Protecting Humanitarian Action: Key Challenges and Lessons from the Field

Brief Author: 
Julia Brooks
Date of Publication: 
October 25, 2016


Executive Summary

This paper presents an overview of key challenges and dilemmas faced in the protection of humanitarian action and aims to provide an initial overview of the legal, policy, and operational trends and issues identified by ATHA through its ongoing research and discussions with practitioners. On the basis of this analysis, this paper addresses three key areas. The first section highlights knowledge and questions regarding security incidents, trends, and causes of violence, including around causes and motives for attacks, and tensions between individual and collective responses. The next section then explores the role of the humanitarian principles, and the perceptions of humanitarian actors, in affecting their security in the field. Building on this, the final section examines the protection of humanitarian action under international law, and the impunity gap resulting from ineffective implementation or enforcement of the law. The purpose of this paper is to provide key background information and to serve as a starting point for dialogue and reflection on the protection of humanitarian action from attack. It is based on research and consultations with experts and humanitarian practitioners in the field.


I. Introduction

The protection of humanitarian action from attack is an increasingly critical challenge for the humanitarian sector. While available data indicate that global incidents of violence against aid workers reached a peak in 2013,[1] concerns over threats and violence against humanitarian practitioners have continued to grow in recent years as agencies internalize this new operational reality. In the process, organizations in the field are confronted with mounting tensions as they seek to maintain access to populations in need while simultaneously ensuring the safety and security of their staff in complex environments.

This paper presents an overview of key challenges and dilemmas faced in the protection of humanitarian action. The paper aims to provide an initial overview of the legal, policy, and operational trends and issues identified by ATHA through its ongoing research and discussions with practitioners. On the basis of this analysis, this paper will address three key areas. The first section highlights knowledge and questions regarding security incidents, trends, and causes of violence, including around causes and motives for attacks, and tensions between individual and collective responses. The next section then explores the role of the humanitarian principles, and the perceptions of humanitarian actors, in affecting their security in the field. Building on this, the final section examines the protection of humanitarian action under international law, and the impunity gap resulting from ineffective implementation or enforcement of the law. It also considers advocacy efforts calling for enhanced protection of humanitarian action with a view to the perceived efficacy and risks of speaking out. The paper ends by outlining key questions and challenges for the future of research and humanitarian action in light of these trends and circumstances.


II. Understanding incidents, trends and causes of violence against aid

The first challenge arises in defining the problem, including in efforts to document incidents and quantify trends of violence against humanitarian action, to understand its causes and motives, and to manage security individually and collectively in the field.

Quantification of incidents and trends

Expanding efforts over the last two decades to quantify and track incidents of threats and violence against aid workers have contributed to a much fuller understanding of security incidents and trends of violence affecting humanitarian aid workers and programs. Nonetheless, a number of challenges remain with regard to data collection, sharing, analysis and governance, which impact upon our understanding of incidents and trends as well as what would be the most effective responses.

Following some early attempts by researchers to quantify violence against aid workers in the late 1990s using mortality figures,[2] several databases have emerged to record and share information on security incidents. The most publically accessible data is provided by the Aid Worker Security Database (AWSD), a project of Humanitarian Outcomes. Launched in 2005, the AWSD “records major incidents of violence against aid workers [killings, kidnappings, and attacks that result in serious injury], with incident reports from 1997 through the present.”[3] The Security in Numbers Database (SiND), launched in 2008 by Insecurity Insight and collaborating humanitarian agencies, provides another source of statistics on security incidents affecting aid agencies and delivery of aid.[4] SiND is not publically available, but is accessible to partner organizations. It records information on incidents ranging from “contexts (e.g. crime, active conflict, denial of access), to the effects on personnel (e.g. threats, injuries, deaths, kidnapping), impact on agency operations (e.g. loss or theft of supplies, damage to property),” as well as “responses to security incidents by agencies (e.g. evacuations, withdrawal) and relevant authorities (e.g. arrest, prosecution).”[5] Both databases have served as the basis for a number of recent studies on trends in violence against aid workers[6], including efforts at data visualization and mapping.[7]

Humanitarian organizations have also increasingly developed their own internal systems for incident reporting and tracking, or contribute to inter-agency databases. These include:

  • For United Nations personnel, the UN Department of Safety and Security (UNDSS) maintains a Security Incident Reporting System and has published annual safety and security reports since 2000.
  • For INGO personnel, many organizations maintain their own internal databases, and some (e.g. ICRC) publish summary reports on a periodic basis.
  • For healthcare workers, databases maintained by the ICRC, the WHO, and others.

Overall, recent data paint a similar picture: that recorded violence against humanitarian aid workers has increased nearly four-fold over the last decade, reaching a peak in 2013, and that the rise in the total number of attacks is not universal, but rather is driven by a few highly insecure contexts – led over the last two decades by Afghanistan, followed by Sudan, Somalia, South Sudan, Syria, Pakistan, Democratic Republic of the Congo, Iraq, Sri Lanka and Kenya.[8] There has also been increased attention in recent years to attacks against healthcare workers and facilities, driven in part by the systematic targeting of healthcare in Syria, as well as attacks on healthcare in other contexts such as the Gaza Strip, Iraq, Pakistan, Libya, Ukraine, Central African Republic and Yemen.[9]

While these attacks are more fully documented today, a number of challenges remain for data collection, analysis, sharing and governance, especially with regard to accuracy, reliability, comparability, and usefulness for informing humanitarian policy, programming and action in the field.[10] There are four key challenges to consider. First is the lack of clear or consistent definitions across data sources of who is considered an ‘aid worker’, and what constitutes a ‘security incident’.[11] This has hindered the comparability of statistics, especially between datasets focusing on humanitarian action or healthcare in conflict. Second, while estimates are improving[12], the continued difficulty of determining the total population of aid workers (the ‘denominator’) raises questions about trends in the actual rate of attack per aid worker in the field. In other words, it remains difficult to empirically determine whether there is an absolute or relative increase in attacks against humanitarian aid workers, taking into account the proliferation of aid actors and organizations in the field. Third is the question of gaps or bias in reporting (i.e. underreporting of incidents biasing the trend downward, or increased reporting over time biasing the trend upward), considering that many incidents continue to go unreported or uninvestigated.[13] This challenge is particularly acute for incidents affecting national staff, as practitioners note that it is often difficult to determine whether national staff are attacked in their professional capacity as humanitarians, or in their personal capacity as civilians. Fourth is our ability to disaggregate the data to learn more about trends within the diverse community of humanitarian practitioners, based on the organizational affiliation (e.g. UN, Red Cross/Red Crescent, INGO, NGO), nationality, gender or other identities of staff, or the areas and profiles of their work.[14] Questions also remain about the extent to which attacks against humanitarians are correlated with attacks against civilians more broadly.

While gaps therefore remain in our quantitative understanding of security incidents and trends affecting humanitarian action, efforts in this area have provided a much fuller picture. This analysis, supplemented by qualitative evidence and narratives about the impact of insecurity on humanitarian assistance and protection, has also helped raise international awareness and concern over the violence against humanitarian action, though practitioners express frustration that this growing concern has yet to translate into significant higher-level policy changes.

Individual and collective security management

Further challenges inherent in the response to violence against humanitarian action arise due to conflicting incentives for individual versus collective management of insecurity. While insecurity is generally collective, that is, affecting multiple agencies that operate in a given area, security is typically handled at the individual program or agency level, especially since security information is often considered too sensitive to share. Moreover, the justification of protecting individuals and organizations affected by attacks is frequently used to silence communication or advocacy at large in the aftermath of attacks.

On the one hand, the nature of insecurity is generally collective, meaning that threats and attacks against one humanitarian agency are likely to impact upon the security of other agencies in the same operating environment. This is applicable to actors expressing hostility or indifference toward humanitarian actors – including armed groups, government officials, media or the general public, which often have difficulty distinguishing between different humanitarian organizations.[15] It is likewise applicable to humanitarian organizations, whose behaviors or security approaches are likely to impact upon the perceptions and acceptance of other organizations. In Iraq, for example, the fortification of compounds or use of armed guards – steps taken by a number of humanitarian organizations in response to security threats and attacks – contributed to the perception of humanitarians as allied with intervening militaries, and in the eyes of many also to the perception of humanitarians as legitimate targets of attack.

As such, there is a high degree of “security interdependency”[16] among agencies operating in the same context. On a positive note, recognition of this interdependency is reflected in the growth of fora for inter-agency collaboration, information sharing and security training, wherein humanitarian organizations seek to reduce the individual costs of security management through cooperation. On the other hand, however, the sensitive nature of security still has many organizations responding on an individual, agency-centered basis. As Stoddard et al. write, “there is a tendency to approach security issues with insularity and a reluctance to share information, which further complicates the security relationship.[17] There are a number of rationales for this, including differing organizational mandates and cultures, scarcity of resources required for coordination, the desire for autonomy/independence, and perhaps foremost, the high level of trust required to share sensitive security information.[18] As Schneiker writes, “even though humanitarians might acknowledge the need to share security-relevant information, they only do so with those whom they trust.”[19] Practitioners note that trust is often higher among international staff than between international and national or local organizations, with reluctance to share information on security incidents compounded by concerns over the loss of funding.

Furthermore, practitioners also note a tendency within some organizations to see protection as the responsibility of growing security departments and managers, rather than something which is mainstreamed throughout operational planning, programming, legal engagement and advocacy. This “siloing” of security risk management then often leads to the perception of security measures as an inhibitor, rather than an enabler, of programming, and fuels a disconnect between the management of security on the ground, and operational, legal or advocacy and policy efforts seeking to improve the protection of humanitarian action at a higher level.

While collective action therefore brings a number of advantages for security management and protection, it can also entail significant challenges and liabilities, for instance if sensitive information is leaked to outside actors. These fears, along with resource and capacity limitations, have tempered participation in these collective efforts, whether in terms of sharing information on security incidents, participating in collective advocacy or policy efforts, or simply connecting across departmental specialties. The dangers of insularity, however, are also clear, especially when information on attacks is not shared with other agencies in the same operating area. As one practitioner noted, “I know of at least two occasions where staff were attacked after other agencies had kept silent about attacks they had previously suffered. So this silence is dangerous.”[20]

Understanding causes and motives

As quantitative efforts have provided an improved understanding of where, when and how security incidents occur, we still largely lack a comprehensive and evidence-based understanding of why such attacks occur in other words, the causes and motives behind violence against humanitarian aid workers. This is important, as Larissa Fast writes, since “The obvious danger is that we develop policies and agendas that wrongly identify the problems and causes, and thus the appropriate and corresponding responses.”[21] Recent research and field engagement is making the picture much clearer, though there is still a significant debate over the actual causes and motives of violence against aid workers, of which there may be many.

One such debate is over the degree to which violence against aid workers is primarily driven by factors external to humanitarians, or factors internal to the humanitarian sector and humanitarian action in the field. External factors often presumed to contribute to the problem of violence against humanitarians include: the proliferation of non-state armed groups and the prevalence of non-international armed conflicts, especially in contexts of weak or failed states; general erosion of respect for IHL and the recurrence of violations for which the perpetrators are infrequently held accountable; the spread of anti-Western ideologies; the intentional targeting of civilians; the blurring of lines between civilian and military actors in conflict settings; or economic or criminal motivations.[22]

Factors internal to the humanitarian sector include: the growth of humanitarian action in insecure settings overall; staff training, behavior, and adherence to the humanitarian principles in the field. Additional factors resulting from both external and internal pressures include: ‘securitization’ of aid; the ‘politicization’ or ‘militarization’ of aid and the corresponding loss of neutrality, impartiality or independence, including amidst the ‘Global War on Terror’ and the blurring of the lines between military and civilian actors; and even measures designed to better protect aid workers in the face of violence, such as armed escorts or the ‘bunkerization’ of humanitarian compounds.[23]

Furthermore, whether the primary threats to aid workers emanate from government forces, non-state armed actors, or others, varies from context to context. Whether attacks result from deliberate targeting, indiscriminate attacks and collateral damage, accidental or mistaken targeting, or criminal or opportunistic violence, also varies by context. While this paper focuses primarily on deliberate violence against aid workers, these other cases imply different motives, and accordingly necessitate different responses.  A common thread among discussions of both external and internal causes of deliberate violence against aid workers is the role of the humanitarian principles, and the extent to which erosion of principled humanitarian action has contributed to growing insecurity in the field. Views on this also differ greatly depending on one’s view from headquarters or the field level. The next section will explore this question in greater detail.


III. The role of the humanitarian principles

Adherence to the principles of humanity, neutrality, impartiality and independence also has a role in protecting humanitarian actors, though how this operates is a matter of ongoing debate. On the one hand, adherence to the principles is seen as a necessary element for the protection of humanitarian action, and part of creating an environment where humanitarian actors are accepted in the field. On the other hand, especially in insecure settings, frontline negotiations may require humanitarian actors to make difficult compromises around the principles in order to obtain or maintain access and presence. Navigating this grey zone in which the principles are negotiated, and where the role of the humanitarian principles is to maintain some semblance of protection and integrity in some of the most difficult environments, remains a great challenge, with implications for the perception and protection of humanitarian actors.

External perceptions

It is a common refrain within the humanitarian sector that recent years have seen an erosion of respect for the norms of humanitarian action and IHL by armed actors. This view is driven not just by attacks on aid workers, but more broadly by the number of highly visible international law violations and other impediments to humanitarian action and civilian protection in recent conflicts ranging from Syria and Iraq to South Sudan. Given the general prevalence of non-international armed conflicts today, much debate has centered on the role of non-state armed groups, and how to promote their compliance with international norms, though many of the higher profile attacks have been attributed to state actors.

On the one hand, many members of armed groups – such as the Taliban in Afghanistan and Pakistan, or al-Shabaab in Somalia – have expressed distrust or hostility toward humanitarian organizations, which they perceive to be collaborators of Western militaries, spies, or profiteers.[24] While some practitioners have blamed this hostility on a misunderstanding of the roles and identity of humanitarian actors, practitioners also note that many attacks are carried out by actors who understand well the identities and missions of humanitarians. As Gareth Price Jones writes, it is fairly clear to many why combatants attack aid workers:

The incentives to attack humanitarians remain much the same – denying assistance to populations that may support opposition groups, removing witnesses to atrocity, allowing greater military freedom of movement and blocking or discrediting independent narratives around levels of humanitarian suffering.[25]

Others have noted that attacks against aid workers do not necessarily reflect a misunderstanding of their mission, since even principled humanitarian action comes at a price. In attempting to provide aid in a neutral and impartial manner, for instance, humanitarian actors may face opposition from groups who see them as providing aid to their enemies. In still other cases, the motives for attacks against humanitarian actors are more criminal and financial than strategic.

What is less well-understood, then, are the potential disincentives for attacking humanitarians. Indeed, many members of other armed groups have expressed more positive, or at least neutral, attitudes towards humanitarians, which can be leveraged in the field. In recent consultations with armed non-state actors, Geneva Call found that many “express overwhelmingly positive attitudes toward IHL, including the rules about humanitarian access.”[26] They found such groups to be “broadly familiar with the core humanitarian principles” and to have a “strong expectation that humanitarians should keep to their principles.”[27]  Nonetheless, it bears noting that these results are based on consultations with groups already engaging with Geneva Call – thereby resulting in selection bias towards those armed groups more inclined towards such engagement – and that they may reflect the views of commanders who express willingness to comply, but that these views may not necessarily filter down to the rank and file of armed groups. When pressed on the issue of violence against aid workers, the interviewees offered explanations of violence against humanitarians which ranged from “blam[ing] other parties to the conflict” and “claim[ing] that they have made mistakes in identifying targets” to “frustration […] directed towards aid workers because they were representatives of the international community.”[28] According to Geneva Call, many members of armed groups felt that “aid workers could lose [their] protection if they collaborate with enemy forces.”[29] This may refer to the situations – such as recently in Iraq and Afghanistan – where the co-location or close coordination of humanitarian actors with invading international military forces provoked hostility among armed actors opposing those forces, and by association, their humanitarian allies.

Thus, while some groups may express categorical opposition to humanitarian actors, the views of many others can be influenced by humanitarian actors’ observed behavior and engagement in the field. Moreover, while some non-state armed actors may actively oppose international norms, many others have incentives to comply, or to appear to comply with international norms, which can grant them legitimacy or other reputational benefits.[30] As one practitioner put it,

In my experience, I’ve felt very much more protected by the ‘hassle factor’ – by paperwork. It is within the power of armed actors to kill you, rob you, or whatever, but it’s a lot of hassle for them to do so. This is where international protection becomes useful as a deterrence. It is not because the perpetrators are likely end up at the [International Criminal Court], but because attacking us will make it more difficult for them to get other things they want.[31]

It is therefore important to examine in greater depth the factors that limit or impede compliance among armed groups – as some may be inclined toward compliance, but lack the training, expertise, resources, command structure or organizational cohesion to do so successfully – as well as the factors that affect perceptions of humanitarian actors, which are considered in the following section.

Internal behavior

While some armed groups may exhibit hostilities toward humanitarian actors for ideological or strategic reasons, as noted in the previous section, the behavior of humanitarians themselves is also critically important to their acceptance in the field. Indeed, a number of humanitarian practitioners point to poor understanding or implementation of the humanitarian principles among field staff – highlighting a tension between headquarters and field staff over understanding and implementation of the principles in difficult settings. Some practitioners, especially at headquarters, have expressed a concern that field staff lack a detailed awareness or understanding of the principles, or how to implement them in concrete decisions in the field. In many cases, however, field staff understand the principles well, but struggle to navigate them in environments where the principles are continuously negotiated, and where compromises need to be made. Different environments raise these issues in different ways. In some contexts, for instance, the strict adherence to the principles is necessary for maintaining presence and acceptance; in other contexts, compromise is necessary. The dilemmas are particularly stark in contexts where both situations apply.

As a result, some see a large gap between knowledge and ownership with regard to principled action, while others see a gap between policy at headquarters and practice in the field. As one practitioner noted:

In some organizations, there is a big disconnect between what they’re saying at headquarters, or in terms of advocacy and policy, and the pragmatism of [field staff] in reaching people in need on the ground, which may counterbalance the principles against other pragmatic concerns.[32]

While headquarters staff may be aware of the pragmatic issues of their application in the field, challenges may still lie in communication and awareness of the realities in the field and the nature of compromises made.

Many have pointed to the conflict in Syria, for instance, and the difficultly for aid organizations of asserting their neutrality and independence when they are only willing or able to operate in regime or opposition-controlled areas, respectively. As one practitioner put it:

It’s true that sometimes in conflict situations we are facing the choice between our security and the principles. Sometimes we stop activities because security isn’t there, or because the principles, e.g. neutrality, cannot be guaranteed. In other cases, it may be quite easy to remain neutral, but independence is more complicated. For instance in Syria, […] if we decide to go to this area and not that area, we can be seen as not fully independent. […] If we decide to cancel activities for security reasons, it raises questions of humanity, and civilians’ access to aid.[33]

It may be hard to establish a direct correlation between a lack of adherence to the humanitarian principles and violence against aid workers, as another practitioner noted:

Overall, when there is violence against us, we’re making it much harder to say that we are neutral and impartial’ when in action we are not, because of how we are seen providing aid. […] There is a certain dereliction of duty to adhere to principles, and it has made it more difficult for us to be seen as providing neutral aid.[34]

A number of sources point to factors such as the perceived loss of impartiality and neutrality, including as a result of politicization, militarization or instrumentalization of aid in recent conflicts in Iraq, Afghanistan and Syria, for instance, as exposing humanitarians to greater hostility and attack.[35]

Another factor highlighted by practitioners is the trend toward the use of unmarked vehicles, taken as a safety measure to lower the profile of aid workers in some contexts where they have been deliberately targeted, such as Afghanistan, Libya and Yemen. In a recent letter to Saudi Arabian defense officials, for instance, MSF noted the use of cars without their logos in Yemen:

Due to the situation on the ground in Yemen, there are zones where MSF judges that from a ground safety perspective it is safer to drive in an unmarked car than a marked one. MSF expects to make such decisions on its own and requires some understanding from the Coalition about this process.[36]

As with other strategies designed to protect aid workers in the face of increased attacks, this strategy may also have perverse effects on protection. As one practitioner noted:

One trend I see as dangerous over last ten to fifteen years is the reduction in use of marked vehicles, etc., especially by NGOs. We have this protected status [under international law], but we are afraid to use it, because we fear being targeted by others. […] Its’s a shocking illustration of how far expectations have shifted […] If we are guilty of not using protected symbols, or of abusing them [e.g. by using ambulances for non-medical transportation of aid workers], then that’s very problematic for our protection.[37]    

Additionally, the inter-agency nature of most operating environments means that if one agency is not acting in a principled manner, for instance, that can affect the perceptions of other agencies, and the extent to which they are accepted and protected. This is because external actors are likely to associate humanitarian actors with each other – again highlighting the collective nature of security and protection in the field.

To reverse this trend, many have reiterated the importance for humanitarian organizations of strengthening and refining their own internal knowledge and understanding of the principles. As one practitioner noted:

[it is important to talk about] ethical decision-making in high risk environments. Organizations take on humanitarian principles, and they think that they can uphold them all simultaneously, in any given situation. Ultimately, you have to make compromises, but often we don’t give enough consideration to how those compromises are made, and what the implications are.[38]

Others have also called for more frank conversations within and across organizations about their acceptable levels of risk. Since risks cannot be eliminated completely, practitioners see a need for more open and transparent dialogue about risk thresholds at the individual, organizational and sectoral levels. It is thus important for humanitarian actors to understand the principles in all their complexity, to make conscious decisions about what compromises they are willing or not willing to engage in, and the implications of those compromises. Furthermore, even when making conscious decisions, it is important to recognize that humanitarian actors in difficult settings may face limited good options in the choices they are making.


IV. Legal protection, the impunity gap, and the risks and efficacy of advocacy

A further important challenge arises from the gap between the formal protection afforded to aid workers under IHL, and the widespread impunity for attacks by armed actors against humanitarians in practice. For some practitioners, this has motivated calls to strengthen the legal protection aid workers, while others have expressed frustration that legal mechanisms are either unavailable or infeasible means of realizing protection in the field. These questions have also taken on central importance in advocacy efforts calling for enhanced protection for humanitarian aid workers, where a tension remains between the felt need for advocacy, and the perceived risks and challenges of efficacy.

Legal protection and the impunity gap

With regard to IHL, the first challenge arises from differing views on the level of protection afforded to humanitarian aid workers under the law itself. On the one hand, IHL protects humanitarians as it does all other civilians: namely, attacks against them are prohibited, and when done deliberately, constitute war crimes. On the other hand, however, IHL also affords special protection to certain specifically mentioned categories of relief personnel, namely: UN and Associated Personnel, and medical services of armed forces, civilian hospitals in wartime, and affiliates of the International Red Cross and Red Crescent Movement. The resulting patchwork of IHL obligations relating to the security of humanitarian personnel in situations of armed conflict can therefore be seen as producing a hierarchy of legal protections that privileges certain categories of aid workers above others, while leaving the majority of aid workers, primarily NGO and INGO staff, with general civilian protection.

The extent to which these different tiers of legal protection actually translate into different levels of protection that organizations experience in the field remains an open question. Many see the existing legal protection of humanitarians as appropriate – with regard to the law itself, at least, as attacks against them are prohibited in either case – and have focused their efforts on enhancing protection for civilians in general.[39] Others advocate for strengthening the special protection for humanitarian aid workers in particular, due to their essential function of providing aid in conflict settings. However desirable it may be to revise IHL though, opening these provisions international law to revision in the current political environment is likely to lead to a weakening rather than strengthening of humanitarian norms. To the extent that existing protections have proven insufficient in practice, then, many practitioners express the need to focus on stronger implementation and enforcement of the law to better protect civilians and humanitarians alike.

A second challenge arises for humanitarian actors from the difficulties of implementing the law to claim protection in the field and justice when attacks do occur. Indeed, very few incidents of violence against humanitarians are ever independently investigated or prosecuted.[40] Due to the insecurity and breakdown of the rule of law typical in many conflict zones, certain difficulties in enforcing the law are to be expected. There may be a lack of follow-up on incidents, for example, due to the difficulty of conducting investigations in remote and insecure conflict areas, or of ascertaining the identity of the perpetrators. In other cases, aid organizations may distrust local authorities, or lack the capacity to follow-up on incidents themselves, given that many attacks are followed by reduced or withdrawn operational presence. Or aid agencies may choose to keep a low profile, and follow up with authorities on a confidential basis only. The result is that many incidents go undocumented or uninvestigated, which limits the reliability of incident data as well as the potential for future accountability. As one practitioner put it, “This issue of lack of solidarity among agencies under the pretext (sometimes) of security is doing more harm than good.”[41]

In many cases, however, humanitarian organizations or practitioners seeking accountability and justice for attacks against them have made little progress. In the case of the 2006 murder of 17 of its staff members in Muttur, Sri Lanka, for instance, Action Contre la Faim (ACF) has spent a decade campaigning for accountability.[42] In another more recent high-profile case, Médecins Sans Frontières (MSF) continues to call for an independent investigation and accountability for the bombing of its hospital in Kunduz, Afghanistan by U.S. forces in October 2015. In this case, MSF has had difficulty finding a competent forum for enforcing the law; it has called for an investigation by the International Humanitarian Fact-Finding Commission (IHFFC) (which has never been used, and would require the consent of the U.S. for activation). It is similarly difficult to find a forum for prosecution; the International Criminal Court (ICC), for instance, is yet to prosecute a case of violence against aid workers, though it has monitored attacks against aid workers and peacekeepers in Sudan.[43] Prosecutions of attacks against aid workers before ad hoc tribunals or national courts are similarly hard to come by.

While it may then be difficult for humanitarian organizations to find effective mechanisms for enforcing compliance with the law, a larger question arises as to whether this should be their role at all. It is natural for the organizations affected by attacks to demand justice, but even where enforcement mechanisms are available, humanitarian organizations also have reasons to stay out of the realm of enforcement. Humanitarian practitioners (especially those working on security management) may be unfamiliar with legal mechanisms, for instance, or reluctant to engage with them due to concerns about undermining their access or acceptance in the field by targeting armed actors for prosecution. They may also decide against engaging in enforcement based on pragmatic concerns about the high costs of engagement and remote likelihood of a satisfactory outcome. As a result, few humanitarian practitioners consulted expressed an interest or willingness to turn to legal enforcement mechanisms in the aftermath of incidents, while those that have pursued accountability for attacks express frustration at the inefficacy of the process.

A number of proposals put forth to address this gap have thus focused on removing the decision from affected staff and individual organizations with operations to maintain in the field, to actors with a safer distance and greater ability to absorb the costs. ACF has proposed, for instance, the creation of a special mandate holder for the protection of humanitarian workers. In his report for the World Humanitarian Summit, UN Secretary General Ban Ki proposed the creation of a “dedicated ‘watchdog’ to systematically track, collect data and report on trends of violations, gaps in compliance, accountability and State cooperation in all conflicts.”[44] While this proposal is aimed at enforcing compliance with international humanitarian law more broadly, it could also help to ease the costs to individual agencies with regard to attacks against aid workers in particular.

With many challenges standing in the way of implementing or enforcing the law to protect aid workers – whether with regard to the structure of the law itself, the paucity of effective mechanisms for its enforcement, or the disincentives for humanitarians to engage in it – the result has been widespread impunity for violence against aid workers. As one practitioner put it, “We’ve been very bad at collectively raising the costs of violations for parties,”[45] even in cases where the parties may be more susceptible to international influence, as in the recent attacks in South Sudan for example. Gareth Price-Jones notes that the costs have clearly shifted in favor of combatants:

Ten years ago the Muttur case made clear to combatants globally that attacking humanitarians came with considerable costs – reputational, diplomatic, political and financial. Today those costs are negligible.[46]

On the other hand, as noted above, ‘raising the costs’ for armed actors is also likely to entail significant costs for humanitarians themselves in terms of access and acceptance, which may not be considered worth it in individual cases. The result in aggregate, though, in the words of another practitioner, is that “this is not seen as a red line anymore.”[47] At most, violence against aid workers generally results not in accountability for the perpetrators, but in a reduction or withdrawal of programming – which may have been the intention of armed groups or governments suspicious of aid organizations in the first place. To address this challenge, Price-Jones and others call for the removing the decision to speak out and seek accountability, from national actors and affected staff, to actors with a safer distance and greater ability to absorb the costs.

A challenge of collective action has played out with regard to the implementation and enforcement of the law, and the advocacy or legal engagement that it requires, due in part to the high perceived individual costs to humanitarian organizations, and the low perceived likelihood of achieving justice. The next section examines this challenge in greater detail.

Advocacy risks and efficacy

Advocacy around the protection of humanitarian aid workers, and decisions over when, and how, to speak out, has also been marked by a challenge of collective action. Here too, we see that efforts to raise awareness of attacks against aid workers, and to call for their enhanced protection, are also seen as entailing significant costs and risks to individual aid workers and organizations, with a low likelihood of reward, and low costs to non-participation. In this context, a variety of organizations and coalitions engaged in advocacy are struggling against the relative normalization of violence against humanitarians.

As incidents of violence have increased in recent years, so too have advocacy efforts and campaigns by humanitarian organizations. At the organizational level, humanitarian organizations have become accustomed to responding to specific instances of violence against their staff, operations or partners with public condemnation and calls, for instance, for unimpeded access and respect for international humanitarian law.[48] At the sectoral level, there are also a growing number of initiatives advocating for improved respect and protection for aid workers, especially in the field of healthcare[49], and for the well-being of practitioners more broadly.[50] The issue has also reached international policy makers to some extent, resulting inter alia in the UN Security Council’s adoption of several resolutions condemning violence against humanitarian personnel and reaffirming Parties’ obligations to protect them, including most recently in Resolution 2286 (2016) on the protection of medical facilities. Nonetheless, many practitioners remain frustrated that this advocacy is still piecemeal, disjointed, or otherwise ineffective. Most organizations struggle to move beyond individual statements in response to specific attacks, to more collective strategies that effectively raise the profile of the issue for the international community in general.

For one, there exists significant debate within the sector as to the appropriateness and efficacy of advocacy on the protection of aid workers. Until more recently, many organizations refrained from public advocacy on this issue – except in the highest profile incidents – due to the confidential nature of their dealings with conflict parties and host governments, the lack of resources to devote to advocacy (especially among smaller organizations), or the desire to keep a low profile. One view is that advocacy should focus on enhanced respect for IHL in general, as noted above. Such narratives shy away from the portrayal of aid workers themselves as ‘vulnerable’ or ‘victims’ of violence, which is are seen as detracting from core aims, namely assistance to and protection of the populations they serve. Another view, however, is that aid workers require enhanced special protection, whether in law or policy, by virtue of their critical role; namely, without secure access for aid workers, there can be no assistance to beneficiaries. This remains an ongoing discussion, including with regard to the well-being of aid workers, although the perception of growing insecurity for aid workers has gradually led more organizations to speak out specially for their protection.

Moreover, some organizations may agree with the aims of such advocacy, but weigh the relatively high potential costs of speaking out (e.g. loss of local acceptance or access) against the relatively low costs of silence, at least for the organization itself. As Gareth Price-Jones notes,

The default has shifted from immediately speaking out strongly, to considering whether to speak out, to staying silent. From the highest echelons of the UN to local NGO staff, no-one is willing to speak out until the costs of not doing so (i.e. widespread disgust at an organization for failing to protect its staff) hugely outweigh the costs of speaking.[51]

This is because organizations tend to focus on the visible risks of speaking out (e.g. security risks, risk of compromising humanitarian principles, reputational and credibility risks, loss of access or resources), while discounting the risks of not speaking out (e.g. suffering of affected people, poor policy choices, risk of compromising humanitarian principles, risk of eroding accepted practice, risk that situation doesn’t change).

As with the lack of accountability, the result has been a relative normalization of violence against humanitarians. Individual incidents may be met by expressions of “grave concern” and “strong condemnation” (at least in public statements) by aid officials, or the withdrawal of programming – which may have been the intention of armed groups or governments suspicious of aid organizations in the first place. Beyond those directly affected, however, incidents are also often met by a collective “shrug of the shoulders” – a sense that this is an inescapable feature of contemporary conflict. In the words of one practitioner, “it has become the new normal.”[52]


V. Conclusion

As threats and violence against humanitarian actors have grown in a number of conflict settings in recent years, so too have concerns over the ability of humanitarian organizations to maintain effective access and programming, while at the same time keeping their staff and partners safe. With the growing concerns over the protection of humanitarian action in insecure settings have come a plethora of efforts from across the sector, ranging from incident tracking, training and professionalized security risk management, to the creation of inter-agency fora and coordination mechanisms, as well as advocacy campaigns. Nonetheless, violence against aid has continued at a rapid pace, and humanitarian organizations struggle to protect their staff and programs in a number of insecure settings.

As this paper outlines, several key challenges have arisen for organizations in the protection of humanitarian action which call for continued dialogue and mobilization across organizations around these shared challenges and how to best address them. While many humanitarian organizations experience the lack of protection and have taken a variety of steps to adapt, one of the key overarching issues to emerge is the challenge of collective action, whether with regard to data collection and information sharing to quantify trends in violence against aid workers, cooperation in security management, implementation and enforcement of the law, or advocacy efforts. Another overarching challenge is the need for humanitarians to better understand and adapt to changing operational contexts, and to highlight and leverage the motives and incentives for armed parties to respect rather than to attack humanitarian actors. A number of questions also remain open with regard to improving legal protection and implementation, navigating principles and pragmatism in the field, increasing the efficacy of advocacy, and learning from non-conventional approaches to humanitarian action in insecure settings. These questions call for continued reflection among researchers and humanitarian practitioners, both within their own organizations and across the sector.

Violence against humanitarian actors and operations has grave consequences for humanitarian staff and organizations, as well as the vulnerable populations they serve. Such attacks represent an assault on individual lives, and the collective norms which enable humanitarian assistance and protection in emergencies. It is imperative that the international community – including humanitarian organizations, states, and other actors – comes to terms with this changing operational reality, and hold perpetrators to account for violations. This urgently requires a deeper understanding of the causes, manifestations and consequences of insecurity, and redoubled efforts to enhance the protection of humanitarian actors.


[1] “The Aid Worker Security Database (AWSD),” Humanitarian Outcomes, accessed August 10, 2016,

[2] See Mani Sheik et al., “Deaths among Humanitarian Workers,” British Medical Journal 321, no. 7254 (2000),; Dennis King, “Paying the Ultimate Price: Analysis of the Deaths of Humanitarian Aid Workers 1997 - 2001,” Text (UN OCHA, January 15, 2002), Dennis King, “The Year of Living Dangerously: Attacks on Humanitarian Aid Workers in 2003” (Washington, DC: Humanitarian Information Unit, U.S. Department of State, 2004), Cate Buchanan and Robert Muggah, No Relief: Surveying the Effects of Gun Violence on Humanitarian and Development Personnel (Centre for Humanitarian Dialogue, 2005),; Elizabeth A. Rowley, Byron L. Crape, and Gilbert M. Burnham, “Violence-Related Mortality and Morbidity of Humanitarian Workers,” American Journal of Disaster Medicine 3, no. 1 (2008): 39–45.

[3] “The Aid Worker Security Database (AWSD).”

[4] “Aid in Danger - Security in Numbers,” Insecurity Insight, accessed September 1, 2016,

[5] “Security in Numbers Database (SiND): Monitoring Actions That Interfere with Aid Delivery” (Insecurity Insight, 2016),

[6] For studies relying on the AWSD data, see e.g. the series of annual Aid Worker Security Reports published by Humanitarian Outcomes since 2011, the most recent being: “Aid Worker Security Report 2015: Figures at a Glance” (Humanitarian Outcomes), accessed January 11, 2016, Abby Stoddard, Adele Harmer, and Kathleen Ryou, “Aid Worker Security Report 2014 - Unsafe Passage: Road Attacks and Their Impact on Humanitarian Operations” (Humanitarian Outcomes, August 2014), See also Abby Stoddard, Adele Harmer, and Victoria DiDomenico, “Providing Aid in Insecure Environments: 2009 Update,” HPG Policy Brief (Humanitarian Policy Group, Overseas Development Institute, April 2009), Abby Stoddard, Adele Harmer, and Katherine Haver, “Providing Aid in Insecure Environments: Trends in Policy and Operations” (Humanitarian Policy Group, Overseas Development Institute, 2006), For studies relying on the SiND data, see e.g. Christina Wille and Larissa Fast, “Aid, Gender and Security: The Gendered Nature of Security Events Affecting Aid Workers and Aid Delivery” (Insecurity Insight, 2011),

[7] See “Mapped: 15 Years of Aid under Fire,” Irin News, accessed June 7, 2016,

[8] “Highest Incident Contexts (1997 - 2015),” The Aid Worker Security Database, 2016,

[9] “Report on Attacks on Health Care in Emergencies.” World Health Organization, 2016.

[10] See Fabrice Weissman, “Violence Against Aid Workers: The Meaning of Measuring,” in Saving Lives and Staying Alive: Humanitarian Security in the Age of Risk Management, ed. Michaël Neuman and Fabrice Weissman, CRASH – Fondation Médecins Sans Frontières (Hurst, 2016),

[11] See Larissa A. Fast, “Mind the Gap: Documenting and Explaining Violence against Aid Workers,” European Journal of International Relations 16, no. 3 (September 1, 2010): 365–89.

[12] See Abby Stoddard, Adele Harmer, and Katherine Haver, “Data Are Not Dangerous: A Response to Recent MSF CRASH Critiques,” ODI HPN, May 11, 2016,

[13] See e.g. Marianne Abbott, “Dangerous Intervention: An Analysis of Humanitarian Fatalities in Assistance Contexts” (PhD dissertation, Ohio State University, 2006),; Dennis King, “The Year of Living Dangerously: Attacks on Humanitarian Aid Workers in 2003”; Sheik et al., “Deaths among Humanitarian Workers”; Leonard S. Rubenstein and Melanie D. Bittle, “Responsibility for Protection of Medical Workers and Facilities in Armed Conflict,” The Lancet 375, no. 9711 (2010): 329–40.

[14] See Julia Brooks, “Humanitarians Under Attack: Tensions, Disparities, and Legal Gaps in Protection” (Advanced Training Program on Humanitarian Action (ATHA), 2015),

[15] See Andrea Schneiker, Humanitarian NGOs, (In)Security and Identity: Epistemic Communities and Security Governance (Routledge, 2016), 5.

[16] See Koenraad Van Brabant, “Managing Aid Agency Security in an Evolving World: The Larger Challenge,” EISF Article Series (London: European Interagency Security Forum (EISF), November 2010), 10,

[17] Abby Stoddard, Adele Harmer, and Morgan Hughes, “Aid Worker Security Report 2012 - Host States and Their Impact on Security for Humanitarian Operations” (Humanitarian Outcomes, December 2012),

[18] Schneiker, Humanitarian NGOs, (In)Security and Identity, 6–9.

[19] Ibid., 9.

[20] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[21] Fast, “Mind the Gap,” 382.

[22] See Fast, “Mind the Gap.”

[23] See ibid.; Larisa Fast, Aid in Danger: The Perils and Promise of Humanitarianism (Philadelphia, Pennsylvania: University of Pennsylvania Press, 2014).

[24] Ashley Jackson, “Negotiating Perceptions: Al-Shabaab and Taliban Views of Aid Agencies,” Policy Brief 61 (Humanitarian Policy Group, August 2014), 2.

[25] Gareth Price-Jones, “The Danger of Keeping Silent” (working paper, 23 August 2016).

[26] Ashley Jackson, “In Their Words: Perceptions of Armed Non-State Actors on Humanitarian Action” (Geneva Call, June 6, 2016), 6, Ashley Jackson and Abdi Aynte, “Talking to the Other Side: Humanitarian Negotiations with Al-Shabaab in  Somalia,” HPG Working Paper (Humanitarian Policy Group, December 2013), 16.

[27] Ashley Jackson, “In Their Words: Perceptions of Armed Non-State Actors on Humanitarian Action,” 5.

[28] Ibid., 21.

[29] Ibid.

[30] See Rob Grace, “Preparatory Review of Literature on Humanitarian Negotiation” (Advanced Training Program on Humanitarian Action (ATHA), September 2014),

[31] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[32] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[33] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[34] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[35]  See Victoria Metcalfe, Alison Giffen, and Samir Elhawary, “UN Integration and Humanitarian Space: An Independent Study Commissioned by the UN Integration Steering Group” (Humanitarian Policy Group / Stimson Center, December 2011),

[36] “Médecins Sans Frontières Recognizes Saudi Coalition’s Help in Yemen,” Al Arabiya English, April 25, 2016,

[37] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[38] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[39] See Camilla Barker, “Aid Workers Could Secure Better Protection under the Protection of Civilians Mandate,” IntLawGrrls, August 15, 2014,

[40] See “Twenty-Second Report of the Prosecutor of the International Criminal Court to the United Nations Security Council Pursuant To UNSCR 1593 (2005)” (International Criminal Court, Office of the Prosecutor, 2015),

[41] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[42] Nita Bhalla, “Justice Elusive for Slain Aid Workers on Front Lines of Crises,” Thomson Reuters Foundation, August 19, 2016,

[43] “Sixteenth Report of the Prosecutor of The International Criminal Court to the UN Security Council Pursuant to UNSCR 1593 (2005)” (International Criminal Court, Office of the Prosecutor, 2012),

[44] United Nations Secretary-General, “Report of the Secretary General for the World Humanitarian Summit” (New York, NY, USA, February 2, 2016), para. 60,

[45] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[46] Gareth Price-Jones, “The Danger of Keeping Silent.”

[47] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

[48] See e.g. “Central African Republic: Humanitarian Coordinator Strongly Condemn Attacks against Humanitarian Organizations and Calls for Free Movement of Aid Workers” (Bangui: UN Office for the Coordination of Humanitarian Affairs, UN Resident and Humanitarian Coordinator for the Central African Republic, September 29, 2015), “WHO Condemns Attack on Benghazi Medical Center in Benghazi, Libya,” World Health Organization, accessed September 17, 2016, “CAR: MSF Condemns Violent Attack and Killing of Staff Member,” MSF USA, May 19, 2016,

[49] These include the including the Healthcare in Danger project of the International Red Cross and Red Crescent Movement, and the Safeguarding Health in Conflict coalition, as well as efforts by the World Heath Organization.

[50] See “Petition: UN, World Governments - Protect Aid Workers Now!,” accessed September 6, 2016,

[51] Gareth Price-Jones, “The Danger of Keeping Silent.”

[52] Confidential ATHA interview conducted via telephone in Fall 2016 with high-level humanitarian practitioner, name of interviewee on file.

Recent Tweets

Our Sponsor

A Program Of

All materials © 2014 Harvard University

Back to Top

Back to Top