UNIVERSITY OF THE WITWATERSRAND
 INSTITUTE FOR ADVANCED SOCIAL RESEARCH
 SEMINAR PAPER
 TO BE PRESENTED IN RICHARD WARD BUILDING
 SEVENTH FLOOR, SEMINAR ROOM 7003
 AT 4 PM ON THE 13 MAY 1996.
 TITLE: 'Nobody's Baby': The Politics and Perceptions of
 Informal Housing in Alexandra Township.
 BY: JUSTINE LUCAS
 NO: 400
'Nobody's Baby': The Politics and Perceptions of
 Informal Housing in Alexandra Township.
 Justine Lucas
 Department of Social Anthropology
 University of the Witwatersrand
Introduction
 In late February 1996, a group of property-owners in northern Johannesburg attacked
 residents of a squatter camp that had sprung up on their doorstep, and burnt down several
 shacks. Local government officials, while not condoning the action, promised to relocate the
 squatters to another site. The squatters, who were not told where they would be rehoused,
 refused to move, prompting further threats from the enraged property-owners. On April 29,
 there was an attempt to move the squatters to Diepsloot, far from their current homes and
 jobs. They resisted, and in the ensuing struggle police fired teargas and rubber bullets.
 This has been a familiar story over the last decade, as local government battled against
 endless waves of 'land invasions' while a new democratic dispensation was being negotiated,
 and mainly white property-owners protested against the rising crime and drop in property
 prices that they feared would result from informal settlement. But there is a twist to this tale.
 In 1996, two years after South Africa's first democratic election brought in a new housing
 policy, and several months after local government was elected to provide urban municipal
 services, the angry property-owners in the above account are residents of Alexandra
 township's East Bank suburb. Many of them are members of the ANC, and are represented
 by the ANC-aligned East Bank Civic Association. The local government officials who have
 promised to remove the squatters are also mainly ANC members; several of them former
 civic activists from Alexandra who championed the rights of squatters in the 1980s.
 Land invasions do of course present a problem for new democratically-elected local
 government, and the East Bank squatters in Alexandra are occupying land that has been set
 aside for low-cost housing development. This confrontation resonates with a nationwide
 concern about the obstacles informal settlements - generally seen as uncontrolled, unstable
 and violent - present to urban planning and development. In a thought-provoking discussion
 of these challenges, Mabin1 asserts that both apartheid and its opposition were characterised
 by modernist conceptions of planning which are inappropriate to today's 'postmodern' urban
 environment.2 Here, Mabin is in part referring to the social fragmentation that is the legacy
 of apartheid and the liberation struggle, and the inappropriateness of totalising ideologies to
 deal with this complexity. But he is also referring to a global shift in the make-up of urban
 space that confounds a modernist conception of urban planning based on the assumption of
 unilinear economic growth. A decline in mass industrial production has led to the
 suburbanisation of new information-based and high technology industries, leaving urban
 centres with a growing informal economy. In the absence of public housing programmes,
 private investment is put into offices and luxury housing, which coexist with slums and
 ghettoes. With this growing polarisation, the unemployed are excluded from the benefits of
 economic growth and forced into informal settlements, leading to new forms of social and
 spatial segregation. In South Africa, Mabin argues, these dynamics undermine the Freedom
 Charter's vision of redistribution, which was formulated in a period of mass production and
 industrial growth. And while the newly emerging planning idea of the 'compact city' aims
 to overcome fragmentation and segregation by bringing together working and living spaces,.
 Mabin suggests this is unlikely to have much impact given the zoning power of a deregulated
 land market.
The growth of informal housing is therefore not simply a consequence of rural-urban
 migration, but is one of the products of changes in the urban economy. This counters
 assumptions not only about unilinear economic growth, but also similarly modernist
 conceptions of urbanisation. Informal housing has however generally been understood as a
 transitional stage of a unilinear process in which people move closer to the city step by step,
 while necessarily maintaining a rural household base for security3. The social, political and
 cultural characteristics of shack-dwellers are then presumed to emanate from this liminal
 position between town and country. Bonner thus talks of a "transitional migrant culture"
 which, instead of assimilating into the broader urban culture, "lives on, in grotesquely
 parodied form",4 and Cole describes practices and institutions in Crossroads, modelled on
 "the invented tradition of bantustan structures", which "served to reinforce and perpetuate
 material and ideological linkages with the rural areas".5 In other words, informal settlements
 are understood to be culturally closer to the rural areas than to the city, but nevertheless
 culturally distinct: neither urban nor rural. Current changes in the urban environment
 however contradict this perception. Increasing numbers of urban-born people are moving
 from formal to informal housing,6 and the continued persistence of circular migration
 challenges the assumption that it is merely a transitional stage in a unilinear process of
 change.
 Population movements between town and country, and between informal and formal housing,
 are in reality flexible and multi-directional, and subject to changing strategies in an uncertain
 future. Recently, there has been some recognition that a modernist framework is inadequate
 to explain these dynamics. The search for a new approach has been described as a need to
 look at the social character of urbanisation. So, for example, Mabin7 argues that
 understandings of urbanisation which ignore social processes such as circular migration are
 not adequate to explain the dynamics of urban settlement, and particularly of informal
 housing. He discusses the contributions of 'rural' activities towards 'urban' households, and
 how variations in 'rural' households affect patterns of 'urban' settlement.8 Mabin concludes
 his review of urbanisation since 1960 with the statement,
 Moving into the 1990s, it appears that some social processes affecting urbanization,
 such as circular migration, may persist for substantial periods and prove little affected
 by policy changes. At the same time, informal urbanization has challenged authority
 and altered the make-up of cities - it may yet challenge their whole structure.9
 Bonner10 also responds to the "hidden assumptions" underlying modernist analyses of
 urbanisation with a historical exploration of its social character. He uses life histories to
 move away from grand narratives to more finely contextualised accounts of urbanisation.
 These life histories demonstrate how poverty and social instability in an "extraordinarily fluid
 and heterogeneous urban environment" produced constant changes in strategies to cope with
 a conditional urban existence.11
 The failure of South Africa's newly elected local government structures to resolve the
 problem of informal housing would seem to be a symptom of their inability to move beyond
 a modernist conceptual framework of urbanisation and economic growth, with the case of the
 East Bank squatters merely one of many similar instances. But while Alexandra is undeniably
 subject to the same forces that are reshaping the rest of Johannesburg, its situation cannot be
 understood only in terms of current economic and political developments. People who are
familiar with the history of Alexandra will not be surprised at the situation today, which
 displays marked continuities with the past. Political tensions around housing, and the
 perception of Alexandra as a problem for urban planners, have been recurrent themes
 throughout its history, which is in many ways unique.
 When the East Bank squatters were attacked, the press was quick to explain this as an
 expression of community anger that outsiders were jumping the queue for housing by staking
 a claim to this land. But the formally-housed residents of East Bank - a middle-class suburb
 built in the 1980s - are not in the queue for low-cost housing, and it was not competition for
 scarce housing resources that prompted their anger. And the squatters are no more 'outsiders'
 than are the bulk of the township's population today, the majority of whom live in zinc
 shacks. Many of these shacks are clustered around brick houses in the yards of 'old
 Alexandra', but in some areas they have taken over the landscape completely. Beside the
 graveyard, along the main roads bordering the township, in stormwater drainage areas and
 in any other under-utilised spaces, Alexandra today has effectively become a squatter
 township. Squatters have lived on the bank of the Jukskei river for years, and the only factor
 preventing the small number of shacks in the bush and scrubland of the far East Bank from
 turning into large-scale settlement was a lack of water. With the promise of housing and
 services in this area, informal settlement has now spread across the river. This is not simply
 a case of opportunistic outsiders usurping the rights of settled urbanites, and thereby
 provoking the outrage of those with a justified claim to housing, but is rather an example of
 the inexorable spread of Alexandra's housing problems beyond its boundaries, and
 encroaching on the previously shack-free suburb of East Bank.
 As a 'squatter township' Alexandra is different both from the rest of Johannesburg's northern
 suburbs and from most other townships or informal settlements. This difference is a legacy
 of its anomalous history as a 'location' - a formerly freehold area where property rights were
 granted a year before the introduction of the 1913 Land Act, and which survived the
 demolitions and removals meted out to other similar areas like Sophiatown and Western
 Native Township. Although usually described as a township, Alexandra predates the
 development of these vast, regulated and uniform 'satellite towns*, far from white areas,
 which were designed to house and control the African workforce.12 For a significant portion
 of its history, Alexandra fell outside municipal authority and remained free of much apartheid
 legislation. As an overcrowded and crime-ridden slum, Alexandra was 'Nobody's Baby'13 -
 situated in the heart of Johannesburg's northern suburbs, but too much of a liability for
 white municipalities to take responsibility for. The modernist conceptions of planning that
 underpinned apartheid thus largely failed to gain purchase in Alexandra. Instead, the state
 repeatedly, and unsuccessfully, tried to remove its population with a view to eventual
 demolition. Ironically, it was precisely this abdication of responsibility that underlay
 Alexandra's history of political resistance. While its population undoubtedly experienced
 some of the worst poverty and oppression of apartheid South Africa, it was always an
 uncontrolled and, from the perspective of the state, disordered space, permitting an autonomy
 of thought and action that undermined the vision of apartheid planners. Through the actions
 of the. ANC, and later civic organisations, Alexandra was constructed as an emancipatory
 space - most clearly defined during the uprisings of 1986 and the declaration of 'people's
 power'. In their attempts to create a new social and spatial order, these organisations were,
 as Mabin suggests, also trying to effect a modernist project. But the manner in which
 political organisations sought to represent the 'emancipatory space' of Alexandra was
continually contested by alternative orderings of space and social relations; not least in the
 growing areas of informal housing. Like apartheid modernism, then, the modernist strategies
 of democratic organisations never completely succeeded in their objectives. This was in part
 related to an inwardly directed political conflict, focused on housing, which has always
 characterised Alexandra.
 In this paper I look at how contestations over the representation of space in Alexandra have
 been shaped by the failure of apartheid's racial modernism, and explore the historical roots
 of current perceptions of informal housing as a problem for planners. Ferguson14 has argued
 that current modernist narratives of urbanisation have their roots in a liberal modernism that
 was developed to counter conservative policies of influx control during the first half of this
 century. Beginning with a historical overview of apartheid planning and resistance, I show
 how continuities from the past have indeed informed the post-1990 relationship between the
 ANC and the Alexandra Civic Organisation (ACO), and their ideas about informal housing.
 Despite the differences between the two organisations, both base their representations of
 space in Alexandra on modernist narratives of urbanisation that employ the dichotomies of
 urban and rural, and modernity and tradition. I argue that this obscures the reality of people's
 lived experiences, and their own constructions of space and social relations, and conclude by
 suggesting how we can reconceptualise the place of informal housing in today's 'postmodern'
 urban environment.
 South African Modernisms: from Garden Cities to Verwoerdian Apartheid
 Several authors have described the development of townships in the 1950s as a moment of
 'high' modernism, emerging from post-war economic reconstruction and fordist mass
 production, and informed by a modernizing vision of progress through scientific, rational
 state planning.15 In practice, of course, the imperatives of control and regulation that
 underlay this 'modernist project' were harnessed in racial oppression and economic
 retardation; a process that elsewhere has been termed 'the dark side of modernism'.16 While
 apartheid planning has in the past been viewed as a response to the needs of manufacturing
 capital,17 Mabin18 has drawn our attention to tensions and ambiguities in post-war housing
 policies, centring around this vision of modernisation and progress. This challenges the idea
 that the development of townships was the result of an apartheid grand plan, and resonates
 with other recent works on the fragmented nature of apartheid state power.
 This apparent paradox of apartheid as both modernist and fragmentary is not, however,
 unique to South Africa. Globally, there was not one unitary 'modernist project' but many,
 characterised by dynamic tensions between internationalism and nationalism, and between an
 Enlightenment-inspired vision of universalism and the efforts of elites to retain power and
 privilege.19 The character of modernism was thus influenced by both time and place. We
 can see this in the way that urban planning - a product of modernism - was exported from
 colonial centres to the peripheries, where it was shaped and transformed by local conditions.
 In the case of South Africa, town and regiopal planning owes its existence largely to the
 Garden Cities Movement, which was introduced from Britain in 1917.20 In Britain, this was
 a response to the 'premodern' slum conditions of nineteenth century cities: an attempt to
 replace disorder and squalor with order and health. The categories of urban and rural were
central to this essentially anti-urban movement, which sought to reconcile the divide created
 by industrialisation, and to restore the relationship of city-dwellers to the countryside by
 replacing the metropolis with smaller and less dense garden cities.21
 When these ideas and categories were exported to colonies like South Africa, they underwent
 a significant transformation. Colonial planning was constructed as a response not only to the
 'premodern', but also to the 'unmodern' or 'traditional' indigenous society.22 The resulting
 tensions are evident in Mabin's23 sensitive analysis of the fluctuations in post-war housing
 policies. Initially, the Garden City approach to regional planning, residential zoning and the
 creation of satellite towns was easily adapted to accommodate racial segregation. Likewise
 the Garden City objective of creating a healthy environment was harnessed in a racialised
 slum clearance programme or 'sanitation syndrome'.24 But the association of the 'unmodern'
 and 'traditional' African with the countryside was later to become a linchpin of the policy
 of separate development. This can be seen in the transfer of responsibility for black
 townships from local government to Verwoerd's Department of Native Affairs, the
 redirection of funds from urban public housing to homeland development, and internationally
 unprecedented attempts to enforce restrictions on African urbanisation.
 Apartheid planning policy was thus characterised by two conflicting strands of thought
 regarding the distinction between town and country. The first, drawing from the British
 experience of industrialisation, was a universal25 vision of modernisation which sought to
 dissipate the dangers of the city by breaking down the rural-urban divide. The second, driven
 by a desire for racial segregation, and drawing on a localised association of the countryside
 with the 'unmodern' that could not be merged with the modern city, sought to enforce a
 racialised division between town and country. This second approach understood African
 modernisation as progressing along a different path from white 'European' society.
 It should be clear that the contested nature of the South African 'modernist project' fed into
 and helped to shape the political debate between liberals and conservatives, specifically over
 the rights of urban Africans.26 The distinction Bozzoli27 makes between a period of liberal
 paternalism and later 'high racial modernism' can therefore be understood both in terms of
 political changes and shifting ideas about modernisation. While we may be able to isolate a
 particular period of 'high' modernism (whether we follow Mabin and Parnell in defining this
 as the 1950s public housing programmes, or whether we use Bozzoli's marker of intensified
 removals and influx control in the late 1960s), conflicting visions of the modernist project
 had been in circulation since the 1920s, and underlay both liberal paternalism and
 Verwoedian apartheid. The ambiguities and contradictions which these contestations wrought
 throughout the history of apartheid are particularly well illustrated by the many successive
 attempts to 'modernize' and control Alexandra.
 Alexandra 1912-1958: an 'uncontrolled' township.
 Alexandra was established in 1912, 1V4 miles from the boundary of Johannesburg, and since
 this municipality wanted no responsibility over the township it was envisaged that it would
 eventually become a self-governing municipality in its own right. The 1916 Health
 Committee, which was appointed to oversee administration, was intended to be the first stage
 in a process which would culminate in the election of a town council. It never, however,
 progressed beyond the second stage of its development, in which elected 'native' and
'coloured' officials joined the previously nominated white committee members. In 1932, a
 report by the Department of Public Health28 led to the committee being dissolved, amid
 charges that the elected officials were irresponsible, incompetent and corrupt. Ineffective
 administration was however inevitable, given the lack of financial resources and growing
 population of Alexandra. A more likely reason for the committee's dissolution was that the
 elected members, who represented the politically assertive and vociferous standholding
 middle class of Alexandra, had the power to overturn decisions made by their nominated
 white counterparts. This conclusion seems to be borne out by the government's refusal to
 publish the findings of the 1929 Young Commission of Enquiry, which stated that the
 problems were caused by inadequate revenue, and recommended that no changes be made
 to the Health Committee. This position was further supported by the 1936 Feetham
 Commission, which called for capital loans to be granted to Alexandra, which could then
 proceed, albeit "cautiously", along the path of self-government. In response to this report,
 the Johannesburg City Council instead established a committee to discuss the possibility of
 demolishing the township.29
 Alexandra's position outside the control of the state was thus already being perceived as a
 threat, although the proposed solutions to this problem differed. In 1926, before the
 Department of Public Health wrote its damning report about the Health Committee, it had
 in fact recommended the incorporation of Alexandra into the Johannesburg municipality; a
 recommendation that was later repeated by the Johannesburg mayor in 1944. The call for
 demolition, however, appealed to white residents of the areas surrounding the township, who
 in 1939 formed the North Eastern Districts Protection League to lobby for this eventuality.
 Their antagonism towards Alexandra was phrased largely in terms of a fear that its rapidly-
 growing and uncontrolled population would lead to the spilling over of crime and disease into
 white areas. This fear of disease and disorder was shared by white officialdom, as can be
 seen from a 1943 newspaper report citing a statement by the Johannesburg municipality on
 the conditions of Alexandra:
 This is a serious state when it is considered that the township is badly controlled, has
 an inadequate health service, the people occupying the rooms are of all ages and
 sexes, the sanitary service is mostly inadequate, the water drawn from the wells
 subject to contamination.30
 That this medicalised conception of disorder acted as a metaphor for moral deviancy is
 evident in the report of a doctor examining Alexandra squatters in 1947, expressing a fear
 of contamination from, among other ailments, venereal disease.31 In the same vein the
 Native Commissioner of Johannesburg was quoted as saying,
 I do not think anyone can question the growing lawlessness, lack of parental control,
 prostitution, defiance of authority, and signs of moral degradation, which are
 becoming features of the life of the Bantu in towns, particularly the younger
 generation. These conditions are known to exist in controlled locations, and it is not
 to be wondered at that in a densely populated and uncontrolled area, such as
 Alexandra, these features should be more marked and undoubtedly receive more
 scope for development. That the township is the refuge and harbourage of Native
 criminals and loafers of every type and degree, that liquor is brewed freely, and that
 vice is rampant, admit, in my opinion, of no argument.32
This theme of anxiety over disorder and a lack of control recurs throughout the successive
 attempts to resolve the problem of Alexandra. The establishment of urban order is of course
 characteristic of modernism in a very general sense, but the divergent responses to this
 question illustrate the contested nature of the South African 'modernist project'.
 The conditions of Alexandra in the 1930s and 1940s were similar to those found in
 Johannesburg 'slumyards'. In these areas, the emphasis of the British Garden City Movement
 on creating a healthy urban environment was accommodated in a racialised slum clearance
 programme. But while the same ideas were often invoked in attempts to 'clean up'
 Alexandra, wholescale removal would have been expensive and would have provoked
 political opposition, not least by Alexandra's Health Committee. By 1942, this body had been
 reconstituted to once again allow for locally elected officials, and included amongst its white
 members Hoernle, a prominent liberal, and the communist Bram Fischer. The state
 nevertheless viewed the growing political militancy of Alexandra with alarm. The bus
 boycotts, anti-pass campaigns and resistance to Bantu Education in the 1940s and 1950s were
 seen as further symptoms of disorder, prompting Verwoerd to announce it was time to bring
 the township "under very strict control".33 The solution was provided by the 1952 Mentz
 Commission, which recommended that Alexandra remain, but that slum clearance and
 removals reduce the population, and that buffer zones be established to separate the township
 from surrounding white areas. In 1958, Alexandra was brought under the control of the Peri-
 Urban Areas Health Board, which was to act on these instructions.34
 The Mentz Commission was also central in developing the planning guidelines for the
 establishment of townships in the 1950s. Although this body was acting on the instructions
 of Verwoerd, the policy that emerged was rooted in earlier British-influenced ideas about
 planning.35 In the decade leading up to the Mentz Commission, there was much debate
 about the applicability of British regional planning in South Africa. The Social and Economic
 Planning Council (SEPC), established in the 1940s, explicitly called for the implementation
 of the British approach, which it argued was the route to achieving social equity, an
 improved standard of living, higher productivity and modernisation. On the question of racial
 segregation, it proposed that urban communities be separated by 'green belts', but at the
 same time emphasised the integrated nature of the urban economy.36 Elements of the
 SEPC's recommendations were supported by Verwoerd's Department of Native Affairs, but
 social reconstruction was here reinterpreted to mean the development of new townships in
 the reserves, with funding for urban development elsewhere to be kept to a minimum.
 Verwoerd sought to have direct control over township planning, and instructed the Mentz
 commission on the cpre principles for planning guidelines, which ignored many of the social
 objectives of the SEPC.37 But this highly conflictual and contested process of developing
 township planning policy was nevertheless informed by a premise that was shared by both
 liberals and conservatives: essentially that "only by the provision of adequate shelter in
 properly planned native townships can full control over urban natives be regained".38 Most
 of the Mentz Commission's guidelines could not, of course, be applied to Alexandra, where
 population reduction and the creation of an industrial buffer zone and open 'green belt1 by
 Peri-Urban were the only available mechanisms of control. In many other respects, it
 remained outside state control, and countered the objective of creating 'properly planned'
 townships.
The question we need to ask here is what it was about Alexandra that was so 'disordered'.
 Was the state's conception of disorder in reality a concern at growing levels of political
 resistance: at the possibility of an alternative order challenging its attempts to impose a
 hegemony? Did residents of Alexandra in fact perceive the township as an ordered space, or
 was there also a localised fear of an underlying chaos? Levels of crime were undoubtedly
 high during this period, with a growing criminal class able to control the lives of both
 standholders and tenants, thus challenging ideas of order and status. But more significant for
 this discussion is the manner in which inwardly-directed political protests over housing were
 also characterised as 'disorder' by the township elite who dominated the ANC.
 Resistance or Disorder?
 When the Health Committee was again reconstituted to include elected officials in 1942, only
 registered standholders39 were eligible to vote or to stand for election. Although the
 Committee claimed to represent the interests of the whole community, it clearly acted in the
 interests of property owners, with whom it held monthly meetings.40 The Health Committee
 supported landlords' practice of building and renting out additional rooms on their
 overcrowded properties. Consequently, it was largely standholders who were concerned about
 threats to demolish the township, since they stood to lose their main source of income. The
 Alexandra Standholders' Association formed an Anti-Expropriation Committee, and the
 Health Committee promised to support them in fighting for the survival of the township.
 Tenants, on the other hand, were subject to exorbitant rents, which were much higher in
 Alexandra than in other freehold areas, due to the exceptionally high interest rates on
 township bonds.41 They thus had little to lose materially from being moved, and many were
 to leave Alexandra voluntarily in the 1960s. It has been suggested that in other townships at
 this time, class relations were 'softened' by other social forces such as patronage.42 In
 Alexandra, many landlords did indeed strive to construct their relationships with tenants as
 paternalistic, with the landlord as a benevolent father figure presiding over the semi-domestic
 space of his stand, or 'yard', which was seen as analogous to a family homestead.43 While
 this construction of yards as domestic spaces was often successful, the paternalistic moral
 authority of the landlord was frequently undercut by far more blatantly exploitative rentals
 and stark inequalities than elsewhere. The landlord-tenant nexus in Alexandra consequently
 became the focus of inwardly-directed political conflicts within the township. As Tourikis
 points out, "The so-called 'housing question' in Alexandra...was the question of the class
 struggle between African petty-bourgeoisie and working class",44 and it has continued to be
 so ever since.
 Standholders were thus able to use the Health Committee to advance their interests. What
 made this a threat to the state was that the same grouping also controlled the powerful
 Alexandra branch of the ANC - one of the largest in the country - thus complementing their
 access to the Health Committee. In the 1940s, the ANC in Alexandra was split between
 'Charterists', who supported the alliance with the working class-oriented South African
 Communist Party (SACP) and South African Congress of Trade Unions (SACTU), and
 'Africanists', who were strongly opposed to these organisations. Both groups were, however,
 comprised of the standholding middle-class; and the Africanists, led by the wealthy bus-
 owner and sometime national treasurer-general of the ANC, R.G.Baloyi, were largely in
 control of the branch, including its women's qnd youth sections.45
In the light of this situation, we need to understand that the rising political militancy in
 Alexandra was multi-faceted. While bus boycotts, anti-pass campaigns and resistance to
 Bantu Education were perceived by the state as a threat, and as a symptom of disorder, a
 significant portion of political resistance at this time was directed inward, at the standholding
 middle class. The first bus boycotts46, for example, were supported by the Alexandra
 Workers' Union, and directed partly against black bus-owners such as Baloyi. Standholders
 had little control over these events, which have been described as spontaneous mass
 movements with little political leadership.47 It was not until PUTCO bought out local bus-
 owners in 1944 that subsequent boycotts were more universally supported, organised and led
 by the ANC.48
 While the bus boycotts illustrate some of the political tensions within Alexandra, the issue
 these tensions arose from and coalesced around was housing. In the mid-1940s, Alexandra
 tenants started organising themselves in opposition to standholders, forming the Alexandra
 Tenants* Association. In 1947 thousands of tenants, led by Schreiner Baduza, protested
 against high rentals and inadequate housing by setting up a squatter camp in the middle of
 Alexandra: the township's first 'land invasion'. Baduza argued that tenants should leave
 Alexandra, and persuaded a number to move to Pimville.49 Not suprisingly, the squatter
 movement was vigorously opposed by the Health Committee and by standholders, who feared
 losing their tenants. While the Health Committee cautioned that the situation posed a threat
 to both Alexandra and Johannesburg, standholders were more forthright in their
 condemnation. In a telegram to the Prime Minister, the Minister of Native Affairs and the
 Administrator, they stated that they could "no longer tolerate the behaviour of the
 squatters...Therefore we demand their removal".50
 This approach was similar to that of the Johannesburg City Council, which during discussions
 on the possible demolition of Alexandra highlighted the role of tenants as presenting "gravest
 difficulties in regard to health, sanitation and law and order".51 In other words, while there
 were political actions during this period which were unambiguously directed at the state and
 universally supported, Alexandra was - and indeed still is - characterised by cross-cutting
 interests that produced shifting patterns of alliance and conflict. Significantly, both
 standholders and the state expressed a common interest in establishing order and control,
 which was threatened by the tenants' squatter movement. We therefore need to view
 resistance in Alexandra from two perspectives. The activities of the ANC, far from being a
 symptom of disorder, were attempting to counter the state's efforts to impose a hegemony,
 and to suggest an alternative non-racial social order. But since the ANC in Alexandra was
 dominated by standholders, whose vision of social and spatial order was based on property
 rights and paternalism, there was an internal contestation between organised political 'order'
 and spontaneous mass 'disorder' within Alexandra itself. While the UDF in the 1980s was
 to try to harness this dynamic as 'ungovernability', the ANC in the 1940s and 1950s tried
 to eradicate it.
 Returning to the question of divergent forms of modernism in South Africa, it is clear that
 the ANC in this period espoused a universal vision of modernisation and progress that was
 close to that of the liberal tendency within (and later outside) the state. The popular
 perception of tenants among both standholders and the liberally-inclined Health Committee
 was that they were recent arrivals in Alexandra, and thus not fully urbanised.52 This
 assumption needs to be questioned53, not least because it has continued to be invoked by
property-owners in relation to squatters for the last fifty years. Here, the notion of rural
 'tradition' was constructed in contrast to an urban 'modernity', with 'disorder' resulting from
 incomplete urbanisation and modernisation. In the discourses of both Alexandra standholders
 and white liberals, the modernity of permanently urbanised Africans was not questioned.
 With the increasing marginalisation of liberals from state power after 1948, a significant
 number supported ANC-led acts of protest in Alexandra, including the 1957 bus boycott and
 the establishment of 'ANC schools'. At this stage, then, the ANC was attempting to effect
 a 'modernist project' which was in line with the British-influenced liberal modernist tendency
 within the state; and as we shall see later, these ideas have continued to influence the ANC
 in Alexandra in the 1990s. In the 1960s, of course, there was a growing eclipse of this vision
 of liberal modernity by Verwoerd's 'high racial modernism'. Bozzoli54 asserts that this
 policy broke down old patronage networks and created a moral vacuum. Alexandra did
 indeed come under increasing state control in the 1960s and 1970s, but, as the following
 discussion will show, this did not succeed in destroying either the status and authority of
 elites, or localised challenges to them.
 The Failure of 'High Racial Modernism *
 The period from 1958 to 1979 was characterised both by attempts to intensify control over
 the Alexandra population, and by increasing conflict and uncertainty within the state over the
 future of the township. In the first few years of its existence, Peri-Urban was relatively
 successful in its objective of reducing the population. Permits were introduced in 1958 for
 those who were entitled to live in Alexandra, and by the end of 1960 25000 people had been
 removed. But despite these measures, people continued to flood into Alexandra. As long as
 freehold property existed, it was impossible for the state to curb this influx, since it was not
 in standholders' interests to reduce the numbers of tenants, and they refused police
 permission to raid their properties. Partly in response to this, Peri-Urban began buying up
 stands. This was also a necessary first step to being able to demolish Alexandra, and there
 was a great deal of uncertainty at this time over whether the actions of Peri-Urban were
 aimed at control, or were in fact a prelude to demolition. In 1960, Peri-Urban established
 a social welfare section, which funded libraries, schools, sporting activities and care for the
 aged. Money was spent on infrastructure, and by the early 1960s there was a marked
 decrease in crime. There was, however, conflict in government over this use of funds, which
 seemed to be predicated on the continued survival of the township.53
 Peri-Urban's success prompted the Minister of Bantu Administration and Development to
 report in 1961 that, "Thanks to the good work performed by the Board order was created out
 of chaos in this notorious town".56 Nevertheless, in the same year, an ad hoc committee
 established by Verwoerd to reassess the situation decided that Alexandra would be
 demolished to make way for a hostel township. In the meantime, new legislation would be
 introduced to intensify control over the existing population. The Better Administration of
 Designated Areas Bill, passed in 1964, served to make the extensive existing legislation on
 urban African residence applicable to freehold areas such as Alexandra.57 In Alexandra,
 extra legislation was introduced which banned meetings, prohibited the carrying of dangerous
 weapons, legislated conditions of tenure, and allowed for removals and state purchase of
 properties.58
 10
This "vigorous and bureaucratised modernism"59 was both a response to rising levels of
 political militancy and an expression of Verwoerd's vision of separate development, which
 entailed a forced separation of town and country. In the years that followed, the criteria
 defining people with the right to be urban became increasingly restrictive. By 1961, housing
 permits were available only to 'qualified' men over the age of 21, who were employed and
 had dependents legally living with them. All others, including single women with the 'section
 10' rights that allowed them urban residence, had either to move to the homelands or live
 a precarious existence as tenants. Even those men who were eligible for housing could have
 their permits cancelled at any time, if they were deemed not to be "fit and proper" persons.
 There is no question that state control intruded into people's lives in a manner which it had
 previously been unable to do. But there was opposition to these policies - both from within
 the state and in Alexandra - which were eventually abandoned without having achieved their
 objectives.
 During the debate over the Better Administration of Designated Areas Bill, the UP vigorously
 contested the NP's method of exerting control over Alexandra. While it was accepted by both
 parties that the township in 1963 remained overcrowded, unhygienic and uncontrolled, the
 UP asserted that many residents had become permanent town-dwellers, and that property
 rights would ensure social and political stability. They argued that the legislation already
 existed to 'clean up* Alexandra through reducing the population and improving housing and
 infrastructure.60 Here we can see the stark opposition between a liberal vision of
 modernisation which accepts African urban residence as permanent, and seeks to control this
 through reducing population, and the NP's 'high racial modernism' which, at the same time
 as it sought to control urban areas, associated Africans with the countryside and thus "issued
 an ideological onslaught upon the ideas of 'the urban1, turning it into an empty, hollow
 category with negative connotations".61
 Initially the NP's approach prevailed, and in 1973 the state centralised control over
 Alexandra. The West Rand Bantu Affairs Administration Board (WRAB), which was directly
 responsible to the Minister of Bantu Administration and Development, took over from Peri-
 Urban, announced that Alexandra would be transformed into a hostel township by 1984, and
 began expropriating properties. While the planned hostel township never materialised, a total
 of over 65000 people were removed and three hostels were built. The multi-storey Alexandra
 hostels, unlike the bungalow dormitories of other townships, were regarded as a unique
 experiment. Intended to be structures which would "be divided into controllable units of 150
 people, divided by electronically operated steel doors...to prevent unrest from spreading",62
 these were clearly envisaged as 'total institutions': "monuments to a mad form of
 modernity".63
 The remaining standholders however resisted, and formed the Alexandra Township
 Residents' Liaison Committee, led by Rev. Sam Buti, The combination of this resistance and
 the recommendations of the Riekert Commission, which proposed consolidating the rights
 of urban 'insiders', led to Alexandra being 'reprieved' in 1979. The plans that followed were
 similar to those proposed by the UP in the 1960s: Alexandra was to be replanned as a
 middle-class family township; and while WRAB continued to buy properties so that it could
 redevelop the area, houses would be made available for purchase under 99 year leasehold.
 The Liaison Committee was formalised, and at its inauguration, Minister Koornhof promised
 that Alexandra would now become a "model township".64
 11
This was an ambitious plan, since the state's attempts to control and regulate Alexandra had
 up to that point failed dismally. Despite the state's attempts to restrict urbanisation, and to
 enforce rigid control over the shrinking number of Africans who were accorded the right to
 call themselves urban, South African cities accommodated many who were excluded from
 urban status, such as migrants and unmarried women, and who were "left to shelter in the
 urban crevices...hidden from the public eye".65 Alexandra, with its overcrowded and
 impenetrable yards and alley-ways, was just such a crevice, where the order and control of
 the state still failed to gain purchase. Unlike the regulated and ethnically-zoned townships
 originating from the 1950s, Alexandra in the 1970s was a potent 'melting pot' of diversity:
 a theme which predominates in Serote's description of the area during this time:
 Alexandra is just that - a terrible stew. The stew bubbles, usually on Saturdays and
 Sundays. Everyone is in Alexandra. The grace of the dances, the variety of both the
 dances and the traditional garb, the many, many languages of the people of South
 Africa, all these, are packed and piled into the pot, Alexandra, and they bubble and
 bubble and bubble.66
 In this space, characterised by numerous distinctions of class and identity, the dichotomy
 between rural 'tradition' and urban 'modernity' was confounded. With cattle and Cadillacs,
 donkeys and Jaguars, fifteen-roomed houses and one-roomed houses existing side by side,
 Serote asks us with desperation, "What the hell is Alexandra?".67 The answer is that
 Alexandra was everything the modernist vision of the city was not: the chaotic, fragmented
 and disorientating labyrinth that European planners and architects sought to banish with their
 'new internationalisms'; their 'Radiant' or 'Garden cities'. Alexandra was the hidden
 underbelly of modernity; the Dark City to Johannesburg's Golden City: two sides of the same
 coin.
 From the centre of the Golden City to the centre of the Dark City is a mere nine
 miles. Where one starts the other ends, and where one ends, the other begins. The
 difference between the two is like day and night. Everything that says anything about
 the progress of man, the distance which man has made in terms of technology,
 efficiency and comfort: the Golden City says it well: the Dark City, by contrast, is
 dirty and deathly.68
 So Alexandra continued to confound all facets of the complex and contested vision of South
 African modernity. In this it challenged both 'high racial modernism' and the liberal
 modernism of the township elite. While a significant number of standholders still held
 property rights by 1980, many of those whose properties had been expropriated still inhabited
 their old yards. But without legal ownership, they struggled to maintain control over the
 increasing numbers of shack-dwellers who were flooding into the township. The state
 recognised the tensions this created, and sought to coopt the old elite into its planning
 strategy: a central component of its 'hearts and minds' campaign to counter political
 opposition in the 1980s.69
 In the early 1980s, there was a concerted attempt to turn Alexandra into a middle-class
 family township, and influx control was tightened to exclude shack-dwelling 'illegals'. This
 was a clear attempt to coopt the township elite, 300 of whom still resisted being bought out
 by WRAB. The 1980 redevelopment plan was however prevented from being fully
 implemented, both by standholders' refusal to sell and thus allow the township to be rebuilt,
 12
and by a lack of funds. Here we see a switch from the state as direct agent of control to the
 Alexandra Liaison Committee, which exhorted standholders to sell their properties and
 threatened 'illegals* with removal. Squatters whose shacks were demolished in 1981 were
 described as "not bona fides of the township. They (were) people who had just streamed to
 Alexandra from nowhere", and should return to the homelands.70 The remaining
 standholders did not oppose this (they continue to express much the same sentiments today),
 but they objected to Buti's stand on property rights. Antagonism to his Save Alexandra Party
 mounted, and two parties stood in opposition to Buti in the 1981 Liaison Committee
 elections. Buti won, but the poll was very low. One of the opposition parties was led by
 Mike Beea: a disgruntled former associate of Buti who fell out with him over an incident of
 expropriation. Beea was to become, and still remains, the chairperson of the Alexandra Civic
 Association (ACA). After the introduction of the Black Local Authorities Act in 1983, ACA
 joined with other democratic organisations in affiliating to the UDF.71 Ideological
 contestations however continued to characterise what has always been, and still remains,
 inwardly conflictual local politics.
 The NP had argued in the 1960s that by removing property rights it would destroy
 inequalities. This argument has often been repeated by scholars who suggest that classes were
 compressed under Verwoerdian apartheid.72 The NP strategy was thus an explicitly
 modernist attempt to break down an old social order and create a new one, countered by the
 UP's argument that breaking down the old order would instead result in disorder. Bozzoli
 similarly refers to the destruction of township elites, and the networks of patronage they had
 established, as creating a moral vacuum that allowed the 'ungovernability' of the 1980s
 township revolt to occur, and which continues to undermine current policies of reconstruction
 and development. In other words, she suggests, the impersonalised bureaucratic order of
 racial modernism, in breaking down an older paternalistic social order, failed in its objective
 of creating a new social order, and thus contained the seeds of its own destruction.
 In Alexandra, I would challenge these claims on several counts. Firstly, there is a significant
 degree of continuity in the social order and patronage networks of the elite right up to the
 present day, as well as the liberal modernist ideas they embodied. Many of the old
 standholders remain in the same yards, where they still attempt to assert a paternalistic moral
 authority over their former tenants, albeit with the same ambiguities and contradictions as
 before. While the voices of the elite were muted during the turmoil of the mid-1980s, they
 never disappeared, and are now being vigorously reasserted by ANC-aligned standholders.
 Secondly, the strategies of youth and civic activists in the mid-1980s to create a new moral
 and social order, informed by socialist rather than liberal ideas, were partially successful, and
 civic structures created in 1986 continue to exist under the auspices of the Alexandra Civic
 Organisation (ACO).73 This means that if we are to understand the current confrontations
 over housing in Alexandra, we need to look at the ambiguities and contradictions these two
 sets of ideas have created in the post-1990 alliance between ACO and the ANC.
 Contested Representations of Urban Space in Post-1990 Political Discourse
 The relationship between ACO and the Alexandra branch of the ANC in the early 1990s was
 characterised by both alliance and opposition: an ambiguity owing much to the ideological
 13
contestations of the 1980s. ACO was originally launched in 1986 as the Alexandra Action
 Committee (AAC), under the leadership of trade unionist Moses Mayekiso. Although AAC
 was aligned to the UDF, it was formed in opposition to the older UDF-aligned Alexandra
 Civic Association (ACA). While much of the ideological contestation during this period was
 framed in terms of the distinction between the 'Charterist' and 'workerist' camps of the
 UDF, there was a high degree of overlap between these categories, which operated as labels
 rather than as distinct groups, and both AAC and ACA claimed a charterist identity.74 The
 social character and ideological standpoints of the two organisations nevertheless diverged.
 ACA leadership was made up largely of professionals and property-owners: the old
 standholding middle class created during Alexandra's freehold era. AAC, on the other hand,
 consisted of youth and union activists who had a more working-class character, and several
 of its most prominent leaders were relatively recent arrivals in Alexandra who had no
 connection to the period of freehold tenure.75 They viewed ACA as bourgeois and populist,
 and instead sought to prioritise working-class interests. AAC's success in establishing yard,
 block and street committees effectively marginalised ACA, and at a mass meeting in May
 1986 AAC was unanimously chosen to represent the people of Alexandra.76 The formation
 of civic structures was a central component in the attempt to replace state control with organs
 of 'people's power': to construct Alexandra as an emancipatory space that rendered apartheid
 ungovernable. When AAC was relaunched as ACO in 1989, it built on this legacy, and was
 widely perceived as the sole legitimate civic in Alexandra. In 1990, however, the newly
 unbanned ANC entered into fraternal relationships with both ACO and ACA. The tensions
 that this produced exposed the existence of inwardly-directed political conflict, which had
 been temporarily overshadowed by a united resistance to apartheid in the 1980s, and which
 revolved primarily around informal housing. Associated with this conflict were significant
 contestations in the representation of space in Alexandra.
 When an ANC branch was formed in Alexandra after the organisation was unbanned, the
 newly-elected secretary was Obed Bapela, a former AAC activist who had been imprisoned
 with Moses Mayekiso when they and other AAC leaders were charged with treason.77 The
 election of a prominent civic leader onto the ANC executive illustrated the common
 perception that the UDF was effectively the internal wing of the ANC while the movement
 was banned, and highlighted the dominance of AAC and later ACO in UDF politics. Bapela,
 however, left civic politics when he was elected, thus potentially weakening the alliance
 between ANC and ACO. From 1990, no ACO office-bearers held office in the ANC, and
 although ACO leaders were all ANC members, the civic claimed to be representative of all
 the people of Alexandra, regardless of political affiliation. They represented the township in
 local government negotiations and played a prominent role in the formation of the national
 civic body, SANCO, to which ACO affiliated.
 In contrast to the growing separation of ACO and ANC, ACA maintained a close relationship
 with the ANC. Mike Beea of ACA was in 1992 chosen to represent the ANC and its allies
 on the Alexandra Interim Crisis Committee (ICC), which was formed to deal with the
 endemic violence in the area. The close relationship between ACA and ANC again illustrates
 the perceived continuity between the UDF and ANC, but here the ANC's association with
 an older charterist grouping that rivalled ACO served to undermine the ANC/ACO alliance.
 In the early 1990s, then, the relationship between ACO and ANC, while overtly one of
 alliance, was underlaid by serious ideological tensions. These were compounded by socio-
 economic differences between the two groups of leaders. Many returned ANC exiles, who
 14
were prominent in both the branch executive and as general members, came from the same
 middle-class social background as the ACA leadership: old Alexandra families who had
 claims to property in the township and were dismayed at the proliferation of shacks. These
 social differences served to underscore the diverging ideological standpoints of the two
 organisations, particularly with reference to housing and the 'squatter' issue. The relationship
 between ACO and ANC, and their approaches to informal housing, were not however static.
 The ambiguity of alliance and competition meant that local politics was characterised by
 fluidity, so that shifts in the ACO/ANC relationship were reflected in changing political
 discourses. While overtly addressing the issues of urbanisation and informal housing, these
 discourses embodied initially complementary and later contrasting representations of space
 in Alexandra. These were social maps through which ACO and ANC sought to make sense
 of the distribution of space in the township and to lay the basis for their claims to represent
 the community.
 ACO discourse:
 ACO's ideological starting point was clear in its adoption of the Freedom Charter slogan
 "Houses, security and comfort for all". It claimed to represent all the people of Alexandra,
 but was faced with an extremely diverse constituency, with differing short-term interests.
 Shack-dwellers, for example, desired access to housing, while potential property owners
 wanted to evict shack-dwellers from their yards. ACO attempted to overcome these divisions
 through the creation of inclusive democratic structures that would bring together residents
 of formal and informal housing. Its basic units of organisation were yard committees, which
 were organised into block and/or street committees, and these were in turn grouped into
 fourteen area committees presided over by an executive. ACO organisation thus rested on
 the concept of a pyramid of interlocking spatially-bounded units. One of its main difficulties
 was therefore how to organise in areas without yards: in large free-standing shack
 settlements, where there were no clearly-bounded units other than individual shacks. ACO's
 response was to make a distinction between shack-dwellers in yards, who participated in
 formal civic structures, and shack-dwellers in shack settlements. A separate shack-dwellers'
 coordinating committee was formed to cover the whole of Alexandra, independent of other
 civic structures within specific areas. This structure was eventually disbanded, due in part
 to charges that it promoted ethnic conflict (many shack settlements were perceived as being
 ethnically homogeneous), and individual shack settlements were then supposed to organise
 shack committees which would function in a similar way to yard committees, and feed into
 broader area committees. As we will see, this approach proved inappropriate for the social
 and spatial organisation of shack settlements.
 Shack-dwellers were often perceived to form the core of ACO's support-base, but civic
 structures were in fact very weak in free-standing shack settlements. Richard Mdakane, the
 then secretary of ACO, recognised the problem, and explained the civic's approach.
 Most people say ACO is based in shack areas, but we are very weak in shack areas.
 The basic structure of ACO is yard committees - tenants and sub-tenants. In shack
 areas, people don't come to meetings. Maybe we neglected to organise shack
 areas...Of course they organise themselves into strong structures, but not necessarily
 in our structures. We failed to give them political direction.78
 15
While making a distinction between shack-dwellers in yards and shack settlements, Mdakane
 recognised that many township residents lumped them together in one category - squatters -
 since people in formal housing had permits from the local council and people in shacks did
 not. Problems arising from the lack of legal access to housing were therefore common to all
 shack-dwellers.
 Because of the location of Alex near work, this attracts people. It would be difficult
 to stop the influx. I don't know how we would deal with it without more land. There
 are some problems in organising these people. Some are very conservative; some are
 members of other organisations; some are migrants and not interested in participating
 in structures. But housing is a serious question for them. Ultimately they can be
 organised by ACO. But short term they are conservative.79
 Mdakane was thus displaying an ambivalent attitude towards shack-dwellers, saying that
 while they had the potential for strong civic organisation, ACO was still weak in shack areas.
 A closer examination of Mdakane's ideas shows how his explanation of this problem rested
 on a particular understanding of urbanisation, which appeared in ACO discourse as a clearly
 modernist narrative. This narrative explained that people arriving in Alexandra from the rural
 areas would become part of the community only if they were organised and transformed by
 ACO into a radical urban working class and abandoned their previously 'conservative' or
 'traditional' ideas and practices, thus breaking with their rural past. Shack-dwellers' need for
 space could act as the catalyst in such a transformation, but this process would have to be
 directed to prevent them following alternative strategies of organising their space which
 would conserve and strengthen their outsiderness.
 Ideally, people were supposed to ask permission from local civic structures before erecting
 shacks. ACO's narrative of urbanisation facilitated and legitimized this process. More
 fundamentally, it described an ordering of space in which the township was represented as
 a construction of interlocking units coming together as a unitary whole which was
 simultaneously ACO, Alexandra and 'the community': an image which represented the civic
 as a hegemonic power. ACO recognised class as a category of social division, and prioritised
 working-class interests, but at the same time social divisions based on class, ethnicity and
 housing type were intended to be overridden by the unity ACO engendered. Any area which
 was not organised into ACO structures, such as Inkatha-controlled hostels,80 therefore
 represented an abberation - a problem and maybe a threat to the whole. In addition to Inkatha
 hostel-dwellers, shack-dwellers from outside Alexandra were the group Mdakane referred to
 as the most likely to organise autonomously in this way, facilitating the development of
 opposition to the hegemony of ACO. ACO's interpretation of the independent organisation
 of space as opposition which posed a threat to the whole 'community' was thus presented
 as real and natural. The media at this time often followed this line with an unquestioning
 acceptance of the violence as taking place between Inkatha-supporting hostel-dwellers and
 'township residents'. An examination of ANC discourse however suggests that the unitary
 category of 'township residents' was open to contestation.
 ANC discourse:
 The ANC's position on urbanisation was rather more complex, due to the shifting political
 terrain since February 1990. Immediately after the ANC's unbanning, the Alexandra branch
 16
of the movement allied itself closely with ACO. ACO executive members were all members
 of the ANC; and ANC structures, or 'sub-branches', coincided with the boundaries of ACO
 area committees. Potential conflicts of interest that might have been caused by overlapping
 structures were initially contained by a clear division of labour between the two
 organisations. Local civic issues such as housing were dealt with by ACO, while the ANC
 mobilised around national political concerns. The ANC did not play a significant role in local
 debates on urbanisation, but officially supported ACO's position.
 There was, however, a growing sense of disquiet among ANC leaders, which may have been
 encouraged by strongly vocalised discontent among returned exiles, about the 'squatter
 problem'. Many returned exiles were absent during the struggles of the 1980s, during which
 the Freedom Charter's statement of the right to housing acquired a new significance as shack-
 dwellers were mobilised in opposition to the state. On returning, exiles found their yards full
 of shacks which they associated with crime and squalor. They blamed ACO for bringing
 outsiders into Alexandra. "We are the bona fide people of the township...We have permits",
 exclaimed one returned ANC exile, ironically echoing the words of the Alexandra Liaison
 Committee - the precursor to the Black Local Authority which generated such antipathy and
 was the catalyst for anti-state opposition in the 1980s.81 The problem was particularly acute
 among those who had come from property-owning families and who planned to buy back the
 yards which had earlier been expropriated by the state. A returnee in this position
 complained, "ACO has brought in shack-dwellers and told them not to pay
 rent...[standholders] are complaining throughout Alex".
 This attitude towards shack-dwellers conflicted with ANC policy, and was officially
 discouraged. It was particularly significant that at a funeral of a member of the ANC's armed
 wing, Umkhonto we Sizwe, in July 1991, one of the coffin-bearers was the chairperson of
 a shack committee in a large free-standing shack settlement. At an ANC general meeting
 in February 1992, however, the tensions between ANC and ACO finally surfaced. In this
 meeting, ANC activists described their organisation as the historically legitimate
 representative of the people of Alexandra, and ACO merely as an interest group representing
 squatters.82 The ANC saw itself as a mediator between ACO and property owners, who
 were represented by the Alexandra Land and Property Owners' Association (ALPOA), and
 challenged ACO's right to represent the people of Alexandra in local government
 negotiations. This dramatic political shift was tied to changing ideas about the local role of
 the ANC. During a discussion I had with Obed Bapela, secretary of the ANC branch, around
 this time, he spoke about the need for the ANC to get involved in local welfare and
 development. The ANC seemed perilously close to stepping on ACO toes. ACO leaders
 expressed fears that individuals in the ANC leadership were allying with it rival, ACA,
 which was anti-squatter and mobilised mainly among residents of formal housing. Personal
 accusations and personality clashes further contributed towards the deterioration in the
 relationship between ACO and ANC, and obstructed mutual debate about their respective
 roles.
 By 1992, the ANC's position on urbanisation was still unclear. On the one hand, there was
 a tendency to view shack-dwellers as outsiders who posed a threat to Alexandrians - this
 narrative of urbanisation emphasised the temporary nature of migrant shack-dwellers'
 residence in Alexandra and legitimised the claims of an urbanised middle class to space.
 These ideas are largely convergent with the 'liberal modernism' espoused by the ANC before
 17
its banning and by liberals within and outside the state. On the other hand, the ANC was still
 trying to mobilise shack-dwellers. On the same day as its February 1992 general meeting,
 the chairperson of the ANC branch addressed a meeting in 'S'tshwetla', the biggest shack
 settlement in the township, and emphasised their unity with all South Africans in the struggle
 for liberation. He was accompanied by an executive member of ACO. In other words, the
 ANC appeared to be vacillating between two ideologically conflicting readings of the spatial
 ordering of Alexandra. The first was of a unitary township which included both shack-
 dwellers and residents of formal housing. This concurred with ACO's position, and made
 political sense if ANC and ACO were in alliance rather than competing for power. The
 second reading described the 'rear Alexandra - ie the Alexandra of the 1950s - as
 unchanging over time but as being invaded by outsiders who, by moving uninvited into yards
 and refusing to pay rent, were upsetting the 'natural' spatial order of the township, embodied
 in the concept of property rights and the paternalistic authority of standholders over their
 tenants. This emphasis on temporal continuity legitimized the claim of the long-established
 ANC to represent the people, thereby contesting the hegemony of ACO. It also presented as
 natural a spatial ordering representing the values and interests of a property-owning middle
 class, a constituency perhaps neglected by ACO.
 This reading clearly demonstrates that there was competition for legitimacy and power
 between ANC and ACO, which took the form of appealing to, and empowering, different
 constituencies. ACO's objectives of creating a unitary organised township thus coexisted with
 the ideas and concerns of a standholding elite that was never destroyed by 'high racial
 modernism'. In areas like East Bank, where the crowded yards of Alexandra were replaced
 in the 1980s by individual houses, ACO has been usurped by newer organisations.
 Elsewhere, many yard committees are still functioning. But where former standholders still
 inhabit these yards, the continuity in patronage networks has led to ambiguous and sometimes
 contradictory ideas about social and political relations. The result today is that localised
 representations of space and social relations in Alexandra yards are highly variable, and are
 often at variance with those put forward by either ACO or ANC. As we shall see, the
 attempts by these organisations to impose a hegemony has been even less effective in free-
 standing shack settlements, where the civic's yard-based model of organisation has proved
 inappropriate.
 Tensions between ANC and ACO continued to characterise Alexandra politics throughout the
 period leading up to the 1994 national election, and they were heightened during the
 preparations for local government elections in 1995. While both ANC and ACO candidates
 were selected to stand on the ANC's electoral lists, there was much disagreement about the
 respective weighting given to the two organisations, and about the role of ACA, ALPOA and
 the recently formed East Bank Civic Association (EBCA): all of which were ANC-aligned,
 had significant property-owning constituencies, and rivalled ACO. ACO's affiliation to
 SANCO, which was a member of the ANC's national 'three plus one' alliance,83 however
 lent weight to its claim for representation on the ANC's list. A solution was reached, and
 today ANC and ACO activists jointly represent the township in local government. The two
 organisations have also converged on the issue of informal housing on the far East Bank.
 After protests by the EBCA and media appearances by ACA's Mike Beea, ACO and ANC
 councillors concurred that the squatters should be moved.
 18
Does this represent a policy shift on the part of ACO? In part, the influence of the civic has
 been weakened by the emergence of other organisations close to the ANC, and it would not
 be surprising if ACO was forced to concede on the issue of informal housing. But we should
 also remember that during the early 1990s, even as ACO sought to organise informal
 settlements, it was concerned at the potential for squatters to organise themselves
 autonomously, and thus to undermine the civic's hegemony. The East Bank squatters have
 organised themselves independently of ACO, and thus pose a challenge not only to a spatial
 order based on property rights, but also to one based on a unitary model of community
 organisation. For both the ANC and ACO, these squatters represent disorder which
 undermines local government's ability to plan a new and equitable urban social order. Both
 organisations employ modernist narratives of urbanisation that view 'uncontrolled' informal
 settlements as less urban and less modern than the broader urban community, and therefore
 as a threat to the goal of modernising Alexandra, whether this refers to the modernity of a
 property-owning middle class or of an urban proletariat. For today's councillors who face
 the challenge of planning a new urban environment, these ideas are reinforced by academic
 literature which is based on a similarly modernist understanding of urbanisation. Here,
 informal settlements are depicted as marginalised from the broader urban community and
 characterised by a 'migrant culture' that legitimises the emergence of a violent form of
 patronage, thus posing serious problems for democratic community-based development.84
 As we have seen earlier, to explain the problems of informal settlements in terms of their
 presumed liminality between rural and urban in a unilinear process of urbanisation is
 misleading. Elsewhere, I have criticised the literature on informal housing for its assumptions
 about both marginal ity and culture, which I have argued overemphasise the separation of
 informal settlements from the broader urban context85. This is especially true for
 Alexandra, where the social and political dynamics of formal and informal housing are
 closely intertwined. Consequently, the modernist discourses which ACO and ANC employ,
 and on which they are basing current policies, do not necessarily reflect people's lived
 experiences. In order to shed light on this, I look at ANC and civic activists1 attempts to
 organise an informal settlement, known locally as 'Mozambique'.86
 Informal Housing in a 'Postmodern' Urban Environment: The Case of
 'Mozambique'
 Mozambique is a densely populated shack settlement named after the large concentration of
 Mozambicans living there. It first came into being when a construction company working in
 Alexandra moved its Mozambican workforce onto a piece of empty land in 1990, and later
 retrenched most of them. Since then, many other people have moved to the settlement:
 mainly Mozambicans and South Africans from the Giyani and Bushbuckridge areas of the
 northern Transvaal. While most of the shack-dwellers still have families in the rural areas,
 many of them are long-term migrants who also have friends and relatives in urban centres
 like Soweto. Consequently, their previous places of residence include urban hostels as well
 as formal and informal housing.
 Prior to 1991, ACO was unsuccessful in bringing Mozambique into its township-wide shack-
 dwellers' coordinating committee. But with the outbreak of violence in March of that year
 19
there were several attempts by civic and ANC activists from the local 'Lusaka' area
 committee to establish a shack committee in the settlement. The two main players here were
 David Shabangu, who held office in both ACO and the ANC, and Philemon Mthembu, a
 former FRELIMO guerilla and senior ANC marshal87 who acted as right-hand man to
 Shabangu. Shabangu was a skilled organiser who had established a powerful support base in
 the area through forming patronage relationships with influential youth activists, or
 'comrades', who in turn performed a coercive policing role for him. He used this to try to
 carve out an autonomous territory for himself, and while he used ACO and ANC
 membership to legitimise his authority, he manipulated differences between the two
 organisations to enhance his personal power. Shabangu's ideas about political leadership were
 quite at variance with those of his organisations. As a sangoma, and claiming to be a member
 of the King's council in Swaziland88, he emphasised the importance of tradition and strong
 individual leadership rather than democratic participation. Eventually this was to lead to his
 expulsion from Alexandra, after he established an ethnically-based vigilante group, extorted
 money for weapons, and was implicated in the murder of a Zulu-speaking couple who were
 ANC members. Mthembu, on the other hand, adhered to democratic principles, and tried to
 combine political education with organisation. Since he did not hold office in local structures,
 he deferred to Shabangu, while at the same time acting independently where possible.
 Shabangu, meanwhile, was dependent on Mthembu for access to self-defence units which
 were controlled by the ANC's marshal structure.
 Shabangu sought to incorporate Mozambique into his area committee, and was supported by
 Mthembu, who explained, "If it wasn't organised it would be easy for the enemy to get in
 and create its own structures". The shack-dwellers were keen to support any organisation
 which could help alleviate the terrible conditions of the settlement, and in particular which
 could provide a solution to high levels of police harassment. Shabangu and Mthembu made
 several successive attempts to establish a shack committee in the settlement. But the yard
 committee model proved inappropriate for the large amorphous shack settlement. With no
 clearly-defined boundaries, Mozambique was continually growing and, unlike in the semi-
 domestic spaces of yards, the shack-dwellers had few social ties. There was therefore little
 social base to their participation in a shack committee. This factor, combined with the
 inability of shack leaders to satisfy the material needs of the shack-dwellers, weakened the
 influence of the committees, which in each case collapsed.
 Organisation was initially facilitated by Mthembu, who encouraged a long-time 'comrade'
 and ANC member resident in the settlement to form a shack committee. When this man
 moved away and the committee collapsed, Shabangu stepped in and chose another leader,
 who was influential in migrant networks. While Mthembu was trying to create a democratic
 structure and Shabangu a more traditional style of leadership, in both cases the chair of the
 shack committee was dependent on these outside actors for what they hoped would be
 material assistance and protection. What emerged was therefore patronage-style dependency
 relationships, which combined ideas about democracy and tradition to differing degrees.
 This illustrates that vertical dependency relationships are not just internal to informal
 settlements. Far from being a causal factor underlying social, cultural and political isolation,
 these relationships are often precisely the means by which informal settlements are integrated
 into the broader urban community, albeit through personalistic criteria which render
 integration uneven and therefore undemocratic.89 Shack leaders can act as brokers between
 20
shack-dwellers and outside actors or patrons who command resources: developers who need
 popular legitimacy for their projects or political activists seeking to increase their support
 base. Informal settlements are, in other words, a resource desired by political actors whose
 own organisational machinery is inadequate to integrate shack-dwellers without establishing
 a clientelistic relationship with key individuals in these settlements. In this way, patterns of
 clientelism originating from outside informal settlements are replicated within such
 settlements rather than being generated solely by the conditions of informal housing.
 This example also challenges the idea of informal settlements as more 'rural' and 'traditional'
 than a more modern 'urban culture'. While culture in informal settlements is often
 understood to be a hybrid mixture of rural and urban influences, resulting in "distinctive sub-
 cultures and political practices": some more 'rural' and others more 'urban' in character,
 depending on people's experiences,90 this is usually understood to emanate from a
 transitional stage in the process of becoming 'urban'. 'Urban culture' however displays many
 of the same features and ambiguities evident in informal settlements. Many of the shack-
 dwellers I interviewed had lived in other types of housing, so that their experiences and
 resources are drawn from a broader environment. The same fluidity also applies to those
 living in the formal township. The leadership of Shabangu is a case in point. Although he
 had a rural homestead which he occasionally visited, he lived in formal housing and was not
 a migrant. He frequently evoked tradition to legitimise his practices, while using urban-based
 democratic political organisations to legitimise his position. As long as he was able to fulfil
 people's expectations of him as a leader, he managed to combine democratic and clientelistic
 practices without overt contradiction. While Mthembu's approach to politics could be
 characterised as more 'urban', 'modern' or 'democratic', he nevertheless worked closely with
 Shabangu; and where Mthembu was concerned with bringing the shack-dwellers into the
 ANC's ideological fold, Shabangu was more instrumental in integrating them into urban
 political structures.
 Conclusion
 In this paper, I have tried to demonstrate the continuity between the current stand-off
 between squatters and ANC-aligned property-owners in Alexandra, and past attempts by the
 township elite to construct a spatial order based on property rights and paternalism. While
 this represented opposition to apartheid, and particularly to the policies of 'high racial
 modernism', it has since the mid-1980s itself been contested by the more socialist-oriented
 ideas of ACO. All of these political players have however approached the issue of informal
 housing in terms of a modernist conception of urbanisation. All assume that squatters are
 recent arrivals from the rural areas, and thus less urban and less modern than the broader
 township community. The state thus attempted to prevent squatting through restrictions on
 urbanisation, coinciding with the interests of property-owners who perceived squatters as a
 threat. For ACO, on the other hand, civic organisation was seen as the tool that could
 transform squatters into a radical urban proletariat. In many of Alexandra's free-standing
 shack settlements, however, organisation was weakened by the attempt to apply a yard-based
 model of organisation to a very different social and spatial arrangement. The resulting
 'uncontrolled' settlements are thus perceived by both ACO and ANC as being outside 'the
 community', and therefore a problem for community-based development. My material,
 21
however, supports Mabin's assertion that informal housing today, far from being a transition
 between rural and urban, is very much part of the urban environment. While the internal
 dynamics of these settlements cannot be characterised simply as 'rural' and 'traditional', or
 as 'urban' and 'modern', they are largely a product of broader political processes which are
 similarly ambiguous. Alexandra has always confounded these modernist contrasts, and
 continues to do so.
 What does this mean for planning strategies? Firstly, any research intended as a preliminary
 to addressing the question of land invasion should explore the linkages between informal
 settlements and their surroundings. The crisis that occurred in Zevenfontein in the early
 1990s, for example, can only be understood by looking at the role of a powerful patron in
 the ANC's regional executive, who provided the opportunity for authoritarian leaders to
 assert themselves, and encouraged them to dismiss the community's lawyers.91 Secondly,
 the recognition that informal housing is a significant feature of the urban environment, and
 that it cannot be addressed solely through policies attempting to direct urbanisation processes,
 means that mechanisms of participation need to be developed to involve squatters, and
 surrounding formally-housed communities, in development initiatives that affect all of them.
 This will not be easy, but it is more realistic than the mistaken idea that we can just
 temporarily remove squatters from the urban environment while development is implemented.
 Informal housing is a product of this environment, and a refusal to recognise and thus take
 responsibility for this 'Nobody's Baby' will not make the challenge disappear. Stadler's
 metaphor of 'birds in the cornfield'92 is particularly apt in the 1990s. If local government
 carries out its ultimatum to move the East Bank squatters on May 3rd, it is highly likely that
 others will simply take their place.
 l.Alan Mabin, 'On the Problems and Prospects of Overcoming Segregation and Fragmentation in Southern
 Africa's Cities in the Postmodern Era'. In S Watson and K Gibson (eds), Postmodern Cities and Spaces.
 Blackwell 1995.
 2.Mabin is using the term postmodern as a socio-economic condition, rather than as an aesthetic or as an
 epistemology.
 3.Alan Mabin, 'Struggle for the City: Urbanisation and Political Strategies of the South African State'. Social
 Dynamics 15(1): 1-21, 1989; P Bonner, 'The Politics of Black Squatter Movements on the Rand, 1944-1952*.
 Radical History Review 46-47:89-113. 1990.
 4.Bonner, 'The Politics of Black Squatter Movements' p93.
 5.Josette Cole, Crossroads. The Politics of Reform and Repression 1976-1986. Johannesburg. Ravan Press 1987
 pl9.
 6.See Hilary Sapire, "Politics and Protest in Shack Settlements of the PWV Region, South Africa 1980-1990'
 Journal of Southern African Studies 18(3), 1992, p677.
 7.Alan Mabin, 'The Dynamics of Urbanization Since I960'. In M Swilling, R Humphries, K Shubane (eds),
 Apartheid City in Transition. Cape Town. OUP. 1991
 8.Ibid p42.
 9. Ibid p44.
 22
10.P Bonner, 'African Urbanisation on the Rand Between the 1930s and 1960s: its Social Character and
 Political Consequences'. Journal of Southern African Studies. 21(1) 1995 ppU5-129.
 Il.lbidpl25.
 12.While the term location is used to describe African urban areas that were built before the public housing
 programmes of the 1950s, Alexandra's freehold status meant that it was not strictly speaking a location either.
 It was proclaimed as a 'Native township' in 1912, but although its freehold property rights were protected under
 the 1913 Land Act, the Native Trust and Land Act of 1936 prevented any other such areas from being
 established (M Sarakinsky, 'From "Freehold Township" to "Model Township" - a Political History of Alexandra
 1905-1983'. Unpublished Honours Dissertation. University of the Witwatersrand 1984, p2). In this paper, I
 follow popular usage in referring to 'Alexandra township'.
 13.T Couzens, 'Nobody's Baby: Modikwe Dikobe and Alexandra 1942-1946'. In B Bozzoli (ed), Labour
 Townships and Protest. Johannesburg. Ravan Press. 1979.
 14.James Ferguson, 'Mobile Workers, Modernist Narratives: A Critique of the Historiography of Transition
 on the Zambian Copperbelt'. Journal of Southern African Studies 16(3-4): 385-412, 603-621.
 15.AlanMabin, 'Conflict, Continuity and Change: Locating "Properly Planned Native Townships" in the Forties
 and Fifties'. Symposium on South African Planning History 1993; Peter Wilkinson, 'A Discourse of Modernity:
 The Social and Economic Planning Council's Fifth Report on Regional and Town Planning, 1944'. Symposium
 on South African Planning History 1993; Susan Parnell, 'South African Cities: Perspectives from the Ivory
 Tower of Urban Studies'. Unpublished paper. University of the Witwatersrand Institute for Advanced Social
 Research 1995.
 16.0ren Yiftachel, "The Dark Side of Modernism: Planning as Control of an Ethnic Minority'. In Watson and
 Gibson, Postmodern Cities and Spaces.
 17.D Hindson, Pass Controls and the Urban African Proletariat. Johannesburg. Ravan Press. 1987
 18.Mabin, 'Conflict, Continuity and Change'.
 19.David Harvey, The Condition of Postmodernitv. 1991
 20. Alan Mabin and Dan Smit, 'Restructuring South Africa's Cities 1900-2000: A Prospectus (Or, A Cautionary
 Tale). Unpublished paper. University of the Witwatersrand African Studies Institute 1992; Parnell, 'South
 African Cities*.
 21.E Wilson, The Sphinx in the City. London. Virago Press. 1991; Parnell, 'South African Cities'.
 22. A D King, Urbanism. Colonialism and the World Economy. Cultural and Spatial Foundations of the World
 Urban System. London. Routledge. 1990
 23.Mabin, 'Conflict, Continuity and Change1.
 24.Maynard Swanson, "The Sanitation syndrome: Bubonic Plague and Urban Native Policy in the Cape Colony
 1900-1909'. Journal of African History 18(3) 1977.
 25.This-existed in tension with ideas about racial segregation.
 26.1 am using the terms liberal and conservative here as tendencies rather than as two distinct camps. There
 were of course many intermediate positions, which is not surprising if we consider the ambiguity of underlying
 ideas about modernisation.
 27.Belinda Bozzoli, 'From Governability to Ungovernability: Race, Class and Authority in South Africa's Black
 Cities'. Unpublished paper. University of the Witwatersrand Institute for Advanced Social Research 1996.
 23
28.The establishment of this department in 1919, with (poorly defined) planning powers, illustrates the centrality
 of die 'sanitation syndrome' in early planning policy, arid the liberal modernist idea that planning was a solution
 to social problems and reconstruction (Mabin and Smit, 'Reconstructing South Africa's Cities'; Wilkinson, 'A
 Discourse of Modernity*).
 29.PNTourikis, "The'Political Economy'of Alexandra Township: 1905-1958'pp7-18. Unpublished Honours
 Dissertation. University of the Witwatersrand. 1981; Sarakinsky, 'From "Freehold Township" to "Model
 Township"' p5.
 3O.The Star 11/1/43, cited in Sarakinsky, 'From "Freehold Township"' p3
 31.Sarakinsky, 'From "Freehold Township"' p4.
 32.The Star 11/1/43, op cit.
 33.Tom Lodge, Black Politics in South Africa Since 1945. Johannesburg. Ravan Press 1983. pl64.
 34.Sarakinsky, 'From "Freehold Township"' pp5-6.
 35.Mabin, Conflict, Continuity and Change'.
 36.Wilkinson, 'A Discourse of Modernity'.
 37.These were: distance from white towns; adjoining the locations of neighbouring towns; industrial and open
 buffer zones; hinterland for expansion; easy distance for rail transport; own road connecting with the city;
 distance from main roads (Mabin, 'Conflict, Continuity and Change').
 38.WWM Eislen 1950, cited in Mabin, 'Conflict, Continuity and Change' p331.
 39.Men or women over the age of 21.
 40.The role of die communist Bram Fischer, who was a member of the 1942 committee, was however rather
 different, since he supported the formation of the Alexandra Tenants' Association.
 41.Tourikis, "The 'Political Economy of Alexandra' p43.
 42.Bozzoli, 'From Governability to Ungovernability' p9.
 43.J Lucas, 'Space, Domesticity and People's Power: Civic Organisation in Alexandra in the 1990s'. African
 Studies 54(1): 89-113 1995; J Lucas, 'Space, Society and Culture: Housing and Local-Level Politics in a Section
 of Alexandra Township, 1991-1992'. Unpublished MA dissertation. University of the Witwatersrand 1995 pp95-
 100.
 44,Tourikis, 'The 'Political Economy' of Alexandra' pi 16.
 45.Tourikis, 'The 'Political Economy of Alexandra' pp36,108-112 ; Sarakinsky, 'From "Freehold Township"'
 p20.
 46.These took place in 1940, 1942, 1943 and 1944.
 47.E Roux, Time Longer Than Rope: The Black Man's Struggle For Freedom in South Africa. University of
 Wisconsin Press 1978 p319; Lodge, 'Black Politics' pl63.
 48.Sarakinsky, 'From "Freehold Township1" p!5.
 49.AJf Stadler, 'Birds in the Cornfield: Squatter Movements in Johannesburg 1944-7'. In B Bozzoli (ed),
 Township. Labour and Protest. Johannesburg. Ravan Press 1979; P Bonner, 'The Politics of Black Squatter
 Movements on the Rand, 1944-1952'. Radical History Review 46-47: 89-113.
 24
5O.Sarakinsky, 'From "Freehold Township"' p4.
 51.Tourikis, 'The 'Political Economy' of Alexandra' p58.
 52.Tourikis, 'The 'Political Economy' of Alexandra' p34.
 53.The impressions of Drum editor Basil Davidson were that many had lived in Alexandra all their lives (Ibid).
 54.Bozzoli, 'From Governability to Ungovernability'.
 55.Sarakinsky, 'From "Freehold Township"" pp26-28.
 56.C E Carter, 'Comrades and Community: Politics and the Construction of Hegemony in Alexandra Township,
 South Africa, 1984-1987'. Unpublished Phd thesis. Oxford University 1991. p50.
 57.The Native (Urban Areas) Consolidated Act, Native Services Levy Act, Urban Bantu Councils Act, Housing
 Act, Native Resettlement Act and the Group Areas Act are among the legislation that now applied to Alexandra.
 58.Sarakinsky, 'From "Freehold Township"', p48.
 59.Bozzoli, 'From Governability to Ungovernability' pl4.
 6O.Sarakinsky, 'From "Freehold Township"', p39.
 61.Bozzoli, 'From Governability to Ungovernability' p l9 .
 62.Sarakinsky, 'From "Freehold Township"' p49.
 63.Bozzoli, 'From Governability to Ungovernability' p20.
 64.Sarakinsky, 'From "Freehold Township"', p63.
 65.Parnell, 'South African Cities' plO.
 66.Mongane Wally Serote, To Every Birth Its Blood. Johannesburg. Ravan Press 1981 p22.
 67.Ibid p36.
 68.1bid p21.
 69.K Jochelson, 'Reform, Repression and Resistance in South Africa. A Case Study of Alexandra Township
 1979-1989'. Journal of Southern African Studies 16(1): 1-32.
 7O.Sarakinsky, 'From "Freehold Township"', p72.
 71Jochelson, 'Reform, Repression and Resistance'.
 72.Hindson, 'Pass Controls and the Urban African Proletariat'; on Alexandra, Couzens, 'Nobody's Baby'.
 73.This did coexist with the 'disorder' of ungovernability, but I would suggest that this emerged from the
 destructive aspect of this attempt to overturn the old social order and replace it with a new hegemony.
 74.The ideological orientation of the 'charterist' camp of the UDF was guided by the Freedom Charter, while
 'workerists' were more concerned to prioritise working-class interests. The charterist camp claimed a more
 direct relationship with the exiled ANC, although this was not necessarily always the case, and viewed the
 workerists with some antipathy. In actual political developments, 'charterist' and 'workerist' tended to operate
 as categories rather than identifiable groups, so that there was often disagreement over who fell into which
 camp. In Alexandra, Mayekiso viewed ACA's middle-class executive as petty bourgeois, and later accused ACA
 chairperson Mike Beea of being 'populist'. Mayekiso described himself as a socialist, and was scathing about
 25
workerists. However, his background in the Federation of South African Trade Unions (FOSATU), his interest
 in the formation of a workers' party, and his stated rejection of the Freedom Charter as a 'capitalist document'
 meant that many UDF activists doubted his charterist credentials (Carter, 'Comrades and Community'
 pp201,206,209).
 75. For example, Moses Mayekiso, his brother Mzwanele, and Richard Mdakane, all of whom were on the AAC
 executive.
 76.Jochelson, 'Reform, Repression and Resistance' pl8.
 77.The charges were dropped in 1989.
 78.Interview widi Richard Mdakane, November 1991.
 79.Ibid.
 80.Before violence erupted in 1991, ACO claims that Inkatha members were involved in ACO structures, and
 that ACO had committees in all three hostels.
 81,Sarakinsky quotes Leepile Taunyane of Sam Buti's Liaison Committee in 1981 describing shack-dwellers
 as "not bona fides of the township. They (were) people who had just streamed to Alexandra from nowhere"
 (Sarakinsky, 'From "Freehold Township"' p72).
 82.This highlights the point made by Mdakane: that residents of formal housing perceived shack-dwellers as
 the back-bone of ACO support, largely because they also believed that ACO had brought shack-dwellers to the
 township.
 83.ANC, SACP, COSATU and SANCO.
 84.Caroline White, 'Makhulu Padroni? Patron-Clientelism in Shack Areas and Some Italian Lessons for South
 Africa'. Unpublished paper. Association for Anthropology in South Africa 1993.
 85.Lucas, 'Space, Society and Culture' ppl83-226.
 86.1 use Pseudonyms throughout this case study.
 87.The ANC marshal structure exists in parallel with other branch structures. In Alexandra during 1991 and
 1992, it played a central role in the formation of self-defence units.
 88.This proved to be untrue.
 89.See White, 'Makliulu Padroni?'. The integrative properties of clientelism have been noted elsewhere (Eg R
 Lemarchand, 'Political Clientelism and Ethnicity in Tropical Africa'. In S Schmidt, J Scott, C Lande and L
 Guasti (eds), Friends. Followers and Factions. Berkeley. University of California Press. 1977).
 9O.Sapire, 'Politics and Protest' p681.
 91.See White, 'Makhulu Padroni?'.
 92.Stadler, 'Birds in the Cornfield'.
 26