Slumdog Cities: Rethinking Subaltern Urbanism


Ananya Roy (, Department of City and Regional Planning, 228 Wurster #1850, University of California, Berkeley, CA 94720, USA.



This article is an intervention in the epistemologies and methodologies of urban studies. It seeks to understand and transform the ways in which the cities of the global South are studied and represented in urban research, and to some extent in popular discourse. As such, the article is primarily concerned with a formation of ideas —‘subaltern urbanism’— which undertakes the theorization of the megacity and its subaltern spaces and subaltern classes. Of these, the ubiquitous ‘slum’ is the most prominent. Writing against apocalyptic and dystopian narratives of the slum, subaltern urbanism provides accounts of the slum as a terrain of habitation, livelihood, self-organization and politics. This is a vital and even radical challenge to dominant narratives of the megacity. However, this article is concerned with the limits of and alternatives to subaltern urbanism. It thus highlights emergent analytical strategies, utilizing theoretical categories that transcend the familiar metonyms of underdevelopment such as the megacity, the slum, mass politics and the habitus of the dispossessed. Instead, four categories are discussed — peripheries, urban informality, zones of exception and gray spaces. Informed by the urbanism of the global South, these categories break with ontological and topological understandings of subaltern subjects and subaltern spaces.


Intervenant sur les aspects épistémologiques et méthodologiques des études urbaines, cet article cherche à comprendre et à modifier les modalités d'analyse et de représentation des villes des pays du Sud dans la recherche urbaine et, jusqu'à un certain point, dans le discours populaire. Pour ce faire, l'attention est portée sur une formation d'idées, ‘l'urbanisme subalterne', qui vise la conceptualisation de la ‘mégacité', avec ses espaces subalternes et ses classes subalternes. Parmi ceux-ci, le ‘taudis' (slum) omniprésent est le plus saillant. Contredisant les textes apocalyptiques et dystopiques sur ce lieu, l'urbanisme subalterne apporte des récits du taudis vu comme un cadre d'habitation, de source de revenu, d'auto-organisation et de réflexion politique. Les écrits explicatifs dominants sur la mégacité sont ainsi mis en question de façon cruciale, voire radicale. Toutefois, l'article s'intéresse aux limites de l'urbanisme subalterne et à ses alternatives. Il met donc en avant des stratégies analytiques nouvelles, avec des catégories théoriques qui transcendent les métonymes habituels du sous-développement comme mégacité, taudis, politique de masse et habitus des défavorisés. Quatre catégories sont présentées à la place: périphéries, informalité urbaine, zones d'exception et espaces gris. Reposant sur l'urbanisme des pays du Sud, elles dérogent aux conceptions ontologiques et topologiques des sujets subalternes et des espaces subalternes.

‘Across a filthy, rubbish-filled creek we enter the slum's heaving residential area, treading carefully to ensure we do not step in human sewage. Live wires hang from wobbly walls; we crouch through corridor-like passages between houses made from reclaimed rubble as the sky disappears above our heads. Behind flimsy doorway curtains we spy babies sleeping on dirty mattresses in tiny single room homes, mothers busy washing, cooking and cleaning.

 The few hours I spend touring Mumbai's teeming Dharavi slum are uncomfortable and upsetting, teetering on voyeuristic. They are also among the most uplifting of my life.

 Instead of a neighbourhood characterised by misery, I find a bustling and enterprising place, packed with small- scale industries defying their circumstances to flourish amidst the squalor. Rather than pity, I am inspired by man's alchemic ability to thrive when the chips are down.’

Crerar (2010)

‘Postcolonial studies, unwittingly commemorating a lost object, can become an alibi unless it is placed within a general frame.’

Spivak (1999: 1)

In the urban imagination of the new millennium, the ‘megacity’ has become shorthand for the human condition of the global South. Cities of enormous size, they are delineated through what Jennifer Robinson (2002: 531) has called ‘developmentalism’. Their herculean problems of underdevelopment — poverty, environmental toxicity, disease — are the grounds for numerous diagnostic and reformist interventions. The megacity can therefore be understood as the ‘constitutive outside’ of contemporary urban studies, existing in a relationship of difference with the dominant norm of the ‘global city’— urban nodes that are seen to be command and control points of the world economy. Following Chantal Mouffe (2000: 12), who in turn relies on Jacques Derrida, the ‘constitutive outside’ is not a dialectical opposite but rather a condition of emergence, an outside that by being inside creates ‘radical undecidability’. The megacity thus renders the very category of global city impossible, revealing the limits, porosities and fragilities of all global centers. Is there a megacity future for every global city? What global city can function without relational dependence on seemingly distant economies of fossil fuels and cheap labor? In this sense, the megacity marks the limits of archival and ethnographic recognition. In this sense, the megacity is the ‘subaltern’ of urban studies. It cannot be represented in the archives of knowledge and it cannot therefore be the subject of history.

This article is an intervention in the epistemologies and methodologies of urban studies. In it, I seek to understand and transform the ways in which the cities of the global South are studied and represented in urban research, and to some extent in popular discourse. As such, the article is primarily concerned with a formation of ideas —‘subaltern urbanism’— which undertakes the theorization of the megacity and its subaltern spaces and subaltern classes. Of these, the ubiquitous ‘slum’ is the most prominent. Writing against apocalyptic and dystopian narratives of the slum, subaltern urbanism provides accounts of the slum as a terrain of habitation, livelihood and politics. This is a vital and even radical challenge to dominant narratives of the megacity. Subaltern urbanism then is an important paradigm, for it seeks to confer recognition on spaces of poverty and forms of popular agency that often remain invisible and neglected in the archives and annals of urban theory.

However, in this article I undertake a study of the limits of such itineraries of recognition by rethinking subaltern urbanism. Drawing on postcolonial theory, I shift the meaning of ‘subaltern’ from the study of spaces of poverty and forms of popular agency to an interrogation of epistemological categories. Following the work of Spivak (1999), the subaltern can be understood as marking the limits of archival and ethnographic recognition; it is that which forces an analysis of dominant epistemologies and methodologies. Meant to be more than epistemological disruption, the article highlights emergent analytical strategies of research. In particular, four categories are discussed — peripheries, urban informality, zones of exception and gray spaces. Informed by the urbanism of the global South, these categories break with ontological and topological understandings of subaltern subjects and subaltern spaces.

The metonymic slum

The megacity is a metonym for underdevelopment, Third Worldism, the global South. As a metonym, the megacity conjures up an abject but uplifting human condition, one that lives in filth and sewage but is animated by the ‘alchemic ability’ (Crerar, 2010) to survive and thrive. And it is the slum, the Third World slum, that constitutes the iconic geography of this urban and human condition. It is the ‘recognizable frame’ through which the cities of the global South are perceived and understood, their difference mapped and located (Nuttall and Mbembe, 2005: 193). Much more is at stake here than Gilbert's (2007: 701) fear of the use of a ‘an old, never euphemistic . . . dangerous stereotype’. If we are to pay attention to what postcolonial critic Gayatri Chakravorty Spivak (1985: 262) has identified as the ‘worlding of what is now called the Third World’, then it is necessary to confront how the megacity is worlded through the icon of the slum. In other words, the slum has become the most common itinerary through which the Third World city (i.e. the megacity) is recognized.

I do not use the term ‘itinerary’ casually. Today, the Third World slum is an itinerary, a ‘touristic transit’ (Freire-Medeiros, 2009). The most obvious example of this is the slum tour, available in the Rocinha favela of Rio de Janeiro, the Soweto township of Johannesburg, the kampungs of Jakarta and the Dharavi slum of Mumbai. One such slum itinerary appears epigraphically as the introduction to this article. In it, The Times journalist Simon Crerar (2010) introduces his readership to the ‘harsh existence of Mumbai's poor’ but also to ‘spirit and enterprise’, where the ‘pace of work’ amidst ‘buzzing flies’ is ‘breathtaking’. From plastic recyclers to the makers of poppadams, Crerar charts his itinerary of a humming and thriving slum. It is in keeping with the ethos of Reality Tours, the ‘ethical tourism’ agency whose guides lead tours of Dharavi. Reality Tours (n.d.) presents Dharavi, ‘Asia's biggest ‘slum’, as a ‘a place of poverty and hardship but also a place of enterprise, humour and non-stop activity’. Proud to be featured in travel guides ranging from Frommer's to Lonely Planet, Reality Tours (ibid.) makes the claim that ‘Dharavi is the heart of small scale industries in Mumbai’ with an ‘annual turnover of approximately US $665 million’. Tour profits are directed towards a nonprofit organization that operates a school for slum children, and slum tourists are discouraged from wielding cameras and photographing slum reality.

Crerar's recent ‘slum and sightseeing tour’ references two dramatic worldings of Mumbai: the ‘terrorist’ attacks of November 2008 with its killing sites of luxury hotels and urban cafes, and the blockbuster film Slumdog Millionaire. Indeed, Crerar's (2010) narrative begins with what is already a well-worn cliché: ‘I've wanted to visit Mumbai since Danny Boyle's Slumdog Millionaire swept to Oscar glory. The film is set in Dharavi , the dusty creek-bed where one million souls live in an area the size of London's Hyde Park, surrounded on all sides by Asia's most expensive real estate’. In Slumdog Millionaire, the various slums of Mumbai are combined into a singular composition that has come to be interpreted as Dharavi, Asia's largest slum. This too is a metonym — Dharavi: Slum.

Slumdog Millionaire is a worlding of the megacity, and of the metonymic megacity Mumbai. The film, with its story of a young slum-dweller and his dreams and aspiration, has been the focus of protests in India for both its apocalyptic portrayal of the ‘slum’ as poverty pornography — we are not ‘dogs’ the slum dwellers of India have bellowed — and its romanticization of a way out of the slum — Salman Rushdie has thus dismissed the film as impossibly unreal (Flood, 2009). Crerar (2010) notes that his guide on the Dharavi tour expressed annoyance at the derogatory use of the term ‘dog’: ‘People were angry with the way they were represented’.

Slumdog Millionaire can be read as poverty pornography. It can also be read as a metonym, a way of designating the megacity that is Mumbai. The film depicts the violent nightmare that is Mumbai: the riots of 1992–3 when Hindu nationalist mobs burned Muslims alive in the alleys of Mumbai's slums; the broken dreams of the migrants who flock to the city but become yet another body in the vast circuits of consumption and capital. Which came first: the fictional and cinematic Mumbai or the real Mumbai of ‘reality’ slum and sightseeing tours? After all, Suketu Mehta's book Maximum City (2004), which itself redraws the line between fiction and ethnography, is an uncanny shadow history of the real Mumbai. Slumdog Millionaire then is only one of the many fictional technologies through which cities like Mumbai are constituted. The film depicts what can be understood as dhandha— entrepreneurial practice akin to street-level hustling. Everyone is out to make a deal — the entrepreneurs of misery who maim children so that they can beg on the sidewalks of Mumbai; the entrepreneurs of space who replace the slums of Dharavi with sky-high condominiums; and the entrepreneurs of dreams who devise game shows as a world of fantasy for the rich and poor. Slumdog Millionaire itself has come to be implicated in a new round of dhandha— from the ‘explosion of slum tourism’ (Crerar, 2010) to the putting up ‘for sale’ of Rubina Ali (one of the film's child actors) by her father. It is this equivalence of cinema and the megacity/slum that Nandy (1999) and Mehta (2008) highlight in different ways. In the wake of the Mumbai killings, Mehta (ibid.) wrote: ‘Just as cinema is a mass dream of the audience, Mumbai is a mass dream of the peoples of South Asia’. Nandy (1999: 2–3) argues that popular cinema in India is the ‘slum's eye view’, with the slum as an entity that ‘territorializes the transition from the village to the city . . . from the popular-as-the-folk to the popular-as-the massified’. Here, categories of equivalence such as ‘popular’ or ‘mass’ connect cinema, slum, megacity and postcolonial nation. It can be argued that this equivalence is the condition of subalternity.

The reception of Slumdog Millionaire in India was marked by protests. Pukar, a Mumbai-based ‘experimental initiative’ founded by anthropologist Arjun Appadurai and concerned with ‘urbanization and globalization’, presented a native refusal of the film's violent narrative of slum dhandha. In particular, Pukar took objection to the word ‘slum’ and sought to reposition Dharavi as a zone of economic enterprise. Here is an excerpt from a Pukar opinion-piece published in The New York Times shortly after the release of Slumdog Millionaire:

Dharavi is probably the most active and lively part of an incredibly industrious city. People have learned to respond in creative ways to the indifference of the state . . . Dharavi is all about such resourcefulness. Over 60 years ago, it started off as a small village in the marshlands and grew, with no government support, to become a million-dollar economic miracle providing food to Mumbai and exporting crafts and manufactured goods to places as far away as Sweden. No master plan, urban design, zoning ordinance, construction law or expert knowledge can claim any stake in the prosperity of Dharavi . . . Dharavi is an economic success story that the world must pay attention to during these times of global depression. Understanding such a place solely by the generic term ‘slum’ ignores its complexity and dynamism (Echanove and Srivastava, 2009).

Pukar's native refusal of Slumdog Millionaire is an example of what I term ‘subaltern urbanism’. Writing against apocalyptic and dystopian narratives of the megacity, it seeks to resurrect the subaltern space of the slum as that of a vibrant and entrepreneurial urbanism. In doing so, it confers recognition on urban subalterns, and perhaps even on the megacity itself as subaltern subject. I am interested in this itinerary of recognition and how it shapes the emergence of what Rao (2006: 227) has described as the ‘slum as theory’— that ‘passage from slum as population and terrain’ to the slum as a ‘new territorial principle of order’. Indeed, the metonymic slum is central to the formation that I am designating as ‘subaltern urbanism’.

Subaltern urbanism

It is a hallmark of postcolonial theory that the Gramscian concept of ‘subaltern’ was taken up by modern Indian historiography, specifically by the group known as the Subaltern Studies Collective (Sarkar, 1984; Spivak, 2005). In this appropriation of Gramsci's ‘Southern Question’, the idea of the subaltern was used to call into question the elitism of historiography (Guha, 1988). ‘Emphasizing the fundamental relationships of power, of domination and subordination’ (Sarkar, 1984: 273), the term came to mean a ‘space of difference’ (Spivak, 2005: 476). Most famously, in Ranajit Guha's (1988: 44) formulation, the subaltern was the ‘demographic difference between the total Indian population and all those . . . described as the “elite” ’. Thus, subalternity came to be seen as the condition of the people, those who did not and could not belong to the elite classes, a ‘general attribute of subordination’ (ibid.: 35). As Spivak (2005: 476) notes, in such usage, the term ‘subaltern’ was closely associated with the idea of the popular. Subaltern politics is thus popular politics and popular culture. Further, in the work of the Subaltern Studies Collective, the agency of change came to be located in this sphere of subaltern politics. In this sense, subalternity became more than the ‘general attribute of subordination’; it also became a theory of agency, that of the ‘politics of the people’ (Guha, 1988: 40). More recently, Partha Chatterjee (2004) has advanced the concept of ‘political society’, a ‘popular politics’ that he distinguishes from ‘civil society’ or the politics of rights-bearing, enfranchised bourgeois citizens. Political society, for Chatterjee (ibid.: 40), involves claims to habitation and livelihood by ‘groups of population whose very livelihood or habitation involve violation of the law’.

I am interested in this shift: from the subaltern marking the limits of archival recognition to the subaltern as an agent of change. As the subaltern is granted a distinct political identity, so this figure comes to be associated with distinct territories. One such territory is the slum. It is also in this way that the idea of the subaltern has entered the realm of urban studies, leading to the emergence of a formation that I call subaltern urbanism. Two themes are prominent in subaltern urbanism: economies of entrepreneurialism and political agency.

Let us return for a moment to Pukar's native refusal of Slumdog Millionaire. Pukar's claim that Dharavi is an entrepreneurial economy is well borne out by the work of various scholars. Nijman (2010: 13) for example argues that the urban slum is more than a warehousing of surplus labor; it is also a space of ‘home-based entrepreneurship’. It is this economy of entrepreneurialism that is on display in the ‘reality tours’ of Dharavi. This too has a metonymic character, for the slum's entrepreneurialism stands in for a more widespread entrepreneurial spirit — perhaps that of the megacity, perhaps that even of the postcolonial nation. Thus, leading Indian journalist Barkha Dutt (2009) writes that Slumdog Millionaire is a ‘masterpiece’ of a movie because it depicts the ‘the energy, entrepreneurship and imagination of the slum kids’. She likens this to ‘the jugadu spirit that is so typical of India’.

Jugadu . . . was originally the word for a marvellous invention — a hybrid automotive that welds the body of a jeep with the engine of a water pump and looks like a tractor. Today it has come to be our shorthand for spunkiness — a we-will-get-the-job done attitude no matter how bad the odds are (ibid.).

In similar fashion, global architect, Rem Koolhaas interprets the urbanism of Lagos as a ‘culture of make-do’ (Enwezor, 2003: 116). In his encounter with Lagos, part of Harvard's ‘Project on the City’, Koolhaas is taken with the inventiveness of its residents as they survive the travails of the megacity. He sees such experimental responses as generating ‘ingenious, critical alternative systems’, a type of ‘self-organization’ creating ‘intense emancipatory zones’ (Godlewski, 2010: 8–9). It is not surprising then that Koolhaas draws the following conclusion: ‘Lagos is not catching up with us. Rather, we may be catching up with Lagos’. In this way, the seemingly ‘alien and distant’ megacity becomes the platform for a ‘neo-organicist’ analysis of urbanism (Gandy, 2005; Godlewski, 2010). As Gandy (2006) has noted, such imaginations turn on the premise of ‘ontological difference’, the African megacity situated outside the currents of world history. There is a lot that can be said about the personage of the star architect and the project of the Third World megacity. But what concerns me here is the emphasis on self-organizing economies of entrepreneurialism, and how this leads us to a theory of subaltern urbanism.

Koolhaas, delirious with the power of his own gaze, is easy to dismiss. But subaltern urbanism far exceeds footloose architects looking for new projects of exploration. Koolhaas' ideas are best paired with those of influential global policy guru, Hernando de Soto (2000), whose libertarian optimism presents the Third World slum as a ‘people's economy’ populated by ‘heroic entrepreneurs’. Here the slum economy is interpreted as a grassroots uprising against state bureaucracy, a revolution from below. For de Soto such economies are rich in assets, albeit in the defective form of dead capital. The ‘mystery of capital’ is how such dormant and defective assets can be transformed into liquid capital, thereby unleashing new frontiers of capital accumulation.

There is a striking resemblance between such arguments of economic libertarianism and the utopian schemes of the Left. For example, in a sketch of ‘post-capitalism’, geographers Gibson-Graham (2008: 613) celebrate the ‘exciting proliferation of . . . projects of economic autonomy and experimentation’. Making a case for the performing of ‘new economic worlds’, for an ‘ontology of economic difference’, Gibson-Graham showcase ‘community economies’ and urge us as researchers to make them more real, credible and viable.

Equally important as a theme in subaltern urbanism is the question of political agency. In his widely circulating apocalyptic account of a ‘planet of slums’, Mike Davis (2004: 28) expresses anxiety about the political agency of slum dwellers, asking: ‘To what extent does an informal proletariat possess that most potent of Marxist talismans: “historical agency”?’. Davis argues that ‘uprooted rural migrants and informal workers have been largely dispossessed of fungible labour-power, or reduced to domestic service in the houses of the rich’ and that thus ‘they have little access to the culture of collective labour or large-scale class struggle’. Against such accounts, subaltern urbanism recuperates the figure of the slum dweller as a subject of history.

Take for example the work of Asef Bayat (2000: 533), who argues that, in Third World cities, a ‘marginalized and deinstitutionalized subaltern’ crafts a street politics best understood as ‘the quiet encroachment of the ordinary’. There is almost a Wirthian quality to Bayat's analysis, for it is the territory of the restructured Third World city that produces this quiet encroachment. More recently, Bayat (2007) has rejected the arguments that link the rise of militant Islamism to the ‘urban ecology of overcrowded slums in the large cities’. The slum, Bayat argues, may not be characterized by radical religiosity but it does engender a distinctive type of political agency: ‘informal life’. For Bayat (2007: 579), ‘informal life’, typified by ‘flexibility, pragmatism, negotiation, as well as constant struggle for survival and self-development’ is the ‘habitus of the dispossessed’. This idea — of a slum habitus — is a key feature of subaltern urbanism.

In a highly sophisticated account, Solomon Benjamin (2008) delineates the differences between three different political arenas: a policy arena penetrated by real estate lobbies and finance capital; a civil society arena that seeks to restrict political activity to those deemed to be ‘legitimate citizens’; and an arena of ‘occupancy urbanism’ through which the urban poor assert territorial claims, practice vote-bank politics and penetrate the lower, ‘porous’ reaches of state bureaucracy. Benjamin's analysis is by no means Wirthian. Indeed, his political-economy account of multiple land-tenure regimes firmly grounds the slum in the circuits of finance and real estate capitalism. But in a manner similar to the Subaltern Studies Collective's conceptualization of popular politics, he grants the poor a distinctive form of political agency: occupancy urbanism. Such urbanism for Benjamin (ibid.: 719, 725) is necessarily ‘subversive’, appropriating ‘real estate surpluses’ and possessing a ‘political consciousness that refuses to be disciplined by NGOs and well-meaning progressive activists’. In this, Benjamin's analysis bears close resemblance to Chatterjee's (2004) conceptualization of ‘political society’ as a space of politics formed out of the governmental administration of populations but escaping such forms of developmentalism.

I am highly sympathetic to the cause of subaltern urbanism. I see it as an important correction to the silences of urban historiography and theory, the ‘sanctioned ignorance’— to borrow a phrase from Spivak (1999: 164) — that has repeatedly ignored the urbanism that is the life and livelihood of much of the world's humanity. Even Koolhaas' encounter with Lagos, as Godlewski (2010: 17) notes, can be seen as a sign of the ‘growing sense that architectural theory should address global practice rather than singular monuments in the Western world’. And it would be naïve to fault subaltern urbanism for the various appropriations of slum entrepreneurialism that today make up an increasingly busy traffic of slum tours, blockbuster films, entrepreneurial NGOs, and globally circulating architects and policy consultants.

In fact, the urgency of such recognition is all around us. Working on the US–Mexico border and tracking the mobile mutations of this militarized urbanism, architect Teddy Cruz (2007: 75) searches for ‘alternative urbanisms of transgression’. In border neighborhoods, he finds ‘a migrant, small scale activism’, what he designates as the ‘informal’. These urbanisms, he argues, inhabit space ‘beyond the property line in the form of non-conforming spatial and entrepreneurial practices’. Here is an effort to imagine a ‘new brand of bottom-up social and economic justice’ amidst the brutal subalternity imposed by the global border. How can such a project be denied sympathy?

Working within, rather than against, subaltern urbanism, I hope to pose some critical questions about this project of recognition and its key analytical themes. In doing so, I draw upon Spivak's critique of itineraries of the ‘subaltern’. Writing against those versions of subaltern studies that seek to identify the subaltern as the subordinate classes, as the ‘demographic difference’, Spivak (1999: 191) casts doubt on ‘conscientious ethnography’ that hopes to study and represent the interests of the subaltern. This ‘produced “transparency” ’, she rightly notes, itself ‘marks the place of “interest” ’ (ibid.: 265). Spivak's work thus challenges us to study how the subaltern is constituted as an object of representation and knowledge — in lieu of the conscientious ethnography that claims to speak for the authentic subaltern. Further, Spivak calls into question the slippage between ‘subaltern’ and ‘popular’ that can be found in the work of the Subaltern Studies Collective and in many renderings of subaltern political agency. In particular, Spivak (2005: 480) draws attention to the metonymic character of such frameworks of subalternity:

Agency presumes collectivity, which is where a group acts by synecdoche: the part that seems to agree is taken to stand for the whole . . . A performative contradiction connects the metonymy and the synecdoche into agential identity.

With such critiques in mind, I present three challenges directed at the formation that is subaltern urbanism. The first is concerned with the economy of entrepreneurialism, the second with political agency, and the third with archival and ethnographic recognition.

Itineraries of recognition

Let me return for a moment to the utopian call for post-capitalist worlds by critical scholars like Gibson-Graham. In keeping with the broader formation that is subaltern urbanism, this call pivots on an ontological vision of the people's economy. Gibson-Graham list a set of community economies marked by the register of difference: squatter, slum dweller, landless, informal credit. But it can be argued, as I have in my recent work (Roy, 2010), that these people's economies are also the active frontiers of contemporary capitalism, the greenfield sites where new forms of accumulation are forged and expanded — in the interstices of the slum, in the circuits of microfinance. It is not surprising that post-capitalist yearnings bear such close resemblance to the frontier ambitions of economic libertarians like Hernando de Soto. De Soto sees in the ecology of the slum a world of dead capital waiting to be turned liquid. He is an important interlocutor in a composition of ideas and practices that I have termed ‘poverty capital’— the conversion of poverty into capital, the mining of ‘the fortune at the bottom of the pyramid’ (Prahalad, 2004). The slum, in its territorial density, represents a crucial space for bottom-billion capitalism, one where poor populations can be easily rendered visible for global capital. It is thus that Dharavi, that subaltern site celebrated in the native refusal of Slumdog Millionaire, is much more than a self-organizing economy of the people. It is also increasingly visible to global capital as an urban ‘asset’ (Tutton, 2009), the ‘only vast tract of land left that can be made available for fresh construction activities’ at the heart of Mumbai (Singh, 2007). Mukesh Mehta, the architect who is leading the controversial Dharavi redevelopment plan, argues that this could be India's ‘Canary Wharf’ (Tutton, 2009). At the frontiers of redevelopment, the spaces of poverty celebrated by subaltern urbanism are transformed into what I have termed ‘neoliberal populism’ (Roy, 2010), a thorough commodification of community economies.

Similar issues are at stake in the conceptualization of subaltern political agency. While Benjamin does not suggest that the terrain of occupancy urbanism is immune from political machinations, he nevertheless presents it as the subversive politics of the poor, autonomous of developmentalism, state action and real estate capital. But in the work of Weinstein (2008: 22), land development is also the domain of ‘development mafias’—‘local criminal syndicates, often with global connections’. Such mafias operate in tandem with real estate capital, the state and the police. This too is an occupancy urbanism — of the powerful — and it exists in complex interpenetration with the vote-bank politics and territorial claim-making of urban subalterns.

To understand the turn in subaltern studies to the theme of political agency, it is necessary to pay attention to the broader enterprise of postcolonial theory. In an effort to erode the imperialism of knowledge, postcolonial theorists, especially those trained in the dependista tradition, have sought to end ‘epistemic dependency’ (Mignolo, 2002: 85). For Mignolo (ibid.: 90), this means confronting ‘colonial difference’, that which marks the limits of ‘theorizing and thinking’, that which made the world ‘unthinkable beyond European (and later, North Atlantic) epistemology’. As an example of ‘liberating reason’, Mignolo turns to Dussel's (2000: 473) idea of ‘transmodernity’, where ‘modernity and its denied alterity, its victims, would mutually fulfill each other in a creative process’.

Subaltern urbanism, with its emphasis on the subaltern as political agent, is a recuperation of modernity's supplement, the colonized Other. Particularly important here is Gidwani's (2006) conceptualization of ‘subaltern cosmopolitanisms’. Writing against a cosmopolitanism that postures ‘its provincial and prejudiced European origins in the name of the “universal” ’, Gidwani (ibid.: 16, 18) draws attention to ‘the connections that are possible between the different disenfranchised’. Similarly McFarlane (2007) documents the worldliness of slum politics in Mumbai, designating these imaginaries and practices as ‘slum cosmopolitanism’. Such interventions lie at the very heart of postcolonial theory: they disrupt, trouble and thereby decolonize ideas of modernity and cosmopolitanism. And in doing so, they grant political/postcolonial agency to the subaltern, although as Jeffrey and McFarlane (2008: 420) note, ‘subaltern cosmopolitanism is often contradictory and limited in its political effects’.

But what is this postcolonial agency? I am taken with the worldliness of the subaltern, with the unbounding of the global slum, with the new solidarities and horizontalities made possible by such transmodern exchanges. But I am also taken with how the ‘colonial wound’ (Mignolo, 2005: 95) is the occasion for a host of postcolonial centerings, for violent practices of domination and hegemony. In this transmodernity, postcolonial experiments inaugurated by emergent nation-states and their megacities generate and stage global value. Such experiments cannot be read as a reversal of colonial power; instead they demonstrate the brutal energy of the postcolony. In some of our recent work, Aihwa Ong and I have sought to analyze how such experimental urbanisms are producing an interconnected and interreferenced Asia, a complex circuitry and hierarchy of expertise, investment and development crisscrossing Mumbai, Singapore, Shanghai, Dubai, Manila and Jakarta (Roy and Ong, 2011). These postcolonial experiments transform the ‘colonial wound’ into the ideology of an ascendant ‘Asian century’, of a history to be dominated by the economic powerhouses of a territory imagined as Asia.

It seems to me that in order to tackle the question of subaltern political agency, it is necessary — as Spivak (2005) has noted — to make a distinction between agency and identity. If the subaltern is inscribed as the popular, then it is possible to ascribe an identity to the people's politics. But if we are to return to Guha's original sense of the subaltern as ‘demographic difference’, then this space of subordination cannot be represented by a coherent identity. At best, subaltern politics can be seen as a heterogeneous, contradictory and performative realm of political struggle. It is this performativity that Spivak (ibid.: 482) seeks to capture through a return to the idea of metonymy: ‘The point is to have access to the situation, the metonym, through a self-synecdoche that can be withdrawn when necessary rather than confused with identity’.

Here the recent work of Craig Jeffrey (2009) on jugaar is instructive. While subaltern urbanism tends to see jugaar/jugadu as the inherent make-do spirit of bricolage that characterizes the slum, Jeffrey presents jugaar as a political entrepreneurship that is strategically performed by various social classes. Studying how student fixers operate within an ‘informal economy of state practices’, Jeffrey (ibid.: 205–6) shows how young men belonging to a rural middle-class engaged in ‘shrewd improvisation’ to tackle a future of unemployment and economic uncertainty. Jeffrey's analysis is important for two reasons. First, it shows how a seemingly subaltern strategy such as jugaar can be taken up by various social classes and deployed in quite differentiated ways (in this case, middle-class students could collude with state officials). Second, Jeffrey highlights the ‘moral ambivalence’ associated with jugaar and how such ambivalence was negotiated through cultural inventions, all of which required drawing upon the social basis of class power. It turns out then that as with the case of occupancy urbanism or street politics, jugaar is not the ‘habitus of the dispossessed’ but instead a flexible strategy wielded differentially by different social classes in the context of urban inequality.

What then is the subaltern? Where then is the subaltern? In her critique of subaltern studies, Spivak (2005: 476) argues that ‘subalternity is a position without identity’. She continues: ‘Subalternity is where social lines of mobility, being elsewhere, do not permit the formation of a recognisable basis of action’. In this sense, the subaltern is neither habitus nor territory, neither politically subversive nor culturally pragmatic. Against ontological and topological readings of the subaltern, and building on Spivak's critique, I argue that the subaltern marks the limits of archival and ethnographic recognition. Such an idea returns us to Guha's (1988) initial interest in challenging the ‘elitism of historiography’ or Chakrabarty's (1988: 179) mandate to study the ‘conditions for knowledge of working-class conditions’, with special attention to the ‘silences’ of ‘ruling class documents’. But while Guha, Chakrabarty and other members of the Subaltern Studies Collective seek to recover the history of subaltern classes, Spivak rejects such an itinerary of recognition. In her work, the subaltern marks the silences of our archives and annals. It is this conceptualization of the subaltern that I believe is most useful to urban studies, for it calls into question the conditions for knowledge through which ‘slumdog cities’ are placed in the world.

Beyond recognition

In my earlier work, I have argued that the study of the twenty-first-century metropolis requires new geographies of theory (Roy, 2009). Subaltern urbanism is indeed one such approach. It is a vital and even radical challenge to apocalyptic and dystopian narratives of the megacity. However, subaltern urbanism tends to remain bound to the study of spaces of poverty, of essential forms of popular agency, of the habitus of the dispossessed, of the entrepreneurialism of self-organizing economies. I am interested in a set of theoretical projects that disrupt subaltern urbanism and thus break with ontological and topological understandings of subalternity. In the broadest sense I am interested in the following question: how can we understand the inevitable heterogeneity of Southern urbanism, that which cannot be contained within the familiar metonymic categories of megacity or slum, and that which cannot be worlded through the ‘colonial wound’? With this in mind, I foreground four emergent concepts: periphery, urban informality, zones of exception and gray spaces. Each concept has a distinctive genealogy and therefore cannot be seen as new or novel. For example, the idea of periphery can be traced to Latin American dependista frameworks of world systems and their geographic polarizations. Similarly, the notion of the informal sector was first put forward in the context of East African economies, and then traveled to explain forms of deproletarianization and deregulation in many other parts of the world. My claim is not that these concepts are new and therefore worthy of attention. Rather I am interested in how scholars, working in a variety of urban contexts, are using such concepts to chart new itineraries of research and analysis.


In a recent treatise on city life, AbdouMaliq Simone (2010) makes the case for the importance of the periphery in urban life. Simone's concept of the periphery is multivalent. By it, he means cities that have been ‘at the periphery of urban analysis’ and whose urbanism has thereby been ignored (ibid.: 14). By periphery, he also means a ‘space in-between . . . never really brought fully under the auspices of the logic and development trajectories that characterize a center’ (ibid.: 40). It is the ‘entanglement’ of periphery and ‘possibility’ that most interests Simone (ibid.: 33). On the one hand, the periphery — not unlike the slum — is a space produced through the interventions of humanitarianism, urban restructuring, capital flows, policing and control. But on the other hand, the periphery is a ‘potentially generative space — a source of innovation and adaptation . . . potentially destabilizing of the center’ (ibid.: 40).

Is the periphery also a place? Simone (ibid.: 41), following Jacquier, rejects topological meanings of the periphery and instead uses the term to refer to a ‘range of fractures, discontinuities, or “hinges” disseminated over urban territories’. But he also identifies the ‘interstitial zone between urban and rural’ as one of several significant peripheries (ibid.: 45). Similarly, Holston and Caldeira (2008: 18), seeking an alternative to the vocabulary of slums, present the autoconstructed peripheries of Brazilian cities as spaces of the invention of citizenship: ‘Sites of metropolitan innovation’, they argue, ‘often emerge at the very sites of metropolitan degradation’. Here the periphery signifies a relationship of interdependence in an apparatus of domination but it also refers to a specific topographical location: the peripheral neighborhoods of the urban poor. Similar ideas can be found in the work of the Los Angeles School of geography, which is concerned with how, in the postmodern metropolis, the hinterlands or periphery now organize evacuated city-cores (Dear and Dahmann, 2008: 269).

The periphery, even in its topological use, is an important conceptual device to decenter urban analysis. But perhaps most significant is the claim that the periphery is both a space in the making and a form of making theory (Caldeira, 2009). Simone (2010: 14), for example, argues that cities at the periphery of urban analysis must be repositioned as an ‘invented latitude’, a ‘swath of urban life running roughly from Dakar to Jakarta’ that has ‘something to do with each other’ and that skirts ‘the usual obligatory reference to cities of “the North” ’. Here then is an itinerary of recognition that is dramatically different from that of the dominant map of global and world cities. Here then is a cartography of transmodernity. But is the periphery as theory a departure from the slum as theory?

I am convinced that the promise of the concept of periphery lies in its ability to transcend territorial location, to demonstrate various foreclosures that complicate political agency and to call into question the conditions for knowledge. Simone (ibid.: 62, 28), for example, highlights how the periphery is also a ‘platform’ for ‘anticipatory urban politics’, one where ‘peripheral status’ can be used as an advantage. However, this is not a habitus of the dispossessed. Indeed, Simone (ibid.: 99) insists that the ‘politics of anticipation is not just a form of resistance or simply a politics from below’ since ‘these very anticipations can also be used by more powerful actors and forces’. Caldeira (2008) goes further, arguing that poor young men of the periphery use cultural tactics such as rap to produce a ‘powerful social critique’. But they also ‘establish a non-bridgeable and non-negotiable distance between rich and poor, white and black, the centre and the periphery, and articulate a position of enclosure’. It is the analysis of such paradoxical forms of social agency that troubles, disrupts and expands the realm of subaltern urbanism.

Urban informality

Subaltern urbanism functions through slum ontologies. Such ontological readings of the megacity and its urbanisms have repeatedly invoked the idea of the informal. Bayat (2007) asserts that informal life is the habitus of the dispossessed. Cruz (2007) sees informal habitation at the global border as an urbanism that transgresses across the ‘property line’. For Hernando de Soto (2000), the informal represents the grassroots rebellion of the poor against state bureaucracy; it is a sign of heroic entrepreneurship. Mike Davis (2004: 24) states that ‘informal survivalism’ is ‘the new primary mode of livelihood in a majority of Third World cities’. He thereby evokes an older usage of the term ‘informal’, that of Keith Hart (1973: 61, 68), who identified a ‘world of economic activities outside the organised labor force’ carried out by an ‘urban sub-proletariat’. In all such frameworks, the informal remains synonymous with poverty and even marginality. It remains the territory and habitus of subaltern urbanism.

Against these various interpretations, in my work I have argued that informality must be understood as an idiom of urbanization, a logic through which differential spatial value is produced and managed (Roy and AlSayyad, 2004). Urban informality then is not restricted to the bounded space of the slum or deproletarianized/entrepreneurial labor; instead, it is a mode of the production of space that connects the seemingly separated geographies of slum and suburb. The splintering of urbanism does not take place at the fissure between formality and informality but rather, in fractal fashion, within the informalized production of space. Informal urbanization is as much the purview of wealthy urbanites as it is of slum dwellers. These forms of urban informality — from Delhi's farmhouses to Kolkata's new towns to Mumbai's shopping malls — are no more legal than the metonymic slum. But they are expressions of class power and can therefore command infrastructure, services and legitimacy. Most importantly, they come to be designated as ‘formal’ by the state while other forms of informality remain criminalized. For example, Weinstein (2008: 23) shows how various shopping centers in Mumbai had been ‘built illegally . . . by the city's largest and most notorious mafia organization, on land belonging to the state government's public works department’. Or, in the case of Delhi, Ghertner (2008: 66) notes that a vast proportion of city land-use violates some planning or building law, such that much of the construction in the city can be viewed as ‘unauthorized’. He poses the vital question of how and why the law has come to designate slums as ‘nuisance’ and the residents of slums as a ‘secondary category of citizens’, while legitimizing illegal and informal ‘developments that have the “world-class” look’. Also in Delhi, Gidwani (2006: 12) characterizes the proliferation of illegal farmhouses as the ‘urban conquest of outer Delhi', a process of ‘unauthorized construction’ that involves ‘cordoning off the few remaining agricultural tracts’.

The valorization of elite informalities and the criminalization of subaltern informalities produce an uneven urban geography of spatial value. This in turn fuels frontiers of urban development and expansion. Informalized spaces are reclaimed through urban renewal, while formalized spaces accrue value through state-authorized legitimacy. As a concept, urban informality therefore cannot be understood in ontological or topological terms. Instead, it is a heuristic device that uncovers the ever-shifting urban relationship between the legal and illegal, legitimate and illegitimate, authorized and unauthorized. This relationship is both arbitrary and fickle and yet is the site of considerable state power and violence. Urban informality thus makes possible an understanding of how the slum is produced through the governmental administration of population (Chatterjee, 2004), as well as how the bourgeois city and its edifices of prosperity are produced through the practices of the state. In this sense, urban informality is a heuristic device that serves to deconstruct the very basis of state legitimacy and its various instruments: maps, surveys, property, zoning and, most importantly, the law.

Zones of exception

The concept of urban informality denotes a shift from slum ontologies to an analysis of sovereign power and its various spatialized negotiations. It also denotes a shift from the territorial imagination of cores and peripheries to the fractal geometries of metropolitan habitation. For a theory of such spatialities, it is necessary to turn to the work of Aihwa Ong. While other theorists have explained the territorial logic of neoliberalism as one of revanchist frontiers (Smith, 1996), spatial dispossession (Harvey, 2005) or the rescaling of ‘state spaces’ (Brenner, 2004), Ong (2006: 7) studies ‘market-driven strategies of spatial fragmentation’. She thus traces patterns of ‘non-contiguous, differently administered spaces of graduated or variegated sovereignty’, or zones of exception. From special economic zones to special administrative regions, these zones both fragment and extend the space of the nation state. Such zoning practices have been particularly visible in China, where liberalization has coincided with ‘zone fever’— the formation of numerous types of zones encompassing economic and technological development zones, high-technology development zones or science parks, bonded zones or free-trade zones, border-region economic-cooperative zones, and state tourist-vacation zones (Cartier, 2001: 455). George Lin (2010) thus reports that the thousands of Chinese zones together cover a territory that exceeds the country's total urban built-up area. Indeed, one may ask: in a territory where zones of exception proliferate, what then is the city?

Ong's work on zones of exception is a crucial counterpoint to subaltern urbanism. Instead of slum entrepreneurialism, she is concerned with what may be understood as the entrepreneurial state. Ong (1999: 215–7) thus argues: ‘I maintain that the nation-state — with its supposed monopoly over sovereignty — remains a key institution in structuring spatial order’. Such order is produced and managed through ‘a system of graduated zones’. What is crucial about such zones of exception is the ‘differential deployment of state power’: that ‘populations in different zones are variously subjected to political control and to social regulation by state and non-state agencies’. Zones of ‘superior privileges’ (ibid.: 219) coexist and contrast with zones of cheap-labor regimes; transnational zones of investment coexist and contrast with transnational zones of refugee administration; zones of neoliberal rule coexist and contrast with zones that are exceptions to neoliberalism.

In Ong's theorization, zones of exception are arrangements of sovereign power and biopower. She is particularly attentive to the ‘technologies of subjectivity’ and ‘technologies of subjection’ (Ong, 2006: 6) that characterize these systems of zones. Here her work has important connections to the theme of exception present in the work of Giorgio Agamben. For Agamben, the space of exception is a state of emergency produced through the sovereign's suspension of the juridical order. It is as Derek Gregory (2010b: 154) has noted a ‘legal–lethal space’. But it is also, as Judith Butler (2004: 98, 67) points out, a state of ‘desubjectivation’, a space where ‘certain subjects undergo a suspension of their ontological status as subjects when states of emergency are invoked’.

But states of exception cannot be seen to stand outside the spaces of metropolitan habitation. Rather they indicate a specific ‘legal–lethal’ logic of rule that is ever present in the seemingly ordinary spaces of the city. Of the various spatial technologies of exception, Derek Gregory (2010a: 84) notes:

The very language of ‘extraordinary rendition’, ‘ghost prisoners’, and ‘black sites’ implies something out of the ordinary, spectral, a twilight zone: a serial space of the exception. But this performative spacing works through the law to annul the law; it is not a ‘state’ of exception that can be counterpoised to a rule-governed world of ‘normal’ politics and power.

Writing against Agamben, Ong (2006: 22, 23) thus notes that it is not enough to trace the ‘logic of exception’ that is ‘invoked against the politically excluded’ and that is measured in relation to a ‘universal norm of humanity’. In the multiple and differentiated zones of exception that she documents, rule unfolds through freedom, rights-talk, virtue, nationalism and many other ‘visions of the good life’.

Gray spaces

In his analysis of the global war prison, Gregory (2010a: 57) interprets such spaces of exception additionally as ‘a potential space of political modernity’. It is a ‘profoundly colonial apparatus of power’ that gives ‘form and force’ to such spaces. But, he notes, the ‘metropolitan preoccupations of Foucault and Agamben more or less erase’ this colonial present. The global war prison as a metonym for colonial violence thus marks the limits of archival and ethnographic recognition. It is also the poignant counterpoint to that other space of political modernity — the popular politics of the subaltern vaunted by Chatterjee (2004).

Gregory's mandate to take up the study of colonialism and its war cultures leads us to the concept of ‘gray spaces’ put forward by Oren Yiftachel. Writing in the context of what he designates as ‘urban colonialism’, Yiftachel (2009a: 88) describes ‘gray spaces’ as ‘those positioned between the “whiteness” of legality/approval/safety, and the “blackness” of eviction/destruction/death’. He notes that these spaces are tolerated and managed but only ‘while being encaged within discourses of “contamination”, “criminality” and “public danger” to the desired “order of things” ’ (ibid.: 89). There are important connections between ‘gray space’ and other concepts that I have presented earlier in this article. ‘Gray spacing’ makes evident the flexibility of sovereign power that is at the heart of Ong's analysis of zones of exception. For Yiftachel (2009b: 247), such ‘gray spacing’ takes place at the ‘periphery of peripheries’, for example the impoverishment of indigenous Bedouins by an ethnocratic Israeli state. At these colonized margins, Yiftachel (2008: 366) argues, ‘bare life’ must be understood ‘as daily routine, not as exception’. And finally, Yiftachel (2009a: 92)is particularly interested, as I am, in analyzing the manner in which state power formalizes and criminalizes different spatial configurations:

The understanding of gray space as stretching over the entire spectrum, from powerful developers to landless and homeless ‘invaders’, helps us conceptualize two associated dynamics we may term here ‘whitening’ and ‘blackening’. The former alludes to the tendency of the system to ‘launder’ gray spaces created ‘from above’ by powerful or favorable interests. The latter denotes the process of ‘solving’ the problem of marginalized gray space by destruction, expulsion or elimination. The state's violent power is put into action, turning gray into black.

Yiftachel's concept of gray spaces both extends and challenges the idea of ‘colonial difference’ and thus the epistemic and political location of subalternity. In settings of colonial difference, can the archives and annals yield the voice of the subaltern? Or is such a voice and existence constantly blackened, constantly erased?

Vanishing points

The elitism of historiography, which sparked the work of the Subaltern Studies Collective, also lurks within the project that is urban studies. Bunnell and Maringanti (2010) have recently designated this tendency as ‘metrocentricity’. Subaltern urbanism is an important intervention in such conditions for knowledge. It calls into question the ‘sanctioned ignorance’ that attends metrocentricity. Subaltern urbanism is also a politics of recognition, one that seeks to make visible what McFarlane (2008: 341) has called ‘urban shadows’, ‘spaces at the edge of urban theory’. This is the slum as theory; this is the periphery as theory.

But in this article I have also called for a disruption of the ontological and topological readings of subalternity, those that celebrate the habitus of ‘slumdog cities’ and assign unique political agency to the mass of urban subalterns. For this I have turned to four emergent concepts — peripheries, urban informality, zones of exception, and gray spaces — that together present the possibility of a different valence of Southern theory. Each of these concepts is, in Mouffe's (2000: 12) sense, a ‘constitutive outside’, an outside that by being inside introduces a ‘radical undecidability’ to the analysis of urbanism. Each then is — to borrow a term from both Mouffe (1993) and Gregory (2010a) — a ‘vanishing point’. For Mouffe (1993: 85), a ‘vanishing point’ is ‘something to which we must constantly refer, but that which can never be reached’. This perhaps is the most productive aspect of the analytic concept of subaltern. With this in mind, the four emergent concepts presented here can be read as vanishing points at the limits of itineraries of recognition.


  1. An earlier version of this article was presented as the IJURR lecture at the 2009 meeting of the Association of American Geographers. I wish to thank IJURR, and especially Roger Keil, for the invitation. Gautam Bhan provided valuable research assistance for this project. The IJURR reviewers helped clarify the concept of subaltern urbanism and the purpose of this article.