When Housing Policies Are Ethnically Targeted: Struggles, Conflicts and Contentions for a “Possible City”

Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible.


Introduction
At the end of December 2014, the former homeless shelter of via Traves, on the outskirts of Turin, was re-inaugurated as a temporary collective housing structure for deprived families living "in emergency situations", within the framework of a local rehousing project called La Città Possibile (The Possible City), funded by the Italian Ministry of Interior. Since then, the building has hosted Romanian Roma families who had been recently evicted from their makeshift barracks on the banks of the Stura River as well as homeless Italian families. This cohabitation between Roma and non-Roma families was not originally planned in the project that was instead strictly addressed to Roma beneficiaries. Indeed, project specifications only envisaged, at most, the possibility of a residential mix of Roma families and "young Italian people interested in experiencing new forms of cohabitation" (Città di Torino 2013: 24). However, the presence of Italian families resulted from their "contentious politics" (Ataç et al. 2016): they occupied the newly refurbished structure on the night of 01 December, just a few weeks before the scheduled inauguration, with the support of militants of far-right groups and the local councillor of the right-wing party Fratelli d'Italia. Their action was both a response to an unsolved housing need and a protest against the assignment of a public structure to "non-Italian" familiesand, what is more, of Roma ethnicity.
The structure, located in the suburban district Le Vallette, was inaugurated a few days before Christmas. 1 However, few elements marked a Christmas atmosphere; among them, red Santa Claus hats worn by some of the prostitutes working in that street. It was a cold, sunny day, and the local NGO AIZO, in charge of the management of the housing structure, had organised some outdoor leisure activities for the children who were to live in the building. Some local authorities attended the ceremony, among them the Councillor for Social Services and the Bishop of the city. Next to them, a tired old man dressed as a clown had just taken off his wig and lit a cigarette. After having entertained the children with some numbers, it was the turn of AIZO's social workers, who activated competitive games with the Italian and Romanian children. Since the toys collected for the occasion were not enough to go around, they organised prize competitions in which only those who came in first place won the toy at stake. In fact, this was an unintended metaphor of the competitive mechanisms at the basis of The Possible City. Around the group of children playing with the AIZO volunteers, each parent supported his or her own child. This grotesque scene, in which each child was incited to come first in order to obtain his or her Christmas present, strengthened the sense of competition against immigrants that moved the Italian families to occupy the building just a few weeks before.
This chapter is based on two ethnographic investigations on Roma housing policies carried out in Turin between 2011 and 2016. The first one focuses on the mismatch between residential strategies of Roma camp dwellers and Turinese local housing policies addressed to Roma (Vergnano 2015), while the second illustrates the impact on Roma families of different housing projects, beyond Roma-camps, promoted by the local administration (Manca 2017). Building on two episodes of public conflict, the aim of this chapter is to highlight the limits of selective, ethnically-targeted housing policies, as well as the reactions these policies may produce in a context of general 'housing crisis', increasingly affecting wider cohorts of population excluded from the housing and labour markets.
We move from the conception of the camp and camp-like institutions as political spaces constituting new political subjectivities, where the acts of citizenship (Isin and Nielsen 2008) performed not only by dwellers, but also by neighbours and activists, trigger new emerging processes of subjectivation and question selective programs of rehousing in a context of welfare cuts and housing crisis. Participant observation has been carried out inside and outside the slum of Lungo Stura Lazio (hereinafter "the slum"), inhabited by Romanian families (mostly of Roma ethnicity) from 2011 until 2015, before and during its dismantling. In addition, we carried out participant observation in two collective housing structures where some of the slum dwellers were rehoused, as well as in a building squatted for 2 weeks by Roma families excluded, for different reasons, from The Possible City rehousing project. Between 2012 and 2015, we collected, recorded and transcribed 58 in-depth interviews with slum dwellers (30 men and 28 women, both those benefitting and those excluded from the rehousing project, all Italian speaking), 17 with civil servants from the Municipal Social Services and Municipal Police, 13 with social workers from private NGOs and 4 with civil-society actors from neighbourhood associations supporting the clearance of the slum.
The pioneering nature of Turin concerning Roma housing policies makes this city an especially suitable choice for detecting new governance models of Roma groups in the Italian context. Indeed, Turin was the first Italian city where an official camp was created on an ethnic basis as a housing solution for a group of Sinti (Italian Roma) families, in 1978. The camp was managed by a special municipal office called 'Nomad Office.' 2 Several years later, in 1982, a special police patrol was created as part of the Municipal Police Department: the "Nomad Patrol". Currently, there are four official Roma camps in Turinese suburbs and several informal settlements or slums, also popularly known as "Roma camps". During the 1980s and 1990s, the camp model introduced in Turin was reproduced in several large and medium sized Italian cities (see Sigona 2002Sigona , 2005a: in fact, in 2000, the European Roma Rights Centre defined Italy as a "campland" (ERRC 2000). In 2008, the Italian government declared a 'Roma emergency' in relation to the settlements of "nomad communities" in different Italian cities (Presidenza del Consiglio dei Ministri 2008), motivated by "public security reasons"; however, at the end of 2011, the Council of the State declared the emergency illegitimate and, in 2012, the new National Strategy for Roma Inclusion up to 2020 (NSRI) was officially approved. Starting in 2013, Turin was the first Italian city to implement a rehousing project addressing Roma families living in makeshift settlements, under the inclusive framework of the NSRI, thus once again confirming its pioneering character in the local governance of Roma groups.

Going Beyond Camps in the "Crisis" Context
The Italian "Roma question" is a longstanding issue which can be defined with more precision as a 'camp question' or, in other words, a complex set of ethnopolicies and social conflicts resulting in and from the spatial segregation of poor Roma families.
The origin of the camp as a housing solution for underprivileged families of Roma ethnicity must be sought in the late 1970s, when prohibitions against the settlement of (mainly) Italian Sinti and Roma in urban and suburban spaces were proliferating in many Italian municipalities (Sigona 2011). The planning and equipping of camps in Italy resulted from the heated debate, beginning in the 1960s, concerning the conflictive presence -at the local level -of mobile stigmatised groups, namely of Sinti and Roma ethnicities, both native and immigrant. Compared with previous municipal bans against "nomad" settlers, the first regional laws of the 1980s for the "protection of Romani culture" (which established the need for creating specific "settling areas") were considered a necessary step by civil and religious organisations for Roma rights. However, camping areas (or camps) soon resulted in marginalised and segregated spaces, which institutionalised "nomadism" as an intrinsic dimension of "Romani culture" (and therefore, paradoxically, permanent temporariness as an intrinsic dimension of Roma housing). Indeed, following Agamben's formulation, Piasere (2006) defined the conditions of Roma camp dwellers in Italy as a form of "inclusion by exclusion".
During the 1980s and 1990s, camps became a very popular residential model for both native and immigrant Roma in several Italian cities. The number of Roma living in camps has increased over the last 20 years, reaching approximately 40,000 (Dalla Zuanna 2013) as the conditions of these camps have continuously worsened, negatively impacting the social and territorial stigma of this already-stigmatised ethnic minority. Political manipulation of such conflicts for electoral purposes culminated in 2008 with the declaration of a "state of emergency" regarding Roma settlements by the Italian government in three large cities: Milan, Rome and Naples. The declaration of emergency provided local prefects with special powers and ministerial funds, 3 mainly used for clearing out informal settlements and building new camps in peripheral urban areas (Daniele 2011;Pasta et al. 2016;Vitale 2008), and was extended to Turin and Venice in 2010.
Since the emergency was declared illegitimate at the end of 2011, the NSRI was approved in 2012, under the EU framework for National Roma Integration Strategies until 2020. The NSRI recognises the substantial failure of a four-decade camps policy and promotes the social inclusion of Romani groups through the design of policies following the controversial common EU principle of an "explicit but not exclusive approach" (European Commission 2012). This sudden switch from "Roma emergency" to "Roma integration" is connected to the need to adapt to European directives and offered unexpected support to different voices from civil society and third sector organisations requesting a new approach to Roma housing. However, it does not result in direct improvement of national and local policies, as the Turinese example of The Possible City illustrates. Indeed, until today, the NRSI appears to be just a formal plan, largely overlooked and mainly used for rhetorical answers and the creation of consensus (Bontempelli 2014).
As a consequence of the NSRI's recognition of the failure of the "camp policy", Roma become an additional category among an increasing number of others -the poor, the young, single-parent families, migrants -that express specific housing needs and have no access to affordable accommodation in the private real estate market or public housing. Indeed, the historic matter of housing access for pauperised groups is now disputed in a contemporary scenario, quite different from the context in which "nomad camps" were conceived at the end of the 1970s. Housing deprivation no longer exclusively concerns traditional low-income families: it also involves new, numerous and heterogeneous population cohorts and groups, previously not affected by this problem (Tosi 2008;Governa and Saccomanni 2009). It is important to note that in 2012 Turin achieved the bitter record of the Italian "capital of evictions", with 4000 eviction notices affecting families with rent arrears, and a long waiting list of 9000 requests for allocation of social housing (Redattore Sociale 2013; Minister of Interior 2013). The same record was once again achieved in 2016, with 3388 executions of eviction orders (Minister of Interior 2017).
The emergence of new forms of housing deprivation is connected to wider dynamics of transition towards a post-industrial and neoliberal city. Indeed, in the last decades, rental deregulation and the reduction of public housing stock (Ascoli and Sgritta 2016) has had a deep impact on the housing supply. The radical restructuring of social housing provision, justified through a hegemonic discourse aimed at undermining the legitimacy of the social housing sector and promoting homeownership, was accompanied by a process of financialising housing that culminated in a global crisis of the housing market in 2008 (Aalbers 2015). Even if in Italy the ratio of household mortgage debt to net disposable income is lower than in other European countries (García Lamarca and Kaika 2016), rising levels of unemployment weigh on the capacity of an increasing number of households to afford their own housing.
In this context, those economically excluded from access to the labour and housing markets have been subjected to specific forms of governance. Recent studies on developments in housing policies (Blessing 2016) point to the emergence of a new marketised and privately financed model of "affordable rental" on a time-limited basis, especially addressed to low-and moderate-income groups with income growth potential in the medium term, such as students or young couples. "Affordable rental may thus be understood as a strategy of "backing winners'", based on the notion of 'economic opportunity' rather than charity, or any universal right to housing" (2016: 18). These developments support a broader agenda of neoliberal social service reforms, according to a new model of citizenship based on a contractualisation of the relationship between the individual and the state, whereby the entitlement to rights is conditional and can be earned after proving one's activation and responsibleness (Clarke 2005;van Baar 2012;Suvarierol and Kirk 2014). In addition, the retrenchment of public sector investments has been partially replaced with private investments on the basis of different welfare mix formulations and decentralisation dynamics towards the local -especially the urban -level.
While recent housing policies aimed at providing temporary "affordable rental" are mainly addressed to low-and moderate-income people, what happens to those vulnerable social groups, not even able to afford an "affordable rental"? As Sassen (2014) has convincingly shown, wider dynamics of expulsions are characterising the current extractive economic logic in many fields usually analysed as separated, from the social realm to the biosphere. "Anything or anybody, whether a law or civic effort, that gets in the way of profit risks being pushed aside-expelled. This switch in economic logics is one major systematic trend not fully captured in current explanations" (ibid.: 213). In this sense, van Baar (2016) introduced the notion of "evictability" to define the permanent possibility of being evicted from a sheltering place that certain groups of 'border crossers' (not only migrants, but also those national citizens perceived as an internal danger, such as homeless, Muslims, Roma or inhabitants of urban areas experiencing gentrification processes) ordinarily experience. This is the context in which the NSRI is being implemented. In a structural context of "housing crisis", however, ethnically targeted housing policies manifest their intrinsic paradoxes between the enlargement of housing demand and the contraction of supply on limited categories. Therefore, both the conflicts among pauperised social groups (such as those that took place in the housing accommodation of via Traves) and the political concerns on electoral results contribute to shape and influence the various levels of implementation of local policies for Roma housing. Moreover, these kinds of ethno-policies trigger new emerging processes of subjectivation among camp dwellers and other groups excluded from housing access. Indeed, as many scholars have argued, Roma camps (and their geographical and social proximities, we may add) are not only "spaces of exception" but also contested spaces of citizenship where processes of political subjectivation take place (Maestri 2017;Maestri and Huges 2017). Roma dwellers, and other non-Roma homeless, are not only passive beneficiaries of exclusionary or allegedly inclusive measures, but they are also -and similar to any other citizen -autonomous subjects, who make decisions about their necessities and preferences between different housing solution and opportunities.
In fact, the individual freedom of being and acting in order to achieve personal objectives can be considered a fundamental aspect for achieving social inclusion and individual well-being (Sen 1985). According to Sen's "capability approach", well-being is intended as agents' capacity to promote their own aims through a production process of real opportunities to choose, with an active transformation of choices and resources into results. In this sense, social actors cannot be considered as simple recipients or beneficiaries of social intervention: on the contrary, they are active parts of the process of evaluation, choice, and construction of their own personal pathway of inclusion.
In fact, the ethnic and selective character of the alleged "inclusive" Municipal intervention on Roma housing has been contested by different social groups, both Roma and non-Roma, even from opposite political areas. The dynamics and purposes of their contestations are illustrated in the following paragraphs.

Seeking the Right to Be Included in the Possible City
How did Roma slum dwellers arrive in via Traves? As mentioned, from December 2013 until November 2015, a massive re-housing project promoted by the Municipality of Turin, La Città Possibile, was implemented in the slum of Lungo Stura Lazio. This settlement, the largest in the city, was defined as an informal Roma camp by local authorities and was the home of approximately 1000 Romanian citizens mostly of Roma ethnicity.
The project was drafted by the local Prefecture and Municipality between 2010 and 2011, during the so-called "Roma emergency". The declaration of illegitimacy by the State Council in 2011 put the city of Turin in a difficult situation, since the ministerial funds assigned to the management of the "emergency" (approximately 5 million euros) had to be returned to the Ministry of Interior. After the approval of the more inclusive normative framework of the NSRI, the same ministerial funds were re-assigned to the city of Turin. The Possible City was aimed at reducing city slums by moving their inhabitants through massive rehousing and 'voluntary' returns to Romania.
A first selection phase of the potential beneficiaries of rehousing took place prior to the implementation of the project (Spring 2013), when a series of expulsions and arrests within the slum was conducted. In that period, the police conducted massive controls inside the settlement: 298 controls in 6 months, approximately 2 per dayoften early in the morning and blocking all access routes. The controls permitted the identification of 1057 people. Seventeen were arrested, 52 had charges pressed against them and 29 were affected by expulsion orders (Prefettura di Torino 2015). Of utmost importance, the police operations were extensively covered by the local press, thus giving rise to a process of criminalisation of the inhabitants of the slum. As a former employee of the Nomad Office told us, They carried out hundreds of police controls in the months before the project, identifying the people living in the slum. Hundreds of censuses. We have no way of knowing, but everything seems to indicate that at the end of that period, having hundreds of different census, they asked themselves: "Which census can we choose for implementing the project?", and they obviously answered "This one!". I mean, the smallest one [the census with fewer people]. [emphasis added] (18/10/2013) According to the official census, there were 800 slum dwellers. However, police data reveals that more than 1000 people were identified in the slum before the implementation of The Possible City. According to our personal observations, we consider the latter data more adjusted to reality. The hypothesis of the interviewed former staff of the Nomad Office partly explains the arbitrary nature of the exclusion of dozens of dwellers from The Possible City: they were missing in the final census simply because they were not in the slum that day. The second selection step was the evaluation by the Prefecture and Municipality of personal legal requirements, which resulted in a "list of beneficiaries" that remained secret (and only in part negotiated with NGO workers) over the entire duration of the project.
The inclusion project was funded with €2,465,859 from the ministerial funds originally allocated for the "Roma emergency" in Turin. Local NGOs participated in a call for tender aimed at implementing the project. One of their main tasks was to contribute to providing private apartments to selected families and promoting "voluntary returns" to Romania. Indeed, they directly managed rental contracts and payments. Rehoused families had to pay rent through growing quotas until the end of the project, while the remaining part was to be paid with project funds. Since the clearance of the slum was carried out progressively throughout the whole project, the families that were rehoused first benefited from a longer period of subsidies compared with those who were rehoused at the end, when the project was almost finished. Once the project was finished, the families were expected to be able to meet the rent of a house at market prices with their own means. 4 On the other hand, families choosing to voluntarily return to Romania received the ticket for the trip and were involved in a 6-month project of labour reintegration in the country of origin, managed by local non-profit organisations selected as partners of The Possible City.
Rehoused families were mostly moved to collective housing (such as the one of via Traves), and some individual apartments in urban and suburban areas. Indeed, difficulties in finding affordable housing were solved by third sector organisations by mean of concentrating dozens of families in the same building, or using private apartments of charity organisations involved in The Possible City as project partners. Collective housing units were composed of minimal spaces -single or double rooms per family -generally equipped with fundamental services (sometimes without utilities such as heating systems, private spaces, electric systems compliant with legal standards), inhabited exclusively by Roma families and managed by third sector social workers for daily activities. It is worthy to note that the largest of such collective housings, the building of corso Vigevano, did not comply with legal habitability criteria (Italiano 2016).
The beneficiaries of The Possible City were requested to sign an "inclusion agreement" with a series of commitments, including the respect of Italian law and the specific rules of the project. The agreement was called patto di emersione, literally, "pact of emergence", referring to the need to "emerge" from the informal sphere of housing and labour to the formal one. Not respecting the agreement was punishable with expulsion from the project -the sanction affecting the entire family. As part of a symbolic process of re-education to the social norms of civilised society, each selected family had to demolish their shack and dispose of the building material as a step prior to being rehoused, while the Red Cross guarded the evicted areas in order to avoid reoccupation by former dwellers excluded from the project. A specific budget of €457,985 was allocated for this surveillance activity.
The clearance of the slum took place progressively from December 2013 to November 2015. While being included in the beneficiary list of The Possible City entailed two possibilities (the acceptance of rehousing, on the one hand, or the acceptance of return, on the other), exclusion was simply tantamount to being evicted without alternatives. In spite of the decisive impact of these different possibilities on the "inclusion process" of each slum dweller, the criteria according to which families were either selected or excluded, on one side, or (in case of having been selected) proposed to either access alternative housing or return to Romania, on the other, were neither transparent or questionable for anyone but the project's heads. NGO workers interviewed during the fieldwork provided different explanations. Some among them recognised that the only criteria employed was the attempt to "match" a family's characteristics with the housing supply available in a given moment.
The families who were selected as beneficiaries but refused to return to Romania or to pay high rents for housing, had no possibility of negotiating the terms of their inclusion in the project and were simply considered "undeserving". Therefore, most of the weakest and poorest families were excluded from the project.
In any case, the lack of transparency, participation and communication contributed to the creation of a tense atmosphere of rivalry between the families during the implementation of the project. As mentioned, their moment of accessing the project would significantly influence the opportunities that were offered to them: families living in the area that was cleared first benefited from a longer rent allowance than the last families to be rehoused; in addition, the last inhabitants removed from the camp, only received proposals for voluntary return to Romania, as at the final stage of the project alternative housing solutions were no longer available. What this meant, on the ground, was that as each day passed, there was a mounting anxiety and anguish felt by those remaining in the slum.
The area was cleared progressively, the shacks demolished and the dwellers were either rehoused, sent to Romania or evicted. At the end of the project, 643 individuals out of 1000 inhabitants were included in the project: according to the final report, 97 "inclusion agreements" were signed for rehousing (equivalent to 378 individuals) and 85 "return agreements" (equivalent to 265 individuals). However, after 5 months from the end of the project 23 families were still relying on rental support, 25 were in single apartments rented on the private market, 14 were in Romania 5 and 13 were squatting the same temporary housings in which they had been rehoused, as they were not able to afford the payment of the full rent. After the end of the project in November 2015, the Municipality extended another year of economic support for rent payment only for 24 families, but after this extra period the Municipality stopped all economic contribution, maintaining just a monitoring of the families from November 2016. At that moment, 1 year after the end of the project, only six of these families were able to afford the rent of their houses on the private housing market. Among the others, some were defaulting debtors in the private housing market, and some were hosted in free accommodations by local NGOs and other charity associations. 6 Their difficulties were, again, the costs of housing payments on the private real estate market, and the lack of regular jobs for economic autonomy. The long-term sustainability of this "emergence" process was, in fact, insecure and questionable.
The remaining 350 dwellers were simply evicted without alternative solutions. Most of them moved to precarious shacks in another city slum. They were defined "not adapted" for the project; however, neither selection criteria were clear for the dwellers and the external observers, nor were they communicated to the families.

Contesting the Ethnic Character of the Project: The Occupation of via Traves
In order to rehouse about 600 beneficiaries, according to project specifications, the Municipality of Turin only provided one public building: the former 24-bed homeless shelter on via Traves. The rest of the housing solutions were found by involved NGOs in the private market. The former homeless shelter was a sheet-metal container, located in the suburbs of Le Vallette, on the outskirts of Turin. Le Vallette is a working-class neighbourhood, traditionally isolated from the rest of the city and strongly stigmatised due to a deep-rooted perception of insecurity and criminality, in which the current unemployment rate is among the highest in Turin, according to the last available census: 33% among adults, 46% among young people (Caprioglio 2011). Even if between the 1970s and 1990s, most of the public housing assets of the neighbourhood were sold very cheaply to their tenants, not all the households have been able to access ownership of their houses. As a result, some of the poorest families are currently facing the impossibility of affording rent.
According to tender specifications of The Possible City, the refurbishment of the shelter of via Traves had to be participative: Roma families involved in the rehousing process had to directly participate in the renovation; moreover, as already mentioned, tender specifications included the possibility of a residential mix of Roma families and "young Italian people interested in experiencing new forms of cohabitation" (Città di Torino 2013: 24). 7 However, upon implementing the project, the involved NGOs operated in a different way. The refurbishment of the building was entrusted to external specialised companies, financed with approximately €100,000 (part of the budget of the CP project). No dweller of the slum took part in the refurbishment.
As already mentioned, on the night of 30 November, 3 weeks before the inauguration of the renovated transitional shelter, four families of Le Vallette, evicted from their houses, squatted the building. As the coordinator of the project explained, They did it in a crucial moment, because [the refurbishment] was just finished. Everything was ready, except for the furniture. The heating system had been inaugurated on Friday, the installations had been tested on Thursday, the only things missing were the baseboards. So it was a very special moment, because if they would have entered in the previous week they would have found a very different situation as for installations and facilities. And if they would have decided to enter the next weekend, they would not have been able to, because they would have found Roma families inside (13/12/2014). Therefore, on Monday morning, when the social worker appointed to open the site arrived, he found the external gate locked with a chain, and two Italian flags hanging on both sides of the gate. The people inside were declaring through the window their intention to not let anybody in: "The police have already been notified, we won't let anybody in", "We have occupied this place because we are Italian families and we don't have anywhere to go".
As mentioned, a few militants of the far-right movement Gioventú Nazionale, together with a far-right municipal councillor of the political party Fratelli d'Italia quickly arrived on the scene, claiming housing rights for "Italian families", thus appropriating the action within a far-right discourse. As an NGO worker told us, The intention of strategically hitting right there… this is political stuff. It doesn't belong to these families. I know them, they are good people, but they are really vulnerable, really in difficulty. I mean, they don't think in a strategic way, about going in a place rather than another. They have a problem of daily survival, every place is good for them. I think that they found somebody who told them "look, that is a good place for many reasons, go there, go on that day". I don't think that it was their target, of their own will. Their aim was just to find shelter. These families had been sleeping in a tent for one month, and then another month in a car, with children between 5 and 9 years old… Even our Roma families, when they knew their situation, said that the decision of squatting that building was right. They were dead tired and exhausted (15/12/2014).
It is noteworthy that the assignment of shelter to Roma families was surprisingly criticised by one of the same NGOs taking part in The Possible City project, "Terra del Fuoco". About 1 month before the occupation, the NGO sent an open letter to the mayor of the city, strongly criticising the ethnic criterion of assignment of housing and highlighting the danger of creating social tensions between pauperised ethnic groups: For public opinion, providing housing with public money is equivalent to allocating social housing without formal requirements. […] Creating competition between Italian families in difficulty and Roma families involved in the project is a source of social tension' (President of Terra Del Fuoco, reported by Guccione in La Stampa, 25/10/2014).
The potentially discriminatory character of local Roma housing policies was especially conflictive in the context of the "housing crisis" affecting Turin, the Italian "eviction capital". As the mayor of Turin admitted in a public meeting about the final outcomes of The Possible City, "it is difficult to present this project to citizens, because it is unpopular". 8 Indeed, the claim of the squatting families of via Traves was immediately recognised as legitimate by public authorities and responsible NGOs. Immediately after the occupation, the Municipality entrusted AIZO, one of the NGOs involved in The Possible City, with mediation tasks. Two of the four squatting families were involved in the same "pact of emergence" proposed to selected Roma families, by which NGOs and rehoused families undertook a series of mutual commitments, including a rent allowance until the end of the project scheduled for December 2015 (starting from this date the two families were expected, like the others, to be able to pay the entire rent on their own). The other two families chose a different solution for their housing needs, since they moved to another building on the outskirts of the city, already squatted by other militants of far-right groups.

Contesting the Selective Character of the Project: The Occupation of via Asti
Neither was the occupation of via Traves the only protest action carried out during the implementation of The Possible City, nor were the four families of Le Vallette, together with far-right groups, the only protesters. Indeed, at the end of 2015, a few days before the end of The Possible City project, 26 Roma families, supported by some local anarchist and anti-racist activists, entered the former military base "La Marmora" in via Asti, located in an affluent neighbourhood close to the city centre, and squatted some empty buildings with the intention of making their home there. Some of these families were excluded from The Possible City (and therefore were going to be evicted from the slum without alternatives), others were temporarily rehoused in some apartments on corso Vigevano as well as in via Traves, but -now that the project was almost finished -they were unable to afford the full rent.
Nevertheless, these families were not the only and first squatters of the abandoned site: indeed, the structure had already been occupied, 6 months earlier by a local NGO claiming the spaces for social and cultural events. And it was not just any NGO, it was "Terra del Fuoco", one of the organisations involved in the implementation of The Possible City, the same organisation that criticised, 1 year earlier, the ethnic criterion of the allocation of housing at the basis of the project. This news received great attention in local media (Rocci 2015;Caracciolo 2015), thus giving visibility to the families and also the opportunity to speak publicly about their reasons and requests. The Municipality and the NGOs involved in the dismantling of the slum were thus compelled to listen to their voices, and their message was strongly critical against them. Among many empty buildings in the city, that group of former slum dwellers chose to enter the only one where they could meet the social workers considered responsible for their eviction. This act, and the following declarations, were visible forms of political acting of usually invisibilised people.
The occupation started with a meeting between the two groups of squatters (the Roma group and Terra del Fuoco group), aimed at finding a common solution in spite of their different objectives and needs. Roma families complained about the destruction of their shacks and the forthcoming evictions from the buildings in which they had been rehoused. They highlighted that, even if their shacks were precarious and unsafe, these were the only homes they had, and they lost them either for not being selected as deserving beneficiaries or, if selected, for not being able to pay rent without regular jobs. The promised alternative homes and the professed inclusion project were, in fact, more precarious than their old shacks. As Marianna wrote in one of her field notes: When I ask George if they know the other occupants, he smiles and answers "do you know Rossella? And do you know where she lives?" I reply that I know Rossella, that she is a social worker for The Possible City project and she lives in Turin. George becomes serious and says, "she lives here, on the other side of the courtyard. She should have found a house for us, but she didn't do it and now she doesn't want us to stay here. They say that we have to leave because there's no water and electricity here, but we lived for years on the riverside" (5.11.2015).
In fact, this was the squat of a squat, a political claim that took the place of a previous claim, a claiming by a group of "deviant squatters" within and against that of a group of "moral squatters" (Piemontese 2016). Indeed, the Roma were strongly critical of the social workers: they claimed their right to a home and contested The Possible City, as they wrote in an open letter 9 addressed to the neighbours. On their side, the "moral" squatters of the NGO appeared embarrassed and uncomfortable with their new Roma neighbours. "We are not the owners of this building -they declared in the meeting -and we are not the ones who decide who has the right to stay here; we can help you, but we are not the owners". "Here we are, we don't have another place to go", a Roma woman replied (Tessarin 2015).
During their stay, the families worked hard to make the spaces comfortable with furniture and domestic objects, also for children and the elderly. They opened it to all external people curious about their situation and supporting their claim. The common space of the structure, a large central courtyard, was physically separated between the two occupant groups, and used for different activities during the day and the evening. The Roma hoped that the presence of the "moral squatters" would protect them from eviction. But this expectation was disappointed: 2 weeks after their arrival, Roma families were forcibly evicted together with the NGO squatters by a large police deployment (Manca 2015). They had very little time to gather their objects and leave the building, and they decided to organise a demonstration in the city centre to express their disappointment and their fear about the future. In the following weeks, the Roma received solidarity and attention from different parts of civil society, but nothing changed in administrative policies. When The Possible City project ended, no solution was provided for most of those families still living in the slum, whose shacks were simply demolished without providing any alternative accommodation for them.

Conclusions
Fieldwork research illustrates the crucial character of the local dimension as a fundamental arena for policies, in which the limits of politics, available resources and different implementation levels directly face local realities and resistances. It is not only recipient and beneficiary potentialities and needs, but also social tensions between different groups of stakeholders (Roma and non-Roma homeless people, far-right militants, anti-racist activists and NGOs, slum dwellers and local authorities) that emerge at this level, and are intensified, in a particularly critical economic phase, by insufficient political responses.
The Possible City is an example of how budgetary cuts affecting social policies and social services, externalisation and processes of fragmentation of the social demand through specific (ethnic) targeting may undermine the inclusive effects of social intervention and their sustainability (Clarke 2005). The contracting of social groups involved, on the basis of specific characteristics (such as ethnicity) is connected to wider logics of individualisation and contractualisation of social protection devices, both for people involvement and governance.
The connection between austerity policies and the dismantling of the welfare state, on one side, and the recognition of the need to promote more 'normalised' housing policies for Roma, on the other, has turned the Roma issue into a sort of taboo, an increasingly unpopular argument for local administrations. The emergency approach here plays a fundamental role, since it modifies quality standards through ethnic definitions (Harrison 1998) and reduces generalist services by means of separation from general assistance (Le Méner and Oppenchaim 2012).
In this context, implementing a rehousing project for slum dwellers of Roma ethnicity requires a series of rhetorical justifications, a discourse aimed at dividing "deserving" and "undeserving" beneficiaries according to morally defined criteria (having a criminal record or stable employment) rather than more "neutral", administrative ones. This kind of framing fabricates "the Roma" as a special object of intervention, by differentiating them from other groups similarly excluded from housing access. The language of crisis, emergency and exception legitimates strongly vertical forms of politics and the demarcation of groups as deviant, unworthy or unfit to be present in public, while it delegitimises some forms of agency and agentive responses. At the same time, the "emergency" or "exception" discourse led to highly specialised forms of humanitarian intervention that may appear incongruous in a social context of rising deprivation and inequalities affecting increasing population cohorts, regardless of ethnicity. Because the shelter of via Traves is located at the centre of a particularly underprivileged area, Le Vallette, the omission of the "indigenous poor" from the rehousing intervention was more obvious. The Possible City, indeed, is a good illustration of the exclusionary effect of the discourse on deservingness, affecting both Roma and non-Roma homeless.
As a collateral effect, vertical, selective and differential policies may catalyse different forms of resistance to/of acts of citizenship (Isin and Nielsen 2008), such as the occupation of via Asti. The controversial character of this kind of "benevolent" social intervention, non-transparent, non-participative and based on previous processes of exclusion, can be easily questioned by excluded social actors for a wide range of reasons and arguments.
As the different accounts from via Traves and via Asti reveal, a competitive situation concerning basic needs can be converted into an opportunity for excluded people to centralise the margins and make visible the invisible through radical acts of citizenship. The recipients of social intervention, even if formally excluded, sought to conquer a space for speech, and in this sense, they managed to occupy a political space denouncing the paradoxical effects of ethnically targeted interventions, retrenchment of public interventions and arbitrary selection processes. At the same time, Roma migrants and other native marginalised groups become political capital and electoral merchandise for local and national powers, as the political character of the two occupations actually received a very different recognition from local authorities: the Italian families were allowed to remain in the former homeless shelter in via Traves they irregularly occupied, while the Roma migrants were quickly and brutally evicted from via Asti, even if a previous occupation of the same building by NGO workers, with much more symbolic and social capital, was tolerated until that moment and was presumed to somehow be able to "protect" and legitimise their presence.
The housing exclusion of impoverished Romanian Roma is not unique or exceptional: it is linked to other forms of exclusion that become stark when differential and selective policies are activated. It powerfully exposes existing social divisions, such as those between native and migrant poor, and temporary alliances (between evicted families and different groups of activists and political militants). These various forms of mobilisation reveal different episodes where excluded groups construct their own housing and biographic paths between constraint and opportunities, starting from their resources and preferences, with different capacities of activation in different conditions. They use public space to express their needs and opinions about personal and public questions, revealing the contradictions of differential welfare housing schemes.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.