Issue 25 - Article 8

In the line of fire: surveying the impact of small arms on civilians and relief workers

December 15, 2003
Robert Muggah, Small Arms Survey

One Sunday evening in August 2003, a group of armed men held eight relief workers at gunpoint at their home in the Burundian capital, Bujumbura.  The assailants, armed with pistols and assault rifles, threatened to kill one of the expatriates, and demanded that the group hand over all its cash. No one was injured but, as the OCHA Situation Report describing the incident observed, ‘the psychological impact on those present, mainly volunteers, is of grave concern’.

It is generally acknowledged that the availability of small arms affects the quality and quantity of humanitarian and development assistance, and an array of UN reports and studies have highlighted the dangers armed violence poses to relief personnel.  But there remains a surprising lack of evidence to prove what is intuitively known: that aid workers are frequently targeted and exposed to a high risk of death and injury in the course of their work. This article considers some of the findings of a recent victimisation survey initiated by the Small Arms Survey and the Centre for Humanitarian Dialogue.  It emphasises the views of workers in the line of fire, and considers a number of entry-points for constructive engagement with the small-arms issue.

The scale of the problem

Over 220 UN civilian staff have died as a result of malicious acts since 1992, and at least 265 have been taken hostage while serving in UN operations.  In 1998 alone, 27 UN staff members were killed in the field, the first year that more UN civilians have been killed than military personnel.  Thousands of relief and development workers have been targeted over the same period, and many hundreds killed.  Following the devastating attacks on the UN in Baghdad in August 2003, in which 23 people died, the UN Security Council voted unanimously in favour of a resolution to increase the protection of aid workers in conflict zones.  The final version of Resolution 1502 condemns violence such as kidnapping, rape and murder against UN and other humanitarian workers, and demands that states prosecute those crimes. It also classifies attacks on aid workers as war crimes, to be punished accordingly.

Over the past decade, considerable energy has been devoted to highlighting the consequences of armed violence, including small arms-related violence, on humanitarian and development personnel.  Even before the introduction of Resolution 1502, the UN had acknowledged that virtually every department of the UN system is exposed to the direct and indirect consequences of the unregulated availability of arms.  Risks of armed violence are not limited to UN workers. According to the UN Secretary-General’s report Safety and Security of United Nations Personnel of 2000: ‘threats against NGO staff can … directly affect UN humanitarian and assistance programmes, especially since conflicting parties often do not distinguish between UN and NGO personnel’.

A number of research studies have examined this issue in more detail.  In 1999, the International Committee of the Red Cross (ICRC) highlighted the implications of unregulated weapons proliferation during and after wars, as well as the long-term psychological effects of victimisation on ICRC delegates.  A range of studies have highlighted the implications of the widespread civilian possession of small arms, the frequent interruption of operations and constrained access to beneficiary populations, and the pervasiveness of intentional violence against aid workers.

Other epidemiological studies have established trends in mortality and morbidity among humanitarian workers and peacekeepers.  A study that investigated the deaths of 375 UN and NGO aid workers and UN peacekeepers between 1985 and 1998, published in the British Medical Journal in 2000, found that ‘weapons, rather than motor vehicles, pose the greatest threats.  Not only do young inexperienced workers die but veterans as well … Both expatriates and national staff share the risks, with death among the latter group probably greatly underreported’.  Some aid organisations have begun to improve their safety and security management, mostly improving incident reporting, research, training, advocacy and coordination.  But there are still too many aid agencies where senior managers fail to see the problem, or do not feel responsible or able to respond.

The study

The Security and Risk in Humanitarian and Development Action Study began in 2002 and will finish in 2006.  It was led by the Centre for Humanitarian Dialogue and the Small Arms Survey, in collaboration with a number of international NGOs (CARE, Oxfam GB, Médecins du Monde, Concern Worldwide, World Vision, UNDP, Handicap International, Merlin and the Save the Children Federation), and local partner NGOs.  Over 600 questionnaires were distributed in 39 countries and territories, in a variety of languages, in hard copy and electronic form.  A particular focus of the study was South-east Asia and the Balkans. The study will be undertaken again in 2004, focusing primarily on the Middle East and the Great Lakes region, as well as Afghanistan and Angola. Special emphasis will be placed on increasing response rates from UN personnel: in order to facilitate their involvement, the Centre for Humanitarian Dialogue and the Small Arms Survey signed a Memorandum of Understanding with the UN Security Coordinator (UNSECOORD) in New York in 2003.

The operational context

In order to generate a clear understanding of how respondents perceive the context in which they work, they were asked to describe the security environment using a four-point scale, from ‘little’ or ‘no’ violence to ‘widespread armed conflict’.  The survey found that respondents worked in a variety of security environments.  Predictably, these were not rated similarly across countries; Sri Lanka and the Philippines were on average rated as being the most violent or conflict-prone, while Thailand and Laos were rated as having the lowest levels of armed violence.  The survey also found that the prevalence of small arms availability and misuse shaped the assessments offered by respondents.

Humanitarian and development workers reported that a large number of groups – including organised criminals, insurgents and civilians – possessed weapons regardless of the security context.  A sizable percentage of respondents estimated that levels of civilian possession were ‘moderate’ to ‘very high’.

Victims of small arms availability and misuse

The survey also found that civilians are deliberately targeted (leading to unintentional death and injury), and that small arms are frequently used for criminal and coercive purposes.  Overall, the highest proportion of weapons-related death and injury among civilians was attributed to handguns. In areas of widespread conflict or war, assault rifles surpassed handguns as the leading cause of weapons-related death and injury among civilians.  Respondents also appear to routinely encounter a variety of small arms – mostly handguns and assault rifles – in and around ‘programme’ areas.

The survey’s two focus regions – the Balkans and South-east Asia – revealed important differences in the impact of small arms on operations, personnel and civilians.  In general, respondents from South-east Asia tended to report working in more violent or conflict-prone environments, and felt that small arms were more prevalent and more widely used. Respondents from both regions frequently reported seeing handguns, but respondents in South-east Asia were much more likely than their Balkans counterparts to report having seen assault rifles.  They were also more likely to indicate assault rifles as the leading cause of death and injury among civilians; to note the targeting of civilians with assault rifles; and to indicate awareness of unintentional death or injury of civilians due to assault rifles.

Relief and development operations were reported as being adversely affected by the availability and use of small arms.  Frequently, obstacles such as evacuations, suspensions or delays, and inaccessible beneficiaries, were associated with violent security environments and higher estimates of small arms prevalence and misuse. Nearly three-quarters of personnel working in areas with ‘very high’ levels of small arms availability reported recent suspensions or delays in operations.  South-east Asian respondents more frequently reported operational hindrances than did respondents in the Balkans. Moreover, they rated armed attacks on relief workers and armed conflict between belligerents as more significant hindrances to operational effectiveness.  Yet respondents from South-east Asia also expressed less negative attitudes towards small arms. It is commonly believed that the Balkans has more of ‘gun culture’ than societies in South-east Asia. It is conceivable that the risk threshold of Balkans respondents is higher than their counterparts in South-east Asia.

Screen Shot 2012-10-24 at 2.24.13 PM


Screen Shot 2012-10-24 at 2.24.21 PM

The overwhelming majority of respondents felt personally threatened by small arms. Perceptions of personal threat are heightened not only in areas characterised by higher levels of violence or conflict, but also where the civilian possession of small arms is seen to be more prevalent.  In addition to perceptions of personal threat, a large number of respondents reported that they or their colleagues had experienced serious security incidents, including armed intimidation, armed robbery, armed assault, detention and kidnapping.  Many respondents reported colleagues having suffered either non-fatal or fatal small arms-related injuries within the previous six months.

Despite working in dangerous environments, many personnel indicated that they had not received any security training within the organisation for which they currently worked.  The frequency of reported security training did not always correspond to the level of violence in a given environment, to the estimated prevalence and misuse of small arms, or to the level of personal threat expressed.  In many organisations, national staff are half as likely as expatriates to receive security training.

The importance of training cannot be overstated, particularly as the survey revealed that those that had received it typically viewed it as being helpful.  Security training or awareness is also associated with an increased tendency for individuals to take precautions, such as walking with others or limiting local travel.  The vast majority of respondents were unfamiliar with basic safety procedures associated with guns and ammunition, such as applying safety locks or safe storage.  Those who received security training were no more knowledgeable about small arms safety than those who had not.


The operational and policy implications of this study are multi-faceted. In addition to encouraging a debate within and between the humanitarian and development communities about ways to confront the unregulated availability and misuse of small arms, an array of pragmatic interventions could usefully contribute to improving the security of staff and civilians alike.  For example, although most agencies report incidents, coverage and analysis of this data should be encouraged. Agencies could also consider internal security reviews to assess how their staff perceive a host of issues – including small arms.  Many agencies would also benefit from the inclusion of small arms availability and misuse as early-warning indicators or as factors in conflict mapping exercises.  Finally, more attention could be directed to small arms in risk assessments, training and debriefing.

This article is dedicated to the victims and survivors of the attack on the UN in Baghdad, particularly internationally-recognised refugee expert and friend, Gil Loescher. We follow his and other’s lead in their efforts to make the world a safer and more humane place.

Robert Muggah is a Senior Researcher at the Small Arms Survey, Geneva.  In the Line of Fire: Surveying the Perceptions of Humanitarian and Development Personnel of the Impacts of Small Arms and Light Weapons, by Ryan Beasley, Cate Buchanan and Robert Muggah, is available at or

Further reading

International Committee of the Red Cross (ICRC), Arms Availability and the Situation of Civilians in Armed Conflict (Geneva: ICRC, 1999).

Dennis King, ‘Paying the Ultimate Price: An Analysis of Aid Worker Fatalities’ Humanitarian Exchange 21, July 2002.

Robert Muggah and Martin Griffiths, Reconsidering the Tools of War: Small Arms and Humanitarian Action, HPN Network Paper 39, July 2002.

Benjamin Seet and Gilbert Burham, ‘Fatality Trends in United Nations Peacekeeping Operations: 1948–1998’, Journal of the American Medical Association, vol. 284, no. 5, August 2000, pp. 598–603.

Mani Sheik et al., ‘Deaths Among Humanitarian Workers’, British Medical Journal, vol. 321, 2000, pp. 166–69.

Small Arms Survey, ‘Caught in the Crossfire: The Humanitarian Impacts of Small Arms’, The Small Arms Survey 2002: Counting the Human Cost (Oxford: Oxford University Press, 2002).


Comments are available for logged in members only.