I want to open by offering profuse thanks to Malini Ranganathan for her generous offer to edit this forum on my book, and to Ananya Roy, Nandita Sharma, and Yousuf Al-Bulushi who (perhaps unbeknownst to them) were crucial interlocutors at various stages of the research that became this book.

Delivery as Dispossession emerged from what was initially a much larger project that involved mapping the various socio-spatial forms of relegation in the post-apartheid city. Yes, there were land occupations, the subject of my book, but there were also established informal settlements, backyard shacks, temporary relocation areas (essentially state-run encampments), freestanding state-provisioned homes (frequently in disrepair, to put it mildly), larger state-run housing developments, subsidized working-class housing, and surely quite a few others. I was particularly interested in how and why postcolonial democracies manage their surplus populations by constantly rearranging them in urban space through some combination of coercion and consent, or, what was the provisional title of my book, delivery and dispossession. This is precisely what Engels (n.d. [1972]: 77) was on about when he argued, “The breeding places of disease, the infamous holes and cellars in which the capitalist mode of production confines our workers night after night, are not abolished; they are merely shifted elsewhere!”

It was this political technology of “shifting elsewhere” that I wanted to investigate with this book. But as I proceeded to do so, two things happened. First, I homed in on one particular socio-spatial form – the land occupation – which seemed to me to be at the root of the larger housing system. Established informal settlements are actually just land occupations that have achieved toleration. Backyarders often attempt to escape the dependency of having to live in someone else’s space by occupying land. People living in “temporary” relocation areas – they have been there for well over 15 years now – are often forcibly relocated from occupations. And finally, shoddy state-provisioned housing is often given to occupiers to clear the land, largely to moderate the shame faced by housing officials in the face of an exploding number of occupations. In addition to this analytic rationale, there was a more contingent one: as ethnographers like to say, I “gained access” to a couple of occupations, though I find that kind of language off-putting. I much prefer to say that I grew incredibly close to a number of occupiers as I began carrying out this research, which was inseparable from the political work I was doing with a social movement called the Housing Assembly (see Levenson 2017). This book is therefore an exercise in what I like to think of as comradely ethnography, not far off from what Orisanmi Burton (2023:8), writing in a very different context, has called a “relation of accountability.”

The second thing that happened that made me focus exclusively on land occupations is that I went into the field assuming that everything would conform to what Ananya Roy calls “a story of real-estate rationalities.” And as she rightly observes, it did not. As I watched occupation after occupation face eviction, one thing became clear to me: the standard narrative of gentrification – or “slum gentrification,” as some authors put it (e.g. Lees, Shin, and López-Morales 2016: 140–71) – was not helping me comprehend eviction patterns. Instead, I uncovered a political logic underlying displacement that is tied (but not identical) to what Roy and her coauthors helpfully characterize as the “ruse of housing.” She is absolutely right that promises of housing vouchers facilitate dispossession – if I did not agree, I certainly would not have titled my book Delivery as Dispossession. But there is also one caveat I would add to the South African case that complicates the analysis. Insofar as a ruse implies a trick, housing officials have largely tricked themselves in Cape Town: they have successfully convinced themselves that clearing new land occupations keeps the housing delivery apparatus functioning smoothly. In practice, it does no such thing. But housing delivery is not only some kind of “trick” here. As I argue in the book, it was crucial to the production of the African National Congress’ legitimacy during the transition from apartheid. Or maybe it is right to think of delivery as a ruse here – but an unintended ruse.

Another ruse Roy raises in her comments is that of property: “how and why,” she asks, did “property bec[o]me the best bet for the postapartheid/postcolonial state”? In an aside in the book’s appendix, I suggest that property was residents’ “best bet.” But she is absolutely right to point out that “such a bet can never pay off.” And indeed, in the land occupation referenced in that line in the appendix, it did not: every single one of the residents was evicted. Franco Barchiesi (2011) wrote a brilliant book about the linking of citizenship to employment in modern South African history. There is a great companion volume waiting to be written about the linking of property ownership to citizenship as well. Both are ruses in a context in which South Africa’s unemployment rate remains the world’s highest (World Bank 2023), and where, as I show in my book, a comparable percentage of South Africans lack access to formal housing. The latter figure skyrockets when we make it not about housing but about the claim to a given piece of property. These are ruses insofar as both employment and property ownership are represented as essential, yet neither is remotely achievable today. On the other hand, the desperation for a home of one’s own is understandable when the entire discourse of good citizenship is predicated upon what Nandita Sharma, citing C. B. MacPherson in her review, calls “possessive individualism.”

Sharma’s review is sharp. She takes me to task for overlooking the role of formal citizenship status in all of this. If my argument turns around the function of the law in relation to hegemony, or what Gautam Bhan (2016) calls the “judicialization of resistance,” Sharma insists that the law excludes non-citizens from access to land and housing. According to the letter of the law, she is right: in order to register for the government’s waiting list for housing, for example, South African citizenship is a requirement. But when it comes to the eviction rulings considered in Delivery as Dispossession, citizenship played no role vis-à-vis how the state saw occupiers. In fact, both occupations included quite a few undocumented participants. Their status never came up in either court ruling. Instead, the politics of citizenship only affected eviction outcomes insofar as it affected how residents self-organized. In the case of Siqalo, it played no role whatsoever, especially after the occupiers expelled representatives of political parties. But in Kapteinsklip, it had disastrous consequences. In the book, I discuss the formation of a faction in that occupation called First People First (FPF) a reference to so-called “Colored” South Africans being indigenous to the Cape, whereas “Africans” were claimed to be “migrants” (see p. 91). FPF enforced their exclusivist politics through violence (see p. 100), and in chapter 3, I write about how they defined isiXhosa-speaking residents as “migrants” and attempted to force them out of the occupation altogether. The same was true for a number of undocumented immigrants living in Kapteinsklip, including both Mozambicans and Tanzanians. FPF mobilized against all foreigners, a category they openly racialized, reducing all “Africans” to “migrants.” But when it came time for an eviction ruling, immigration status was never even mentioned in the courtroom. What mattered was how residents factionalized themselves around the politics of citizenship, which irked the judge, who would dismiss them as “queue jumpers” jockeying for a handout.

I think this raises larger questions about imposing one’s own political framework onto the quotidian struggles of the urban poor. Certainly Sharma is right that lacking citizenship excludes non-citizens from housing delivery. But as I demonstrate in the book, housing delivery is not reliable in any case, with the average wait time in Cape Town now exceeding 60 years. Land occupation is people’s – dare I say it? – best bet, and as the comparison in the book suggests, citizenship status is not a crucial driver of evictions.

Finally, Sharma castigates me for paying “insufficient attention to the possibility (and practice) of being political outside of the terrain occupied by the state.” I was under the impression that that was the entire focus of my book! In fact, the very reason I chose to study land occupations had to do with a limit in Gramsci: if in the Prison Notebooks, he wants to understand how people are incorporated into state projects through the private apparatuses of hegemony, why does he select “so-called private organizations like the Church, trade unions, or schools” (Gramsci 1994 [1931]: 67), which remain formally tied to the state? Land occupations constitute a limit case: residents self-organize while attempting to avoid the gaze of the state, yet they still wind up in dialogue with that state. That is the guiding problem in my book, and residents are absolutely seeking to flee the state. That said, Sharma proceeds to clarify what she actually means: “some social movements actively call for the dismantling of states and reject national citizenship as the legitimate basis for political membership.” Indeed they do. But I want to make two important points here. First, I did not encounter any politics along these lines beyond the confines of the university, as supportive of these demands as I might be. And second, I would argue that even these sorts of demands – much like occupiers’ (failed) attempt to fly beneath the state’s radar – have a political society articulation. In other words, insofar as the demand constitutes a program around which to self-organize – a civil society articulation, in the language of the book – that articulation necessarily has a political society articulation, which is to say that it becomes visible to the state. How residents see the state – including rejecting that state – still informs how they are seen by the state.

As Yousuf Al-Bulushi puts this point, “even in cases when land occupiers want nothing to do with the state, or when they assert an autonomous anti-state politics, they are inevitably likely to be drawn into dialectical relationship with the state.” And he makes a brilliant intervention in relation to ongoing debates on the left by way of the book: if the contemporary left is regularly calling for “meaningfully interven[ing] on the terrain of state power,” my book demonstrates the inverse: political parties have consistently “failed to understand the particular dynamics of poor people’s movements.” On the other hand, maybe it all turns around the key term “meaningfully,” insofar as being absorbed by the state is hardly a meaningful intervention on the terrain of state power. Al-Bulushi asks, “how exactly should movements relate to the (integral) state?” I am hesitant to provide anything approximating a road map, but I do think that the book serves as a cautionary tale in one important sense. The lesson is that all collective organization will necessarily be read by the state, often in ways that those self-organizing do not intend. I am not dogmatic about the need to retain autonomy, but insofar as external entities like political parties and NGOs tend to fragment this self-organization, they should be avoided at all costs. That said, context is crucial. Residents did not see themselves as making any demands, whereas the government read an implicit demand: the right to stay put. But where people are seeking something more transformative, as in Sharma’s plea for an anti-nationalist politics, it is far less likely that a specific demand will be imputed to them.

On the other hand, this reading requires understanding the state as a delivery apparatus, but as Al-Bulushi rightfully points out, this is only part of the story. Given the extent of post-apartheid housing delivery, how to square distributive democracy with some of the state’s other faces: “‘state capture,’ violent repression of the poor, and a virulently nationalist anti-immigrant politics”? One possible answer would invoke dualistic political theory, from Machiavelli’s Chiron through Gramsci’s coercion/consent couplet, Althusser’s RSA/ISA distinction, and Bourdieu’s left and right hands of the state. But I am also hesitant to provide an answer that verges on functionalism, in which the welfare and police achieve a perfect balance so as to reproduce the capitalist relations of production. One of the reasons I wrote this book was to accord some relative autonomy to the political and break with functionalist models of population control. My real answer, which is admittedly only provisional, has to do with the way that these two modes of governance are always articulated. We can separate them analytically, but that does not mean that they are not intertwined in practice. To riff on a line from Gramsci (1971: 160) that I quote in the book, the distinction appears “as an organic one, whereas it is in fact methodological.”

So what does this mean for us? First, as I have argued elsewhere (Levenson 2018), it is futile to separate out the “neoliberal” from the “social democratic” aspects of the state. The government may be funding housing delivery, but only through granting contracts (or “tenders”) to private-sector providers, often with partisan connections. The result is that the profit motive remains, as tenderpreneurs attempt to build on the cheap. And so what looks like state-provisioned housing winds up allowing private sector actors to pocket the difference. Second, Al-Bulushi raises the more violent aspects of the state, including its seeming tolerance for xenophobic vigilantism. Here too, I think we need to explore how the two faces of the state are articulated in different ways in different conjunctures. There is another opposition I might add to the list of political theories above: delivery and dispossession. If after apartheid, the book argues, dispossession enables delivery, we can use this as a model for understanding violent forces beyond eviction. Many government actors – including leaders of the Cape Town’s Anti-Land Invasion Unit – believe themselves to be acting in the name of preserving the possibility of delivery. Extending this peculiar rationale, we might say that officials uttering all kinds of xenophobic nonsense still see themselves as defending a welfare state – even if, as Sharma rightly points out, it remains exclusivist, nationalist, and abhorrent. It would be really generative, I think, to apply the framework developed in the book to other instances in which repression and distribution are articulated, tracing how this articulation has shifted over time. In the case of housing, I show how the apartheid model – dispossession through delivery – is inverted after the transition, with delivery coming to enable dispossession: delivery through dispossession. But how might similar articulations play out in relation to other South African state projects, let alone in other postcolonial contexts?

In the end, my book is a plea for conjunctural analysis. It is admittedly pessimistic insofar as even the “success” case is not particularly desirable. Once tolerated, Siqalo’s residents continued in the well-worn channels through which they negotiated with government actors over eviction, this time demanding access to essential services. Toleration is a far cry from liberation, though it certainly beats eviction. But this is a necessary pessimism. The alternative is projecting our political desires onto collective actors who in all likelihood do not hold the politics we presume they do. The celebration of any mass action “from below” is not a political strategy. Without an adequate understanding of the absorptive capacity of the capitalist state today, the dispossessed don’t stand a chance. Here’s to radical experimentation – but under the conditions we actually face.


Barchiesi F (2011) Precarious Liberation: Workers, the State, and Contested Social Citizenship in Postapartheid South Africa. Albany: State University of New York Press.
Bhan G (2016) In the Public’s Interest: Evictions, Citizenship, and Inequality in Contemporary Delhi. Athens: University of Georgia Press.
Burton O (2023) Tip of the Spear: Black Radicalism, Prison Repression, and the Long Attica Revolt. Oakland: University of California Press.
Engels F (n.d. [1972]) The Housing Question. New York: International.
Gramsci A (1971) Selections from the Prison Notebooks. New York: International.
Gramsci A (1994 [1931]) Letters from Prison, Vol. 2. New York: Columbia University Press.
Lees L, Shin HB and López-Morales E (2016) Planetary Gentrification. Cambridge: Polity).
Levenson Z (2017) Social movements beyond incorporation: the case of the Housing Assembly in post-apartheid Cape Town. In: Paret M, Runciman C and Sinwell L (eds) Global Resistance in Southern Perspective: The Politics of Protest in South Africa’s Contentious Democracy. London: Routledge: pp. 89–104.
Levenson Z (2018) The road to TRAs is paved with good intentions: dispossession through delivery in post-apartheid Cape Town. Urban Studies 55(14): 3218–33.
World Bank (2023) Unemployment, total (% of total labor force). Available here (accessed 17 December 2023).

Zachary Levenson is an Assistant Professor of Sociology in the Department of Global and Sociocultural Studies at Florida International University and a Senior Research Associate in the Department of Sociology at the University of Johannesburg. He is the author of Delivery as Dispossession: Land Occupation and Eviction in the Postapartheid City (Oxford University Press, 2022) and the co-editor of the forthcoming volume The South African Tradition of Racial Capitalism (Routledge, 2024).