Abstract
The Sahel region has become a focal point for externalisation policies seeking to prevent the onward travel of migrants bound for North Africa and Europe. In this context, humanitarian actors are increasingly called upon to implement migration control policies or to address migrants’ vulnerabilities during journeys characterised by ‘extended temporariness’. This paper builds upon the concepts of the humanitarian border and institutional trust to develop a typology of organisations and to explore how migrants in transit perceive these forms of humanitarian borderwork. By employing the concept of trust as an analytical prism to examine social navigation, risk-assessments and the felt externalisation of migration control, we argue that transit migrants’ mistrust in humanitarian organisations is a ripple effect of the externalisation process. This leads to cumulative ripples, such as protection gaps for individuals travelling from West to North Africa. The article compares two key locations for West African transit migration in Niger and Mali, exhibiting similar characteristics but differing in the extent and visibility of external migration-related programming. Through qualitative content analysis of interviews conducted with 90 transit migrants between 2020 and 2021, we identify the perceived involvement of organisations in migration control and the phase in the migration trajectory as key factors influencing (mis)trust in organisations. We further trace the effects of (mis)trust on migrant strategies, including exit, risk-acceptance, and self-protection. The results highlight the far-reaching impacts of different externalisation environments on migrant safety and strategies, as well as on the ability of humanitarian organisations to provide protection in vulnerable migration situations.
Introduction
The Sahel region has become a focal point for European Union (EU) policymakers seeking to prevent the onward travel of migrants bound for Europe. This approach, known as externalisation, involves implementing policies addressing border and migration management in countries of origin and transit, dismantling smuggling networks, or promoting return policies (Molenaar, Tubiana and Warin 2018). The consequence of these efforts has been the creation and expansion of borderspaces in Africa, where migrants face increased risks along routes to North Africa and Europe. Humanitarian and development actors are increasingly called upon to implement migration control policies or to address migrants’ vulnerabilities during their journeys, aiming to mitigate the negative ‘externalities of externalisation’ (Ferrer-Gallardo and Gabrielli 2022, 645).
A significant body of literature in critical border studies has examined the role of humanitarian organisations, including international and national non-governmental organisations (NGOs), in borderwork and the externalisation of migration control (Geiger and Pécoud 2014; Pallister-Wilkins 2017; Frowd 2018). However, there is limited in-depth research on how migrants perceive the role of humanitarian organisations, the extent to which this perception affects their trust, and consequently, their decisions and risk assessments (but see Lambert 2023). Given the growing, but sometimes unclear, role of organisations in external migration control and the possibility that transit migrants may avoid approaching organisations to avoid return or detention, this represents a significant research gap.
This article proposes using trust as an analytical prism for investigating the rippling effects of externalisation and humanitarian borderwork in the Sahel (see introduction to the special section ‘Unpacking the rippling effects of European migration governance in Africa’), particularly on migrants’ risk assessments and protection during increasingly precarious journeys. Following Hardin (2004) we understand trust as both a cognitive and relational category associated with expectations of others’ behaviour, which involves an element of risk and risk assessment. Yet trust can also be extended towards institutions, things, and complex systems (Sztompka 2006). Trust is thus an essential element of any social situation and an important prerequisite for social action. Anthony Giddens (1990, 92) describes trust as ‘a sense of the reliability of person and things’, arguing that modernity is characterised by a rise in importance of the exercise of trust towards more impersonal and abstract objects, such as organisations or the state. The latter is usually conceptualised as political (Uslaner 2018) or institutional trust (Egger et al. 2023), as opposed to social or interpersonal trust that relies on concrete interactions and relationships with individuals. We argue that an analysis of institutional trust is central to understanding migrants’ choices to access aid organisations during transit journeys, while also shedding light on the factors shaping migrants’ (mis)trust 1 in aid organisations. By examining migrants’ varying degrees of (mis)trust, we move beyond a state-centric approach and uncover the felt consequences of externalisation policies on individual migrants, shedding light on implications for the humanitarian system.
To trace the rippling effects of externalisation, we take the implementation process as a point of departure. The article focuses on Niger and Mali, two countries in the Sahel region that have been referred to as ‘laboratory’ (Venturi 2017) and ‘playground’ (Dünnwald 2016) for European migration control approaches. With the Sahel being a space of both externalisation and transit, these countries have witnessed the implementation of practices and policies aimed at curbing the movement of migrants and refugees towards Europe, such as border management and return programmes. Through qualitative content analysis of interviews conducted with 90 migrants in these locations, the article provides a grassroots perspective by examining international actors and their interventions as perceived by migrants in transit. Furthermore, it explores how a migrant's position on the continuum of trust and mistrust impacts how they navigate systems of humanitarian borderwork and protection during transit.
In addition to concrete empirical insights, this article advances theoretical understandings of the rippling effects of externalisation by introducing the concept of trust as a proxy for assessing ‘felt externalisation’ (Chemlali 2023). This contributes towards recent calls to rethink externalisation in migration and border studies, particularly to de-centre research perspectives and focus on multi-actor engagements beyond the state (Cobarrubias et al. 2023). We argue that felt externalisation engenders mistrust in humanitarian organisations, consequently giving rise to vulnerabilities and protection gaps for migrants in transit.
The article's first section presents the analytical framework, laying the groundwork for exploring the rippling effects of externalisation in the Sahel region. It examines the role of humanitarian organisations 2 in this context, concurrently refining the conceptualisation of trust as a pivotal gauge for understanding how organisations and their relationship to externalisation policies are perceived. The second section develops and traces the cumulative ripples of factors of (mis)trust and consequent migrant strategies. The conclusion emphasises the relevance of using trust as an analytical lens, providing new insights into the intricate processes of externalisation and its far-reaching implications.
Trust as an Analytical lens: Understanding the Dynamics of Externalisation and Transit Migration
While a growing body of literature examines beneficiaries’ perceptions of humanitarian organisations (Dijkzeul and Wakenge 2010), most studies on trust focus on migrants with a longer-term presence in a specific area (Gandolfo 2021), trust in institutions of host countries (Röder and Mühlau 2012), or refugees in camp structures with specific attention to the UNHCR (Hynes 2017; Ozkul and Jarrous 2021; Essex et al. 2022). The impact of external interventions on the trust relationship between migrants and humanitarian organisations, particularly in transit situations, has received limited research attention, although Schapendonk (2018) has argued that the winning of migrants’ trust is a common interest for all actors in the migration industry.
In the context of irregularised migration and externalised migration control, long and perilous ‘fragmented journeys’ (Collyer 2010, 273) and involuntary immobility have become common features of global migration systems. Transit migration, as we define it, transcends rigid time demarcations or clear-cut differentiation between countries of transit or destination. Instead, it denotes a period characterised by ‘immobility in a process of movement’ (Schapendonk 2012, 579), encapsulating what we term ‘extended temporariness’. Therefore, our focus is on the dynamic, fragmented, and often precarious nature of migrants’ journeys and the role of trust during them.
When it comes to (irregularised) transit migration, trust becomes ‘vulnerable, fleeting, precautious, and essential’ (Grønlykke Mollerup and Sandberg 2022, 150). Following Lyytinen (2017), we view trust as a central element of the trajectory, encompassing the giving and receiving of protection before, during and after the journey. Across a migration journey, trust exists along a continuum between trust and mistrust and evolves over time (Hynes 2017). On such journeys, degrees of trust can determine how and with whom migrants choose to interact. However, especially in transit situations, migrants may find themselves in dire circumstances where choices are constrained. Therefore, as a cognitive and affective concept, (mis)trust is particularly relevant in situations marked by risk, uncertainty, and bounded rationality (Bertsou 2019). Trust may therefore not be a linear, straightforward choice, but can take different forms and dimensions, such as trust out of necessity, transactional trust in economic exchanges or communal trust within established networks. 3
Previous research in the domain of migration and refugee studies has examined both social trust and trust in institutions (Hynes 2003). Given our specific focus on trust in organisations, our research aligns with the latter category. Trust in organisations may be shaped by interactions with individual aid workers, but there is an institutional dimension that cannot be reduced to such interactions. We conceptualise trust as a cognitive expectation and as the behavioural intention to take risks or rely on other actors, aligning with Li and Betts’ definition (2011, 103) of trust as ‘a choice, a decision to place one's confidence in others’. We therefore move beyond ‘the emotional basis of trust’ (Barbalet 2009, 368) by exploring how trust influences behaviour. This aligns with a more recent operationalisation of trust as a decision or actual intent initiated by the trust agent (Heyns and Rothmann 2021). As individual risk assessments are typically followed by conscious or subconscious actions, degrees of (mis)trust are therefore a form of ‘social navigation’ (Triandafyllidou 2019). This heuristic tool allows for capturing the different actors involved and the dynamic, sometimes ambiguous, nature of trust.
Another challenge in investigating the ramifications of externalisation policies lies in precisely identifying where these effects materialise and how they shape socioeconomic and political realities in specific locations (Stock, Üstübici and Schultz 2019). Exploring the dimensions of (mis)trust helps to uncover personal and individually experienced impacts of externalisation, which Chemlali (2023) has termed ‘felt externalisation’. This approach enables us to understand perceptions, beliefs, and thoughts associated with actors involved in migration control. These perceptions may or may not align with objective reality or be rooted in specific knowledge and interactions; nevertheless, they influence individuals’ actions and decision-making processes. By focusing on the concept of trust, we move beyond conventional policy analysis, redirecting our attention to the impact on the lives of people on the move. This scrutiny of trust provides nuanced insights into the nature and dynamics of externalisation. The following sections introduce our conceptualisation of externalisation and its rippling effects, along with a detailed exploration of the specific local contexts in Mali and Niger.
The Rippling Effects of Externalisation and Humanitarian Borderwork
Externalisation can be defined as ‘the process through which destination countries promote, support, delegate, impose, or directly carry out activities related to migration and border management outside their territories to prevent unwanted arrivals at their territorial borders’ (Cuttitta 2020, 2). This approach involves extensive outsourcing of migration control practices and resources beyond European territory, shifting responsibilities to countries of origin and transit (Bialasiewicz 2012; Lemberg-Pedersen 2019), as well as to international organisations (Lavenex 2016) or private actors (López-Sala and Godenau 2022). Consequently, externalisation encompasses a range of practices, from rescue operations and direct or remote border controls to measures addressing migration drivers. As a result, it blurs the boundaries between humanitarian aid, development assistance, border control, and security interventions.
Scholars have examined the role of humanitarian organisations in migration control through the lens of the ‘humanitarian border’ (Walters 2011) and ‘humanitarian borderwork’ (Pallister-Wilkins 2017). This growing body of literature analyses the coexistence and mutual reinforcement of care and control logics in migration management (Williams 2015; Little and Vaughan-Williams 2017), with specific attention given to the role of NGOs (Cuttitta 2020). Acknowledging the heightened vulnerability of individuals within border spaces (Weber and Pickering 2011), often exacerbated by the criminalisation of migration, humanitarian organisations play a substantial role in these areas. The notion of a humanitarian border then characterises the borderland as ‘a biopolitical space where mechanisms of border enforcement that may cause death are paradoxically intertwined with policies that aim at preserving life’ (Sahraoui 2020, 908–909).
Humanitarian borderwork transcends being merely a response to border violence and securitisation; it is an integral component actively contributing to the production of both the border and the apparatus of migration control (Pallister-Wilkins 2017). For example, when funded, either directly or indirectly, by European donors, medical aid facilitated by humanitarian organisations allows European authorities to provide care at a distance to individuals they confine outside their territories. As such, functioning within the humanitarian apparatus involved in ‘managing the undesirables’ (Agier 2011), healthcare provision through these organisations is intertwined with the externalisation of borders. We conceptualise their role in externalisation not so much in terms of contributing to a shift in the borderline but rather as actively contributing to the containment of onward movements.
Various approaches exist for analysing the role and agency of international organisations in migration governance. Perspectives range from characterising these organisations as ‘captured’ by EU interests (Spijkerboer 2021) to differentiating between their role as implementers, transmitters, and counterweights (Lavenex 2016). Alternatively, some views position them as autonomous actors within the international system (Barnett and Finnemore 1999). Indeed, organisations operating in the field of migration and forced displacement, such as the International Organisation for Migration (IOM) and the UN Refugee Agency (UNHCR), can assertively shape policy design and implementation (Geiger and Pécoud 2014; Pécoud 2018). We consider humanitarian organisations as actors operating within highly politicised contexts. However, we refrain from making definitive claims about the specific engagement of these organisations in migration control, as explored elsewhere (see Koch 2014; Scheel and Ratfisch 2014; van Dessel 2019). Our primary focus is to investigate how varying environments of externalised migration control affect migrants’ perceptions of these organisations, influencing their trust and decisions to seek assistance in times of need.
Expanding on the concept of ‘constitutive effects’ (Cold-Ravnkilde and Lindskov Jacobsen 2020), we systematically trace rippling effects stemming from the proliferation of externalisation policies. This approach shifts the focus to the ‘effects of effects’, unpacking the diverse consequences of externalisation, transcending the binary division between intended and unintended consequences. Externalisation practices in the Sahel do have various (unintended) consequences. While interventions might seem successful in deterring arrivals on EU territory, the prevailing trend indicates that migration movements are not reduced but redirected (Bøås 2021; Hahonou and Olsen 2021). Moreover, adverse effects extend to the manufacturing and adaptation of the human smuggling industry (Brachet 2018), increased economic and security destabilisation in the region (Donko, Doevenspeck and Beisel 2022), strained citizen-government relations (Abebe 2019), and the undermining of regional free movement (Castillejo 2019). Additionally, externalised and securitised migration control policies can amplify migrants’ vulnerability (Bergmann et al. 2017), exposing them to risks in transit countries (Bastide 2017) and the peril of abandonment or death in the desert (Ahmet Tchilouta et al. 2023).
Externalisation and Humanitarian Borderwork in the Sahel: A Comparative Study of Agadez and Gao
Our research focuses on two distinct regions: Gao in Mali and Agadez in Niger. While these localities share similarities as regional hubs for migrants in transit, they differ in volume, form, state capacity, and visibility of external migration-related programming. Notably, Niger has played a considerably larger role in external EU migration governance. Examining these two cases allows us to explore contextual differences in the occurrence and experiences of (mis)trust towards humanitarian organisations.
Both Niger and Mali are situated along the Central Mediterranean Route, a major passage for migration from West Africa to the Maghreb and onward to Europe. Beyond this migration corridor, trans-Saharan trade networks, along with return, seasonal, circular, and intra-regional migration, significantly contribute to shaping the mobility dynamics of the Sahelian region (Molenaar and El-Kamouni-Janssen 2017). Migration within the Economic Community of West African States (ECOWAS) is governed by a Free Movement Protocol, and some of these countries are both origins and destinations for migrants, including refugees. Despite this multifaceted reality, both countries have experienced a notable transformation described by Frowd (2020, 342) as the ‘production of transit states’. This term, from a Eurocentric perspective, presupposes that migrants passing through these countries are inherently labelled as irregular ‘would-be asylum-seekers’ (Oelgemöller 2011, 208). 4 Consequently, the externalisation process in the Sahel region has led to the construction of ‘immigrant illegality’ (de Genova 2002) thousands of kilometres before the physical borders of the EU. These changes in regional migration policies and patterns have been facilitated through various cooperation tools established as part of the 2016 European Agenda on Migration (Bisong 2020), building upon earlier policies and programmes (Brachet 2018).
Notable differences exist in the politics of migration and their economic implications in the two locations, which allow us to make claims regarding the effects of increased border management. Before the coup in 2023, Niger has been a more active recipient and implementer of migration control activities compared to Mali, resulting in a more prominent humanitarian borderspace and a ‘Europeanisation of migration policies’ (Boyer and Mounkaila 2018). Under the influence of both support and pressure from the EU, Niger enacted Law 2015–36 in 2015, targeting the ‘illegal trafficking of migrants’, thereby criminalising activities associated with smuggling or transporting migrants (Brachet 2018; Hahonou and Olsen 2021). This legislative shift resulted from a complex interplay of interests between Nigerien and European authorities, driven by their shared objectives in regulating trans-Saharan mobility. For Niger, this policy change presented an opportunity to improve its international standing and derive financial benefits by imposing restrictions on certain citizens’ movements. Meanwhile, the EU aimed to demonstrate proactive measures in managing migration in and from the African continent (Boyer, Ayouba Tinni and Mounkaila 2020; Ahmet Tchilouta et al. 2023). However, the ramifications of this criminalisation have had a profound impact on migrant journeys and local communities, disrupting historical mobility patterns and criminalising local and regional economies. Migrants face heightened physical risks as they are forced to take more dangerous routes and reside in concealed compounds or houses (referred to as ‘ghettos’ by locals and migrants) 5 without access to external support (Molenaar, Tubiana and Warin 2018; Ahmet Tchilouta et al. 2023).
In contrast, the Malian government has been more hesitant to implement projects and policy changes under EU ‘cooperation’ (Dünnwald 2016; Spijkerboer 2021). Unlike Niger, the transportation of migrants was not criminalised in Mali, and there was less external pressure to do so, as the country and the international community were primarily focused on stabilisation responses and the peace process (IMREF 2021b). Despite EU border interventions in Mali aimed at enhancing control and confinement, the country has demonstrated more resistance compared to Niger. This resistance is partly attributed to the larger number of Malians migrating to Europe and the perception of migration as a development asset. Additionally, Mali benefits from a strong network of civil society organisations supporting migrants (Idrissa 2019). 6 For example, returnees’ organisations engage in countering securitised return management practices (Sylla and Cold-Ravnkilde 2022).
In the context of a rise in the humanitarian government of migration in Mali and Niger, organisations are involved in external migration control in various capacities. Some organisations may solely provide food, shelter or medical assistance to migrants, while others, such as the IOM and UNHCR, are involved in a mix of humanitarian protection, border and migration management (Moretti 2020). Migration governance has become a conflation of protective and protectionist policies, while humanitarian and development programming has become more ‘migration-targeted’ (Frowd 2018; Collyer 2019). Therefore, migrants may find it challenging to distinguish between care and control, given that organisations often work on similar projects and are funded by the same donors (Cuttitta 2018). For this analysis and the specific context of externalisation in Niger and Mali, international organisations that provide services to migrants can be categorised into four types which distinguish between 1) their overall stated objectives; and 2) the nature of activities they engage in:
Organisations of category one are those whose stated objective is to manage (mixed) migration or refugee protection, but that include humanitarian and development services in their portfolio, such as IOM or UNHCR. The literature highlights the ambiguous role of IOM in the governance of mixed migration, whose core activities centre on assisted voluntary return and reintegration (AVRR) or capacity building for border management, which mainly supports governments rather than migrants (Kortendiek 2018). Although UNHCR's original mandate is protecting the rights of refugees, it is increasingly conceived of as an organisation that engages in governing global mobility (Koch 2014; Scheel and Ratfisch 2014). 7
Category two organisations are those whose stated objective is to provide humanitarian services, but may receive funding from donors that aim to manage migration. This includes faith-based relief organisations or independent medical humanitarian organisations (i.e., Médecins du Monde (MDM), Médecins sans Frontières (MSF) or the International Federation of Red Cross and Red Crescent Societies (IFRC)). MSF explicitly refuses to take funds from the EU or member states since 2016, ‘in opposition to their damaging deterrence policies on migration’ (MSF 2024).
Category three includes organisations whose stated objective is to provide development services, but receive funding from donors whose goal is to manage migration. The difference with category two might be gradual given the aim to bridge the gap between long-term development and short-term humanitarian support for populations of interest, including migrants and refugees (Interagency Standing Committee 2018). This pertains, for example, to national development agencies of countries in the Global North.
Finally, under category four fall smaller, local organisations or associations whose stated objective is to provide humanitarian services or information to migrants and who may or may not receive funding from donors whose goal is to manage migration. In contrast to the other types they include solidarity support or migrant self-organisation, as well as a stronger focus on advocacy and political work for migrant rights and against migration control (e.g., the Alarmphone Sahara (APS) in Agadez).
Our analysis highlights that migrants perceive differences between various organisations but may also group them together. When it is clear which category is meant, we distinguish as such.
Methodology
This research is based on semi-structured phone-based interviews conducted between December 2020 and February 2021 with 90 transit migrants. In addition, 16 interviews with humanitarian workers informed the background for this research but were not analysed in depth. Data collection was carried out as part of the IMREF project. 8 Among the 90 migrants interviewed, 45 were located in Agadez and 45 in Gao. The sample included 30 women, 14 in Agadez and 16 in Gao. Of the respondents, 49 had stayed in the locality for less than 6 months, while 41 had stayed for more than 6 months. The length of stay ranged from 1 week to 36 months, highlighting what ‘extended temporariness’ can mean in practice. Despite the notable variation in the duration of stay, all respondents can be regarded as being in a transit situation, as transits are known to extend for a considerable time, with waiting periods ranging from weeks to, at times, years (Castagnone 2011). Furthermore, all interviewees expressed the wish to continue their journey, although there were eleven in Agadez that had been refouled from Algeria versus only two in Gao. The age of respondents ranged from 18 to 41 years. In Agadez, the majority of respondents were from Cameroon (20), followed by Benin (4), Senegal (4), and Ivory Coast (4). In Gao, most respondents were from Guinea (11), Ivory Coast (8) and Burkina Faso (7).
To ensure inclusion of a diverse spectrum of migrants residing in the research areas, the study employed a snowball sampling strategy with multiple entry points. 9 The qualitative sample does not aim to be representative of the nationality, age, and gender breakdown of the migrant population in both locations, but includes diversity by including individuals who have received assistance from organisations and those who have not. The primary objective is exploratory; to understand individual perceptions of the role of humanitarian organisations and their influence on behavioural decisions throughout the migration trajectory in different environments of externalisation.
In the interview guide, trust was operationalised in a multidimensional manner, with questions exploring migrants’ trust in various actors during their journeys, 10 identifying whom they would seek help from in certain situations and their expectations (i.e., regarding risks or benefits). Interviews were conducted by researchers based in Bamako and Gao (Mali), as well as Agadez and Niamey (Niger). They were conducted over the phone due to COVID-19 safeguarding concerns and were either in French or a local language. The enumerators were trained to clearly explain the study's objectives and impartial nature, assuring interviewees’ anonymity to encourage the expression of their opinions about organisations. While the impersonal nature of phone-based interviews does have drawbacks (e.g., lack of trust relations with interviewers, prevention of collection of supplementary contextual data), stranded migrants are typically open to speaking to researchers, because it is a channel through which they can make their voices heard. 11
The interview transcripts were analysed using structured and focused comparison (George and Bennett 2005), allowing for a systematic accumulation of narratives across the two cases. Specific aspects of the cases were examined in-depth, such as the impact of migration control and externalisation on migrant (mis)trust in humanitarian organisations. The qualitative content analysis enabled the identification of migrants’ meaning-making practices related to (mis)trust and risk assessments, providing a novel foundation for assessing migration journeys in the Sahel region.
Factors Impacting (Mis)Trust in Humanitarian Organisations in the Sahel
Migration, particularly irregularised migration, occurs within highly politicised and precarious contexts, where mistrust is often pervasive from the outset. Throughout the interviews, participants consistently expressed a general lack of trust towards other people, except for family and friends, and, occasionally, their fellow travellers. It is imperative to contextualise our findings within this overarching atmosphere of low trust and keeping in mind that the mistrust expressed in the interviews is a justificatory narrative that may misrepresent how organisations and migrants interact. Moreover, perceptions of organisations may be influenced by pre-existing notions, particularly if these organisations are active in migrants’ countries of origin. Additionally, these perceptions may evolve through interactions with organisations, fellow migrants, and information provided by actors within the migration industry. Consequently, our results can only provide a snapshot of the situation during the research period, which was further complicated by the impact of the COVID-19 pandemic.
Migration Control and (Mis)Trust
‘I think that these organisations are the causes of all this refoulement and deportations; that's why I'm so suspicious of them and I don't want anything to do with them’. (Aga_16) 12
This response by a Cameroonian female migrant in Agadez underlines the prevailing belief in both research locations that accessing aid services may expose them to risks, primarily the threat of deportation to their countries of origin. Many respondents expressed their mistrust through feelings of suspicion, influenced by general observations and accounts from other migrants rather than personal experiences. A male Ivorian respondent in Gao explained: ‘What makes me suspicious of organisations is that every year we see many migrants who are turned away by the authorities. And finally, one wonders if humanitarian organisations are not at the root of these expulsions; maybe humanitarian organisations share our information with the local authorities’. (Gao_5)
The general migration control environment involving deportations, returns and refoulements leads to migrants questioning whether their access to organisations will increase their risk of being exposed to these practices. The perceived collaboration between humanitarian organisations and national authorities emerged as the most common factor fuelling mistrust due to fears of deportation.
13
Migrants also expressed strong mistrust of police or local authorities, who are frequently implicated in extortion and harm against migrants in Mali and Niger, along with other armed or criminal groups (Molenaar, Tubiana and Warin 2018; IMREF 2020). This history of abuse contributes to the negative perception of collaboration between organisations and the state. Some migrants in both Gao and Agadez believed organisations had provided information about them to the police, resulting in detention or deportation. A 25-year-old man from Benin stated that, ‘So many migrants have been expelled from Algeria in 2019. Those of us who have not gone to Algeria yet think that it may be because they have been denounced by humanitarian organisations’. (Gao_9)
This underscores the direct link migrants establish between refoulements and the perceived contribution to them by humanitarian organisations. However, this link is often based on general perceptions of the migration control environment along the route rather than concrete, personal experiences. Especially in uncertain contexts, such rumours contribute to shaping migrants’ and refugees’ understanding of how international organisations function (Ozkul and Jarrous 2021) and information mismanagement can lead people to develop deep mistrust in government and aid organisations, and instead turn to informal brokers like smugglers (Carlson, Jakli and Linos 2018).
As most transit migrants wish to remain undetected during their journeys, they perceive the visibility of organisations as a risk. It was common among migrants in Agadez to observe that when organisations visit their neighbourhoods, local authorities would follow shortly after and thus know where migrants are located, leading to suspicions of collaboration between organisations and authorities. A 21-year-old Ivorian respondent explained, ‘Let's take the general case of IOM. When IOM comes to a ghetto, you will see the police come just 30 minutes after their departure. So, we cannot trust them’. (Aga_7)
Another factor influencing (mis)trust in organisations is their involvement in awareness-raising campaigns and interventions migrants perceive as discouraging their migration plans: ‘They can discourage us, put ideas in our heads that are against our ideals […] So we risk compromising our dreams and in the end we'll be wasting our time […]’ (Aga_17)
‘Organisations should respect their [humanitarian] commitment and do their work properly, and they should avoid wasting their time with awareness-raising on the risks of migration because we are fully aware of the risks, and we are responsible for our own choices’. (Aga_1)
Migration-information campaigns were framed as a waste of time for migrants and organisations alike by many of the respondents. This aligns with previous studies indicating that awareness campaigns are unlikely to change migrants’ migration plans (Savio Vammen et al. 2021) and emphasises the active agency of individuals in their migration projects. A female Nigerian respondent in Gao directly connected the discouragement of migration to mistrust, ‘As soon as migrants suspect that organisations follow government directives and help discourage migration, mistrust will become great. People will try anything to get to their destinations, which will promote clandestine and criminal practices’. (Gao_F15)
This supports the notion that stricter control measures do not limit migration but redirect it, making journeys more dangerous (Molenaar, Tubiana and Warin 2018). Concerns that organisations may discourage migration were the most common reason for being suspicious towards them and were sometimes linked to the belief that organisations have hidden agendas aligned with European migration policies. A male Cameroonian respondent in Agadez explicitly made this connection, stating, ‘Migrants are wary of humanitarian organisations because they are funded by the European Union. All of us know that the EU is against irregular migration. Today, it is the EU that is at the origin of all the repatriation; it is the root of all our ills’. (Aga_44)
Between Care and Control: (Mis)Trust by Mandate and Migrant Trajectory
As argued previously, the distinction between care and control activities within humanitarian organisations is often blurred (Cuttitta 2018). Migrants must navigate between organisations as aid providers and as part of broader global migration control initiatives. The previous section highlighted how migrants who mistrust organisations perceive them as feeding into and, in some cases, directly participating in a larger system of containment. However, it is important to note that respondents did not universally express a general mistrust towards all organisations. Many differentiated between organisations based on their specific mandates and roles on the ground, particularly in Agadez. We term this phenomenon ‘(mis)trust by mandate’.
One significant aspect of mistrust by mandate is the heightened mistrust towards organisations associated with return programming. Migrants perceive these actors as directly involved in deportations in both Agadez and Gao. Specific concerns were raised about the return assistance programme of the IOM, with some migrants believing that seeking aid would result in their return, as this was perceived to be the true objective of the IOM: ‘I don’t trust them […] IOM's goal is to return migrants to their countries’. (Aga_33)
In line with the results from the previous section, the perceived collaboration between the IOM and authorities was particularly prominent in shaping this mistrust. For instance, a Senegalese male migrant in Agadez expressed, ‘If you talk with the IOM, it's like confiding directly to the police because they tell them all, they even contribute to the arrests. […] I think that migrants prefer the services of MDM and Red Cross because they take into account the most urgent needs, namely health, food, and cash more than those of IOM which only intervenes to repatriate us. They do not care about our projects. If they approach you, it is to convince you to return to your country of origin’. (Aga_29).
Simultaneously, humanitarian rescue operations (HRO) at the Algerian border, mainly conducted by IOM in collaboration with the Niger General Directorate for Civil Protection (DGPC), illustrate the ambiguity inherent in the dual nature of care and control, as migrants express both suspicion and appreciation for such activities. A 23-year-old Malian woman provided insights into these diverse opinions: ‘[i]t is quite normal for migrants to develop different opinions about the services that organisations offer. If we take search and rescue, it is the first assistance that migrants receive from organisations. At this time, they think of nothing but the ordeal they experienced before arriving in Assamaka.
14
They are, for them, like a kind of ‘messiah’ who saves their lives’. (Aga_25) ‘IOM is encouraged and applauded by migrants wishing to return to their country, but for migrants who are turned back against their will, they do not have a good appreciation of this service because even I think that IOM is at the origin of our refoulement otherwise how it can know that we are abandoned in the middle of the desert to come and help us’. (Aga_39)
Due to IOM's involvement in return, some migrants also view the life-saving activities of this organisation as a testament to their involvement in migration control, demonstrating the deep-seated mistrust towards IOM. This is further underlined by the fact that organisations such as MSF and APS also conduct such HROs at ‘Point Zero’ near Assamaka, but were not met with the same suspicion.
Trust and mistrust can therefore come and go and even coexist, not least also due to the mandate ambiguity of some organisations. These nuances emphasise the significance of considering the temporal context and accounting for migration trajectories when examining trust dynamics. For example, those people who do wish to return may appreciate the return programmes offered by IOM. Migrant trajectories involve continuous re-routing, adjustments, and navigations, influencing their perceptions and trust dispositions (Schapendonk et al. 2020). Therefore, the stage of the journey a migrant finds themselves in shapes their needs, expectations of aid, and willingness to accept it. Outbound migrants may be more suspicious of organisations as they strive to continue their journey, but they could also have less mistrust as they have not yet encountered migration and border control structures. Moreover, they are likely to still have more access to resources and fewer immediate needs. Returned or refouled migrants on the other hand may have different perceptions: ‘The feelings are different for a migrant who comes from a neighbouring country and is about to continue his journey to Libya and Algeria, and a migrant who has been turned away from those countries. The migrant who just arrived does not trust organisations because he has not witnessed what those organisations do’. (Aga_22)
Positive encounters can lead to a reassessment of suspicion and mistrust which was based on second-hand accounts, as exemplified by a Guinean male migrant who shared his evolving opinion during his journey in Gao: ‘Yes, my opinion about humanitarian organisations has changed since I arrived in Gao. Some of our migrant friends have told us from Guinea that there are organisations in complicity with certain people that make sure that migrants cannot migrate. Without explaining why, or how. But once at the station, I understood that it's far from what I'd heard before. I have seen that organisations seek to do only good and not harm migrants’. (Gao_19)
Overall, the interviews highlight that migrants tend to trust organisations less visibly involved in migration control and focus more on strengthening migrants’ capacities and resilience during their journeys, such as through solidarity and community building or training for skills that may be valuable when living in foreign countries. Trust is thus influenced by the alignment of organisational objectives with migrants’ objectives. When objectives align, trust is more prominent, but when objectives conflict, mistrust dominates. Most migrants aspire to continue their journey, and organisations seen as unwilling or unable to assist in this goal are met with mistrust. For onward migrants, perceptions about the relevance of support were consistently linked to how the support related to their travel plans. This perspective is captured by a female Beninese respondent who stated: ‘No matter the nature, quality, or quantity of aid, if it does not contribute to the continuation of our project, it is minimal. We certainly appreciate what they do for us, but it is not enough’. (Aga_4)
Coming back to the typology of organisations developed earlier, our results underline that migrants do distinguish mostly between organisations in the first category (organisations whose stated objective is to manage migration, but that include humanitarian and development services in their portfolio) and category two (organisations whose stated objective is to provide humanitarian services but receive funding from donors that aim to manage migration). While the work of organisations of category four (local associations) was appreciated due to their grass-roots character and social justice agenda, it was acknowledged that they cannot intervene in many aspects (be it in a positive or negative way) due to their limited size and resources.
Consequences of (Mis)Trust on Migrant Strategies and Protection
Previously, we explored the factors contributing to migrants’ suspicions of organisations and their impact on their perception of organisations. In this section, we delve into the influence of varying levels of mistrust on migrant choices, as it can serve as both a coping mechanism and a means of protection (Hardin 2004). Mistrust and suspicion are often survival mechanisms in the context of migration, allowing individuals to avoid persecution or exploitation (Hynes 2017) and respond to risks that threaten their migration plans. Coupled with rational suspicion and past experiences, trust-informed risk assessments shape migrants’ actions and behaviours. Therefore, mistrust as a result of navigating the migration control environment can trigger various responses, ranging from crafting exit strategies to making compromises and engaging in self-protection.
Exit Strategies: Evasion or Refusal of Services
Exit strategies involve migrants choosing to withdraw from protection relationships or refusing offered protection (Vigneswaran 2020). In the preceding section, we identified two instances where such exit strategies become relevant. The first scenario arises when migrants believe that organisations may contribute to their deportation, prompting them to actively prevent humanitarian staff from entering their communities or to refuse engagement. The second occurs when migrants suspect that organisations might manipulate them into returning to their country of origin, thereby nullifying the investments they have made in their journeys. In these cases, migrants deliberately avoid interactions with humanitarian actors as a harm-reduction strategy. Additionally, the perception that organisations discourage migration prompts migrants to actively avoid their services, safeguarding the investments they have already made in their journey. As one Cameroonian respondent in Agadez expressed, ‘You know, I have friends who have travelled several times. It was through them that I received a lot of information about IOM, who are there to send migrants home […]. Friends made me understand that if I see an IOM officer in front of me not to listen to him because he will discourage me to the point of no longer wanting to continue my adventure’. (Aga_6)
Difficult Decisions: Between Compromise and Self-Protection
In the context of irregularised migration, trust practices seldom align neatly with distinct roles of being either trusted or trustful (Grønlykke Mollerup and Sandberg 2022, 148). Most migrants described their willingness to seek assistance from organisations as a trade-off between their level of need and perceived risks to their journeys. As a result, many migrants stated that they would turn to organisations if they had no alternative. However, even in situations of dependency, trust can be seen as a risk assessment and harm-reduction strategy, as exemplified by a Cameroonian woman in Agadez: ‘Yes, if I am in despair, I will be forced to return to them [organisations] so as not to end up in prostitution’. (Aga_17)
Migrants are aware of the connection between violence and vulnerability resulting from border control, and the potential of some aid organisations to mitigate these effects. A male Cameroonian respondent in Agadez reflected on the risks migrants face during their journeys and the possible reliance on humanitarian aid: ‘If organisations like them do not exist, the lives of migrants in transit countries will take a different turn. You really have to live this adventure to see the ordeal we are going through for the sole purpose of realising our dreams and ambitions. These organisations are in a way insurance for us in the field of protection and accompaniment, which shows their assistance, and their help is essential because our survival depends on it’. (Aga_44).
At times, the decision not to opt for an ‘official’ protection relationship with an organisation meant that migrants had alternative forms of support: ‘[E]ven in case of need of help, I do not need to turn to an international or local organisation, we help each other, and my boss supports me in my migration project’. (Gao_16)
Tracing and Comparing the Cumulative Ripples of Externalisation in the Sahel
We have underscored the far-reaching impacts of European migration control on shaping migrants’ perceived risks when accessing organisations and support services. The refusal and evasion of humanitarian aid represent consequential ripples with potentially harmful effects on migrant safety. The lack of trust leads to a strategic avoidance of aid to mitigate perceived risks, such as fear of detention, deportation or discouragement. Still, it may simultaneously result in protection gaps, as migrants often wait until they reach extremely vulnerable situations before seeking support. Gendered implications arise, as some women mentioned accepting aid to avoid resorting to prostitution. Overall, we find particularities of protection in situations of extended temporariness, as significant protection gaps exist for irregularised transit migrants not falling within the mandates of any humanitarian actors. Furthermore, the prevalence of mistrust due to migration control highlights two critical issues: the inability of organisations to provide transit migrants with information and resources for safer journeys, and the fact that destitute migrants and returnees are often left with no choice but to accept assistance from organisations like IOM, which may be conditional on accepting voluntary repatriation.
Clear differences emerged between Agadez and Gao across all indicators, underscoring that contextual disparities between the two locations influence attitudes towards humanitarian organisations. This variation is crucial for comprehending and contextualising the consequences of increased external actor involvement and visibility in migration governance in the Sahel. The political-economic implications of anti-smuggling policies, new interdiction practices, and the scale of migrant-targeted programming have expanded the humanitarian borderspace in Agadez to a much larger extent than in Gao. As discussed earlier, the visibility of some organisations already represents a risk and reason for migrants not to access them.
Mistrust in organisations in Agadez is therefore much more closely linked to the criminalisation of migration, particularly due to Nigerien Law 2015-36. Awareness of the EU's involvement in migration control is also more pronounced in Agadez: ‘Here we are in Niger, right? The European Union finances the Nigerien government. […] They will try to repatriate you because Niger takes a lot of money from the European Union’. (Aga_9)
Conclusion: Fragmented Journeys – Fragmented Trust
This article explored the experiences and perceptions of transit migrants as they navigate access to humanitarian organisations and grapple with mistrust on precarious migration journeys in the Sahel. Our emphasis on the pivotal role of perceptions in shaping protection outcomes contributes to the conceptualisation of how issues of (mis)trust can be indicative of felt externalisation. The analysis of 90 interviews conducted in Gao (Mali) and Agadez (Niger) reveals that the extent and visibility of externalisation activities significantly influence migrants’ mistrust and their willingness to engage with humanitarian organisations, in turn resulting in protection gaps or practices of exit and self-protection. This finding is reinforced through a comparative approach between the two locations. In contexts where externalisation is less extensive or visible, such as in Gao, migrants expressed fewer needs, possessed less knowledge about organisations, and exhibited less mistrust. Overall, key factors contributing to mistrust included perceptions of organisations aiming to compel migrants to return, collaborating with police for denunciation and deportations, or engaging in campaigns discouraging migration. While these effects were more prevalent in Niger than in Mali, it is important to note that the humanitarian border is not fixed but fluctuates with the shifting routes of migrants, in line with the concept of ‘itinerant borders’ (Ahmet Tchilouta 2023). The recent coups in the region are also likely to affect how Mali and Niger continue to implement or contest EU externalisation policies. 15
Our analysis sheds light on the interplay between the dialectic of care and control within the humanitarian border and how it is encountered and contested. We have shown how (mis)trust is a part of migrants’ social navigation that then gives rise to different types of agency, as it plays a role in migrants’ risk assessments and strategies regarding engagement with organisations, highlighted by acts of exit, compromise, and self-protection. However, as migrants increasingly avoid accessing humanitarian aid to remain unnoticed amid heightened border controls, this triggers cascading effects of externalisation, such as increased human insecurity. Viewing externalisation through the lens of trust and migrants’ risk-avoidance strategies provides insights into how such diffuse policies manifest and impact individuals at the local level, unravelling felt consequences on the lived realities of people on the move.
Our conceptualisation of (mis)trust as choice and risk-assessment calls for further exploration of the interplay between agency and uncertainty during migratory journeys. The question is not only whether mistrust can be seen as a form of agency, but also how strategies of refusal and resistance then relate to various de- or re-bordering processes. While our focus on migrant experiences and perceptions was essential to highlight how externalisation is felt, a dynamic approach incorporating and shifting between different perspectives of migration facilitation and control as suggested by Schapendonk (2018) could further disentangle externalisation processes and their rippling effects.
In addition to exacerbating vulnerable situations for migrants, mistrust due to the externalisation of migration control also presents challenges for humanitarian actors, as the level of trust in an organisation is a measure of its legitimacy. In the Sahel, mistrust is selective, and many organisations are still perceived as providing essential aid, while organisations like the IOM are not regarded as legitimate aid providers by many migrants in transit. They are aware of the European agenda against irregular migration, and any awareness-raising or voluntary return activities are seen as implicit support for this agenda. Organisations must acknowledge the negative effects of migrant-targeted programming, reflect on their ambiguous role in bordering, scrutinise the donors they engage with, and consider the messages they convey to migrants and related communities. If organisations do not advocate for a right to stay or safe pathways, their ability and willingness to protect those at the sharp end of migration control is limited.
Footnotes
Acknowledgements
The paper draws on data originally collected and coded by Seefar as part of the IMREF project. The authors would like to thank the research teams led by Ibrahim Nabiou in Mali and Abdoulkarim Maazou in Niger. We are also grateful to the project leader, Xavier Rodriguez (Integrity) for supporting this paper from its very inception and to FCDO for permission to re-analyse the original empirical data collected for this paper. The paper benefited greatly from comments received at various occasions drafts have been presented, such as at the workshop ‘De-centring and Contesting Externalisation in West Africa and Beyond’ at SOAS (London). We also would like to thank the two guest editors of the special section ‘Unpacking the rippling effects of European migration governance in Africa’, Cathrine Talleraas and Ida Maria Savio Vammen, and the anonymous reviewers for their helpful comments in the revision process.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship and/or publication of this article: The original empirical research underpinning this paper was conducted in the framework of the Independent Monitoring and Rapid Research and Evidence Facility (IMREF), funded by the Foreign, Commonwealth and Development Office (FCDO), under FCDO’s ‘Safety Support and Solutions Phase II Programme’. The secondary analysis of data for this article was co-financed by the Asylum, Migration and Integration Fund and the Austrian Federal Ministry of the Interior (Grant no 10 ST AMIF 2319F – 230/20 BS).
