Scholarly article on topic 'Governing Mobility through Humanitarianism in Somalia: Compromising Protection for the Sake of Return'

Governing Mobility through Humanitarianism in Somalia: Compromising Protection for the Sake of Return Academic research paper on "Social and economic geography"

0
0
Share paper
Academic journal
Development and Change
OECD Field of science
Keywords
{""}

Academic research paper on topic "Governing Mobility through Humanitarianism in Somalia: Compromising Protection for the Sake of Return"

Development

and Change Published on behalf of the Institute of Social Studies, The Hague

Governing Mobility through Humanitarianism in Somalia: Compromising Protection for the Sake of Return

Cindy Horst and Anab Ibrahim Nur

ABSTRACT

This article aims to contribute to an increased understanding of the importance of migration in humanitarian and 'post-humanitarian' contexts, by exploring the interlinkages between protection and displacement. It argues that the strategies by which conflict-displaced populations protect themselves are largely based on mobility. Yet, humanitarian approaches to displaced populations do not take sufficient account of the mobility needs of those they assist. Furthermore, the actual location at which aid is provided is affected by funding realities and donor priorities. This article discusses the case of protracted displacement realities of Somali refugees and internally displaced people in Kenya, Somaliland and south-central Somalia. Based on in-depth semi-structured interviews and focus group discussions with humanitarian aid workers and displaced people, the article offers an analysis of the recent 'stabilization discourse' that fuels programming directed at the return of displaced Somalis. The authors argue that humanitarian protection is compromised by immobile aid practices and by humanitarian programmes that are guided by states' interest in refugee return.

INTRODUCTION

In this article, we argue that humanitarian policies and practices do not take sufficient account of the mobility of those they assist. At the same time, funding realities and donor priorities often impact the actual location at which aid is provided. Studies of forced migration have focused on the 'politics of mobility' (Hyndman, 1999, 2000, 2012) to show that mobility is imbued with power relations that enable some and restrict others from moving. In the (analysis of) practices of humanitarian actors, limited

Earlier versions of this paper have been presented by Cindy Horst at the international workshop Mobilities: Immobilities. Revisiting the Forced Displacement Conundrum (CMI Bergen, 5-7 September 2013) and the Protection of Civilians (PoC) seminar (4-5 November 2013). We would like to thank participants of these workshops as well as the anonymous reviewers of the journal for their constructive feedback.

Development and Change 47(3): 542-562. DOI: 10.1111/dech.12233

© 2016 The Authors. Development and Change published by John Wiley & Sons Ltd on behalf of International Institute of Social Studies.

This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

attention is paid to the issue of mobility. Yet most of those receiving assistance in humanitarian contexts are or have at some point been displaced, often with considerable implications for their level of vulnerability and protection needs. Similarly, the role of humanitarian aid practices in the politics of mobility also deserves further analysis. For people caught up in 'man-made' crises, changing conflict trends and humanitarian realities, in conjunction with increasingly restrictive migration regimes worldwide, have impacted on their ability to make the decision to move or to stay, with further implications for humanitarian realities. Involuntary immobility leads to people in search of protection being trapped in conflict zones (Lubkemann, 2008). In 2014, there were approximately 19.5 million refugees and 38.2 million Internally Displaced People (IDPs) worldwide (UNHCR, 2015a: 8). At the same time, 1.5 billion people continue to live in countries affected by fragility, violence or conflict (see World Bank, 2011).

This article aims to contribute to an increased understanding of the importance of migration in humanitarian and 'post-humanitarian' contexts, by exploring the interlinkages between protection and displacement. It does so by focusing on the protection realities of displaced Somalis in Kenya and the Somali regions in the period following the establishment of the Federal Government of Somalia in late 2012. This was the first permanent central government in the country since the start of the civil war and its installation was an integral part in the official 'Roadmap for the End of Transition'. The hope and belief that south-central Somalia is transiting from conflict to post-conflict is what fuels the 'stabilization discourse' that has become increasingly dominant in recent years. This discourse includes a focus on the desirability of assisting displaced populations to return. We argue that humanitarian activities that draw on the stabilization and return discourse do not sufficiently problematize underlying perceptions of home, displacement and return.

The neat evolution from conflict to post-conflict, from humanitarian aid to development assistance, and from displacement to return which is implied in this discourse is far from reality in the Somali context. In most contemporary conflict and post-conflict situations, violence can erupt anywhere, at any time, leading to a state of radical uncertainty (Horst and Grabska, 2015). Types of violence and conflict that are neither war nor peace, neither criminal violence nor political violence, are increasingly the norm in conflict regions (Richards, 2005; Suhrke and Berdal, 2012), and Somalia is no exception. Present-day conditions in Mogadishu and other parts of south-central Somalia are very different from those in the early 1990s at the onset of the war, yet the increased securitization of the Somali conflict in light of regional and global security concerns contributes to continued violence and uncertainty locally.

Similarly, the idea that 'displacement' simply ends when people 'return' has long been challenged within refugee studies (Black and Koser, 1999; Grabska, 2014; Hammond, 2004). In the case of Somalia — where the Somali people live in five nation states and have, throughout history, moved

back and forth across the borders of these states for a range of reasons, with family members living in different nation states — it becomes clear just how problematic such binary understandings of mobility are. Where would a young Somali woman 'return' to if she was born in Kenya, of parents who fled the Somali conflict in 1991, with grandparents who moved from the North Eastern Province of Kenya to south-central Somalia in the 1980s, while a number of their older children (the woman's aunts and uncles) remained in Kenya? Where, then, is she displaced and where is she home? Categories of 'refugees' and 'locals' are highly fluid here (Ikanda, 2014) and complicate understandings of displacement and return considerably.

This article draws on a study that focused on the interrelations between protection and displacement.1 Fieldwork was conducted in Nairobi, Mogadishu and Hargeisa between December 2012 and March 2013 and focused on how humanitarian aid workers and those displaced understand protection in Somalia.2 A total of 26 humanitarian aid workers were interviewed in the three fieldwork locations, while 30 internally displaced people were interviewed in Hargeisa and Mogadishu. Seven focus groups were conducted with displaced in both locations. The research team also attended a number of protection meetings, and collected and analysed reports and documents from the United Nations and NGOs.

Data were analysed along a range of topics that were drawn from the research questions or presented themselves inductively. Topics included informants' understandings of protection; understandings of who is responsible for providing protection; the main protection challenges faced by the displaced (including evictions and sexual and gender-based violence); community-based forms of protection; humanitarian challenges (including 'remote control' and the development-humanitarianism gap); and the stabilization discourse. In this article, we present a number of well-formulated quotations that we identified through this analysis, in order to illustrate more general trends in our data. Given the small size of the humanitarian community working on protection in Nairobi, we have chosen to provide limited identifying details of our informants in order to avoid compromising their anonymity.

The next section of this article provides a brief overview of regional protection realities for Somali refugees and IDPs. We then argue that protection lies at the heart of the humanitarian agenda but international protection mechanisms in reality often can only focus on mitigating the consequences of abuses rather than preventing them from happening in the first place. Individuals and communities are ultimately always central to their own protection. We claim that mobility is crucial to the self-protection strategies

1. Research conducted for the 'Protection of Civilians. From principles to practice' project, funded by the Research Council of Norway, from 2012 to 2016.

2. Data in Nairobi were collected by Cindy Horst, while Anab Nur was responsible for data collection in Mogadishu and Hargeisa.

of displaced Somalis, and show how international protection efforts fail to recognize and relate to this mobility. Discussing current humanitarian approaches in the region, we show that much programming happens through 'remote control' due to security concerns, while at the same time, the international discourse on south-central Somalia centres around stabilization and return. In a situation where the lives of many Somalis are highly mobile due to continued uncertainty and a long history of mobile livelihoods, humanitarian aid cannot offer protection if it remains immobile or disproportionately focused on one type of movement: return.

REGIONAL PROTECTION REALITIES FOR SOMALI REFUGEES AND IDPs

With the collapse of the Somali state in 1991, over a million Somalis are estimated to have left the country while numbers of internally displaced are estimated to have exceeded a further million. The Somali region has not been in constant crisis since 1991; rather, conflict and displacement have fluctuated greatly in different geographical locations and over time (Menkhaus, 2006,2009). Lindley (2011) distinguishes between three phases in south-central Somalia: 1) a major displacement crisis in the early 1990s, caused by the onset of the war; 2) a period of localization and stabilization between 1996 and 2006 which entailed much less new movement; and 3) the transformation of the Somali conflict since 2006, in light of the global war on terror, which has caused new patterns of flight from political violence and persecution, as well as from hunger, following the devastating drought of2011.

In this section, we briefly highlight some of the protection realities for three groups: Somali refugees in the Dadaab camps of Kenya, the displaced in Hargeisa in Somaliland, and IDPs in Mogadishu in south-central Somalia. Different groups of displaced face different kinds of protection challenges: while those with refugee status have protection through the UN 1951 Refugee Convention, IDPs are not guaranteed the same status from the Guiding Principles on Internal Displacement. The responsibility to protect those who have not crossed state borders lies firmly with the government in question; however, it is undermined when the government in question is unable or unwilling to provide protection. International humanitarian action towards refugees and IDPs often differs considerably because of the limited levels of access that humanitarian organizations have to IDPs due to insecurity or government restrictions.

Dadaab, Kenya: Refugee Protection under Threat

Kenya has hosted Somali refugees since the start of the conflict, with large influxes in refugee numbers in the early 1990s and the late 2000s coinciding

with conflict patterns in Somalia. At the time of writing, 423,244 Somali refugees were registered in Kenya, amounting to 59 per cent of the total number of registered Somali refugees and asylum seekers in the East and Horn of Africa (UNHCR, 2015b). The majority of Somali refugees originate from Lower Jubba, Banadir (Mogadishu area) and Middle Jubba (ibid.). Most Somali refugees in Kenya live in the Dadaab camps in the North Eastern Province of Kenya, a province with mostly Somali-Kenyan inhabitants. While refugees are expected to live in Dadaab or in the Kakuma camps in the North, many have found their way to urban centres because of the potential opportunities there and due to their experience of poor conditions in the camps.

The increased securitization of the Somali conflict is intrinsically linked to regional dynamics and in particular to Kenya's role in the conflict. In October 2011, Kenya Defence Forces (KDF) launched a military offensive in Somalia against the Somali militant group al-Shabaab. In June 2012, Kenyan forces were integrated into the African Union Mission in Somalia (AMISOM). Kenya experienced an increase in security incidents, including the attack on the Westgate shopping mall in September 2013 in which 67 people were killed, which made international news headlines. The relationship between Kenyan police and defence forces and (Kenyan-)Somalis residing in the country has deteriorated, in a downward spiral of attacks, abuse and violence. As one donor employee points out: 'The Kenyan government is a tricky duty bearer, as it plays a double role. It is in Kismaayo with its army while at the same time hosting refugees. So it is a delicate situation for us'.

Concerns over the protection of Somali refugees have intensified — and their security situation in Kenya has worsened — as a consequence of the return agreement that was signed on 10 November 2013 by the Kenyan government, the Somali government and the United Nations High Commissioner for Refugees (UNHCR). In a recent public expression of Kenya's willingness to return refugees from Dadaab to south-central Somalia, Kenyan Deputy President William Ruto threatened on 11 April 2015 that if the UNHCR would not close the Dadaab refugee camp and provide alternative arrangements for its residents, Kenya would take the initiative to relocate the refugees back into Somalia. While repatriation has not yet taken place on a large scale, thousands of Somali refugees have been evicted to Mogadishu since the return agreement was signed. Repatriations are carried out amid worsening security conditions in Mogadishu. Many of those returned face dangers such as extortion, violence and sexual abuse (Amnesty International, 2014). Yet at the same time, more than half of camp inhabitants report feeling unsafe in the camps (RCK, 2012), and conditions in Dadaab amount to 'inhuman and degrading treatment' according to a UK ruling.3

3. Sufi and Elmi v UK (2011).

Hargeisa, Somaliland: Distinguishing Internally Displaced and Illegal Aliens

After a period of conflict, Somaliland has had almost two consecutive decades of relative stability. In the late 1980s, the region faced major repression under Siyaad Barre's regime. With the collapse of the state in 1991, Somaliland declared its independence, claiming the borders of the British protectorate — although, internationally, it is recognized only as an autonomous region of Somalia. Since an internal conflict among Isaq clans in 1994-96 was resolved, Somaliland has been relatively stable. The main threat to this stability derives from disputes over the eastern part of the territory, which is contested by Puntland (Lindley, 2013). According to figures from UNHCR (2015 c), only 84,000 internally displaced remain in the whole of Somaliland, the majority residing in Hargeisa and Burao.

In 2006, national policies were developed concerning IDPs in Somaliland, which distinguished between those originating from Somaliland and those 'from neighbouring countries such as Ethiopia and Somalia'. While the first group is recognized in the policy as internally displaced, the second is classified as 'illegal aliens or immigrants'. A considerable number of the displaced in Somaliland originate from south-central Somalia, and many of them reside in IDP camps. In these camps, there are relatively few security concerns. For most displaced people in Hargeisa, the main threat is eviction caused by disputes over the land on which a number of the IDP camps are located. For those from south-central Somalia, forced repatriation is an additional concern. Furthermore, due to limited clan protection and an insufficient legal apparatus in the camps, those from south-central Somalia are more vulnerable to attacks, abuse and assault.

Mogadishu, Somalia: Continued Security Concerns for Displaced Populations

Since the late 1980s, south-central Somalia has faced the greatest level of violent conflict and the largest displacement levels in the region. The current Federal Government of Somalia came to power in August 2012, ending Somalia's eight-year transition period. While this led to cautious optimism, the security environment remains challenging in various parts of south-central Somalia, including Mogadishu. There are currently still an estimated 893,000 IDPs in this region (UNHCR, 2015c) of which 369,000 are in the Mogadishu area (UNHCR, 2015d).

The security situation, as well as periods of severe drought and starvation, create ongoing displacement in south-central Somalia. IDPs are particularly vulnerable in this environment, and multiple displacements are not uncommon. In Mogadishu, IDPs are spread throughout the city, living in a number of designated camps or as squatters in public buildings. The general security environment in Mogadishu is difficult in itself, and IDPs face additional risks. While information flows are quite restricted from IDP camps, there

are reports of severe human rights abuses in the camps, in part due to the powerful position of 'gatekeepers' (Human Rights Watch, 2013). Furthermore, displaced people run an increasing risk of being evicted from where they live, both because previously unused government buildings are being reclaimed and because thousands of people return to Mogadishu each week (HIPS, 2013b). Moreover, the Federal Government considers the presence of displaced people in the city a security risk, and plans for relocations are underway. While it is ultimately the Somali government, with its police and judiciary, who should provide protection, there is still an environment of impunity, and priorities for international humanitarian actors may be very different from those of the government. A protection officer working for an international NGO, whom we interviewed in Nairobi, highlighted this:

The government of course says the right thing. After over 20 years they know how to deal with the international community. But they still need to pay the police and that is their priority, rather than Sexual and Gender Based Violence (SGBV) and other issues that the UN and NGOs come with, because SGBV follow-up and prevention is just providing redress to a select few rather than contributing to the stabilization of the country.

INTERNATIONAL PROTECTION MECHANISMS AND THE IMPORTANCE OF MOBILITY FOR SELF-PROTECTION

In a normative sense, the protection of civilians from the worst effects of violent conflict, human rights abuses and persecution — as well as from famine and loss of livelihoods — lies at the heart of the humanitarian agenda (Collinson et al., 2009). As Elizabeth Ferris points out, protection is a recognized responsibility of states towards their citizens and only when states are unable to provide protection does international law provide for protection by others (Ferris, 2011: 1). A commonly used definition of protection is that of the International Committee of the Red Cross (ICRC), which refers to obligations and rights in accordance with the letter and the spirit of the relevant bodies of law, in particular International Humanitarian Law, International Human Rights Law and Refugee Law: 'Protection aims to ensure that authorities and other actors respect their obligations and the rights of individuals in order to preserve the safety, physical integrity and dignity of those affected by armed conflict and other situations of violence' (ICRC, 2008: 752).

Humanitarian practice can entail preventative and mitigating measures, which differ in their time perspective and in their target populations. Pre-ventative measures are efforts that deal with the causes and circumstances of violations and abuses, preventing, putting an end to, and avoiding the recurrence of violations and human suffering (Gnaedinger, 2007). But to what extent can humanitarian action really prevent violations and abuses, given that prevention often requires long-term societal transformation? In the words of a protection officer working for the UN, whom we interviewed

in Nairobi: 'what is it that humanitarian aid workers, with their three months to one year perspective, can possibly do that would change behaviour?'. He illustrated his point with an example from Dadaab:

Prevention is not a humanitarian issue. Prevention is not simply about awareness-raising sessions. I stopped that in the camps. I said 'find me three women who do not know that firewood collection is risky'. But the conditions need to change: why is it that the men are chewing [khat] in the market while the women are exposed to that risk? We do not engage in single awareness-raising projects anymore, it has to be linked to a service or we won't do it. Telling people about their rights is problematic if no one upholds them.

In south-central Somalia, with its perceived environment of impunity, genuine preventative work rarely takes place. Humanitarian workers do not consider prevention to be a humanitarian task. The few preventative measures that do take place tend to focus on providing people with material and infrastructural support that lowers personal security risks. Nevertheless, a number of the humanitarian aid workers we spoke to feel dissatisfied with the gap between material aid and the 'real issues' at stake. In line with Abdelnour and Saeed's (2014) critique of the 'rape-stove panacea', a human rights professional working for an international human rights organization in Nairobi stated that:

There is no accountability for protection issues, the international NGOs do not focus on this but the donors do not either. Instead, there are very reactive measures taken, such as street lights, firewood-efficient stoves and such. I can see that putting up street lights in Mogadishu is a good step to improve security there, sure, but it does not address the main issues. As to those stoves, I do not see the point. Most women are raped in their houses, at night — it has nothing to do with firewood collection. There is definitely a role for humanitarian actors to do more. But many of the real underlying structural problems, and thus solutions, are political. The diplomatic community then has the main responsibility.

Mitigating measures, on the other hand, are efforts that address the consequences of violations and abuses by mitigating human suffering and helping people overcome the negative effects of abuses. This is clearly a more appropriate area for humanitarian action, but those we spoke to in our study were generally sceptical of the value of such types of protection measures. Mitigating measures were often seen to be of limited value if they were not accompanied by preventative measures that addressed the underlying causes of abuses and violations. Yet, funding for long-term protection initiatives in chronic emergencies is limited. Results are slow and very difficult to measure, while societal transformations are also difficult to attribute to particular interventions. A donor representative in Nairobi elaborated on this challenge:

In the end, protection issues do not end up in joint assessment reports as major gaps or areas of future focus, since we like to give ourselves tasks that we can achieve. It is much easier to talk about amounts of food, numbers of people receiving education, than it is to talk about protection. That is far less tangible. We focus on measurable results. In order to make a difference in protection issues you really have to engage the community. We love quick fixes, but what is needed is long-term community engagement . . . Another aspect is

that NGOs are all competing for scarce resources. Maybe people are no longer shooting kids, but at the same time a donor could fund a shelter for malnourished women and children and have their logo on the building.

These are clear examples that show just how complicated it is to aim to put normative protection frameworks in practice considering the humanitarian realities. In the end, protection may be seen not as a field of activity but rather as a rallying call or a (hard-to-fund) budget line.4 Ultimately, humanitarian practice can only play a limited role in providing displaced individuals with protection.

Self-Protection: The Central Role of Mobility

Research shows that individuals and communities are always central to their own protection. In most conflict situations, the majority of people have no access to international protection. As Slim and Bonwick (2005: 32) indicate, 'protection is not just a commodity or service that can be delivered like food or healthcare. It is also something that people struggle for and achieve within a given situation, or secure more widely in the politics of their own society'. As such, in order to understand how displaced Somalis cope with the many protection challenges they face in Somalia and in the region, it is crucial to start by exploring the strategies they use to protect themselves and others.

One of the most efficient ways in which civilians who face persecution or are caught up in violent conflict can protect themselves is by leaving. But while mobility is central in people's protection strategies, at the same time displacement often challenges the safety, dignity and integrity of the individuals involved in specific ways. In order to illustrate people's complex individual motivations as well as the contextual triggers that influence the decision to migrate or not, we present two examples of how decisions to move are being made and acted upon by those in protracted displacement situations.

Halima is a 45-year-old woman originating from Mogadishu who we spoke to in an IDP camp in Hargeisa. She had lived in Hargeisa once before, for four years, deciding to return to Mogadishu because she felt she was not welcome in Hargeisa and she found it difficult to survive there with her children. Although, upon returning, she preferred life in Mogadishu over life in Hargeisa, circumstances forced her back to Hargeisa again. Halima explained her second move from Mogadishu to the IDP camps in Hargeisa as follows:

I was happy to stay in Xamar [Mogadishu]. Although at the time the security situation was not so good, there were better opportunities for me to provide for my children. But the problem was that al-Shabaab had their eyes set on my son. He was only a teenager and he did nothing wrong. But they had accused him of being against Islam and they wanted to take

4. Thanks to Oliver Bakewell for this comment.

him to sentence him. This is what forced us to flee back to Hargeisa. I would not have been a good mother to my son if I had let al-Shabaab take him. I had to protect him, so I made the decision for us to come back here. If it were not for my son, I would have remained in Xamar.

Halima's explanation of her return to IDP life in Hargeisa, with all its hardships, illustrates a number of points. First, mobility is a central strategy for Halima in guaranteeing her family's survival. The decision to migrate or not is something she reflects on constantly, while in Mogadishu, which is a conflict zone, and also in Hargeisa, where she and her family live as displaced. Second, in her migration decisions, Halima balances security needs against livelihood survival, demonstrating the point that, for those affected, 'protection' encompasses not just physical safety but also basic needs (Jaspars et al., 2007). Finally, Halima's descriptions clearly highlight the fact that decisions to migrate or not are rarely based on individual protection concerns. Rather, such decisions are taken on a family level since different family members face different risks and have access to different strategies to protect themselves and others.

Hassan, a 24-year-old man we spoke to in Hargeisa, had already moved several times between different locations — something that is quite common among Somali displaced. He explained his various moves as follows:

I am originally from Xamar. But I fled to Kenya when the war worsened and the situation in Xamar was really bad. This was about four years ago and I made my way to the border between Somalia and Kenya. I lived in Hagadhere refugee camp in Dadaab. I decided to leave Kenya in January this year. I left because the situation in Dadaab was very bad. It was just too dangerous to stay there so I decided to return to Somalia. I travelled through Xamar. When I left Hagadhere I went first to Xamar and I stayed there for 15 days before deciding to travel to Hargeisa. I found it too difficult to stay in Xamar. It was too hot and there were no real opportunities. I am a young man and I need education. This is very important to me. So I travelled to Hargeisa because I thought I would have better opportunities here.

Hassan's story gives further insight into the use of mobility as a protection strategy. First, movement in and from conflict regions is motivated by a range of factors; including protection needs. Hassan's first move from Mogadishu was related to conflict and insecurity, while the second move was guided by the lack of opportunities he experienced and his eagerness to get an education. Second, a lack of physical safety can be experienced in displacement as much as in conflict contexts. Increased insecurity has become a general concern in Dadaab, and among young men there is also the fear of (forced) recruitment into organizations like al-Shabaab. This fear combines with a concern for anti-terror activities undertaken by the Kenyan police and military. Third, Hassan's motivations for his various moves illustrate that life cycle changes play an important role in migration choices. Being a young man not only puts him at risk of military recruitment; his wish to be able to build a livelihood in order to establish a family is a central drive for Hassan to search for education and/or job opportunities (Grabska and Fanjoy, 2015).

The migration choices and considerations of Halima and Hassan clearly illustrate the importance of mobility in protection strategies among Somali displaced. However, while migration plays a crucial part in self-protection mechanisms for those living in conflict environments, mobility is highly constrained in contemporary conflict zones. The movement of people is often deliberately restricted by imposing curfews, enforcing roadblocks, closing borders or forcibly returning people to unsafe areas. Restrictions of movements can also be 'self-imposed', as a consequence of the extreme fear caused by conditions of violence and insecurity. This particularly applies to women, who are more often targets of sexual violence (Skjelsbaek, 2011). Furthermore, increasingly restrictive migration regimes worldwide reflect a shrinking protection space for potential refugees. Finally, as we argue below, humanitarian approaches also restrict and/or aim to control people's mobility in various ways.

HUMANITARIAN APPROACHES TO PROTECTION FOR SOMALIS

Given the importance of mobility for self-protection strategies, how does it feature in humanitarian approaches to protection? How is mobility impacted by such approaches? Humanitarian assistance to displaced populations is generally understood as a temporary activity that lasts until people return 'home'. Yet, significant advancements in the theorization about the relationship between people, place and identity have been made in anthropology (Clifford, 1997; Malkki, 1992). Mobility plays a central part in many people's livelihood strategies and life histories, with 'routes' complicating 'roots' (Clifford, 1997). The implications of these insights have been explored in the fields of development studies (Bakewell, 2008), refugee studies (Hyndman, 2012; Hyndman and Giles, 2011) and mobility studies more generally (Hannam et al., 2006; Urry, 2007). The debate has evolved through the decades, exploring inter alia how the geographic mobility of people forces them to redefine their homes and their identifications with places, or deterritorialized spaces, that they might or might not corporeally inhabit (Sinatti and Horst, 2015). In particular, those who conduct research in places where mobility is a central aspect of many people's livelihoods have challenged a static, sedentary perspective on people dealing with conflict and crisis (Horst, 2006; Monsutti, 2004). In humanitarian studies the implications of such perspectives and the politics of mobility have not been discussed extensively.

Many types of humanitarian aid are tied to an individual's status as a refugee (or IDP) and cease once that status has ended — or has been proclaimed to have ended by the host state and/or international humanitarian actors (Omata, 2013). Funding realities and donor preferences play an important role here, as is clearly illustrated by the stabilization discourse that fuels the strategy of return currently advocated in the Somali context. This

strategy takes place in the context of a limited humanitarian space, in which access to populations is restricted and, simultaneously, in a situation where mobile forms of assistance are not very common.

A Limited Humanitarian Space

As already discussed, Somali displaced populations face challenging protection realities, particularly in south-central Somalia. In this environment, it is also quite difficult for aid agencies to operate effectively. The kidnapping of aid workers is a security concern (Program on Humanitarian Policy and Conflict Research, 2012), as are military operations such as those against al-Shabaab in 2014. Another challenge is the increasing blurring of lines between UN political and humanitarian affairs, which gives the perception that humanitarian operations are a tool to achieve political ends (Rotelli, 2014). While 24 UN agencies and over 100 national and international NGOs continue to operate in Somalia, most humanitarian partners rely on national staff to implement programmes, due to high levels of insecurity (OCHA, 2014). Attacks and threats against humanitarian staff increased in 2014, which has affected the movement of humanitarian partners. As a result, as Hammond and Vaughan-Lee (2012) point out, humanitarian space in Somalia can be considered a scarce commodity.

This lack of an international presence on the ground, due to the security situation, is one of the greatest hindrances to carrying out protection work in Mogadishu. Many organizations operate through remote management via local organizations and the UN presence is also limited. In October 2010, the UN operated out of Mogadishu airport and in 2012 it moved 800 meters from the airport to its own compound. Yet, as most informants point out, it is impossible to guarantee protection if one has no authority to speak of, and there are hardly any local NGOs with sufficient capacity on the ground. Operating through 'remote control' is always problematic for a host of reasons, and particularly in the case of protection activities, as a protection officer working for the UN in Nairobi pointed out:

One concern is how to do protection when you are not there. It cannot be done remotely, despite what the literature says. The monitoring aspect cannot be done remotely. How can you guarantee that the rights of IDPs are respected when you are not there, especially when the government is also not there? Mitigating activities are important and can be done to some extent, but the preventative part is very difficult.

Aid diversion is another commonly acknowledged concern; it is widely recognized that aid has reached, and will continue to empower, people who are contributing to insecurity in the country (Human Rights Watch, 2013; Yarnell, 2012). As a protection officer indicated, 'aid without presence is worse, as it is fuelling conflict'. The UN Monitoring Group (Security Council Committee, 2010: 60) highlights an example of this, describing

how World Food Programme ground transporters and local implementing partners diverted up to 50 per cent of aid, of which 5 to 10 per cent went to al-Shabaab. Others have reported that agencies had to negotiate with al-Shabaab over access, were required to pay 'taxes', and were subject to checks over their operations (Jackson and Aynte, 2013a, 2013b). These negotiations were done by local staff, at times far from the control of head office.

One of the main challenges is monitoring and evaluation. As preventative forms of protection activities are, to a large extent, about changing attitudes and behaviours, the effects of such activities are notoriously difficult to measure. If these activities are carried out by local organizations, through projects that cannot be visited, then international humanitarian aid workers will have additional concerns about how to guarantee that protection activities do in fact take place and have an impact. Donor regulations and anxiety over the possible funding of terrorism have led to increasing reporting demands, as well as to concerns amongst organizations that this information will be used in counterterror operations (Program on Humanitarian Policy and Conflict Research, 2012).

Mobile Lives: Immobile Aid

Mobility plays a central part in historic and current livelihood strategies, life choices and a sense of belonging among many Somalis. The underlying assumption that, ultimately, people are rooted in a certain fixed place and that movement needs to be prevented, or reversed through return, is problematic considering actual lived realities (Sinatti, 2015; Sinatti and Horst, 2015). As the stories of Halima and Hassan illustrate, when individuals move they juggle a range of security and livelihood concerns, for different family members over time and in different geographical locations. The challenge for humanitarian actors is to create much more mobile forms of assistance that impact mobility only minimally, or in ways that positively affect the choices in staying or moving for people affected by conflict; examples include advocacy that increases freedom of movement, and the provision of cash, which is portable and could be accessed through a country-wide remittance infrastructure.

At present, such mobile forms of assistance are rare and humanitarian assistance is largely unable to relate to the mobile strategies that individuals use to maximize protection and guarantee livelihoods. A particular incident that was mentioned by several of the humanitarian aid workers that we interviewed during our study in Nairobi illustrates the problem posed by mobility. According to our informants, large numbers of people 'suddenly' disappeared from the Afgoye corridor into Mogadishu. The international community was taken by surprise, as the Afgoye corridor emptied virtually overnight. A protection officer for an international NGO, whom we interviewed in Nairobi, expressed his concern:

This was kind of embarrassing for the international organizations working there as nobody saw the move coming. There used to be 100,000 to 200,000 in the Afgoye corridor, and they moved to Mogadishu... Two years ago, we thought that Afgoye was urbanizing and about to become a suburb of Mogadishu. That didn't happen. People move all the time, to the desperation of the NGOs — this is very inconvenient for us.

Current humanitarian initiatives are investing in attempts to relate to what the NGO protection officer quoted above refers to as the 'inconvenience' of mobility. These initiatives include the tracking of movements of people, such as Protection Monitoring Networking (PMN) and Population Movement Tracking (PMT) headed by UNHCR and implemented by its partner organizations. While such mobility tracking initiatives are presented as facilitating humanitarian work, there is a risk that they could be used to enable migration management and control. Furthermore, they do not necessarily imply the creation of more mobile forms of aid.

Not only is the mobility of displaced populations understood to complicate humanitarian work, there is an additional concern that aid has an impact on mobility patterns and in particular makes people move away from home areas. Thus, the location of aid is much debated. In several conversations, humanitarian aid workers voiced the need to protect 'the right to remain'. This discourse is a reminder of the 'de facto containment approach' to human displacement, known at the UN refugee agency in the 1990s as 'preventative protection' (Hyndman, 2012: 245). A UN protection professional in Nairobi argued:

By what we are doing, we are making people stay in locations where they would not have resided otherwise. This is detrimental, creating these hubs of services. During the last famine, we were trying to get resources to the Bay and Bakool areas where al-Shabaab has a great presence, so that people do not move from there. We should not be in Dobley [border town], but in Kismaayo, in Afmadow... We should be closer to their own areas, so that the moves are shorter.

Having the choice of remaining in or returning to locations of socio-cultural and politico-economic value is important, as is the possibility of living together as a family. Nonetheless, the quote above merits further reflection. While it is important that people have the option to remain, underlying assumptions of why people leave are crucial. Do the inhabitants of Kismaayo leave the city because the international community is not providing humanitarian aid there, or are they leaving because of the continuous bouts of violent conflict and the implications of these? Throughout the war, control over Kismaayo has been contested because it has one of the country's largest sea ports; it is one of Somalia's most diverse cities in terms of clan composition; and it is the natural entry point to Somalia's most fertile region, the inter-riverine area (HIPS, 2013a). In this context, the humanitarian space is extremely limited, but aid that is provided 'remotely' runs a range of risks, as discussed above. We might question whether the presence or absence of international humanitarian aid is really relevant in determining

the migration decisions of individuals and families when basic safety and physical security are such a central concern.

The humanitarian focus on the right to remain might link closely with an interest to contain. Similarly, as Hyndman (2012) points out, there is a risk that individual protection concerns are insufficiently taken into account. Attempts to govern mobility in the Somali region are not restricted to border control and migration measures; they are also implemented through development aid and humanitarian aid. Laura Hammond convincingly argues that in the 25 years after the 1984-85 Great Sahelian Famine, the Ethiopian government used the management of mobility as a tool to control the poorest of its citizens, all in the context of humanitarian crises (Hammond, 2011). We argue that similar patterns can be observed when analysing international humanitarian aid, even though it is governments rather than international humanitarian actors who have a political interest in population management.

The Stabilization Discourse: An Increasing Focus on Return

As part of the stabilization discourse, the ideal of return to south-central Somalia is promoted quite strongly by a range of actors in the region as well as internationally. Since 2012, Kenya has increased its focus on return for Somali refugees, developing and presenting plans to repatriate more than half a million people, despite continued instability in Somalia (HIPS, 2013b). These plans have largely been a response to the increasing vulnerability of Kenya to terrorist attacks, whether directly related to the invasion of Somalia or not, which have cost lives and had major implications for Kenya's economy, especially its tourist industry (ibid.: 8). Surprisingly, perhaps, there has not been a serious international outcry in response to these plans. The response may in fact be complicated by the fact that several European governments are eager to start returning Somali asylum seekers to south-central Somalia. As a human rights professional in Nairobi, working for an international NGO, pointed out:

The Kenyan government then says it is safe to return, and talks about 'liberated areas', which is rubbish... But it is actually DFID and ECHO pushing this as well — DFID has their stabilization programmes that will receive a lot of money if Somalia is considered ready for the next phase . . . Sweden changed its policy on returning, and removed the prima facie status. Return to Mogadishu is now no longer considered unsafe. It is a highly political subject at the moment, and the humanitarian aid workers are under pressure because return is the main goal.

Simultaneously with the debates on return of the displaced, governments and INGOs have started considering the relocation of their own activities and of aid provision. Several organizations have discussed relocation from Nairobi to various 'safe' parts of south-central Somalia in light of substantially reduced funding streams into Kenya since 2012 (see Figure 1) and donor requests for programmes on return of displaced populations. During

Figure 1. Humanitarian Aid Spending in Kenya and Somalia

USD 1 600 USD 1 400 USD 1 200 USD 1 000 I USD 800 USD 600 USD 400 USD 200 USD 0

2009 2010 2011 2012 2013 2014 Year

Source: UN Office for the Coordination of Humanitarian Affairs (OCHA) Financial Tracking Service

our fieldwork, both UN and NGO employees discussed the need for their organizations to position themselves in relation to the plans of action of the Kenyan and Somali governments to return displaced people, and the increasing interest among donors in stabilization and return. They also talked about dilemmas. Should they collaborate if they felt that stimulating return was premature? If they would not collaborate, would the situation of the displaced be made even worse? Should they develop programmes on return because there were increasing numbers of Somali refugees, for example, reportedly interested in returning, or because this seemed to be a very fundable activity? How could an organization remain critical and relevant at the same time?

While donor spending in south-central Somalia has been relatively stable since 2012, humanitarian aid to Kenya has reduced substantially during the same period. Humanitarian agencies, if they want to remain relevant, need to follow suit by focusing their operations more in Somalia. It is a well-documented practice that assistance in refugee camps is reduced in order to encourage people to go 'home'. Nao Omata discusses how Liberian refugees in Buduburam refugee settlement in Ghana found themselves living on dwindling international support (Omata, 2011). Omata convincingly demonstrates that decreasing aid was not just a consequence of the limited funding available but was part of a strategic approach to encourage Liberians to return 'home'. When humanitarian actors in practice become implementers of migration-stimulating policies, they need to take into account the long-term security implications of their decisions on where to

provide aid — even if those decisions are guided by funding realities and the policy priorities of donor and host states (Horst and Sagmo, 2015).

CONCLUSION: GOVERNING MOBILITY THROUGH HUMANITARIANISM?

In this article, we have argued that humanitarian policies and practices do not take sufficient account of the importance of mobility in the lived realities of the people they aim to assist. We have illustrated how this gap hampers the provision of protection to those displaced by the Somali conflict, as humanitarian aid provision and withdrawal can be used as instruments in preventing displacement or encouraging return. The importance of such policies and practices is particularly visible in Somalia's 'transitional phase' that has been in place since 2012 and is characterized by a shift from an emergency approach to stabilization and return programming. This shift is accompanied by substantially reduced humanitarian aid budgets to Kenya and stable support to Somalia, where insecurity is still rampant and the newly established government faces considerable challenges. While an active and massive relocation across the border is less likely in the short term, reduced assistance to the Dadaab camps and increased funding support for refugee return is challenging protection needs on a range of levels. Ultimately, this shift in programming reduces people's freedom to choose the mobility strategies that address their individual and family protection concerns in the way they themselves judge best.

Humanitarian actors are not necessarily initiating attempts to govern mobility, but are caught up in daily realities in which they become imple-menters of the mobility-managing attempts of governments. Their dependency on funding from governments has an impact on the location of their programmes, as does their ability to physically access certain areas and their wish to remain relevant. Furthermore, particular perceptions of home, displacement and return prevent international humanitarian organizations from challenging the assumptions behind the return efforts of those governments. The expectations that humanitarian actors have about the mobility patterns of displaced people and their understandings of home, influence the nature and location of protection activities. Yet in contexts like Dadaab, the definition of 'home' is debatable, for instance when the children and grandchildren of those who fled a certain conflict have been born in a refugee camp and have never been to their 'homeland'. While protracted encampment is deeply problematic, understanding return as the solution to displacement is just as problematic when violent conflict is ongoing and humanitarian aid is largely provided through remotely-managed operations. There is a clear risk that individual protection is compromised for the sake of the stabilization discourse in south-central Somalia and the interests of many governments in encouraging return.

When envisioning new forms of humanitarian assistance that cater for the mobile protection realities of recipients of aid, both humanitarian technologies and advocacy are important ingredients. New forms of assistance could potentially provide people with greater flexibility, for example in deciding where they wish to make use of the aid. Aid recipients could be offered cash-based forms of assistance through (mobile) money transfer systems which allow returnees to access the money at any location. The Somali hawala money transfer system and the Kenyan MPesa mobile banking system are examples of services that can enable such cash-based assistance that is not tied to a certain location. Increasingly, this mobile cash can be used to pay directly for goods and services in the region. At the same time, advocacy and lobbying for greater freedom of movement for displaced populations may be a necessary part of a humanitarian approach that addresses mobile protection realities more effectively. The UNHCR's stance in relation to the Economic Community of West African States (ECOWAS) protocol on free movement in West Africa is an interesting example. Rather than governing one type of movement — return — a focus on enabling free movement allows displaced populations to move to a place where they can find protection. By taking the mobile realities of displaced populations as a starting-point, such alternative types of humanitarian aid can enable international organizations to uphold their core responsibility: the protection of civilians from the worst effects of violent conflict.

REFERENCES

Abdelnour, S. and A. Saeed (2014) 'Technologizing Humanitarian Space: Darfur Advocacy and the Rape-stove Panacea', International Political Sociology 8(2): 145-63.

Amnesty International (2014) 'No Place like Home: Returns and Relocations of Somalia's Displaced'. London: Amnesty International.

Bakewell, O. (2008) '"Keeping Them in Their Place": The Ambivalent Relationship between Development and Migration in Africa', Third World Quarterly 29(7): 1341-58.

Black, R. and K. Koser (eds) (1999) The End of the Refugee Cycle? Refugee Repatriation and Reconstruction. Oxford and New York: Berghahn.

Clifford, J. (1997) Routes. Travel and Translation in the Late Twentieth Century. Cambridge, MA: Harvard University Press.

Collinson, S., J. Darcy, N. Waddell and A. Schmidt (2009) 'Realising Protection. The Uncertain Benefits of Civilian, Refugee and IDP Status'. HPG Report 28. London: Overseas Development Institute.

Ferris, E. (2011) The Politics of Protection. The Limits of Humanitarian Action. Washington, DC: Brookings Institution Press.

Gnaedinger, A. (2007) 'Protection of Civilians in Conflict: The ICRC Perspective'. Statement Address by Angelo Gnaedinger, ICRC Director-General, Humanitarian and Resident Coordinators' Retreat, Geneva (9 May 2007).

Grabska, K. (2014) Gender, Home and Identity. Nuer Repatriation to Southern Sudan. Woodbridge, Suffolk: James Currey.

Grabska, K. and M. Fanjoy (2015) '"And When I Become a Man": Trans-local Coping with Precariousness and Uncertainty among Returnee Men in South Sudan', Social Analysis 59(1): 76-95.

Hammond, L. (2004) This Place Will Become Home: Refugee Repatriation to Ethiopia. Ithaca, NY and London: Cornell University Press.

Hammond, L. (2011) 'Governmentality in Motion: 25 Years of Ethiopia's Experience of Famine and Migration Policy', Mobilities 6(3): 415-32.

Hammond, L. and H. Vaughan-Lee (2012) Humanitarian Space in Somalia: A Scarce Commodity. London: Humanitarian Policy Group.

Hannam, K., M. Sheller and J. Urry (2006) 'Editorial: Mobilities, Immobilities and Moorings', Mobilities 1(1): 1-22.

HIPS (2013a) 'The Kismaayo Crisis: Options for Compromise'. HIPS Policy Briefing Issue (4 June). Mogadishu: Heritage Institute for Policy Studies.

HIPS (2013b) 'Hasty Repatriation. Kenya's Attempt to Send Somali Refugees Home'. HIPS Report. Mogadishu: Heritage Institute for Policy Studies.

Horst, C. (2006) Transnational Nomads. How Somalis Cope with Life in the Refugee Camps of Dadaab, Kenya. Oxford and New York: Berghahn.

Horst, C. andK. Grabska (2015) 'Flight and Exile: Uncertainty in the Context of Conflict-induced Displacement', special issue of Social Analysis 59(1): 1-18.

Horst, C. and T. Sagmo (2015) 'Humanitarianism and Return: Compromising Protection?'. PRIO Policy Brief (03 2015). Oslo: Peace Research Institute Oslo.

Human Rights Watch (2013) 'Hostages of the Gatekeepers: Abuses against Internally Displaced in Mogadishu, Somalia'. HRW Report (28 March). https://www.hrw.org/ report/2013/03/28/hostages-gatekeepers/abuses-against-internally-displaced-mogadishu-somalia

Hyndman, J. (1999) 'A Post-Cold War Geography of Forced Migration in Kenya and Somalia', Professional Geographer 51(1): 104-14.

Hyndman, J. (2000) Managing Displacement. Refugees and the Politics of Humanitarianism. Minneapolis, MN: University of Minnesota Press.

Hyndman, J. (2012) 'The Geopolitics of Migration and Mobility', Geopolitics 17(2): 243-55.

Hyndman, J. andW. Giles (2011) 'Waiting for What? The Feminization of Asylum in Protracted Situations', Gender, Place and Culture 18(3): 361-79.

ICRC (2008) 'ICRC Protection Policy. Institutional Policy', International Review of the Red Cross 90(871): 751-75.

Ikanda, F. (2014) 'Kinship, Hospitality and Humanitarianism: "Locals" and "Refugees" in Northeastern Kenya'. Doctoral Dissertation, Social Anthropology, University of Cambridge.

Jackson, A. and A. Aynte (2013a) 'Al-Shabaab Engagement with Aid Agencies'. Policy Brief No. 53. London and Mogadishu: Humanitarian Policy Group and Heritage Institute for Policy Studies.

Jackson, A. and A. Aynte (2013b) 'Talking to the Other Side: Humanitarian Negotiations with Al-Shabaab in Somalia'. London and Mogadishu: Humanitarian Policy Group and Heritage Institute for Policy Studies.

Jaspars, S., S. O'Callaghan and E. Stites (2007) 'Linking Livelihoods and Protection: A Preliminary Analysis based on a Review of the Literature and Agency Practice'. Humanitarian Policy Group Working Paper. London: Overseas Development Institute.

Lindley, A. (2011) 'Between a Protracted and a Crisis Situation: Policy Responses to Somali Refugees in Kenya', Refugee Survey Quarterly 30(4): 14-49.

Lindley, A. (2013) 'Displacement in Contested Places: Governance, Movement and Settlement in the Somali Territories', Journal of Eastern African Studies 7(2): 291-313.

Lubkemann, S. (2008) 'Involuntary Immobility: On a Theoretical Invisibility in Forced Migration Studies', Journal of Refugee Studies 1(4): 454-75.

Malkki, L. (1992) 'National Geographic: The Rooting of Peoples and the Territorialization of National Identity among Scholars and Refugees', Cultural Anthropology 7(1): 24-44.

Menkhaus, K. (2006) 'Governance without Government in Somalia: Spoilers, State Building, and the Politics of Coping', International Security 31(3): 74-106.

Menkhaus, K. (2009) 'Somalia: "They Created a Desert and Called it Peace(building)"', Review of African Political Economy 36(120): 223-33.

Monsutti, A. (2004) 'Cooperation, Remittances, and Kinship among the Hazaras', Iranian Studies 37(2): 219-40.

OCHA (2014) 'Somalia 2015 Humanitarian Needs Overview. Annex: Operational Environment'. New York: UN Office for the Coordination of Humanitarian Affairs. http://reliefweb. int/report/somalia/somalia-humanitarian-needs-overview-2015 (accessed 30 June 2015).

Omata, N. (2011) 'The Livelihood Strategies of Liberian Refugees in Ghana. The Significance of Refugees' "Social Worlds" for their Economic Survival'. Doctoral thesis, London School of Oriental and African Studies.

Omata, N. (2013) 'The Complexity of Refugees' Return Decision-Making in a Protracted Exile: Beyond the Home-Coming Model and Durable Solutions', Journal of Ethnic and Migration Studies 39: 1281-97.

Program on Humanitarian Policy and Conflict Research (2012) 'Countering Terror in Humanitarian Crisis: The Challenges of Delivering Aid to Somalia'. Cambridge, MA: Harvard University.

RCK (2012) 'Asylum under Threat: Assessing the Protection of Somali Refugees in Dadaab Refugee Camps and along the Migration Corridor'. Nairobi: Refugee Council of Kenya.

Richards, P. (2005) No Peace, No War. An Anthropology of Temporary Armed Conflict. Oxford: James Curry.

Rotelli, M. (2014) 'Humanitarian Actors' Struggle for Access, Impartiality and Engagement with Armed Non-state Actors'. Geneva: Professionals in Humanitarian Assistance and Protection (PHAP). https://phap.org/articles/humanitarian-actors-struggle-access-impartiality-and-engagementarmed-non-state-actors-soma (accessed 30 June 2015).

Security Council Committee (2010) 'Report of the Monitoring Group on Somalia. Report Submitted in Accordance with Resolution 1853 (2008), S/2010/91' (10 March). New York: UN Security Council

Sinatti, G. (2015) 'Return Migration as a Win-Win-Win Scenario? Visions of Return among Senegalese Migrants, the State of Origin and Receiving Countries', Ethnic and Racial Studies 38(2): 275-91.

Sinatti, G. and C. Horst (2015) 'Migrants as Agents of Development: Diaspora Engagement Discourse and Practice in Europe', Ethnicities 15(1): 134-52.

Skjelsbsk, I. (2011) The Political Psychology of War Rape: Studies from Bosnia and Herzegovina. London: Routledge.

Slim, H. and A. Bonwick (2005) Protection. An ALNAP Guide for Humanitarian Agencies. London: Overseas Development Institute.

Suhrke, A. and A. Berdal (eds) (2012) The Peace in Between: Post-War Violence and Peace-building. London: Routledge.

UNHCR (2015a) UNHCR Statistical Yearbook 2014. Geneva: UNHCR.

UNHCR (2015b) 'Somalia: Areas of Origin (Regions) of Registered Somali Refugees and Asylum-seekers in the East and Horn of Africa'. Geneva: UNHCR. http://reliefweb. int/report/somalia/areas-origin-regions-registered-somali-refugees-andasylum-seekers-east-and-horn (accessed 30 June 2015).

UNHCR (2015c) 'Somalia Refugees in the Horn of Africa and Yemen'. Geneva: UNHCR. http://reliefweb.int/map/somalia/somalia-refugees-horn-africa-and-yemen-april-2015 (accessed 30 June 2015).

UNHCR (2015d) 'Total IDPs by Region (April 2015)'. http://reliefweb.int/map/somalia/ somaliatotal-idps-region-april-2015 (accessed 30 June 2015).

Urry, J. (2007) Mobilities. Cambridge: Polity Press.

World Bank (2011) World Development Report 2011: Conflict, Security and Development. Washington, DC: World Bank.

Yarnell, M. (2012) 'Gatekeepers and Evictions: Somalia's Displaced Populations at Risk'. Field Report. Washington, DC: Refugees International. http://www.alnap.org/resource/12957

Cindy Horst (corresponding author; e-mail: cindy@prio.org) is Research Director at the Peace Research Institute Oslo (PRIO), Norway, and Research Professor in Migration and Refugee Studies. Her current research interests include: mobility in conflict; diaspora; humanitarianism; refugee protection; transnational civic engagement; and theorizing on social transformation. For recent publications see: www.prio.org/staff/cindy.

Anab Ibrahim Nur (e-mail: anab.nur82@gmail.com) is a Human Rights Lawyer currently based in Mogadishu, Somalia. She has been involved in various research projects for Heritage Institute for Policy Studies and is now focused on governance, working to strengthen Somali government institutions.