I. Introduction

The business of providing humanitarian aid has become increasingly dangerous. In virtually every part of the world, those providing aid to distressed populations have been robbed, beaten, raped, abducted and murdered. In recent years, in fact, more people have been killed in action working for the Red Cross than for the U.S. Army. The causes for increased risk to relief workers are complex and varied, and operate on a number of levels. While this is not a new problem – relief workers have always confronted dangerous environments – the number, scope, and nature of threats to those engaged in humanitarian assistance have become much more severe in recent years, consistent with the proliferation of internal conflicts. United Nations humanitarian operations in conflict areas grew from five in 1985 to a total of 28 complex emergencies around the world in 1995.1

The humanitarian aid community, to include the United Nations, the Red Cross, non-governmental organizations (NGOs) and private voluntary organizations (PVOs), has been struggling to come to terms with the increased risk their personnel face in conducting relief operations. A recent survey of the humanitarian community concluded personal security was a major source of stress for expatriate field staff working in dangerous situations. Only 6% of those interviewed - which included development workers - recorded no security problems at their work location.2

Donor governments have also recognized that physical security is a critical component in making it possible to deliver relief aid to those in need in conflict environments. As recently as 1996, the international community actively contemplated intervention by a multi-national military force in the Great Lakes Region "to establish security and get humanitarian relief access".3 In December 1992, a multi-national mission with a similar objective to "secure relief operations"4 was launched in Somalia, with results that caused many in the United States to question the advisability of having U.S. military forces become involved in humanitarian emergencies at all.

As we survey the world situation, there seems little doubt that humanitarian crises will continue, that the international community will continue to wish to intervene to reduce the suffering caused by such crises, and that therefore severe risks to those who provide relief in complex emergencies will also manifest themselves. There is little doubt that, in the vast majority of cases, donor nations’ preferred response will continue, at least initially, to consist in funding NGOs, the UN and the Red Cross to provide relief as opposed to conducting direct unilateral or multilateral operations. It also remains clear that the U.S. and its major partners will retain the option of providing direct support to relief operations in insecure environments, with military force if necessary. As the U.S. National Military Strategy for 1997 observes, "armed forces can assist with the pursuit of humanitarian interests when conditions exist that compel our nation to act because our values demand U.S. involvement".5 Similarly, UK Secretary of State for Defense Michael Portillo, announcing Britain's willingness to send a contingent of troops to Zaire to facilitate relief operations, responded to concerns about a lack of ‘vital national interest' in that conflict with the statement, "...Britain is a civilized nation. We can see that people are about to die in the thousands and we are one of the few nations on earth who have the military capability to help at least some of them."6

In keeping with this, the UN Security Council has also demonstrated increasing willingness to envisage the use of coercive military action in support of humanitarian space or in order to enforce international humanitarian law (IHL). Resolution 751 on Somalia specifically refers to the need for "safety and security of the staff of humanitarian organizations"; and there are numerous parallels in other texts.7

U.S. national security policy, most recently expressed in Presidential Decision Directive 56, directs the U.S. military to plan for providing security for relief agencies in hostile, or "non-permissive," environments.8 Contrary to widely-held assumptions, the relief community, in recognition of the increased complexities of the security environment, has often called for military intervention to provide security in extreme situations so that relief could be administered and the protection of endangered populations ensured. Despite these calls, there is a measure of skepticism as to the effectiveness of military intervention in practice.9

The scope of the paper is nonetheless somewhat restricted. We are not attempting to look at the whole complex of issues which may have a security character and merit a military response, particularly insofar as these concern distressed populations facing intimidation and threats from hostile or indeed indigenous militias. These are clearly challenges of another order which merit, and must receive, separate treatment10.

The problem of security for relief operations has, over recent years, become widespread and pervasive. The UN Security Coordinator (UNSECOORD) notes that, prior to 1992, security was "not a major issue" for the UN, that "UN policy was that staff and dependents should not be exposed to dangerous situations," and that "it was almost unheard of for a staff member to be killed."11 By way of contrast, in June 1997, the UN considered 53 countries to be insecure to some degree; 28 were considered wholly or partially to be at phase three (requiring relocation of all non-essential staff and all dependents) or above. UNHCR calculates that at the same date it had over 3,000 staff working in areas designated insecure, around 10,000 if associated NGOs are added into the equation. The working environment of UNHCR staff is said to have "altered dramatically" over the last five years. 12

The brief recitation of recent security incidents below illustrates the scope of the problem: 13

• In Somalia, more than a dozen relief workers have been kidnapped and killed since 1991. In 1995, an Italian doctor was murdered while working for the NGO Caritas/Italy.
• In Burundi, Tutsi extremist bands have repeatedly attacked relief convoys, compounds, and personnel. In 1995, at least 10 international and Burundian relief workers died in politically or criminally motivated attacks, while in 1996, three ICRC delegates were killed while conducting relief operations.
• In Sudan, four Sudanese working for Save the Children, Norwegian Peoples Aid, and German Agro Action died in three separate war-related incidents in 1995.
• Also in Sudan during 1995, 41 international relief staff were taken hostage and held for various lengths of time by various parties to the Sudanese conflict, while an additional 29 were seized in 1995. There were 35 evacuations of relief personnel between mid-1995 and mid-1996 for security reasons.
• In October 1995, an American relief worker lost her legs and another was injured when their car hit a land mine near Goma, Zaire.
• A French aid worker was killed in November, 1997, during efforts by security forces in Tajikistan to free her and a colleague from their abductors.
• Six International Committee of the Red Cross (ICRC) workers were murdered inside a compound protected by security guards in Novy Atagi, Chechnya, in December, 1996;
• In January, 1997, three Spanish medical personnel working for the NGO MDM were murdered in northwest Rwanda, while in the same attack an American was severely wounded and lost a leg;
• In February, 1997, five personnel representing the UN High Commissioner for Human Rights Operations in Rwanda (HRFOR) were ambushed and murdered in southwest Rwanda.

In terms in overall scope, the UN Security Coordinator reported in April, 1997, that between 1 January 1992, and 1 March 1997, there were 131 UN staff members killed, of which 28 (21 percent) were engaged in humanitarian operations, and 68 (52 percent) were killed in areas where there was no de facto government. Between 1 January 1994 and 17 March 1997, there were 119 individuals taken hostage in 35 incidents, with staff members being held for varying amounts of time ranging from several hours to five and a half months. It is important to note that of the 119 individual kidnapped, 67 percent were internationally recruited staff members. Regarding rates of incidents, UN statistics state that deaths amounted to one staff member per month in 1992, one every two weeks in 1993 and more than one a week in 1994; the figures were reduced to one a month in 1995 and 96, but there were nine fatalities in the first two months of 1997.14 The International Committee of the Red Cross (ICRC), one of the relief agencies which takes security most seriously, reports that in 1996 its delegates suffered 153 security incidents, to include staff members killed or wounded, a figure which has risen consistently over recent years. As in the case of the UN, numbers wounded or killed peaked in 1993/94 were lower in 1995 and 1996, but there is no sign that this represents any sort of long-term down-turn.

Understanding the nature of threats faced by relief workers, and their potential causes, is an important element in formulating responses. For present purposes, a four-fold distinction is proposed: accident, criminality, banditry and targeting.15

Much evidence, but again largely anecdotal, is available to establish a causal link between increased risk and a number of factors which characterise present-day conflicts, including:

• the economic value of relief supplies;
• the impact of relief on the dynamics of conflict;
• the potential political capital which aggressions against relief workers may attract;
• the possibility of collecting significant ransom money; and
• collateral killing as a result of the intrinsically anti-humanitarian character of the war goals of parties to internal conflict.

The particularly brutal nature of ethnic or religious internal conflicts and the dynamics to which they are subject undermines respect for the legal status of relief workers,16 blurs the distinction between combatants and non-combatants, and overcomes the notions of war objectives inherent in the Western tradition of ius ad bellum.17 These factors, which have been discussed in detail elsewhere,18 all contribute to a climate where relief workers become valuable targets to subvert the war effort of an adversary. The important point to note is that these factors are an intrinsic part of internal conflict and they will be a given in most relief contexts. Further, when relief is deployed it alters the conflict dynamics and becomes inextricably linked to and subject to the war economy; even if it is intended to be value-neutral, it will always be impact-loaded. Only an analysis which recognises this fact will be useful in mitigating its consequences.

There may be another structural factor contributing to increased casualties and risk. In a climate where donations to aid agencies have increased dramatically for humanitarian operations while development expenditure has declined,19 a phenomena which some suggest is caused by governments attempting to display their concern without having to implement direct political or military solutions, 20 competition between relief organizations for funding has intensified and new, inexperienced agencies are enticed into the market. It has been argued by one of the authors elsewhere that this competition leads to increased risk taking between agencies. From this perspective, donors unwittingly force existing players to discount risk - in effect to reduce the value they place on the security of their staff - in order to remain "competitive". In other words, agency and donor funding behavior can also unwittingly give rise to risk.21

It is important to bear in mind when considering the causal factors above that these conditions induce increased risk, but they do not directly lead to incidents as such. Humanitarian security is not analogous to military security. It is multi-dimensional and carries a multi-dimensional response in terms of programming arrangements as well as security precautions. This notwithstanding, many of the risks relief workers face can, however, be limited or mitigated through security procedures and practices, training and awareness. 22 Unfortunately, in many cases, serious security incidents have resulted from a lack of even basic precautions, awareness and understanding of the threats field workers face. In recognition of these shortfalls, a number of agencies have become aware of the need to significantly enhance their security capabilities over the past five years, and several have indeed moved to improve them. We will briefly review several of these programs below.

In providing relief, the UN plays a leading role, not only in the conduct of operations but also in sponsoring NGOs and PVOs as "implementing partners." With very few exceptions, the UN has by far the most organized security system and capabilities in place during relief operations, and in most situations the only "hard" security capability on the ground. Often, the UN provides the only structured security system, the only coordinated evacuation plan, and the only means to carry out evacuations when necessary.23

Within the United Nations system, the United Nations Security Coordinator (UNSECOORD) is tasked with coordinating, planning and implementing inter-agency security and safety programs and acting as the focal point for inter-agency cooperation on security issues. UNSECOORD’s operating costs are funded on an inter-agency basis, and its mandate covers all UN staff and staff on UN mission, whether involved in humanitarian assistance or other UN activities, with the exception of staff of UN peacekeeping missions. 24 UNSECOORD is a relatively small office, consisting of the UN Security Coordinator, who also holds the post of Assistant Secretary General for Conference and Support Activities, five professionals and three administrative staff.

In high threat areas, UNSECOORD provides field security officers (FSOs) funded on an inter-agency basis as required to support the Designated Official (DO), normally a high-ranking official from a UN agency who reports on security matters to UNSECOORD in New York. The DO is responsible for developing overall UN policies on the ground, formulating security and evacuation plans, forming a security management team (SMT) of senior agency and sometimes NGO representatives, and for establishing security phases on the basis of the perceived threat to UN operations and personnel. The DO, in consultation with the SMT, coordinates UN security measures with the host government, with embassies, and with NGOs. 25 As of April 1997, there were 27 UNSECOORD FSOs funded on an interagency basis. 26

The UN relies in large measure on the host government (assuming there is one) for ensuring the protection of staff involved in relief operations. The UN recognizes the host nation’s responsibility to maintain order and, in particular, to protect officials and the property of international organizations, which is an obligation under the United Nations Charter. UN staff members also have specific legal protection by virtue of the Convention on the Privileges and Immunities of the United Nations of 13 February 1946 and, in the future, the Convention on the Security of United Nations and Associated Personnel adopted by the General Assembly on 9 December 1994, but at the time of writing not yet in force.27 As UNSECOORD observes, however, "Whereas in the past personnel were assured protection by virtue of their association with the work of the United Nations, this is no longer the case. On the contrary, personnel are increasingly at risk because of such association." 28 Needless to say, when the UN is simultaneously engaged in coercive military action in response to a crisis, or even just issuing strongly-worded declarations from New York, the perception of the neutrality and impartiality of its humanitarian agencies on the ground is likely to take a serious beating. Relying on this form of protection then becomes naive to put it mildly.

The primary UN agencies engaged in front-line relief operations -- UNHCR, UNICEF, and WFP – have established their own security capabilities in line with their operational requirements. UNHCR and UNICEF have established security offices at their respective headquarters in Geneva and New York, and have hired their own field security officers to work in high risk areas. These officers are primarily focused on the work of their own agencies (although UNHCR is also interested in concluding security arrangements with its implementing NGO partners), and coordinate their activities with the UNSECOORD FSO. They develop specific operational plans for their agencies within the guidelines established by the SMT, conduct facility surveys to improve building security, establish emergency procedures at the country level and at specific sites, coordinate emergency communications, oversee contract security operations, coordinate specific operations with military officials where appropriate, provide training and situation awareness support to staff operating in the field and to headquarters, and respond to specific incidents and other operational requirements as they occur. FSOs are almost always unarmed, 29 and are often retired or former military officers with limited direct experience in relief operations.

The creation of agency-specific FSO positions and complementary security arrangements appears to be a rather natural consequence of the fact that UNSECOORD has limited resources and confines itself to a limited definition of security which captures only a part of field agency needs. This response is not, however, necessarily yet a very optimal one. The performance of the UN security system has come under close scrutiny in recent years, and there are those within the system who believe that it is being overwhelmed by the current security challenges agencies face. 30 Discussions with current and former agency FSOs have confirmed the concern that relief organizations have different operational requirements than those normally encountered by the UN system as a whole, and that if the agencies were expected to fund the UNSECOORD FSOs, they would prefer to have a more focused and dedicated capability outside the host capital city. Regardless of the degree to which agencies implement their own security arrangements, the benefits of having a high-level official representing security issues within the Secretariat, and a single chain of command where necessary, are recognized as significant strengths of the current system.

Although the primary UN relief agencies have published manuals and security operations guidelines, and try to provide some limited training to staff members prior to deployment to the field, most agencies acknowledge that their staff members are under-prepared for the degree of risk they face in high threat situations. 31 UNHCR, for example, notes that training "is a crucial element in assuring staff safety and security" but that while a training program exists "very few people have benefited from it." 32 Training provided prior to deployment is often augmented with country and site-specific training and drills once a staff member arrives in-country, but the degree to which this supplemental training is provided or approaches adequacy varies from agency to agency, and from FSO to FSO.

The security relationship between UN agencies and NGOs in the field is complex, and while specific some guidelines exist on paper, often implementation is situation-specific. In terms of formal guidance, under an MoU circulated by UNSECOORD in January 1997, NGOs who are implementing partners of organizations of the UN system may request to be included in UN security arrangements. To participate in the UN system, NGOs must agree to pay their share of the related costs, using the same formula that is applied to UN agencies, and to abide by UN Security guidelines and directives. With regard to evacuation, assistance is provided to NGOs who have not entered into such an agreement, or are not implementing partners, on an as possible and to the extent feasible basis. Local staff of such NGOs, just as local UN staff, would generally not be considered eligible for evacuation. 33 The option of concluding an agreement with the UN to be included in security arrangements is not available to NGOs which are not UN implementing partners. The UN limits it, moreover, to those employees which are directly engaged in contracted relief activities. This situation could conceivably give rise to two categories of security cover within the same NGO – one for staff carrying out UN work and another for all other staff.

To date, NGOs have been sceptical regarding participation in the UN security system. When UNHCR raised the issue with its main operational partners, "none of [them] had previously seen the document, and all of [them] stated they would not be signing it"; revisions were called for.34 UNICEF, on the other hand, has argued that NGO objections to the cost and loss of autonomy associated with "buying-in" are not convincing.35

The UN is acutely aware that its security system and practices need to be strengthened, and is taking a range of steps to improve its overall capabilities. Recent demands by the UN employees’ union to improve security conditions for staff members has further increased pressure on the UN leadership to take more decisive steps.36 Several recent measures are presented below, but in almost all cases security enhancements have been limited by funding constraints as well as a limited conceptual understanding of the problem. In recognition of the severity of the problem, UN General Assembly Resolution 52/45 calls on the Secretary General to present a report on the "security situation of all humanitarian personnel and measures to be taken to improve it".

There is even more variance within the non-UN relief community in actual practice and capabilities. The ICRC is perhaps the most advanced non-UN relief organization in terms of its security training program, guidelines, and overall capabilities. The organization has gone so far as to create a training site in Switzerland which simulates a contested eastern European village, and provides its delegates with up to two weeks of realistic pre-deployment training.37 The ICRC also provides specific security training to "security delegates," which it recruits externally and from within its own ranks. Security delegates are assigned to high risk operations on an as needed basis.38 The ICRC has also developed specific guidelines for delegates, which provide practical guidance on different operational security situations. It incorporates suggestions on behavior which address its special status, including a prohibition against using armed escorts except in exceptional situations.39

In terms of NGOs, individual agency preparation of their staff members varies, primarily in direct proportion to the size of the agency and its experience. Save the Children Fund - UK, for example, has published and recently updated a detailed manual providing operational and personal security guidelines and procedures to staff members.40 A few other agencies have undertaken similar efforts to provide security handbooks to staff members prior to deployment. Security-specific training, however, is more the exception than the rule in most NGOs. Some of the larger NGOs have formed security offices, or have appointed agency-wide security coordinators. One of the first agencies to take this step was World Vision, although the individual in charge of security coordination would be the first to admit that he had no specific security experience or capabilities prior to assuming the position.

In an effort to redress the lack of security doctrine and trainers for NGOs, the U.S. Office of Foreign Disaster Assistance (OFDA) sponsored a project to develop a training curriculum for NGO cadre trainers. Conducted by the U.S.-based NGO umbrella organization InterAction, cadre-trainer training was provided to NGO representatives at a workshop held in January, 1998. Topics included personal security, situation awareness, planning and communications. Trainers were drawn from the relief community, and the course was specifically developed to address the needs of NGOs. InterAction plans to hold another training session later in 1998.41

At the field level, NGOs in high risk areas will frequently appoint a staff member to oversee security arrangements and to coordinate any security planning or emergency responses with UN FSOs. In some cases, NGOs operating in a country will nominate a representative to serve on the UN SMT.42 However, with few exceptions, the staff members involved have little experience in security practices, techniques, and capabilities, other than those gleaned from prior operations. In many cases, they lack a framework at all to assess and plan for specific security risks. This is due in part to a shortage of individuals with the multi-disciplinary profile which, ideally, is needed for a field security officer working in a humanitarian relief context. It has become apparent in previous operations that it is not sufficient to take a relief generalist and appoint him or her in charge of security, any more than to take a security professional without assessing the difficulties he or she may face in adapting to the relief culture. This problem is certainly not unique to NGOs, as security professionals within the UN have observed in conversations with the authors.

As is well known, the NGO community is highly diverse, and the evolution of NGO security capabilities will no doubt largely depend upon the specific character of the agency concerned. While this is inevitable, efforts such as the InterAction training program are badly needed if agencies are to achieve a minimum standard of capability. Several other efforts to develop codes of conduct and standards of practice have emerged through inter-agency efforts. Notable among these is the People in Aid "Code of Best Practice in the Management and Support of Aid Personnel." Principle 7 of the Code calls on signatories to "take all reasonable steps to ensure staff security and well-being," and sets indicators addressing pre-mission risk assessment, pre-deployment training, emergency evacuation provisions, and regular briefings to personnel on security procedures.43 As guidelines such as the above gain wider acceptance throughout the NGO community, overall security practices will hopefully improve and the incentive to engage in risk-taking behavior to achieve a competitive position will be reduced.

Second, in general agencies prefer to take what has been referred to as an "anthropological" approach to security, believing that close and sustained association with the distressed population, as well as adherence to humanitarian principles, will afford the best degree of protection. This protection can take a number of forms, primarily consisting in establishing a conduit for information as a relationship is built between the community and the aid agency. In cases where potential attackers are from the benefiting community or well known to it, local pressure from the populace may act as a deterrent to criminals or paramilitary groups. Local communities may assist in early warning and, in extreme situations, in evacuation by alerting departing staff to the safest exit routes. The anthropological approach is also consistent with practices that emphasize reducing the material scope or incentive for abuses to take place, for example through the use of local merchants and service providers for delivery and distribution tasks, and, more generally, integrating the local population into the implementation of emergency assistance. 44

Third, the culture of relief agencies and the nature of their work heavily influences their approach to security. Relief organizations are committed to providing assistance to those in extreme distress, and in most cases the risks associated with these operations are acknowledged as being part of the package. This is not, of course, to suggest that there is any enthusiasm for risks which are disproportionate or unnecessary. The creation of "humanitarian space"45 is a critical concern from the twin perspective of gaining access to those in need and providing a secure environment for agencies to operate. Risks associated with establishing and maintaining this space are inevitable in extreme circumstances, and the willingness to take risks is often rewarded within aid organizations and by those who fund them. Many of the most experienced relief workers began their careers in circumstances more benign than the current environment, and the "work ethic" which has become a cultural part of the relief community is rooted in a more forgiving past. This is a large part of the challenge relief agencies face today – a significant cultural re-orientation toward risk reduction and control as opposed to risk-taking to achieve results. Risks need to be assumed consciously and intelligently: as the adage in the community has it, "when you are busy saving lives, don’t forget your own".

Fourth, the need to maintain an independent character is also an integral part of NGO culture, and all the more so in the context of relief operations where the need to maintain impartiality is uppermost. There is very rarely a hierarchical relationship, even between UN agencies and implementing partners, or anything even remotely resembling "command and control" in the military sense. The term "implementing partners" is far different than the contractor-client relationship in government-commercial or strict commercial relationships. This implies a collaborative relationship as opposed to an employee-employer one, and it is a relationship that is tied to the nature of the work of providing relief. While this "collaborative" relationship may have been somewhat eroded by what the International Federation of Red Cross and Red Crescent Societies (IFRC),46 among others, has called the emergence of "contract culture" in the relief community, and NGOs are in reality somewhat less independent than they would claim or wish,47 the underlying sense of a cooperative venture among independent entities at the service of civil society or human rights still exists.

A TFSOC would collect security incident reports and other relevant information, and facilitate or carry out operations coordination, planning, field security and facility security assessments, threat assessments, communications, and information management and analysis. These data would be overlaid on organization locations and operations, staffing, equipment and capabilities. This "fused" information would be centrally located to support planning, crisis management, evacuation operations and risk assessment, which would be shared among all NGOs operating in the country. It could also provide or support training, dissemination of information and expertise, and sharing of labor intensive tasks such as facility surveys and convoy operations. It would also provide a venue for agency leaders to develop unified policies on security conditions, operations and plans as argued in the next section. Over a period of several months, the center would plan on moving over to full agency staffing. Over time, NGO and/or UN personnel would be rotated through the center, providing "on the job" training in security operations management and coordination 48.

The quality of security threat assessments, and resulting situation awareness, is dependent not only on the existence of such an information management system but also on the quality of security officers and the experience of personnel in the field. Agencies currently dispose of very few staff members trained in conducting threat assessments, particularly when evaluation of military operations is required. It has recently been observed that "Assessing risk and determining risk reduction behavior is a skill that few staff may have, particularly those without professional military training."49 While the requisite knowledge of local conditions, and an awareness of trends and patterns, can be developed over time, often this awareness departs as experienced staff members rotate to new postings.

Coordination: Coordination between agencies in the field is a recurring problem area. Agencies have differing and sometimes competing mandates, and are often reluctant to coordinate the type of detailed information regarding their operations that is required to plan and conduct effective emergency security operations. Moreover, even if such information sharing is deemed desirable, there is no overall unifying body to pull together the security efforts of NGOs, PVOs, UN, and unilateral donor operations. While there are partial exceptions – Operation Lifeline Sudan for example – extensive "unity of effort" is rare to non-existent. Coordination is informal, often in weekly meetings, if it occurs at all. It is argued that the "contract culture" dynamic referred to above may influence cooperative efforts on the ground as agencies compete for resources within the context of a specific emergency. At the same time, however, if there is one area where relief agencies tend to agree (at least in theory) on the need for cooperative efforts, it is in the realm of security 50.

The institutional framework which a strengthened and more comprehensive UN system requires probably needs some more thought. Insofar as it is possible, the central mechanism afforded by UNSECOORD should be built upon, expanded in scope and provided with additional resources and more autonomy of action. At the same time, there is a need to strengthen the capabilities of UNHCR, UNICEF and WFP to enable them to provide better security support to their partner NGOs and others at the field level; cooperation between these field agencies similarly needs to be enhanced. Of course, any greater role for the UN may be limited by the readiness of NGOs to agree to be bound by UN security policies and guidelines, a condition which, as mentioned, many NGOs resist.51 This notwithstanding, greater support to NGOs even on an à la carte basis, for example in the fields of emergency communications and evacuation planning, ought to be an option.

A major lack in existing arrangements is that there is no mechanism which integrates the operational aspects of security which are specific to humanitarian aid, provides a standard platform for broad inter-agency cooperation on this issue, offers appropriate scope for NGO input in setting its parameters, and feeds into overall planning. This has two distinct manifestations. One is at the global planning level, which is clearly the responsibility of OCHA, or, in the presence of UN military support, of the UN secretariat as a whole (OCHA plus DPKO). Moves to strengthen the strategic character of the UN Consolidated Appeals Process, which is coordinated by OCHA, need to include the security dimension. The second is simply the UN version of the TFSOC idea referred to above. This could be hosted either by OCHA (in a similar way or even as part of their existing on-site operations coordination centre (OSOCC) concept, which is run out of OCHA’s Geneva Office but mainly deals with natural disasters), subject to the guidelines established the framework of the IASC, or by a lead field agency. In either case, it should be based on integrating security within general field support services so as to afford a comprehensive platform for planning. 52 Such a structure would support UN security decision-making, without necessarily affecting the formal chain of decision-making.

In terms of recruiting security personnel externally, military personnel who have conducted extensive mobile training team (MTT) operations may be the best source of recruits from the ranks of the military, along with personnel with significant on the ground civil affairs experience. There are several examples within the UN of retired special forces personnel, who are frequently tasked with MTTs in developing countries as part of their duties, making excellent security officers.53 This is also the case with military personnel who have served in attaché positions, particularly in the region where they have worked in the recent past. 54 Civil affairs personnel are typically more attuned to the relief culture, particularly if they have experience working in civil-military operations centers (CMOCs) alongside relief agencies. Special forces may also be more attuned to issues of diffuse threat by virtue of having worked on terrorist issues or in other hostile contexts distinct from open warfare.

Infrastructure: Almost all agencies in the field face significant shortfalls in the infrastructure required for effective security operations, including vehicles, communications, information management technologies, and physical security enhancements. The equipment and tools available to relief agencies has not kept pace with clear operational requirements for even basic necessities. 55 As a result, field personnel are required to accept even greater risks, while technologies and equipment are readily available commercially that could partially or completely mitigate threats. This shortfall is attributable both to lack of awareness on the part of agencies and to lack of resources.

Communications shortfalls is perhaps the most critical infrastructure problem area. Many agencies lack even basic, reliable communications capabilities beyond local VHF radios. Satellite communications capabilities are limited outside the UN agencies, and even then different agencies are often using different systems, which prohibits inter-agency communication. More importantly, there is rarely an established emergency information net in place that that includes all organizations engaged in relief activities. Even within the UN, "the issue of standardized effective telecommunications remains to be resolved ... when a crisis occurs, staff of different organizations are [often] unable to talk to each other".56 This issue needs to be sorted out, and it is reasonable to call for the UN to play a lead role in defining emergency frequencies and ensuring their coverage in the field, including for NGOs.

Resolving communications shortfalls will also require trained personnel, in the form of technicians who can configure equipment appropriately, and of operators to monitor emergency nets and provide emergency management support. As K. von Brabant observes, "very few (agencies) have a radio technician permanently in the field or train their staff in basic radio techniques or repair." 57 Ideally, NGOs would be able to participate fully in existing UN systems, but in situations where this is not possible, it may be desirable to establish a 24-hour emergency communications capability within the NGO TFSOC discussed above or at the field office of a particular NGO.

Summary: Before turning to potential cooperative areas between the military and relief agencies, it may be helpful to underscore a recurring theme in the preceding paragraphs. Cooperation and creation of synergy between relief agencies, both within the structure of the UN security systems and as an adjunct to it, is an essential precondition to improving security capabilities. The economies of scale associated with sharing information, resources, practices, and personnel is essential to ensuring minimum acceptable security standards. In addition, cooperation and cost sharing eliminates any "competitive advantage" an agency might accrue from opting out of cooperative security arrangements. Detailed "best security practices" should be established in the NGO community, for many of the same reasons discussed in the previous paragraph. 58 Initial steps taken in this direction should be expanded, and should cover training, planning, information sharing, and communications.

Even with this degree of consensus and cooperation, the donor community must recognize its role in ensuring the security of relief workers. This point argues for a "best practices for donors" approach. The donor community must put incentives in place for agencies to reduce the risks they currently take. Funding must be provided to support the development of security capabilities, and a structured approach to security should be a precondition of contract and grant awards. Agencies should include the costs of security operations and personnel in their contracts, and donors should stipulate security requirements in RFPs. For those agencies working with the UN, participation in a cooperative security network in high threat areas should be set as a requirement for becoming an "implementing partner." The donor community can have a tremendous influence on the creation of safe operating practices.59

Since the early 1990s, military and humanitarian organizations have more frequently entered into an uneasy alliance to support the provision of relief to distressed populations. The concept of using the military to achieve humanitarian objectives is not new. As Jim Whitman of Cambridge University has pointed out, "as early as the year 1000, French princes of the Church declared their willingness to wage ‘war against war’ by the intervention of collective military forces under religious leadership." 60 More recently, most military organizations have developed capabilities to provide relief to civilian populations affected by war. This type of expertise most frequently resides in Civil Affairs Units, 61 or units engaged in "civic action"; however, recent peacekeeping and peace enforcement operations have blurred these doctrinal distinctions as conventional units become more directly involved in complex emergencies.

The types of operations described above undoubtedly imply an innovatory approach to international military deployment; this evolution is underway, although not without its problems, as we move from traditional peacekeeping operations into "peace support" and "peace enforcement" activities under the pressure of the need to intervene in intra- as opposed to inter-state conflicts and increased willingness to countenance doing so. These types of operations may involve the use of troops in offensive roles "to combat banditry and restore popular confidence in law and order." 62

Stabilization operations of this type are most frequently conducted in what national security analysts have termed "gray areas," or situations where conditions such as widespread population dislocations, ethnic and religious conflict, and terrorist organizations and agendas, among others, exist. 63

Not surprisingly, the "gray area phenomena" overlaps directly with the security conditions relief workers face as discussed above. Jean de Courten, for instance, has observed that at the end of 1994 there were 17 operations in which ICRC and UN peacekeeping forces were concurrently present. 64 Military analysts are currently expending significant energy in grappling with the problem of how to apply forces trained, organized and equipped for direct force-on-force conflict to missions conducted in this environment. The U.S. Central Command (USCENTCOM) mission for Operation Restore Hope / UNOSOM II, to " provide a safe environment for humanitarian operations while operating under a UN mandate," 65 fell squarely within this environment, and the military organizations involved have been coming to terms with the lessons learned ever since.

Several other examples of recent military operations that involved security operations in conjunction with humanitarian assistance activities include: Operation Provide Comfort in Northern Iraq; the UN Assistance Mission for Rwanda (UNAMIR); Operation Turquoise conducted in Rwanda by the French under UN sanction; 66 and Operation Uphold Democracy in Haiti. Other significant, but more traditional peacekeeping operations that involved extensive interaction with humanitarian organizations include UN Protection Force (UNPROFOR) operations in the former Yugoslavia; the UN Transitional Authority in Cambodia (UNTAC); and the multi-national force (Operation Guardian Assistance in U.S. terminology) contemplated for deployment to eastern Zaire in late 1996. 67 As numerous other articles in this volume acknowledge, these operations illustrate that the types of missions in support, wholly or in part, of humanitarian assistance will largely be conducted under some international mandate, and in most cases will be conducted by coalitions.68

The nature of gray area conflicts by definition calls for a multi-dimensional, as well as a multi-national, response. As Larry Minear observes, "Post-Cold War crises are no longer simple affairs of single cause or single response. The political, military, human rights, and humanitarian dimensions, as well as the economic and development implications, now all come together like an accordion." 69 Cooperation between relief agencies and military organizations, in this view, is a given, particularly at the most dangerous front-end initial intervention stages of international action or in situations where conditions deteriorate dramatically.

Several initiatives have been undertaken to clarify the military’s role in providing security support to humanitarian operations. The Dutch, for example, in a non-paper from 1995, envisage the creation of a UN Rapid Deployment Brigade which could act "as an advance party for agencies providing humanitarian relief or ... in cooperation with such agencies, for example by providing military protection to humanitarian convoys or by pacifying refugee camps so as to facilitate adequate humanitarian emergency relief." 70 Recent British Army doctrine on peace operations observes that "A PK (peacekeeping) force may be tasked with specific protective tasks for a humanitarian operation. However, a PK force will not be able to ensure the protection and delivery of aid against widespread opposition. Such a task and the guaranteed protection of human rights can only be achieved by a PE (peace enforcement force)." 71 The British Army assumes that "Whilst (peace support operations) are not conducted exclusively by the UN... they will be at least authorized by the UN, joint, involving two or more of the armed services, combined and/or multinational and multi-agency."72 As noted above, U.S. PDD 56 recognizes the need for "humanitarian intervention" operations to provide security for relief operations, and that such operations may not have the consent of local military organizations and forces.73

While generic military doctrine for humanitarian operations is starting to be developed, the "political doctrine" is far less clear. Moreover, the track record is not good: as Prof. Adam Roberts has observed, "there has been remarkably little serious thinking about military protection, and the record of outside military involvement ... is full of instances of vacillation and retreat, poor coordination, a reluctance to make serious commitments and take serious risks, and achieving at best only temporary results". He goes on to suggest that any such intervention is at risk "due to certain inevitable features of the way the UN goes about its collective business".74 In concrete, we have seen in the past that the process of negotiation in the Security Council all too often leads to a lowest common denominator mandate which allows insufficient latitude to deal with the problems on the ground and may even lead to operations which are militarily untenable.

Recent evidence suggests that, in addition to political difficulties, the degree of "fog" surrounding complex emergency situations, and the speed at which they evolve, may pose an insurmountable obstacle to existing response modes. This was apparently the case with Operation Guardian Assistance, where the fluid nature of the situation confounded military planners. The mission was never clearly articulated for personnel on the ground, and it was obvious that the creation of "humanitarian corridors" would have potentially brought the MNF into direct conflict with the warring factions. As a result, the complexity and degree of the severity of the situation was minimized, with the American and British Governments arguing that there were "only" 50,000 refugees affected, and that many of these might in fact have been Zairian IDPs.75 The reality, as became apparent after the MNF had packed up its forward base at Entebbe, is that the relief community’s estimates of 250,000 refugees was much closer to reality.

Noting past collaborations in Cambodia, Northern Iraq, Somalia, the former Yugoslavia and Rwanda, UNHCR has observed that, "these common endeavors demonstrate the value of cooperation between civil and military bodies, and timely, well-planned, coordinated military participation in humanitarian ventures." The agency has developed a manual for military units and commanders, to "facilitate maximum coordination and cooperation between UNHCR and the military, and thereby to ensure the success of the UNHCR emergency operations in which the military take part, by helping members of the military to understand the nature, context and style of UNHCR activities in the field."76

In a sense at the other end of the spectrum is, for example, the ICRC, which argues that "humanitarian work must be disassociated from military operations aimed at ensuring security and restoring law and order in regions affected by conflict." As a consequence, the ICRC rules out "the use of armed military escorts to protect humanitarian convoys or any other humanitarian activity" for its operations.77 Armed escorts for relief convoys are seen as compromising the ICRC’s humanitarian identity and, quite possibly, even the viability of the whole operation which may as a result be seen as non-neutral and a "justified" target for attack. Nonetheless, "they remain an option for confirmed banditry."78

In order to combat criminality, on the other hand, armed guards at workplaces and residences may be used if there is no other option available. In this case, ICRC considers that "it is preferable to call on an established security firm or the police rather than the army."79

At the same time, however, the Red Cross movement recognizes that the military may play a role in establishing "humanitarian space," which the International Federation’s 1997 World Disasters Report defines as "a classic distancing technique employed to carefully separate military and humanitarian action, [which] offers considerable potential for effective military-humanitarian synergy when working in conflicts." The author of that report goes on to clarify that, on this model, "the role of armed forces ... is to provide security and freedom of movement for all, such as keeping airports and roads open, and carrying out mine clearance".80 In short, military intervention may be required to stabilize the security situation, but ICRC at least wishes carefully to distance itself from direct operational collaboration.

The military "foot print," or the size of the force and the speed with which it is committed to the operation are apparently significant factors in determining success, and agency, political and military decision makers should consider these when contemplating humanitarian intervention. Citing several operations over the past eight years that "exemplify success," (although with admittedly narrow criteria), Donald Daniel and Bradd Hayes observe, "The common thread throughout these examples is the quick deployment of robust forces which, possibly through shock effect, implicitly if not explicitly deliver the message that they mean business."81 These factors argue for rapid employment of an adequately sized force with a well defined mission – no military commander is likely to dispute this, but very often the politics of forming a military coalition prohibit such a response.

One alternative is the possibility of providing a rapid intervention capability from a central location (inside or outside the affected country or area as necessary), which could accomplish many of the same objectives of a dedicated physical protection force with reduced visibility. The utility of having a force available to intervene "surgically" if required has been raised in the Somalia context.82 Such a force would be charged with responding to specific attacks or threatening situations, but would in non-response mode remain physically and, more importantly, visually disassociated from relief agencies. There are a number of potential drawbacks to the rapid reaction force concept, including the response time to emergencies, response planning by both military organizations and relief agencies, and alert procedures. The conditions under which a rapid reaction force would respond would have to carefully defined, and the concept may work best in the context of a larger PK operation.

"Information" and threat assessment support: Providing intelligence from national sources, military or otherwise, could be a great asset to relief agencies on the ground attempting to respond to emergency situations and to understand rapidly developing conflict situations. The same can be said for national threat assessments as to capabilities and intentions of participants in the conflict. Such information could easily feed into the relief agencies’ assessment and planning efforts.83 There are a number of recent examples of this type of information being provided to the UN, 84 and it has been seen that the creation of a coordination mechanism on the ground, such as a CMOC, greatly facilitates information sharing.

The development of information support relationships have been hampered in the past by the failure of civil-military cooperation (CIMIC) arrangements, which are currently subject to extensive review in NATO. The possibility of governments’ utilizing their embassies to provide information to relief agencies has so far been insufficiently exploited. The notion of governments providing security-related information to agencies has some support in the humanitarian community. UNICEF has commented: "we strongly agree with this idea, noting that this already occurs between some governments and DPKO. Much information which is readily available to governments, including aerial photography, is not sensitive or can be made unclassified and still be of great value to humanitarians."85

Evacuation support: Providing military support to evacuation operations is clearly a humanitarian activity, and presents the least confusion in terms of potential problems with impartiality. The type of quick, in-and-out non-combatant evacuation operations (NEO) such as that conducted in Liberia is well within the capabilities of a number of military organizations. Several countries have shown, by word and deed, that they see this as a potentially important function of their militaries, and it is included amongst the Western European Community’s (WEU) task lists. 86  Military organizations can also be indispensable in assisting relief agencies with evacuation planning even if they do not intervene to carry out the evacuation itself.

At the risk of providing a laundry list of prescriptive actions, we offer the following for consideration:

• NGOs must take steps to further address the shortfalls in security staff capabilities, general staff security awareness, deficiencies in communication and coordination. The concept of a technical field security operations center (TFSOC) presented in this paper is a key element in improving the situation on the ground. At the same time, current collaborative efforts such as those undertaken by the members of the People in Aid project should be encouraged and, where possible, strengthened. The mechanism of establishing a lead NGO for security should be investigated, and appropriate cost sharing mechanisms and operational agreements structured. Serious thought should be given to the security relationship NGOs desire with the UN – failure to adhere to UN guidelines should be recognized as potentially excluding them from UN-assisted evacuation if the worst happens.
• UN relief agencies should continue efforts to strengthen their internal security capabilities in many of the same areas, and work on mechanisms to improve the security situation of both implementing partner organizations and other NGOs operating in the same area. Efforts to strengthen and adapt the UNSECOORD structure should be intensified, particularly in areas that directly support the operations of the security management team (SMT). Overall analysis of security incidents occurring throughout the relief community should be undertaken as a matter of urgency, and information analysis, assessment and dissemination systems should be instituted and shared. Inter-agency synergy in this respect needs to be a conscious objective, and the notion of building a TFSOC into on-site coordination mechanisms such as OSOCC should receive serious consideration.
• Military organizations can provide a range of support to relief organizations in the area of security, but this support must be provided in the context of agency concerns for impartiality and neutrality. Specific procedures for information sharing should be developed, training opportunities investigated, and various mechanisms for providing direct protective support explored. Doctrinal projects should be furthered and shared. Forces should be carefully structured to maximize their deterrent value with credible force, while at the same time containing elements to support civil-military operations.
• Donor governments have tremendous influence over the relief security situation, and should assume responsibility for doing more to protect aid workers in the field. As is abundantly clear from the preceding pages, efforts to improve security capabilities have been severely hampered by the lack of funds for security. Security needs to be supported, indeed required, when considering grant awards. Donors should make security expertise and relevant information available to relief agencies, and should actively sponsor and encourage combined military-agency exercises.

While the suggestions above provide an initial agenda for action, much remains to be developed, defined and implemented. Relief workers’ commitment to their vocation is legendary, and has been witnessed by millions around the world. The community as a whole must assume a similar degree of commitment to its own safety . It is similarly incumbent on those who fund relief operations to recognize the price of insecurity, and take determined steps to halt the suffering of those dedicated to stopping suffering.

* * * * * * * * * * * *

ENDNOTES

1. Bradbury, M (1995), "Aid Under Fire: Redefining Relief and Development Assistance in Unstable Situations", Wilton Park Paper with DHA, ODI and ActionAid, HMSO, London; cited in Slim, H.; "Planning Between Danger And Opportunity: NGO Situation Analysis In Conflict Related Emergencies"; presented at the ActionAid workshop on emergency preparedness at Kiboswa, Kenya in March 1995.
2. Macnair, R. (1995), "Room for Improvement", ODI RRN Network Paper n° 10
3. U.S. Department of Defense Background Briefing, Attributable To: Senior Military Official; Subject: Zaire; 15 November 1996.
4. Major General S.L. Arnold, U.S. Army, "Somalia: An Operation Other than War," Military Review, December 1993, page 27. MG Arnold was Army Force commander for Operation Restore Hope.
5. National Military Strategy of the United States of America – Shape, Respond, prepare now: A Military Strategy for a new Era. U.S. Joint Chiefs of Staff, September, 1997, page 6. Italics in original.
6. Cited in: Whitman, Jim, "`Those That Have the Power to Hurt but Would Do None': The Military and Humanitarianism"; presented to the `Aspects of Peacekeeping' Conference, Royal Military Academy, Sandhurst, January 23, 1997; available at: http://www-jha.sps.cam.ac.uk/a/a023.htm posted on 4 July 1997
7. See for instance Bettati, M. (1995), "L'accès aux victimes: droit d'ingérence ou droit d'assistance?", in "Law in Humanitarian Crises", vol. II, pp. 13-66, Office for Official Publications of the European Communities, Luxembourg. (1995) and idem in Université de Genève (1996), "Les Nations Unies et le Droit International Humanitaire – actes du colloque international", pp. 287-296
8. Office of the Assistant Secretary of Defense for Strategy and Requirements (OASD/S&R), background briefing, "Presidential Decision Directive-56: Managing Complex Contingency Operations," September, 1997.
9. See for instance de Waal, A. (1997), "Famine Crimes: Politics and The Disaster Relief Industry in Africa", ch. 10; and Roberts A. (1996), "Humanitarian Action in War".
10. See the comments by the working group chairs in conclusion to this volume
11. United Nations Security Coordinator (UNSECOORD, 1997), "The United Nations Security Management System", presentation to the HLWG, Geneva, April 1997. Statistics on non-fatal events are not given. M. Arsanjani in Université de Genève (1996) gives much higher figures, including presumably military personnel: 456 fatalities amongst UN personnel in the period 1991-95. It is reasonable to assume that (rightly or wrongly) the UN is both more security-conscious and more risk-averse than many NGOs, so these figures probably under represent the phenomenon in the wider aid community. Arsanjani also argues that "no person responsible for any of these attacks has been identified, let alone held accountable" (p. 117).
12. UNHCR (1997), "Joint Staff-Management Committee on the Security and Safety of Staff in the Field - Report on Security and Safety of Staff in the Field", UNHCR Geneva.
13. U.S. DOD, Global Humanitarian Emergencies, 1996; Section 3; 01 Oct 1996; Levine, I. (1997), "Promoting humanitarian principles: the southern Sudan experience", ODI RRN Network Paper n° 21, May 1997, page 9; and various new reports.
14. UNSECOORD (1997). Statistics on non-fatal events are not given nor is an incident typology put forward. It should be noted that 82% of casulaties involved staff recruited locally. The figures do not include military personnel. It is not clear what definition of "staff" is used. M. Arsanjani in Université de Genève (1996) gives much higher figures, including presumably military personnel: 456 fatalities amongst UN personnel in the period 1991-95. It is reasonable to assume that (rightly or wrongly) the UN is both more security-conscious and more risk-averse than many NGOs, so these figures probably underrepresent the phenomenon in the wider aid community. Arsanjani also argues that "no person responsible for any of these attacks has been identified, let alone held accountable" (p. 117).
15. This partly follows ICRC’s practice.
16. In conflict situations, relief workers are protected under international law (i) by virtue of their status as non-combatants under the Geneva Conventions; (ii) more specifically, under Additional Protocol I 1977 to the Geneva Conventions; and (iii) by virtue of specific rules relating to the UN, and the Red Cross. Specific protection may also be foreseen by means of bilateral instruments such as the Headquarters agreements that the Red Cross movement has negotiated with host governments.
17. Hunter, COL Horace L., U.S. Army, "Ethnic Conflict and Operations Other Than War;" Military Review, November, 1993, pages 18-19.
18. See for example, Robert Kaplan’s controversial "The Coming Anarchy," The Atlantic Monthly, February 1994.
19. UNHCR’s budget, for example, more than doubled between 1990 and 1995, from $544 million to $1.3 billion.
20. Ron Redmond, "Humanitarian Workers Easy Targets," The Washington Times, 5 may 1997, page 14. Redmond cites statements by Mrs. Ogata, calling relief operations a "humanitarian alibi" and a "fig leaf," and former UNPROFOR spokesman Michael Williams, now a senior fellow at IISS, who observes that "Humanitarian actions by civilian aid agencies is the easy option for governments today… (humanitarian action) is always easier than finding a political solution or mounting preventative military action." As was noted recently in a presentation by UN humanitarian agencies, relief workers "are often working in areas of operations where governments are seemingly reluctant to deploy their peace-keeping soldiers." (UNICEF et al. (1997), "Humanitarian Assistance – Security and Stress Concerns and Challenges", presentation to the Humanitarian Liaison Working Group, Geneva, April 1997 (co-authored with WFP, UNHCR and DHA), page 2.)
21. See, "Security of Relief Workers and Humanitarian Space," ECHO Working Paper, forthcoming.
22. This is of course not the only factor which can be acted upon to reduce serious security incidents. For a fuller treatment see ECHO (op. cit.)
23. This was the case during the evacuation of relief workers from Bukavu, Zaire in October 1996, conducted by the World Food Programme. Of 128 aid workers evacuated from Bukavu, five were WFP staff members and the rest from other United Nations agencies and NGOs. WFP News Release - 26 October 1966.
24. UNSECOORD’s role is of course limited to the technical/procedural aspects of security. When it comes to negotiating humanitarian space or setting the parameters of the relief environment other parts of the UN come to the fore – either OCHA (formerly DHA), operational agencies such as Unicef in the context of Operation Lifeline Sudan, and/or the Secretary-General or Security Council. It would be a great mistake to assume UNSECOORD either capable of or mandated to handle all aspects of humanitarian security.
25. UNSECOORD, "United Nations Field Security Handbook," 1 January 1995, pages 6-7, 12-14.
26. UNSECOORD, (1997), "The United Nations Security Management System", presentation to the HLWG, Geneva, April 1997 page 10. UNSECOORD note that there are an additional nine chiefs-of-security at peacekeeping and other field missions. They note that they would like to provide FSOs to all high risk duty stations, but that inter-agency funding limitations have prevented hiring additional personnel.
27. A call to states to become party to the 1994 Convention was included inter alia in UNGA Resolution 52/45. The Convention is discussed by M. Arsanjani in Université de Genève (1996), pp. 131 ff. and in ICRC (1998), "Respect et protection du personnel d'organisations humanitaires", Document préparatoire du CICR pour la 1ère réunion sur le DIH, Genève, 19-23 janvier 1998, pp. 20-22
28. UNSECOORD, (1997), "The United Nations Security Management System", presentation to the HLWG, Geneva, April 1997 page 1.
29. Although UNSECOORD field officers in Burundi and perhaps other locations have been authorized to carry firearms.
30. Cf. e.g. the report of the UNHCR Staff Management Committee (1997), p. 1, "There is serious concern that the size and structure of both UNSECOORD and of UNHCR's own (security system) are not presently adequate to meet the challenges faced by staff in the field"; p. 13 "The Designated Official System leaves much to be desired", and a review should address the need for UNSECOORD to have "the appropriate resource structure and management capacity to meet present, and future, security challenges". UN sources consulted informally by the one of the authors, while they appreciate that at least a UN-wide system does exist and support that in principle, have almost without exception confirmed this analysis: UNSECOORD needs to be strengthened, and its status may need review. Nonetheless, UNICEF has commented that "the UN security management system is working much better than it did in 1994" and affirmed its belief in the inter-agency approach.
31. Within the UN, security training has been recognized as a significant shortfall. UNSECOORD notes that "limited numbers of staff on an ad hoc basis" receive training, citing resource constraints, (United Nations Security management System, April, 1997, page 9); while UNHCR, UNICEF and WFP state that "it would be impossible to overstate the need for additional training" ("Humanitarian Assistance – Security and Stress Concerns and Challenges", April 1997)
32. UNHCR, "Joint Staff-Management Committee on the Security and Safety of Staff in the Field – Report on Security and Safety of Staff in the Field", UNHCR Geneva, 1997, pp. 2, 13. UNSECOORD notes that they have developed a training program jointly with UNHCR covering Personal Security Awareness, and have developed Security Management Training for limited numbers of personnel which is "provided to limited numbers of staff on an ad hoc basis as funds become available." UNSECOORD, presentation to the Humanitarian Liaison Working Group, Geneva, April 1997.
33. Thus, for example, they were left behind in Bukavu in 1996.
34. UNHCR, "Joint Staff-Management Committee on the Security and Safety of Staff in the Field – Report on Security and Safety of Staff in the Field", UNHCR Geneva (1997), p. 13
35. Source: comments to the authors.
36. Reid, Robert, "UN Union urges Staff safety," the Associated Press - New York, 12 December 1997
37. Colin Nickerson, "Relief Workers Shoulder A World Of Conflict: Aid Agencies Encounter Growing Dangers As Nations Withhold Peacekeeping Troops," The Boston Globe, 27 July 1997, page 1.
38. Jean de Courten, "Statement on Security Environment," presented to Humanitarian Liaison Working Group, 25 April 1995.
39. Schmidt, Frank, "Recommendations for improving the security of humanitarian workers," International Review of the Red Cross no 317, 1 March 1997, p.152-155
40. Cutts, M. and Dingle A. (1995), Safety first: Protecting NGO employees who work in areas of conflict, Save the Children – London.
41. Inter Action, "Monday Developments," August 1997, and conversations with InterAction personnel.
42. This was the case in Rwanda during 1996-1997, when a representative from Lutheran World Federation was selected to represent the interests of the NGO community to the SMT.
43. "The People in Aid Code of Best Practice in the Management and Support of Aid Personnel", ODI RRN Network Paper n° 20, Feb. 1997, page 23.
44. cf inter alia Prendergast, J. (1996), "Frontline Diplomacy: Humanitarian Action and Conflict in Africa"
45. The term "humanitarian space" has come into widespread recent use without agreement as to its precise definition. The origins of the term lie in the idea of a consensual space for humanitarian actors to do their work. Until recently, the fact that physical security is an intrinsic and necessary part of humanitarian space has tended to be downplayed. It is becoming increasingly clear, however, that this needs to change and, indeed, is changing.
46. IFRC, "Conflicts, contract culture and codes of conduct," June 1995.
47. cf. inter alia De Waal, "Famine Crimes", African Rights, London 1996, p. 141
48. There are nascent systems evolving in the UN community (UNICEF’s SoftRisk initiative, for example) and amongst NGOs for structuring this kind of information flow, but the field-HQ model and field collaboration issues have yet to be convincingly addressed.
49. van Brabant points out in, "Security Guidelines: No Guarantee for improved Security," Relief and rehabilitation Network Newsletter #7, page 3.
50. Signatories to InterAction’s NGO Field Cooperation Protocol agree to consult with other NGOs in order to reach consensus in dealing with various issues, including a number related both directly and indirectly to security. The technical/procedural aspects of security are referred to under a specific heading.
51. A recent example was the vocal criticism amongst NGOs of the UN’s decision to categorise half of Rwanda as unsafe in 1996/97, in opposition to their own views following the murders of five HRFOR personnel and three members of MDM. The UN is clearly in a difficult position when it is viewed as the sole arbiter of the security situation in any given country and its views need to be interpreted in function of the specific situation. Nonetheless, by virtue of its experience and size, it does and should play a role beyond the borders of the UN family stricto sensu.
52. OCHA, formerly known as the Department of Humanitarian Affairs (DHA), is responsible for coordinating UN responses to disasters. There is on-going debate within the UN community about the organization’s role in complex emergencies. Germane to the present discussion is DHA’s role in establishing the UN Rwanda Emergency Office (UNREO), which coordinated UN relief activities within Rwanda during the 1994 emergency. UNREO’s information coordination functions covered many of the same security-related activities that are discussed in this paper. See Taylor B. Seybolt, "Coordination in Rwanda: The Humanitarian Response to Genocide and Civil War," Journal of Humanitarian Assistance, available at http://www-jha.sps.cam.ac.uk/a/a027.htm posted on 5 July 1997, as well as Gregory Alex’s article in this volume. The DHA Information Centre which essentially operated as an OSOCC in Kigali during late 1996 and early 1997 did in fact provide a rudimentary capability in these areas.
53. The authors are aware of cases where UNHCR and UNICEF have recruited from the special forces community (both U.S. and UK) with excellent results.
54. Although an important cautionary note is that, at least in the U.S. case, attaché duties include intelligence collection activities, which may make personnel unacceptable to host nations.
55. An example of widely available technologies is the "emergency activation" switch in use on most radios carried by police and fire departments in the U.S. The EA switch is flipped when the staff member is in distress, and sends an emergency signal to the communications center. The radio operator is not required to speak, and there is no emission from the radio to alert people in that area that an emergency signal has been transmitted. The radio transmitting the signal is stamped with the agency and unit, and after confirming that the signal is a real emergency, the dispatcher sends units to the rescue. This system requires a full time communications center, and a standardized capability within the radios and base station, and operations coordination to monitor the unit’s last location.
56. UNSECOORD (1997), § 44.
57. K. van Brabant, "Security Guidelines: No Guarantee for improved Security," Relief and rehabilitation Network Newsletter #7, page 3.
58. At the same time, however, the impact of guidelines should not be overstated, nor should they be considered a complete panacea, as K. van Brabant points out in, "Security Guidelines: No Guarantee for improved Security," page 3.
59. These points, and several others addressing steps donors can take, are presented in much greater detail in ECHO, "Security of Relief Workers and Humanitarian Space," ECHO Working Paper, Section 3.1 (forthcoming)
60. Whitman, J., "`Those That Have the Power to Hurt but Would Do None': The Military and Humanitarianism"
61. Wright, LTC Walter E., U.S. Army; and Fiegle, MAJ Ronald L. U.S. Army; "Civil Affairs Support in Operations Other than War," Military Review, October, 1993 page 27.
62. Dobbie, Charles, "A Concept for post-Cold War Peacekeeping," Survival, vol. 36, no. 3, Autumn 1994, page 121.
63. Turbiville, Graham H. Jr., "Operations Other Than War: Organized Crime Dimensions;" Military Review, January 1994, page 35.
64. Université de Genève (1996). De Courten goes on, however, to underline ICRC's wish to keep as far from them as possible.
65. Freeman, Major General Waldo D. U.S. Army; Lambert, Captain Robert B., U.S. navy; and Mims, Lt. Col. Jason D., U.S. Army; "Operations Restore Hope: A USCENTCOM Perspective;" Military Review, September 1993, page 61. MG Freeman was deputy commander in chief / chief of staff for U.S. Central Command during Restore Hope, Capt. Lambert and LTC Mims were planners on the CENTCOM staff during the operations. Mims served as a liaison officer to the U.S. DART team.
66. Daniel, Donald C.F.; and Hayes, Bradd C.; "Securing observance of UN Mandates through the Employment of Military Forces," U.S. Naval War College, Newport, RI, 1995. Daniel and Hayes note that in Operation Turquoise, "The French wanted to act much like traditional peacekeepers whose function is to interpose themselves between parties who seek or might be tempted to massacre their opponents. They were quite ready, however, to enforce the UN’s writ (under the UNIMIR mandate) if necessary in the protective zone they established." At the same time, however, they note that: "The operation proceeded without securing the consent of factional leaders, or local military or paramilitary elements. page 19.
67. ibid., page 6.
68. Operations that involve only logistical support for humanitarian operations, without a security component, such as Operations Restore Hope (Eastern Zaire / Rwanda, 1994); and Operation Sea Angle (Bangladesh, 1991), to mention just two U.S. examples, fall beyond the scope of this paper.
69. Minear , Larry "Humanitarian Action and Peacekeeping Operations," Background Paper for the UNITAR/IPS/NIRA Singapore Conference, Feb. 24-26, 1997. Journal of Humanitarian Assistance Available at: http://www-jha.sps.cam.ac.uk/a/a024.htm posted on 4 July 1997.
70. Nuiten, F., Dutch national practice research, "ICRC Study on customary rules in international humanitarian law," The Hague, 1997.
71. Peace Support Operations, Joint Warfare Publication 3-01, (2nd Study Draft) Chapter 6 - Operational Tasks; Journal of Humanitarian Assistance, available at: http://www-jha.sps.cam.ac.uk/a/a423f.htm posted on 20 December 1997. The document defines Peacekeeping (PK) as operations "undertaken under Chapter VI of the UN Charter with the consent of all the major parties to a conflict to monitor and facilitate the implementation of a peace agreement." And Peace Enforcement (PE) as operations that "are coercive in nature and undertaken under Chapter VII of the UN Charter when the consent of any of the major parties to the conflict is uncertain. They are designed to maintain and re-establish peace or enforce the terms specified in the mandate." See Chapter 1 - Strategic Background To Peace Support Operations.
72. Ibid., Chapter 1.
73. OASD/S&R, background briefing, "Presidential Decision Directive-56: Managing Complex Contingency Operations," September, 1997, page 6.
74. Roberts (1996), pp. 8, 29; see also pp 35-44 for a detailed analysis of experience with peacekeeping troops in a humanitarian role and the notion of "safe zones" and pp. 65-69 on armed protection.
75. These estimates were based on aerial reconnaissance flights flown by British, Canadian and U.S. aircraft; while the aid agencies were getting reports from relief agencies and missionaries on the ground. The inability of the reconnaissance aircraft to penetrate the triple canopy jungle covering most of the affected areas was more severe than most military analysts thought at the time.
76. UNHCR, "A UNHCR Handbook For The Military On Humanitarian Operations," Geneva, First Edition - January 1995 page 1.
77. de Courten, Jean; "ICRC Statement on Security Environment," presented to the Humanitarian Liaison Working Group, 24 April 1997, page 2.
78. Dind, P., "Safety of Humanitarian Workers in Conflict Situations", ICRC, Geneva, 1997 § 6;
79. ibid., § 7.1
80. IFRC (1997), World Disasters Report, sec. 1, ch. 2
81. Daniel and Hayes, "Securing observance of UN Mandates through the Employment of Military Forces," page 7. The authors cite Operation Turquoise, UNITAF, Provide Comfort, and Uphold Democracy as successful as successfully inducing cooperation from belligerents.
82. Petrie, Charles, "Learning from Operation RESTORE HOPE and UNOSOM II: Somalia Revisited," presented to the Africa Studies Program at Princeton University, April, 1995. Petrie’s reference was specifically directed to the possibility of using U.S. Marines to intervene to break negotiation deadlocks.
83. This possibility is also raised by Van Brabant, in "Cool Ground for Aid Providers," page 8.
84. The authors are aware of several situations where such cooperation has occurred, including Somalia, Bosnia and Zaire.
85. Comments to one of the authors.
86. e.g. Belgian, French and US deployments to evacuate Zaire/Congo and US deployments to evacuate Liberia.