Skip to main content

Kidnapping and the limits of acceptance


For many humanitarian agencies, acceptance—gaining the trust and protection of local communities—is the preferred security management tool for reasons of perception, ease of access and cost (both real and opportunity costs). Humanitarian agencies have long been uncomfortable with the contradiction of using deterrence mechanisms in humanitarian operations, although the increased use of armed guards has been a noticeable trend over the last decade or so. Protection—‘bunkerisation’—has also become the norm in many highly insecure contexts, with similar contradictions and feelings of discomfort associated with this strategy. But in hyper-insecure contexts, is acceptance a viable option? This paper argues that in some contexts, the acceptance strategy no longer works. The primary cause of this is the increasing severity of the kidnapping risk which has overwhelmed the usefulness of ‘normal’, non-deterrence and non-protection-oriented security measures such as acceptance. The dangers of relying on deterrence measures for humanitarian organisations in such sensitive contexts will be reviewed. As a case study, the experience of one particular humanitarian organisation working in northern Nigeria and Syria in the 2012–2014 period is elaborated upon. A ‘zone of exception’ framework is proposed based on the work of Carl Schmitt. Issues for future reflection by organisations working in such contexts are introduced.


Humanitarian agencies have long had to concern themselves with managing insecurity. The myth of a golden era when humanitarian agencies could work freely and need not be concerned with their security is exactly that. Though contested,Footnote 1 over the last few years, a general perception has formed that humanitarian contexts have become increasingly dangerous for humanitarian agencies. This paper focuses specifically on the growing kidnapping risk. But first, it is important consider what we know about the current state of aid worker security.

Data from the Aid Worker Security Database (AWSD) indicates that since 2009, kidnappings have become the most frequent means of violence against aid workers of all types—national NGOs, INGOs, the UN, ICRC/Red Cross and Red Crescent Societies, internationals and locals included. In 2009, it was estimated that 85% of kidnapping cases did not end in death. There has also been a year-on-year upward trend for major attacks of all types (resulting in death or serious injury) (Stoddard et al. 2012, p. 2). A 2013 report also paints a disturbing picture of the developing security situation for humanitarian workers, stating that the number of major attacks rose again in 2012 to 167 major incidents (Harmer et al., 2013, p. 1) for the fourth time in a decade. That year, kidnappings accounted for a quarter of all major attacks on aid operations and 36% of aid worker victims (ibid, p. 3). In 2014, a subsequent report confirmed that 2013 was ‘another record-breaking year for violence against aid workers’ with 460 victims (Stoddard et al. 2014, p. 2). Both the number of attacks and number of victims were the highest ever—a rise of 66% in number of victims and 48% rise in number of separate attacks over the previous year (ibid, p. 2). Kidnappings again were amongst the most prevalent types of incident—out of 460 major incidents, 134 involved kidnapping (ibid, p. 4).

In 2013, 155 aid workers lost their lives, more than double the previous year. By far, the largest number of these (87%) were the national staff. The other 59 victims (13%) were internationals, which though a far smaller overall number illustrates a greater rate of attack when compared to their numbers in the field (it is estimated that less than 8% of humanitarian staff in the field are internationals) (ibid, p. 3). Local NGOs and Red Cross and Red Crescent Societies have borne the greatest burden of attacks. The top five contexts remain consistent, with Afghanistan, Pakistan, Somalia, South Sudan, Sudan and Syria appearing routinely.

In analysing the effects of insecurity on humanitarian action, particularly the threat of kidnapping, this paper takes a snapshot of the 2012–2014 period and builds a case study around one humanitarian organisation’s response to two insecure environments during that period—northern Nigeria and Syria.Footnote 2 Three key points are highlighted:

  • Although the number of major attacks fluctuates year-by-year, the data up until the end of 2014 paints a grim picture of the contemporary security context, indicating that providing humanitarian assistance has indeed become increasingly more dangerous. There are consequences to this besides making security management more difficult. The situation fundamentally affects the ability of humanitarian organisations to implement their activities, impacting on populations in danger and their critical needs through their decreased access to assistance.

  • Kidnappings in particular are on the rise. Kidnappings are not a new threat. They were prevalent in the North Caucasus in the interwar period in the 1990s and during the second Chechnya war in the early 2000s. Columbia also saw a high prevalence in the same era. The sense currently though is that the kidnapping threat has become generally more critical in a larger number of contexts, which greatly concerns the aid community.

  • The majority of attacks have occurred within a small set of contexts, but this should not mask the fact that serious attacks, including kidnappings, can occur anywhere. Even one serious incident in a context can have disastrous consequences—for the individual, the organisation, the aid community and especially, the populations needing assistance.

The Aid Worker Security Database on which the paper draws is subject to several limitations. The first involves the country’s specific nature of the data collection. The data shows the number of attacks within countries where aid workers operate but does not easily capture regional dynamics. Threats which originate in a context where aid workers are not present may impact where and how aid agencies operate in neighbouring countries to the point of forcing evacuations. In this case, due to lack of presence, and therefore incidents, neither context would appear to be dangerous. The second limitation is that data collection only reflects attacks on aid workers and not others. This would not be relevant except that non-aid worker security incidents can negatively affect neighbouring contexts where aid workers do work. Both of these aspects are explored in more detail in the first case study.

The paper first discusses the operational practises of humanitarian agencies and the tools available to them in managing and mitigating kidnapping risk, focusing on the acceptance strategy. This is further explored through analysing one humanitarian organisation’s response in the two case study locations of northern Nigeria and Syria. Finally, the paper proposes a framework for understanding the kidnapping phenomenon in relation to humanitarian agencies.

Security management tools—promise and limits

The toolkit of security management approaches available to humanitarian agencies include three basic strategies—protection, deterrence, and acceptance. Protection refers to hardening the target, which reduces vulnerability but does not address the threat.Footnote 3 It involves reducing exposure, decreasing visibility, and utilising protective devices and procedures, such as walls and barbed wire. This strategy is common in many contexts, and its extreme form is often referred to as bunkerisation. Isolating an agency’s personnel and materials in compounds with guards, defensive walls and limited access has in fact long been a normal response in many areas of insecurity.

Deterrence aims to actively counter threats and intimidate those behind them by use of armed protection, injunctions, withdrawal and/or suspension of activities. One form of deterrence which has been widely debated in recent years has been the use of armed protection. In some contexts, such as Somalia, this has been the norm for decades. Armed guards and escorts were also common in the North Caucasus for many years, and humanitarian agencies have begun to use armed protection in a larger number of contexts, such as in Iraq. One argument against their use is based on the perceived incompatibility between the use of armed personnel and the nature humanitarian action, with the threat of force in the implementation of humanitarian operations seeming contradictory to the ethos of humanitarian action, which is based on the alleviation of suffering.

The third option is acceptance, which aims to increase the consent for the agency to implement its operations.


An acceptance strategy tries to reduce or remove threats by ‘increasing the acceptance (the political and social ‘consent’) for an organisation’s presence and work in a particular context (politicians and the military call this ‘winning hearts and minds’)’ (van Brabant 2000, p. 58). Acceptance involves making and using contacts (networking), managing one’s image and showing cultural respect. The nature and appropriateness of programming also plays a key role. Highlighting an agency’s good works in the media, or discussing the value of an agency’s presence with local communities, are key tactics. The goal is to be more than tolerated, but rather, to be accepted into the communities where humanitarian operations are being implemented.

Acceptance is widely considered to be the preferable strategy for reducing or removing threats. As a strategy, it does not involve using weapons or the threat of force. It also does not rely on bunkerisation, though some aspects of protection may be used even if the primary strategy followed is acceptance, such as walled compounds and the use of unarmed guards. Even agencies which state that acceptance is at the heart of their security management strategy will often harden their compounds. For many, there is a difference between using protective elements in the construction of compounds and limiting uncontrolled access, and bunkerisation, which entails the isolation of an agency from the community and is more difficult to reconcile with the concept of acceptance.

Acceptance can therefore be paired with selected elements of other strategies, even the use of armed escorts for travel between locations. But organisations will quite often strive to put an acceptance strategy at the heart of its security management policies. As a way to achieve maximum proximity to the population—to know what their needs are and to be able to provide well-targeted services in a sensitive manner—it is considered the best strategy. In fact, if it works properly, it is a self-reinforcing process. The better the knowledge an organisation has about a population and context, the better the programming, which ensures improved acceptance and security. In contrast, reliance on protection and especially deterrence will often create a negative feedback loop which will decrease the effectiveness of an agency’s operations. Recognising this, organisations almost always try for at least basic consent form and contact with communities; rarely is acceptance given up on entirely.

The promise of acceptance is great. Its theoretical utility and minimal costs, including decreased opportunity costs, makes it the gold standard.Footnote 4 But there are also limits to its promise and a variety of reasons why it may not work. These mostly revolve around cases of extreme insecurity or an incompatibility between an organisation’s background and mandate and a local community’s worldview. There can also be tensions between agencies and local power structures. Acceptance requires a working relationship based on respect, trust, and mutual benefit. But building such a relationship, and in certain circumstances, having any form of positive relationship at all, is sometimes not feasible, given the political and security environment.

A key weakness of the acceptance strategy, if less acknowledged, is that even in situations where acceptance may work at a local level, other influences can interfere with the development of a constructive working relationship between humanitarian agencies and local actors. There is a common view that ‘all security is local’, necessitating the contextualisation of security data (Stoddard et al. 2012, p. 1). To a certain extent, this is true, as acceptance is primarily a local interaction between international agencies and local actors within the communities in which assistance is being provided. But external developments—transnational threats—may also adversely affect this localised acceptance. These include, for instance, anti-Western sentiments based on religious and political developments globally; the presence of transnational terrorist/criminal organisations; or the use of international agencies as ‘political footballs’ by national or regional actors. There are therefore contexts where acceptance may not work because of forces outside the control of both international agencies and local actors.

Acceptance also has limits that relate to specific types of security risks. Relying on acceptance to mitigate the risks of humanitarian personnel being kidnapped is highly challenging. In a situation where the chances of kidnapping are significant, it would be a high-risk strategy to rely solely on local acceptance for protection—particularly since in some situations, it is not the local actors who are the source of the threats. Given the serious implications of a kidnapping, lower risk strategies, such as decreasing presence, are preferred.

Understanding and evaluating the usefulness of acceptance requires an element of measurement that is not straightforward. One very simplistic method is to gauge how freely aid workers can travel outside their compounds without trepidation, thought to security concerns, and pre-planning. This can range from completely free access and movement, to full bunkerisation with little movement outside, or to the use of deterrence in the form of armed protection. At a certain point, acceptance ceases being a viable strategy, and organisations must consider relying more fully on protection or even deterrence (or not, as some organisations refuse to do so on ethical grounds). Whilst it is possible to chart the free movement of aid workers, this involves a high degree of subjectivity, as prescriptions on aid workers’ movement are ordinarily determined by the judgement of those responsible for security management within the organisation. Numbers can help with giving an impression of what the threats are and what the level of risk is, but a comprehensive understanding of a strategy such as acceptance requires more than this. It requires in-depth, qualitative analysis to access ground truths and understand the ‘hearts and minds’ of people.

This qualitative analytical process is based on deep context analysis, wide-spectrum networking, empathy based on proximity to populations in need, the utilisation of local knowledge and the understanding of transnational threats and regional dynamics, which is especially vital in mitigating against the risk of kidnapping. The tools to engage in such a process are available and known but often ignored or underutilised, as such process takes more time, skill and resources than the performance of simple mathematical equations which assign a number to a risk.

With this background discussion concerning acceptance in mind, the two case studies will be considered to see how these issues play out in practise.

Case study 1: northern Nigeria

The northern Nigeria context has changed dramatically over the last few years. Whilst the security challenges are many and complex, the focus here is on the armed non-state groups Boko Haram, Ansaru and AQIM.

Boko Haram was founded in 2002 by Mohammed Yusuf. The group is more properly called Jama’atu Ahl asSunnah li-Da’awati wal-Jihad (People Committed to the Propagation of the Prophet’s Teachings and Jihad). In 2003, it sought to establish an encampment on the Niger border, called Afghanistan, but this was soon destroyed by the security forces. For the next few years, the group remained fairly passive, but 2009 marked a transition for the group into a more active military role. Yusuf was killed, and violence between the remaining Boko Haram militants and the Nigeria state security services became common, particularly in Bauchi, Borno, Yobe and Kano states. Abubakar Shekau took over the leadership in 2010, and the group has since been responsible for thousands of civilians killed and a large number of attacks on military targets (Simonelli et al. 2014, pp. 1-2). In essence, Boko Haram is a Sunni jihadist group attempting to clear away what they consider to be the corrupt power structure in place in northern Nigeria and replace it with a regime following a fundamentalist interpretation of Islam.

The armed group Ansaru split with Boko Haram in January 2012, reportedly over disagreements about what constituted legitimate targets. Ansaru claimed Boko Haram was not acting correctly in killing innocent Muslims. Similarly, Boko Haram did not like Ansaru’s focus on kidnapping foreigners for ransom, as it wished to focus on the Nigerian government and security forces in the effort to rid the north of corrupt elite power structures (Simonelli et al. 2014, p. 4). Ansaru was involved in the kidnappings of a British and an Italian engineer in the north-western state of Kebbi in May 2011, which ended in the death of the hostages who had been held in Sokoto. The captors detected the presence of a foreign military surveillance team and followed standing orders to kill the hostages if there were an attempt to free them (Zenn 2013, p. 2). Hostage killing became a recurring pattern. A German engineer was kidnapped in Kano in January 2012 and died during a Nigerian-led rescue mission (ibid, p. 3). In February 2013, seven more foreigners were kidnapped from a construction site in Bauchi, north-eastern Nigeria, and on 9 March 2013, they were reportedly killed (Zenn 2013, p. 4).Footnote 5 Common wisdom within humanitarian organisations at the time said that the kidnappers were under pressure by the security forces and were thus forced to kill the hostages.

Far from being solely an indigenous Nigerian terrorist group, Ansaru has conducted its operations with the support of Al-Qaida in the Islamic Maghreb (AQIM), which has tried to link with indigenous sub-Saharan groups (Zenn 2013, p. 4). AQIM traces its roots back to the Algerian civil war of the 1990s and the Armed Islamic Group (GIA). The group became affiliated with Al-Qaida in the early 2000s and took on a more regional approach (Laub and Masters 2014).

Cooperation developed between Nigerian groups and AQIM in the kidnapping of two Frenchmen in Niamey, Niger, in January 2011. This incident ended with their death the day after their kidnapping in a rescue attempt involving the French military (Zenn, p. 5). It has been widely reported, and has been an operating assumption for agencies working in northern Nigeria, that AQIM gives support to Ansaru with the agreement that the group would take advantage of the presence of foreigners in Nigeria by kidnapping them and handing them over to AQIM operatives in the Sahel for money, arms and training (ibid, p. 6). In February 2013, Boko Haram appeared to have changed its tactics by kidnapping a seven-member French family in northern Cameroon, followed later by other kidnappings of priests and nuns in November 2013 and April 2014 (Cummings 2014).

During this period, AQIM’s intervention in local grievances in northern Mali provoked France to intervene militarily in the country in January 2013 (Annis 2013). This was a relatively short intervention as France began drawing down its forces in April and handed over responsibility to a UN force in July–August 2013 (BBC 2014a). The key point here is that this foreign intervention heightened anti-Western tensions in the region and provided greater impetus for the kidnapping of Westerners. The AQIM statements were clear at the start of the Mali intervention:

  1. 1.

    We send a clear warning to the government of the aforesaid countries; France, Britain, Holland and Sweden, that their stances a military attack (sic) on the mujahideen in north Mali will be taken as their approval for executing their citizens…

  2. 2.

    If the military operation causes the death of the hostages, their blood will be divided on everyone who participated in this [inevitable] aggression, and the European governments and their followers of governments of the coastal region will bear the full responsibility for what happens.

  3. 3.

    The Mujahideen do not wish for such a tragic ending, and they are careful to find a peaceful and just resolution for the affair of the hostages. And based on it, we direct an urgent call to the families of these hostages; we say to them: you must pressure your governments to avoid…this military operation that will surely cause the death of your relatives.

  4. 4.

    We demand from the media of the European countries in question to shed the light on this risk and political adventure the governments are going to wage, careless toward the blood of their citizens (Flash Point Partners 2012 , p. 1).

The humanitarian organisation under consideration, in its overall policy document for the mission, stated that it was working in Nigeria in order to provide emergency response in many locations of the north, as this area featured frequent outbreaks of disease. The organisation was also positioned in the north in case of degradation of the security and civil situation in northern Nigeria. In the event of such violence, ‘it would be very difficult to negotiate access to the north, especially as Western organisations are likely to be perceived as siding with the Christian south. It is therefore desirable to have a base of acceptance, knowledge, and activities in the north should the need to arise respond to a civil crisis’. An acceptance policy was then at the heart of the access strategy.

Though much of Nigeria has faced serious security issues over the years—especially in the Niger Delta, where kidnapping has long been a serious issue for foreign workers, the north had been relatively safe for aid workers throughout the noughts. For the most part, an acceptance strategy had successfully been used by the organisation until the rise of the kidnapping threat. Some protection tactics had always been used, such as the use of unarmed guards and fences around compounds, but focus had remained on an acceptance strategy. The situation changed quickly and fundamentally over a short period, however, beginning in 2012. Internal security documents from 2012 detail a concern that kidnappings could occur by criminal groups with the intention to sell hostages to AQIM, but this was not yet considered a high risk. But by 2013, the kidnapping risk was considered to be so serious that one of the far northern projects, located on the Niger border, was closed due to the high potential for kidnappings. The threat analysis was clearly stated in internal security documentation from the time:

The biggest threat that the mission faces is related to the activities of national and transnational militant groups and the subsequent threat of the kidnapping of expatriates. The rise of the militant group Ansaru over the last year has become one of the biggest security threats to the mission. The intensity of their links with AQIM can be debated, but what cannot be debated is the fact that Ansaru has been involved with a number of kidnappings of expatriates over the last year. … The concern remains in place, and in fact has been heightened, that criminal groups may kidnap expats on speculation to be sold to AQIM. The composition and extent of local/regional criminal groups is to a large part still unknown and needs further research. A continually improved security related network will assist in finding out more about such groups (Security Report, 2013).

The primary concern then was the activities of Ansaru, their links with transnational groups, and the threat of expatriate kidnappings. But how new was this threat for the organisation and how serious was it really considered to be for the organisation at that point in time (spring 2013)?

Kidnapping has always been a significant risk; there are internal Nigerian groups that may wish to abduct expats for ransom or for political reasons. However, since the war began in Mali in late 2012, the risk of abduction driven by external actors has shot up. As of early 2012, it was rumoured that AQIM would pay $200,000 for a European hostage. … Since the war in Mali began, it seems clear that AQIM’s appetite for foreign hostages is increasing. This means that not only ideologically motivated persons may be motivated to kidnap expats, anyone seeking money may see the opportunity to abduct expats for sale to AQIM. In particular, the ‘incursion-style’ abductions in early 2013 indicate the probability of highly-motivated actors with transnational connections (principally to the north, into the Sahel) proactively seeking out Western hostages (Nigeria Country Security Plan March 2013).

The ‘incursion-style’ abductions mentioned relate to a tactic used in some incidents where massive force was applied to overwhelm a target’s defences. The ‘direction’ of these incursions was from north to south, indicating a northern, even cross-border, origin of the kidnappers. The fear was based on real data as since between 2003 and 2012, 68 Westerners had been kidnapped in the Sahara region, 50% between 2008–12. It was estimated that $30 million had been paid in ransom since 2008. The ‘going rate’ was considered to be $2 million per expat (Kravitz et al. 2014). These figures do not exactly match the organisation’s estimates in detail, but the general impression was shared—there had been many kidnappings in the wider region, foreigners were the focus, each expatriate was valuable, and a large amount of ransom money had been collected by kidnapping groups. Most of the people kidnapped were in fact not aid workers. But who the hostages were was in many ways irrelevant, outside possible financial considerations, as it was only important that they were foreigners and preferably from a politically relevant country. If the objective was ransom, what the expatriate was doing was less important than who the expatriate was.

The presence of transnational criminal/terrorist groups operating in the region became a dominate theme for the organisation. Security management in the project area remained based on acceptance when considering local security issues, and this strategy had been quite successful in managing the project’s security. But the introduction of regional threats fundamentally changed the operational context and opened an internal debate on whether acceptance was still sufficient. In the end, the decision was made that emerging risks could not be managed through an acceptance strategy and that the at-risk project on the border should be closed. Three key points informed the organisation’s decision to close the project.Footnote 6

The first involved the seriousness of the threat—kidnapping. Based on the context and the analysis that the organisation had made, it was determined that the kidnapping risk was extremely serious. The consequences of any kidnapping are frightening, but then, recent history of kidnappings in northern Nigeria and the region had shown how alarming the incidents had become.

The second involved the organisation’s inability to monitor the context properly with regard to transnational terrorist/criminal groups operating at the regional level. This goes to the core of the issue—the lack of space to build a relationship or even trust in the possibility of contacting regional groups. In fact, the organisation determined that there was no feasible way to contact the groups that would potentially be involved in a kidnapping, at least in any meaningful way. This was partly due to the perceived incompatibility of the organisation’s objectives as an international humanitarian agency and a transnational terrorist group. Except in a case of a kidnapping for purely financial gain, it was also thought unlikely that a group responsible for a kidnapping would want to enter into negotiations with a foreign agency which had no official mandate and little formal political power, especially when the motive for the kidnapping would most probably be political in nature.

And the third point was the judgement that an aggressive deterrence strategy would not provide an acceptable level of security, even if the organisation decided to use one. In this situation and given the low level of emergency needs exhibited in the project area, such a strategy would both not be viable and not indicated.

Other factors were also considered besides these three main points. Organisational reputation was a concern—kidnappings can have a very negative effect on how an organisation is viewed. In addition, there are long-term effects of a kidnapping on the ability of agencies to continue to work in a given context. But it was the context-specific considerations discussed above that directly informed decision-making.

The Nigeria case is one example of a situation where emerging transnational/regional types of security threats can very quickly overwhelm an organisation’s security management options. Even if on a local level, acceptance works as a strategy in the face of transnational threats, especially as they relate to the risk of kidnapping; an organisation can be pushed very quickly from acceptance to deterrence and can ultimately determine that no strategies are viable. In this case, understanding the local context was not sufficient; acceptance with the local communities was not sufficient. It was the regional dynamics which made the organisation’s presence untenable. One of the important issues was the assumed lack of ability to discuss and negotiate with the relevant actors. Tracking these regional issues was essential to evaluating risk, but there were severe limitations on the organisation’s ability to mitigate against these threats.

Case study 2: Syria

This is a case study of the situation in 2014.

A second example of how the kidnapping risk affects the implementation of humanitarian operations concerns beheadings by the Islamic State of Western hostages in 2014. There are two issues to discuss in the Syrian context. The first involves the lack of local knowledge that agencies had to deal with from the beginning and which made acceptance less certain. In a sense, it is debatable whether there has been any ‘local’ at all in the Syrian context. And the second is that the context became dominated by kidnappings with increasingly very poor outcomes.

From August to September 2014, four Western expatriates were beheaded in Syria, nominally by the Islamic State. This Sunni jihadist group traces its roots to the Iraq conflict. The group Tawhid wa al-Jihad was established in 2002 by the Jordanian Abu Musab al_Zarqawi. But in 2004, a year after the US invasion of Iraq, Zarqawi re-formed the group into al-Qaida in Iraq (AQI) which was allied to Osama bin Laden’s al-Qaida. Zarqawi was killed in 2006 but the group remained active and in time created an affiliated organisation called the Islamic State in Iraq (ISI). After a period of being weakened militarily by other Sunni groups, ISI was rebuilt and by 2013 it had become a formidable fighting force in Iraq. By this time, the group had joined in the insurgency against President Bashar al-Assad in Syria under the name al-Nusra Front. In April 2013, the Iraqi and Syrian groups merged to form the Islamic State in Iraq and the Levant (ISIS). Dissension caused some from al-Nusra and al-Qaida to reject this formulation, but those loyal to the leader al-Baghdadi maintained their cohesion. At the end of 2013, ISIS moved back to Iraq, taking over substantial territory. By June 2014, a caliphate had been declared and the group became the Islamic State. Islamic State’s involvement in Syria has continued to grow. Many other groups, including al-Qaida, have shunned the group and condemned its brutality (BBC 2014b).

The Islamic State emerged as a brutal jihadist group operating in the conflict zone where Western journalists and aid workers were also attempting to operate. The group’s willingness to use extreme violence was emphasised by the beheading of two journalists (both American) in August and September 2014 and then two aid workers (both British) in September and October, with two others (at the end of 2014) also threatened with beheading (one American and one British). One of the aid workers beheaded worked for the French INGO ACTED and the second was an aid worker working privately with an aid convoy. ‘The motive changes according to circumstances. Sometimes the unlucky captives are insurance against attack or, worse, a target for retribution’, and to explain the captures ‘their captives are spies, they say, or prisoners of war, or bargaining chips for the exchange of ‘hostages’ unjustly held in Western prisons’ (Economist 2014, p. 40).

The year 2013 was the height of ISIS’s kidnapping—several dozens were taken. After that time, there was little expatriate presence in northern Syria. A common view by aid agencies concerned about kidnappings was that ISIS only started to ask for ransom when they were put under military pressure and required funding. They had already kidnapped locals for ransom but not yet internationals. Once targeted, though, internationals had ‘netted multi-million euro ransoms from several European governments’ (ibid). ‘Islamic State had been very successful in raising funds through criminal activities and is a richly resourced group who made money from natural resource extraction (oil), as well as kidnapping for ransom. They are often referred to as the world’s richest terrorist group’ (Lock 2014).

The Syria context could be interpreted as a local context where extremely violent groups operate, or a local context dominated by transnational oriented groups with a larger agenda. Either way, acceptance was not a viable strategy. As opposed to northern Nigeria where a formerly ‘safe’ area had been threatened by transnational jihadist groups, the Syria context was dominated by groups which had both local and transnational characteristics. The timing of the involvement of humanitarian agencies had also been against the negotiation of safe access. In the case of Syria, access before the insurgency was for the most part blocked by the Syrian regime. When cross-border access became possible in the midst of the civil conflict, it was already too late to build a viable acceptance strategy.

Whereas in Nigeria, an acceptance strategy had been slowly built through developing a relationship over time which was then disrupted, in Syria, relationships needed to be built with a large set of different militant actors, local and international, in the midst of active fighting. Yet, it was not inevitable that the context would slide into such brutality, where kidnappings were common and had such negative outcomes. This was partly determined by how agencies dealt with such a context.

The underlying question is: Are any organisations really equipped to work in such contexts? Large numbers of average Syrians, religious leaders, human rights activists and diplomatic personnel have also been kidnapped during the conflict (Giovanni 2014), demonstrating that even those with sufficient local knowledge and contacts have been affected. Organisations must have a good idea of the risk environment in any given context, but in an extremely insecure context such as Syria in the midst of a brutal civil war, it is even more imperative. And with this knowledge comes a responsibility to make informed decisions. As Ackerman (2014) statesFootnote 7: ‘It is effectively negligent to go into a place you don’t understand and expose vulnerabilities. To think you’ll be left alone because you are doing good is wrong…a kidnapping can become a geo-political event’. In many of the areas affected by the Syrian civil war, acceptance is highly unlikely to be a viable strategy, and indeed, it is difficult to see how protection or deterrence strategies would work any better.

An informed decision may be to abandon the context as the risk is too high and outcomes too poor, even given the massive humanitarian needs are present. The organisation under consideration in fact abandoned Islamic State areas after suffering a kidnapping. Another option may be to implement a remote management system, as did the organisation under consideration once the levels of insecurity for expatriates had become too high.Footnote 8 The remote management system put into place was an extreme form—for the most part, no formal linkages existed between the organisation and those implementing aid programmes. Remote management operations bring forward many ethical and operational dilemmas related to risk transference, accountability, monitoring, as well as challenges to the ability of organisations to speak out about what they do not witness themselves. This case is not unique—for in most aid agencies, many areas of northern Syria context have become off-limits, at least in any traditional sense of humanitarian programming.

Zones of exception?

Sovereignty can be defined according to prerogative: ‘The sovereign is he who decides on the state of exception’ (Schmitt 1922/1934, p. 5). A state of exception, in essence, is a period of emergency where normal rules do not apply. The rules still exist as a reference point, but they are no longer applied given the emergency situation which must be dealt with. This idea of deciding on a ‘state of exception’—and the related prerogative to make decisions based on it—is at the core of the interaction between governments and humanitarian agencies. It can also apply to the relationship between humanitarian agencies and armed non-state actors.

A state of exception is a crisis point. In this paper, it is correlated to both a context of civil conflict and a context of transnational terrorism. As Schmitt stated, ‘what characterises an exception is principally unlimited authority, which means the suspension of the entire existing order. In such a situation, it is clear that the state remains, whereas law recedes. Because the exception is different from anarchy and chaos, order in the juristic sense still prevails even if it is not of the ordinary kind’ (Schmitt 1922/1934, p. 12). A state—often a failed state—remains, whilst other actors attempt to institute a new juristic order on the territory. This is precisely the type of context in which humanitarian agencies often attempt to operate. Without an adequate understanding of this concept, humanitarian agencies will not truly understand the security risks they face.

In the case of armed non-state actors, the question is less about sovereignty per se, but rather an armed non-state actor’s prerogative to decide on a state of exception and, more importantly, what to do about it. In practise, the concept of the state of exception should not be confused with an actual declaration of a state of emergency, as an armed non-state actor cannot declare one. But a state of exception can be considered as any point in time where a normal legal and administrative system—rules and procedures—are fundamentally changed or ignored. In this case, the legal and administrative order relates to the local and international norms surrounding the working of humanitarian agencies. In a state of exception, these rules are not followed. In most cases, a government explains the changes on the basis of national security, a perceived threat to sovereignty and the like. An armed non-state actor in such a situation is concerned with achieving its objectives, in reference to other actors, but outside their rule structure. In many cases, it is the very rejection of the enemy’s norms and rules that is the objective.

The essential element to be understood by aid agencies is that the rules have changed and that the government, or increasingly a non-state armed actor, has entered into a new way of conceptualising the boundaries of its action and the acceptability of action by external actors. One critical way in which rules change or are ignored is in relation to the International Humanitarian Law, and the legal norms relating to respect for humanitarian personnel. There has been a long-standing assumption that these rules should always apply (even if this has never really been the case in practise). But in many contexts, these rules clearly no longer apply. It could be argued that these places are ‘zones of exception’. This is even the case for non-state armed actors. In the midst of conflict, such actors, in the same way as states, make a determination who is friend and who is enemy and based on this determination decide on when to declare that the rules concerning the proper treatment of enemies need to be discounted.

Within such an environment, there is a constant battle between humanitarian agencies and armed non-state actors, as there is with governments, on the parameters—moral, legal, and political—of their relationship. Humanitarian agencies must ask: What is the limit of an armed non-state actor’s prerogative? What laws and legal codes apply to the humanitarian agencies and in relation to the context? Is a ‘state of exception’ in place and how does this inform their relationship with the armed non-state actors and the population? Once a decision is made that a context becomes a zone of exception, certain actors within this zone are considered enemies and will be treated outside accepted norms. In such a situation, acceptance is no longer a viable security management strategy. Then, the discussion needs to be had whether protection and deterrence will be able to sufficiently mitigate the risk. Unfortunately, there are few, if any, steps to take to improve security in such situations, at least in the most egregious cases. The worst-case scenario is the abandonment of such contexts.

Are there any solutions, or is the situation hopeless in such contexts? Some within the humanitarian community think that ‘de-Westernising’ Western humanitarian aid agencies may work. The idea is to somehow strip away the Western identity of aid agencies, such as through altered recruitment policies—favouring non-Western employees, or by shifting the geographical locations of their management centres to non-Western settings. These strategies to change an organisation’s identity are often referred to as ‘diversification and internationalisation’. Such a strategy may possibly work over the long term and would bring with it many other benefits beyond decreasing security risks, such as improved local knowledge by managers. But in the short term, it is unlikely to have much effect. In particularly insecure contexts, there has been a trend to use national staff or expatriate co-religionists for the field level operations. But these staff are still representing a foreign organisation and therefore remain at risk. In addition, deflecting risk on to national staff is morally questionable. A search for other methods to re-build a set of operational norms for such zones of exception must be launched.


In summary, acceptance is not a viable strategy in all circumstances and for all security threats—and in fact it never has been. The question is whether acceptance as a security management strategy has lost ground in an increasingly number of contexts and in relation to such security threats as kidnapping. The concern is that in particular circumstances, such as in northern Nigeria, threats—especially the threat of kidnapping—come from transnational groups operating regionally, and therefore, the threat theatre has been delocalised. This delocalisation creates a potentially untenable situation for humanitarian agencies, as their security management options have become severely limited. And in the Syria for example, it is debatable whether there is even a ‘local’ to speak of, as the transnational aspects of the context are so dominant. Acceptance as a strategy was probably doomed from the start. And in both cases, there has been a disturbing trend for kidnapping victims to be killed.

This paper uses the concept of ‘zones of exception’ to describe the case study situations. These are areas of extreme violence where normal rules by which humanitarian organisations and states work no longer apply. Such a view allows an organisation to label a context, but more importantly it is a way for organisations to analyse and respond to such contexts of extreme risk. It is critical that humanitarian organisations have at their disposal methods to both conceptualise the contexts within which they work and upon which to base concrete responses. The fear is that this trend will continue and affect a greater number of contexts and regions of the world. There are no short-term fixes for this situation. More thought needs to be put into how the humanitarian enterprise can re-shape itself to take into account these disturbing trends.


  1. See for example: Neuman M. and F. Weissman (eds.).

  2. In the case studies, the organisation requested to remain anonymous and is therefore not mentioned by name.

  3. From a technical standpoint, the terms threat and risk have specific meanings. Threat refers to the potential event or incident that may result in harm or injury to staff, loss or damage of property, or negative impacts on programmes or reputation. Risk refers to the likelihood (probability of event occurring) plus the likely impact of the threat (without mitigation and contingency measures). This paper, though, follows a more common, looser usage of the terms.

  4. See Fast et al. Acceptance is not only a security tool—it can apply to development programming also and is sometimes not even used as such in humanitarian operations.

  5. Although, there was only documented evidence of four bodies.

  6. Much of this section is based on personal interviews with officials from the organisation.

  7. As stated by Steve McCann, CEO of Safer Edge.

  8. See Howe et al. for a discussion about partnerships between international and local organisations engaging in humanitarian action in remote management and insecure settings, particularly the northern Syrian case.



Al Qaida in Iraq


Al Qaida in the Islamic Maghreb


Aid worker security database


International non-governmental organisation


Islamic State in Iraq and the Levant


Medecins Sans Frontieres


Download references


I would like to thank the organisation under consideration for kindly allowing the use of internal documentation in this research as well as the anonymous reviewers for their input.


Not applicable.

Availability of data and materials

Available for review under certain terms, e.g. non-disclosure of the source.

Competing interests

I formerly worked for the organisation under consideration.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Andrew J. Cunningham.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Cunningham, A.J. Kidnapping and the limits of acceptance. Int J Humanitarian Action 2, 4 (2017).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI:


  • Nigeria
  • Syria
  • Humanitarian security
  • Acceptance
  • Humanitarian agencies
  • Kidnapping
  • AQIM
  • Carl Schmitt