Securing the Home: Crime, State Sovereignty and Social Housing in Medellín

In urban Latin America social housing developments have become a strategy to (re)centre territorial sovereignty with the state, by relocating low-income populations from informal settlements controlled by organized criminal groups. Yet criminal groups wield significant influence in new social housing developments, and states’ monopoly on violence continues to be contested. While studies of urban housing in the region have largely disregarded matters of urban security, research on urban security has ignored houses as material agents. Bridging studies of housing and studies of urban security, I promote a broad understanding of security that conceives the cumulative effect of diverse threats to residents’ livelihoods. By doing so, I further develop a material approach to state sovereignty in which a house operates both as a material referent and as an affective “object of desire” (Berlant, 2011) in urban security politics. By promising a secure home, social housing developments materialize the state’s responsibility to protect its citizens. However, the deficient construction and inadequate design of many new homes expose residents to climatic, health and crime-related threats. I conclude by outlining three interrelated sets of question that arise from conceiving houses as politicized materiality. I base my argument on an ethnographic case study of a social housing development in peripheral Medellín.


Introduction
In a global comparison, Latin America shows the highest homicide rates in non-war zones (UNODC, 2013), besides other types of violent crime. In the region's cities, the complex interplay of rapid urbanization (resulting in eroded social ties) paired with frequent police abuse and low police presence (resulting in public security actors' low legitimacy) as well as continuously high rates of social inequality (Davis, 2012;Segura, 2018), broadly explain the high impact that violence has on urban populations (Arias & Tocornal-Montt, 2018). Academic debates have long situated the region's security challenges within discussions about institutional failures to guarantee sovereignty (O'Donnell, 1993), suggesting that this situation should be understood as resulting from a "violent pluralism"  rather than as loss of the monopoly on violence (Pearce, 2010).
In this context, governments have deployed social housing as a crucial tool to strengthen State-citizen relationships in Latin America, yet often undermine rather than supporting attempts to overcome social inequalities (Angotti & Irazábal, 2017). State-subsidized housing programs have improved the livelihoods of low-income populations in Latin American cities for decades, by offering access to property titles and urban services (Gilbert, 2004). They have, however, failed to change people's living situation in more general terms, since they contribute to economic and social exclusion (de Duren, 2018;Rolnik, 2016) due to the remote location of social housing developments on cities' outskirts in Latin American countries (Janoschka & Salinas, 2017), despite at times improving the material conditions of urban populations' homes (de Duren, 2017).
Most Latin American governments have included the United Nations' standards to define "adequate housing" within their urban agendas and developed social housing programs to achieve them as part of the Sustainable Development Goals (Bredenoord et al., 2014;UN Habitat, 2018). As protection from environmental threats and crime, forced eviction and violent dispossession ( UN Habitat, n.d.) are among these standards, the objectives of agendas promoting adequate housing programs have expanded to incorporate urban security concerns. These are particularly relevant in cities with high crime and homicide rates and a long legacy of paramilitary activity. By dedicating housing policies' aims to fulfilling those standards, governments have assumed the responsibility to protect their citizens and thus, albeit implicitly, express the will to claim their monopoly on violence. Thus, social housing developments in Latin American cities from Mexico City, Santiago de Chile, Córdoba, São Paulo, Rio de Janeiro to Bogotá and Medellín have become a strategy to centre territorial sovereignty on the state, moving low-income populations away from informal settlements controlled by organized criminal groups, including drug traffickers and paramilitaries. Yet criminal groups continue to wield significant influence in new social housing developments, and states' monopoly on violence continues to be internally contested.
This article aims to invite (academic) debate about how social housing programs and their implementation interfere with state sovereignty. I conceive of social housing programs as the central means to fulfil the state's constitutional obligation to protect. Colombia's second biggest city, Medellín, is a case in point: it is characterised by, on the one hand, high dynamism and complexity in terms of authority exercised by non-state actors, ranging from paramilitary, guerrilla, militias and organized gangs to relatively splintered criminal groups (Rozema, 2018, p. 539f.). On the other hand, the city government has partially remedied the city's deficit in adequate housing for marginalised populations. In Medellín, ongoing territorial conflicts between criminal actors compromise the state's expression of its monopoly on violence to guarantee protection (Abello Colak & Guarneros Meza, 2014). Despite demobilization accords, first and foremost the "Justice and Peace Law" in 2005, criminal gangs continue controlling people's movement and (economic) activity (Martin & Martin, 2015), thereby undermining what in a European tradition has become the quasi natural character of states (Agnew, 2009). As part of a set of policies that have attempted to transform the city into an inclusive, safe place under the motto of "Social Urbanism", social housing programs have increasingly been designed to interrupt the cycle of forcibly displacing vulnerable populations living at risk (Moncada, 2013;UN Habitat Medellín, 2011). Thereby, resettlement to social housing developments has become the central strategy to fulfil the state's responsibility to protect and to regain territorial sovereignty.
In this article, I study how criminal actors undermine the social housing programmes' assumed primary goal, i.e., to regain state sovereignty by offering low-income populations a safe home in the city. I examine the obstacles posed, by criminal actors' presence, to impede residents' own attempts to improve their livelihoods in state-subsidized, albeit deficient, social housing developments. 1 The article contributes to an anthropological understanding of sovereignty with an ethnographically-informed case study. Similar to what Zeiderman (2016) has shown for Bogotá, in Medellín being a member of a "population at risk" means for some individuals that state policy will protect them by relocation. I will focus on the house as a material site of political contestation rather than a confined object in relation to security policies. Consequently, the house rather than a neighbourhood or an urban area becomes the scale at which to analyse security policies. In order to look at how local gangs interfere with these security policies through the lens of an individual house, I will focus on the material house's role in securing and protecting residents' livelihoods. The next section locates my approach within theories of urban security. These theories, I suggest, should consider the house as a material site where state sovereignty can be adequately addressed. I present state sovereignty as a material process that depends on successful claims to provide a substantively secured home, a term which is informed by Giddens' concept of ontological security. I will then introduce the social housing development Ciudadela Nuevo Occidente (CNO) as a case of the local state's attempt to regain state sovereignty by securing homes, given that since its early existence CNO has been a site of criminal groups' activity. Based on a range of ethnographic materials (drawings, interviews, and observations), gathered during fieldwork in Medellín in early 2018, I will demonstrate how different threats are linked and reinforce each other. Concentrating on securing homes shall suffice to outline the effects that living in proximity to gangs has on residents' livelihoods, shaping their ability to respond to material lacks and geological hazards threatening their new houses. As I will illustrate, effectively securing homes is contingent on deficient constructions, social mistrust, criminal activity, and a fragmented presence of public institutions. I will conclude by outlining three interrelated sets of questions that will advance studies of houses as a material site of contested State sovereignty in Latin American cities.

Securing homes
Many Latin American cities are characterized by a housing deficit (Magalhães, 2016), the presence of organized crime (Koonings & Kruijt, 2013) and low accountability by public security forces. At the same time, it is widely acknowledged that cities play a "critical role in processes of state consolidation, transformation and erosion" (Beall, Goodfellow & Rodgers, 2013). Policies aiming to centralize sovereignty within urban contexts have generally taken either a coercive ("strong-arm") or a distributive ("friendly hand") approach. On the coercive end, scholars have observed militarization of urban space, manifesting a punitive turn in urban security governance (Gledhill, 2015;Graham, 2009;Müller, 2016). On the distributive end, securing home tenure ( UN Habitat, 2015) has been an important way in which governments have strengthened territorial control over marginalized populations and hence gained sovereignty. Tools to do so have included resettlement to social housing projects, legalization of land ownership, and credit schemes for low-income populations (Gelder, 2009). However, the continued authority of organized crime, drug gangs or militias in state-subsidized housing developments in the region's urban peripheries is best described as "fragmented sovereignty", understood as the presence of "multiple, localized, and relatively autonomous cores of power" (Gazit, 2009: p. 1, cited in Davis, 2010. Further developments of this approach to sovereignty, I argue in this section, need to take into account the interplay of diverse material conditions of the homes in social housing developments. In turn, the dynamic of correlating various material layers and elements of the house's role in urban security urges us to grasp security in more encompassing terms and to approach the effects of composite threats. Conceiving the house through its multiple, dynamic material layers furthers debates within Critical Security Studies (Aradau et al., 2015) that have recently claimed to go beyond the "cultural, linguistic and discursive constructedness of security" (Schouten, 2014, p. 24). The house plays a central role not only as a material referent of political or media discourses but within arrangements for security in which social and material elements together compose the conditions of fragmented sovereignty.
Despite this paper's focus on Medellín, it is concerned with the materialization of sovereignty and its territorial contestedness that more generally characterize urban security issues in the region's cities. In Medellín, organized crime, including diverse constellations such as paramilitary self-defence groups and hierarchically organized drug traffickers, control large parts of the city and enter the real estate market to install and control illicit trade networks (McDermott, 2013). In Rio de Janeiro, estimates suggest up to two million residents live under the control of militias, that is, former military or police personnel organized into diverse security and urban services (Exame, 2018;Cano et al., 2012). Militias are particularly present in peripheral low-income settlements and increasingly active in purchasing apartments from Brazil's federal housing program Minha Casa Minha Vida (My House, My Life) (Pública, 2019). In housing developments belonging to Mexico's National Worker Housing Finance Agency (Infonavit) the active, often repressive presence of organized crime has caused 250,000 families to leave their property (Excelsior, 2013) with gangs invading and renting out abandoned houses below market price (El Financiero, 2014). (In)security in these diverse contexts across Latin America is produced through hybrid assemblages in which state agents coexist with armed non-State actors, such as private security industries and organized criminal groups (Jaffe, 2013). In addition, in all these cases the material conditions of the respective developments, houses, and apartments, their isolated location, poor infrastructure and bad construction quality interplay with their exposure to the influence of non-state armed actors. While moving (or being resettled) to a social housing development might increase an individual's or family's security of tenure, the presence of violent actors can push families to again abandon the development (Unruh, 2007). Thus, these examples call us to conceive of social housing developments, in particular houses or apartments within them, as material security concerns and hence material elements through which sovereignty is territorially contested.
Regaining sovereignty in Colombia's major cities has also occupied a central place in the country's ongoing transition to peace among, in general terms, the military, paramilitary groups and guerrillas (Giraldo Ramírez & Preciado Restrepo, 2016). In this urban transition, largely initiated by the post-millennial national policy ("Democratic Security") by conservative former President Álvaro Uribe, "security" is being discursively mobilized as fulfilling the State's obligation to protect (Zeiderman, 2016), particularly in those urban territories where criminal actors directly challenge the State monopoly on urban violence (cf. Davis, 2010). As Zeiderman argues, securing the urban "population at risk" homes also includes protecting them against the climatic threat of heavy rainfall and subsequent landslides. Since the discursive mobilization of security and pacification is directed against gangs (combos) or larger criminal structures (bandas criminales, BACRIMs) new social housing developments are critical territories to manifest State monopoly on urban violence.
For some scholars of urban governance and security, the ongoing "cycle of violence" (MacLean, 2015) in Medellín is a showcase of state institutions' "fragility" (Abello Colak, 2015). From this Weberian perspective, the failure of social housing developments to (re)instate territorial sovereignty and fulfil the state's responsibility to protect its citizens frustrates the normative assumption according to which states ought to centralize the monopoly on violence . In consequence, non-state violence such as extortion, forced recruitment and direct threats to life and health authored by criminal actors, are deviant forms of authority. However, where "the state and the shadow state become intertwined" (Humphrey & Valverde, 2016, p. 171), as in the case of Medellín (Sotomayor, 2017), a more adequate understanding of urban security conceives of it as a contested territorial expression of power to express and enact violence (Hansen & Stepputat, 2006;Das & Poole, 2004), and as the ability to draw and enforce social borders (Diphoorn, 2016). These anthropological studies of sovereignty acknowledge that urban security, from the perspective of a territorially contested monopoly on violence, emerges from onthe-ground interconnectedness of various violent and criminal actors aiming to control economic or political territories. They thereby overcome the normative ideal of state-centred sovereignty territorializing a monopoly on violence.
However, these studies largely disregard the materiality of sovereignty, to which this articles wishes to contribute. Scholars of such entangled bottom-up urban security have largely focused on neighbourhood levels, drawing relatively little attention to the role of a house in securing people's livelihood. Those who base their critical review of statehood and territorial sovereignty on "selfhelp security" (Denyer Willis, 2015) and vigilante groups in marginal settlements  implicitly reduce houses to a confined category and an object of protection. Where discussions of urban security do include issues with home-based security politics, they are mainly located in areas of the urban upper middle classes or elites (Coy, 2006;Caldeira, 2000) which study (with few exceptions) gating among marginalized populations (Plöger, 2011;García Peralta & Hofer, 2006) or note "fortressification of the home" as a trend of wider Euro-American society to set up one's own house as a place of protection against crime and related insecurities (Blandy & Atkinson, 2016).
Pointing to the multi-layered process of securing one's home, and highlighting both its social and material conditions, Hiscock, Kearns, MacIntyre, and Ellaway (2001) have framed tenure security in Anthony Giddens' terms as being inseparably tied to "ontological security". Ontological insecurity emerges when residents live in a state of worried uncertainty regarding their future and in which their home lacks a protective function. The numerous studies of "affordable"/adequate urban housing in the Global South have, however, largely underestimated how diverse dimensions of urban security are co-productive in securing the home. This is surprising, given the recognizably influential role of criminal actors in local governance in most of the regions' cities.
By unpacking the role that securing house and home has in state sovereignty, I wish to further a substantial understanding of urban security. Therefore, I follow up on recent approaches that have convincingly rejected Weberian ideal forms of state-centred sovereignty, by further developing the approach to fragmented sovereignty. While the monopoly on violence in a given territory, in a classical, Western understanding, is primarily associated with the modern nation-state, ethnographic work on policing has relativized the role that state institutions have in exercising the means of violence (Müller, 2016). From the perspective of political anthropology of the state, sovereignty then refers to "a tentative and always emerging form of authority grounded in violence that is performed and designed to generate loyalty, fear, and legitimacy from the neighbourhood to the summit of the State" (Hansen & Stepputat, 2006, p. 297). Where the monopoly on violence is contested, sovereignty loses its stable ground, that is, a clearly demarcated territory and population. Fleshing out the theoretical consequences from a Southern perspective, Comaroff and Comaroff elaborate that, "[h]ere the reach of the state is uneven and the landscape is a palimpsest of contested sovereignties, codes, and jurisdictions -a complex choreography of police and paramilitaries, private and community enforcement, gangs and vigilantes, highwaymen and outlaw armies." (Comaroff & Comaroff, 2006, p. 9). Adding a material perspective to this human-centric understanding of emerging, fragmented sovereignty then means that, "sovereignty appears as a field that is worked on through an agency distributed between human beings and material devices." (Navaro-Yashin, 2012, p. 43). As Navaro-Yashin brilliantly delimits the scope of such a turn towards materiality, power must be located in the relations of human actors that engage with each other, yet "within a terrain of materialities and physical properties" (Navaro-Yashin, 2012, p. 43).
Rethinking the role of social housing within such emerging, material sovereignty then enables us to locate human actors' political engagement in urban security around and in houses. The house is not only an object that needs protection and that protects, but also a material agent in securing the home: The house is the material site, location of emotions and thus, "an object of desire" (Berlant, 2011, p. 23) from which citizens enact protection within the neighbourhood they perceive as their home. In this process of securing the home, in which the house is materially embedded, urban security politics can be studied. In other words, the house itself becomes a material site which needs to be protected against and which protects residents from external threats. Thus, I suggest we need to take a closer look at the house as a material site of security policies. Different forms of matter -different things, e.g., water, bricks, electricity, doors, and property titles -are instrumental to securing urban residents' home, and therewith, their livelihood. They affect what I call substantive home tenure security, taking a perspective inspired by Giddens (1991). This form of security, which includes yet goes beyond legal home ownership, is threatened from both within and outside the home.
In his conceptualization of security, Giddens concentrates on the psychology of an individual self. To address the power relations underlying the "feeling of ontological security" (Giddens, 1991, p. 168) in social environments, however, requires us to account for the material correlations at work in securing a home. For the present context, in which "criminal organizations influence social regulation" (Duncan, 2014, p. 19), this means locating gang activity and urban security politics between the protective effects of housing construction and human efforts to protect the house. In a similar vein, Bennet proposes an alternative model of policies and "thing-power" (2010, p. xxvi) and promotes analyses of the effect of matter -both paralyzing and activating. Such "vitality of matter" (Bennet, 2010, p. 6) assigns crucial importance to the corporal affectivity of human actors and their built, material environment, and in the present sense, to the protective, and to-be-protected house -the twofold function which substantiates home tenure security.
While from a radical "New Materialist" standpoint (Bennett, 2010; Coole & Frost, 2010) matter and inanimate things per se foretell political processes, I follow a more cautious interpretation of materiality's role in actively shaping human-political agency. To condition human agency, emotions are affectively embedded in things; things, e.g., the apartment and its stable walls, can motivate feelings of protection or, otherwise, alertness to potential deterioration and loss. Yet, things are also deliberately included into speech acts and political articulations and demands, e.g., of a secure home. Following a recent critique of Bennett's "vitality of matter" (2010) by Lemke (2018), I assert that political processes (e.g., securing the home) can be fruitfully expanded to include material "thing-power", but must not disregard interrelations with power structures. Whereas social housing developments allegedly materialize the state's responsibility to protect its citizens, their deficient construction and inadequate design expose residents to climatic, health and crime-related threats. Moreover, and locating the analysis of state sovereignty in supposedly "formal" social housing developments, I argue that exploring the ways in which a home generates material and discursive threats can stimulate a rethinking of state sovereignty: as a material effort "to bring the state back in" through housing provision. In other words, I understand state sovereignty as a material process that, in the face of non-state armed groups' parallel efforts to reclaim their share in "criminal en-terprises and local governance" (Arias, 2017), centres on houses as material referents and as affective objects of desire. I wish to outline below the perspective on the materiality of State sovereignty that the lens of houses contributes to these observations. First, I will introduce the case of Ciudadela Nuevo Occidente; second, I will discuss details of property titles as a materialized social relationship to study contested State sovereignty. I will third summarize how gangs manifest their territorial presence and describe how this presence affects residents' attempts to improve their livelihoods.

The case of Ciudadela Nuevo Occidente
Due to this research's interest and subject, my first fieldwork experiences in Ciudadela Nuevo Occidente (CNO) on the slopes of the Aburrá Valley demanded great care: Instead of naming and explaining my interests regarding drug traffickers' and/or militias' activities in CNO to the interviewees (residents) from the beginning, I had to engage in trust-building conversations with residents. Seeking to understand residents' life stories by conducting narrative interviews of three to four hours, instead of directly approaching questions around (feelings of) security, soon demonstrated the necessity to envisage the mutually reinforcing, dynamic constellation of multiple threats jeopardizing CNO dwellers' livelihoods. They are confronted with problems that derive from partially decaying apartments and that, due to the presence of criminal actors and widespread distrust in neighbours, must mostly be resolved individually. The geological and construction-related deficiencies of CNO on residents' livelihoods, which will be described in greater detail in the remainder of this article, are amplified by the presence of said actors. Thus, my initial reluctance to address topics of local security turned from a tactic to protect my interviewees and me into a methodological tool to penetrate the material layers of urban security as they entangle with the presence of non-state authorities.
Criminal groups appropriating apartments in social housing developments, threatening residents and establishing illicit economies is not a specifically Colombian phenomenon -instances are reported from Caracas (Desdelaplaza, 2015), and Rio de Janeiro (O Globo, 2018), among other Latin American cities where gangs have consolidated their hold on entire social housing developments. However, the chosen case is particularly interesting since the statesubsidized CNO megaproject is part of a set of policies that together comprise Medellín's famous "social urbanism". This holistic approach countered prevalent urban violence with a technical, bureaucratic form of governance, reducing the territorial power of competing gangs and their paramilitary meta-structures, and reinstate Medellín citizens' confidence in public institutions. Social urbanism thus framed social housing as a security concern more explicitly than housing policies in other cities of the region. CNO provides "housing solutions" 2 for 100,000 inhabitants, mostly for intra-urban displaced populations. It does so through a public-private partnership among the Housing Institute of Medel-lín (ISVIMED), real estate developers, and loan-granting banks, predominantly for lower-income populations resettled for various reasons. 3 CNO, "aiming at the urban, construction and ownership-related legalization" (Alcaldía, 1997, p. 90) promises access to health care, education, leisure, sports and cultural facilities, retail economies and transport infrastructure. CNO can be read as a governance intervention that implements a "substitutive security governance" (Schuberth, 2016, p. 4), a strategy which aims at driving off residents' support for local gangs by replacing the functions that gangs perform in marginalized urban areas.
CNO sits high above the centre of Medellín, capital of Antioquia, Colombia. It was developed with the objective of urbanizing the western peripheries of Medellín, thereby "legalizing illegal land use and ownership" for the intraurban displaced populations in the area to be developed (Alcaldía, 1997); the area had been inhabited by approximately 2,000 people, living in two invasiones ('informal' squatter settlements). The residential project was geostrategically built in the pathway of one of the strongest flows of illicit objects, that is, drugs and weapons, in a corridor that runs along the city's western periphery (Patiño Villa, 2014, pp. 276-278) and connects it to transnational drug and weapon trafficking (Quijano, 2017). Controlling this corridor has for long been a major strategic objective of the paramilitary group Urabeños and criminal groups pertaining to the Oficina de Envigado. CNO can therefore be considered a material intervention to regain State sovereignty in Medellín.

Matter and security
Análida 4 lives on the fifth floor of a six-floor building in "H", one of the 13 sub-districts of CNO. She visually portrays her memories of the days when she first arrived at "H" and moved into the apartment block, five years earlier, in a drawing. 5 The new apartment, she explains, has a firm structure, unlike the place she inhabited before. She draws friends, benches, the nearby forest, and the adjacent barrio, "a dangerous place. They are the ones who don't like to see us around there." Who "they" are remains unspecified. I ask whether and what particular events have changed her perception of her neighbourhood, and how she feels about them. "Bad." She chooses the blue colour, paints lines. "When it rains, the water gets in". She fills the apartment with blue horizontal lines. "The balcony wall has got fissures, here the water pours out again on the other side." The drawing highlights the threatening, yet also protective condition of the apartment building. She makes it clear that the humidity, caused by insufficiently sealed walls and inadequate window frames, has informed her wish to leave. While the humidity in itself was not the single most destabilizing factor of Análida's house, it does matter in frustrating her expectations that the new apartment would firmly protect her; yet, despite the failed promise to protect her, she holds on to her apartment, evidencing the paralyzing "precariousness" (Berlant, 2011) of her home.
Unlike Análida, Rinna 6 has no property title yet, even five years after moving into the apartment. She has lived through multiple resettlements, from rural Antioquia during heavy paramilitary confrontations, to a peripheral neighbourhood in Medellín, where she lost her house during a major landslide. The constant humidity in the apartment that she received as compensation has let fungus spread all over the wall, causing chronic respiratory health problems. To be allowed to sell, Rinna had to prove fuerza mayor [force majeure, an act of nature beyond her control] (Ley 1537): in her case, severe, chronic health problems. Although she has succeeded in obtaining the doctor's certificate, she still hesitates to sell since this sale, although including ISVIMED as mediator, is in danger of exploitation by criminal organizations: as soon as they learn of the transaction, they (living one floor above her apartment) would not hesitate to extort their share. While Análida visualizes her apartment as a lost place of protection, Rinna endures on a threshold of prolonged uncertainty in which she has reached out for help to various neighbours and human rights organizations to support her formal permission to sell the apartment.
With Jane Bennett, one could argue that Rinna's apartment has the power to shape political affinities, bringing together residents, public institutions, and the Observatory of Passion for CNO. The latter is formed by a group of residents of CNO who were concerned about all kinds of threats to the development, particularly geological hazards, but also the perceived lack of public security actors. Rinna has, in addition to activating said help from (external) groups, become active in another sense: due to the (perceived) threats from the gang, she had to stay, and thus also actively protect her apartment from potential appropriation by the gang. Rather than equalling human and non-human matter in shaping politics, this suggests "to take into account the material preconditions and contexts necessary for political change" (Lemke, 2018, p. 17). Unable and unwilling to continue her trajectory of forced resettlements, Rinna has decided to improve her life circumstances by steeling her will to remain in the apartment, and to oppose the criminal practices of her feared neighbours.
In "H" many families have not received property titles, often being unable to pay the charge (one minimum monthly salary, about 220€), for this "service". Meanwhile, faced with unbearable living situations-construction deficiencies, personal insecurity and threats, besides economic hardships-many families have opted to sell to a local gang. In an interview, a community leader estimates that 150 out of the 983 apartments in "H" have been sold or sublet, handing the asset over to the local gang for a price of about 50 per cent of the original amount. 7 Property titles, a social relationship materialized on paper, become enacted by different human actors to accumulate power in a situation of disputed jurisdiction among the State, ISVIMED, and combo gangs (Moncada, 2016). They further prolong the cycle of forced displacement, with life in jeopardy and looming indebtedness. ISVIMED pressures residents to form an assembly, as a precondition for property title certificates; residents share knowledge about how to finally receive the title, and social leaders base their authority on their particular know-how and personal ties to ISVIMED; gangs take advantage of the not-yet titled ownership by offering prices below market value to those who can no longer endure the situation.
Criminal organizations, via local gangs, are deeply involved in everyday routines: In attempts to stabilize their territorial control, gangs extort protection money from owners of motorcycles and cars, the bus company, shops, and owners of "dangerous" dogs. 8 Residents are obliged to integrate illegal payments to criminal gangs into their everyday life (Secretaría de Seguridad, 2018). Following the lead of private security guards who were initially recruited by ISVIMED to prevent invasions, gang members are using empty apartments as "offices", to sell and store drugs. 9 They are threatening to recruit young male family members, a practice well known from the comunas (Moncada & Lopera, 2017); however, family ties with the gangs are kept secret. 10 The resulting general mistrust in one's neighbours prevents many residents from taking part in the neighbourhood association (junta de acción communal, JAC), and many are not willing to form a homeowners' assembly. As a result, some entire buildings have not been formally handed over to the community and thus remain the property of ISVIMED, which demands the prerequisite of a formally constituted assembly. For that to happen, families need to collectively sign off on the building's conditions, which several have refused to do, because of a second set of quality issues.

Frustrated expectations
Many residents of CNO used to live in high-risk areas, and amidst threats from local gangs, prior to resettlement. In order to process their claim to an apartment, residents agreed to borrow from a private bank, to be paid back in monthly instalments over the next 20 years, in addition to handing their previously inhabited plot over to ISVIMED. A general expectation of progress, in economic terms and regarding future stability through the new home, has been unmet. ISVIMED's promise to residents, to finally own an apartment that nobody would be able to contest or take away; the Administrative Department of Risk and Disaster Management (DAGRD) promise that the new building would be a material improvement compared to the former house; and more generally the sustainable, optimal living conditions, announced by the Municipality's planning department (DAP, 2010) have largely fallen short, or at least are perceived as being so.
CNO features several material shortcomings; additionally, some of the buildings in this "city" are located in high-risk areas (DAGRD, 2016). Facing geological and construction-related risks, members from the National University, Habitat and Planning Department, public officials of the Personería resettlement section, and social leaders from CNO have formed an observatory. Together with hired engineers, they have documented geological hazards, since some of the buildings are prone to landslides; architectural damages, e.g., cracking floors, leaking outer walls; as well as construction mistakes, such as a poor sand-cement mix, or inadequate foundations. While these studies' findings on architecture and geology are convincing, legally enforcing warranties would require a costly judicial process, which residents cannot afford. 11 Fearing other leaders' cooperation with gangs (due to family ties, or under duress) some social leaders have refused to agree to activate the participatory budget. The participatory budget, which could be used to finance the process, has remained untouched since 2012, due to fears that gangs would press for extortions, as soon as they find out that JAC holds such financial resources (which would be unavoidable, due to their direct vicinity or family ties). This fear, in turn, has prevented them from starting any judicial process to demand fulfilment of these promises.
The home, rather than providing a secure, stable socio-material environment, alters the ways residents become active demanders, increasingly towards public institutions. I interpret the new home's agency as a prolongation of the generalized insecurity: Despite resettlement to a social housing development, residents remain in an inescapable condition in which the home changed from a promise of constancy to unceasing socio-material insecurity. In sum, established power relations strongly condition the ways in which matter and things, property titles and the apartment, can be mobilized to politically engender improvements to residents' own livelihoods and social and/or economic benefits.

Securing homes amidst territorially contested sovereignty
In conclusion, I propose to approach territorial sovereignty in Latin American urban peripheries as a material process which centres on securing the home. Such an approach aids investigations of how different threats-from the copresence of criminal actors, material deficiencies, and geological hazardscondition the state's attempts to territorialize sovereignty via social housing programs. My findings suggest that inhabiting an apartment in a social housing development can hold residents in a permanently uncertain livelihood. Facing parallel threats from criminal gangs, climate effects, and inadequate construction, resettled populations' livelihoods remain insecure. The apartments, while fulfilling an instrumental necessity for resettled populations, are at the same time vital in transforming the conflict-ridden relationships among public institutions, non-state armed groups, and residents. The continuously insecure home of "populations at risk" (Zeiderman, 2016) is an expression of a contested material sovereignty and thus a place to study the power relations inherent in urban security politics. For Latin America's marginalized populations, their home is therefore less a place of predictability and constancy (Dupuis & Thorns, 2002) and rather a material site from where and through which diverse actors manifest political claims. In crediting a "thing-power" (cited above) to the house, I propose to conceive of the house as both agent and object in urban security politics. This, I summarize, points us to three interrelated sets of ques-tions for understanding the house as, first, politicized matter which, second, occupies a central role within the dynamic relationship between state and nonstate (armed) actors and therefore, third, can be a vantage point to analyse urban pacification.
Understanding houses in social housing developments as politicized materiality combines concerns about criminal groups' influence on land-use allocation, contract awarding to construction firms, and the location of residential projects in the urban fabric, 12 on the one hand; and the agency role of things and matter in securing the home, on the other hand. Granting the local gang an apartment on the buildings' top levels (a vantage point for the area) becomes part of the bargain between real estate developers and criminal organizations. 13 Generalizing from these observations, matter, things, and deteriorating homes can be conceived of as "material signifiers" (Ivasiuc, forthcoming) and as affective "objects of desire" (Berlant, 2011, p. 23) that influence residents', public institutions', and criminal actors' positions in urban (housing) politics. Instead of conceiving of houses as objects and tools to regain territorial sovereignty, public institutions thus need to consider "material resistances" (Kemmer, forthcoming) and the unintended effects that poor infrastructure, deficient construction, and the specific design of social housing developments have on residents' livelihoods. How do planners, politicians, and criminal actors use homes to strengthen their discursive positions? And how do matter, apartments and constructions affect enduring attempts to change security policies in Latin American urban peripheries?
These ongoing uncertainties also reproduce interrelationships and mutual dependence between state and non-state actors. Due to citizens' limited normative expectation from the state, public institutions are not the naturally prioritized guarantors to submit such claims and demands to. In neighbourhoods where gangs are acting as the local moral authority, called to intervene in cases of interfamily fights/violence, or disturbances by neighbours, i.e., to monitor the neighbourhood and intervene in social conflicts, criminal actors take over crucial functions of maintaining order that are otherwise ascribed to public institutions. In addition, real estate developers, when proposing the budget, include a share of bribery to the gangs for "protecting" the construction site. 14 Generalizing from Medellín motivates us to ask how the political ties of such a materialized "hybrid State" (Jaffe, 2013) can be qualified on a continuum from competition to coproduction in controlling urban peripheries, and securing homes. How does this contested territorial sovereignty affect improvements to resettled populations' livelihoods?
A last set of questions relates to the current historical juncture of postconflict/postaccords Colombia. An "urban peace" is far from being achieved, with 350 gangs active in the larger metropolitan area of Medellín, and two competing major paramilitary conglomerations, which altogether control an estimated 70 per cent of its territory. 15 Due to Medellín's geostrategic location between two oceans and close to Central America, control of its future-planning struc-tures has national and international significance. Criminal organizations' influence in local governance and security production and their ability to drain economic benefit from social housing developments depends on and furthers their ability to effectively engage in the process of securing the home; and thereby to strengthen their discursive positions in urban planning politics. How will this appropriation of social housing developments and of urban planning procedures by organized criminal groups play out in future "urban peace" talks and inform novel governance arrangements between the State and criminal organizations? *** 10 Interview Jhon and Análida (No. 9 and 5). 11 Interview Ángela (No. 10 3).