Tracing Colonial Maternalism Within the Gendered Morals of Humanitarianism: Experiences of Migrant Women at the Moroccan-Spanish Border

By bringing together two sets of qualitative fieldwork conducted in 2016 and 2017 with humanitarian organizations and migrant women on the two sides of the Eastern Moroccan-Spanish border, this article examines the ways in which women humanitarians exercise power over women’s lives, bodies and mobility. Though humanitarianism in the border context has been researched at European and United States-Mexico borders, the specifically gendered implications of humanitarian governance at borders needs further investigation. Drawing on interviews and participant observation with a religious humanitarian organization on the Moroccan side of the border and with medical humanitarians and social workers on the Spanish side, we researched women humanitarians’ interventions toward migrant women from West African, North African and Middle Eastern countries. Beyond the differences that characterize these religious and socio-medical humanitarian settings and the different migration regimes in which they are inscribed, we argue that the power exerted by women humanitarians reproduces a form of maternalism underpinned by gendered moral beliefs regarding women’s bodies, mobility and family life. We foreground that such maternalism represents a cornerstone of women’s humanitarian engagement across time and we identify continuities between colonial maternalism, contemporary forms of humanitarian care carried out by women and maternalist integration politics in Western postcolonial societies. Rooted in colonial maternalism, racialized beliefs justified women’s (religious, medical, social) prominent role in intervening in the intimate spaces of women casted as radically Other. Our contemporary case studies demonstrate how the practices of women humanitarians impact on racialized migrant women’s daily lives, intruding on their intimacy, imposing controls over their bodies and impacting on their possibilities for mobility. The article explores how the racialization of migrant women, articulated with moral ideas around women’s reproductive health and mothering responsibilities, produce varied forms of disciplining and control on both sides of the border Moroccan-Spanish border.


INTRODUCTION
Studies of the Moroccan-Spanish border around the enclaves of Ceuta and Melilla have foregrounded the militarization of migration control at these borders over the past three decades and the related heightened levels of violence, especially with regard to so-called "Sub-Saharans", nationals of Central and West Africa (Pian, 2009;Moffette, 2013;Andersson, 2014;Ferrer-Gallardo and Gabrielli, 2018). The lived experience of violence at these borders has also been recounted by Central and West African migrants who undertook that crossing (Mbolela, 2017;Mboume, 2018 among others) as well as documented by NGOs (for instance Migreurop, 2007;Medecins Sans Frontieres, 2013;GADEM, 2015). This "Western Mediterranean" route to Europe represents one of the three main routes to Europe with the Central Mediterranean one (Libya-Italy) and the Eastern one (Turkey-Greece). Over the past few years, the relative significance of these routes has varied according to changing geopolitical contexts and specific measures of migration control and deterrence. In 2016 for instance 9,900 persons arrived in Spain while over 180,000 persons risked the central Mediterranean crossing and as many undertook the Eastern Mediterranean one. In 2018 however a little over 23,000 persons arrived through the central route, almost 56,000 via the Eastern one and the Western route to Spain grew rapidly with over 56,000 arrivals 1 . Over the past few years, the eastern Moroccan-Spanish land and sea borders have thus witnessed varied numbers of border crossings and attempted crossings against the background of the broader geopolitics of the Mediterranean routes to Europe.
The experiences of migrant women at the Moroccan-Spanish borderland have been relatively little researched (Laacher, 2010;Freedman, 2012;Tyszler, 2018;Tyszler, 2019a;Tyszler, 2019b;Tyszler, 2020;Sahraoui, 2020a;Sahraoui, 2020b), the dominant figure of the migrant at these increasingly militarized borders being mostly depicted under the traits of the single Black man. However, a plurality of people try to pass through, other than those from Central and West Africa, including Moroccans, Algerians, Syrians or even Indians, Bangladeshis or Yemenis depending on the period. While men do indeed form the majority of migrants at the Moroccan-Spanish borders, women are also present, and their experiences of these borderlands are impregnated with gendered forms of violence. In 2017 for instance over 7% of the persons who reached (mainland) Spain by sea were women and in the Center for the temporary Stay of Immigrants (CETI) in the Spanish enclave of Melilla (on the African continent), women accounted for 17% of residents (while minors represented around 20%) in September 2016.
In this article we are concerned with the experiences of illegalized 2 migrant women at the Eastern Moroccan-Spanish border with the Rif region on the Moroccan side and the Spanish enclave of Melilla on the other. The article results from two distinct sets of fieldwork carried out by the authors independently, while sharing a focus on the daily experiences of women in humanitarian settings. Importantly, we want to clarify that we each interviewed women of different nationalities on the two sides of the border: Elsa Tyszler met women from West and Central African countries (notably from the Congos, Cameroon, Ivory Coast, Mali, Senegal and Guinea) and Nina Sahraoui met mostly with Syrian, Algerian and Moroccan women. While some women from West and central African countries were present in the CETI in Melilla at the time of fieldwork (19, mostly from Guinea), none was pregnant or accompanied by small children, which was the initial entry point of the research conducted in Melilla. Overall, by September 2016, 58 pregnant women went through the CETI that year and the overwhelming majority were from Syria, Algeria and Morocco. Yet as Tyszler's research demonstrates, pregnant women from Central and West African countries were present on the Moroccan side of the border. The fundamentally racialized and gendered modes of border crossing account for these disparities. Black pregnant women were indeed unable to attempt crossing the border by land and were constrained to a sea crossing in an attempt to reach the Spanish mainland (and not the enclave of Melilla). In 2017 for instance, 161 pregnant women reached Spain by the sea (APDHA Asociación pro derechos humanos de Andalucía, 2018). Women from Arab countries could in contrast attempt to cross the border on foot by mixing with the daily passages of many Moroccans into and out of the enclave 3 . While our study illuminates certain similarities in the gendered humanitarian governance on both sides of the border, it is however important to highlight that owing to the deeply racialized migration control implemented at the border, producing racialized border crossing tactics (Tyszler, 2019b), our case studies actually concern women of different nationalities. In addition, this border region is characterized by a continuum of migration control and restrictions as well as ruptures owing to the different national migration regimes to which the research sites pertain. The two sets of data are brought together to shed light on certain similarities yet the chosen vignettes also illustrate some of the diverging contextual elements observed on each side of the border.
In this article we explore migrant women's experiences with humanitarian organizations on both sides of the border and we are attentive to gendered dynamics that concern the religious and medical humanitarians we study. By bringing together two casestudies around humanitarians' involvement in this border region we seek to uncover the specifically gendered morals that underpin certain strands of humanitarianism. While some attention has been dedicated to formal gender-related programmes implemented by humanitarian actors (Olivius, 2016), we are here interested in the implicit and latent ways in which humanitarian interventions, notably those carried out by women, bear gendered implications. We argue that the observed attitudes and the attached moral ideas are inscribed in broader relations of power imbued with coloniality (Quijano, 2000) and in particular with colonial maternalism.
We first situate our contribution in relation to the growing field of literature around humanitarianism and engage with the notion of humanitarian border. Concerned with the actions of women religious missionaries and of medical professionals, also women in their great majority, we foreground in the first section the shared colonial genealogy of religious and medical forms of humanitarianism, which brings us to emphasize the colonial maternalism that underpins this humanitarian assistance. Following an overview of the methodology that governed the two case studies brought together here, we discuss how women's intimacy was curtailed in the respective humanitarian settings and the ways in which women humanitarians exerted control over migrant women's bodies, health and mobility. The fourth section is concerned with the many forms of resistance and contestation that the women we met displayed and enacted daily in spite of the constraints imposed.

HUMANITARIANISM, COLONIALITY AND MATERNALISM
The governance enacted by humanitarian actors in a postcolonial world is enmeshed in moral ideas, notably through the higher moral grounds of saving lives that is the justifying principle of humanitarian intervention. However, as researchers concerned with humanitarianism have argued, the life that is to be saved under the humanitarian predicament needs to be recognizable in humanitarian terms, which means embodying the innocent victim by being construed first and foremost as a suffering body (Ticktin, 2011). To demonstrate her point, Miriam Ticktin explores the work of Médecins Sans Frontières (MSF, Doctors without Borders) in the Congo Republic during the civil war in the late 1990s, observing that MSF's analysis of their own practices, in an attempt to account for the organization's neglect of war rapes, concluded that "women who had experienced sexual assault were not the ideal subjects of aid, since they could not be easily identified with images of innocence" (Ticktin, 2011: 259-260). In the early stages of humanitarianism's engagement with sexual violence, moral understandings of gender-based violence indeed remained imbued with traditional gender stereotypes prevalent in Western societies. This is but one possible example of the ways in which the supposedly apolitical rationale of humanitarian governance cannot in fact escape politicization, rather humanitarianism is caught up in intersectional relations of power which include racialized hierarchies inherited from colonial endeavors. Didier Fassin argues that there is a need to understand the "humanitarian language", that is a language of "affects and values" in order to "explain why people often prefer to speak about suffering and compassion than about interests or justice" (Fassin, 2010, 3). Humanitarianism is also a political enterprise because, as a matter of fact, and often despite the intentions of humanitarians themselves, it represents mostly Western organizations engaging in non-Western regions in a world dominated by Western countries. This is not to say that there are no non-Western humanitarian organizations but within the politics of humanitarianism Western organizations are more easily recognized as humanitarian actors and benefit from greater visibility (Barnett and Weiss, 2013). Ultimately, humanitarianism is political because it has become a mode of governance for certain populations, usually those situated in one way or another outside the political framework of the nationstate.
Humanitarian actions and humanitarian discourse are key tools of governance toward those deemed in need of saving and rescue. Yet, a fundamental characteristic of humanitarian intervention is to rely on a hierarchical relation of victims and saviors (Perkowski, 2016). These dynamics also play out in border contexts, whereby the increasing involvement of humanitarians prompted William Walters to identify "the birth of the humanitarian border", emphasizing the "reinvention of the border as a space of humanitarian government" because "border crossing has become [for some] a matter of life and death" (2011: 138). Studying United States and EU border police respectively, Williams (2015), Pallister-Wilkins (2015a), and Pallister-Wilkins (2015b) show how humanitarian logics constructing people as "at risk" have intertwined with border police practices that more traditionally treat migrants "as a risk", resulting in complementary care and control practices that transform both police practices as such, and relations between certain humanitarian actors and police authorities. The humanitarian border "re-orients border practices around the provisions for particular forms of life and introduces explicitly humanitarian actors into the borderwork assemblage" (Reece et al. 2017, 6). As Ruben Andersson (2017) argued in relation to the Spanish-African frontier, there is a need to understand certain humanitarian measures as part of a broader set of mobility and border controls. Focusing on the past work of Doctors Without Borders in the Moroccan-Spanish context, Lorena Gazzotti (2019, 18) wrote that, "operating in the interstices of border containment, humanitarianism therefore becomes a tool to protect a life that has become structurally degraded", pointing to the bestialization of Black lives. Following these analyses, we consider humanitarian interventions at the border not as a sideline, but as central to border processes (Cuttitta 2015). The central role humanitarianism plays at the border results however from the functional interdependence of the care and control nexus: "care functions as a technology of border enforcement as it is contingently provisioned in ways that pull the sick and injured into the enforcement regimes in new ways" (Williams, 2015, 17). As analyzed by Sahraoui (2020b) and Tyszler (2020), the logic of care and control at the border brings about gendered constraints for racialized migrant women. Here, we further unpack the genealogies of the moral ideas that sustain these practices of discipline and control of migrant women's lives.

Genealogy of Religious and Medical Forms of Humanitarianism
Even though the scope of humanitarian actors is diverse, we argue with Antonio Donini that "humanitarians perform functions inherited from colonial administrations and religious institutions" (Donini, 2010, 5227). Historically, medical and religious bodies have played a prominent role within colonial interventions. Whereas the contemporary language of humanitarianism developed primarily as a mode of intervention in post-colonial contexts, early forms of humanitarianism hark back to the colonial period. Humanitarianism is thus to date imbued with colonial legacies and humanitarian interventions tend to reproduce old-fashioned tropes of moral superiority (Brauman, 2005). The hierarchy of donor and recipient at the core of humanitarianism thus needs to be analyzed in the light of certain continuities as to how humanitarian actors exert disciplining and controlling power over the beneficiaries of their actions.
Christian missions were part of European colonial enterprises (Prudhomme, 2004) and in various contexts collaborated or instigated the control of the bodies and sexuality of the colonized (Stoler, 2002). It is important to take into account the missionary and colonial genealogy of today's humanitarianism, and notably medical humanitarianism. As Guillaume Lachenal and Bertrand Taithe argue: "Recent historians of the missionary phenomenon have (...) highlighted the important medical role played by Protestant and Anglican missions. Although significant, the role of Catholic missions has been largely underestimated, both from a medical and development perspective (...) the relationship between the Church and medicine was clarified from 1925 onwards by Pope Pius XI himself, who recommended the use of medicine in missionary work. 'Heroism is no longer enough; science must be used to support the missions' (...). The duration of the missionary presence, its nature and the gradual inculturation of its practices make it one of the most interesting actors in the history of humanitarianism." (Lachenal and Taithe 2009, 45-46, authors translation).
The critical literature on migration and borders has largely neglected the study of the role of religious organisations in border management. Yet, the Catholic Church needs to be taken seriously as a major actor in the migration/border regime (Tyszler, 2020). Migration has long been a concern of the Catholic Church but it was from the 19th century onwards that migration pastoral care was established, becoming a major papal concern in the 2010s (Geisser, 2018). Studies have already shown, in some contexts, a "Christian monopoly of assistance" to "Sub-Saharans" trying to reach Europe (see Etienne and Picard 2012 on Egypt). The Catholic Church is an obligatory stopover on the clandestine routes to Europe. Religious organisations, especially Catholic ones, are often present where other actors are not. Thanks to a close and extensive network through the implantation of churches all over the world, they are present where NGOs cannot be because of sensitive security conditions. Thus, from Gao to Tangier via Tamanrasset, the Catholic Church sees and assists Africans trying to reach Europe. Christian organisations, while demonstrating their own dynamics, are fully integrated into the humanitarian borderscape: in the field, they can stay while the more militant NGOs leave or cannot establish themselves, as it is easier for them to get along with local authorities and they are therefore tolerated as long as their repertoire of action remains humanitarian, focused on beneficiaries' physical survival and (formally) non-political (Fassin 2005;Ticktin 2011). Using a sociological and ethnographic approach, Elsa Tyszler's work (2019bTyszler's work ( , 2020 shows that the religious humanitarianism practiced in a border town in Northern Morocco (near the enclave of Melilla), mainly engaged in interventions concerning medical issues, leads to even greater constraints on the mobility of people who are supposedly being "helped", and thus contrary to what is claimed, reproduces a racialized and gendered order at the border. Her research within a religioushumanitarian organization reveals how its missionary members constituted a "police of morals and intimacy" (Tyszler, 2020) especially with migrant women "beneficiaries", reconnecting this contemporary violence to history and underlining the coloniality of such humanitarianism.
As noted by Alison Bashford in her volume "Medicine at the Border": European powers were "in practice intervening through a range of colonial and occupying structures, and through a discourse of "civilization" whereby another state's incapacity to be "civilized" in a "sanitary" sense was a justification for intervention" (2014:3). In tracing the genealogy of the French humanitarian NGO MSF (Doctors Without Borders), Ticktin uncovers an (a priori) paradoxical relation between colonial humanitarianism and a universalism that continues to justify humanitarian interventions. She argues: "drawing on the universalist rhetoric of 1789 regarding the right of all people to basic freedoms, republican ideology inspired the French to take measures to liberate Africans from indigenous forms of oppression they believed to exist, which included not only forms of African slavery and "feudalism", but also disease" (2014: 126-127). Although (partially) defeated from the justification of its civilizing mission, contemporary medical humanitarianism continues to rely on the idea of a superior moral obligation to rescue certain populations construed as "Others".

Humanitarianism, Gender and Maternalism
As briefly reviewed above, a growing literature explores the many dimensions of humanitarianism and a critical strand of the literature has engaged with its colonial genealogies. While the study of the gendered implications of humanitarian intervention toward migrants and refugees remains limited (Phillips, 2009;Sahraoui, 2020a;Sahraoui, 2020b;Tyszler, 2019b;Tyszler, 2020), building on these previous studies we seek to further trace the moral ideas that shape humanitarian care toward racialized migrant women. In this regard we see maternalism as a cornerstone of women's humanitarian engagement across time and we identify continuities between colonial maternalism and contemporary forms of humanitarian care. Colonial maternalism was partly embodied by European women missionaries. Claude Prudhomme notes the little interest given to women missionaries: "The writing of the history of the Christian missions has long reserved a secondary place for the role played by women (. . .) Whether religious or lay, the woman missionary is assigned in literature to play an auxiliary role. A paradoxical situation, as has often been noted, since the 19th century, women have been in the majority in the missions (. . .)." (Prudhomme, 2014, 3, authors translation). With women's missions taking shape from the 19th century onwards, missionary work took a different turn, with humanitarian activities focusing on women and children. As historians explain, priests found themselves dependent on the nuns for accessing women (Bouron, 2014) and mothers whom they believed could facilitate family conversion (Curtis, 2010). For European governments who justified colonization by a 'civilizing mission', women missionaries were a perfect alibi: their schools, pharmacies, clinics and charity offices made the argument that the French were improving local living conditions indisputable (Ibid). The sisters often specialized in medical activities with colonized women and in their education. Looking at the activities of the Congregation Notre-Dame d'Afrique in colonies (Haute-Volta and Gold Coast) of West Africa between 1912 and 1960, Bouron (2014) mentions the responsibility of the women missionaries in preparing colonized Christians and catechumens to become faithful wives and devoted mothers. They trained the women in housekeeping, gave advice on childcare and suggestions on conjugal life. "Whatever the case, missionary ethnocentrism gives the nuns the task of embodying a family and maternal model largely inspired by the European bourgeoisie. In spite of their "suspect social identity", linked to their single status, the White Sisters imposed themselves as paragons of an ideal femininity, made of devotion, charity, application and obedience." (2014: 56, authors translation) Ethnographic examples in this article show that this type of colonial maternalism has persisted in the post-colonial present through religious humanitarianism. But colonial maternalism, however, is not limited to the issue of women missionaries. During colonial times some European feminists actively took part in the civilizing mission aiming at "emancipating" colonized women from their seemingly backward practices (Farris, 2017, 211).
The historian Margaret D. Jacobs (2009) explores the role of white mothers in advocating for and organizing the removal of indigenous children in the United States and Australia at the end of the 19th and beginning of the 20th century. She demonstrates how maternalist activists supported women's role as primary carer in the family (defending for instance white women's custody rights) while at the same time they believed indigenous mothers to be incompetent at mothering and advocated for the removal of these children from their families. Colonial maternalism was thus anchored in racialized beliefs and used to justify intervention into the most intimate dimensions of indigenous peoples' lives. As noted by Jacobs, the ideas behind such maternalism did not disappear with the end of formal colonial policies: "Like colonizing women around the world, white maternalists in the United States and Australia reduced the heterogeneity and intricacy of indigenous gender relations to a homogenized image of indigenous women as the oppressed victims of their tyrannical men. This appears to have been an early variant of and historical precedent for what the feminist theorist Chandra Mohanty calls "the Thirld World Woman", a creation of Western feminists in the 1970s and 1980s that similarly positioned non-Western, non-white women as always and everywhere the powerless and dependent victims of male violence, patriarchal families, and male-dominated religions" (2009: 117-118). These frames and imaginaries continued to weigh upon social policies and practices targeting racialized women in Western postcolonial spaces. Aihwa Ong's ethnography carefully unpacks how American health, police and social workers intervened in gendered ways in the lives of Cambodian refugees in the US. She argues: "Feminist-infused refugee love-which goes back to the work of good church women among poor people at home and abroad -is merely the latest transformation of the civilizing mission to educate and "uplift" populations considered not yet quite American" (Ong, 2003, 133). This is also what decolonial feminist philosopher Maria Lugones suggests conceptualizing the coloniality of gender (Lugones 2008), by denaturalizing gender as well as race, which Anibal Quijano does not do when he elaborates the notion of the coloniality of power. Based on the work of the Nigerian researcher Oyèrónké Oyewumi (1997), who analyses the Western imposition of gender as having entrenched a strongly binary understanding of gender in the world, the concept of coloniality of gender emphasizes that non-white women are animalized, hyper-sexualized, and that their supposed lack of emotion justified sexual violence during the colonial enterprise, with these stereotypical constructions having consequences to this day. Across the myriad of colonial, humanitarian and postcolonial interventions carried out by religious actresses, women humanitarians and women social workers toward women construed as radically "Other", there appears to be a continuity in the moral ideas that justified and legitimated such actions. Though crucial differences also characterize this set of actresses -for instance some drawing their moral assumptions from religious sources and others from medical scientific knowledge as our article illustrates -we aim here to foreground the relevance of the concept of maternalism to grasp the specifically gendered power relations that continue to unfold between humanitarian women and racialized migrant women.

METHODOLOGY: STUDYING BORDER ACTRESSES AT THE HUMANITARIAN BORDER
The data on the Moroccan side of the border emerges from PhD dissertation fieldwork (Tyszler, 2019b) carried out for two and a half years, between 2015 and 2017, mainly in Morocco, and in Ceuta and Melilla. Elsa Tyszler used observant participation in several NGOs working with migrants from Central and West Africa. In addition, she conducted about 160 interviews with a plurality of actors around the border: about 80 with migrant women and men mostly from Central and West Africa; the rest with NGOs workers, activists; staff of governmental and international organisations and institutions. The data used for this article in particular is the result of an ethnography conducted within a humanitarian-religious organization working at the border that she joined as a volunteer-with the explicit purpose of research-in summer 2017. During this fieldwork, she was able to observe in particular the interactions between missionary sisters and the migrant "beneficiaries" of their humanitarian interventions. The differential treatment of migrant women (in comparison with male migrants) by the nuns became apparent very quickly. In addition to observation, semi-directive interviews were conducted with religious and humanitarian actors and actresses involved in the organization's project. The biographical approach revealed more specifically the interweaving of gender and race in the experience of illegalized migrant women and men at the border. Collecting life stories was indispensable in order not to limit the analysis to their experiences of violence and to examine the production of political subjects.
It is useful to present some details relating to Tyszler's positionality as a young French white woman scholar. In Morocco, she was aware of her dominant status in relations of race and social class, and of the asymmetry of power in the investigator and respondent relations that she could institute whether she liked it or not. Her French and white privileges allowed her access to the border-field without physical risk when others were at risk of violent repression at any time. This also enabled her to take up positions (paid and unpaid) in NGOs-mostly managed by Europeans -on several occasions and with ease. Her female gender was also an asset for staying in a highly monitored field, as it was associated with less suspicion and considered non-threatening, both by local authorities and by the various actors at the border. Despite the absence of an ethics committee in her university (and more generally in France), throughout the research, she strived to maintain high ethical standards by ensuring that her presence in the field and her interactions caused no harm to research participants.
The data upon which the analysis relating to the Spanish side of the border relies stems from fieldwork conducted between August and October 2016 and in January 2017. This fieldwork was carried out by Nina Sahraoui in the framework of the European project EU Border Care exploring the politics of maternity care at European borders. Fieldwork inside the Center for the temporary Stay of Immigrants (CETI) was authorized by the Ministry of Employment after the authorization request was submitted by the director of the Center. At the academic level, the research project received ethical clearance from the Ethical Assessment Committee of the European Research Council Executive Agency based on the application submitted by the project's PI.
During her stay in Melilla and in addition to fieldwork conducted with various NGOs and within the maternity ward of the hospital (see Malakasis and Sahraoui, 2020), Nina Sahraoui spent six weeks visiting the CETI, usually remaining for several hours inside the Center, at different times of the day. During these weeks she conducted 12 interviews with healthcare professionals providing primary care to residents of the Center as well as additional interviews with several social workers including socalled "integration workers" (integradoras), social mediators (mediadoras) and interpreters. These workers were women in their overwhelming majority. Elsewhere Sahraoui refers to the healthcare professionals working in this setting as medical humanitarians (Sahraoui, 2020b). Here, we are particularly attentive to humanitarians' own gender and their gendered roles, given that the role of "humanitarian actresses" at the border has been little researched so far through the lens of gender.
In parallel, Sahraoui conducted 18 interviews with migrant women residing in the Center, most of them were either pregnant at that time or had recently given birth. These interviews usually took place in the public spaces of the Center, at times inside the residents' room to have some privacy if these were empty (this however occurred rarely given that rooms were occupied by up to eight persons). The difficulty in finding a space for interviews illustrates one of the points made below as to the absence of spaces for intimacy in the daily life of the Center. This absence was all the more salient in that the immediate surroundings were not fit for finding such spaces either: situated a few hundred meters from the international border with Morocco, neighboring some private golf courses, and about an hour walk from the city, the Center and its surroundings were situated in a military zone. The women interviewed on the Spanish side of the border were from Syria (7), Algeria (7), Morocco (3) and Yemen (1) and they had been living in the Center on average for about 3 months (between four days to 11 months). A French citizen with a Moroccan surname, Nina Sahraoui's positionality was in some ways plural. As a researcher involved in a European research project, she was able to access different local administrations and NGOs. At the same time, while she clearly found herself in a privileged position -by being in Melilla for work purposes (and thus being economically independent) and, importantly, by being able to cross borders at liberty-her position was not entirely that of a white European owing to her surname and the use of some Arabic during interviews which affected how she was perceived by the research participants. To ensure anonymity all names appearing in the article are pseudonyms.

THE POLITICAL POWER OF ALLEGEDLY UNPOLITICAL MORALS: GENDERED DIMENSIONS
Through ethnographies from the interior of two different organisations, we seek to highlight the gendered control of mobility performed by the humanitarian regimes in place at the Moroccan-Spanish border. In this section, we explore how these organisations shape the border experience of racialized migrant women, hindering their intimacy and imposing embodied forms of control, in multiple ways.

Spaces of intimacy and humanitarian control
Spaces of privacy are often scarce or non-existent for illegalized immigrants in border areas. In our cases, the women we interviewed were living in self-built migrant camps located in the forests (before being hosted by a religious-humanitarian organization) on the Moroccan side of the border, and in a Center for so-called "temporary stays" on the Spanish side. In these contexts, intimacy was strongly constrained, in different ways for these women. Our investigations reveal how humanitarian interventions contribute to restricting women's intimacy, despite the aid provided and even sometimes the prowomen rhetoric. On the Moroccan side of the border, the organization under investigation is an entity of the Northern Diocese of the Catholic Church. Its "project for migrants"-which focuses solely on "sub-Saharan migrants" (in fact from Central and West Africa) -was launched at the end of 2012 following, on the one hand, the proposal to take over MSF's activities in the border region and, on the other hand, the increased involvement of the Catholic Church and its various branches in migration issues and in particular in direct assistance to refugees. A Spanish priest assigned to the Catholic Church in the city appealed to the different congregations of missionary sisters who would send people to work on the project. During an interview, a nun explained how the missionaries were recruited to work on the project in the border town: " They don't send African sisters because otherwise they risk becoming part of the clan of victims, along with the migrants. The wounds of colonization must be healed". Without claiming to study missionary dynamics, this statement suggests a form of ethnic-racial selection in sending sisters for this project. There is the fear that Black African nuns would identify too much with migrants who are repressed at the border and would thus lose any sense of distance, which would make it impossible for them to carry out their religious/ humanitarian work, unlike Western nuns whose neutrality is supposedly obvious in this context. An annex to the church was built over time to house the offices and some facilities for temporary accommodation; several people were hired to make up the staff of the organization. The team consisted of several missionaries (mainly Spanish but also South American 4 ), two young Spanish aid workers in positions of responsibility, some local Moroccan workers (a social worker, two drivers, a cook/housekeeper for the reserve and a security guard) and finally two West African workers hired as "mediators" with "sub-Saharan migrants". The staff carried out different types of interventions, including "medical" interventions in the camps based in the forest, which consisted of visiting people who called the organization's emergency number in the early morning for health problems and possibly taking them to hospital. In addition, "social/humanitarian" interventions were deployed in the same camps, but to "map"-i.e., to see where the migrants were, who they were, in what proportions-to visit them, to identify needs and/or to distribute clothing or food from time to time. Follow-up visits to the hospital were carried out, as well as activities with the people accommodated in the (small) residence, which included seriously injured people, women who had just given birth or very sick people who could not stay in hospital. It is precisely in this residence that could be observed the interaction between the missionaries and the women being "helped". The rooms in the residence were supposed to be the only safe space for these women at the border, places where they could find privacy, regain intimacy. But the control of the missionaries could turn them into yet another place of violence and not places of healing.
An Ivorian woman who had given birth to dead twins was brought back to the residence. She was visibly in a state of deep sadness and exhaustion. The sister in charge of the emergency rooms accompanied her to the one that would be hers. They entered the room: there was a bed and...two empty baby cots. When I mentioned to the sister the fact that the two cradles could make this lady feel worse, she said 'ah yes... but we don't know where else to put them'. Later, the sister said to this bereaved woman who had no appetite and who preferred to call some of her relatives on the phone in her room rather than join the others for dinner: 'you must come and eat now. If you don't respect the rules of the house, you leave. I'm going to take the phone away from you.' (Field notes, border town in Northern Morocco, June 2017).
This scene of symbolic violence illustrates well the dynamics of the missionaries' power relations over the migrant women admitted to the emergency rooms of the residence. Here, not only does the nun not respect the double mourning experienced by this woman, but she also reinforces the pain by not caring about what is for her a logistical detail of furniture storage. This extract also shows the processes of infantilization through the reference to the rules to be respected and the threat of the confiscation of the telephone. The racialized infantilization can be observed more generally on a daily basis and with regard to all their "Sub-saharan beneficiaries" (men and women combined) as this extract of conversation with a nun illustrates: "They need to be kept busy (...), you can't have them sitting around doing nothing. You can play games with them (...), they like to play a lot. Dinner must be served at 7 pm. At seven o'clock everyone has to be there. They have to respect the rules, here it's not just anywhere (...) Often the Sub-Saharans complain because there is not enough or they don't like what we give them, but they have to make do with what we have, here it's not a hotel (...) Then you prepare their milk, they love milk with chocolate. Then they can help you with the washing up. Then when you leave you close the door of the Residencia, ok?" (Conversation with sister Anita, border town in Northern Morocco, June 2017).
However, the maternalism of the sisters seemed even more violent toward women beneficiaries, through attitudes that are both racist and sexist. Daily observations highlighted the fact that the sisters showed sustained empathy for men injured at the barriers or attacked by civilians, while a certain disdain or even contempt seemed to be reserved for women who had just given birth (or death) at the border, echoing what Ticktin (2011) wrote about women victims of sexual violence in the DRC, as not appearing innocent enough in the eyes of humanitarians. "She's going to get over it, these women are strong you know" said the sister about the bereaved woman, continuing her day's activities. This kind of rhetoric was also heard about the sexual violence experienced by migrant women at the border, considered to be part of the African culture and/or the bad morals of these women traveling alone, the nuns drawing in this way the contours of a shameful femininity of these migrant women on the border (Tyszler 2020). The missionaries' words quoted above echo the colonial figure of the African woman, naturally strong and resistant, linked to an animalizing conception of Black women's bodies (Dorlin, 2006;Pourette, 2010). These attitudes illustrate the coloniality of gender (Lugones 2008), in which non-white women continue to be seen as incapable of the same emotions as white women, minimizing the violence, even sexual, and pain they experience. This research highlights that the nuns never questioned their practices and attitudes around the women they "helped" at the border, thus performing the a priori unshakeable legitimacy of their work toward their "beneficiaries". The issues of femininity and motherhood are central in producing the racialized and gendered figures of the Western missionnary-humanitarian vs. the African migrantbeneficiary. This was also illustrated by the many scenes of informal "maternity lessons" given by the missionaries to the women housed with their newborn babies: "Don't go in the living room with your baby"; "you have to put your baby back in his cradle"; "go and feed your baby he's crying"; "stop carrying your baby all the time it's no good", said the missionaries, in the corridors or when entering the women's rooms. In this way, the maternalism performed by the missionaries was not only hindering the little space of intimacy that the women could have had in this border space, but also attempting to influence the ways in which mothers relate to their newborns and how these women deal with motherhood. These "maternity lessons" were not perceived by women as respectable advice from elders, but as a way of infantilizing them ("they treat us as if we were children") and "imposing" their "Western way of doing" on them (in their own words). These ethnographic observations and the women's comments are reminiscent of what Bouron (2014) writes about missionaries in the colonies of Upper Volta and the Gold Coast, responsible for turning African women into "devoted mothers" and who constantly deplore the alleged parental laxity that, according to them, is at the root of the deviant upbringing of African children. The following paragraphs reveal how humanitarians on the other side of the border performed some strikingly similar patterns of disciplining and infantilization, in spite of the a priori very different identities of the two humanitarian organisations studied here.
On the Spanish side of the border, the humanitarian actors involved in managing and providing assistance in the CETI were social workers, medical personnel, lawyers as well as NGO workers from partner organizations that intervene inside the Center for specific activities (kindergarten, health-related workshops, Spanish classes, sports activities). The longstanding collaboration of several humanitarian organizations with the administration of the Center, under the supervision of the Spanish Ministry of Employment, illustrated the embeddedness of humanitarian activities in border management and represented a case in point of the workings of a "humanitarian border" (Walters, 2011). Many of these workers were long-term employees, with several years of professional experience, and thus their humanitarian intervention was characterized by a certain routine, notwithstanding the usual connotation of emergency and adhoc intervention that the notion of humanitarian action entails. Against this background, managing life inside the Center relied on discipline as a technology of power (Foucault, 1975(Foucault, /1995 whereby humanitarian workers were granted material and symbolic influence over various aspects of the residents' lives. A commonly heard response to the interviewed women who raised some concern with social workers asserted this hierarchy clearly: "It's not a hotel here, you can leave if you want", echoing what some of these women might have heard on the Moroccan side already. Requests to change rooms following a conflict with some of the women in the room, or any logistical matter around daily life in the Center were addressed or neglected at the social workers' discretion. Some of the social workers were specifically designated by the title of "integration workers" and tasked with organizing life inside the Center through tasks such as managing the distribution of rooms. These logistical activities were nevertheless understood in highly symbolic terms and performed with educational overtones by being framed as a preparation to integrate into European societies. While rooting the social and medical workers' supervisory power over residents' lives in moral values, the narrative of integration had little to do with residents' actual daily lives and perspectives (see Sahraoui, 2020a for an analysis of the functions performed through the integration discourse). All residents experienced uncertain administrative statuses, and some faced the risk of forced removal after being transferred to the peninsula if they had been issued an expulsion order.
Against this background, the control exerted over migrants' daily lives and mobility was framed by humanitarian actors in benevolent terms in order to be morally justifiable. For instance, if cooking was absolutely forbidden inside the rooms it was "for the residents own good" to avoid any accident happening. Beyond the paternalism/maternalism of dispossessing individuals of their autonomy, some mechanisms of control fulfilled a symbolic function aiming at the display of humanitarians' moral commitment to general principles such as gender equality. Yet as a matter of fact families were systematically separated in the Center and children slept with their mothers, often in the same bed, thus increasing room occupancy rates; as one integration worker stated: "babies sleep with their mothers obviously". Women and children shared rooms (with eight single beds, and potentially more persons if several children were also in the room) on one side of the camp and men shared even more crowded spaces by being accommodated under big tents on another side of the Center. Notwithstanding this initial separation, integration workers foregrounded that they tried to reunite families after the birth of a baby in order to foster the father's involvement and support to the mother after the birth. Paradoxically, while the Center organized the separation of the family in the first place, the intention to reunite families for a brief period of time was presented as a measure to promote gender equality. The promotion of a nuclear family model is part and parcel of the practices of social workers toward communities construed as "others" in the colonial period and its aftermath. In the context of the French colonization of Algeria for instance, Neil MacMaster's study of the politics of women's emancipation during the war argued: "the campaign by the French army for the emancipation of Algerian women offered to displace the 'traditional' Muslim family and gender roles by a particular western model of the couple and companionate marriage" (MacMaster, 2009, 178). These normative beliefs continue to permeate social policies toward racialized migrants, Ong analyzed for instance the meanings underpinning policies and interventions within the Cambodian community in the US: "Implicit in these social-work strategies was the construction of undeserving patriarchal subjects and deserving female and child victims of patriarchy, who should be taught such American values as autonomy in decision making, gender equality and nuclear family relations" (Ong, 2003, 133). Going back to the CETI, these statements -though inscribed in such a genealogy -seemed to serve a discursive purpose rather than a reformist objective. The women interviewed complained that the criteria for being granted such a temporal family reunification remained unclear and rather than a general rule this practice hinged in their eyes on the integration workers' relations with individual families. Indeed, while one integration worker proudly stated that she gave a room to a couple with a newborn for 40 days, stressing that this would ensure the husband's participation in caring for the baby, another integration worker mentioned much shorter periods of time, one or two weeks, granted according to rooms' availability. Even though the reunification policy was far from being systematically implemented, it served to assert the integration workers' educational role against the background of a discourse that characterized Muslim migrants as difficult to relate to, one of them stating: "The Syrian women you can't communicate with their husbands, they have a low cultural level. The pregnant Syrian women they have their first child at 15 or 16, they don't know how to attend to their baby so we monitor them". Another integration worker specified that their educational mission included making sure that the women were adequately dressed and with their hair in order when outside of the bedroom, for instance when joining the collective meals distributed in the Center. To further build on the parallel with Jacobs' historic enquiry into white maternalists' interventions among indigenous peoples, she analyzed: "The intimate lives of indigenous people-the ways they cared for and raised their children, their dwellings, their sexuality, their marriage practices, their gender relations, even the ways they adorned their bodies and styled their hair-eventually came under the scrutiny and condemnation of their colonizers" (2009: 24).

Mobility, Bodies and Reproductive Health Under Control
Their significant differences notwithstanding, by bringing together our two case studies it appears that both religious and medical humanitarians exerted a certain control over women's bodies and lives on both sides of the Moroccan-Spanish border.
Pushing their control practices even further into women's intimacy, the research on the Morrocan side of the border also shows how the missionaries can exert strong moral pressure on pregnant and postpartum women (Tyszler, 2020). In both cases, the missionaries try to influence women in the making of decisions about their bodies and lives, and to orient their mobility away from the border.
"Of the girls I had here and who gave birth here, only one, a Cameroonian, I managed to convince her to go home with her baby (...) But at the beginning she didn't want to go home at all (...) she told me 'it's that my mother, what is she going to say', and blah, blah, blah, blah... but face it! Tell her, 'Look, Mom, it's not my fault, they raped me', it's not her fault! Even if the custom... listen, is that the African custom? And it is the same Africans who rape them. I tell her: 'If it doesn't get out, if nobody knows, how it's going to end?' So, she talked to her mom and she called too. I told her 'Yes, ma'am, see, it's not just that your daughter is deviant, not at all, it's rape, that's what happened'. And she understood it, and she said: 'let her come, I will take care of her'. The girl didn't want to go back at all at first, she wanted to go to Spain! And then she finally convinced herself that... it was difficult, to go to Spain with a child... and finally she returned to Cameroon. We helped her with a little savings (. . .). IOM repatriated her (...) But... they are rare those who dare to take a step like that!" (Interview with Anita, missionary, border town in Northern Morocco, June 2017).
This excerpt from the interview reveals the multiple logics interwoven into the sister's behavior toward the women accommodated in the residence and, more broadly, provides information on the purpose of the humanitarian interventions carried out at the border by the organization in question. We see here that the nun, through a racist grid of analysis, uses the observation of patriarchal violence perceived as "cultural" and the repressive context of the border as a means of encouraging women to return to their country of origin. The sister's speech about the only woman she allegedly succeeded in sending back to her country invites us to think that the organization's interventions are not simply humanitarian in the sense of material and medical assistance to the most "needy" migrants at the border. The objective of encouraging return is highlighted in this excerpt, i.e. a practice in line with the migration control policies in place. These observations raise the question already asked by other researchers such as Jennifer Hyndman in her study of the work of the United Nations High Commission for Refugees (UNHCR) in Somalia and Kenya during the 1990s: "At what point do charitable acts of humanitarian assistance become neocolonial technologies of control?" (2000: 147). Although hidden behind the moral and religious values by which they define their actions with migrants, their practices are eminently political, engaging with the dynamics of race and gender relations governing mobilities and the modalities of blocking and crossing the Moroccan-Spanish border. In other words, the humanitarian-religious organization also ensures that the established migratory order is respected, in its own way. Just as the role played by women missionaries in colonial history has always been underestimated, including by religious women themselves (Lachenal and Taithe 2009), the sisters we met always seemed to minimize the weight of their actions toward the targeted migrants, behaving both as simple witnesses and as little hands providing necessarily legitimate help. However, nuances must be maintained so as not to essentialize the figures of these missionaries whose positions remain plural, dynamic and ambivalent. While violent practices emanating from those encounters have been described, the research also reveals the politicized discourse they may hold, denouncing the "murderous migration policies" and the "hypocrisy of governments" that ally themselves in the "massacre" they can observe, in part, daily at the border. The sisters may encourage the return of women because of the saturation that the missionaries may feel in the face of the situations of extreme distress that they observe daily at the border. But they never seem to take into account the violence they themselves generate. In the course of the research it was found that the sisters never go very far in discussions with the migrant women (and not only because of language difficulties), quickly deciding on their situation, as if they knew better what is right or wrong for them anyway. Another problem that remains is that these actresses often simplify the understanding of the border situation for many people and organisations, as they are one of the few humanitarian organisations operating on the border. The partial and biased vision given by these missionaries is often reflected in the media or political positions, especially on the Spanish side, which depoliticize and exoticize the multiple violence suffered by migrant women from Central and West Africa on this border. On the other side of the border, medical humanitarians also engaged in forms of mobility control on the basis of their own understanding of what was appropriate and thus moral for the migrant women they attended.
In the accommodation Center in Melilla, while medical humanitarians ensured access to primary healthcare and organized perinatal and maternal care for pregnant women, they were also involved in managing women's mobility (Sahraoui, 2020b). With mobility framed as a risk for pregnant women, medical personnel at the Center could delay a transfer to the peninsula by several months. It remained unclear what was the exact stage of pregnancy after which travel was not possible anymore, since different actors provided different answers, yet it seems that it could have been as early as the 7 th month of pregnancy. Adding to that a mandatory period of 40 days after birth, a pregnancy could significantly prolong the time spent in the Center, while all women interviewed were hoping to leave as soon as possible. For these migrant women, pregnancy led to a biopolitical control of their bodies, one however that was enmeshed with the administrative control of their mobility. In her research at Australian borders, Kristen Phillips argues that 'There are the bodies that cannot have a political status, that are involved in the reproduction of life, and biopower's task is to make them live by taking direct control of these bodies ' (2009: 143). An illustration of this approach is nurses' decision not to allow transfer for pregnant women in a relatively advanced stage of pregnancy since "transfer" involves spending several hours on a boat. One of them commented: "Because what we would do is to put the baby in danger, even though there are many mothers who prefer to take that risk and leave, but we won't put them on a boat for 8 h [sic, the crossing lasts about 6 h], something could happen." As a matter of fact, the women interviewed were deeply affected by such decisions in the light of the material conditions of life in the Center and their desire to be able to continue their migratory journey rather than being stuck for several months in the enclave. Spending several hours on a boat, when still weeks ahead of their due date, did not represent a major risk in their eyes. While for medical humanitarians their decisions were strictly medical, they were taking on a responsibility with decisive implications for these women's mobility, performing a humanitarian maternalism that deprived these women of the possibility to decide for themselves. A certain understanding of motherhood underpinned these practices, carried out mostly by professional women in their quality as humanitarian helpers. The concept of maternalism helps us grasp the gendered dynamics at play between women humanitarians and racialized migrant women. Indeed, Western/White women have played a crucial role in reaching out to "other" women, in colonial contexts, humanitarian settings and postcolonial politics of integration (Stoler, 2002;Ong, 2003;MacMaster, 2009). The shared traits of these policies reside in an alleged moral superiority of a certain version of femininity and gender roles to which other women need to be introduced, denying these women the legitimacy of making decisions for themselves. Admittedly ferry companies, similarly to airlines companies, follow certain guidelines for transporting pregnant women (a medical certificate might be required over 36 weeks or a limit applied at that stage), yet contrary to regular travellers who can travel up to that limit, migrant women in the Center could be made to wait at different stages of their pregnancy. They were indeed dependent on a formal police authorization that could be influenced by the Center's administration based on different grounds, including pregnancy. Coated in benevolence and protection, these decisions reproduced an infantilization of those who were deemed in need of protection, and this dimension of humanitarianism harks back to its colonial genealogy of an intervention grounded in "knowing what is best" for the beneficiary. While constraints were pervasive, migrant women we met carved out some space for intimacy and sought to achieve as much control as possible over their lives and mobility, as explored in the following section.

MIGRANT WOMEN'S RESPONSES TO HUMANITARIANS' GENDERED CONTROL OVER THEIR DAILY LIVES
On both sides of the border, despite their illegalization and humanitarianization, i.e., their subjugation to a humanitarian governance based on highly unequal relations of power, the women we met displayed agency by creating opportunities to regain spaces and moments of intimacy, and keeping control of their bodies and their lives -playing or outwitting the humanitarians' game.

Claiming Back Spaces of Intimacy
Given the architectural configuration of the residence on the Moroccan side, the people staying there -unless they are in a bedroom with the door closed -can be observed at all times, since narrow windows in the parish space overlook the main corridor where people stay most of the time; and they can also be observed from the floor above. The sisters enter the rooms without knocking, especially women's rooms. Moreover, the kitchen space is closed outside meal times and people are not allowed to cook themselves. The whole residence is locked at night "Because otherwise they can run away, and then it's to avoid trafficking" explained one of the missionaries. These observations illustrate the permanent tension between the dynamics of care/ protection and surveillance/control (Pallister-Wilkins, 2015b;Pallister-Wilkins, 2017;Williams, 2015;Isleyen, 2018) structuring the activities of this organization. In spite of a space that is monitored and constrained by strict and infantilizing rules, keeping the people who are "helped" in subordinate positions, the research identified strategies deployed by some women to thwart some of the rules set. The example of the purchase of a chicken illustrates this. Marie B., a Liberian woman staying in the residence explained: "You know here they give us food that we are not used to, it even gives some people digestion problems. So sometimes we each contribute to buy a chicken. We give the money to one of the Moroccan drivers who is nice and goes to buy it for us outside because we can't go out. Then we cook it during the weekends when the sisters are not there".
In this humanitarian setting, it is considered that the people being helped should consider themselves lucky to be given food. The choice of food is considered a luxury and not a right or a necessity resulting from different dietary practices or health reasons such as for women who have just given birth, or sick people who need/want to eat in a certain way. Buying a chicken and cooking it "African style"-as she said-is a way for Marie B (and those who taste her dish) to regain some control over her daily life managed by the religious humanitarians who act as if they know better what is good or necessary for her and her peers. The "chicken trick" is thus a way of confronting the racist humanitarian maternalism that demeans them. Choosing what to eat and being able to cook are a way of regaining a certain intimacy with oneself.
Similarly, in the Center in Melilla, crowded rooms, shared showers and toilets, a few tables and benches for hundreds of residents, offered little space for families' and individuals' intimacy. In this context, the practices of disciplining and control described in the previous section added further incursions into migrants' intimacy. And yet, residents' daily practices created spaces of intimacy at the margins of these intrusions. Though neither kitchen appliances nor food was allowed in the rooms, many had at least a boiler. This allowed residents to drink tea together and mothers to prepare bottles for their babies. On one occasion for instance a group of Syrian women was having tea together when Fadwa, a 25-year-old Syrian woman, came back to the Center after giving birth at the hospital. Without bending the rules, there would have been no space for intimate encounters among women. Again while the restrictions were always framed as in the best interest of the residents-so that they "enjoy the well-being provided by a clean room" as one integration worker put it-it also supposed the impossibility of most basic social encounters, having tea or coffee with someone inside the Center. Similarly, an informal "market" inside the Center, held by some women, created autonomous spaces of interaction that escaped the control of the administration. Isolated from the city and its shops, this very limited system de facto escaped the Center's strict regulations. Securing some moments or spaces for intimacy thus asserted these women's autonomy. They also demonstrated their agency by taking decisions and making their own choices within and in spite of the constraints that weighed upon them.

Resisting Humanitarian Governance, Making Choices
The research also reveals how women have been able to resist missionaries' control practices regarding the issue of motherhood and projects of mobility. On many occasions scenes of migrant women laughing with each other and discreetly avoiding the informal "maternity lessons" given by the sisters that seemed irrelevant to them were observed at the residence. There were also situations in which women turned down help from the organization when the rules seemed too rigid and inappropriate to them. The case of Aya C., a young Malian woman who had just given birth, is an interesting example of resistance faced with humanitarian maternalism. Aya wanted to give her child up for adoption for various reasons. In addition of being afraid of what her family would say if they would find out that she had a child out of wedlock, she explained that she was afraid not to be able to cross the border with the child, that she had already "lasted too long" in Morocco, that she couldn't stand life in the forest camps anymore. She said she couldn't take care of the baby, as her milk was cut off because of a misdiagnosis of HIV at the hospital. She stated that if she had known soon enough that she was pregnant she would have had an abortion, that she didn't know who the father was because she had been raped. If one of the nuns seemed understanding, another one did everything to dissuade her, arguing to Aya that since the baby was Black, he would be mistreated in Morocco, "This is your first child, it's a grace from God, you have to take care of him". Despite all the pressure that the missionary put on her every day, Aya did not change her mind. After a while, her decision was taken, and a social worker began the process of having the child put in an orphanage. Aya's example shows that despite an extremely constrained situation, the women assert their right to make decisions about their bodies and their lives, as they also did on the other side of the border in many ways.
Not only were women creating spaces that somehow escaped the regimented control that internal regulations imposed on daily life in the Center, they also found ways to make and assert their choices, even though these remained constrained by their precarious statuses and limited financial means. While registered in the Center, Amina, a young Moroccan woman, pregnant at that time, and her husband, managed for instance to rent out a room outside of the Center so that they could continue to live together while they would have been separated inside the Center. They had to come every three days to pass their electronic cards at the entrance's turnstile so that they remained registered. Running out of funds after a few weeks they had to stay inside the Center, yet Amina was glad she had managed to spend some time with her husband and had been able to choose what she ate for some time. Most pregnant women in the Center deplored the lack of diversity in the foods served in the canteen, and notably the absence of fresh fruits or vegetables. Similarly to the views of religious humanitarians, here too the provision of food was considered as a favor good enough in and of itself, and complaints were met with incomprehension at best, disdain at worst. Amina was the only person interviewed who had spent some time outside as it seems that such arrangements only concerned a very small minority of migrants who go through the Center. As Amina and her husband were Moroccan, and thus had only crossed one border, they had some funds upon arrival in Melilla, which might not be the case for the great majority of migrants who by the time they arrive in Melilla have already traveled for months, some for years. Syrians had for instance often already lived outside of Syria in refugee camps in Lebanon or elsewhere before they attempted the Western Mediterranean route and had thus left their homes for several years already. Amina's case illustrates however that in spite of the costs endured by living partially outside the Center, this option was favored as long as it was possible in order to avoid the strict regimentation of life in the CETI.
To take another example, some women also sought additional medical resources outside of the care provided in the Center. Medical personnel inside the Center provided primary care to residents but they also managed and coordinated specialized care with doctors outside the Center and with the city's hospital. Importantly, this gave residents the opportunity to access healthcare free of charge. Yet in the context of the power of medical personnel on women's mobility, this appeared to be particularly problematic in some instances. Khawla for instance, a Yemenite woman, was informed upon arrival that if she was 8 months pregnant, she would not be able to leave, not until after giving birth. She knew however already that she had been granted a police authorization to be transferred to another Center in mainland Spain and thus if retained in Melilla this would be because of the pregnancy. As she had to wait for the hospital appointment given through the Center, she was seeking obstetricians in the private sector so that she would have a chance to leave before it was too late. Again, while such a solution could only be available to those with some financial means, it shows how some residents sought resources outside of the humanitarian assistance in order to escape decisions that those managing humanitarian services made over their lives and their mobility.

CONCLUSION
This article brought together two ethnographies initially conducted separately by the two authors in the light of the striking similarities observed within gendered patterns of humanitarian governance on both sides of the Eastern Moroccan-Spanish border. Even though the authors met with women of different nationalities owing to the racialized logics of migration control and crossing at this border as recounted in the introduction, humanitarians on both sides of the border exerted power over illegalized migrant women's bodies and mobility by curtailing their intimacy and interfering with their family life, notably by organizing their daily life in their places of accommodation and by imposing or influencing their decisions and possibilities for mobility. We argue in particular that such a humanitarian "police of morals and intimacy" (Tyszler 2020) harks back to the missionary and colonial genealogies of humanitarianism owing to the fundamentally unequal power relations that define the humanitarian relationship. Importantly, the moral discourses mobilized by women humanitarians, both religious and medical, revealed gendered and racialized assumptions constantly reproducing the figure of the bad mother who needs to be educated for the good of her (future) child. We trace this form of power exerted by humanitarian women over women who are othered and marginalized to the logics and representations already encapsulated in colonial maternalism. At the same time, migrant women's agency shone through in both our studies through a myriad of practices of resistance. The disciplining and controlling power of humanitarian assistance was resisted and contested in both small daily interactions and through strategies seeking to secure resources and support outside of the humanitarian settings.
By unpacking the moral underpinnings of a postcolonial and humanitarian maternalism, we foreground the moral hierarchies that come to justify intrusive interventions into African and Middle-Eastern migrant women's lives. On the Moroccan side of the border, the maternalism of the missionary sisters revealed a conception of women not as subjects of rights, i.e., as an end in itself, but rather as the means of an intervention whose ultimate beneficiaries are the (future) children. As a result, in the context of these interventions, women are subject to the interest of humanitarians mainly as mothers, and their rights and duties are also seen as dependent on this naturalized role. Humanitarian assistance on the Spanish side of the border, while embedded in a discourse of benevolent integration contrary to the situation encountered in Morocco, equally enacted and reproduced stereotypical tropes as to racialized migrant women's inadequate family and mothering practices. In both cases, this humanitarian maternalism does not escape the historical weight of colonial maternalism that continues to shape certain representations on the one hand and that is to be found in the unequal power relationships enacted by the humanitarian hierarchy of the donor vs. the recipient of aid on the other. The humanitarian maternalism performed on migrant women from Africa and the Middle East on both sides of the border reinforces the naturalization and hierarchization of the gendered and racialized categories of humanitarian actress/migrant woman through the constant updating of the dichotomies of good or bad motherhood, of rightful or shameful femininity. Keeping migrant women "beneficiaries" in a subordinate position allows humanitarians to constantly reaffirm their raison d'être and maintain their privileged positions in the social order. These case studies are thus inscribed in a continuity of representations and practices from colonial maternalism to femonationalism (Farris, 2017) whereby the figure of the racialized woman is depicted as oppressed, non-agentic and in need of benevolent assistance. While humanitarian assistance does provide in the short term crucial material support to migrant women, it is important to critically examine the moral ideas upon which these interventions rely in order to promote relations that would be more respectful of migrant women's intimacy and decisions in terms of reproductive health, family life and mobility.

DATA AVAILABILITY STATEMENT
The datasets presented in this article are not readily available because Data sharing is not applicable to this article as no quantitative datasets were generated during the study. The qualitative datasets are not publicly available for the respect and protection of the research subjects. Requests to access the datasets should be directed to Nina Sahraoui, nina.sahraoui@ cnrs.fr; Elsa Tyszler, elsa.tyszler@cnrs.fr.