uTshani BuyaKhuluma-The Grass Speaks: The political space and capacity of the South African Homeless People's Federation

Abstract

The point of departure for this article is the contemporary tendency towards localisation of politics in the context of neo-liberal globalisation. Mediated through institutional reforms, political discourses and localised struggles, this localisation of politics produce new and transformed local political spaces. The purpose of the article is to examine the capacity of popular movements to use and transform such political spaces within the South African housing sector. This analysis is done through a combination of conceptual examination of political space and actor capacity and a concrete case study of the political strategies and capacities of The South African Homeless People’s Federation. The article argues that the Federation has utilised political relations at different scales to mobilise resources such as land and subsidies for housing for its members. It has also influenced the formulation of housing policies through its discourses and practical experiences with people-driven housing processes. In consequence the Federation’s ability to function as a civil/political movement has granted them a certain capacity to participate in the complicated process of turning de jure rights to adequate shelter into de facto rights for the urban poor as citizens of a democratic South Africa.

Full text

uTshani BuyaKhuluma––The Grass Speaks:
the political space and capacity of the South African
Homeless PeopleÕs Federation
Marianne Millstein
a
, Sophie Oldfield
b
, Kristian Stokke
a,*
a
Department of Sociology and Human Geography, University of Oslo, P.O. Box 1096 Blindern, 0317 Oslo, Norway
b
Department of Environmental and Geographical Science, University of Cape Town, Private Bag, Rondebosch 7701, South Africa
Received 13 February 2002; received in revised form 10 February 2003
Abstract
The point of departure for this article is the contemporary tendency towards localisation of politics in the context of neo-liberal
globalisation. Mediated through institutional reforms, political discourses and localised struggles, this localisation of politics
produce new and transformed local political spaces. The purpose of the article is to examine the capacity of popular movements to
use and transform such political spaces within the South African housing sector. This analysis is done through a combination of
conceptual examination of political space and actor capacity and a concrete case study of the political strategies and capacities of The
South African Homeless PeopleÕs Federation. The article argues that the Federation has utilised political relations at different scales
to mobilise resources such as land and subsidies for housing for its members. It has also influenced the formulation of housing
policies through its discourses and practical experiences with people-driven housing processes. In consequence the FederationÕs
ability to function as a civil/political movement has granted them a certain capacity to participate in the complicated process of
turning de jure rights to adequate shelter into de facto rights for the urban poor as citizens of a democratic South Africa.
Ó2003 Elsevier Ltd. All rights reserved.
Keywords: South Africa; Democratisation; Local politics; Social Movements; Housing; South African Homeless PeopleÕs Federation
1. Localisation of politics in the age of neo-liberal
globalisation
The contemporary world is characterised by both
globalisation and localisation, i.e. simultaneous tenden-
cies towards global economic and cultural integration
and local embeddedness (Castells, 1996, 1997; Cox,
1997; Held, 2000; Held et al., 1999). This global–local
tension has generated a growing literature on local
economic embeddedness and local governance in the
context of neo-liberal globalisation, as well as discus-
sions about the production of scale and inter-scalar re-
lations (Collinge, 1999; Cox, 1998; MacLeod and
Goodwin, 1999; Marston, 2000; Peck, 2002; Sheppard,
2002). A less examined trend in conjunction with glob-
alisation is localisation of politics. Local-scale politics
are mediated through institutional reforms towards de-
centralisation, local democratisation and good gover-
nance, discourses on local participation and civil society,
as well as localised struggles over local, national and
global issues. Simultaneously, recent development the-
ory and practice have been marked by various attempts
to compensate for the negative effects of globalisation by
turning Ôthe localÕinto a prime site of socio-economic
development and political democratisation (Mohan and
Stokke, 2000; Nederveen Pieterse, 2001). The normative
assumption is that mutually enabling relations between
decentralised state institutions, local businesses and civil
society will generate economic growth, poverty allevia-
tion and good governance. These tendencies towards
localisation of politics and localism in development call
for critical analyses of the making and transformation of
local political spaces and the strategies and capacities of
different actors (e.g. local elites and political parties,
trade unions and popular movements, local state insti-
tutions and governments) to utilise rights, institutional
*
Corresponding author.
E-mail addresses: marianne.millstein@admin.uio.no (M. Millstein),
oldfield@enviro.uct.ac.za (S. Oldfield), kristian.stokke@sgeo.uio.no
(K. Stokke).
0016-7185/$ - see front matter Ó2003 Elsevier Ltd. All rights reserved.
doi:10.1016/S0016-7185(03)00046-0
Geoforum 34 (2003) 457–468
www.elsevier.com/locate/geoforum

channels and discourses to promote their instrumental
and democratic aims. The purpose of this article is to
examine the ways in which popular movements may use
and transform political channels and discourses within
the South African housing sector. This is done through a
combination of conceptual examination of political
space and actor capacity and a concrete case study of the
political strategies and capacities of The South African
Homeless PeopleÕs Federation (SAHPF). The aim is to
analyse their capacity to make use of local and non-local
political spaces and promote the social and political
interests of the communities they seek to represent.
2. Local politics and democratisation in South Africa
In South Africa, institutional reforms towards local
administration and government have been on the po-
litical agenda for at least two decades. The National
Party (NP) initiated various reforms, including the
construction and operation of local authorities, towards
the end of the apartheid period in an attempt to over-
come economic and political crises and to split, depo-
liticise and contain township struggles (Coovardia, 1991;
Parnell, 1992). Simultaneously, local civic organisations
and trade unions sought to make cities and townships
ungovernable and ultimately to make a democratic
transition unavoidable (Bozzoli, 1996; Parnell, 1992;
Seekings, 2000). Politicisation of local single issues and
specific interests, e.g. land rights and housing, were
central to the ideological struggle for liberation and
democratisation. Thus, local politics, dialectically
related to national politics, was an integral part of
the anti-apartheid struggle in the 1980s and early 1990s
(Maharaj, 1996; Mayekiso, 1996).
South AfricaÕs democratic transition in the early
1990s produced a radical constitutional reform that
granted extensive formal rights for all citizens and nu-
merous institutional reforms, including national elec-
tions from 1994, to ensure their actual implementation.
These changes at the national scale have been followed
by local elections, extensive local government reforms
and political discourses endorsing good governance and
popular participation (Atkinson and Reitzes, 1998;
Cameron, 1999). These transformations mean that his-
torically well-organised political and civic associations
have been placed in a situation with radically trans-
formed and widened local, regional and national polit-
ical spaces (Neocosmos, 1998; Smit, 2001). This
combination of a vibrant civil society and a conducive
political environment should, it seems, provide an ideal
case for substantial democratisation, i.e. a situation
where ordinary citizens have both the possibility and the
capacity to make use of democratic rights, institutions
and discourses to address their instrumental and demo-
cratic aims (T€
oornquist, 2002). Unfortunately, in prac-
tice, the post-apartheid political and socio-economic
conditions have proven to be more complex and con-
tradictory (Bond, 2000a,b).
One major obstacle for substantial democratisation in
South Africa remains the persistent and increasing social
inequalities and absolute poverty. While the immediate
post-1994 period was characterised by a remarkable
political liberalisation, the ensuing post-apartheid pe-
riod has been marked by a transition in macro-economic
policy with important bearings on the fulfilment of so-
cio-economic rights and, thereby, the question of sub-
stantial democratisation. The early post-apartheid
period was characterised by the state-led Reconstruction
and Development Programme (RDP) designed and
concomitant with other restructuring processes to rectify
systematic socio-economic differentiation and discrimi-
nation. The macro-economic context on which the RDP
built was, however, constrained and circumscribed by
the structural imperatives of the South African domestic
and global economy. Thus, RDP state-led reconstruc-
tion and transformation battled with and, eventually,
gave way to neo-liberal policy in the governmentÕs pol-
icy for Growth, Employment and Redistribution
(GEAR) (Marais, 2001). Current macro-economic pol-
icies, while designed to enhance economic competitive-
ness in the long run, have perpetuated and deepened
absolute and relative poverty in the short run (Adel-
zadeh, 1996; Nattrass and Seekings, 2001). Although
South African citizens have been granted extensive de
jure socio-economic rights, the translation of these
rights into de facto socio-economic empowerment has
proven to be extremely complicated and persists as a
fundamental problem of substantial democratisation in
South Africa.
Post-apartheid political participation has also proved
complicated in practice. While democratic elections have
placed the tripartite alliance from the anti-apartheid
struggle––the African National Congress (ANC), the
Congress of South African Trade Unions (COSATU)
and the South African Communist Party (SACP)––in a
hegemonic political position at the national level, the
actual political participation of popular forces is de-
batable. The influence and power of organs of civil
society such as civic associations and trade unions,
instrumental in the struggle against apartheid, have
generally been on the decline, not the least as a result of
co-optation of civic leaders, a depoliticisation and nor-
malisation of politics and the dominance of technocratic
approaches to development (Adler, 2000; Adler and
Steinberg, 2000; Smit, 2001). Although these organisa-
tions continue to be active in cities and towns across
South Africa, their activities are uneven geographically
and tend to resonate on the local level in specific area-
based issues (Habib and Kotz
ee, 2002; Huchzermeyer,
2001). The translation of local interests and demands
into national policy and ideological struggles has been
458 M. Millstein et al. / Geoforum 34 (2003) 457–468

caught up in the multifaceted and fragmented ways in
which policy and finance for services and basic infra-
structure are directed from the national state, financed
through provincial bodies, and implemented by munic-
ipalities. Furthermore, the uneven capacity and power
of state institutions, organs of civil society, and indi-
viduals makes local power sharing arrangements prob-
lematic in practice (Oldfield, 2000a,b, 2002). Although
South Africa has made substantial progress towards
formal democratisation at both the national, provincial
and municipal levels, the experiences of ÔeverydayÕSouth
Africans and the diverse movements that represent them
are mired in the complex ways in which the unequal
legacies of the apartheid past are reinvented in the post-
apartheid present. The majorityÕs transformation from
apartheid subjects to democratic citizens remain bogged
down in the fragmentation of political spaces and indi-
vidual and community-based groupsÕuneven capacities
to utilise and transform democratic rights and institu-
tions (Reitzes, 1998).
These challenges of substantial democratisation are
especially visible in the case of social rights such as
employment and shelter. Implementation of these rights
for the urban poor remains a fundamental challenge of
substantial democratisation in South Africa. The con-
stitution guarantees a right to adequate shelter for all
citizens, but it remains a daunting task to translate this
into actual houses for marginalised groups. In this sit-
uation, new popular movements have emerged around
issues of housing provisioning (e.g. the SAHPF) and
municipal practices of evictions from homes and dis-
connections of municipal services (e.g. Western Cape
Anti-Eviction Campaign and Soweto Electricity Crisis
Committee) (Millstein, 2001; Oldfield and Stokke,
2002). This raises conceptual and empirical questions
about the political capacity of these new housing
movements and about the nature of political spaces
within the post-apartheid housing sector.
3. Conceptualising political space and movement capacity
Recent studies of social movements have drawn
attention to the importance of political opportunity
structures, mobilising structures and cultural framing
for the emergence and success of collective actors (della
Porta and Diani, 1999). These terms refer to political,
social and cultural conditions for collective action, but
also draw attention to the collective actorsÕability to
comprehend, utilise and transform such conditions. One
way of specifying the relevant contextual conditions––
beyond ambiguous notions of conducive or hostile po-
litical environments––is through the concept of political
space suggested by Webster and Engberg-Pedersen
(2002). This term, which refers to a political terrain that
offers uneven possibilities for different political actors
and strategies, contains two main mechanisms for po-
litical influence: institutional channels and political dis-
courses.
1
3.1. Institutional channels
The first dimension concerns the cluster of constitu-
tional rights and the institutions that uphold, promote
or prevent the actual implementation of such rights.
Equally important are the formal and informal proce-
dures for affecting policy formulation and implementa-
tion. In the case of social movements, this refers to what
is commonly described as the political opportunity
structures for collective action (Tarrow, 1994). Such
opportunity structures are usually taken to include ‘‘the
stability or instability of that broad set of elite align-
ments that typically undergird a polity; the presence or
absence of elite allies; the stateÕs capacity and propensity
for repression’’ (McAdam et al., 1996, p. 27). While
rights and institutions provide a formal framework for
participation, such political channels may be decisive for
actual access to and transformation of rights and insti-
tutions (Holland and Mohan, 2000; McEwan, 2000).
3.2. Political discourse
The second dimension in Webster and Engberg–Pe-
dersenÕs conceptualisation regards the discursive fram-
ing of rights and responsibilities, institutions and
popular actors, political injustices and goals. This means
that there is a dynamic and competitive process of
strategic framing in a variety of arenas, within and be-
tween the political sphere and civil society (della Porta
and Diani, 1999; Johnston and Klandermans, 1995:
Melucci, 1996). At the most general level it can be ob-
served that we are now, in the context of neo-liberal
globalisation, seeing a powerful global development
discourse that emphasises various institutional reforms
in favour of democratisation, human rights, decentrali-
sation, good governance and civil society. Such dis-
courses and their institutional manifestations, define
political spaces for various individual and collective
actors who claim to be the legitimate expressions of
these good causes and Ôthe peopleÕ. The vital question
then regards the capacity of diverse actors to claim po-
sitions as representatives or to challenge competing
claims, and thereby make use of and possibly transform
political discourses.
Shifting the focus from political space to actors, it can
be observed that the last two decades have seen a new
1
A third dimension is identified as social and political practices
which may be a basis for influencing policy and programme agendas,
decision-making and implementation. We will examine this as a
question of movement capacity (see below).
M. Millstein et al. / Geoforum 34 (2003) 457–468 459

interest among development researchers in questions of
political capacity. One key contribution to this shift has
been analyses of the relative autonomy and capacity of
developmental states to formulate goals and implement
policies vis-
aa-vis the economy. Migdal et al. (1994)
correctly observe that this approach has yielded an
overemphasis on state institutions at the expense of so-
cial forces and advocate a relational approach to the
autonomy and capacity of actors within both state and
society (Oldfield, 2000a,b). Unfortunately, they provide
few analytical guidelines beyond this general observa-
tion. Some useful pointers may, however, be derived
from recent studies of movement politics. Based on
observations of popular movements in Indonesia and
Kerala, T€
oornquist (1999, 2000, 2002) proposes three
questions that he sees as especially central to the analysis
of movement political capacity. These regard (a) where
in the political terrain the actors choose to work; (b)
what issues and interests they promote and politicise;
and (c) how people are mobilised into political move-
ments and the political sphere.
3.2.1. Location in political terrain
T€
oornquist (2002) presents a schematic map of the
political terrain for movement politics. This approach
revolves around three major ways in which societal ac-
tivities are organised at different scales: state and local
government;self-government units (e.g. neighbourhood
associations, co-operatives, ethnic and religious com-
munities, clans, and families), and; business units. Pop-
ular movements may be active within a relatively
autonomous public space between these spheres. In this
public space, actors form three main kinds of associa-
tions: political societies (e.g. political parties, pressure
groups, lobbying groups), which seek influence within
the political arena; civil societies, which organise in re-
gard to either business units (e.g. trade unions and
peasant organisations) or to self-government units (e.g.
religious movements that relate to various churches, or
womenÕs organisations against domestic violence), and;
civil-political societies that combine or link the activities
of political and civil societies (e.g. labour movements
with a political party in addition to popular unions,
women groups, co-operatives etc.; human-rights groups
that both support victims of violence and seek to influ-
ence state policies, and; peasant movements that mobi-
lise against landlords but also campaign for public land
reforms). T€
oornquistÕs argument is that reflexivity and
strategic decisions regarding where and how to be active
in this political terrain may provide movements with
vital political capacity.
3.2.2. Politicisation of issues and interests
The second dimension in T€
oornquistÕs framework re-
gards the content of movement politics, i.e. the issues
and interests that actors choose to bring up to be in-
cluded into politics. More specifically, T€
oornquist (2002)
argues that the politicisation of issues and interests may
be analysed in terms of whether it is based on single
issues and specific interests; ideologies and collective in-
terests; or moral and spiritual values and communal
loyalties. Social movement scholars have emphasised
that collective actors play a vital role in constructing and
communicating identities, grievances and political al-
ternatives. This cultural framing is obviously located
within a discursive field with a dynamic relationship
between movement discourses and the populations they
intend to mobilise and between movement discourses
and those of political authorities (Alvarez et al., 1998).
In this sense, successful construction of issues and in-
terests constitute both a precondition and an outcome of
collective action.
3.2.3. Political inclusion of people
The third dimension regards the mobilisation of
people into politics. Here T€
oornquist follows Mouzelis
(1986) in his distinction between integration on the basis
of broad popular movements generated by comprehen-
sive economic development, and elitist incorporation of
less solid organisations into comparatively advanced
politics in economically late-developing societies. In the
latter case, Mouzelis makes a further distinction be-
tween clientelism and populism. Clientelism, on the one
hand, refers to networks of patrons at different levels
with capacity to deliver some concessions in return for
services and votes. Populism, on the other hand, gen-
erally goes with charismatic leaders who are able to
express popular feelings. Political inclusion of people
also refers to the mobilising structures of collective
movements, i.e. the organisational form within a
movement and to networks and institutions in society
that may serve as arenas for collective mobilisation.
Such social infrastructure can facilitate communication,
co-ordination and solidarity prior to and during col-
lective actions. This means that a movementÕs ability to
organise civil society and ensure a degree of political
inclusion may be an important source of movement
strength and political influence.
4. The post-apartheid political space for housing struggles
South AfricaÕs democratic transition, with negotiated
compromises and social compacts as key features, has
radically transformed the political space for housing
struggles (Lodge, 1999; Marais, 2001). This topic is too
complex to provide any kind of comprehensive exam-
ination here. However, a few remarks about key char-
acteristics of relevant political discourses, institutional
architecture and political opportunity structures are in
order as a precursor to the concrete analysis of the
SAHPF.
460 M. Millstein et al. / Geoforum 34 (2003) 457–468

Contemporary political discourses within the housing
sector must be understood against the background of
the National Housing Forum, which was established in
the transition period as an arena for negotiations over
future housing rights and strategies to resolve the crises
of inadequate and segregated housing. The Forum,
which included multiple interests and stakeholders, was
marked by a polarisation between the interests of eco-
nomic elites in the housing sector, expressed through the
Urban Foundation, and the interests expressed by ANC,
the South African National Civics Organisation (SAN-
CO) and progressive non-governmental organisations
(NGOs) (Bond, 2000a,b; Huchzermeyer, 2001; Lalloo,
1999; Wilkinson, 1998). Whereas the coalition of ANC,
SANCO and NGOs emphasised the stateÕs role in re-
ducing the apartheid legacy of social and racial in-
equalities, the Urban Foundation was unwilling to grant
the state a prominent role unless this was deemed nec-
essary to ensure economic development. The Urban
Foundation held a clear majority within the Forum and
possessed a superior organisational capacity (Lalloo,
1999). Consequently, the Forum came to favour market
rather than state-led housing policies. As housing be-
came a politicised issue during the election campaign in
1994, the political discourse on housing strategy turned
towards short-term goals of speedy delivery of large
numbers of housing units (Bond, 2000a). In the end,
these constellations yielded a hegemonic discourse on
housing strategy that was product- and delivery-ori-
ented rather than process- and participation-centred,
and where self-help and market mechanisms were
granted a central role (Hart and Parnell, 1999). The key
mode of state intervention in the housing process be-
came housing subsidies to identified target groups, for
houses that would be delivered through a partnership
between state planning agencies and private developers.
These post-apartheid discourses and development
mechanisms have to a large extent framed housing
policies and strategies in the post-apartheid period (Hart
and Parnell, 1999; Huchzermeyer, 2001; Oldfield, 2000b;
Parnell, 1992). Shortly after assuming power, the ANC
expressed its own vision for the housing sector through
The White Paper on Housing (Republic of South Africa,
1995). The overall aim was to: ‘‘Establish a sustainable
housing process which will eventually enable all South
AfricaÕs People to secure housing with secure tenure,
within a safe and healthy environment and viable com-
munities in a manner that will make a positive contri-
bution to a non-racial, non-sexist, democratic and
integrated society, within the shortest possible time
frame’’ (Republic of South Africa, 1995, p. 20). More
specific targets were to provide 1 million houses within
five years and allocate 5% of the annual national budget
to the housing sector. The chosen provisioning mecha-
nisms were largely similar to those prescribed by the
National Housing Forum. An alternative discourse on
peopleÕs participation in housing development was
subordinated to this hegemonic discourseÕs emphasis on
top-down technocratic planning and commercial hous-
ing contractors. Thus, Kotz
ee observes that: ‘‘the origi-
nally progressive notion of people’s participation in
development, once so central to the RDP, seems to have
gone the same way as the notion of self-help in that it
has often become a mere gesture’’ (Kotz
ee, 1998, p. 98,
emphasis in original). One of the main achievements of
the SAHPF, which will be discussed in the next section,
has been a discursive shift away from top-down tech-
nocratic development management to a participatory
people-centred housing process in the late 1990s.
Regarding institutional architecture, the democratic
transition has yielded an overall reorganisation of state
institutions at different scales with an increasing degree
of democratic decentralisation to the provincial and
local levels. Following The White Paper on Housing,a
division of labour has been established between policy
making and overall resource allocation (e.g. state land
and housing subsidies) by national government and
decentralised implementation by provincial and local
state institutions. The provincial level, which has limited
autonomy in regard to national government, is granted
significant powers in regard to the local state, both in
terms of decision making on resource allocation and in
terms of intervention in actual implementation at the
local level (Republic of South Africa, 1997). The Pro-
vincial Housing Boards, for instance, can make strategic
decisions on minimum standards for housing and pri-
oritise among development projects within parameters
set by national policies and budget allocations, but have
limited power to influence the fiscal mechanisms that
determine the availability and allocation of economic
resources (Kihatoo, 2000). The present regime has ac-
tively sought to redesign the ensemble of local state in-
stitutions and government as developmental local states,
with an embedded autonomy in regard to local civil
society and economy as well as to political institutions
and actors at provincial and national levels (Oldfield,
2000b; Ramutsindela, 2001). The expectation is that
such local developmental states will generate efficient
and relevant economic and social development and also
contribute to a substantial democratisation from below.
More time and research is clearly required before any
conclusion can be drawn about these experiences. What
seems clear at the present conjuncture is that the polit-
ical autonomy of local institutions is constrained by
political processes and decisions at the other levels and
that this may have prohibited the local state from
playing the anticipated developmental role. Thus, Hu-
chzermeyer observes that ‘‘the centralised nature of
urban policy in South Africa in which local government
acts merely as implementer, has discouraged the explo-
rations of alternatives from within local government’’
(Huchzermeyer, 2000, p. vii). On the other hand, critical
M. Millstein et al. / Geoforum 34 (2003) 457–468 461

insight is also required regarding the local embedded-
ness of the local state. The national and local politics of
land allocation may provide an illustrative example.
Land is a crucial component in any integrated housing
process and, thus, for the de-racialisation of South Af-
rican cities (Republic of South Africa, 1995, 1997). The
White Paper on Housing (Republic of South Africa,
1995) and The Housing Act (Republic of South Africa,
1997) called on provincial and local governments to
utilise needed means to ensure a supply of land for
housing and established the principle that unoccupied
land should be transferred to the local state.
2
This
gradual transfer of land to local authorities may con-
tribute to more flexible, speedy, simple and ultimately
more efficient land allocation. However, it also opens up
the local politicisation of land issues. It seems likely that
elite interests may ensure that land in desirable locations
will be allocated to other purposes than social housing,
but local control also opens up channels for popular
struggles for land. In any case, it is obvious that the
democratising and developmental effects of decentrali-
sation cannot be taken for granted.
South AfricaÕs democratic transition has radically
transformed the system of formal and informal political
channels that are available to popular movements. Civics
were, prior to 1994, generally opposed to the apartheid
state and operated with a high degree of political au-
tonomy vis-
aa-vis state institutions and government.
Maharaj (1996) demonstrates that the local reality was a
bit more complex as certain local political channels did
in fact exist in the 1980s. Civics involved in housing
struggles used these channels and thereby contributed to
the transformation of the local state and to emerging
tensions between the local and the national state. Nev-
ertheless, this pattern does not fundamentally alter the
understanding of state/civics relations as principally
antagonistic prior to the democratic transition. Post-
apartheid political opportunity structures are clearly of
a very different nature.
The South African democratisation process has ob-
viously created new opportunities for access to and
strategic alliances with political elite groups, but also
tendencies towards co-optation and de-politicisation. A
few key mechanisms should be mentioned here. First,
the character of the democratic transition in South Af-
rica, with a strong emphasis on negotiations and com-
promises between diverse stakeholders, means that
political participation of civics has been institutionalised
through various advisory boards and forums. This re-
structuring has opened up a number of formal and in-
formal channels for political influence. However, several
observers have also pointed out that this has depoliti-
cised civics and placed them in a role as stakeholders
rather than as a critical political force (Bond, 2000b;
Jones and Datta, 2000; Lodge, 1999; Marais, 2001;
Napier, 1998; Smit, 2001). Second, the ANC has
throughout the 1990s had a somewhat ambivalent re-
lation with civics. On the one hand, civics have been seen
and treated as useful political allies. On the other hand,
they have also been perceived as instrumental commu-
nity organisers within a technocratic and top-down de-
velopment process. This view has rendered a role for
civics as parastatal development agencies or state-sup-
ported NGOs (Marais, 2001; Napier, 1998; Smit, 2001).
Finally, several civics leaders and activists have been
integrated into formal politics through political and
bureaucratic positions. Although civics sought to build
popular democratic structures in local communities,
they often remained somewhat hierarchical and top-
down. In practice the loss of key leaders and activists
has constituted a significant weakening of their organi-
sational capacities (Levine and Weiner, 1996).
In sum, the identified post-apartheid changes in po-
litical discourses, institutions and channels may have
created new opportunities but also reduced the necessity
and ability of civics to function as independent watch-
dogs. This shift means that the political transition has
been an ambiguous political process for popular move-
ments––producing new and widened political spaces but
also tendencies towards de-politicisation.
5. SAHPF as a post-apartheid civil/political association
The SAHPF is a popular movement whose aim is to
realise fundamental socio-economic rights for the urban
poor.
3
The SAHPF emerged from a conference of
homeless people from informal settlements and backyard
shacks, held in Broederstrom in 1991. This conference
identified the primary needs of the urban poor to be
shelter and land for houses, and argued that access to
credit was a critical precondition for fulfilment of the
right to adequate shelter. Following from this, local
saving groups started to appear in October 1992 and were
consolidated as a co-ordinated movement––uMfelan-
dawonye waBantu BaseMjondolo (The SAHPF)––in
2
While discussing the political mechanisms for land allocation, it
should be kept in mind that the market remains the main mechanism
for land allocation in urban areas. High land prices and relatively
modest housing subsidies, combined with the requirement that land
must be in hand before the housing subsidy can be made available to
the builder, contribute to the reproduction of urban segregation with
housing schemes for poor people located at the urban periphery and
other less desirable areas.
3
The analysis is based on fieldwork in Cape Town in the fall of
2000. Primary data was gathered through qualitative interviews with
representatives from the PeopleÕs Dialogue, the SAHPF and from local
government institutions in the Cities of Cape Town and Tygerberg
(now incorporated in the Cape Town UniCity), the South Peninsula
Municipality and the Western Cape Province.
462 M. Millstein et al. / Geoforum 34 (2003) 457–468

March 1994. As one of the fastest growing and most
visible social movement in the 1990s, SAHPF has been
upheld as a paradigmatic case of post-apartheid popular
mobilisation (Billy and Ismail, 1998; Bond, 2000a; Smit,
2001). The Federation is, in several critical ways, different
from the civics that emerged under apartheid. First,
whereas civic associations mobilised around local socio-
economic demands to form a front against apartheid,
post-apartheid popular movements like the Federation
strive for the realisation of socio-economic rights within
the political context of the new democratic state. Second,
while the civics grew out of domestic experiences of op-
pression and exploitation under apartheid, the Federa-
tion has also relied on inspiration from international
discourses, experiences and practices regarding social
housing. Third, whereas the Federation is a loosely
connected network of autonomous local groups with a
high degree of political autonomy, the civics have close
political affiliations with the ANC through the SANCO.
Fourth, there are distinct differences in organisational
form. Every community-based organisation (CBO)
within the Federation has a high degree of autonomy
with a decision making process based on equal partici-
pation of all members. Civics, on the other hand, have
commonly been hierarchically organised with elected
representative committees holding extensive powers to
make strategic decisions and ensure their implementation
(Levine and Weiner, 1996). These differences mean that
SAHPF and the civics represent two organisational
models, with potentially different capacities to mobilise
society, exert political influence and advance the political
inclusion of the urban poor.
5.1. Location in the political terrain
T€
oornquist (1999) argues that analyses of movement
politics should pay close attention to the collective ac-
torÕs strategic deliberations over where and how it is
meaningful to be active. Prior to 1994, the Federation
chose to concentrate on building a network of local
groups in informal settlements that would eventually
yield a strong popular movement of the urban poor
(Bolnick, 1993, 1996). The Federation has gradually
transformed itself from such a civil association to a civil/
political movement, i.e. from concentrating solely on
community mobilisation to gradually combining com-
munity work with political engagement with state actors
at different scales. A decisive factor for the FederationÕs
capacity to influence relevant political processes, both
locally and nationally, has been exactly their capacity
building in local civil society (Joel Bolnick, personal
communication). Thus, it can be argued that the com-
bination of social and political mobilisation has ensured
a degree of political inclusion for socially and politically
marginalised groups. Conversely, their political relations
with state actors and NGOs at different scales have
granted access to vital resources such as housing subsi-
dies for the empowerment of SAHPFÕs constituencies of
urban poor.
The Federation advocates a people-driven housing
process (Bolnick, 1993, 1996; Heldal, 1997). They em-
phasise that the poor are the foremost experts in
building their own houses and that their latent capacities
should be captured in housing strategies (PeopleÕs Dia-
logue, 1998a,b, 1999a,b, 2000a,b,c). Still, the actual
construction of houses is only seen as an element in a
larger development process that also involves income
generating activities and overall community building
(Bolnick, 1993, 1996) (Qxoliswa Tiso, personal com-
munication). The core of the SAHPFÕscommunity mo-
bilisation is the local saving groups. These saving groups
are different from conventional micro credit schemes in
the sense that saving is a collective activity that is de-
signed to strengthen collective identities and organisa-
tional capacities. Since the act of saving is considered to
be as important as the actual amount of money saved,
loans are granted on the basis of active and long-term
engagement in saving groups rather than the actual ac-
cumulation of funds (Joel Bolnic, Qxoliswa Tiso, per-
sonal communication; Heldal, 1997; Huchzermeyer,
2001). While some groups deposit their savings in
commercial banks, a growing number are using the
FederationÕs own regional funds (Inqolobane), which
also mobilise funds and offer loans for other purposes
than housing.
The SAHPF is occasionally accused of being little
more than a conventional NGO that idealises self-help
in civil society and allows the state to shift the respon-
sibility for adequate shelter onto the poor communities
themselves (Bond, 2000a). Leaders and activists that we
have interviewed counter this critique by pointing to the
aforementioned collective and ideological nature of the
local groups. The organisation of saving as a group ef-
fort rather than as individual resource mobilisation
makes the saving groups a mechanism for collective
mobilisation and identity formation (Huchzermeyer,
2000, 2001). This idea is expressed in the slogan: ‘‘When
we collect money, we collect people’’ (Qxoliswa Tiso,
personal communication). Such community mobilisa-
tion forms the basis for political inclusion and negotia-
tions with state actors. Furthermore, SAHPFÕs people-
driven housing process is also about developing and
demonstrating the sustainability of an alternative to the
hegemonic state discourse and housing strategy. Com-
munity mobilisation thus constitutes a critical precon-
dition for both material and discursive political
inclusion of homeless people, i.e. a strategy of political
engagement rather than disengagement in regard to the
state. Expressed in the following way by one informant
within PeopleÕs Dialogue: ‘‘The aim is to get the state on
the hook, not off the hook’’ (Joel Bolnick, personal
communication).
M. Millstein et al. / Geoforum 34 (2003) 457–468 463

5.2. Political channels for resource mobilisation
A key feature of a collective actorÕs strategic delib-
erations around location in the political terrain is an
active interpretation of institutional architecture and
political opportunity structures. There were few and
weak links between the South African Federation of
Homeless People and the state prior to 1994. Federation
leaders saw it as meaningless to develop relations to
state institutions that remained embedded in the apart-
heid regime and chose instead to concentrate on ca-
pacity building in poorly organised informal settlements
(Pieterse and Simone, 1994; Seekings, 2000). The dem-
ocratic transition created a political space that made it
possible to establish a network of CBOs. However,
the Federation still chose to abstain from the negotia-
tions in the National Housing Forum and other similar
arenas.
SAHPFÕs strategy regarding where to be active
changed after the first democratic elections in 1994,
when there was a shift towards critical and selective
engagement with the state (Joel Bolnick, personal
communication; Bolnick, 1996). Thereafter, the Feder-
ation has developed informal links to centrally placed
political actors and has been engaged with key institu-
tions like national and provincial housing boards. They
have utilised new policies, like the housing subsidies
programme, to mobilise resources (e.g. land and fund-
ing) for members, while also seeking to influence policies
on land and housing and on the role of poor people in
the housing process. Access to alternative sources of
credit has always been an important issue for the Fed-
eration simply because the members are unable to ob-
tain loans from formal financial institutions. Even
though most Federation members are entitled to subsi-
dies from the government, it has been a common expe-
rience that the funds are insufficient and normally
received after the project is started or the houses are
built. Access to bridging finance is therefore impor-
tant. An important breakthrough came in 1994/1995,
when the Federation established its own revolving
fund (uTshani). The uTshani fund grants the members
access to small loans for housing. uTshani has also
been used as a channel through which national and
provincial governments release subsidies and other re-
sources to local Federation groups. The institutionali-
sation of uTshani and subsequent uTshani agreements
with the national or provincial state have granted the
Federation control over vital economic resources that
they can use to enable local people-driven housing
processes.
This engagement with the state has not, according to
our informants, yielded any extensive co-optation and
de-politicisation. The Federation has avoided political
affiliation with any political party but has built flexible
alliances across party boundaries. Their emphasis on
building community capacity has given the Federation
considerable bargaining power towards the state. The
Federation had already built functioning community
structures and houses through a practical and people-
driven strategy that was consistent with new state dis-
courses on participatory development. Whereas civics
within SANCO have focused on citizenship rights and
have made political demands about their realisation, the
Federation has focused on the practical fulfilment of
formal rights through resource mobilisation and com-
munity mobilisation (Charles Croeser, Joel Bolnick,
personal communication). Thus, SAHPF has something
to offer vis-
aa-vis a state troubled with economic and
institutional constraints within the housing sector
(Charles Croeser, personal communication). In dealing
with the state, the Federation could legitimately claim
that: ‘‘We do what you say you would like to do but
cannot handle; if we get sufficient resources we can handle
this together’’ (Joel Bolnick, personal communication).
In effect the Federation has utilised their organisational
capacity, created through mobilising structures in local
communities, to strengthen their political relations and
capacities vis-
aa-vis other central actors in the housing
sector.
The Federation is also involved in an active part-
nership with a registered non-governmental organisa-
tion called the PeopleÕs Dialogue (PD). The role of
NGOs in South Africa has undergone marked
changes. Many NGOs, who organised and supported
local struggles against apartheid, now function as non-
governmental service providers in local communities,
i.e. as the bottom tier in a relatively top-down and
centralised development process controlled by the
national state and international aid donors (Napier,
1998; Smit, 2001). Partnerships between NGOs and
community organisations are often characterised by
uneven power relations and the NGOsÕability to rep-
resent the poor in efficient, transparent and accountable
manners may be called into question. Our informants
assert that the partnership between SAHPF and PD
deviates from this pattern, describing it as a key asset for
the Federation. PD plays a key role as a mediator be-
tween the Federation and the formal political sphere,
making formal political procedures, politics and bu-
reaucracy intelligible to local Federation groups while
also communicating and explaining the FederationÕs
grievances and practices to politicians and bureaucrats
(Michael Hoffman, personal communication). They
handle all matters that require formal legal status, in-
cluding fund raising from domestic and international
sources, and thereby facilitate actual people-driven
housing processes in informal communities (Heldal,
1997; Michael Hoffman, personal communication). By
using the PD as a channel into the formal political
system, local Federation groups can maintain their in-
formal and flexible character.
464 M. Millstein et al. / Geoforum 34 (2003) 457–468

5.3. Political discourses and politicisation of issues
Another important meeting point between SAHPF
and the state is in the field of housing discourses. The
White Paper on Housing introduced a state discourse on
peopleÕs power, participatory development and em-
powerment which to a large extent has defined the post-
apartheid discursive field in housing. The state discourse
on housing has been characterised by a firm belief in a
non-conflictual process of negotiations, compromises
and social compacts involving all relevant stakeholders,
including influential private economic interests. This
discourse is also marked by a tension between the need
for speedy delivery of a substantial number of houses
and the desire to involve people and communities in
their own housing development (Bond, 2000a; Lalloo,
1999).
The state discourse on housing stands in contrast to
the Federation discourse which persistently advocates a
truly bottom-up, people-driven housing process where
housing is a means towards actual participatory devel-
opment, self-reliance and empowerment rather than the
principal end itself. Federation activists have accused
the state of promoting a top-down and delivery-oriented
housing approach that reifies the power of state insti-
tutions and private developers while civics and NGOs
are assigned the role of suppliers of services. The state
approach, it is argued, rests on the assumed existence of
well functioning community structures with community
interests articulated through democratic civic associa-
tions (Thami Maqelena, personal communication). In
opposition to this understanding, the Federation insists
on the need to build inclusive and representative com-
munity structures as a precondition for a participatory
housing development process (Joel Bolnick, Qxoliswa
Tiso, personal communication). This means that both
state and Federation discourses on housing revolve
around notions of participation, local civil society and
empowerment. However, this apparent convergence
hides real differences in the conceptualisation and im-
plementation of participatory housing development.
These differences can be summarised in the Federation
slogan that housing development should not just be peo-
ple-centred but also people-driven and controlled.
The FederationÕs counter-hegemonic discourse has
been an important source of inspiration for the gov-
ernmentÕs strategy for a People’s Housing Process that
was launched in 1998 (Joel Bolnick, Charles Croeser,
Belinda Fortune, Nicky Sasman, personal communica-
tion). This strategy reiterates the importance of partici-
patory development and insists on the active
participation of poor people, in partnership with local
state institutions, private developers, NGOs and civics.
This conceptualisation represents a significant discursive
concession to the Federation (PeopleÕs Dialogue, 1998b,
1999b, 2000c). Nevertheless, the Federation maintains
that the implementation of the process remains within
the neo-liberal frame with an emphasis on delivery of
houses, participation as consultation and a continued
neglect of the importance of building community ca-
pacity (Joel Bolnick, Michael Hoffman, personal com-
munication). They argue that systematically uneven
power relations between marginalised communities and
their state, non-governmental and private partners in
the housing process remain unaltered (Huchzermeyer,
2000, 2001). Consequently, the Federation continues to
work for land and funding for its members, despite the
fact that the stateÕs PeopleÕs Housing Process is sup-
posed to obtain the same results and through mecha-
nisms that are presented as identical to those of SAHPF.
5.4. Scale and the politics of land
The point of departure for this article was the ob-
served general and South African tendency towards lo-
calisation of politics in the context of neo-liberal
globalisation. The post-apartheid institutional architec-
ture in South Africa is organised in three tiers of gov-
ernment and administration; national, provincial and
local, with different responsibilities and degrees of au-
tonomy and capacity (Atkinson and Reitzes, 1998;
Cameron, 1999; Oldfield, 2000b). This structure creates
a situation where popular movements like SAHPF have
to make strategic decisions about where to be active in
terms of spheres, as already discussed, but also with
regard to scale. The latter can be briefly illustrated by
the contemporary politics of land allocation in Cape
Town (Millstein, 2001).
The Federation is a key urban social movement and
actor within the housing sector in the Western Cape and
Cape Town. Approximately 400 saving groups are cur-
rently operating within the province, most of them
within the Cape Town metropolitan area. These groups
organise more than 30 000 members, mainly women.
The Federation has constructed more than 5000 houses
in Western Cape during the period 1995–2001 and close
to half of these have been erected without state subsidies
(Joel Bolnick, Qxoliswa Tiso, personal communication;
PeopleÕs Dialogue, 2000b). The FederationÕs most pub-
licised housing scheme, Victoria Mxenge, is located in
Philippi. Victoria Mxenge, which also functions as a
regional network node and education centre (ufundu
Zufe), has been expanded through two new housing
schemes (Hazeldean and Vukuzenzele). These three
Federation projects illustrate three modes of land ac-
quisition. Victoria Mxenge was built on land that was
owned by the Catholic Church and acquired after pro-
longed negotiations with church and local state repre-
sentatives; Hazeldean was constructed on land that was
purchased from a private land owner, and Vukuzenzele
is developed on land that was taken over after a col-
lective land occupation and subsequent negotiations
M. Millstein et al. / Geoforum 34 (2003) 457–468 465

with the registered land owner (Joel Bolnick, personal
communication; PeopleÕs Dialogue, 1998b).
Contemporary South Africa is marked by a complex
system of ownership and institutional responsibilities in
regard to land as well as relatively weak integration of
land issues in housing policies. This system is to a certain
extent a product of the apartheid era, when the admin-
istrative responsibility for state land was fragmented
between diverse institutions and tiers of the state. It
constitutes a major impediment for state and popular
housing development initiatives, for instance through
the difficulties involved in identifying the owner of a plot
of land. Billy and Ismail (1998), for instance, describe
one case where land negotiations ensued for two years
between a local Federation group and the City of Ty-
gerberg before it was discovered that the plot of land
was the property of the Western Cape Province. Fur-
thermore, successful negotiations about land acquisition
require functional relations with a diversity of political
and private actors operating at different scales. The
Federation has traditionally built strong formal and
informal political relations at the national and provin-
cial levels. Such relations have granted a certain influ-
ence on policy formulations, but have also been crucial
for the instrumental interest of acquiring land. Direct
interventions by the Minister of Land and the Ministry
of Housing ensured, for instance, that the Vukuzenzele
group gained access to housing subsidies which could be
utilised to purchase the land that had already been il-
legally occupied by the group. Federation groups in the
Western Cape have on several occasions been able to
purchase land at subsidised cost from the Provincial
government (Michael Hoffman, personal communica-
tion). Land purchases have also been facilitated by
uTshani agreements at the national and provincial
levels.
An on-going process of transferring responsibility for
state land to the local state means that while land
ownership should become less convoluted, the political
opportunity structures for housing movements are being
recast and new needs for local political relations im-
posed. At the local level in the Western Cape, the Fed-
eration has had quite different experiences with the cities
of Tygerberg and Cape Town (now merged within the
Cape Town UniCity). Several informants observe that
local authorities in the City of Tygerberg subscribed to a
significant degree to the Federation discourse on hous-
ing and also provided organisational and technical as-
sistance (Thami Maqelena, Qxoliswa Tiso, personal
communication). SAHPF was, at the time of our field-
work, involved in three major partnerships with the City
of Tygerberg, where local groups had been allocated
land from the city. This arrangement stood in contrast
to the much more problematic relationship to the City
of Cape Town, with numerous conflicts over the nature
and extent of popular control and participation in the
housing process, over allocation of land and regarding
the construction and operation of infrastructure and
public services (Qxoliswa Tiso, personal communica-
tion). Thami Maqelena (personal communication) as-
serts that these conflicts with the City of Cape Town
were generated in the meeting of established political
actors and constellations and the SAHPF as a relatively
new political actor at this level. Strong local Federation
groups, which advocate and implement an alternative
and participatory housing process, pose a challenge to
existing institutions and actors within the housing sec-
tor. Such groups demonstrate that the current institu-
tional architecture and channels for participation are
insufficient to ensure substantial local democratisation
within the housing sector (Thurman, 2000). In conse-
quence SAHPF groups now find themselves within a
local political context that is characterised by changing
institutional responsibilities and capacities. This dy-
namic poses challenges in terms of developing mean-
ingful political relations and strategies in regard to land
and other issues.
The difficulties involved in gaining access to land
have initiated new strategic deliberations within the
movement. SAHPF and PeopleÕs Dialogue advise local
groups to avoid land occupations since negotiations and
agreements with local governments are seen as more
likely to succeed than confrontation. Land occupation is
only recommended as an instrumental mean for initi-
ating such negotiations. However, the difficulties expe-
rienced in getting access to land combined with recent
legal reforms that should reduce the propensity of local
governments to retort to violence, may lead to more
frequent use of force by local groups (Joel Bolnick,
personal communication). The future is likely to hold an
intensified politicisation of land at the local level in-
volving diverse state and non-state actors in complex
systems of conflict and alliance. The local political
spaces and capacities of key popular actors such as the
SAHPF will be crucial in determining the outcome of
this potentially conflictual situation.
6. Conclusion
Contemporary tendencies towards localisation of
politics call for critical analyses of the making and
transformation of political spaces and the political ca-
pacities of local actors to utilise rights, institutional
channels and discourses to promote their instrumental
and democratic aims. In South Africa, post-apartheid
changes in political discourses, institutions and channels
have created new opportunities for popular movements
but also disabling structures of co-optation and de-
politicisation. While the political spaces and experiences
of local groups are diverse, we believe that the SAHPF
has had a certain capacity to make use of and transform
466 M. Millstein et al. / Geoforum 34 (2003) 457–468

local and non-local political spaces within the housing
sector. The Federation has utilised political relations at
different scales to mobilise resources such as land and
housing, but it has also influenced the formulation of
housing discourses and policies through its practical
experiences with people-driven housing processes. The
prime source of the SAHPFÕs political capacity has been
their ability to mobilise local communities and achieve
results through an alternative housing development
model. These achievements and their own housing dis-
course have been crucial for successful political negoti-
ations with state actors at different scales. Mediated
through the partnership with PeopleÕs Dialogue, these
political relations have granted access to state resources
that could enable further community work. Thus the
SAHPFÕs ability to function as a civil/political move-
ment has granted them a certain capacity to participate
in the complicated process of turning de jure rights to
adequate shelter into de facto rights for the urban poor
as citizens of a democratic South Africa.
Acknowledgements
We would like to express our deep gratitude to
movement activists, community organisers and scholars
in Cape Town who have shared their knowledge with us.
We are also grateful to friends and colleagues in Cape
Town and Oslo for inspiration, support and suggestions.
The observations and arguments in the article remain
our sole responsibility.
References
Adelzadeh, A., 1996. From the RDP to GEAR: The gradual
embracing of Neo-Liberalism in economic policy. Transformation,
66–95.
Adler, G. (Ed.), 2000. Public Service Labour Relations in a Demo-
cratic South Africa. Witwatersrand University Press, Johannes-
burg.
Adler, G., Steinberg, J. (Eds.), 2000. From Comrades to Citizens: The
South African Civics Movement and the Transition to Democracy.
Macmillan, London.
Alvarez, S.E., Dagnino, E., Escobar, A. (Eds.), 1998. Cultures of
Politics, Politics of Cultures: Re-visioning Latin American Social
Movements. Westview, Boulder, CO.
Atkinson, D., Reitzes, M. (Eds.), 1998. From a Tier to a Sphere: Local
Government in the new South African Constitutional Order.
Heineman, Sandton.
Billy, A., Ismail, S., 1998. Build houses, people and communities! In:
Barberton, C., Blake, M., Kotz
ee, H. (Eds.), Creating Action Space:
The Challenge of Poverty and Democracy in South Africa.
IDASA, Cape Town.
Bolnick, J., 1993. The peopleÕs dialogue on land and shelter:
Community-driven networking in South AfricaÕs informal settle-
ments. Environment and Urbanization 5 (1), 91–110.
Bolnick, J., 1996. uTshani Buyakhuluma (The Grass Speaks): PeopleÕs
Dialogue and the South African Homeless PeopleÕs Federation
(1994–1996). Environment and Urbanization 8 (2), 153–170.
Bond, P., 2000a. Cities of Gold, Townships of Coal: Essays on South
AfricaÕs new Urban Crisis. Africa World Press, Trenton, NJ.
Bond, P., 2000b. Elite Transition: From Apartheid to Neoliberalism in
South Africa. Pluto, London.
Bozzoli, B., 1996. From Governability to Ungovernability: Race, Class
and Authority in South AfricaÕs Black Cities. Seminar Paper,
Institute of Advanced Social Research, University of Witwaters-
rand, Johannesburg.
Cameron, R., 1999. Democratisation of South African Local Govern-
ment: A Tale of Three Cities. J.L. van Schaik Academic, Pretoria.
Castells, M., 1996. The Rise of the Network Society. Blackwell,
Oxford, UK.
Castells, M., 1997. The Power of Identity. Blackwell, Oxford, UK.
Collinge, C., 1999. Self-organisation of society by scale: a spatial
reworking of regulation theory. Environment and Planning D:
Society and Space 17 (5), 557–574.
Coovardia, C., 1991. The role of the civic movement. In: Humphries,
R., Shubane, K., Swilling, M. (Eds.), The Apartheid City in
Transition. Oxford University Press, Cape Town.
Cox, K.R., 1998. Spaces of dependence, spaces of engagement and the
politics of scale, or: looking for local politics. Political Geography
17 (1), 1–23.
Cox, K.R. (Ed.), 1997. Spaces of Globalization: Reasserting the Power
of the Local. Guilford Press, New York.
della Porta, D., Diani, M., 1999. Social Movements: An Introduction.
Blackwell, Oxford, UK.
Habib, A., Kotz
ee, H., 2002. Civil society, governance and development
in an era of globalisation. Paper, Conference on ‘‘Local Politics
and Democratisation in Developing Countries’’, October 17–19,
University of Oslo, Oslo.
Hart, D., Parnell, S., 1999. Self-help housing as a flexible instrument of
state control in 20th century South Africa. Housing Studies 14 (3),
367–386.
Held, D., 2000. A Globalizing World? Culture, Economics, Politics.
Routledge, London.
Held, D., McGrew, A., Goldblatt, D., Perraton, J., 1999. Global
Transformations: Politics, Economics and Culture. Polity Press,
Cambridge.
Heldal, I., 1997. The South African Homeless PeopleÕs Federation and
PeopleÕs Dialogue Alliance. SIDA, Stockholm.
Holland, J., Mohan, G., 2000. Citizenship and rights-based develop-
ment: The new hegemonic discourse on (local) development. Paper,
Workshop on ‘‘Local Politics and Democratisation in Developing
Countries’’, November 16–18, University of Oslo, Oslo.
Huchzermeyer, M., 2000. The exploration of appropriate informal
settlement intervention in South Africa: Contributions from a
comparison with Brazil. Ph.D. dissertation, Department of Soci-
ology, University of Cape Town, Cape Town.
Huchzermeyer, M., 2001. Housing the poor? Negotiated housing
policy in South Africa. Habitat International 25, 303–331.
Johnston, H., Klandermans, B., 1995. Social Movements and Culture.
UCL Press, London.
Jones, A.G., Datta, K., 2000. Enabling markets to work? Hous-
ing policy in the ‘‘new’’ South Africa. International Planning
Studies 5.
Kihatoo, C., 2000. A critical look at local government planning in
South Africa. Paper, International Conference on Democratic
Decentralisation, May 23–27, Thiruvananthapuram, India.
Kotz
ee, H., 1998. Civil society: Setting the record straight. The problem
of defining development roles within the new paradigm. In:
Barberton, C., Blake, M., Kotz
ee, H. (Eds.), Creating Action
Space: The Challenge of Poverty and Democratization in South
Africa. IDASA, Cape Town.
Lalloo, K., 1999. Arenas of contested citizenship: Housing policy in
South Africa. Habitat International 23, 35–47.
Levine, R., Weiner, D., 1996. The politics of land reform in South
Africa after apartheid. Journal of Peasant Studies 23 (2/3).
M. Millstein et al. / Geoforum 34 (2003) 457–468 467

Lodge, T., 1999. South African Politics since 1994. David Philip
Publisher, Johannesburg.
MacLeod, G., Goodwin, M., 1999. Reconstructing an urban and
regional political economy: on the state, politics, scale, and
explanation. Political Geography 18 (6), 697–730.
Maharaj, B., 1996. Urban struggles and the transformation of the
apartheid local state: The case of community and civic organiza-
tions in Durban. Political Geography 15 (1), 61–74.
Marais, H., 2001. South Africa. Limits to Change: The Political
Economy of Transition, second ed. Zed, London.
Marston, S., 2000. The social construction of scale. Progress in Human
Geography 24 (2), 219–242.
Mayekiso, N., 1996. Township Politics: Civic Struggles for a new
South Africa. Monthly Review Press, New York.
McAdam, D., McCarthy, J.D., Zald, M.N. (Eds.), 1996. Comparative
Perspectives on Social Movements: Political Opportunities, Mobi-
lizing Structures, and Cultural Framings. Cambridge University
Press, Cambridge, UK.
McEwan, C., 2000. Engendering citizenship: Gendered spaces of
democracy in South Africa. Political Geography 19 (5), 627–651.
Melucci, A., 1996. Challenging Codes: Collective Action in the
Information Age. Cambridge University Press, Cambridge, UK.
Migdal, J.S., Kohli, A., Shue, V. (Eds.), 1994. State Power and Social
Forces: Domination and Transformation in the Third World.
Cambridge University Press, Cambridge, UK.
Millstein, M., 2001. uTshani BuyaKhuluma (Grasrota snakker). South
African Homeless PeopleÕs Federation og post-apartheidstaten
(thesis). Department of Sociology and Human Geography, Uni-
versity of Oslo, Oslo.
Mohan, G., Stokke, K., 2000. Participatory development and empow-
erment: The dangers of localism. Third World Quarterly 21 (2),
247–268.
Mouzelis, N., 1986. Politics in the Semi-Periphery: Early Parlamen-
tarism and Late Industrialisation in the Balkans and Latin
America. Macmillan, London.
Napier, M., 1998. The struggle for shelter in a changing South Africa:
Towards an understanding of new roles for non-governmental
organizations in a time of transition. Paper, Seminar on Associ-
ational Life in African Cities, Urban Governance in an Era of
Change, Uppsala, Sweden.
Nattrass, N., Seekings, J., 2001. Distribution in highly unequal
societies: the case of South Africa. Journal of Modern African
Studies 39 (3), 471–498.
Nederveen Pieterse, J., 2001. Development Theory: Deconstructions/
Reconstructions. Sage, London.
Neocosmos, M., 1998. From peopleÕs politics to state politics: Aspects
of national liberation in South Africa. In: Olukoshi, A. (Ed.), The
Politics of Opposition in Contemporary Africa. Nordic Africa
Institute, Uppsala, Sweden.
Oldfield, S., 2000a. The centrality of community capacity in State low-
income housing provision in Cape Town, South Africa. Interna-
tional Journal of Urban and Regional Research 24 (4), 858–872.
Oldfield, S., 2000b. State restructuring and urban transformation in
South Africa: A negotiation of race, place and poverty. Ph.D.
dissertation, University of Minnesota, Minneapolis, MN.
Oldfield, S., 2002. Embedded Autonomy and the challenges of
developmental local government. In: Parnell, S., Pieterse, E.,
Swilling, M., Wooldridge, D. (Eds.), Democratising Local Gov-
ernment: The South African Experiment. University of Cape Town
Press, Cape Town.
Oldfield, S., Stokke, K., 2002. Western Cape Anti-Eviction Campaign
Report. Department of Environmental and Geographical Sciences,
University of Cape Town, Cape Town.
Parnell, S., 1992. State intervention in housing provision in the
1980s. In: Smith, D.M. (Ed.), The Apartheid City and Beyond:
Urbanization and Social Change in South Africa. Routledge,
London.
Peck, J., 2002. Political economies of scale: Fast policy, interscalar
relations, and neoliberal workfare. Economic Geography 78 (3),
331–360.
PeopleÕs Dialogue, 1998a. Backyard Fax. PeopleÕs Dialogue, Cape
Town.
PeopleÕs Dialogue, 1998b. uTshani BuyaKuluma. PeopleÕs Dialogue,
Cape Town.
PeopleÕs Dialogue, 1999a. Backyard Fax. PeopleÕs Dialogue, Cape
Town.
PeopleÕs Dialogue, 1999b. uTshani BuyaKuluma. PeopleÕs Dialogue,
Cape Town.
PeopleÕs Dialogue, 2000a. Backyard Fax. PeopleÕs Dialogue, Cape
Town.
PeopleÕs Dialogue, 2000b. SA Homeless PeopleÕs Federation and
PeopleÕs Dialogue. PeopleÕs Dialogue, Cape Town.
PeopleÕs Dialogue, 2000c. uTshani BuyaKhuluma. PeopleÕs Dialogue,
Cape Town.
Pieterse, E., Simone, A.M., 1994. Governance and Development: A
Critical Analysis of Community Based Organizations in Western
Cape. Foundation for Contemporary Research, Cape Town.
Ramutsindela, M., 2001. Down the post-colonial road: Reconstructing
the post-apartheid state in South Africa. Political Geography 20
(1), 57–84.
Reitzes, M., 1998. Democratic consolidation: Local government and
civil society. In: Atkinson, D., Reitzes, M. (Eds.), From a Tier to a
Sphere: Local Government in the new South African Constitu-
tional Order. Heineman, Sandton.
Republic of South Africa, 1995. The White Paper on Housing.
Department of Housing, Johannesburg.
Republic of South Africa, 1997. The Housing Act. Department of
Housing, Johannesburg.
Seekings, J., 2000. The UDF: A History of the United Democratic
Front in South Africa, 1983–1991. David Philips, Cape Town.
Sheppard, E., 2002. The spaces and times of globalization: Place, scale,
networks, and positionality. Economic Geography 78 (3), 307–330.
Smit, W., 2001. The changing role of community based organizations
in South Africa in the 1990s: With emphasis on their role in
development projects. In: Tostensen, A., Tvedten, I., Vaa, M.
(Eds.), Associational Life in African Cities: Popular Responses to
the Urban Crisis. Nordic Africa Institute, Uppsala, Sweden.
Tarrow, S., 1994. Power in Movement: Social Movements, Collective
Action and Mass Politics in the Modern State. Cambridge
University Press, Cambridge, UK.
Thurman, S., 2000. An Evaluation of the Impact of National Housing
Policy in the Western Cape. Development Action Group, Cape
Town.
T€
oornquist, O., 1999. Politics and Development: A Critical Introduc-
tion. Sage, London.
T€
oornquist, O., 2000. Dynamics of Indonesian democratisation. Third
World Quarterly 21 (3).
T€
oornquist, O., 2002. Popular Development and Democracy: Case
Studies with Rural Dimensions in the Philippines, Indonesia, and
Kerala. University of Oslo, Oslo, Centre for Development and the
Environment in collaboration with UNRISD, Geneva.
Webster, N., Engberg-Pedersen, L. (Eds.), 2002. In the Name of the
Poor: Contesting Political Space for Poverty Reduction. Zed,
London.
Wilkinson, P., 1998. Housing policy in South Africa. Habitat
International 22 (3), 215–229.
468 M. Millstein et al. / Geoforum 34 (2003) 457–468