This essay series explores the human costs and policy challenges associated with the displacement crises in the Mediterranean and Andaman Seas. The essays explore the myths or misconceptions that have pervaded discussions about these two crises, as well as the constraints or capacity deficiencies have hampered the responses to them. See more ...

As the Syrian refugee crisis enters its fifth year, major shortcomings of the international humanitarian response, particularly in the first several years, have begun to emerge. Important among these is the near-exclusive focus on emergency-based responses which largely ignored the needs — arguably even the presence — of the host communities, which in many instances were among the first to provide support to incoming refugees.[1] In part due to a broader move from an emergency to development-based approach and due to pressure from central Lebanese government authorities, the humanitarian effort has now been coupled, since mid-2014, with one that takes into greater account the needs of local host communities alongside those of refugees.

This shift has manifested itself both in greater support to national programs that provide aid to vulnerable Lebanese, such as support to the National Poverty Targeting Programme (NPTP), as well as community-level support through “ projects aimed [at] enhanc[ing] public services, mitigat[ing] the negative effects of the refugee presence and prevent[ing] tensions between populations.”[2]

This essay focuses primarily on the latter of these two efforts and traces the way in which tensions between hosts and refugees have become increasingly central to the development and execution of aid projects aimed at community-level support. I argue that this has important consequences that may actually incentivize the tensions it aims to alleviate. First, the notion of “host-refugee” tension is premised on, and reinforces, a false assumption that the categories of hosts and refugees, particularly but not exclusively in Lebanon, are ‘naturally’ distinguishable and discrete communities. Second, in an attempt to make “legible,” to use Scott’s (1998) terminology,[3] complex social dynamics, it renders perceived levels of tension and conflict as yet another layer of comparison for the selection of aid recipients. This bears the potential for severe unintended consequences by incentivizing local authorities and communities to, at minimum, display grievances, and, even more dangerously, to encourage conflict, with the aim of attracting development funding.

The International Aid Response to the Syrian Refugee Crisis in Lebanon

The international aid response to the Syrian refugee crisis in Lebanon has gone through, at least, two distinct phases, and it appears to be on the brink of a third. The first began in the immediate period following the influx of displaced Syrians into the country, and can be characterized by the fact that it was internationally-led and implemented, with little to no involvement from Lebanese authorities. It was primarily an emergency-based response that focused exclusively on immediate needs of the Syrian refugee population. In this period, as one senior staff person at an international development organization stated, “the most simple do no harm issues were not applied in Lebanon [such that…] calling it embarrassing does not begin to describe the situation.”[4] At this stage, the presence of local communities was perceived as largely irrelevant to the provision of aid.

As signs of increased tension among communities became more apparent, international aid providers and national authorities “woke up” to the significance of ignoring host community needs. As a result, “[while] the pendulum swung 100% to the Syrians in the first two to two and and half years of the response, […] when it swung back, it swung back to an extreme level because people wanted to compensate for their failures.”[5] This second phase was largely characterized by the increased involvement of the national government and the greater emphasis on development-focused programs, reflected in the Lebanon Crisis Response Plan (LCRP) produced jointly by the Government of Lebanon (GoL), U.N. Development Program (UNDP), and the U.N. High Commissioner for Refugees (UNHCR).

A close reading of the LCRP reveals an effort to emphasize the role of large and small-scale development projects as direct contributors to “social cohesion” (and later, “social stability”).[6] As one senior staff person at an international peacebuilding non-governmental organization (NGO) told me, “the government has an incentive to use the social cohesion framework to opportunistically tap into crisis funding in order to develop the country.”[7] To a large extent, that has been successful: the amount of aid directly going to Lebanese public institutions more than doubled between 2014 and 2015 (from US$77.9 million US$171.5 million),[8] at a time when the overall funding to the country increased by approximately 20 percent.[9]

Due in part to the hesitation on the part of donors to directly fund the notoriously corrupt national institutions, a significant portion of the funding was immediately dedicated to supporting municipalities directly. In 2014, 46 percent of this funding was dedicated to supporting national institutions, while 54 percent was allocated to municipal support.[10] The emphasis on support to municipalities was strengthened in 2015, with a 50 percent increase in the number of projects,[11] and further reinforced after the GoL’s Statement of Intent to the London Conference in February 2016, which articulated investment in municipalities — at an estimated 1 billion dollars — as one of its central recommendations.[12]

Defining and Ranking

This decentralized approach required the development of a set of tools to select areas of intervention. Integral to this process, U.N. agencies and partner organizations have made use of simplifying schema to organize, categorize, and prioritize areas of intervention. These maps, both literal and figurative, guide the selection of municipalities for designated projects. The first, and most prominent, of these is the list of “Most Vulnerable Localities,”[13] referred to as a “bible of sorts” for the selection of project sites for international NGOs in Lebanon.[14] In its latest iteration, this ranking is based primarily on material factors, notably the Multi-Deprivation Index (MDI) that focuses on the resources available within a given locality, and the ratio of Lebanese citizens to refugees. Underpinning the use of these criteria is an assumption that competition over services and resources is a driver of tensions between communities.

In what I characterize as the second phase, tensions were included indirectly within the selection process, which was primarily underpinned by a locality’s position within a materially-based ranking. Under this logic, the selection criteria are less open to interpretation, or even manipulation, by actors. However, as evidence of the effect of decreased material competition on tensions remained unclear, at best, interest in identifying more clearly the “sources” of tension grew. As of January 2016, this interest has increasingly been formalized through a process of consultations aimed at defining sources of tension and ranking levels of tension across localities, through the assessments of primarily international, and to a lesser extent local, organizations working within different localities across the country. This exercise, whose first phase is now completed across the country, aims to “add an extra layer on the vulnerability map, a more qualitative version” that “primarily helps to better target funds as well as [to foster] joint analysis and ensuring early warning.”[15] This shift has the potential to herald a third phase, where perceptions of host-refugee tension become more directly drivers of selection, and where local leaders and municipalities may have a greater range of maneuver in determining the “ranking” of their respective localities.

As practitioners, primarily aid providers, assess what they believe to be the primary “causes” of  “host-refugee” tension, they assign a number (or an X)[16] to each cause they perceive as relevant in a given locality. While the formalization of this practice is very recent, it is clear that, in practice, levels of tension (or the perception thereof) were often taken into account in the selection of sites for municipal support projects. However, the transformation of this practice from an informal, dynamic, and context-driven assessment of locally-based teams (which of course, are not without their own challenges) to a formalized, relatively fixed, and categorical map makes it much more potent. This becomes particularly true as the potential for material benefit to a municipality or locality becomes increasingly tied to its “measure” along this criterion.

Just Another Layer?

To be clear: driving this focus on host-refugee tension is a realization that solely relying on material realities to understand the emergence of conflict between communities is insufficient, and perhaps misleading. This, in many ways, is supported by evidence (e.g., UNDP Conflict Reports and Mercy Corps 2014). Therefore, central to the aims of this process is a desire “to understand more clearly the relationship between the ‘inputs’ — tensions — and ‘outputs’ — violent incidents.”[17]  Despite its otherwise laudable aim, however, it, like many other schema, “[i]llustrates the dangers of dismembering an exceptionally complex and poorly understood set of relations and processes in order to isolate a single element of instrumental value.”[18]

I focus here on the first of two dangers: the emphasis placed on addressing causes of “host-refugee” tension is premised on, and in turn reinforces, a false notion that hosts and refugees are ‘naturally’ distinguishable and discrete communities. In the Lebanese context, this notion needs to be complicated in at least three ways: (1) important sources of tension divide hosts among themselves, and therefore they do not necessarily represent a coherent nor uniform community; (2) Syrian refugees often have pre-existing ties within the communities they settled in; (3) hosts can also be refugees themselves, and therefore can face some of the same (or at times, deeper) challenges in terms of state protection.

As one mayor in a small locality in Central Bekaa expressed to me: “We don’t have social cohesion among ourselves, how do you expect us to have it with them [“refugees”]?”[19] While his statement can be superficially read as simply a rejection of projects that, he believes, aim to “integrate” refugees within his community, it more significantly points to the real danger of reifying tension among “hosts” and “refugees” and dislocating it from broader dynamics within these communities. While, in Lebanon, this is often read as a need to be sensitive to inter-sectarian dynamics, this argument goes beyond a desire to add yet another “layer” of group differentiation. Rather, it is a call to move beyond an understanding of tension that relies upon fixed and discrete notions of community.

This point is best illustrated through an example. I was privy to this incident during one of my visits to a mid-sized locality in Western Bekaa in March 2016. Seated in the mayor’s office, in the follow-up to our third meeting, he was visited by a number of residents who came to get his advice, or more precisely ask for his intervention, on a wide array of issues. One of these meetings was with three men from the same family, who sought the mayor’s advice on a familial dispute over land. The family owns a shared plot of land that has not been separated. The younger man, B, was looking to get married and needed to start building a home. For this, he required his share of the land to be separated. His family member, K, who was not present at the meeting, was refusing to agree to the separation, despite attempts by multiple mediators from the community and family to convince him. His family was there, then, to tell the mayor that they aimed to “punish” him by evicting the Syrian families who lived on the plot of land, and from whom K collects rent.

To my knowledge, in this particular case, the eviction did not happen. However, had it occurred, it would have likely been interpreted as yet another manifestation of “host-refugee” tension. A grounded understanding of this incident demonstrates that this reading would be, at best, limited, and perhaps even misleading. What is clear is that the relationship of the “host” — who is also a landlord in this case — to the “refugee(s)” is mediated by a wider set of social relations and dynamics. Notable among them: the intra-familial dispute and inability to find resolution within the family, the profound link between access to land and independence for B (and many other young men), the extreme power imbalance between tenants and landlords, and finally the perception of the mayor as a mediator of last resort. Importantly, of course, refugees, due to their legal status and limited access to justice, are much more vulnerable to the arbitrary decisions of landlord, employers, as well as municipalities, and makes them prone to being used in the settling of scores that have little to do with their presence. Therefore, it is critical that tensions, and even overt conflict, between Lebanese hosts and Syrian refugees not be employed as a ‘master frame’ that obscures other pre-existing and emergent sources of tension within host communities themselves.

As Estella Carpi carefully demonstrates in her contribution to this series, there is also a danger in seeing host and refugee communities as inevitably separate. Based on her fieldwork in the Lebanese governorate of Akkar, Carpi argues that “[s]ome Syrian nationals, in a sense, were also Akkaris, insofar as Akkaris themselves strived to get access to Syrian welfare, crossing the border to reach the nearby province of Homs rather than driving all the way down to Lebanon’s Tripoli to benefit from a scant welfare system.”[20]  In this context, hosts and refugees, historically, may form one community, and understanding their grievances as separable and distinct may encourage the emergence of fault lines that were not preexistent.

Even in contexts where ties between Syrians and Lebanese are not as deep as in Akkar (where intermarriage and family connections are common), Syrians are rarely entering contexts where they have no ties at all. For example, in a Christian village in Northern Lebanon where I conducted ethnographic interviews, the approximately 1,200 Syrians who had settled there generally had at least one male family member who previously worked in the village as a  seasonal laborer. Therefore, the relationship to the community was primarily one of mutual familiarity and benefit. As was told to me, despite its high rank on the “Most Vulnerable Localities” map, by a bureaucrat in the North governorate office, “there are no refugees [in that village].”[21]

Finally, hosts may themselves be refugees. This is particularly evident in the case of Palestinian refugee communities that are hosting both Palestinian and Syrian refugees from Syria, as well as the reality of refugees in urban contexts, particularly in Beirut and its suburbs, who share their neighborhoods with refugees and vulnerable migrants from a wide array of countries such as Iraq, Sudan, the Philippines, and Ethiopia. In these contexts, as Fiddian-Qasmiyeh argues, “[t]he overlapping nature of displacement leads to a blurring of the categories of ‘displaced person’ and ‘host.’” [22]As she demonstrates, in the Badawi camp in Northern Lebanon, grievances have emerged due to the disparity between the dwindling levels of assistance available to Palestinians (including Palestinians from Syria) through UNRWA and the “support from an expanding range of international and national organizations.”[23] Moreover, the increasing focus on working with municipalities, as Lebanese institutions, has the potential to further marginalize those hosts who are already neglected by and deliberately excluded from their protection.

In addition to its potential to reify divisions between hosts and refugees, the formalization of perceptions of tension and their elevation to an additional criteria of selection runs the risk of incentivizing the very behavior it aims to alleviate. This is particularly true in a context where (1) funding is limited, falling well below the stated needs, (2) where the process of selection of communities is opaque, and (3) that predictability and expectation of future funding is unclear. Under these conditions, competition among local communities for funds is high.

It is critical to understand that this is all occurring within a context of limited funds. As of the latest assessment, “119 out of the 251 localities identified as most vulnerable are not receiving tangible support to alleviate resource pressure or mitigate tensions.”[24]  In this context, municipalities and municipal unions are rightfully aware that they are competing over scarce resources.

This has already led some local leaders to speculate on the criteria being used to determine which municipalities are provided support. One mayor recalled how the locality “only has 600 or so Syrians, but [he] alway[s] say[s] 1200 because under 1000, and you can’t get help.” [25] Another mayor, from the governorate of Baalbek-Hermel, believed that “perhaps [his municipality was not provided support] because [he doesn’t] bark loud enough.”[26]


As stated above, the link between aid provided to the municipality and the level of tension within it is not entirely new in Lebanon. In many cases, support to a municipality was given to — quite literally — buy time and cooperation from local leaders. Importantly, however, this remained an ad hoc and often informal decision. In contrast, isolating host-refugee tension so explicitly as a criteria of selection, and potentially “mainstreaming”[27] it, makes it much more likely that local leaders see an incentive to make their grievances well known, perhaps even by intensifying restrictions on refugees or otherwise increasing the level of tension within their localities, that is to say to ‘bark louder.’ As such, by making explicit the relationship between the perceived level of tension within a locality and their access to international aid, this process runs the great danger of incentivizing the very actions it aims to alleviate.

It is important to note that this argument should not be read as discouraging engagement with municipalities as partners in the response. While many municipalities have instituted discriminatory policies against refugees, and contributed to the vulnerability of refugees,[28] many have also demonstrated a greater level of pragmatism and effectiveness than national authorities. For example, even as the central government further restricted refugees’ right to work, municipal authorities across Lebanon continued to facilitate Cash-for-work programs that provided stipends to refugees in exchange for work within the locality, even as it became unclear whether these conformed to central government policy regarding the right to work for refugees. In some cases, municipalities signed waivers taking responsibility for this practice when international humanitarian actors were hesitant to take on the responsibility for ignoring central government directives.[29] However, in the process of providing greater support to municipalities, it is critical that greater attention be given to the ways in which the selection of municipalities is interpreted by local actors, and its potential effect on the dynamics of conflict.

Note: This work was carried out with the aid of a grant from the International Development Research Centre, Ottawa, Canada. Information on the Centre is available on the web at

[1] For example, see: Marwa Boustani, Estella Carpi, Hayat Gebara, and Yara Mourad, “Responding to the Syrian crisis in Lebanon:  Collaboration between aid agencies and local governance structures.” Working Paper. International Institute for Environment and Development. September 2016.

[2] United Nations Inter-Agency Coordination, “Support to Lebanese Public Institutions: Under the Lebanon Crisis Response Plan 2014,” 3.

[3] James C. Scott, Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed (New Haven: Yale University Press, 1998).

[4] Interview with Andras Beszterczey, Social Cohesion and Governance Program Manager, Mercy Corps, Beirut, February 2016.

[5] Ibid.

[6] In the final draft of the second LCRP, all references to the term “social cohesion” were changed to “social stability,” following a review by the GoL. This reflected a general fear on the part of the government that the term “cohesion” implied a program of local integration. As a result, the “Social Cohesion” sector was also changed to the “Social Stability” sector” (which now includes Livelihoods as well). More broadly, among interveners in the response, the term social cohesion became increasingly taboo, although most recognized that the name change was largely cosmetic.

[7] Interview with Ilina Slavova, International Alert, Beirut, December 2015.

[8] United Nations Inter-Agency Coordination, “Support to Lebanese Public Institutions: Under the Lebanon Crisis Response Plan 2015,” 2.

[9] United Nations Office for the Coordination of Humanitarian Affairs (OCHA) Financial Tracking Service, “Lebanon: Funding Received 2016,” , accessed November 15, 2016,; and OCHA Financial Tracking Service, “Lebanon: Funding Received 2015,” accessed November 15, 2016,

[10] United Nations Inter-Agency Coordination, “Support to Lebanese Public Institutions: Under the Lebanon Crisis Response Plan 2014,” 3.

[11] Ibid., 16.

[12]Government of Lebanon (GoL), “London Conference  – Lebanon Statement of Intent,” February 4, 2016, accessed November 22, 2016,

[13] United Nations Inter-Agency Coordination, “Most Vulnerable Localities in Lebanon,” March 2015.

[14] Andras Beszterczey, Social Cohesion and Governance Programme Manager, Mercy Corps, Beirut, December 2015.

[15] Interview with Bastien Revel, UNDP Social Stability & Livelihoods Sectors Coordinator, October 2016. This should be placed also within a broader context of what Mark Duffield calls “the new humanitarianism,” which moves away from a purported neutrality of aid and intervention towards a linking of security concerns with development. This approach was underpinned by a recognition of the failings of “duty-based ethics, where actions are regarded as being right in themselves (deontological ethics), [and a move towards] goal-based ethics, where good must be seen to come out of actions (teleological ethics).” See Mark Duffield, Global Governance and the New Wars: The Merging of Development and Security (New York: Zed Books, 2001) 92. Thanks to Dr. Estella Carpi for this recommendation.

[16] Most of the assessments rely on a scale (1-3) for each perceived “cause,” while in one office a dichotomous approach (yes/no designated by an X) was used to designate the existence of a cause. The set of options available are: political/religious difference; competition for jobs; weak municipality/local authority; pressure on resources; negative perception/cultural differences & lack of interaction; concentration of refugees and gatherings/ITS; sense of insecurity; negative perceptions of international assistance. The author was present for many of the meetings where these assessments were discussed and compiled, in the Bekaa, South, and Beirut/Mount Lebanon governorates.

[17] Interview with Bastien Revel, UNDP Social Stability & Livelihoods Sectors Coordinator, October 2016.

[18]  Scott, Seeing Like a State, 21.

[19] Interview with mayor, municipality in Central Bekaa, October 2016.

[20] Estella Carpi, 27 October 2016, “Against Ontologies of Hospitality: About Syrian Refugeehood in Northern Lebanon,” Middle East Institute, accessed November 15, 2016,

[21] Confidential Interview, senior staff person, North Governorate Office, Tripoli, August 2016.

[22] Elena Fiddian-Qasmiyeh, “Refugees Hosting Refugees,” Forced Migration Review 53 (2016): 26.

[23] Fiddian-Qasmiyeh, “Refugees,” 27.

[24] United Nations Inter-Agency Coordination, August 2016, “SOCIAL STABILITY Quarter 3 Dashboard.”

[25] Informal conversation, mayor from Bekaa governorate, Zahle, September 2016.

[26] Confidential interview with mayor, municipality in Baalbek-Hermel governorate, April 2016.

[27] Interview with Bastien Revel, UNDP Social Stability & Livelihoods Sectors Coordinator, October 2016. The areas of tension analysis has been presented at a wide array of sector coordination meetings, such as those focused on: Protection, Sexual and Gender Based Violence, and in some cases, WASH (Water, Sanitation, and Hygiene). It has also been recently presented at an Inter-Agency meeting.

[28] See Maya El Helou, “Refugees under curfew: The war of Lebanese municipalities against the poor,” The Legal Agenda, 22 December 2014; Elham Barjas, “Restricting Refugees: Measuring Municipal Power in Lebanon,” The Legal Agenda, 30 September 2016; Elham Barjas, “Municipal Regulation of Syrian Refugees in Lebanon: The Case of Kfar-Rimman,” The Legal Agenda, November 21, 2016.

[29] Confidential interview with NGO staff, Beirut, February 2016.


The Middle East Institute (MEI) is an independent, non-partisan, non-for-profit, educational organization. It does not engage in advocacy and its scholars’ opinions are their own. MEI welcomes financial donations, but retains sole editorial control over its work and its publications reflect only the authors’ views. For a listing of MEI donors, please click here.