Abandonment, Agency, Control: Migrants ’ Camps in Ventimiglia

: This article examines the governance mechanisms regulating migrants ’ presence in Ventimiglia — an Italian city on the north-west border — and migrants ’ agency, following the reintroduction of French border controls in 2015. The study looks at different kinds of camps: the formal one run by the Red Cross and the makeshift camps and settlements that sprang up over time. This work contributes to the academic debate on the interplay between migrants and border regimes by arguing two fundamental points. First, migration governance changed the nature of the makeshift settlements, pushing them from being spaces of autonomy, however precarious, to ones characterised more by exclusion and uncertainty. Second, the relationship between formal and informal spaces and migrants ’ agency is ambiguous and changes over time. Using a diachronic perspective and without underestimating migrants ’ agency, this paper denounces the inadequacy, fragility and violence of the border regime which migrants move into and within.


Introduction
Ventimiglia is a small Italian town located on the border with France.Although less well known than other internal "checkpoints" (Tazzioli and Garelli 2020) of the system for controlling migrant flows within the EU, like Calais, Ceuta, or Melilla, this town is one of the symbols of the EU "migration crisis" (De Genova and Tazzioli 2016) which started in 2015.From that year the place became a bottleneck (Katz 2016) for migrants who attempt to move irregularly out of Italy, due to the suspension by France of the Schengen Treaty and the resulting rise of police controls on both sides of the border.From 2015, a new border regime arose in the area, created by different control mechanisms used by the French authorities to push back migrants' irregular arrivals and by the Italian authorities aiming to manage the increased migrant presence in the city (Amigoni et al. 2021a).
This paper investigates the tense and complex interplay between border regimes and migrants (Mezzadra 2010) in Ventimiglia, by looking closely at the evolution of formal and makeshift camps and arrangements in the area between 2015 and 2018.The idea of observing the reconfiguration of camps arose from the "disappearance" in 2018 of the migrants' makeshift camps on the edge of the city that formed the core case study of my project written in 2016.At the end of 2018, when I carried out my fieldwork, the makeshift camps I included in my project no longer existed, following the last of many evictions.In their place are informal living arrangements in the form of small scattered settlements spread over the area between the beach and the course of the river Roja which here runs into the sea.The following brief fieldwork notes show how much is at stake in this place: Mansa [the research intermediary] and I meet two young men on the beach, we guess they are from Pakistan.They are smoking, they are clearly freezing, they are extremely thin and their eyes look lost ... We try again and again but we can't get them to understand that there is a camp."Camp", "camp", we say repeatedly, but for them it is a meaningless noise.(Fieldnotes, 20 November 2018) Two initial observations emerge from this quote.First, there is the very precarious and lonely condition in which migrants often find themselves in Ventimiglia.Second, even to the eyes of a critical migration scholar, the camp appears at least temporarily as a preferable solution to the young men's intolerable situation, which is characterised by extreme poverty and a complete absence of essential services.Expanding upon these personal and empirical observations, the analysis aims to intervene in the debate on border regimes and migrants' agency.It adopts a diachronic perspective, which enables the spatial and functional reconfiguration of the various camps over time to be highlighted, revealing the ambiguous and unpredictable nature of the relationship between migrants' agency and border governance.
My argument is twofold.First, I argue that Italian migration governance changed the nature of the makeshift settlements at the border, pushing them from being in part spaces of autonomy, albeit precarious ones, to becoming spaces characterised more by exclusion and uncertainty.In this regard, the paper connects the transformations of makeshift settlements to concomitant migration policies based on a complex mixture of control and active forms of abandonment (Pinelli 2018), both of which have affected migrants' lives differently.Second, my research shows the ambiguous relationship between formal and informal spaces and migrants' agency.In that respect, it shows how migrants' agency can operate in the formal camp just as it can be "disempowered" in the makeshift temporary arrangements.In order to unpack the changes in makeshift camps and the relationship between them and migrants' agency I use the concept of "active abandonment", understood as the withdrawal of the state's responsibility to provide migrants with minimum levels of safety and support.Following Leshem (2017), abandonment is a complex political technology that has become ever more present in the governance of migration.By looking at the issues at stake through the lens of "active abandonment" and by using a diachronic perspective, the paper offers an original contribution to the debate on migration regimes and migrants' agency (Squire 2017;Strange et al. 2017) and, indirectly, to the vibrant body of research on "camp geographies" (Ek 2006;Levy 2010;Minca 2015).In particular, the diachronic analysis here sheds new critical light on the ways in which regimes reconfigure themselves constantly, reinforcing and combining different strategies against migrants.To take as an analytical focus the detrimental effects of border regimes on people on the move is not to overlook the importance of migrants' (re)actions.Rather, this paper highlights the importance of analysis and practice that take into proper account migrants' increasing precarity, marginality and reduced room for manoeuvre in order to denounce the structural violence of the border regime.
The paper is structured as follows.The first section develops the conceptual and methodological perspectives, emphasising in particular the importance of focusing on active abandonment seen as a government tool which can shed new light on the issues under study.The second section then traces the history of border governance in Ventimiglia by showing how the Italian authorities have employed and refined different control mechanisms to deal with migrants' presence during different phases.In this context, the Red Cross Camp takes on a fundamental role in governing migrants' mobility, in that its opening in 2016 had a significant impact on border governance, as well as on the evolution of the makeshift camps and other arrangements, as discussed in the third section.The fourth section explores migrants' agency both in the Red Cross Camp and in makeshift camps, by revealing the complex and nuanced relationship between "agency" and its constraints.Finally, the concluding section synthesises the key findings and states the political imperative to develop further analysis on abandonment seen as a political technology (Leshem 2017) that is increasingly present in EU migration and border governance.

Looking at Ventimiglia: Theoretical and Methodological Underpinnings Migration Governance and Migrants' Agency
My argument draws on a very lively debate in critical migration studies, focused on the interplay between migrants' actions and migration regimes that are increasingly violent, especially at the borders (Casas-Cortes et al. 2015).In order to engage with this debate, it is worth clarifying what migrants' agency means (Squire 2017), while interrogating how bordering works.
First, as has convincingly been argued by Mezzadra (2010) and Squire (2017), the relationship between migrants' actions and border regimes is based on dynamics of power-resistance "which are not separable but intertwined or coconstitutive" (Squire 2017:260).Thus, as Mainwaring (2016) notes, agency refers not merely to "choices" but to migrants' decision-making, strategies, counterconducts and actions that occur under conditions of constraint within modern nation states.This relational approach replaces an oversimplified conception of people on the move as simply victims of the violence of border regimes or, on the contrary, as heroes who succeed in their fight against it (Walters and L€ uthi 2016).In line with this conceptual lens, this paper emphasises how varied migrants' actions can be.As we show further in later sections, migrants (re)act to the restrictiveness of EU migration policies in different ways related to their particular situations in terms of legal status, social and economic capital, and individual wills (Welfens et al. 2020).Even when migrants do not intend to resist the border regime deliberately, their practices can open up what Jones (2012) calls "spaces of refusal", which are de facto actions against it (e.g., crossing the EU internal border without knowing it is not legally permitted).At other times, as Strange et al. (2017) eloquently put it, migrants can also accept regimes of power or authority while navigating their journey.
For their part, border regimes also act by adopting different strategies (Gill 2010), aimed at managing migrants' mobility (Mountz et al. 2013) by reducing their autonomous "room for manoeuvre" (Mainwaring 2016).On this battleground (Ambrosini 2020), regimes use different mechanisms to block, contain and expel people on the move, which are interconnected and developed according to migrants' actions.In this respect, for instance, Van Isacker (2019) notably foregrounds the relationship between, on the one hand, the evictions of migrantgenerated spaces, which he calls "acts of domicide" or "domicide practices" and, on the other, the creation of official camps.These complementary mechanisms simultaneously combine to discipline migrants' bodies and behaviours through official camps (Abourahme 2015) and undermine migrants' collective social power by destroying makeshift camps (Katz 2015).Within this peculiar forced "formalisation" of camps, indeed, state power bends selected elements of social reproduction of migrant-generated spaces (such as accommodation) to control strategies (De Angelis 2019), while delegating migrants' basic needs to the "humanitarian sector" (NGOs, aid organisations, and private actors).It is no coincidence that for some time the debate on camps has taken on a sort of "binary perspective" (Abourahme 2015), with both perspectives proceeding in parallel.On one side, official camps have been conceptualised as spaces where the suspension of legal rights annihilates the migrants' power to deal with their coercion and violence (Ferris 2007;Loescher and Milner 2004).Conversely, makeshift camps have been seen as "spaces of political birth and hope", because they are "an outcome of the self-initiated spatial actions of their residents/fabricants" (Katz 2015:85).As has been widely debated (Ramadan 2013;Sanyal 2011;Sigona 2015), this binary perspective has not properly taken into account migrants' agency in official camps.However, the representation of makeshift camps and settlements in relation to migrants' agency also needs to be explicitly improved; what is still lacking is a more nuanced analysis which points out and explains the increasing precarity, marginality and structural violence characteristic of the makeshift situation and its causes.
In order to fill this gap and further the debate, this paper looks at the concept of abandonment (Hyndman and Mountz 2007;Leshem 2017), which refers to the tactical non-presence of the state (Gill 2010) as a means of controlling and excluding migrants.In the EU context, abandonment can be associated with different actions by subtraction.First, it is associated with the hardening of migration law and its implementation, which undermines migrants' rights while constraining their "respective positions and possibilities to act" within the EU migration regime (Eule et al. 2019:66).This process has led to a growing number of people being "irregularised", meaning they are no longer admitted to asylum accommodation centres and other basic facilities (Ambrosini 2012;Darling 2009Darling , 2016;;Kalir 2017).Abandonment can also be linked to the recent continual cutbacks in fundings for NGOs and humanitarian associations while an increasing amount of funding is directed to deportation centres (Kalir and Wissink 2016).According to Squire (2015:513), abandonment is productive for the state because it provides an "exploitable pool of excess labour and of political scapegoats" while producing very detrimental effects on migrants and their lives.In fact, it exhausts people (de Vries and Guild 2019:2157), pushing them to go elsewhere, or causing them to die of hunger, disease and loneliness (Castro 2015;Davies et al. 2017;Mbembe 2003).The state abandons migrants literally on the street while ensuring that it is impossible for this action to lead to their personal and/or collective empowerment.In this regard, abandonment acts after and because the other actions we have already seen such as "domicide practices" remain in place.
Although scholars have already highlighted how acts of abandonment affect informal camps (Martin et al. 2020) and migrants' agency, this article indicates the need to interrogate its consequences over time in greater detail, given its increasing weight in the EU.From this perspective and through makeshift spaces, the interplay between migrants' actions and migration regimes will appear in all its complexity.

Beyond Concepts: My Time at the Border
During my empirical research in Ventimiglia, between September and December 2018, I decided to live in the downtown area.Given its small size, it was easy to quickly gain a mental map of the main places to investigate in more detail.At first, I daily observed social practices in the public spaces, in particular at the train station, in the Gianchette district and along the beach, where migrants are seen more frequently along with the police, while documenting this co-presence and its variability (checks, chases and migrants' negotiations and (re)actions; situations of apparent calm and mutual indifference).
A few days after my fieldwork started, people could meet me casually in the streets and squares and in the two main places in town dedicated to migrants: the Eufemia Infopoint and the Caritas Centre.In particular, I was involved in the collective activities at the Eufemia Infopoint, set up by the Collective 20K 1 in 2016 in the Gianchette district, taking part in weekly meetings with all the "solidares" (activists and NGO employees) present in the city, while monitoring migrants' situations in the public spaces of Ventimiglia, right at the border, and on the trains going to France; all situations in which Italian patrols and French pushbacks occur.At the Infopoint, I was able to speak with many migrants, collecting more than 20 stories in my fieldwork diary, while taking part in the main activities that took place there, such as giving information, providing internet access, and handing out clothes and blankets.Although I never hid the fact that I was (also) a researcher, I deliberately chose not to record any of those stories so as not to ruin the informal atmosphere of the Infopoint, by suddenly placing a recorder between my interlocutors and me.The Eufemia Infopoint played a pivotal role in sorting out my research contacts.There, in fact, I ran into two people who gave me easier access to a broader migrant network: Reis, a Pakistani volunteer at the Infopoint, and Mansa, a Malian volunteer at the Caritas Centre, both of them asylum seekers living in the Roja Camp.Reis, with whom I spoke English, included me in his group of friends, enabling me to spend time with them during the day, going with them by train to the Prefettura of Imperia, where they had to check their asylum procedures.Both Reis and Mansa helped me understand the situation at the official camp before I was allowed to enter it, while I was able to help them understand some aspects of Italian procedure.In November, Mansa formally became my research assistant, supporting me as a translator during the interviews I conducted in a spare room at the Caritas twice a week.Thanks to Mansa and his linguistic competencies (in French, Bambara, Mandingo, Arabic, Italian, and Spanish), I recorded ten interviews, all with asylum seekers from sub-Saharan Africa.
In the Caritas, in the Eufemia and in the public spaces, I came into direct contact with migrants stuck in Ventimiglia in an extremely marginalised situation.Most of these people were homeless, and had set up informal shelters in small groups along the river or on the beach.I saw these precarious settlements in person.In this case, rather than acting as a translator, Mansa facilitated my access as a white woman to places characterised by the presence of almost exclusively black males.Approaching these informal arrangements was not easy since people's physical and mental suffering created difficulties in communication that were not due simply to language differences.The other most difficult place to access was the most formalized one, namely the Roja Camp, although for different reasons.In November 2018, after a wait of more than six weeks, the Prefettura of Imperia finally gave me authorisation to enter it five times.At the Camp, I carried out 15 formal interviews with both migrants and NGO employees, while observing the internal dynamics of the Camp at different times, and chatting informally with different people I met.
All the qualitative methods used, from the least to the most formalised, were aimed at understanding people's migration paths, their difficulties before and after reaching the EU, their situation at the border and their hopes for the future.Along with secondary data, the interviews and conversations with politicians, policemen, NGO employees and local people addressed the main changes that had occurred in Ventimiglia since 2015, providing the basis for my diachronic reading.
To sum up, during my fieldwork I adopted what Brambilla (2015:28) defines as a "performative, participatory and political" approach.This approach structured my presence in Ventimiglia and the methodologies adopted, both designed to counter the idea that migrants are "objects to be governed" (Mainwaring 2019:27) as it frequently appears both in policy and the media.

A Geographical and Political Gaze on Border Governance in Ventimiglia between 2015 and 2018
Ventimiglia is the last town on the north-western stretch of the Italian Riviera before France, close to Nice and the Principality of Monaco.The geography of this area-the sea on one side and the Maritime Alps on the other-facilitates its border function, allowing for only two routes into France using transport infrastructure (rail and road).Hence it was easy for the French police to start monitoring these routes, turning back to Italy an increasing number of migrants who were not permitted to cross irregularly the border.While in 2015, 1193 2 people were refused entry to the Maritime Alps department-the one bordering on Ventimiglia-in 2017 the figure rose to 44,433 (including 13,500 minors).Ever since the first French blockade at the end of 2015, the Italian authorities have developed different and constantly evolving ways of dealing with migrants' presence in the town.Their evolution can be assessed by identifying three different phases (Figure 1).
The first phase coincides with the period until July 2016, when the Roja Camp was opened.Before that time, the measures taken by the authorities were chaotic and characterised by growing tensions between the police, migrants, and activists who came to support migrants along with the local people.During this phase, "acts of domicide" (Van Isacker 2019) against migrants' makeshift camps become the norm while a first aid camp run by the Red Cross near the railway station was closed after few months as no longer "fit for purpose", not least because its dilapidated condition was regarded as being harmful to the image of the tourist city (Interview with the Mayor, 20 September 2018).Following the umpteenth eviction, the migrants were welcomed into the church of Sant'Antonio (Interview with Mujahid, 3 a person from Sudan under humanitarian protection, 2 December 2018) where men were allowed to stay until the opening of the official Camp in July 2016, and women and minors until August 2017.In the prevailing institutional vacuum, the basic function performed by the church was recognised informally by everyone; not just by the Mayor, but by the police, who would send migrants to Don Rito, the local priest, to get help (Interview with M.N., Caritas doctor, 19 November 2018).In May 2016, the Minister of the Interior of the time, Angelino Alfano, visited the city.Following his visit, the institutional approach clearly moved towards normalising the situation in the city 4 giving way to a second phase of the border management.This normalisation process implied two joint and interconnected actions.First, the Prefettura identified a site more than four kilometres outside the city where it could open the so-called Roja Camp, entrusting its running to the Red Cross.In this phase, evictions aimed at transferring migrants to this place, which started to be seen as the only "legitimate space" for migrants by the authorities and by some inhabitants (Member of the "Gianchette" Neighbourhood Committee, 22 November 2018), especially after the total closure of the church to women and unaccompanied minors in August 2017.Second, the national government started to use forced mobility, through the so-called "decompression strategy", which consists literally of "rounding people up and taking them somewhere else", 5 as explained by the Head of the Italian police Franco Gabrielli.In particular, migrants who following specific sweeps turned out not to be seeking asylum in the province were taken by coach to hotspots in southern Italy, mostly Taranto. 6The objective of this decompression strategy was twofold.The first was to discipline migrants' bodies in space; the bodies of those who were actually moved, as well as the bodies of those who-for fear of being transferred-were more likely to be obedient to the discipline being imposed.The second objective of the strategy was "environmental" (Bonnin 2017); namely, reducing the number of migrants in the area by rounding them up and moving them away.
In April 2018, the eviction of the last main makeshift camp in the city marked the transition to a third phase in border governance of Ventimiglia, closely related to what was happening at the EU's external borders and at the Italian level.First, a new deal between the Italian government and the Libyan Government of National Unity in 2017 led to a decrease of around 80% in the numbers of arrivals in Italy via the Central Mediterranean Route.Second, the installation of a conservative government in March 2018 made the use of active abandonment as tool of migration governance more systematic in various ways: by tightening the restrictions around reception (Villa 2018); by not providing real help in the form of lodging for people with refugee status (ASGI 2014); by reducing public funding for asylum centres and integration programmes (Interview with J., NGO worker, 1 December 2018); and, finally, by reinforcing the criminalisation of solidarity networks (Tazzioli 2018).These governance mechanisms led to a change in migrants' motivations for staying in Ventimiglia.During 2018, a growing number of people arrived in the area in order to enter the Roja Camp (Territorial Assembly, 16 October 2018), the last one in Italy open to migrants regardless of their legal status.Its new "attractiveness" for those migrants who aim to find a place to stay (Interview with J., NGO worker, 1 December 2018) has driven the authorities to deny it to a growing number of people.This decision seeks to drive forward the "normalisation" of Ventimiglia (Interview with S., local representative of the League Party, 6 November 2018) in order to prevent the place remaining a magnet for migrants regardless of their intention to cross the border.In this last phase, indeed, the use of the Roja Camp as a mechanism of confinement for people in transit and its closure as a mechanism of abandonment for those in search of accommodation were two complementary forces that inevitably operated both inside and outside this particular space.

Camps in Ventimiglia
The Roja Camp: The Ontology of a "Transit Camp" It was surprising to learn that the Red Cross Camp is commonly known in Ventimiglia as the "transit camp".Anyone who is familiar with the Italian asylum system knows that there is no formal category of camps of this type (Interview with NGO workers, 1 December 2018).As the National Guarantee for those Deprived of their Freedom clearly shows (Garante Nazionale 2016), the centre's functional purpose does not conform either to the Italian legal system, 7 or to that of the EU.It comes into the category of "a centre for temporary assistance to support migrants who are in transit along a route ... presenting grave risks to life" (Garante Nazionale 2016).This ambiguity emerged clearly in the Mayor's words in response to my direct question about its nature: [The Roja Camp] was atypical from the start.It's still atypical because in order to encourage people to go to the camp rather than remaining on the streets, the camp was open to all, because otherwise no one would have gone ... This however was risky for everyone, because, let's put it like this: you created a reception centre that does not exist in law.That is, there were no alternatives: I am putting you into certain conditions, while you wait for your passage or prepare for your journey.(Interview with the Mayor of Ventimiglia, 20 September 2018) The transitional character of the camp therefore has a dual aspect.On the one hand, it refers to the theoretically limited duration of the migrants' stays, since they "do not stay here very long" (Interview with I., Head of the Roja Camp, 22 November 2018).On the other, the focus is on migrants' movement, and the role played by the camp itself in supporting their attempts to cross the French border.
The Roja Camp has at least three main functions.First, it manages migrants' presence at the urban level, by containing migrants far from the city centre and, second, it allows the police to the space with ease.The placement of migrants in the camp and its displacement from the city centre are thus two coexisting forces.While the distance in itself is not great, reaching the camp remains challenging; to get there, one has to travel along a busy road that is not illuminated at night.Although it was not set up in order to detain people, this camp nevertheless operates as a way of controlling bodies in space (Menghi 2018;Mountz et al. 2013;Tazzioli and Garelli 2020).This aspect becomes even more apparent if we look at the changes concerning the rules for entry.After an initial phase lasting a few months in which the camp was completely open, it was subsequently guarded by police 24 hours a day.Those entering the camp are photographed and identified, and their profile is then checked against the AFIS (Automated Fingerprint Identification System) to see if they have a criminal record.Some offences, including sexual violence and acts considered socially dangerous, bar an individual from entering the camp.Anyone who has not previously been identified in Italy or another EU country is taken to the police station and fingerprinted so they can be entered in the EURODAC system.This procedure, from the moment of its introduction, has tended to scare away migrants who have not yet been identified, especially those who came via the Balkans, who were generally not registered when they entered Italy unlike the others.However, until 2018, the camp policy was to encourage people already identified to enter, because it protects them from the decompression strategy: anyone who resides there cannot be sent to hotspots.This sort of "untouchability" accorded to those living in the camp derives from a tacit agreement between the Red Cross and the police.The camp's raison d'être is firstly the desire to keep the situation in Ventimiglia peacefully under control and presupposes that the police cannot enter "in order to round people up" (Interview with I., Head of the Roja Camp, 22 November 2018).Third, despite the assurances given to the French authorities, it attempts to manage migration at the international level, notably by indirectly supporting border crossings by migrants.In 2018, for the first time since the camp opened, there was an increase in rejections by the police of a growing number of migrants (Interview with Mansa, asylum seeker, 1 December 2018; Territorial meeting, 16 October 2018).Not only was the camp turning some people away; from 2019, it allowed camp residents to be transferred to hotspots in southern Italy (Conversation with Mansa, asylum seeker from Mali, 1 March 2019).

Outside the Roja Camp: From Makeshift Camps to Informal Arrangements
Let's reflect further on how the different phases of border governance (Figure 1) affected the evolution of makeshift camps and arrangements over the period considered.
At the end of 2015, as soon as the French border blockade increased the number of people on the move who were stuck in Ventimiglia, makeshift camps immediately sprang up to shelter migrants, supported by local people and an ever-growing international body of activists (Interview with J., NGO worker, 2 December 2018).In particular, migrants-along with the No Borders activists-up a protest camp on the rocks at the Balzi Rossi, which is an area in front of the border checkpoint (Interview with Mujahid, 2 December 2018).A wide variety of individuals joined the protest while "migrants' political subjectivity [was put] at the core of the action" (Trucco 2021:182).After Alfano's visit, the protest camp was evicted, while other makeshift camps begun to arise in the urban area, particularly in the Gianchette district, in the underpass of the major road access to the city (Interview with S.S., the Mayor's assistant, 18 September 2018).The survival of the makeshift camps became ever more precarious, even though they were recreated-after every eviction-always in the same area of Ventimiglia.In this battle between "domicide acts" and "makeshift camps", the mediation and presence of No Borders and volunteers were essential in supporting migrants' actions, as Mujahid highlighted: [T]hey were with us all the time.They slept with us, they did security for us, because the police come there, and then we don't speak the language, we can't even talk to the police and say what our problems were.They were there to help us, when the police arrive, they help us with all the things they can help us with.Obviously with language, if there is violence from the police, they can also help us.(Interview with Mujahid, 2 December 2018) After every eviction, migrants were methodically identified, and activists were served with expulsion orders.In 2016, the systematic nature of evictions and expulsions of the No Borders activists created a paradoxical situation: to the increasing precariousness of the makeshift camps was added a real "institutional vacuum" in support for migrants (Interview with C., a social worker at Oxfam, 1 December 2018), which was partially balanced by the church of Sant'Antonio following the umpteenth eviction.Mujahid clearly remembered the growing role played by the church when the No Borders presence in Ventimiglia became less stable: They cleared us out after four months.Then we tried to find another space ... and they took us away again ... Don Rito said: "Come into the church".So we went into the church.Obviously, the police had served all the No Borders people with expulsion orders, so they all left, only a few remained ... they gave them expulsion orders so they all went home.(Interview with Mujahid, 2 December 2018) According to Mujahid's account, from the summer of 2016, the expulsion of people on the move from the city centre toward the Roja Camp following the closure of the church and the evictions of other makeshift camps provoked strong reactions from the migrants.
[T]hey took us, migrants, to the Red Cross Camp, but obviously, we refused to go in, because we would have been in prison, we have seen what life is like there.We didn't come here [to Europe] to go [to prison].We thought that there was freedom in Europe, we want to be free, to live as we choose.(Interview with Mujahid, 2 December 2018) In this second phase, the decompression strategy was particularly targeted at "rebels", that is, those who were playing a prominent role in mobilising others to get the border with France reopened (Bonnin 2017).
Since the start of the third in border governance, from April 2018, it has been impossible to recreate makeshift camps in the city as in the past; in their place there are a series of physical enclosures that prevent access to spaces.The contemporaneous presence of acts of control and abandonment (Pinelli 2018) has indeed had substantial repercussions on this border area and, in particular, on makeshift settlements.They are smaller than their predecessors and scattered over the area between the beach and the course of the river Roja.Their structure and duration are even more ephemeral than those of the previous makeshift camps (Minca and Umek 2019); this makes it harder to map them (Davies et al. 2017), but also to support people staying there (Interview with C., NGO worker, 1 December 2018).In Ventimiglia active abandonment is also effective because it proceeds in concert with the police's pervasive control over migrants and solidarity movements, and-as previously highlighted-with the external border reinforcement which limits the number of arrivals.We therefore need to turn our attention urgently to migrants' agency.
Migrants Before the Border: Questioning Migrants' Agency in Formal and Informal Camps Migrants' Agency Inside the Camp Coming and going from the camp on foot is slow and dangerous, and even by car it is not easy to find the entrance.Despite this, there is a constant coming and going of people-mainly men-who make this journey into Ventimiglia sometimes several times a day.The centre has capacity for 360 residents, but it often takes in many more, housing them in tents that are available for this purpose.Each living area, a little less than 15 square metres (Menghi 2018), consists of a single room in which around six people live, each of them having a camp bed and a blanket (Fieldnotes).There are differentiated areas within the camp: one with prefabricated bathrooms, another housing the Red Cross offices where legal support is given; a camp kitchen in a large tent where meals are served; a children's area; and a computer room with a dozen PCs that (at least in December 2018) were not yet connected to the internet.Following the admission of women and children in 2017, areas were created for families and single women, and for unaccompanied minors, in accordance with Italian and EU law.The division between these areas is minimal, and the different areas can easily be accessed by anyone who decides to do so.For "safety" reasons, given the high turnover of migrants in the camp, the rooms cannot be locked (Interview with I., Head of the Roja Camp, 22 November 2018).The lack of a safe place for belongings and the continual thefts contribute to the feelings of dispossession and lead to attempts to find alternative solutions to the problem "for example by taking the cables off the air conditioning units ... to shut the door, so they do try to obtain some privacy" (Interview with C., NGO worker, 1 December 2018).This shows that, even in this anonymous and standardised space, some degree of appropriation of the camp's spaces has been attained.For instance, Muslim worshippers have carved out space under the bridge for an open-air mosque, with beautiful carpets spread on the ground.Clothes and blankets are used to cover whose shutters are broken.When the weather permits, the large central space is used to play football and cricket (the latter played mainly by Pakistani men).Competitive matches are organised in the evenings, while the central space serves as a corridor for walking and chatting in the daytime.
The long-drawn-out nature of the asylum process, the uncertainty about how long it will take and the continual rebuffs at the border can cause many residents to feel discouraged or even wholly demoralised.Life in a precarious place such as the camp at Ventimiglia, which is also characterised by a highly mixed population with resulting problems of communication and coexistence (Agier et al. 2018), is often tense and challenging (Interview with I., Head of the Roja Camp, 22 November 2018).In the period since the camp opened, those who have worked there have noted an increase in mental health problems among the residents (Fieldnotes).These are connected not only with the traumas experienced during their migration but also with the difficult situation created by the reception process itself: "people who perhaps were OK but have become victims of the system" (Interview with E., legal aide at the Roja Camp, 1 November 2018).For both men and women, some mental health problems may indeed be the result of a situation of deprivation and severe psychological stress; however, they can also spring from the contrast between the experience of life in the camp and different learned ways of behaving.For instance, in several interviews, migrants complained about the need to ask camp workers for every little thing, stripping them of the autonomy of normal adult life.This infantilisation process (Altin and Minca 2016), which can also be detected in the words of some of the camp workers, not only places the individual in a position of submission and supplication, an unnatural one for an adult, but also makes their actions subject to the decisions of those in charge of the camp (Interviews with Abayomi, 29 November 2018, Asad, 29 November 2018, and Ahmad, 1 November 2018; people waiting to reenter the asylum procedure).As fieldwork shows, the Roja Camp works as a space of containment and control and has negative consequences, in line with other mechanisms of the asylum system, in worsening the mental health of migrants (Beneduce 2015).At the same time, what also emerges is the power of the migrants' agency and their capacity for including the camp in their strategies as temporary accommodation in diverse situations: during their attempts at crossing the border, when they wish to re-enter the asylum system or, for those already rejected, while they decide what their future should be.In all these cases people seem to agree, however, that the camp meets their present basic needs.The Roja Camp certainly does not allow the creation of any kind of group mobilisation, as happened in the first makeshift camps; here, individuals carve out autonomous spaces just for themselves (Interview with Mujahid, 2 December 2018).For some people, the camp is the sole alternative to life on the streets.Abayomi's words are very clear in this respect: Staying in the camp is the only possibility.While we stay here there are no problems ... They give us food, somewhere to sleep, a shower ... If you have nothing, you're better off here than on the streets.(Interview with Abayomi at the Roja Camp, 29 November 2018)

Informal Settlements and Migrants' Agency
It be wrong to simply associate the evolution of informal camps in informal settlements to a "zeroing" of migrants' capacity to navigate the border at Ventimiglia.My fieldwork, in line with the theoretical background of this paper, demonstrated the opposite, alongside a general deterioration of lives in informal settlements.In truth, highly differentiated situations can be related to the informal camps, beyond a general decline of their materiality and collective social-power (Territorial meeting, 16 October 2018).There are always small groups in transit that create hidden and informal settlements while attempting to cross the border (Fieldnotes).In this context, the different strategies and actions carried out by those attempting to cross the border were related to their personal socioeconomical capital and its capacity to give them access, above all, to smugglers in the city (Amigoni et al. 2021b).In particular, people from the Balkan route, who in 2018 were often still unidentified, refused to enter the Camp (Fieldnotes); but so did people in transit who did not wish to stay too far from the railway station.
However, as the previous sections explained, increasing numbers of people stay in Ventimiglia but not with the aim of crossing the border.Some of them are exhausted by their n th border rejection; others are excluded from the Italian asylum system and, for different reasons, were not welcomed in the Roja Camp either.In these cases, people lived in makeshift arrangements, put together from beach umbrellas, mattresses and a few blankets in groups of varying number from three to a maximum of ten individuals.The "increasing number of people to whom access to the Camp is denied" (D., NGO officer, 17 November 2018) was one of the main points under discussion in the Territorial meeting in mid-November 2018.During this event, while some social workers suggested informing the competent authorities of migrants in need, others feared indirectly triggering an intensification of forced transfers to hotspots or deportation centres.Some migrants even preferred to stay in the Roja Camp to avoid this situation of extreme marginality; one of these was Abayomi who decided to remain in the camp in order to avoid suffering from the situation of migrants in the downtown area: I stay in the Camp, I don't want to go into Ventimiglia any more.Because if I see someone who is not OK or is doing bad things to survive, I feel my heart "bursting into tears".If I see someone who is living badly or has problems getting by, it breaks my heart.I don't want to see it.(Interview with Abayomi, asylum seeker at the Roja Camp, 29 November 2018) Abayomi's words pointed indirectly to the detrimental effect of the active form of abandonment on migrants' situation in Ventimiglia.The essential absence of the state (Davies et al. 2017;Gill 2010) from provision of minimum levels of safety and support to migrants in the city was written on their bodies, which carry the signs of the exhaustion and fatigue of the journey and of the waiting, but also the cold and salt air to which they are exposed on the beach or among the bushes on days when it rains (Fieldwork).Again, even in such situations of great marginality, people act or try to act in different ways.For instance, on the beach, some migrants offered paid help to support other people in crossing the border; others were waiting there while their asylum applications were examined.During my fieldwork, I tried to suggest to some people that they should go in person to the Prefettura, because court summons for the asylum hearing would never reach them on the beach.However, in this case too, their physical and mental fatigue emerges, as noted in my fieldnotes: "They're confused, tired; they were mostly cold at night.They offer[ed] us a mandarin, they're happy that we got closer" (Meeting with a group of four Kurds from Iraq, 19 November 2018).
Of course, this physical and mental fatigue is related to the system's violence as well as to the amount of time needed to constantly reconfigure new strategies to cross the border or, depending on the case, to find ways to survive daily in very precarious conditions.In this latter case, even so-called "voluntary return" appears to some as an alternative, as shown by the story of Chokri, a Tunisian man without any accommodation, who I met at the Caritas Centre.With no job, at the age of about 60, he "felt he was dying" living on the streets in winter (Caritas, 23 November 2018) so he was seeking a free passage back to Tunisia after 20 years spent working in Italy.At the Caritas, others were searching for accommodation in Ventimiglia, having been thrown out of the Camp.However, the center offered the eight beds available to homeless Italians, because migrants were supposed to get that kind of support at the Camp.
In some cases, I admit I did not know the situation of migrants I met, neither their actions nor their plans.Nevertheless, even this silence is somehow meaningful, as my fieldnotes indirectly indicate: On the way to the Camp, we meet a 30-year-old man, unsteady, evidently drunk.Mansa holds him up promptly before he falls to the ground unconscious.A police patrol comes up to us, asking me, "Ma'am, is there a problem?".The policeman later explained me that they were afraid that I had become indirectly involved in a fight between migrants.They were about to get back calmly into the car when they realised I was calling an ambulance.In an attempt to calm me down, the policeman said: "Now the A&E workers will be angry![chuckling] Doctors can't bear to see these drunk migrants hogging the emergency room.Once he regains consciousness, they won't do anything to help him".(Fieldnotes) Migrants' "ability to formulate strategy (whether explicitly recognised as such or not)" (Squire 2017:261) seems ever narrower and more restricted.As this section has attempted to show, this reduced "room for manoeuvre" (Mainwaring 2016) is related to a border regime which has continually found new ways to reinforce its structural violence (Tyner 2017).

Conclusion
This paper examined the interplay between migrants' actions and migration regimes in the Italian town of Ventimiglia, close to the French border, by looking in particular at the evolution of official and makeshift camps between 2015 and 2018.The paper starts from 2015 when the restoration of checks on the French border turned Ventimiglia once again into a border city where migrants' movements "collide" with the more rigorous French controls (Queirolo Palmas and Rahola 2018).Through a diachronic reading thanks to extensive empirical research conducted in 2018, this work aims to further the debate on border regimes and migrants' agency by arguing two main points.
First, a close and clear relationship emerged between Italian migration governance and the deconstruction and elimination of makeshift settlements.The paper critically related this process to a series of acts of control and abandonment embedded in the border governance and developed in a complementary way since 2015.In particular, following Van Isacker (2019), the analysis linked the deterioration of makeshift camps to "domicide practices" which disempowered migrants by transferring and containing them far from the city centre in the Roja Camp, by closing the spaces where makeshift camps persisted and, finally, by criminalising support from activists.The analysis revealed how, after the Camp opened, the authorities mobilised specific parts and powers of the state's apparatus (mainly the police) rather than others (the welfare state), delegating key responsibilities in support for migrants to other actors (e.g.locals, the church, and, subsequently, NGOs).However, during 2018, an active form of abandonment triggered a deterioration in migrants' situation at the national level and, therefore, at the local one too.As a result, a growing number of undocumented people without any public support and a slim chance of finding work headed for the Roja Camp in search of a minimal level of support.By following the evolution of the border governance, it was possible to point out how situations of extreme marginality and loneliness have arisen, becoming almost the norm outside the Camp.
This account leads to my second argument.This paper showed how migrants' agency can act in the formal Camp just as it can be "disempowered" in the makeshift temporary arrangements.In focusing on this perspective, I responded to the invitation to broaden the study of contemporary camps in Europe (Martin et al. 2020:746) by casting a light, as never before, on the ambiguous and unpredictable relationship between formal and informal camps, and migrants' agency in Ventimiglia.This case study helped to better problematise the detrimental effect of the border regime without underestimating migrants' power to deal with it both inside and outside the official camps.In particular, the diachronic reading allowed the paper to bring out how, in a context characterised by different forms of abandonment by the Italian authorities, the Roja Camp was actively included by people in their plans thanks to the minimal level of services offered.By contrast, it also stressed how the systematic destruction of the former makeshift camps increased the part played by migrants' individual socio-economic capacity, while leaving all of them in a situation of increasing difficulty outside the official camp.
In August 2020, in the middle of the COVID pandemic, the Roja Camp closed officially because of the fall in migrant arrivals; it had been the last official camp in Italy open to all migrants.This latest act unfortunately supports the argument put forward by my work, as did the demonstration by activists which took place in opposition to the authorities' decision to close the camp.It is clear that the somber consequences of this closure are evident not only to academia but also in social movements.Obviously, neither this protest nor my work are intended in any sense as a defence of the official camp in itself.On the contrary, both clearly indicate that the dynamics of power-resistance underpinning migrants' agency border regimes are increasingly unbalanced at migrants' expense (Eule et al. 2019).From this perspective, the state's active withdrawal from even minimal support and its constant presence through acts of control are expressions of a border regime that constantly implements different strategies and different combinations of strategies in opposition to migrants' autonomy and rights.In such a volatile setting, this work provides unique insights into the power relations between migrants and border regimes thanks to intensive fieldwork and a diachronic approach.In doing so, this work contributes to a growing body of work that (both inside and outside academia) politically denounces the inadequacy, fragility and violence of the political structures and institutions which migrants move into and within.Furthermore, this article urges academic debate to probe more deeply into abandonment mechanisms and their increasing role in reconfiguring migration and border governance in the EU.

Figure 1 :
Figure 1: Border governance in Ventimiglia from 2015 to 2018 [Colour figure can be viewed at wileyonlinelibrary.com]