Introduction

Large-scale humanitarian emergencies are increasingly stretching the international community’s ability to meet critical humanitarian needs. This includes contexts such as Yemen, Syria, Myanmar, Bangladesh, Iraq, Afghanistan, South Sudan, Nigeria and Somalia, as well as many others.

In many of these complex emergencies, humanitarian aid workers, medical workers and healthcare facilities are themselves targets of attack, which not only puts aid workers at risk, but can threaten the provision of humanitarian assistance when resources are either destroyed, diverted, or programs have to be scaled down to minimise risk to personnel. However, whether in complex emergencies or in response to natural disasters, militaries often play an important role in humanitarian relief efforts, sometimes by providing search and rescue and airlift capabilities or by restoring damaged infrastructure. Indeed, in most of today’s crises, humanitarian organisations operate in the same environment as a range of military and non-state armed actors.

Coordination is often easier in natural disaster settings than in conflict, as there is a time-sensitive, common goal (saving lives, particularly in the crucial first days after a disaster) (Forestier et al., 2016). However, that does not mean that there are no challenges to civil–military coordination, not least when a large number of actors respond. An example of this is the 2008 Indian Ocean earthquake and tsunami in Aceh, Indonesia, to which a reported 14 UN agencies, 16 foreign militaries and 195 foreign humanitarian organisations were involved in humanitarian assistance (Wiharta et al., 2008). However, challenges in disaster responses are different and revolve more around how to organise effective coordination than around trade-offs in engaging with armed actors in an active conflict zone (Bollettino and Anders, 2018). Humanitarian organisations have to balance potential benefits from working with militaries (e.g. access to hard to reach locations, protection for staff and assets) with potential risks (such as risks to reputation and access if they are seen to associate themselves with an armed actor, particularly if the military is also involved in the conflict).

Effective engagement between militaries and humanitarian aid agencies can be beneficial for the timely delivery of aid. This engagement can take many different forms, from mere coexistence to communication, coordination and direct cooperation. It is also often unavoidable when trying to gain access to areas controlled by military or non-state armed actors. However, such engagement also comes with risks. Sometimes these are security risks, or risks to the humanitarian principles of impartiality and neutrality. Crucially, it is difficult to assess these risks and/or potential benefits in the absence of metrics to measure civil–military engagement. Previous literature on the subject has described some of the benefits and potential risks of different types of engagement between military and humanitarian actors. Numerous case studies on engagement in specific emergency settings also exist. To date, however, quantifiable data on how civil–military engagement unfolds and which factors influence the effectiveness of coordination is lacking.

This paper aims to make a contribution to addressing the gap in the data by outlining a framework of indicators that can be used to measure effectiveness in humanitarian civil–military engagements – meaning engagement between civilian and military actors either in a humanitarian response or for humanitarian purposes. Three key questions drive this research: (1) When and under what circumstances is civil–military coordination effective? (2) Which metrics are best suited to understanding civil–military engagement? and (3) Can examples of effective civil–military coordination be replicated in different environments? To address these questions, the paper outlines a framework of indicators to measure the effectiveness of coordination in both complex emergencies and natural disasters. The remainder of the paper is structured in the following way. The first section will look at existing literature on the subject and clearly identify the gap this paper is attempting to fill. The second section will introduce the indicator framework (IF) with nineteen descriptive level and twenty perception and effectiveness indicators and explain how and by whom these can be used. A brief conclusion is then drawn, before ‘The Humanitarian Civil–Military Coordination Indicator Framework’ is presented in full in the Annex to the article.

What Civil–Military Relations Literature is Missing

Humanitarian civil–military relations literature has grown in recent years, and studies vary from assessing coordination in specific emergencies from a military or humanitarian practitioner perspective, to evaluating cooperation in specific aspects of a response such as logistics, and generalised assessments of the status quo of civil–military coordination for humanitarian purposes. This section provides a brief overview of relevant literature and explains which gap the IF is attempting to address.

Heaslip and Barber (2016) focus on humanitarian logistics and supply chain management. They have argued that improvements in coordination should not only take place in a response phase of natural disasters, but also during post-disaster reconstruction and during the handover of operations to the affected nation. In a separate work, Barber (2012) makes the point that despite many differences between humanitarian and military organisations, they are actually very similar in their approach to logistics.

In their study of civil–military engagement in the 2010 floods in Pakistan, Madiwale and Virk (2011) find that national militaries are frequently first responders in disasters, rather than only supplementing civilian responses. Forestier, Cox and Horne (2016) provide an analysis of civil–military coordination in the 2014–16 Ebola crisis in Sierra Leone and draw an overall positive picture of coordination, which can in part be explained by military and humanitarian responders having closely aligned goals in that response. This is one of the few publications that is explicit about positive outcomes of civil–military coordination, whereas many others focus on outlining shortfalls and problems of current civil–military engagement only.

Several authors specifically discuss guidelines on civil–military engagement, including the Oslo Guidelines (for natural disaster) and the Guidelines on the Use of Military and Civil Defence Assets (for use in complex emergencies), as well as country-specific guidelines or the lack thereof (Madiwale and Virk, 2011; Bollettino, 2016; Svoboda, 2014; Haysom and Jackson, 2014). Lloyd and Van Dyk (2007) provide interesting insights into the work environment and required skills of civil–military coordination (CIMIC) officers. They argue that CIMIC officers face cognitive, emotional and social stressors, which makes their job requirements different to that of other officers, as civil–military coordination is a particularly challenging task.

Rietjens, Soeters and van Fenema (2013) studied cooperation between the Dutch military and civilians, including non-governmental organisations (NGOs), in Afghanistan between 2006 and 2010. Their paper suggests a ‘compass’ for militaries on how to establish and maintain relationships with civilian organisations. They also point out that civilian or non-governmental organisations should not be seen as one coherent sector, but that views on engagement with militaries are diverse. The goal of the IF is related to the compass for civil–military engagement suggested by Rietjens et al., although the IF is designed specifically for assessing interaction between military and humanitarian organisations and thus not, for example, civilian agencies active in the field such as foreign ministries.

As Zyck has pointed out, one key concern of humanitarian organisations is that sharing any information with militaries might violate humanitarian principles and put them at greater risk of attack (Zyck, 2013). This is echoed by others, and not just in relation to information sharing. In the commentary to the 2010 updated version of the Humanitarian Policy Group’s Good Practice Review 8, Operational Security Management in Violent Environments, Fast and O’Neill point out that actions by military actors can undermine the security of humanitarian organisations, even those who have overall acceptance for their work.

While an individual organisation may well have established an effective acceptance-based approach, this hard-won acceptance can be undone by the behaviour, affiliation or other attributes of another, unrelated organisation. Thus, in places like Afghanistan and Chad, where military and civilian actors work in close proximity, the actions of non-humanitarian organisations can undermine the safety and security of humanitarians. (Fast and O’Neill, 2010).

Kleinman and Bradbury, writing in the same issue, find that aid projects carried out by militaries themselves, such as the US military in Iraq and Afghanistan, but also in countries not currently affected by active conflict, serves to ‘win the hearts and minds of local communities’, thus ultimately serving military strategic interests (Kleinman and Bradbury, 2010). What they highlight here is essentially the fundamental difference between military aid and humanitarian aid: while both provide aid to people in need, military aid very rarely does so in a neutral and impartial manner, but always with a strategic objective behind it, such as winning support for the military’s presence in a region, or gathering information about enemy activities. These are some of reasons civil–military engagement is difficult to execute and also challenging as a subject of study.

The Indicator Framework

The IF was developed in several stages to fill the gap in the data outlined above, and enable researchers and practitioners alike to start collecting quantifiable data with a common tool. An initial review of literature identified previously developed questions and metrics on civil–military engagement. The second stage involved creating a questionnaire designed to understand humanitarian agencies’ experience with civil–military coordination during the international response to Typhoon Haiyan in the Philippines. Sixty-four completed surveys provided insights into challenges in coordination and cooperation in that response and informed a first IF draft. This draft was consulted on in a multi-sector, multi-disciplinary review by a group of experts representing the UN, international NGOs, academics and military staff familiar with humanitarian missions. Consultations were conducted simultaneously to an in-depth review of literature from academic sources and think tanks and research institutes deeply involved in humanitarian studies, including the Overseas Development Institute and the Stockholm International Peace Research Institute, which was touched upon in the previous section. The final step was to integrate feedback and information from these separate steps, eliminate duplication and provide clarification where needed to develop the full IF, which is annexed to this paper.

The IF provides a set of questions that may be used at any stage of a response to a humanitarian emergency from mission planning and assessment through the various stages of a response and post-response assessment. The full set of questions, or a more targeted subset of these questions, may also be used as periodic polls to actively monitor developments in theatre as a means of identifying gaps in the response and areas where civil–military engagement could be enhanced.

The IF is designed in two sections. The first includes questions that aim to establish baseline information, such as the extent of civil–military engagement, level of formality of the engagement, and the actors involved. It contains nineteen indicators. These are designed to provide descriptive level information on humanitarian civil–military engagement. The second part contains more detailed indicators on perceptions of the effectiveness of humanitarian civil–military engagement. This section contains twenty indicators. One of the functions of the IF aims at assessing how well the current practice of civil–military coordination works – or not. Examples are indicators on the UN Civil–Military Coordination (UN-CMCoord) process and interaction with the Logistics Cluster, the presence and practice of multi-national military coordination centres (MNMCCs), and the staff involved in civil–military engagement and their training.

The IF is based on several key assumptions derived from the sources consulted in different steps of its development, including practitioners with field experience in civil–military coordination from both the military and the NGO side, as well as relevant scholarly literature. The first assumption is that established specific civil–military coordination structures are one of the factors determining effectiveness of coordination, providing for more effective coordination than scenarios with any formal arrangements for civil–military interaction. While unstructured coordination – meaning without established coordination protocols between organisations – can also be effective, this would depend more on the personalities and communication skills of the staff members in question than on a predictable, structured process (Rietjens, 2008). The second assumption is that trained and designated civil–military coordination staff could be a contributing aspect for effective engagement. This seems particularly likely as civil–military coordination staff need specific skills that often go beyond the skills needed for coordination with organisations within their own sector (Lloyd and Van Dyk, 2007). The third assumption is that the factors that determine effective coordination differ between contexts, particularly between natural disasters (which generally provide a more permissive environment for coordination) and complex emergencies (which tend to make engagement between civilian and military responders more challenging, especially if military actors are involved in activities related to the conflict and in activities related to relief). While these assumptions should be made explicit here and led to such indicators being included, that does not mean that the IF presumes a positive effect. Rather, it is a tool to assess the real impact of these and other assumptions that are often implicit in some of the literature on the subject but that are rarely backed up by data. To give an example, the IF might be used for an assessment which then shows low numbers of dedicated civil–military coordination staff but with engagement still being perceived as overall effective by the respondent. In such a case, in this hypothetical study, further indicators can then be used to triangulate which other factors could have contributed to an effective engagement.

Conclusion

The IF can be used not only by researchers, but also by practitioners in the field to evaluate past responses and plan for future ones. When using the IF for research and field level evaluations, only indicators developed for the specific context (natural disaster/ complex emergency/ both) need to be included. As humanitarian agencies and militaries increasingly operate in the same operational environment in emergencies across the globe, effective and well-coordinated operations are critical both for the recipients of humanitarian aid and for the safety and security of the actors involved in providing relief. We hope this IF serves as a first step towards formal analysis of field operations that ultimately creates the evidentiary basis for assessing the effectiveness of humanitarian civil–military engagement.

Annex: The Humanitarian Civil–Military Coordination Indicator Framework