Skip to main content
Open Access Publications from the University of California

UC Berkeley

UC Berkeley Electronic Theses and Dissertations bannerUC Berkeley

Representation and Recognition: The Politics of Housing in South Africa


How do postcolonial regimes manage the sudden urbanization of surplus populations in the years following democratization? In post-apartheid South Africa, the government has delivered more free, single-family homes than any other democracy in modern history; yet over the same quarter century, the number of informal settlements has grown more than nine-fold. During the apartheid period, the South African state could simply shift populations at will. But the post-apartheid state does not have this option, as it must simultaneously resolve its housing crisis and reproduce its own legitimacy as a democracy in the eyes of its newly integrated, racialized subjects. As new informal settlements emerge – what I call land occupations – city governments must manage the rapid urbanization of surplus populations without appearing authoritarian. My dissertation explores municipal strategies for managing land occupations in post-apartheid Cape Town. I conducted 17 months of fieldwork combining participant observation, interviews, and archival research in two such occupations in Mitchell’s Plain, Cape Town’s second largest township. Through a careful study of eviction targeting, I demonstrate empirically how squatters’ informal politics affect the outcome of municipal urban policies.

One of these occupations, Rivenland, began with a thousand Colored squatters erecting shacks on a publicly owned field far from any major thoroughfare. They did so in a Colored area, and many of them were supporters of the majority political party in Mitchell’s Plain. No nearby neighbors demanded their removal. By contrast, a second occupation, Holfield, began just a couple of kilometers down the road on two contiguous plots of private property. After a few dozen squatters built shacks, hundreds more arrived every day until there were soon 6000 residents. Most of them were Black in a Colored area, and many of them were presumed to be hostile to the ruling party. Holfield sits along the road connecting one of Mitchell’s Plain’s middle class neighborhoods to the city center, and this neighborhood’s residents mobilized continually to demand Holfield’s eradication. After a year, Rivenland was evicted, but Holfield was allowed by the High Court to stay put. Today it contains more than 8000 people by the City’s count. How should we understand this counterintuitive outcome?

This is where I turn to residents’ own politics as a means of explanation. In Holfield, residents were able to organize a coherent settlement committee prior to their eviction hearing. This largely had to do with the way that their leaders framed the occupation as a social movement, with unified action articulated as the most strategic approach to obtaining official toleration. By contrast, the Rivenland occupation was mired in factionalism, with residents aligning with outside organizations – charities, NGOs, political parties – and competing with one another for access to their lawyers and the court. They did this because their occupation was framed as the distribution of plots of land to potential homeowners; this is what I call the politics of petty proprietorship. The extent of this infighting prompted judges to view the Rivenland occupation as opportunistic. The same court ruled the Holfield occupation legitimate, describing the occupiers as “homeless people in need.”

In order to explain this contrast, I develop the concepts of struggles over representation and struggles over recognition. Without the resolution of struggles over representation and the formation of a unified settlement committee, factionalism will persist, and this, I argue, means that eviction is the most likely outcome. But these factions do not merely reflect preexisting divisions along lines of race, religion, or neighborhood; it is precisely through the formation of representative committees – through the process of representation – that divisions emerge and are concretized. Struggles over representation directly impact how occupations are viewed by the municipal government and High Court judges. When struggles over representation are resolved, judges are likely to recognize occupiers as part of a legible and legitimate population. But when struggles over representation are left unresolved as in Rivenland, judges will fail to recognize occupiers as having any legitimate moral claim to the land. Instead, they will likely view them not as a coherent population, but as individual opportunists attempting to bypass the government’s housing distribution program. In short, the moralizing distinction between homeless people in need on the one hand, and opportunistic queue jumpers on the other, emerges from struggles over representation.

In bringing the insights of political sociology to bear upon urban studies, I break with the prevailing explanation that evictions are most likely in sites planned for development and are driven solely by profit motive. Instead, I conceive of the state not as a coherent institutional entity that simply enacts policies upon populations, but instead as a social relation. The government did not simply design eviction policies and then implement them upon populations; it was through complex relations with residents that eviction outcomes were determined. Only in this way – that is, by seeing the state as a relation, as the condensation of a relationship of forces – can we begin to understand how it was that squatters were evicted from Rivenland and not from Holfield.

Main Content
For improved accessibility of PDF content, download the file to your device.
Current View