Behind the Curtain of the Border Spectacle: Introducing ‘Illegal’ Movement and Racialized Profiling in the West African Region

: The introduction of ‘illegal’ migration in West African countries represented a major conceptual policy shift for societies that were historically characterized by intra-regional free movement. However, this transformation went along with severe allegations of racialized proﬁling of undocumented migrants in many West African societies. De Genova’s concept of the ‘border spectacle’ describes how the presumed ‘illegality’ of migrants is made spectacularly visible in Europe, thus producing a criminalized and racialized portrayal of migrants. Nonetheless, this work argues that today’s illegalization through a racialized representation of migrants has been extended beyond Europe’s boundaries and behind the spectacle’s curtain towards countries of migration origin. Drawing on the cases of Mauritania and Mali, this paper considers their fundamentally opposite reaction to the introduction of ‘irregular’ movement and illustrates the inherent problematics of transferring the ﬁgure of a racialized migrant into the West African region. Particularly successful in countries with a history of ethnic conﬂicts, this process essentially externalized European border practices of racialized proﬁling. On the contrary, this analysis concludes that the presence of established patterns of regional movement and cross-border habits made it undesirable to either introduce the policy concept of ‘illegal’ migration or to adopt its potentially racialized portrayal.


Introduction
In recent decades, the illegalization of migrants in their countries of origin became a new and powerful tool of Europe's border externalization policies. One of the main approaches used by European authorities comprised an unconsidered introduction of the concept of 'illegal' migration in West Africa, a region traditionally linked to the free movement of its inhabitants, which went along with severe allegations of creating a racialized profile of the newly criminalized migrants. Depending on the countries' history of ethnic conflicts and the overall relevance of migration in society, the illegalization of migrants may risk becoming a dangerous instrument in the hands of local elites against ethnic minorities. This raises the important question of how the cooperation on migration, between European and West African countries, aimed to introduce the idea of border controls and to illegalize migrants in a cultural context where these elements were historically absent. The processes that governed these policy goals, however, did not produce the same results in all countries and were subject to contrasting outcomes. The two neighboring West African countries of Mali and Mauritania have been selected to highlight the fundamental differences in how local authorities and societies acted in response to the possibility of cooperating with European states. The marked divergence between the reactions of these two countries provides, moreover, interesting insights into the role played by local civil societies, diasporas, and ethnic relationships in the context of West Africa and its migration-related customs.
During the 1990s, the European Union pursued a strategy of strengthening its external borders and increasing surveillance mechanisms with the aim of reducing arrivals of African migrants on its territory. Despite these efforts, while borders were progressively closed and the paths for legal entry restricted, European governments were soon forced to recognize the failure of traditional migration control policies as migrants continued to subvert these limitations by shifting their itineraries (Papadopoulos et al. 2008). Thus, European policymakers identified the need for a new and broader approach to overcome the sole fortification of its external boundaries. Authorities progressively manifested their interest in the itineraries used by migrants to design the appropriate actions to intervene on their journeys before these could reach Europe's external borders. As a result, European Member States quickly begun to involve third countries in this new migration control strategy. Ultimately, this project, made of deterrence and containment, aimed to reduce the numbers of migrants arriving in Europe and asking for asylum (Ruhrmann and FitzGerald 2016).
The overarching narrative behind migrant's journeys, as told by media and politicians, is commonly one of small, overloaded boats with 'desperate' migrants. The European public has been fueled with a restrictive and securitized view on migration and asylum. This narration describes migrants as illegal border trespassers, invaders, and criminals who are only in search of their personal economic gain. Furthermore, this image of the 'illegalized' migrant is reinforced through the so-called 'border spectacle', where their presumed 'illegality' is made spectacularly visible by producing a criminalized and racialized portrayal of the sub-Saharan migrant (De Genova 2002). Migration itself has come to be viewed as a threat, urging the deployment of a growing number of measures to hinder people from arriving on European territory. Hence, an increasing combination of police and military means went along with a new strategy of international cooperation to externalize border control functions and establish a powerful 'global' migration regime.
In the first decade of the 2000s, the European migration strategy focused on the externalization of borders and outsourced its border patrol activities to third countries such as Mauritania, Mali, and many other West African states. These countries begun to cooperate with the new European migration regime by accepting varying degrees of different measures such as readmission agreements, joint patrol activities, and the introduction of regulations aimed to control and restrict cross-border movements. The outcome of these instruments caused some countries to implement a dangerously racialized illegalization process far away from the European 'border spectacle'. In fact, seen from a Eurocentric point of view, processes of illegalization now are produced, but not enacted, behind the border spectacle's curtain in countries of migration origin and of transit.
The article will begin with a description of how processes of illegalization and securitization had a significant socio-political influence on European societies and policymakers. In this sense, the first section will be dedicated to the representation of migrants as a criminalized threat represented through a racialized portrayal of the sub-Saharan migrant. Successively, to provide the necessary background of the processes behind the European border externalization, the first part will focus on the origin and the main passages which led the European border control cooperation with third countries. Nonetheless, while it is not the scope of this study to retrace the complete evolution of the European policy framework, attention will be given to the aspects which are helpful in highlighting the history and the instruments which paved the way for the European border externalization in the West African region.
Finally, the second part of this article offers a more detailed focus on the impact of the European border externalization on West African countries and migrant's itineraries. This will be achieved by examining the very different response of the states of Mauritania and Mali to the cooperation on migration control with the EU and its Member States. The contrasting reaction of these states to the introduction of the concept of 'illegal' movement draws the attention to the importance of several factors which are at play in third countries. In particular, the presence of a considerable diaspora, the relevance of migration in the local culture, and the previous history of ethnic conflicts may represent decisive factors in the decision-making of third countries. In conclusion, this investigation suggests that the European efforts to outsource border controls-through the introduction of policies contrasting 'illegal' movement-were especially successful in those countries where the 'illegalization' of migration was domestically paired with a racialized portrayal of migrants.

European Borders between Securitization and Illegalization
The development of border externalization and of 'remote-control' policies has been examined from many analytical perspectives which looked at the relationship between the EU's external dimension and the cooperation with third countries (Schöfberger 2019;Gabrielli 2016;Adepoju et al. 2009;Boswell 2003), on state and non-state actors involved (Ostrand and Statham 2020;Andersson 2014), the impact of border externalization instruments on the integration process of the Economic Community of West African States (ECOWAS) (Idrissa 2019;Uzelac 2019;Robin 2009), or the complicated negotiation of Mobility Partnerships (Reslow 2012). As scholars pointed out, the issues associated with border externalization practices and with how we frame their impact have predominantly been approached from a Eurocentric gaze, which produced a prejudice when investigating these complex structures (Cuttitta 2020;Adam et al. 2020;Natter 2018). Undoubtedly, while it is necessary to analyze the developments that occurred on the side of the externalizing states, without the response and decisions of the 'receiving' side, any research in this field would be affected by a strong bias in its premises. Indeed, as will be discussed in the last section, the agency and the interests of West African countries played an active role in determining the outcome and the shape of border externalization policies. States might very well refuse to agree to most of Europe's conditions or, instead, accept them willingly in order to foster their own domestic and international position, as occurred in the case of Mauritania. West African countries, in fact, negotiate EU migration policies while aiming to obtain an advantage from the circumstances that this cooperation may generate (Adam et al. 2019). Certainly, this does not mean to ignore the unbalanced power relations that resulted from the end of the Cold War order, which transformed the nature of political negotiations between Europe and Africa into what Bauman (2002, p. 89) described as a constant "bargain-by-force".
Today, nevertheless, it is generally accepted that West African countries are not simply passive recipients of the decisions adopted by European governments. Rather, these states pursue their own agenda and have the power to negotiate with Europe on specific terms that they agree upon. It is necessary, from this point of view, to keep in mind that the interest of West African countries in EU migration policies can have conflicting and dangerous outcomes. As will be discussed in more detail in the final section, the EU-while being concerned only of its security-oriented agenda-maintained a neutral position in relation to the potential misuse of their policies in third countries. In particular, it will be argued that European states have been responsible for not considering the role of fragile and unbalanced domestic power relations in third countries, such as a previous history of ethnic tensions and conflicts. As a consequence of this dangerous behavior, European migration policies and their resources have been reported to cause, in certain countries, an increase in intra-African refoulement practices and deportations, as well as locally instigating racialized human rights abuses against foreigners and local minorities (Andersson 2014;Trauner and Deimel 2013). In certain circumstances, such as in the case of Mauritania, cooperation with European countries resulted in what has been termed the 'numbers game', which aimed to demonstrate the efficiency of migration control activities against alleged migrants in countries of origin or transit. As discussed in the section dedicated to Mauritania, the search for 'illegal' migrants built upon previous racialized tensions and led to arbitrary deportations of local citizens of the Haratine ethnic group (Andersson 2014). On a regional scale, the increase in deportations which followed cooperation with European Member States is threatening the construction of an ECOWAS area for free movement (Uzelac 2019).
The assessment of the potential impact of EU migration policies, however, depends on a country-by-country examination. The following analysis discusses two divergent responses to Europe's goal of introducing the concept of 'illegal' migration into the West African region. Before continuing in this direction, it is important to provide the context behind the development of European policies and of the increasing role of border externalization policies in the cooperation with third countries.
The concept of border externalization encompasses a broad spectrum of activities and policy instruments which, in general terms, can be framed as a process where a state's migration policy is expanded to actively involve a third country's engagement. Hence, this article adopts the definition, which describes border externalization as "based on the direct involvement of the externalizing state's border authorities in other countries' sovereign territories, and the outsourcing of border control responsibilities to another country's national surveillance forces" (Casas-Cortes et al. 2015, p. 73). In this sense, externalization has the purpose to 'stretch the border' through the multiplication of institutions and actors responsible for stopping and controlling movements in countries of migration origin and of transit. Restrictive visa policies, joint border patrols, police cooperation agreements, economic and material support of border capacity building efforts, and the introduction of the concept of 'illegal' entry and presence constitute only a representative sample of the growing number of instruments adopted to enforce border externalization and the outsourcing of border control activities. It would be a mistake, however, to imagine these externalized European borders as devices aimed only at reinforcing exclusion-or as impassable walls. The EU is not sealing itself off. Rather, it created a differentiated access system to supply the needs of the European labor market, which restricts certain movements while easing others (Bojadžijev and Karakayali 2010). The procedure of differentiating movements refers to the function of borders as 'filters' (Bauböck 2017) characterized by a steady increase in their "selective permeability" (Walther and Retaillé 2015, p. 192). Instead of losing importance, Mezzadra and Neilson (2013) notably argued that we are assisting in the multiplication of borders and in their capacity to select between desired and undesired movements.
From this perspective, the concept of 'liminal porocratic institutions' fits very well with the current development of European borders and its definition provides a particular analytical value to the extension of regimes of control to cooperating third countries (Papadopoulos et al. 2008). The 'liminality' of these institutions expresses their capacity of being flexible and, when required, to continuously adapt its border configuration. Whereas 'porocratic' indicates their primary function as regulatory means of borders' porosity and to operate the crucial distinction between desirable and undesirable movements. In the context of externalizing border functions, these institutions are adaptive and fluid systems of control which aim to follow and govern the mobility patterns of migrants.
Framed as a security issue, over the last few years, migration has been portrayed as a threat to the integrity of European communities. Migrants arriving via boats and pateras from the southern shore of the Mediterranean became the dangerous actors of an alleged 'invasion of aliens' in Europe. This negative view of migration was fueled through an extensive use of the so-called securitization rhetoric: a discursive strategy used by media and politicians to create the general perception of a threat coming from outside-while lacking the evidence to support such claims-which, in turn, produced an emotive response demanding an increase in 'security' (Buzan et al. 1998). Thus, if we look at the development of new means to control migration, the process of securitizing migration shifted the domestic perception of risk from the national to a transnational scale and transformed the state's security objectives from the local to the external level (Popescu 2015).
The portrayal of immigrants as a danger originating from outside the European community is reinforced by their description as subjects whose presence is automatically assumed as illegal. Media representation constructed a narrative around what the anthropologist Nicholas De Genova (2002, 2013 called the 'border spectacle', a concept which describes how the 'illegality' of migrants became an immediate label for most immigrants-especially in the case of sub-Saharan migrants in Europe. De Genova (2002, 2013 highlighted how the enactment of a racialized migrant, whose 'illegality' is made visible and perceived as a 'natural' representation, is produced through a 'spectacle of enforcement' at the border. Therefore, the spectacle of border infringements has created a mediatic performance to induce the public opinion to feel a need for protection and security against the arrival of migrants (De Genova 2013).
The criminalized representation of people as de facto illegal for their physical presence plays an important role in defining the social and political response of the public opinion and policymakers. Scholars like Harald Bauder (2014) highlighted the need to focus on the language and suggested that we use the term 'illegalized' in relation to the institutional mechanisms which made the sole presence of migrants 'illegal' while obscuring the variety of their conditions and personal histories. Indeed, the success of labels like 'illegal' and 'irregular' has raised significant linguistic concerns, as they imply a pejorative and stigmatizing trait on those who enter a country without legal permission (Cernadas 2016). Moreover, these terms suggest the feeling of crime and of being wronged, causing powerful repercussions on the perception of the public opinion. The constant use of 'negative' migration-related terminology, as made by media and politicians, fueled the description and narratives in relation to migrants' crossings of the Mediterranean to reach Europe. Among the consequences of this kind of rhetoric, the resulting border spectacle occurring at Europe's frontier produced a specific form of 'illegalization', which led to a racialized portrayal and criminalization of sub-Saharan migrants (Gazzotti 2021;De Genova 2018).
In the process of externalizing border control functions to third countries, the introduction of the concept of 'illegal' movement-and, in parallel, of promoting a new process of 'illegalization'-caused a significant problem which entailed several risks in those countries who were culturally used to free movement. As Ruben Andersson (2014) described in his work, Illegality Inc., what he calls the 'illegality industry', border externalization, and the 'illegalization' of migrants constitute a 'value chain', where migrant illegality is not only produced but also made lucrative. Besides the direct interests of states, economic benefits from this cooperation go to a number of actors like police forces, institutional authorities, and political elites, which all contend this increasing flow of money coming from the EU and its Member States.
The representation of the border spectacle between African and European shores is only the most visible and outstanding part of the illegalization process, which, in the meantime, expanded far beyond the external borders of Europe (Bialasiewicz 2012). As Ruben Andersson (2014, p. 6) notably recognized, today, the border spectacles' "vigilantes are but a sideshow: instead powerful border regimes seek to keep the undesirables out". While the 'spectacle' is still ongoing, especially on the Mediterranean Sea, and perpetuating the image of a racialized illegal sub-Saharan migrant, it is behind the curtain of the European border spectacle that new processes of illegalization are unfolding. Indeed, the metaphor of the curtain provides a useful perspective as it helps us to shift the focus to what is hidden from the spectacle's spotlight, namely what happens to migrants during their journeys and experiences before reaching the Mediterranean shores. From a European perspective, the curtain conceals the new spaces and forms in which the process of 'illegalization' is being enforced-but not enacted-through the externalization and outsourcing of the European border regime in most North and West African countries. Besides the expansion of the condition of 'illegality', migrants in West Africa also face a status of permanent deportability, a concept originally developed by De Genova (2002), with the risk of being arrested and deported at any time and in any place during their journeys.
The European narrative behind these voyages produced a general perception of migrants moving along direct routes from their point of origin till they are able to reach their country of destination. However, as Crawley et al. (2016, p. 5) pointed out, framing these itineraries "as linear, singular uninterrupted journeys or flows of people heading toward Europe is grossly misleading". In contrast with the common idea of the African continent of being on the move, actual migration from sub-Saharan countries towards Europe is minimal and proportionally negligible (de Haas 2008; Andersson 2014). Those politicians and media who argue that African and sub-Saharan migration is fundamentally directed towards Europe fail to recognize the pre-eminence of intra-regional mobility between West African countries. In fact, West African cross-country migration significantly surpasses the movements to Europe, and, moreover, West and North African countries are themselves important destination countries (Andersson 2014). These movements mainly take the form of seasonal migrations-for example, of itinerant merchants and workers in the agricultural sector-and thus provide an important contribution to family livelihoods in the region. Moreover, with the emergence of specific manufacture productions, such as the fishing industry in Mauritania, these movements also provide the necessary supply of skilled workforce for many countries (Dünnwald 2015). Many of this work-related seasonal movement originated under French colonial authority and remained part of the local custom also after the independence of West African countries. Border crossings were not regulated and informal migration was never perceived as 'illegal', neither by the population in the region nor by the states themselves (Bensaâd 2008). It is worth noting that a state like Mali, with a population of 12 million people, has an estimated number of 4 million citizens living outside its territory. However, of this apparently large amount-from a Eurocentric point of view-more than 3.5 million emigrate for seasonal work to other West African countries and only a small amount of around 200.000 Malians live in Europe (Trauner and Deimel 2013). Indeed, it is essential to acknowledge that migratory movements in the West African area have their destination in the region itself. Afterwards, unemployment or dangerous circumstances may constitute some of the many reasons which can lead individuals to induce a successive mobility towards neighboring countries. Authors such as Hannah Cross (2013, p. 8) called this phenomenon a 'stepwise' migration, which is often produced by a "contradictory mixture of coercion, compulsion and repression combined with choice, opportunity and mobility".
The complexity of intra-regional mobility, as we will see in the following sections, highlights the severe misunderstanding of how European authorities perceive movements in West Africa. Eventually, the mistaken belief of migrants being primarily on the road to Europe led to the controversial policy response which adopted the so-called 'migration routes' approach (Poutignat and Streiff-Fénart 2010). Recognizing this inherently problematic issue, the distinction offered by Casas-Cortes et al. (2015) between 'routes' and 'itineraries' will be used in the next pages to provide for conceptual clarity. The former, as used by the European Global Approach to Migration and Mobility (GAMM), pertains to the control strategies implemented by the European migration regime to capture and organize movements into linear and definite routes. Instead, itineraries represent the non-linear movements and the agency of migrants that constantly obliges European border authorities to be flexible and recalibrate their spatial frameworks. Thus, 'routes' is used to express the perspective and the actions of the European migration control regime, whereas 'itinerary' and 'movements' relate directly to the migrants' practices.

European Border Control Externalization in Third Countries
Before moving to the cases of Mali and Mauritania, in this section, the border externalization will be analyzed from the point of view of the externalizing agents, the European Union, and, more importantly, of its Member States. This will provide the required preliminary background to the different stages of the frequently incoherent relationship between European and West African countries in the field of migration control cooperation.
In 1997, the European Treaty of Amsterdam established for the first time a supranational competency over migration and asylum policies. Prior to this major event in European history, migration was a sovereign domain of Member States and the main instrument available for cooperation with third countries to control international migration was to negotiate bilateral agreements (Bialasiewicz 2012). In 1999, the European Council of Tampere was tasked to define the content and the approach to be implemented by the EU for its new competencies. The Council of Tampere, therefore, highlighted the necessity of a 'global vision' aimed at actively including cooperation with third countries to manage migration at all levels. Moreover, the global vision also recommended to reduce the reasons to move by focusing on what European authorities called the 'root causes' of migration-such as poverty, unemployment, and underdevelopment-in countries of origin. Underlying this process, the EU embraced the 'migration and development nexus' as its key conceptual framework for its future European migration policy strategies. Based on 'push-pull' neoclassical theory, this approach identifies international migration as an outcome of wage and development imbalances between countries (Papademetriou and Martin 1991). From the problematic 'push-pull' point of view, the 'root causes' of migration may 'push' people to migrate and search for employment in another country. To address the conditions in third countries of migration origin, since the Council of Tampere, cooperation in migration control became officially a central part of the EU's strategy (Harding 2012).
After the Council of Tampere, the EU began to structure its approach and envisage the future of its strategy. In 2002, the informal meeting of the Ministers of Justice and Home Affairs held in Santiago de Compostela promoted the development of a 'global plan to combat illegal immigration' which considered the inclusion of countries of origin and transit as a fundamental objective (Araujo 2011;Papadopoulos et al. 2008). First actions consisted in the deployment of Immigration Liaison Officers (ILOs) tasked with the collection of information on migration itineraries in North and West African countries, as well as providing technical and economic support to fight human trafficking in these states. During these years, however, the main debates in Europe concerned the relationship between the requests made to third countries to be more active in controlling their emigration movements and the provision of development aid-a tool which has been increasingly misused as a bargain to force negotiation with third countries. This debate peaked at the Council of Seville in 2002, where authorities from Spain and the United Kingdom proposed the reduction of development aid to those countries who were deemed not conforming with borders control duties as laid out in bilateral agreements (Araujo 2011). While this proposal was rejected, the Council agreed to regularly assess compliance with these agreements and to consider sanctions which, nonetheless, were not to be targeted at development aid.
During the following years, migration from third countries lost its prominent position in the domestic agenda of most European countries, leading to a significant delay in the development of a common approach to migration. In 2005, this situation was due to change when the so-called Asalto Masivo marked a crucial turning point in the future of European migration management. Fourteen migrants were shot dead by police forces as thousands of migrants tried to jump the border fences of the Spanish exclaves of Ceuta and Melilla (Andersson 2014). After this episode, Morocco raided the migrant camps near Ceuta and Melilla and brutally expelled their inhabitants to the desert or to their countries of origin. The Asalto Masivo and the inhuman repression of Moroccan authorities caused a general outcry in the European public opinion. Regardless, in the meantime, Morocco displaced thousands of migrants southward along the Atlantic coast and caused a major shift in the itineraries taken by people who tried to reach the Canary Islands (Casas-Cortes et al. 2016). From that moment, most migrants chose to embark further south on fragile Cayucos and Pirogues from Mauritania, as alternatives to the Western Sahara coast, a much longer and life-threatening journey.
During the same year, the events of Ceuta and Melilla convinced the European Council to hold an informal reunion at Hampton Court (European Council 2005). This meeting, in light of the emotional response to the Asalto Masivo, led to the development of the first guidelines of a new migration policy, which resulted in a formal draft of the European Commission for the next European summit (European Commission 2005). At that point, authorities in the EU were mostly convinced of the importance to develop a border policy capable of tracing migrants' itineraries and translating them into manageable routes (Casas-Cortes et al. 2016). Eventually, the proposal was approved in 2005 and became the comprehensive and holistic migration policy framework known as the 'Global Approach to Migration', later renamed as the current 'Global Approach to Migration and Mobility' (European Commission 2011;European Council 2005).
For the first time, the Global Approach to Migration introduced a common policy framework that aimed to replace the previous bilateral agreements with a new form of multilateral cooperation with third countries of migration origin and transit in order to efficiently transfer the responsibility of border control activities away from Europe's external boundaries (Casas-Cortes et al. 2016). The vision behind this approach was to implement a flexible and rapidly adaptive system to control migration flows according to the shifts in migrants' itineraries. To achieve this objective, migrants' itineraries were extensively mapped and turned into governable routes, where their mobility patterns became divided into categories of countries of transit, origin, and of arrival (Poutignat and Streiff-Fénart 2010). The overarching framework of the European Union can be synthesized in two main pillars (Boswell 2003). On one side, it envisaged a security-oriented approach whose main goal is the overall reduction of-undocumented-migrant arrivals on its territory. On the other side, it was also designed to promote a preventive approach through economic development in countries of origin as, by doing so, Europe sought to tackle the root causes behind migration.
To implement the new strategy involving third countries, the EU started to negotiate 'mobility partnerships' (MPs), a new form of multilateral agreement in the field of migration control. These partnerships entailed the European security agenda, requiring cooperating countries to sign readmission agreements, strengthen their border control capacities, and to increase their effort against undocumented migration. Besides providing the necessary training, resources, and equipment to implement this strategy, the EU on its part offered third country citizens new legal migration channels for work and study purposes in Europe (Reslow 2012).
Spain became an early precursor and the leading country in the development of agreements for the externalization of borders and the outsourcing of border controls. Till the 1990s, the Spanish border regime with Morocco resembled very much those with its European neighbors. However, the Schengen agreement, which established a zone of free movement between European states, required the harmonization of the Member States' immigration policies towards third countries (Nessel 2009;Gabrielli 2011). The EU was aware "that Ceuta and Melilla are vulnerable flanks of Fortress Europe" and Spain begun with building fences along a two-fold barrier around its exclaves on the African continent (Harding 2012, p. 148). In the early years of implementation of the Schengen area, academics such as James Hathaway (1993) warned that the harmonization of migration policies and the increased difficulty to cross borders would make it challenging for refugees to receive international protection on European territory. However, the EU continued to strengthen its external borders and, in 2005, created the European Agency for the Integrated Border Management (Frontex).
After implementing controls at the Spanish-Moroccan border, itineraries across the Strait of Gibraltar became increasingly difficult and eventually led migrants to choose longer and more dangerous journeys. Soon, migrants moved south, embarking on small Cayucos and Pirogues from the Mauritanian shores to the Spanish Canary Islands (Nessel 2009;Cross 2013). Consequently, following the arrival of the first Cayucos on the Canary Islands, Spanish authorities negotiated bilateral cooperation with Mauritania to establish joint patrols with Frontex in the Atlantic. From that moment, Spain begun to largely involve third countries in the West African region in its migration strategy. In 2005, Spanish authorities officially adopted the so-called 'Plan Africa', an approach aimed at setting up cooperation agreements in the area of migration policy with North and West African countries who in turn benefited from increased funds for development aid and of quotas for temporary or circular migration schemes (Dünnwald 2015).
Over time, the EU adopted various policies and soft tools to enhance its capacity to manage migration movements before these were able to arrive at its external borders. Framed under the concept of border externalization and the outsourcing of border controls to third countries, it is useful to highlight the most significant instruments used by Europe in the process of transferring control functions to West Africa and, more importantly, shifting the focus away from the European borders.
First, it is important to remember how the adoption of visa requirements is perhaps the main and most powerful tool available to a country for implementing immigration control (Cuttitta 2015). The establishment of a rigid visa system is consistent with the border reworking process and is directly aimed at constructing a selective and differentiating access to the EU by allowing entry only to 'desirable' individuals. Thus, the EU drew Schengen lists of third countries to define whose citizens require or do not require a visa permit to enter Europe. In this way, while a potential migrant is still in their country of origin, the Member State's consular offices become responsible for examining each application and to decide whether to grant or deny the possibility of accessing Europe.
The negotiation of readmission agreements has progressively gained a fundamental role in European migration policies as it is a mandatory requirement for Member States to deport undocumented migrants. While readmission agreements are commonly bilateral contracts, the EU's effort to create a multilateral approach aimed at implementing readmission into its new Mobility Partnerships. In general, readmission agreements have evolved over the years and, today, depending on the type of obligation, they are commonly divided into two categories. In the first typology, the signing country is committed to accepting only the return of its own citizens. The second typology, instead, the so-called 'second generation' agreements, are negotiated between the EU and countries of migration transit. These new agreements introduced the specific obligation to readmit also individuals who hold a different citizenship of the signing state, as long as it can be proven that the migrant has passed through its territory. The first readmission agreements were signed between Morocco and Spain in the early 1990s, whereas from the early 2000s, these arrangements have been extended to other sub-Saharan and African countries (Nessel 2009).
Alongside these soft tools, the EU promoted border police cooperation with third countries to support patrolling activities of migration routes, to control travel documents, and to train local border forces (Papadopoulos et al. 2008). Until the early 2000s, migrants' most common itinerary from sub-Saharan countries targeted Morocco. This changed soon after, when Spanish authorities decided to sign an early bilateral agreement on police cooperation with Morocco (Gabrielli 2011). Since 2004, Spanish and Moroccan police also begun to conduct regular joint patrols and coordinated their efforts to contrast undocumented border crossings. After migrants' itineraries shifted southward, Spanish and Mauritanian authorities signed an agreement to deter undocumented migration and enhance migration controls (Andersson 2014). In a short time, airplanes and radar technology were supplied by the Spanish Government to the Mauritanian police in order to monitor international waters between Mauritania and the Canary Islands (Nessel 2009).
Hence, a new defensive and lethal border control system was established with the deployment of Frontex joint operations to prevent vessels from reaching European territorial waters near the Canary Islands. These joint operations are characterized by the leading position of one European Member State, with Frontex serving as the coordinating agency, and employ a wide array of technological instruments, including satellite and drone surveillance, to control and stop migrants (Dünnwald 2011). Soon, many joint operations were deployed in the area between the Canary Islands and the African shores, in the Strait of Gibraltar, and several others in the Mediterranean Sea. In particular, as highlighted in the next section, from 2006, the operation HERA strengthened surveillance over the Atlantic Sea between the Canary Islands and the Mauritanian-Senegalese coast (Cross 2013). During these activities in international waters, international sea law obliges vessels to rescue migrants who are in distress and conduct them to the nearest safe harbor. When people are rescued in the European territorial waters, or if they disembark on European territory, migrants can claim for asylum while being protected by the nonrefoulment principle. To avoid this outcome, Frontex operations were mainly deployed in Senegalese-Mauritanian sovereign waters as then migrants could be directly handed over to African countries' responsibility. Notably, the case of the 'Marine One' boat, which was refouled to Mauritania, represents a significant example of the Spanish motives behind these operations. The Marine One was rescued in Spanish territorial waters while carrying around 400 migrants; the boat did not transit Mauritanian sovereign waters but, regardless, these migrants disembarked and were detained in Mauritania for several months (Bensaâd 2008). The European narrative, however, continued to justify these operations as being concerned with saving human lives. In this sense, Europe sees itself as a humanitarian actor and defends the necessity of aerial and maritime patrolling of the Senegalese and Mauritanian shores (Harding 2012). Nonetheless, it is arguable whether forcing migrants to abandon sea journeys for the more dangerous itineraries through the desert could be justified as a humanitarian act.

Shifting Itineraries and Border Externalization in North and West Africa
The impact of the fence jumps of 2005 at Ceuta and Melilla also had significant repercussions on the migrant itineraries and migration policies of the entire North and West African region. In the aftermath of the Asalto Masivo, the Moroccan police reacted by proceeding with violent roundups of sub-Saharan migrants and deported them to the village of Oujda at the border with Algeria (Andersson 2014;Dünnwald 2011). These deportations to the desert, which continued for several years, were strongly criticized and considered as "forced returns that take place outside any legal framework" (Lambert and Clochard 2015, p. 122). Furthermore, Moroccan forces have been alleged to rob, beat, and rape migrants before abandoning them in the desert of Oujda (Harding 2012). Sandra Gil Araujo (2011) gave a wider account of how deportations have been carried out by the Moroccan authorities. Police forces regularly executed raids in the Medinas, especially of Tangier, targeting the quarters where sub-Saharan migrants settled. Afterwards, migrants were treated with brutal force and were denied food and water before being loaded onto the trucks and sent to the closed border with Algeria. In other cases, people were sent south in the desert and left without supplies (Andersson 2014). The repeated appeals made by humanitarian organizations, who denounced these serious racialized violations of human rights suffered by sub-Saharan migrants, have fallen on deaf ears.
As a consequence, migrants soon considered this path as increasingly hazardous to choose and moved their itineraries further south, away from the Strait of Gibraltar and Ceuta and Melilla. Hence, Mauritania rapidly entered the spotlight of European migration management (Gabrielli 2016;Harding 2012). Migrants attempted to reach by boat the Spanish Canary Islands as these were located in the Atlantic Sea and 100-400 km away from the African coasts (Dünnwald 2015). First, migrants departed from the Moroccanoccupied Western Sahara coast, but the surveillance of the coastline against migrants was soon extended also to this area (Cross 2013). Therefore, between 2005 and 2007, migrants moved the starting point of their journeys to the Canary Islands farther south near the cities of Nouadhibou and Nouakchott in Mauritania (Nessel 2009). Pirogues and Cayucos were again the little boats used by thousands of migrants for their long and hazardous sea-crossings.
In the meanwhile, as the EU failed in its attempts to negotiate the first Mobility Partnerships with third countries, the European Commission and Member States begun their efforts to introduce the concept of 'illegal' movement in West African countries. Historically, the West African region has been characterized by its habitual cross-border movements, which still represent an important voice in the economy and the culture of the area. The introduction of a concept like 'illegal' border crossings, thus, did not exclusively regard the policy-level of adopting new regulations. The specific context of West African countries meant that implementing new border control systems and actively regulating movements encountered significantly varying degrees of interest. From this point of view, West African countries were able to exercise their agency and approached European pressures by evaluating the potential benefits or problems associated with their demands. However, as highlighted by the contrasting cases of Mauritania and Mali, the response could be subject to substantial differences and misuse. While the EU did not take into account the local background and, in particular, the potential impact of its policies on West African societies, the 'illegalization' of migrants risked exacerbating previous ethnic tensions in the region. In fact, as will be discussed in the case of Mauritania, the illegalization of migrants was achieved by implementing a racialized profile which targeted the Haratine minority. Nevertheless, ethnic tensions may not automatically lead to such outcomes. In the case of Mali, in the years preceding the civil war, while it was suffering severe ethnic hostilities, the country's specific relationship with migration and its active civil society represented some of reasons for which the concept of 'illegal' movement was not regarded as desirable.

Mauritania
Mauritania was one the very first countries in West Africa to become a policy laboratory for the introduction of the European border externalization. The change in migration itineraries and the country's specific position as an obligatory passage from West Africa to the Maghreb region and its relative proximity to the Canary Islands made Mauritania a strategic partner for Europe. Before this major transformation, however, migration in Mauritania has always been linked to traditional customs as, until the 1970s, the country was characterized to a large extent by nomadism (Dünnwald 2014;Bensaâd 2008). This aspect of the Mauritanian culture remained, during more recent times, as the state neither formally controlled migration nor considered undocumented border-crossing as something 'illegal'. When the itineraries of migrants moved south and the boats begun to depart from the ports of Nouadhibou and Nouakchott, Europe rapidly shifted its security-oriented attention to Mauritania's porous borders and the freedom of movement that it granted to foreigners (Poutignat and Streiff-Fénart 2010). The EU and the Spanish government recognized, therefore, the importance of promoting an artificial illegalization of the informal migration existing in Mauritania (Bensaâd 2008). In contrast to other neighboring countries-in particular, to Mali-the lack of a diaspora abroad influenced the behavior of Mauritania's governments, its population, and civil society organizations, who did not perceive migration as a relevant issue.
Mauritania was a member of ECOWAS, which created an economic space of free movement for goods, services, and to be progressively extended to people. However, Mauritania left ECOWAS in 1999, but maintained free movement agreements with Senegal and Mali. This choice had its well-founded reasons as informal and seasonal movements were fundamental in the lives of West African populations and the borders inherited by the former colonial powers arbitrarily divided pre-existing societies (Andersson 2014;Tamburini and Vernassa 2010). Thus, border crossings in this region were not perceived as illegal by the local populations, as they were part of normal life for many West African countries (Dünnwald 2015).
Mauritania is commonly categorized as a rentier state because the country's economy mostly relies on the rents deriving from its natural resources and its strategic contact position between North and West Africa (Cross 2013;Tamburini and Vernassa 2010). Furthermore, Mauritania's society is historically characterized by strong ethnic segmentation due to its history of widespread slavery, which still affects the current socio-political relationships (Dünnwald 2014;Bensaâd 2008). The country's population is divided between the black Haratine minority, who live mostly in the south of the country along the Senegal river, and the majoritarian Beidanes of Arab ethnicity, who live in the cities and the north (Andersson 2014). The Beidanes-Moors represented the political elite of the country and forcefully imposed their dominance of the poor population of the Haratine minority, which was increasingly pushed southward along the Senegal river. More recently, a conflict between Senegal and Mauritania in the 1990s became an opportunity for the racialized displacement of tens of thousands of Mauritanian Haratines to Senegal and Mali. While the conflict soon calmed down, the introduction of a figure of an 'irregular' migrant, which immediately became identified with people of black ethnicity, was considered to be at extremely high risk of inflaming once more ethnic tensions (Bensaâd 2008).
In the framework of the 'Plan Africa' and the wider European GAMM, the Spanish authorities coordinated their efforts to stop the journeys to the Canary Islands by offering development aid and equipment to Mauritania in exchange for their cooperation. In 2005, the Spanish Government increased the number of sea patrols around the Canary Islands, which, in 2006, became the cornerstone of the new HERA I operation coordinated by Frontex. Soon after, the operations HERA II and III followed and Mauritania was supplied with technological equipment to enhance its border surveillance and interception capacities (Cross 2013). Nonetheless, initially these measures failed to obtain their goal as arrivals increased to their maximum height of 30,000 people in 2006 (Dünnwald 2014;Harding 2012). Therefore, Spain quickly decided to extend its control activities to the Mauritanian territory and found there a conveniently favorable political context. During these years, authorities in Mauritania were particularly open to cooperation as the country was eager to regain international recognition after the two coups d'état of 2005(Poutignat and Streiff-Fénart 2010. In 2006, Spain and Mauritania negotiated two agreements which, moreover, also entailed the deployment of 250 Guardia Civil police forces in Nouakchoot and Nouadhibou (Dünnwald 2015). Additionally, Spain supported the Mauritanian Gendarmerie by sending surveillance equipment, such as helicopters and night vision goggles (Andersson 2014;Cross 2013). In particular, a detention center was established by Spain in the dismissed school compound no. 6 in Nouadhibou and was called Centro de Estancia Temporal de los Inmigrantes. Rather, this center was largely known and referred to as 'Guantanamito' by Mauritanian citizens and migrants themselves (Andersson 2014;Dünnwald 2015;Harding 2012). In fact, Mauritania at that time had no laws against illegal migration until 2010, thence, unfortunately making the name 'Guantanamito' fitting for this out of law condition. The former Malian minister for Culture and Tourism, Aminata Traoré (2007), argued that Africa was in the process of becoming a prison due to these detention centers built or financed by Europe on the continent. Furthermore, the Mauritanian shores were controlled by joint patrols between the Mauritanian Gendarmerie and Spanish forces. In particular, the EU enhanced Mauritania's border surveillance and migration control activities with EUR 8 million through the European Development Fund and the more security-oriented Instrument for Stability between 2008 and 2013 (Frowd 2014).
In 2009, the number of arrivals to the Canary Islands reduced substantially as the HERA operation was said to be responsible for halting almost entirely these journeys (Dünnwald 2011). HERA was praised for its ability to patrol and intercept migrant vessels and has been described by Frontex as its most successful joint operation to date (Frontex 2009). However, the self-proclaimed 'success' of the operation did not interrupt movements to Europe and did not save the lives of migrants. Instead, the only result achieved by Frontex's intervention seems to have made, overall, itineraries more perilous (Van Houtum 2010). Furthermore, the HERA operations and the bilateral agreements signed successively with Senegal continued to broaden the scope of surveillance and interception activities. Consequently, this made also the Senegalese coast troublesome to embark for the Canary Islands, and migrants partly moved further south on even more dangerous and longer itineraries from Gambia and Guinea to reach these Islands.
The European Union managed to successfully incorporate Mauritania into its migration regime for its strategic position as the country became the desired "buffer zone between Africa and Europe" (Cross 2013, p. 90). The expansion of migration control over the Mauritanian territory also entailed routine raids in Nouadhibou's migrant quarters. These roundups filled 'Guantanamito' with those presumed to be migrants intending to move towards Europe. Successively, these-alleged-migrants were deported to the Senegalese and Malian border, respectively, in Rosso and Nioro (Andersson 2014;Cross 2013). The latent racism between Haratines and Beidanes, as described before, was fueled by the introduction of a criminalized 'illegal' migrant (Dünnwald 2014(Dünnwald , 2015. This situation, caused by the Spanish and European intervention, led to a racialized surveillance regime which arbitrarily criminalized black Haratines as potential illegal migrants. In fact, the normal presence of informal migration from neighboring countries in Mauritania made it impossible and unreasonable to determine who, eventually, was planning to migrate towards Europe. Therefore, this migration regime was set to target all those who might look like a 'black' sub-Saharan migrant, as in the case of the Mauritanian Haratines. This was a consequence of creating also in Mauritania an image of the illegal migrant as a black individual from sub-Saharan Africa, worsening even more the already racialized relationships in the Mauritanian society (Cross 2013).
Stephan Dünnwald (2015) conducted in 2012 fieldwork in Mauritania, which acknowledged the increased violence and harassment against foreign citizens, even if they were legally residing in the country. An atmosphere of fear reigned among 'legal' migrants of black ethnicity as arbitrary arrests were rolled out on a regular basis. These people, who possessed the required legal documents, suffered temporary detentions in overcrowded prison facilities which even lacked food and water. Indeed, this procedure was defined by Andersson (2014, p. 114) as the "numbers game": when 'real', irregular migrants were too scarce, numbers were inflated through arbitrary arrests of legal migrants in order to show European donors the utility of their activities (see also Harding 2012). However, if the police deemed an individual as being illegally in Mauritania, she or he was, in most cases, deported to Rosso, a Senegalese border town. From there, migrants usually just re-entered Mauritania but were deprived of all their possessions by the Mauritanian police.
Further, through this approach, Mauritanian forces arrested migrants originating from Mali and Senegal. Hence, the country broke the bilateral contracts on free movement previously signed with its neighbors, by detaining their citizens in 'Guantanamito' and finally deporting them to the respective borders (APDHA/AME 2009; Cross 2013). Some NGOs denounced the participation of European actors in these violent abuses of human rights perpetrated against migrants, but allegations were rejected by Spanish authorities and attributed to Mauritanian authorities (APDHA/AME 2009). Meanwhile, the EU continued to finance this migration regime even after 2009, to expand the control of migration flows in Mauritania. In particular, in 2010, approximately 45 new border posts were built along the borders with Senegal and Mali, thus becoming the only legal entry points to the country (Frowd 2014). Eventually, whereas 'Guantanamito' was praised by the Spanish authorities, by 2010, the Mauritanian Gendarmerie dismantled all furniture and equipment for their own use, leaving it ruined and abandoned (Andersson 2014).

Mali
With the expansion of the border externalization and the outsourcing of border patrol to countries along the West African coast, migratory itineraries shifted to another sea, the internal sea of Africa: the Sahara (Andersson 2014;Gatti 2007). Following the movements taken by migrants, European efforts begun to focus on Mali with the aim of integrating the country into its externalized border regime. The nature and the outcome of this process, however, is of particular interest if it is compared with the racialized migration regime put in place in the neighboring Mauritania. For a long time, the democracy in Mali has been considered erroneously as an outstanding case of institutional best-practice in West Africa and the outbreak of the civil war in 2012 was often viewed as an unforeseen event (Sangaré and McSparren 2018). However, the rapid political disintegration caused by the war was the outcome of a protracted condition of insecurity and tensions rising in the country.
Ethnically divided among their historical roles as pastoralists and cultivators, droughts and the effects of climate change worsened resource-based conflicts by reducing available fertile lands (Pelckmans 2015). Often inexistent at the local level, the authority of the state served to increase the feeling of insecurity and the emergence of militias with the aim of defending their scarce livelihoods. Moreover, desertification made access to water for agriculture and livestock a pressing issue, forcing pastoralists to abandon their nomadic customs and settle down permanently where they found a suitable place with water sources. The population in Mali is segmented between the Mande ethnicity, which represents the majority of the country, with over 50%; the Peuhl or Fulbe, Voltaic, Tuareg, and the Soninké, who together constitute the complex ethnic mosaic in Mali (Sangaré and McSparren 2018). Traditionally separated along the distinction between nomadic pastoralists and sedentary farmers, ethnic-based conflicts have repeatedly occurred in Mali's history. However, ethnic tensions had no impact on the country's relationship with migration and a process of racialized illegalization, in contrast to Mauritania, did not occur. Following the path of Europe's border externalization efforts in Mali, it is remarkable to note the role played by its diaspora and their remittances, by the very active civil society organizations, as well as by its strong need for regional integration. All these elements combined made migration a sensible topic and the introduction of policies aimed at restricting movement neither desirable nor useful from the Malian point of view.
Mali represents a peculiar example for its socio-political reaction towards the European border externalization: less compliant, yet open to negotiate if their demands were met. The country occupies a strategic position as it is both a significant point of origin and of transit for people coming from Congo, Cameroon, and other African states. However, movements from-and passing through-Mali towards Europe, and particularly to France, started comparatively late in the mid-nineties and only witnessed an increase between 2002 and 2006 (Dünnwald 2011). Nevertheless, an early attempt to introduce in Mali, and other West African countries, the concept of 'illegal' migration was conducted by an Italian NGO called CISP (International Committee for the Development of Peoples). The CISP, financed by the European Commission, has cooperated with six West African countries and the IOM since 2004 by organizing awareness campaigns to warn over the risks of undocumented migration towards Europe. Till the events of Ceuta and Melilla in 2005, these campaigns had a rather small audience. In particular, Mali is a straightforward case of the inherent difficulties of introducing the notion of 'illegal' movements, as its society is historically bound to migration in the West African region (Idrissa 2019). It is estimated, in fact, that a third of the country's population habitually works and resides on a seasonal basis in the neighboring states such as in the case of the Ivory Coast.
Nonetheless, the events of Ceuta and Melilla in 2005 had meaningful consequences also in Mali. First, Morocco deported and returned 400 Malian citizens as an immediate reaction against sub-Saharan migrants. In the following months, Mauritania also started to return migrants to the Senegalese and Malian border. However, while Senegal quickly refused to accept these indiscriminate deportations, Mali continued to tolerate this practice and became critically labelled for this reason as the 'African dustbin' (Vincent 2010). Successively, in 2006, the EU and other international agencies, such as the UNHCR and IOM, begun to influence the Malian government by attempting to create awareness of the risks of undocumented migration (Dünnwald 2015). This task, however, needed to face the fact that Malian society is characterized by its considerable diaspora, with more than a third of its population residing in other West African countries (Idrissa 2019). Because of the prominent role of its diaspora, Mali has been one of the most convinced supporters of the ECOWAS community. While its diaspora mostly depends on seasonal work in the West African region, Mali also requires the qualified immigrants who habitually arrive from the neighboring Senegal and the Ivory Coast. Consequently, in order to realize the GAMM guidelines, the EU needed to deploy a significant effort to convince the Malian government of the usefulness of its border policies. Considerable persuasion work was needed to demonstrate the existence of benefits to negotiate agreements entailing surveillance, readmission clauses, and border controls. In fact, the large diaspora and the strong economic dependence on maintaining free movement with its neighbors represented a considerable disincentive for Mali to introduce restrictive migration policies (Trauner and Deimel 2013).
Till the 1970s, the former colonial power of France granted the right of free movement to Malian citizens. From that moment, however, France adopted progressively constrained immigration policies and introduced specific visa requirements. After the 2005 events of Ceuta and Melilla-and the following European pressure on countries of origin and of transit-France attempted to negotiate a bilateral readmission agreement with Mali. The French Government linked the agreement to the extension of development aid projects, whereas Mali, due to the value attributed to its diaspora, demanded an annual regularization program for 1500 Malian sans papiers living in France (Trauner and Deimel 2013).
The negotiation quickly became a major topic in the public debate, with many civil society organizations campaigning against the approval of the readmission agreement. When the agreement was finally due to be approved, the increasing popular criticism led the Malian President Amadou Toumani Touré to refuse to sign the agreement (Gatti 2007). As a retaliation, France interrupted the Co-développement projects and reduced development aid to Mali (Dünnwald 2011). In the meantime, the Spanish authorities sought to create new ties with Mali and likewise try to establish a bilateral readmission agreement. Spain's history of diplomatic relations with Mali commenced only in 2006, when, in conformity with its 'Plan Africa', it opened an embassy in Bamako (Andersson 2014). In this case, the two countries managed straightaway to sign the readmission agreement, which included a quota of 800 Malian temporary migrants per year allowed to work in Spain (Trauner and Deimel 2013). In contrast to France, Spain profited from the fact that its image was not that of the former colonial power in the Malian public opinion. Furthermore, no significant Malian diaspora lived in Spain. Nonetheless, the Spanish Government soon interrupted the program for temporary workers after accepting only 26 Malian citizens, but insisted on continuing repatriations in the first years after the agreement (Dünnwald 2015 Dünnwald (2011Dünnwald ( , 2015 illustrated the curious history of this center, which opened in 2008 and was financed by the European Commission with EUR 10 million from the fund for development aid (Commission 2008). Its main duty was to pave the way for a new migration policy and to organize guestworker programs towards Europe-a task which, however, was quickly rejected by European states (Idrissa 2019). Afterwards, the center was limited to elaborate awareness campaigns over the risks of illegal migration and offered counselling to returnees (Trauner and Deimel 2013). However, none of these programs achieved particular success. The CIGEM was closed without notice in 2014, and its website removed was from the internet. The EU also funded a project to establish 17 new border zones along the Malian boundaries, partly co-financed with Spain. However, as Dünnwald (2015) reports from his fieldwork in Mali, rumors argued that most of the donated materials for this project, such as 4 × 4 vehicles and computers, never reached their destination and, eventually, that previous border crossing habits never changed.
The comparison between Mali and Mauritania highlights the significant differences in the European engagement with third countries in its efforts that aimed to implement border externalization measures. First, the EU itself lacked a real intention to push for a strong cooperation with Mali, as the case with the CIGEM demonstrated, and potential agreements were mainly negotiated by individual Member States. A plausible explanation is that concurrent circumstances occurring on the Mediterranean and the Atlantic Sea were more pressing and shifted the political priorities of European institutions (Cross 2013). Moreover, Mali's geographic position played a crucial role as the country is particularly integrated with its neighbors. Migration, thus, represents an essential issue because of the presence of a significant diaspora and the importance of seasonal migration for the countries' economy-characteristics which were absent in the case of Mauritania. While negotiating migration cooperation agreements, Malian authorities needed to carefully consider the consequent social and economic costs of, for example, accepting readmission obligations for its diaspora. The impact of these policies on the transnational ties of its community abroad is also reflected, as we have seen, by the very active Malian civil society and its protest against previous negotiations with France. On the other hand, countries like Mauritania, who lack the existence of a consistent diaspora, seem to have not experienced the same concerns regarding the negative consequences of readmission agreements. In this sense, Mauritania was able to favor other aspects which were deemed to be more beneficial by cooperating on migration control, such as improving its international image after the coups d'état of 2005 and 2008. The presence of Arab ethnic majorities holding the political power has reportedly caused the common use of racial profiling during migration control activities which targeted and criminalized black sub-Saharan migrants. Historically, the Beidan political elite in Mauritania fostered its dominant position while the Haratine community and other ethnic minorities faced severe discrimination and recurrent displacements. In an effort to demonstrate the efficiency of Mauritania's commitment to its European partners as a West African 'gatekeeper', migrants and local communities were increasingly subjected to arbitrary arrests and deportations only because of the color of their skin. Instead, in a non-Arab country such as Mali, this was obviously neither possible nor desired. While ethnic hierarchies and tensions affect Malian society as well, the strong link with regional movements and its diaspora seem to have excluded migration from the ongoing confrontation. Eventually, the negligent process behind the European border externalization did not take into consideration the very different country-based contexts in West Africa, exposing, unfortunately, local societies to dangerous consequences, as revealed by the racialized illegalization that occurred in Mauritania. West African countries, nonetheless, are not just passive actors and do retain their own agency in this field. Despite the repressive reaction of France, which suspended Co-développement projects after the Malian refusal to sign a bilateral cooperation agreement, it is difficult to determine a clear response of European actors in case of rejection. A wider analysis, which includes more countries in the area, could investigate the indirect consequences of cooperation on migration control and provide some interesting insight into the policy implications both for migrants and local societies.

Conclusions
In the late 1990s, European countries recognized the inherent limits of a migration strategy based only on strengthening Europe's external borders. During the following years, the European Union developed a far more articulated approach and began to actively engage third countries in a broad process that aimed to externalize Europe's borders and outsource its border patrol functions. Generally viewed as a migrant-sending region, West Africa soon became a key partner in the new European project to intervene on migrant itineraries before these were able to reach Europe. Among the many instruments deployed by the EU in its cooperation with third countries, this article examined the major role played by policies which implemented efficient border controls in countries of migration transit and of origin. In particular, what emerged from the analysis of the cases of Mauritania and Mali is that the 'success'-from a European perspective-of introducing the concept of 'illegal' movements depended on the social and ethnic background of each country. Moreover, in a geographical area which is historically characterized by free cross-border movements, such as West Africa, the illegalization of migration is exposed to the risk of harming ethnic minorities through a racialized portrayal of sub-Saharan migrants.
The European Union, with Spain as its forerunner, established a growing number of policies and agreements to stop migrants before these could leave or pass through the West African region. The Global Approach to Migration and Mobility gradually designed a set of multilateral mechanisms which offered the required flexibility to constantly adapt to the shifting itineraries of migrants. Thus, the pathways taken by migrants were divided into manageable and governable routes through the categorization of countries in places of 'transit', 'origin', and 'destination'. New agreements, called Mobility Partnerships, were proposed to third countries and entailed a dual approach backing the main objective of externalizing migration control. On one side, the EU required third countries to accept a series of security-oriented measures, such as readmission agreements, the deployment of joint patrols, and to increase border control capacities. On the other side, the use of development aid and the creation of new opportunities for legal-but temporarymigration was considered to tackle what Europe judged to be the root causes of migration, notably poverty and unemployment, and to bargain the favor of third countries' authorities.
In Europe, the processes of illegalization were commonly connected to the extensive use of rhetorical strategies employed by media and politicians to depict and identify migrants as an external threat. Over the years, the enforcement of Europe's external borders in the Mediterranean and the Balkans performed the function of producing the socalled 'border spectacle', which made the confrontation between undocumented migrants and the border police a powerful representation of migrant's 'illegal' act of trespassing borders. The media visibility of such a narrative served as a crucial step in the construction of a racialized image of the 'illegal' sub-Saharan migrant. Nonetheless, the progressive implementation of border externalization agreements in West Africa went along with a new and less spectacularized illegalization of migration. Seen from the perspective of the European public, the illegalization and the mechanisms to combat undocumented migration have become increasingly hidden behind what this article called the border spectacle's curtain. Away from the spotlight, the curtain conceals the dangerous impact of new approaches to stop migration from West African countries arriving at Europe's borders.
Mauritania and Mali have been selected as case studies to highlight the consequences of the intricate implementation of European migration control strategies in West Africa. The analysis of the specific context of only two countries represents a limit of this research in regard to the wider implications for the entire West African area and the cooperation in the field of border control policies. Moreover, the focus and the reflection offered by this contribution has been circumscribed to aspects which pertain directly to the introduction or refusal of border control activities in a region where these were not present before. These experiences might not apply to other countries in the area, nor do they aim to cover the more complex nature of cooperation agreements and the role played by development funds in the field of border security and migration control.
The main insight resulting from this investigation suggests the impossibility and the dangers of applying a one-size-fits-all policy approach to the countries in the West African region. The complexity behind the dynamics of new border control policies cannot point to a uniform assessment of the potential consequences as each country requires a specific analysis. This issue should specifically inform researchers who investigate the impact of the European border externalization in West Africa or in other areas of the African continent. More importantly, policymakers should be aware of the socio-political contexts which are the object of their policies as several factors influence the outcome of introducing the concept of 'illegal' movement. As acknowledged in this study, a major role was played in this regard by different elements, such as the presence of a strong diaspora, a previous history of ethnic conflicts, and of how migration is perceived in the public opinion, by civil society organizations, or by political elites. Whereas in Mauritania, the illegalization of migration has been implemented upon pre-existent ethnic tensions in its society, the Malian context was less inclined to accept new limitations to international movement due to the peculiar characteristics of Malian society, which is strongly defined by its diaspora. Eventually, the externalized and outsourced European border strategy in the West African area only achieved cooperation with those countries who saw a political and economic benefit from the introduction of illegalization policies. A question which remains open concerns the broader implications for the social behavior and work patterns as a result of introducing restrictive border controls and limiting allowed movements. It is unclear to what extent this will change the landscape of regional West African trajectories and if it will affect the ECOWAS area of free movement. In the meantime, as border capacity building projects continue to be implemented, an interesting area of research regards the analysis of the consequent transformation of local livelihoods which are affected by these developments. While it is difficult to assess whether European governments succeeded in achieving a reduction in arrivals on their territories, this strategy has certainly served to further conceal the increasingly deadlier and hazardous journeys of migrants.
Funding: This research received not external funding.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest:
The author declares no conflict of interest.