirations and one’s determination to join the urban rat race. In this category one really deals with the urban space with a sense of appropriation, and by doing so adopts and internalizes such meanings as the state (the main distributor of building plots) and capitalism have to divulge. Such investment presupposes security of occupancy, and people in this category would rarely direct their building and decorating energies at a squatter plot ?if starting out from such a plot because of sheer housing shortage, they would seize every opportunity to more for a more secure plot and reserve their investments for the latter.

With regard to the bureaucratic, planological structuring of the outlying residential areas (with the exception of un-upgraded squatments, of course) one finds a number of recurrent patterns. All residential areas have their area of vending stalls, more or less solid structures of corrugated iron where groceries and vegetables are offered for sale at slightly higher prices than in the town centre but with attractive credit facilities; access to one of these stall depends on possession of a vending license, and is thus largely controlled by the municipal authorities. Similarly, all residential areas have what is called a ‘Freedom Square?? in most new or upgraded residential areas this is in fact the only street-name people use and recognize for a particular part of their area. It is here that the councillor (member of the city council) responsible for the ward in which the residential area finds itself[xxxviii] addresses his constituency, and where the several political parties on the Botswana scene take turns to canvas their prospective voters. Thus the residential area is rudimentarily articulated by its economic and political central places, and there conveys a (limited) degree of meaning within the context of the post-colonial state and capitalism, highlighting differences in political and economic power between inhabitants of the residential area, and suggesting possible reference groups and strategies for status advancement.

          ?          ? Central places at the level of the residential area or neighbourhood may, moreover, include a neighbourhood school, the local SHHA office, an urban clinic, a bottle store around which customers may gather and sip their canned beer, a beer garden retailing Chibuku (a manufactured imitation of African home-brew) and any number of private ?but perfectly legal ?shebeens where a variety of drinks (including home brews) is offered for sale. The condensation of social interaction occurring around these central places in general fails to generate a sense of community at the residential-area level. In the absence of community centres the residential area’s several churches would have a role to play here, but in the denominationally and therapeutically fragmented society of Francistown (most independent churches offer healing as well as salvation) they do so mainly for their own members and sympathisers. The minimum neighbourly participation in life crises such as marriage and death; the low level of neighbourly social control and hence frequent recourse to formal, geographically and socially distant law-and-order agencies such as the police and the urban customary court for conflict regulation; the ineffectiveness of the ward councillor to mediate in major conflicts; the invisibility and often non-existence of the ward development committee and other politically-affiliated bodies at the residential-area level ? all this suggests (cf. van Binsbergen 1991a) social life in Francistown’s residential areas to be dispersed and individualized. It is meaningless from the perspective of a historic, rural cosmology, and at the same time derives but a shade of meaning from the three cosmopolitan repertoires of meaning and organiza­tion.[xxxix]

6. Creating meaning as a specialism: Churches, sangoma lodges and the sacralization of urban space

Architectural structures which dominate the townscape are those associated with the state and capitalism. This means that in Francistown only an inconspicuous role is reserved for one category of structures which in European and American towns is so visually dominant: churches. Imposing church halls and spires which tower over a town’s main throughways and squares ?this sort of thing is entirely absent in Francistown. The only church to be found in the old town centre, at the boundary between the Central Business District and the Central Area, is the Anglican St Patrick’s Church, a brick structure whose modest dimensions and miniature spire reminds one of a village chapel in England; the back street on which it is situated, behind the civic centre, is named ‘St Patrick Street? and this is virtually the only case of a Francistown street-name having religious (and thus universalist) connotations.

          ?          ? The only other church building which stands out near the town centre is the large, Zimbabwean-orientated Apostolic (Vapostori) Church on the main road passing Area W going to the Tati River. Several major cosmopolitan church missions, such as the Catholic Mission and the Seventh Day Adventist Mission, are located outside the African residential space and form little enclaves on their own. The outlying residential areas contain scores of church buildings. Some of them are of rather elaborate modern brick architecture,[xl] but a far greater number is built on a self-help basis, in an architecture scarcely standing out among the surrounding houses, and displaying considerable reticence in self-advertisement: no sign, or a small hand-painted one.[xli] Churches may attain a certain conspicuity at the residential-area level, but they are far from a marked presence in the townscape as a whole.

          ?          ? Basangoma (spirit mediums) and baprofiti (‘prophets? i.e. leaders of African Independent churches) are the African Francistownian community’s principal specialists in the articulation and manipulation of meaning. They are the only ones to actually sacralize the urban space in its own right through the creation of shrines and the staging of ritual and sacrifice in the urban context. It is true that the sangoma shrines to a considerable extent evoke a viable rural social and cosmological order revolving on ancestors;[xlii] but at the same time items charged with cosmopolitan meaning (the lodge leader’s relatively luxurious town house, modern furniture, emphasis on cash and on cash-bought paraphernalia and sacrifices, reliance on manufactured food and drink, even the ubiquitous plastic shopping bags) are far from shunned, and they are sacralized in the ritual process continuously going on at the sangoma lodges. The baprofiti’s position is related but somewhat different: their reference is to the cosmopolitan repertoires of meaning much more than to the historic rural repertoire, and they impose severe limitations upon the selection from the modern society that their adherents are allowed to indulge in, yet they too offer ritual and symbolic ways in which the suffering and temptation engendered in that modern world can be alleviated and a person can return to it without being overwhelmed by it ?ways which make that world once again an inhabitable place. In this way the baprofiti, too, sacralize and to some extent rehabilitate the urban space itself. Both types of ritual specialists offer a way out from the alienation which for most other Africans in Francistown is both an accepted fact and a major factor in their strictly utilitarian approach to the town as intrinsically devoid of (historic, rural) meaning, as anything but home.

          ?          ? At this juncture, we should mention Richard Werbner’s (1985, 1989) perceptive attempt to define aspects of the symbolism of African Independent churches in Zimbabwe and North East Botswana by reference to two axes: personhood, which he argues can be either framed or unframed, and space, which can be either bounded or unbounded. In this way he is able to pin-point specific differences in church idiom, and also to construe these differences as elements in an ‘argument of images?which essentially addresses the dilemmas of displacement, movement and alienation in the context of migrancy, urbanization and an eroded rural cosmology and economy. Werbner’s approach certainly illuminates the specific form certain Independent church buildings take ?as well as the absence of such buildings in other cases ?but his post-structuralist abstraction from concrete social forms and situations renders it less applicable in the present context.[xliii] Nor does Werbner’s interpretation exhaust the spatial symbolism at hand in the African Independent churches. His static insistence on doctrine and architecture fails to capture the spatial and bodily dynamics of group interaction. In many Independent churches this takes the form of a dancing chorus, a circular dancing movement, or even a ‘planetary?movement with the dancers (as detached, impersonal ‘atoms? once again?) turning both around their own axis and, jointly, around a common centre, where often the congregation’s newborn children, novices, baptismal candidates, sufferers or sick are placed as if to have maximum benefit from the energy unleashed by the frantic yet carefully orchestrated movement of the congregation. Here, and in the not unrelated sangoma dancing ritual, we can see the (attempt of a) group-wise appropriation and hence transformation of a small ritual space inside town, as an active way of confronting and exorcising the alienation which is paramount in the everyday living experience of the African workers in the urban space outside the ritual situation.[xliv]

          ?          ? Sometimes this ritual appropriation is of an amazing directness. For instance, in the Francistown Morningstar church which I studied in detail, the nocturnal dancing worship would take place in a barely roofed shelter of corrugated iron, in Somerset West, immediately adjacent to the railway line. The congregation’s planetary dance would be accompanied (as is often the case in ecstatic ritual, wherever in the world) by respiratory exercises, here meant to release the Holy Spirit. It took the form of a concerted loud hissing which, combined with the stamping of feet and the rustling of ceremonial robes, was strikingly reminiscent of a stream locomotive. Towards midnight, the increasing outside noise and the beam of a headlight penetrating into the dilapidated church building would announce an approaching train, the trembling of the floor under the impact of the dancers would increase by the vibrations of the train passing, and the two sounds, so strangely alike, would merge into a cacophonic yet eloquent statement about images of space, movement and meaning in a Southern African railway town. Yet on the conscious level I seem to have been the only one to notice, among those present ? even though forms of spirit possession imitating locomotives have been endemic in South Central Africa throughout the twentieth century.

          ?          ? The basangoma and baprofiti also specifically mediate between rural spatial symbolism and that in town. I have already explained how the town’s river beds, in the African pedestrians?perceptions, are convenient passage-ways rather than boundaries. This is the place to point out that in the symbolism of the urban landscape they feature also in other capacities. The rivers (whose connotations of liminality may be obvious) have retained their historic rural symbolism as the abode of the ancestors, of territorial spirits and of the Great Water Serpent ?even if they are dry most of the time. The urban rivers play an important part in the ritual of the town’s sangoma lodges in that every novice has to be chased across one of them, dropping sacrificial coins and being beaten by the senior lodge members. Lodge members ritually wash their bodies outdoors in the thicket adjacent to the stream, on the occasion of initiation and bereavement. The rivers also play a role in the baptismal rites of the Independent churches, whose symbolism is historically African at least as much as it is biblical.

          ?          ? Hills are in a category akin to rivers. Nyangabgwe Hill does not only visually dominate the town. The etymology of this place-name contains virtually the only bit of shared historic collective consciousness among the local population: many Francistownians can tell you that the name derives from the Kalanga words for ‘rock?and ‘to stalk? and offer the nutshell myth of a hunter mistaking a rock for a prey he thought to be stalking. This hill is only the tallest of a system of about ten hills around the confluence of the Tati and Inchwe Rivers, and on the tops of several of these hills there are archaeological sites, with zimbabwe-type fishbone-pattern brickwork revealing these places to have been residences of regional minor rulers incorporated in a powerful state (closely associated with the Mwali cult and the Kalanga language) encompassing much of northwestern Botswana and Zimbabwe until only a few centuries ago.[xlv] The contemporary ethnic consciousness of the Kalanga in and around Francistown lacks awareness of this glorious historical past and concentrates on their humiliation at the hands of the Ngwato (a Tswana sub-group) mainly in the colonial period. I suppose that in the technical legal sense these archaeological sites are national monuments, but in the sociological sense they are certainly not, since very few Francistownians are aware of their presence and significance. The hills do however feature in the ritual of the (mutually closely related) Mwali and sangoma cults and in that of the, somewhat more distantly related, African Independent churches, as places of theophany comparable with the rivers.

7. Discussion: Competing repertoires of meaning in the structuring of urban space

Exhausted by our guided tour of Francistown, let us withdraw from descriptive detail and seek to complete, on a more abstract note, our task of making sense of urban space in Francistown.

          ?          ? At first, when thinking about my contribution to the present symposium, my leading idea was: ‘how is it possible for an urban migrant to live in a place which has no meaning? Gradually, however, I came to appreciate that capitalism, the state, and Christianity also are major complexes creating, organizing and upholding meaning, and that the real problem of social life in Francistown is not the absence of meaning, but the fragmented, incoherent, alienated, mutually contested and contradictory nature of the various repertoires of meaning (all of them projected onto the urban space) seeking to capture the Francistownian’s mind and actions.

          ?          ? I have already stressed a number of factors in the light of which this pattern can begin to be understood:

?the heritage of residential segregation between African and White population segments in the town;

?in the past and to a large extent (if we substitute ‘White and African elite?for ‘White? at present, the above factor is itself a manifestation of these two segments?different power positions within the capitalist mode of production and the colonial, subsequently post-colonial state.

          ?          ? Francistown, although a thriving town in a eminently successful independent African country, yet displays patterns of class and racial contestation of the urban space that one would associate rather with conditions of White minority rule in the Southern African sub-continent. Because of its history of mining, monopoly capitalism and land alienation by the Tati Company, Francistown and the North East district are more reminiscent of South Africa than any other part of Botswana, whose benevolent Protectorate rule in general did not lead to conspicuous White presence nor oppression.

          ?          ? But this basic structure, which would apply in most African towns in the colonial period, does not really explain the peculiarities which, to my mind, make Francistown rather unique. Other factors would add more specificity:

?The role of this town as a railway head and as a shunting point in the international transfer of labour migrants between their rural homes and their place of work on the South African Witwatersrand. Francistown and the surrounding villages have absorbed large numbers of stranded male labour migrants from Zambia, Malawi and Zimbabwe. Therefore Francistown society has been less continuous with the surrounding countryside than might have been the case if it had been exclusively populated from there.

?The continued partial reliance, on the other hand, on rural, and partly non-capitalist, modes of production pursued outside town, makes for a relative lack of insistence on the signification of the urban space among those many Francistownians who feel that their ‘home?(as the hub of their meaningful universe) is elsewhere anyway. One can afford to live in a town without street-names, monuments or other publicly articulated symbolism, if one has an eminently more meaningful social space elsewhere.

?Therefore there is a continued projection of a diffuse rural spatial symbolism onto the rivers, hills, and (e.g. in the the low mud fence, the sangoma lodges, the symbolism of African Independent churches) the residential plots of the town.

?The point about these ‘traditional? (i.e. non-capitalist, non-statal, non-White) elements in the town, however, is not so much that they are there, but that they have not been negotiated to become conspicuous elements in the publicly and centrally constructed symbolism of the urban space. They are to a considerable extent underground, tacitly acknowledged and utilized by the African population as major components of their ‘collective private?domain, but never articulated in the latter’s interaction with powerful representatives of the state and of capitalism: civil servants and employers. They help to constitute the invisible but virtually impenetrable spatial and conceptual boundary which divides Francistown society, even despite the superficial appearance of both Whites and African elites, and African working-class migrants, as being increasingly united in a relentless pursuit of mass consumption and the other spoils of an economically viable more or less democratic post-colonial state.

          ?          ? The discovery of a ‘collective private?symbolic domain ? full of historic rural references ?among African working-class urbanites, virtually inaccessible for Whites/elites, should not make us close our eyes for the extent to which the urban space is symbolically structured by the internal dynamics within and between the three cosmopolitan repertoires: the state, capitalism and Christianity. The interpreta­tion of elements of meaning in the Francistown urban space should include an analysis of the principal contradictions and power blocks within Botswana society. Without aiming at an exhaustive discussion, let us try to highlight significant aspects by a number of strategic questions.

          ?          ? Why is it that (in the face of the boom, especially in the most recent years, of prestigious architecture associated with commerce and industry) even in the quarter of a century since Independence, the Botswana state has been rather reticent to leave its imprint on the local townscape? The new civic centre is small-and-beautiful, most of the other offices of the central and municipal administration are bleak colonial-type structures failing to appeal to the imagination, and so is ? tucked away in the Light Industrial Site across the railway line ?the office of the ruling party. And the Tatitown urban customary court, one of the two main outlets of a (totally routinized and police-dominated) version of customary law in Francistown, was until recently crammed into the few square metres of a circular zinc temporary shelter! The main exception is formed by the military, a rising but still reasonably well-contained force within the Botswana constitutional democracy. The signal lights and the pylon on Nyangabgwe Hill, and the air-force hangar, loom large over the townscape, but their utilitarian form can hardly be said to be inspired by the desire to display state power.

          ?          ? Part of the answer seems to lie in the increasing disjuncture, in the Francistown context, of capital and the state. Although both cosmopolitan complexes represent a departure from local tradition, they can no longer be treated as the obvious allies they were in the colonial period. Then the White element pervaded both, whereas now we see only the most entrenched non-African local elites (the Haskins’s, the Ebrahim’s) bother to penetrate the Botswana state (as BDP politicians, Speaker of the National Assembly, Mayor of Francistown), whereas the building and industrial boom has relegated the White category in town to a position of power which the state is eager to please rather than to control or co-opt. The very paucity of monuments of Francistown also largely stems from this disjuncture: as a White creation, the obvious thing for Francistown Whites before 1966 would have been ?like happened in Zimbabwe in the 1950s-1970s ?to raise reminiscences of the colonial and mining past to monument status. However, in modern times it is the state rather than a town council or an ethnic group which defines, erects and protects monuments as a legal category, and after Independence the Tswana-dominated state of Botswana has seen little reason to raise the oldest colonial buildings and other vestiges of the White-dominated founding period to monument status. Despite its Tswana name (which means ‘Display Culture?), White elements prevail in the Supa Ngwao Museum, a small Francistown middle-class voluntary association[xlvi] seeking to promote cultural activities and dreaming of a Francistown local museum, to imitate the success of the Serowe Museum. The latter has been rather successful, but it presented an obvious case: situated in the traditional capital of the dominant Ngwatoland region (the home of King Khama III, Regent Tshekedi, and the first BDP President and State President Sir Seretse Khama), and even housed in Khama III’s state villa. There is no national political reason to invest in White monuments in Francistown, and moreover, due to its colonial and post-colonial history of Kalanga, anti-Tswana identity, there is every reason for the state to limit the monumental repertoire of Francistown to a few brass plaques in the facades of prestigious buildings, commemorating their being among the blessings of the central government and their having been opened by a senior government official.

          ?          ? This brings us to the ethnic dimension[xlvii] in the articulation of meaning in the Francistown urban space. Why, with the general trend of Africanizing colonial place-names, has it proved impossible to substitute ‘Nyangabgwe?for ‘Francistown??even if the former name is already colloquially used? The region in which Francistown arose has for centuries been mainly occupied by Kalanga speakers, with minorities of speakers of Tswana, Ndebele and San. Because of far-reaching land alienation by the Tati Company, and the town’s peculiar history of migrancy which made Kalanga women (catering for international migrants) rather more conspicuous in town than Kalanga men, the Kalanga failed to articulate themselves as the typical ‘host tribe? and Kalanga was never the town’s lingua franca. After Independence many Tswana-speakers from all over Botswana came to swell the ranks of the urban African population, and Tswana hegemony consolidated itself within the state and the ruling party. However, it was in Francistown that Botswana first major and enduring political party (Botswana People’s Party, BPP) was founded in 1960 (by P. Matante, K.T. Motsetse and M. Mpho), and (like other Francistown-based opposition parties gradually to emerge) it strongly appealed to Kalanga ethnic and linguistic sentiments.[xlviii] During the first twenty years, Francistown was a BPP stronghold, and largely in control of the Town Council and the Mayorship. The Tswana centre’s strategies of populist co-optation and avoidance of open conflict, in addition to the largely locally-confined resources of a BPP municipal administration now supplanted by a BDP one, explain both the reticence of state manifestations in the townscape, and the failure to articulate unmistakable expressions of Kalanga ethnic identity on the public scene of Francistown.

          ?          ? Beside White associations and Kalanga-Tswana ethnic conflict, there is another aspect to the relative reticence of the state in the articulation of the Francistown urban space. Despite its inevitable shortcomings,[xlix] Botswana is a functioning democracy. Populism has been one of the strategies of the ruling, Tswana-dominated elite to gain the kind of grassroots-level support that has gained the country international admiration. The prevailing model of socio-political order revolves around consensus, consultation and participation, much more than around state oppression and obtrusive manifestations of state hegemony. In this respect the Botswana state is really a remarkable exception to the pattern most African post-colonial states have displayed in the 1970s and 1980s. Such an orientation of the state does not tally with an intimidating state presence in the townscape through megalomaniacal architecture. The BDP elite’s conservative conception of Botswana as a largely rural, pastoral, peaceful society in which towns are almost undesirable blemishes scarcely eligible for state subsidy, implies that any fundamental cleavage or contradiction between state and civil society is denied: the state is almost considered a large-scale village, and its subjects, as Batswana, are supposed to share in that imaginary village’s neo-traditional values and thus stick together. An emphatic state presence at the local urban scene through intimidating buildings would be incompatible with that ideology. In this respect it is very instructive that the most impressive and massive recent building the state has erected near the town centre of Francistown had to be a hospital: the new Nyangabgwe Hospital which reminds the population that state power is benevolent and coincides with their own self-interest. A peaceful struggle attended the naming of this building: Kalanga activists insisted that the hospital should be named after Kalanga and BPP politician Matante who died in 1979 [check]; shunning the ethnic and party-political divisiveness implied in such a name a compromise was reached by giving the superb building the Kalanga name of the hill which dominates Francistown ?but after which it could not yet be renamed itself.

          ?          ? Meanwhile this emphasis on state aloofness[l] also means that the state has far from realized its full potential to mediate such contradictions as exist within the civil society, and particularly, to limit White and capitalist dominance; hence its inability, more than twenty years after Independence, to erase the invisible boundary that runs through the system of urban meaning, perception and interaction. A greater input of unifying symbolic engineering on the part of the state might have done Francistown a lot of good. The state’s reticence is partly responsible for the relative absence of publicly articulated carriers of meaning, which makes Francistown such a bleak place.

          ?          ? All this suggests that, ultimately, ‘making sense of urban space in Francistown??the interpretation of patterns and contradictions of meaning as expressed in the urban space ? consists in highlighting the struggles, on the part of the state and capital, to control both each other and civil society, including such vital ideological expressions as represent repertoires of meaning which are independent from and cannot be reduced to, the logic of the populist state, wage labour and mass consumption. A continued reliance on historic, rural forms is one such repertoire, and its main strategy turns out to go underground, to make itself invisible in the urban space except for those who already know. Kalanga ethnicity constitutes a closely related repertoire of meaning, and as I am arguing elsewhere (van Binsbergen, in press) it has great difficulty of articulating itself in the face of Tswana linguistic, cultural and political hegemony, and of the rewards the latter offers to those who accept to be Kalanga only in their inner rooms, while publicly submitting to Tswana dominance.

          ?          ? Christianity is another such repertoire. Why is St Patrick’s street a short back-street, and are churches (even cosmopolitan, main-line ones) so inconspicuous in the townscape, despite the fact that the number of Francistown urban Christian congregations would be somewhere between sixty and a hundred? African Independency is in terms of numbers of adherents, number of congregations, and the variety of services offered to the population (including therapy and social belonging) by far the major Christian expression in Francistown. For obvious reasons the links between Independent churches and the White-dominated capitalist complex are limited. Independent churches?strategies vis-a-vis the state vary between accommo­da­tion and submission on the one hand, and on the other strategic evasion in the appropriate awareness of the churches?own hold on civil society (van Binsbergen 1990b). Here again the townscape (characterized by the aloofness of both state and churches at the overall level of the town as a whole) confirms the dilemmas of power and meaning existing in the local and national society. In this context, moreover, Francistown occupies a peculiar place: along with the surrounding North East (formerly Tati) district, it formed a White-controlled refuge for Independency[li] while, during most of the colonial period, elsewhere in the Bechuanaland Protectorate the chiefs, far from having their powers eroded by the colonial administration, could selectively favour specific cosmopolitan church missions.

          ?          ? Finally, there is one major aspect of the creation of meaning in Francistown which for reasons of space and economy of argument has regrettably to remain outside our present scope: the mediation of cosmopolitan, increasingly global images of identification and evaluation through modern electronic media, literacy, and formal education. The pursuit of education, not only for children and adolescents but also for adults in night classes, Bible courses and on-the-job training, is a major activity in Francistown, and no longer one confined to an intellectual or economic elite. The state, capitalism and world religions are each conspicuous in this domain, as agencies organizing and financing formal education,[lii] offering contexts in which formal education is a key to status advancement and power, and structuring this domain increasingly in the direction of commoditification ?of market forces and commercialization. The impact of these complexes on the articulation of meaning in Francistown’s urban space is considerable, but while my data do extend into this domain, I cannot go into a specific analysis here.

8. Postscript

When I returned to Francistown for additional field-work in April-May 1992, I found the process of naming the town’s streets and putting up street signs to be suddenly in full swing. The Francistown Town Council’s attempt to involve the local population in the naming has led to a few names being rejected and a few popular suggestions being adopted. Yet the overwhelming emphasis on street names that are non-local, non-Kalanga, and conceptually alien (like a number of street names evoking Ngwaketse regiments ?only meaningful to Batswana from Kanye district in the south of Botswana; or street names referring to the English or Latin names of plant species and to English names for astronomic constellations) by and large support my argument, which has stressed the absence of publicly articulated carriers of meaning in Francistown. Regrettably I must reserve a fuller analysis of the new Francistown street names for another occasion. Further field-work prompted by having written the present paper also opened my eyes to the conceptualization of the intimate and personal space in the village situation, and to the considerable continuities that exist in this respect between village and town; I regret, again, that it is impossible to incorporate this insight in the present paper already gone to print.


Beach, D.N., 1980, The Shona and Zimbabwe, 900-1850, Gwelo: Mambo Press.

Chirenje, J. M., 1977, A history of Northern Botswana 1850-1910, London: Rutherford.

Chretien, J.-P., & G. Prunier, red., 1989, Les ethnies ont une histoire, Paris: Karthala/Agence de Cooperation Culturelle et Technique.

Cooper, D., 1980, ‘How urban workers in Botswana manage their cattle and lands? National Migration Study Working Paper 4, Gaborone: Central Statistical Office.

Cooper, D., 1982, ‘An overview of the Botswana urban class struggle and its articulation with the rural structure: Insights from Selebi-Phikwe? in: Hitchcock & Smith 1982: 245-54.

Daneel, M.L., 1970, The God of the Matopo Hills ?An essay on the Mwari cult in Rhodesia, The Hague/Paris: Mouton for African Studies Centre.

Egner, E.B., 1971, ‘Interim report of the Tati settlement project, Francistown? Francistown: s.n.

Epstein, A.L., 1958, Politics in an urban African community, Manchester: Manchester University Press.

Epstein, A.L., 1967, ‘Urbanization and social change in Africa? Current Anthropology, 8, 4: 275-95.

Feddema, H., 1987, Overleven door samenleven: Naledi, een armenwijk in Botswana, Amsterdam: VU [Free University] Uitgeverij.

Furnivall, J.S., 1944, Netherlands India: A study of plural economy, Cambridge: Cambridge University Press.

Gerold-Scheepers, T.J.F.A. & W.M.J. van Binsbergen, 1978, ‘Marxist and non-marxist approaches to migration in Africa? in: W.M.J. van Binsbergen & H.A. Meilink, Migration and the transformation of modern African society, African Perspectives 1978/1, Leiden: African Studies Centre, pp. 21-35.

Geuijen, K., 1992, ‘Postmodernisme in de antropologie? Antropologische Verkenningen, 11,1:17-36.

Gluckman, M., 1945, ‘Seven-year research plan of the Rhodes-Livingstone Institute of Social Studies in British Central Africa? Rhodes-Livingstone Journal, 4: 1-32.

Gluckman, M., 1960, ‘Tribalism in modern British Central Africa? Cahiers d’Etudes Africaines, 1: 55-70.

Hellman, E., 1935, Rooiyard: A sociological survey of an urban native slum yard, M.A. thesis, University of Witwatersrand; reprinted, Rhodes-Livingstone Paper No. 13, London: Oxford University Press for Rhodes-Livingstone Institute; reprinted, Manchester: Manchester University Press for Institute for Social Research, 1969.

Hitchcock, R.R., and M.R. Smith, (eds), 1982, Settlement in Botswana, Gaborone: Botswana Society.

Holm, J., & P. Molutsi, 1989, eds, Democracy in Botswana, Gaborone: Macmillan.

Kapferer, B., 1988, ‘The anthropologist as hero ?Three exponents of post-modernist anthropology? Critique of Anthropology, 8, 2: 77-104.

Kerven, C., 1976, ‘Migration and adaptation to Francistown, N.E. Botswana? Botswana Notes and Records, 8: 301-302.

Kerven, C., 1982, ‘Rural-urban interdependence and agricultural development? in: Hitchcock & Smith 1982: 234-244.

Krijnen, E., 1991, ‘PWD: Een squatterwijk tusen staat, stad en platteland: Verslag van een onderzoek in Francistown, Botswana? M.A. thesis, societal history, Erasmus University Rotterdam.

Kwelegano, C.M.T., 1978, ‘The history of Francistown to the 1960s? B.A. thesis, Department of History, University of Botswana.

Lagerwerf, L., 1982, ?i>They pray for you...’’: Independent churches and women in Botswana, Leiden: Interuniversity Institute for Missiological and Ecumenical Research (IIMO), 1982.

Lesholo, J.M.O., 1982, ‘The new towns of Botswana? in: Hitchcock & Smith 1982: 293-302.

Masale, G., 1985, ‘Ethnicity and regionalism in Botswana and their impact on elections: A case study of the North-East and Francistown? paper presented at the workshop on preliminary results from 1984 election study project at the University of Botswana, Gaborone, 16th to 19th May 1985.

Meillassoux, C., 1975, Femmes, greniers et capitaux, Paris: Maspero.

Mitchell, J.C., 1956, The Kalela dance: Aspects of social relationships among urban Africans in Northern Rhodesia, Manchester: Manchester University Press.

Mitchell, J.C., ed., 1969, Social networks in urban situations, Manchester: Manchester University Press.

Molamu, L.L., 1989, First draft report on Francistown Phase IV development area study, Gaborone: Applied Research Unit, Ministry of Local Government and Lands.

Molokomme, A.L., 1991, ‘Children of the fence? The maintenance of extra-marital children under law and practice in Botswana, Leiden: African Studies Centre.

Morton, F., & J. Ramsay, (eds), The birth of Botswana: A history of the Bechuanaland Protectorate from 1910 to 1966, Gaborone: Longman.

Murray, A., H. Nengwekhulu, & J. Ramsay, 1987, ‘The formation of political parties? in: Morton & Ramsay 1987: 172-186.

Nas, P.J.M., 1992, ‘Jakarta, city full of symbols: An essay in symbolic ecology? paper read at the symposium ‘Urban images: Cities and symbols; symbols and cities? Leiden, January 1992; revised version in the present volume.

Nencel, L., & P.Pels,. eds., 1991, Constructing knowledge: Authority and critique in social science, London: Sage.

Nengwekhulu, H.R., 1979, ‘Some findings on the origins of political parties in Botswana? Pula, 1, 2: 47-76.

Picard, L.A., 1987, The politics of development in Botswana: A model for success?, Boulder & London: Lynne Rienner Publishers.

Ramsay, J., 1987, ‘Resistance from subordinate groups: Babirwa, BaKgatla Mmanaana and BaKalanga Nswazwi? in: Morton & Ramsay 1987: 64-81.

Republic of Botswana, 1974, A social and economic survey in three peri-urban areas in Botswana, Gaborone: Central Statistics Office.

Republic of Botswana, 1983, Guide to the villages and towns of Botswana: 1981 population and housing census, Gaborone: Central Statistics Office.

Republic of Botswana, 1984, Francistown development plan: Report of Survey, s.l.

Republic of Botswana, 1986, Report of the Francistown Planning Area Committee, Ministry of Local Government and Lands, s.l.

Schapera, I., 1943, ‘The native land problem in the Tati district? unpublished report to the Bechuanaland Protectorate Government (cf. Schapera 1971).

Schapera, I., 1971, ‘Native land problems in the Tati district? Botswana Notes and Records, 3, 219-268.

Tapela, H.M., 1976, ‘The Tati district of Botswana, 1866-1969? Ph.D. thesis, University of Sussex.

Tapela, H.M., 1982, ‘Movement and settlement in the Tati region: A historical survey? in: Hitchcock & Smith 1982: 174-188.

Tlou, T., & A. Campbell, 1984, History of Botswana, Gaborone: Macmillan.

Tshambani, M.M.W., 1979, ‘The Vapostori of Francistown; the study of an African independent church from 1951 to the present? B.A. thesis, Department of History, University of Botswana.

Tyler, S., 1987, The unspeakable: Discourse, dialogue, and rhetoric in the postmodern world, Madison: University of Wisconsin Press.

van Binsbergen, W.M.J., 1981, Religious change in Zambia, London/Boston: Kegan Paul International.

van Binsbergen, W.M.J., 1989a, ‘Housing procedures and urban social patterns: A preliminary statistical analysis of applications for site-&-service (SHHA) plots in Francistown in the years 1984-1988,?Francistown: Applied Research Unit Ministry of Local Government and Lands, Republic of Botswana/ African Studies Centre, Leiden.

van Binsbergen, W.M.J., 1989b, (with the assistance of E. Krijnen) ‘A preliminary quantitative analysis of plot ‘‘owners’’ in the PWD squatter area, Francistown, 1989? Francistown: Applied Research Unit Ministry of Local Government and Lands, Republic of Botswana/ African Studies Centre, Leiden.

van Binsbergen, W.M.J., 1990a, ‘Church, cult, and lodge: In quest of therapeutic meaning in Francistown, Botswana? paper presented at the 6th Satterthwaite Colloquium on African Religion and Ritual, Cumbria (U.K.), 21-24 April 1990; also seminar paper, University of Cape Town, August 1990, and University of Louvain, January 1991.

van Binsbergen, W.M.J., 1990b, ‘African Independent churches and the state in Botswana? paper presented at the conference on Power and Prayer, Institute for the Study of Politics and religion, Free University, Amsterdam, 10-14 December 1990.

van Binsbergen, W.M.J., 1991a, ‘De chaos getemd? Samenwonen en zingeving in modern Afrika? in: H.J.M. Claessen (ed.), De chaos getemd?, Leiden: Faculteit der Sociale Wetenschappen, Rijksuniversiteit Leiden, pp. 31-47.

van Binsbergen, W.M.J., 1991b, ‘Becoming a sangoma: Religious anthropological field-work in Francistown, Botswana? Journal on Religion in Africa, 21, 4: 309-344.

van Binsbergen, W.M.J., in press, ‘Minority language, ethnicity and the state in two African situations: the Nkoya of Zambia and the Kalanga of Botswana? paper read at the ‘Conference on African Languages, development and the state? SOAS, London, April 1991; in press in: R. Fardon & G. Furniss, ed., Language and languages [provisional title].

van Binsbergen, W.M.J., & Geschiere, P.L., eds., 1985, Old modes of production and capitalist encroachment, London/Boston: Kegan Paul International.

van Binsbergen, W.M.J., & J.M. Schoffeleers, 1985a, Theoretical exploration in African religion, London/Boston: Kegan Paul International.

van Binsbergen. W.M.J., & J.M. Schoffeleers, 1985b, ‘Introduction: Theoretical explorations in African religion? in: van Binsbergen & Schoffeleers 1985a: 1-49.

van Waarden, C., 1986, Places of historical and other interest in and around Francistown, Francistown: author’s edition (printed Stenographix Francistown).

van Waarden, C., 1988, The oral history of the Balakanga of Botswana, Occasional Paper no. 2., Botswana Society, Gaborone: Botswana Society.

Werbner, R.P., 1970, ‘Land and chiefship in the Tati concession? Botswana Notes and Records, 2: 6-13.

Werbner, R.P., 1971, ‘Local adaptation and the transformation of an imperial concession in northeastern Botswana? Africa, 41, 1: 32-41.

Werbner, R.P., 1977, ‘Continuity and policy in Southern Africa’s High God cult? in: R.P. Werbner, ed., Regional cults, New York: Academic Press, pp. 179-218.

Werbner, R.P., 1982b, ‘The quasi-judicial and the experience of the absurd: Remaking land law in north-eastern Botswana,?in: R.P. Werbner, ed., Land reform in the making, London: Rex Collins, pp. 131-150.

Werbner, R.P., 1985, ‘The argument of images: From Zion to the Wilderness in African churches? in: van Binsbergen & Schoffeleers 1985a: 253-86.

Werbner, R.P., 1989, Ritual passage sacred journey: The process and organization of religious movement, Washington/Manchester: Smithsonian Institution Press/Manchester University Press.

Wilson, G., 1942, The economics of detribalization in Northern Rhodesia, I & II, Rhodes-Livingstone Papers 5-6; reprint, Manchester: Manchester University Press for Institute for Social Research, 1968.

Wirth, L., 1938, ‘Urbanism as a way of life? American Journal of Sociology, 44: 3-24.



[i]               An earlier version of this paper was read at the Symposium ‘Urban Images: Cities and Symbols. Symbols and Cities? Leiden, January 6-9, 1992. I am indebted to the organizers for a most stimulating conference, and to Robert Buijtenhuijs, H.D. Evers, Vernon February, Peter Gutkind, Bruce Kapferer, Peter Nas, and Aidan Southall for comments on an earlier version.

[ii]            ?Field-work was undertaken in November 1988 to October 1989 and during shorter visits in August-September 1990 and June-July 1991. I am indebted to the African Studies Centre, Leiden, for the most generous encouragement and financial support; and to research participants, assistants, Botswana government officials, the Applied Research Unit of the then Ministry of Local Government and Lands, and members of my family, for invaluable contributions to the research.

[iii]            ?And not a displaced villager or tribesman ?but on the contrary ‘detribalized?as soon as he leaves his village (Gluckman 1945: 12); the latter reference shows that these ideas have circulated in African urban studies long before 1960.

[iv]              ?Meillassoux 1975; cf. Gerold-Scheepers & van Binsbergen 1978.

[v]            ?Cf. Chretien & Prunier 1989; for the history of the Kalanga, the major ethnic group around Francistown, cf. Werbner 1971, 1989; van Waarden 1988; Ramsay 1987; Tapela 1976, 1982; van Binsbergen, in press; and references cited there.

[vi]            ?Cf. Geuijen 1992; Kapferer 1988; Nencel & Pels 1991; Tyler 1987; and references cited there.

[vii]             Cf. Mitchell 1956, 1969; Epstein 1958, 1967.

[viii]              ?Multiplicity of meaning within a social formation consisting of fundamentally different and mutually irreducible sub-formations constitutes a condition for which postmodernism is not the only, and deliberately unsystematic, analytical approach. As a paradigm that preceded postmodernism by a decade in the circulation of intellectual fashions, the notion of articulation of modes of production is in principle capable of handling such a situation (e.g. van Binsbergen 1981; van Binsbergen & Geschiere 1985), were it not that its emphasis on enduring structure and a specific internal logic for each constituent ‘mode of production? renders it difficult to accommodate the extreme fragmentation and contradiction of meaning typical of the urban situation. The various cosmopolitan and local historic repertoires of meaning available in the Francistown situation as discussed here cannot convincingly be subsumed under the heading of a limited number of articulated modes of production. Yet while deriving inspiration from the postmodern position, my argument in the present paper is a plea for rather greater insistence on structure, power and material conditions than would suit the convinced postmodernist.

[ix]            ?Kalanga: nsha; Tswana: lolwapa.

[x]            ?The name Somerset occurs in a considerable number of toponyms in Southern Africa, and harks back to Lord C.H. Somerset (1767-1831), Governor of South Africa. In this particular case, the name Somerset was first given to a local mine, after which the mine gave its name to the various Somerset residential areas.

[xi]            ?E.g. along the lines proposed by Nas 1992.

[xii]             E.g. Werbner 1977; van Binsbergen 1981: ch 3.

[xiii]          ?Cf. Kerven 1976; Kwelegano 1978; Molamu 1989; Krijnen 1991; Republic of Botswana 1974, 1983 (section on Francistown), 1984, 1986; van Waarden 1986; Tshambani 1979; van Binsbergen 1989a, 1989b, 1990a, 1990b, 1991a, 1991b and in press.

[xiv]              ?Witwatersrand Native Labour Association.

[xv]             Schapera 1943, 1971; Tapela 1976, 1982; Werbner 1971; van Waarden 1986, 1988.

[xvi]          ?Cf. van Binsbergen 1990a; 1991b

[xvii]          ?Cf. Werbner 1989; van Binsbergen 1990a

[xviii]           In 1991 an additional track was completed from Francistown northwest to the brand-new Sua Pan soda-ash plant, 130 km along the Francistown-Nata road. This rail connexion is mainly used for bulk transport and has not yet had an impact on the symbolic physiognomy of the town.

[xix]          ?Named after the twenty-five-years jubilee of King George V of England (1910-1936).

[xx]             Van Binsbergen c.s. 1989; Krijnen 1991.

[xxi]          ?Another, more spacious military establishment is situated south of the road to Impala Ranch (a government nursery catering for the horticultural needs of Francistown Town Council, among others) and adjacent to the Donga residential area.

[xxii]          ?Its most prominent member was the late Hon. John Haskins, M.P., who combined his commercial enterprise with the office of Speaker of the National Assembly; he died in 1990.

[xxiii]           The ingredients for which are almost invariably bought, however, instead of being cultivated by the sellers themselves.

[xxiv]           John Mackenzie was a pioneer missionary in Botswana in the first half of the nineteenth century; Clifton I have been unable to identify so far.

[xxv]          ?In addition to the Francistown Club, established by White initiative in the 1910s, complete with adjacent golf course, Mopane club was launched in the early 1980s as an integrated alternative; however, after financial and managerial set-backs it had to be closed in 1987. The name of the alternative club was not without significance. Mopane is a beautiful tree common in Botswana; it houses the mopane larvae which are a welcome source of protein on the local diet but which Whites consider inedible. Thus the name Mopane conveys Botswana pride and resistance to alienness.

[xxvi]           The complementarity between White/elite’s and African workers?use of the river (a natural boundary for the former, a convenient passage-way for the latter) goes even further. When Tati River is closed as a passage-way for Blacks because it is flowing, White elites occupy it for a few days for recreative purposes, sailing on it on inflated rubber mattresses! Incidentally, because of their lining of shrubs and trees, the rivers are also convenient places for outdoors sexual intercourse ?no mean advantage in a town where, among working-class Africans, housing shortage and lack of privacy often create a shortness of convenient places for casual or illicit sex.

[xxvii]        ?Southern Africa’s widespread cult of the land, whose central figure is the High God Mwali, and whose many local shrines empower ritual status within the cult, and advice on personal and community matters. Cf. Daneel 1970; Werbner 1977, 1989; van Binsbergen 1991b; and references cited there.

[xxviii]              ?Interview with Mr E.B. Egner, Gaborone, 16 August, 1990; cf. Egner 1971.

[xxix]           Who qualified for a plot in these new areas regardless of their socio-economic status ?which given the virtual absence of decent housing for Africans could be considerable; the negative stereotypes concerning squatters as destitute and lawless are largely inapplicable in Francistown.

[xxx]          ?P=Pula, the Botswana currency; P1 ? US$ 0.50. At the time this maximum was perhaps twice the average income of Francistown households. Today it would be rather closer to average. Of course, in a society like Botswana where economic ties with rural areas are taken for granted among urbanites, the formal-sector income is not necessarily an urban household’s only source of cash.

[xxxi]           The Self-help Housing Agency (SHHA) policy discussed here was frozen in mid-1990.

[xxxii]        ?The names of Francistown residential areas as displayed in diagram 1 suggest a similar lack of shared discourse. Of more than thirty names of urban districts and residential areas, many reflect a meaningless bureaucratic classification (e.g. Area W, Block 2); others refer to the state, capitalism and Christianity as cosmopolitan repertoires of meaning (Government Camp, BDF, PWD, Central Business District, Monarch ?i.e. Monarch mine ? Catholic Plots); and only a handful are names in African languages with identifiable meanings such as Kgapamadi (‘Eye Socket full of Blood? Tswana) or Madzibalori (‘Uninhabited Marsh-land? Ndebele).

[xxxiii]        ?When it was announced, in the late 1980s, that PWD was finally to be removed, those having occupied the place for a considerable number of years were made to expect compensation for the demolition of their permanent structures, while all occupants were made to expect a SHHA plot in the nearby relocation area. When registration of PWD was already in full swing, people who did not reside in PWD would build flimsy houses overnight in an attempt to qualify for relocation plots.

[xxxiv]        ? Source: van Binsbergen c.s. 1989. Only plot ‘owners?are identified; the number of ‘structures?(dwellings) per plot owner may be higher than one, and tenants (who are relatively few in PWD) have a different household and possibly a different village home from the plot ‘owner? therefore it is important to note that the diagram ignores those heads of household who are not plot ‘owners?but tenants. Note that the vertical scale is not identical to the horizontal scale, due to the plotting routine used. The letters refer to the plot owner’s home village (all in Botswana):

A = Tonota         K = Shoshong         U = Sefhare  ?         % = Siviya    ?        

B = Serowe         L = Ramokgwebane         V = Pilikwe  ?         * = Marapong                       

C = Mathambgwane         M = Moroka         W = Mmadinare                      + = Mabeleapodi                                         ?

D = Mahalapye         N = Gulubane         X = Maunathlala          ?         @ = Sehope  ?         

E = Mochudi         O = Francistown         Y = Matsiloje                  ? = village home occurs

F = Bobonong         P = Tshesebe         Z = Mandunyane        ?                              ?   ?only once, therefore

G = Tlokweng         Q = Sebina         ?/span> = Maun    ?                              ?   ?no clustering possible

H = Kalamare         R = Palapye         # = Rakops  ?        

I = Senyawe         S = Tutume         $ = Gweta              2 missing cases (= 2 plots)

J = Borolong         T = Tsamaya         & = Makaleng              ?         excluded from diagram

[xxxv]        ?E.g. van Binsbergen 1981; van Binsbergen & Geschiere 1985.

[xxxvi]        ?Cooper 1980, 1982; Kerven 1982.

[xxxvii]         On this topic, cf. Molokomme 1991 and references cited there.

[xxxviii]      ?Often along with ?parts of ?one or more other residential areas.

[xxxix]        ?In this respect the socio-political organization at the neighbourhood level in Francistown is considerably more amorphous and ineffectual than that found in Botswana’s historic ‘tribal?capitals such as Serowe, Mochudi and Kanye. Although the latter have of course taken on increasingly modern urban characteristics (considerable heterogeneity of the population in occupational, class, ethnic, religious and political respects; and the adoption of central functions for not only the ‘tribal?administration but also the national state and the capitalist economy), much of their internal segmentation into wards has survived and continues to function in the judicial domain, in kinship and marriage, etc. For an interesting study of the vital social process in a Gaborone residential area, cf. Feddema 1987. For a general discussion of ‘new?towns in Botswana, cf. Lesholo 1982.

[xl]            ?Like the Lutheran Church of Southern Africa in Somerset East Extension, or the spectacular, bird-shaped church hall of the Guta Ra Mwari church in Area S; cf. van Binsbergen 1990b.

[xli]             The same reticence in self-advertisement is found among herbalists and other traditional healers, including the basangoma. Francistown has scores of healers, many of whom are officially certified as members of state-registered professional organizations, but very few of them display painted signs. People find their way to them by personal recommendation from kinsmen, friends, neighbours and colleagues at work.

[xlii]          ?Van Binsbergen 1990a, 1991b.

[xliii]          ?Cf. van Binsbergen & Schoffeleers 1985b for a specific discussion.

[xliv]          ?In this respect I can now see sangoma and Independent Church ritual to be far closer to each other in the confrontation of urban alienation than I suggested in an earlier analysis (van Binsbergen 1990a).

[xlv]          ?Van Waarden 1986, 1988; Beach 1980; Tlou & Campbell 1984.

[xlvi]          ?A driving force behind this association has been the archaeologist C. van Waarden, whose 1986 publication could be read ?not by accident ?as a catalogue of eligible Francistown monuments.

[xlvii]           Cf. Masale 1985; van Binsbergen, in press.

[xlviii]              ?Nengwekhulu 1979; Murray et al. 1987; Ramsay 1987.

[xlix]          ?Holm & Molutsi 1989; Picard 1987.

[l]              ?Reticence of the state as an implicit policy does not automatically mean reticence of state personnel. One of the striking contradictions of Francistown life is that within the local-level offices which serve as outlets of state power, the legal authority of the state is expounded by junior clerks with emphatic display of vicarious power and inflexible insistence on the letter rather than the spirit of official regulations, while civil conventions of greeting and respect are suspended in the process. This renders a particularly alienating dimension to bureaucratic interactions in Francistown. We should not expect publicly negotiated spatial symbolism to offer us a complete picture of power relations in the urban society; the contradiction between officially cultivated reticence and practical intimidation of clients inside the bureaucracies points to the paternalistic mechanisms by which constitutional order and elite dominance are maintained in contemporary Botswana.

[li]            ?Chirenje 1977; Lagerwerf 1982.

[lii]              ?Incidentally, the night-school variant is called ‘informal?in Botswana.




page last modified: 10-04-02 10:17:52 Easy Submit  
  Add Me!