Commander's Guidance: A Challenge of Complex Humanitarian Emergencies


© 1996 Andrew S. Natsios

From Parameters, Summer 1996, pp. 50-66.

United States military forces have now conducted operations in five complex humanitarian emergencies[1] beginning with Northern Iraq, followed by Somalia, Goma in Zaire, Haiti, and now Bosnia. None of these interventions required a conventional or even unconventional war on the battlefield. The military found itself supporting humanitarian relief operations, imposing order in chaotic circumstances, or implementing political accords to which the United States was a party. It would not be an exaggeration to suggest that the rank and file in the uniformed services are uncomfortable with this new role they have been directed to perform. Until recently they have had limited experience or training in the non-military disciplines needed to carry out these missions. Many are unsure why the military has been given duties it was never created or trained to perform.

The military should know it is in good company in its discomfort. Anyone who believes that all of the other actors who work in these emergencies--diplomats, economists, and humanitarian relief managers--are prepared for these complex humanitarian emergencies would be mistaken. Soldiers are not the only ones playing catch-up. None of the participating communities was particularly well prepared for this challenge of the post-Cold War world. Most non-governmental organizations (NGOs) have limited experience working in wars, let alone in conditions approaching anarchy. Most NGOs focus on development work. Those that do relief work have focused on natural disasters, which are relatively simple operations of short duration with few if any security risks, or on famines, where hunger is the central problem. Interstate wars and civil wars were left to the International Committee of the Red Cross, Doctors without Borders, and a small number of highly specialized humanitarian organizations. Diplomats--by training, temperament, and experience--are accustomed to dealing with other diplomats or with political leaders in the traditional meaning of the word. All that has now changed. Diplomats, relief managers, and development officials of donor aid agencies must deal with warlords, a glorified term for the leaders of gangs of looters who control ever-changing and quite unstable areas of troubled states. The warlord represents the new archetypal figure of the post-Cold War world: a militia leader of little conviction or even ideology, with some military training but little skill at governing, who rules by brute force and cunning.

Every discipline--diplomacy, humanitarian relief, development economics, and military science alike--must learn about their counterparts in the complex emergency response system. While such knowledge will be important at the tactical level and significant at the operational level, it is indispensable for developing a strategic plan for responding to an emergency. Each of the topics addressed in this article bears on that challenge: to know enough about each participating agency that all can be induced to cooperate in the planning and operational coordination that will reduce the risks to all participants while increasing the prospects for solving the central problems responsible for the emergency.

This essay presents some of the challenges every outside actor faces in complex humanitarian emergencies, with particular emphasis on how the interposition of outside military force affects or is affected by the conflict and how a commander might deal with challenges related to--sometimes even created by--the presence of his unit in the region. If commanders are to carry out their missions, they must know how their operations affect the agreed strategic plan for the emergency in order to avoid falling victim to the one characteristic common to all complex humanitarian emergencies: the law of unintended consequences. The greater the degree of disarray, even dissolution, within a social order, the more unpredictable the outcome of any intervention. That does not mean we don't intervene; it does mean we must think carefully about what we will do, how we can do it, and with whom we must coordinate our activities. Commanders and planners must learn to anticipate unanticipated consequences which might obtain if a given course of action is followed.

Two terms frequently heard in national security policy circles--mission creep and exit strategy--underscore the discomfort of the military with peacekeeping missions, humanitarian operations, and armed interventions short of war. The analysis that follows will necessarily conclude that the two concepts, rather than being complementary, are inversely related. The more narrowly defined the military mission in a complex emergency and the more rigorously mission creep is avoided, the more difficult it becomes to design an exit strategy that carries out the political objectives of American foreign policy. Conventional war, unconventional war, and operations short of war all share one common characteristic: they are, as Clausewitz reminds us, politics by other means. Some of the implied missions the military must undertake in complex humanitarian emergencies may appear political in nature, when in fact most military operations in conventional conflicts make political statements; those "statements" may be more subtle than in conventional conflicts but they are political nonetheless.

The first four of the challenges presented in this essay are intrinsic to the conflict itself; the remaining five deal with problems associated with external intervention.

Uncontrolled Population Movement

Nearly all of the nearly two dozen complex emergencies now in progress involve large-scale movements of civilian populations away from their homes in search of either food or security. In most cases between 25 and 50 percent of the populations of the countries have been driven from their homes. Two of the five emergencies into which the US military has been ordered--Goma, Zaire (1994), and Northern Iraq (1991)--have been refugee or displaced-person emergencies. Indeed, these two emergencies have involved the largest movements of civilian populations over a short time period in this century.

When people leave their homes they become exceptionally vulnerable to hunger and to epidemics of communicable disease. The traditional social safety net to which they turn when they are in trouble--extended families, ethnic or tribal group, and neighbors--collapses or is seriously impaired overnight in these tumultuous population movements.[2] No one leaves the safety and familiarity of home and neighborhood without good reason. All population movements of this magnitude have massive humanitarian relief implications: how are these people to survive without sanitation, clean water, food, health care, and shelter? The physical implications, which are evident to anyone in the relief discipline, also have profound political implications. Even in famines where there is no insecurity, people will move in search of food when their stocks are depleted: the great majority of the people who died in the Ethiopia famine of 1984-85 had left their villages. People usually move to a neighboring country which generally does not want them. However, under international humanitarian law to which most nations are signatories, they must be accepted under the doctrine of safe haven. These groups can be quite destabilizing politically to receiving countries. Their presence also may invite military action against the harboring state from the home countries they have just left.

When people move under these circumstances they invariably form squalid refugee or displaced-person camps which are in every respect undesirable. Disease in them is rampant, and sanitation and water--if available at all--are in short supply. Women and young girls are often the objects of terrible violence when they are separated from the men in their families. The camps themselves are sometimes the scene of violence from governments which don't want them there for political or military reasons. The camp residents become instant international paupers who can neither support nor protect themselves; they become wards of humanitarian agencies, sometimes for decades, at enormous cost.

The camps also become breeding grounds for political extremism as the young men in them have no useful work. These men quickly become demoralized and angry, ready recruits for militia armies and terrorist groups. People cannot return to normal lives if they are in camps. If elections in the home country are part of a political settlement, they will not be representative if half the population does not participate because they are in refugee camps in a neighboring country.[3]

These camps are a social, economic, and political nightmare--yet for the people who inhabit them, they sometimes provide a better alternative than the prospect of returning to their homes. Most important, the camps are protected under international law. Such camps should be avoided if at all possible, and if they must be created, they should be designed for use over as short a time as is feasible. The principle must always be that people will return to their homes as soon as they can safely do so.

Relief workers have learned from years of experience that humanitarian relief in a chaotic situation can measurably increase the rate of population displacement. Centers created by relief organizations to feed people who are already displaced will invariably attract many more people who were not originally displaced. Such centers act as huge magnets to insecure or hungry populations. Consequently, feeding centers are now used only as a last resort in relief operations. The preferred response is to bring food aid to people in their home villages if it is logistically possible and if security permits. In several recent emergencies, notably Somalia, US military and later UN force commanders were unwilling to protect convoys into remote areas to deliver food, which has meant that humanitarian agencies could not deliver it, precipitating population movements.

American military units can also attract refugee populations, particularly when they are deployed for humanitarian purposes, but sometimes even when they are engaged in combat operations during a conventional war. The phenomenon has occurred several times in recent years. In the immediate aftermath of the Gulf War, large population movements began toward the Kuwaiti border as Iraqis and foreign nationals sought refuge in allied-controlled territory. The operational plan for deploying US military units in Somalia increased the displaced population in Mogadishu by at least 25 percent: people came in from the countryside because they understandably thought they would be safer and better fed in an area controlled by US troops than in their own villages. General Aideed used this displaced population to increase the size of his militia army, recruiting among the unemployed and demoralized young men in these newly formed camps. In this sense the US military operations plan was a political and military blessing to the very warlord who tried later to kill American troops. Senior personnel from the Office of Foreign Disaster Assistance in the US Agency for International Development (USAID) presented alternative plans which proposed moving units simultaneously into nine cities, with a smaller presence in Mogadishu. For logistical reasons US Central Command chose a Mogadishu strategy. Thus an operational decision for entry into Somalia had profoundly undesirable strategic effects on the humanitarian, security, and political situations.[4]

In one of the most successful refugee emergency responses in the post-Cold War period, US military operations in Kurdistan successfully moved over a million Kurds out of the mountains and back to their home villages over a period of several months. Here a potentially devastating emergency with long-term implications was ended through the concentration of military and humanitarian resources with the deliberate strategic intention of creating the conditions necessary for the Kurds to feel safe enough to return to their home villages. Thus US military forces working with humanitarian agencies and USAID quickly reversed a population movement through a deliberate and well-coordinated strategy.[5]

Three operational principles apply here. First, avoid military actions that will encourage population movements and the subsequent creation of displaced camps; second, work with humanitarian relief organizations to develop a mix of incentives so people will not leave their home villages in the first place; and third, if camps are already formed, work with humanitarian relief groups--as the military did so successfully in Kurdistan--to return people voluntarily and as soon as practical to their homes.

Fragmented Political Authority

One of the distinguishing characteristics of all post-Cold War complex humanitarian emergencies has been the deterioration or complete collapse of political authority in the country in crisis. During the Cold War the superpowers--by their distribution of military, economic, and humanitarian assistance--discouraged more than two sides to any conflict while avoiding the direct involvement of either of the competing superpowers. Since 1989, however, political authority in many states has fragmented. Most internal conflicts now have more than two sides; some, such as in the Balkans, attract many interests in frequently changing combinations. During the war, warlords governed parts of Bosnian Serbia where Milosovich, Karadich, and Mladich have limited influence, a fact that complicates emergency response and the negotiation and implementation of political settlements. The warlords' power issues from their weapons, not from their political skills, nor through government services they might provide, nor by popular consent. While it would be an exaggeration to suggest they have no popular support (most of them have some following or they would not be warlords), they have no particular claim to legitimacy.

Most complex emergencies cannot be settled permanently without some sort of political agreement, yet political settlements become more elusive almost by mathematical necessity as the number and power of the warlords and other interested parties increase. The conduct of outside military and humanitarian operations can profoundly influence the power of the warlords and other non-state actors in any complex humanitarian emergency, as was the case with General Aideed and the displaced population in Mogadishu.

American commanders in complex emergencies tend to select one of two tactics for dealing with warlords: either work with them visibly and regularly, or ignore them completely. Either solution can create grave dangers to the accomplishment of the mission. Humanitarian relief organizations and commanders alike seek to facilitate their own work by trying to reach some accommodation with the warlords. The latter do have power; some conclude that it would be better to neutralize them (better still, make them allies) rather than prepare for a great deal of mischief from them. A US commander's interaction with warlords, whatever the seemingly practical reasons, can send an unfortunate and unintended message to the population: the US military (and therefore government) is supporting the warlords and so should you.

Commanders are more comfortable dealing with other military commanders; most warlords have some military background, which is why they are effective in using weapons and organizing men in paramilitary units in the first place. Dealing with them presents the danger of giving them legitimacy to the exclusion of other elements of civil society, such as traditional clan or tribal chiefs, religious leaders, professional groups such as doctors and teachers, the merchant class, and women's groups--all of which have a much greater claim than the warlords to legitimacy and form the foundation for the reconstitution of civil society. They are the very groups that must be nurtured, cultivated, and strengthened over the long term even as the power of the warlords is diminished.

However, if commanders refuse to deal with the warlords, misunderstandings about such matters as ground rules affecting security (including disarmament efforts), security incidents, and rumors could increase the threat to US forces and humanitarian workers. The military cannot simply assume away the problems related to direct contact with warlord militias.

Experience suggests a compromise between these two extremes that would be most prudent and ultimately offer some chance for success. Commanders must talk with warlords or they risk incidents, but these conversations could be held at subordinate staff levels rather than at the command level. Commanders themselves should meet regularly with the other elements of civil society which have some claim to political, social, or economic legitimacy, and those meetings should be visible and public. While none of these groups can formally claim to represent their society, they can give commanders some idea of current public opinion, however vague or imprecise it may be.

Commanders will inevitably have to meet with the warlord himself; in those circumstances, clan or tribal elders, or others who represent traditional authority structures, should always attend as well. Their opinions should be sought by the US commander with a deliberate show of consideration for each response. The rule I followed while managing the humanitarian relief efforts of the US government in Somalia in 1992 was this: meet with local militia or warlord figures only when the Somali clan elders (most of whom despised the warlords) were physically in the same room. The warlords did not like it, but they had no choice; this arrangement sent a strong message to the community about US government support for traditional authority figures.

The Economics of Chaos

Another abiding principle of complex emergencies is that invisible economic forces drive conflicts more than is commonly understood. Consequently, the merchant class--because they control both wealth and the marketplace--have political power that cannot be ignored. Most warlords have their own indigenous bankers and financiers. Even in chaos the economy of a country, however disrupted it may appear to be, functions at some primitive level. In the absence of a viable local currency, either foreign currency will be used as the medium of exchange (e.g., Deutschmarks in Bosnia or US dollars in Zaire) or there will be a flourishing barter system. The absence of a national government or a central bank is not an impediment to commerce; indeed, given the sorry condition of some national currencies in the developing world, that may even be an advantage: witness the stability and value of the Somali pound after three years with no national government.[6]

Under near anarchic conditions in a complex emergency, the aggregate wealth of the country invariably declines, sometimes precipitously, which means that outside resources introduced into the conflict take on increased value. What may have been of passing economic interest during stable and prosperous times takes on profound importance during unstable and economically depressed times. It is a principle common to most complex emergencies that when humanitarian relief agencies introduce computers, vehicles, food aid, and to a lesser degree pharmaceuticals to such an environment, they profoundly alter the economic calculus of political power in the country. When international military forces introduce their own vehicles, weapons, and other equipment, they attract this same attention; they, however, have the advantage of being able to protect what they own, something that NGOs can seldom do.

Food frequently replaces currency as the medium of exchange and becomes a mechanism for merchants and warlords alike to amass and hold wealth. All complex emergencies produce massive increases in the price and value of food as the established agricultural and economic systems of the country are disrupted. Food stocks become the bank accounts of merchants and warlords. Since food is the one common, immutable human necessity with economic value, particularly as it becomes a scarce commodity, it will invariably attract the attention of local, regular military forces (which tend to get paid irregularly during complex emergencies), warlords (who buy more weapons and the loyalty of their own militias and local population groups with looted food), and gangs of thugs (usually young, unemployed men who support themselves and their families through their looting, a kind of Robin Hood redistribution of wealth). It is not an exaggeration to suggest that those who control food aid possess a significant advantage in a civil war.

The strategy of the USAID's Office of Foreign Disaster Assistance (OFDA) in Somalia was to harness the power of the merchant class as a force for peace and stability. Fred Cuny, who designed this strategy, argued that food had become a source of instability in that the merchants were encouraging the looting of food convoys and NGO warehouses to supply their own markets. The OFDA strategy was to monetize food, that is to sell food assistance from donor governments to Somali merchants at an attractive price so that they had a reliable source of supply. OFDA reasoned that this policy would separate the merchants from the warlords. The monetization effort eventually did have this effect. It took a year and a half after the program started, though, for the merchants in Mogadishu from opposing clans to unite in the effort to separate themselves from the warlords and insist on more stability in the city so they could conduct trade. Commanders of international military forces were not asked to support this monetization scheme. Had they been requested to support it with logistics or security elements, attainment of strategic objectives might have accelerated and stability in the country might have been improved.

As much as commodities, the service contracts that humanitarian relief organizations and military forces negotiate with indigenous merchants can be of major economic consequence. Refuse disposal contracts, trucking, housing rental, fuel, food catering services for international personnel, and the services of local laborers can become temporarily the principal economic activities in the country. Control of these contracts creates heated political issues in complex emergencies, for the contracts can provide sustained cash flow with which to prolong the war effort. More than one relief organization and military force has unknowingly given such contracts to the most disruptive and violent of the indigenous factions, and the warlords who dominate them, thereby sustaining the very forces destroying the country. In chaos merchants either must purchase protection from anyone who can guarantee the security of their assets, usually the local warlord or regular indigenous military commander, or they must employ their own private armies. Determining which merchants have alliances with which factions is essential to avoid invoking the law of unintended consequences through service contracts.

Relief commodities such as food and service contracts for housing do create local jobs and, however temporarily, can stimulate the local economy. Local purchase of food in surplus areas for distribution to food insecurity areas by relief agencies makes good economic sense, particularly if the average farmer is the beneficiary. There are many good arguments for purchasing goods and services locally as long as the contracting agents understand the political implications of what they are doing.

Unintended consequences associated with purchasing and contracting decisions can be forestalled in several ways. Commanders (or relief agencies) should purchase goods and services locally only if they know whom they are subsidizing and have determined that their decisions will not contribute to the conflict. If external actors cannot do that, they should be as self- contained as possible during the early stages of the intervention, relying on external resources to support themselves. Another approach used in some conflicts by the International Committee of the Red Cross (ICRC), consistent with its ideological commitment to neutrality, is for contracts for goods and services purchased locally to be distributed as fairly as possible among all local factions and ethnic groups. This policy also states that all contracts will be suspended if conflict breaks out among these factions. The ICRC makes the defensible claim that conflict immeasurably complicates the management of their relief efforts and that they must suspend both these contracts and the relief effort because the violence necessitates it. Though the ICRC would be offended by the notion they are interfering in anyway in the conflict, including using their resources to promote peace and reconciliation, a larger political purpose is served by this tactic: it creates a set of strong incentives for the merchant class to support stability and peace.

The Problem of Retributive Violence

Physical devastation and human casualties are the most visible long-term consequences of complex humanitarian emergencies. There is, however, the potential for a much more damaging and politically significant outcome: the psychological trauma and emotional wounds carried for decades by the civil population whose members and families have been the object of atrocities. Given the general breakdown of the social order characteristic of complex emergencies, it is not surprising that the restraining influences of social mores and traditional authority structures also deteriorate under the stress of economic collapse and massive population movement. When restraining influences are relaxed, the darker longings of the human character grow more powerful. The human appetite for revenge is insatiable: each atrocity, real or rumored, by one group is followed by other, even more egregious, human rights abuses in retaliation for the original offense. The downward spiral accelerates, and the lives of the survivors are permanently scarred. The greater the scale and extent of atrocities which take place before the conflict ends, the more tenuous any political settlement will be. Skillful diplomacy and military force cannot erase these memories. In some societies, where the atrocities have reached sociopathic levels, peace may not be possible for generations. Preventing atrocities by separating populations that have been committing them against one another could become an implied mission of future military interventions.

The best way to avoid retributive violence is to deal with it through a carefully planned strategy. A commander could initiate an aggressive public affairs campaign using radio broadcasts and newspapers to warn against any resort to violence for revenge. Religious and political leaders could be encouraged to make similar broadcasts. One powerful way of reducing the frequency of these incidents is to encourage the establishment of a legal mechanism for the redress of grievances by those who have been victims themselves, whose family members have been killed, or who have witnessed atrocities. This will provide a sanctioned alternative for justice to be done which might otherwise have expressed itself in revenge killings. Finally, under no circumstances should at-risk populations be placed in displaced camps or closed-in areas to which there is limited access or from which escape would be difficult: these are the very areas where atrocities most frequently take place. Instead, these people should be encouraged to stay in their homes or in dispersed areas where they will not attract attention.

Atrocities, particularly revenge killings, are most likely to take place in the month following the conclusion of hostilities, when tempers are raw and people are furious as they contemplate what they have just been through. Atrocities are most likely to take place against those minority population groups, wealthy or poor, that cannot defend themselves and against whom there has traditionally been discrimination or hostility. The neighborhoods of these groups should be patrolled intensively; military presence will be welcome if it is made clear to neighborhood and community leaders, before the forces are deployed, why they are there. As a member of one of these patrols in Palestinian neighborhoods in Kuwait City in the month following its liberation, I was astonished at the enthusiasm with which my civil affairs unit was greeted by a group of people I thought would be quite hostile to any US military presence. Our effort was successful in restraining human rights abuses and is one of the unexamined successes of US military operations in Kuwait.[7]

The Problem of Strategy

The absence of comprehensive strategies for dealing with complex humanitarian emergencies is one reason that the international response to them has been so troubled and so often frustrated. The strategic void is a manifestation of the complexity of these emergencies; the absence of a set of shared objectives among the actors; the disparate disciplines required to respond to these emergencies, which have not in the past routinely had much operational contact with each other; and finally, the dysfunctional response system which has evolved to deal with these conflicts. Perhaps the most successful recent emergency response was in Northern Iraq, while the most troubling and potentially explosive one was in Goma. In the former, careful execution of a clear, long-term strategy allowed the return to some degree of normalcy among the Kurds. In the latter, the absence of a strategy led to the international community's becoming quartermaster to the Hutu militia, which carried out the genocide against the Tutsis in 1994 and which may well attempt more violence.[8]

Military mission statements and operational plans are no substitute for a comprehensive strategy because in every one of the cases, including Northern Iraq, these plans have failed to grasp what was taking place in the emergencies. Military operations in complex humanitarian emergencies are not combat operations or field exercises which have as their objective the destruction of the enemy's combat capability, the capture of a particular piece of territory, or even a show of force. One does not win a complex emergency response. Military force in such cases always supports non-combat objectives; it is never an end in itself. It is possible for the military mission to be completed and the campaign lost; it is equally possible for the military mission to be incomplete and the campaign won. No military mission statement or operational plan can ever substitute for a comprehensive strategy for a humanitarian intervention because it would not necessarily include humanitarian relief, developmental interventions, political negotiations, and diplomatic initiatives. The military portion of the response is only one part of a much larger effort.

Comprehensive strategies are needed in each emergency in order to change the course of the conflict for the better--that is, toward resolution--and to achieve some degree of normalcy for people caught in the conflict. If a political settlement is impossible or unlikely in the short term, a comprehensive strategy should at least ensure that the humanitarian response does not unintentionally exacerbate the conflict. The next section describes the objectives on which any complex emergency response must focus, and the following one explains how the coordinating structure within the intervention should be designed to implement a comprehensive strategy.

Setting and Measuring Objectives

How does the international community determine when its work in complex humanitarian emergency has been successful? How does one measure success? Because political leaders and military strategists alike have had great difficulty in designing defensible exit strategies in these emergencies, they have instead settled on arbitrary measurements, usually by setting fixed time deadlines as the United States has done in Bosnia.[9]

Conflicts end only when one of two conditions obtain: either one side wins a military victory or both sides accept a negotiated settlement. This proposition is as true in complex emergencies as in conventional war. Absent either of these two outcomes, the strategy for an intervention should focus on intermediate goals. The military does not need to stay until the conflict is over if reducing mortality and morbidity rates are the essential objectives. Each complex emergency is different, so proposing one strategy for all emergencies makes no sense. But it is possible to identify some general principles that can guide the development of a strategic design in most complex emergencies. Below are five measurable objectives which will return most societies to some degree of normalcy and self-sufficiency.

Of these five objectives the most difficult is the last one, not because of anything intrinsically unachievable but because none of the external actors--UN, NGOs, donor aid agencies, diplomats, or the military--will accept responsibility to support local efforts to create a sustainable system of maintaining order. It is the United Nations that should take on this duty, but commanders can support the effort logistically if needed.

Most of the objectives are achievable over the short term and can be tied to the agricultural harvest. Exit strategies should be linked to the achievement of these five objectives as part of a negotiated settlement. Furthermore, a proposed time-determined exit strategy--such as that for Bosnia--ought to include the achievement of these five objectives. Obviously many of the functions being measured within the objectives are not military in nature, but most do require security and logistical support. For example, displaced persons and refugees will not return to their homes unless they believe they will be safe during the journey and when they arrive home. The military can complement NGO and UN resettlement efforts by providing this security. All other objectives are dependent on this first one. If commanders regard this as mission creep, and refuse to carry out the implied mission, they will only postpone the reestablishment of normalcy in the social order of the society in crisis and increase the risk the intervention will fail over the long term.

The Problem of Coordination

The large number of independent external actors--NGOs, UN agencies, ICRC, donor aid agencies, and military forces--working in complex emergencies increases significantly the difficulty of designing and implementing a comprehensive strategy. When a large number of participants have a veto power, or even appreciable influence, over design or implementation decisions, several consequences follow. Anyone in the group can slow or stop either process, force endless compromises which will result in plans that reflect the lowest common denominator rather than address the real issues, and ensure that overlapping geographic and sectoral responsibilities continue unchanged. All of these consequences can lead to endless quarrels among the participants.

Most organizations involved in emergency response call for more coordination as a result of this anarchic organizational situation. Unfortunately, coordination means different things to different people. Even within an organization the term changes depending on how the organization's autonomy and agenda are affected by the quarrels mentioned earlier. Coordination is perhaps the most abused term in the relief vocabulary, yet most objective observers argue that some surrender of autonomy by all of the actors is essential to rationalize the planning and conduct of relief work in complex emergencies.

Commanders unfamiliar with the independent nature of the response structure may have some trouble adjusting to it--such independence violates military training and experience, every principle of war, every instinct of command. Commanders are not in charge of complex emergency operations; they are only one of many participants. Institutions such as the UN Department of Humanitarian Affairs, the European Community Humanitarian Office (ECHO), and the Office of Foreign Disaster Assistance in USAID can influence the other humanitarian groups, but influence is not command and control. The absence of central command authority does not prevent commanders from influencing other participants in the operation through persuasion, good leadership, and innovative ideas.

The most practical mechanism for ensuring that some coherent strategic design and planning does take place is the system of civil military operations centers (CMOC), developed to establish and maintain operational contact among the military and humanitarian participants in a complex operation. While the CMOC system was not intended for the purposes of strategic design, its operational usefulness is clear to most of the humanitarian agencies, a usefulness which could advance the notion of joint planning. Perhaps the most important effort that commanders could make to encourage the coordination of strategy and its execution would be to include a few representatives of the major humanitarian agencies in their operations planning process. It is here that some degree of coherent interagency coordination might take place. Commanders do not have the authority, expertise, or resources to correct the dysfunctional organizational response structure, but they can make an effort to contribute in a modest way to its solution through the CMOC and their own operations planning process.[10]

The Problem of Accountability

As suggested in the discussion of the economics of chaos, although the security situation in many countries distorts the functioning of markets, trade still goes on under changed rules. Regardless of how these distorted economic rules operate, relief food aid, supplies, and equipment will always have value in the economy. Supplies and equipment will always be of considerable political and security consequence in any humanitarian emergency.

The accountability of humanitarian agencies for relief resources has military implications and so must attract the attention of every commander. When warlords, gangs of criminals, or merchants can loot with impunity, insecurity in the country increases. All know that the stolen resources will inevitably be used to buy more weapons, the support of other factions, and the loyalty of irregular militia troops. In any society where order has deteriorated or collapsed, people will search for security, usually by acquiring weapons themselves or by allying themselves with those who already have weapons. For most regular military units, let alone irregular militia forces, pay is a tenuous matter; most of these units live off what they steal. Stolen relief resources can become the basis for a military pay system of sorts. The more weapons your faction has, the safer you and your family will be, the more likely you will get paid your "salary," and the more likely you will eat. And while there is a certain inevitability to this imperative in complex humanitarian emergencies, outside security forces can provide protection from much of this arbitrary violence.

When commanders are asked to provide security for relief operations they should understand that their reluctance to undertake it will put their own men and women at risk. Accountability is not simply a matter of protecting American taxpayer resources in complex emergencies, nor is it only a matter of ensuring that those who are hungry eat. It can become a contributing factor to stability (or instability) in the country, the likelihood of a political settlement, and the amount of weaponry available to unpredictable regular and irregular military units in the country.

The Problem of Neutrality

For many years it was an article of faith that humanitarian assistance was neutral, which is to say that it was provided solely on the basis of human need, regardless of the politics of the situation. During the Ethiopian famine of 1985, President Reagan suggested a dictum that is well known in the relief discipline: a hungry child knows no politics. While that was and remains true, a more recent corollary to the Reagan doctrine is even more consistently relevant: the child's parents frequently do know politics, and even if they don't, the manner in which the hungry are fed has profoundly political consequences. Consider the behavior of the current Ethiopian government, which defeated the Mengistu regime in 1991. It is now asking a chilling question of NGOs working in its country: did you feed hungry people during the Mengistu regime, which allowed the government to buy more weapons instead of more food, which they in turn used to kill more of our soldiers? By doing humanitarian work in a state controlled by a repressive regime, humanitarian agencies--NGO, UN, and ICRC--may be prolonging the life of these regimes, a profoundly political consequence of a supposedly neutral policy.

The US military helped to control the cholera epidemic in the Rwandan Hutu refugee camps in Zaire in the aftermath of the genocide against the Tutsis in 1994. Of the more than one million refugees in those camps, several hundred thousand are regular and irregular Hutu military forces which carried out the genocide. They are spread throughout the camps with their wives and children. Rumors abound that they intend to reenter Rwanda to take back what they lost when Tutsi-dominated political factions took control of the country. Their return, should it occur, will precipitate another bloodbath. The international community continues to provide assistance to the refugee camps in Goma. While intended to meet humanitarian needs which are quite real, the assistance may also provision the Hutu extremist elements which killed nearly a million people. Neutral intentions are not good enough if the consequences are anything but neutral.

This does not mean that humanitarian relief should not be provided, only that the actors in a complex emergency can no longer hide behind the veil of neutrality as a defense for their work when the consequence of their work is anything but neutral, noble aspirations notwithstanding. Some NGOs have abandoned any pretense of absolute neutrality in favor of the principle of independence. The latter holds that the NGO should make a deliberate decision to take a certain action even if it appears to favor one party to an internal conflict over the others. People and politics cannot be so easily separated.

Commanders need to be aware of both the risks and difficulties of neutrality, indeed its near impossibility in most complex emergencies. To the extent that the principle of neutrality can be followed by any of the actors, it certainly is desirable; it just may not be achievable. If neutrality is impossible, then one must determine which side one is favoring through any given action, and what the perception may be of all indigenous groups to that favoritism. In some circumstances neutrality is indefensible if, for example, one local group is trying to exterminate another, or other massive violations of basic human rights are being committed. Neutrality is an admirable, but not a necessary, condition of humanitarian operations. Indeed, it is increasingly unachievable in many complex humanitarian emergencies.


The instability that characterizes complex humanitarian emergencies is a growing phenomenon within the developing world and among former Eastern bloc countries. Humanitarian agencies and US military forces are increasingly being called upon to respond to such emergencies. Military commanders are confronting complex sets of strategic and operational issues with which they may feel somewhat unfamiliar. This article presents some of the problems of humanitarian response in complex emergencies. The issues discussed here are not new, nor were they unfamiliar to commanders during the Cold War or during the Allied occupations of Japan and Germany following World War II. They are issues with which the current generation of military officers lacks experience. Commanders need to be aware of these challenges, or they and the men and women they lead may fall victim to the law of unintended consequences which, during times of chaos, operates at exponential levels.


1. A complex humanitarian emergency may be defined as a national crisis in which: (1) political authority and public services deteriorate or completely collapse; (2) internal ethnic, tribal, or religious conflict occurs, with widespread atrocities against noncombatants; (3) massive population movements take place, with people escaping violence or searching for food; (4) widespread food insecurity appears, frequently deteriorating into starvation; (5) a public health emergency causes epidemics of communicable disease; and (6) the chaos leads to macro-economic collapse with massive unemployment, destruction of the currency, and negative GNP growth.

2. In the Hutu exodus from Rwanda to Goma, Zaire, after the genocide against the Tutsis in 1994, entire villages moved intact to the refugee camps.

3. If quantitative data were collected, I suspect it would show that the Palestinian refugee camps in Gaza and Lebanon produced most of the political extremists in the Palestinian political movements worldwide rather than the Palestinian Diaspora, a diaspora many times more populous than these camps.

4. Fred Cuny, at the time employed as an advisor to USAID's Office of Foreign Disaster Assistance, predicted this outcome in dispatches to Washington prior to the arrival of US troops.

5. See Carnegie Endowment for Peace papers on the Kurdish crisis in northern Iraq.

6. For a fuller analysis of this economic phenomenon, see my chapter, "Humanitarian Relief Intervention in Somalia: The Economics of Chaos," in Learning from Somalia: The Lessons of Armed Humanitarian Intervention, ed. Walter Clarke and Jeffrey Herbst (Princeton, N.J.: Princeton Univ. Press, 1996). The Somali pound continued to be used as the currency even in the absence of a national government or central bank. During this time, the informal market exchange rate between the United States dollar and the Somali pound remained quite stable.

7. For an in-depth description of this effort, see my chapter, "Preventing Human Rights Abuses in Kuwait," in Civil Affairs in the Persian Gulf War: A Symposium, Fort Bragg, N.C., 25-27 October 1991.

8. See the especially critical multi-donor evaluation of the international response to the Hutu refugee emergency in Goma, Zaire, in the fall of 1994, The International Response to Conflict and Genocide: Lessons from the Rwanda Experience, ed. David Millwood, 4 vols., Danish Ministry of Foreign Affairs, March 1996.

9. See Frederick M. Burkle, Andrew Natsios, et al., "Complex Humanitarian Emergencies: Measures of Effectiveness," Prehospital and Disaster Medicine, 10 (No. 1, January-March 1995).

10. Editor's note: Many observers and members of the relief community have noted the problems inherent in developing strategy for managing world response to a complex humanitarian emergency, establishing and measuring progress toward supporting objectives, and coordinating the activities of participants. Two experienced practitioners, Walter S. Clarke and Arthur E. Dewey, have recently circulated in draft a concept for organizing responses in such operations. Their paper, "A Humanitarian/Political/Military Partnership for `Total Asset' Planning in Complex Humanitarian Emergencies," April 1996, introduces the concept of a "comprehensive campaign plan" as a planning and coordinating mechanism.

Andrew S. Natsios (Lieutenant Colonel, USAR Ret.) is Vice President of World Vision U.S. and Executive Director of World Vision Relief and Development, the technical arm of World Vision U.S. From 1991 to 1993 he was Assistant Administrator of the Bureau of Food and Humanitarian Assistance in the US Agency for International Development; in that position he was appointed by President Bush as special coordinator of Somalia relief to manage US government relief efforts in the Somalia famine. In 1989-1991 he served as director of USAID's Office of Foreign Disaster Assistance. In his Army Reserve capacity as a civil affairs officer he served during the Gulf War in Riyadh, Saudi Arabia, and Kuwait City as executive officer of the Kuwait Task Force which planned for relief and reconstruction requirements post-liberation. In early 1993 he served on the Joint Staff at the Pentagon working on Somalia and Bosnia. He is a graduate of Georgetown University, holds an M.P.A. degree from Harvard, and from 1975 to 1987 was a member of the Massachusetts House of Representatives.

Reviewed 28 May 1996. Please send comments or corrections to