GLOBAL DIALOGUE Volume 7 ● Number 1–2 ● Winter/Spring 2005—Humanitarian Intervention

The Complexity of Military Intervention in Humanitarian Crises

JAMES F. MISKEL

James F. Miskel is a professor of national security affairs and associate dean of academics at the US Naval War College in Newport, Rhode Island.

here can hardly be more disturbing images than those painted in Sudan since 2004, in the Democratic Republic of Congo (DR Congo) from the mid-1990s, or in Rwanda in 1994. In each of those instances—and unfortunately others—large numbers of innocent people were, or are being, killed, maimed, and driven from their homes, farms and places of business. Such abuses call out for action to protect prospective victims and prevent future suffering, if not to bring the perpetrators to justice. But as Sudan reminds us, this call is not always answered, and even when it is, effective military action is often delayed until after the violence has crested and most of the damage has already been done.

It would be a mistake to attribute these delays to callousness on the part of leaders or voters in North American and European states towards the suffering of people in other parts of the world. The phenomenon of delayed intervention or non-intervention is too widespread to have been caused by the attitudes of individual statesmen or swings in volatile public opinion. Indeed, the very fact that delayed military action and inaction are the rule rather than the exception suggests that the explanations involve systemic or structural features of humanitarian response. Very broadly speaking, the structure consists of three tiers or levels: diplomatic pressure, economic pressure/inducements, and military intervention. Regarding the second tier, trade sanctions are an example of economic pressure; foreign aid and disaster-relief donations are examples of economic inducements.

Weighing the Risks

From the perspective of a responding state—for example, a North American or European state considering its options with regard to a humanitarian crisis in Africa—the costs, risks and responsibilities are dramatically lower for the first two tiers or levels than they are for the third. The costs and risks associated with military intervention will be discussed later in this essay. The point here is that these costs and risks are not only much greater than they are for diplomatic and economic gestures, but also that the magnitude of the cost differential is increasingly visible to voters and their elected representatives in national legislatures.

Media coverage of humanitarian suffering is widely acknowledged as having had a motivating influence on military relief efforts in Somalia, East Timor, and—albeit late—Rwanda. Television and newspaper coverage informed opinion leaders and the general public about the extent of the suffering and gradually helped create political support for more forceful action than speeches and donations. There is, however, is another side to the coin of media influence. Press and television coverage of violence directed at peacekeepers, relief workers, and even United Nations officials has taught the general public in many countries lurid lessons about the risks and costs that can be incurred by military forces that intervene in humanitarian crises. That much of the widely publicised violence in post-war Afghanistan and Iraq has been directed against local government officials, aid workers and military personnel engaged in peacekeeping or “nation-building” activities only amplifies the public’s wariness about military humanitarian intervention.

Since the United Nations and United States intervened and then withdrew from Somalia in the early 1990s, it has been generally assumed that television news, in particular “twenty-four hour” media outlets such as CNN, can influence government policies in the United States and other advanced democracies by mobilising public sympathy for the victims of a humanitarian emergency. It is possible that this effect may have been reversed. Intensive, vivid and repetitious news media coverage of violence against peacekeepers and aid workers seems to be offsetting sympathy for victims with concerns about costs and risks.

Overriding the public’s wariness in the name of humanitarian principles would not only be bad politics for elected officials, it would also conflict with one of the structural principles of democracy: that the people should have a voice in decisions about how their government spends the nation’s most precious resources. This principle deserves respect, even if it does sometimes result in tardy and/or inadequate action by the United Nations and its leading member states.

Indeed, respect for this principle is what causes believers in democracy everywhere to accept the judgement of the French and German electorates which, polls suggested, opposed the invasion of Iraq in 2003, as well as the judgement of Spanish voters in choosing a prime minister who would withdraw Spanish forces from peacekeeping duties in Iraq in 2004. Wisely or not, the citizens of France, Germany and Spain judged the likely risks and costs of military engagement in Iraq—even in post-war peacekeeping and nation-building—to be greater than the uncertain benefits.

The same calculation of costs and benefits is made by elected officials who, being politicians, keep their fingers on the voter’s pulse with respect to the level of response (diplomatic, economic, or military) for humanitarian emergencies in places like Rwanda, DR Congo, and Sudan. As a result, even in situations where military intervention is widely acknowledged as the most effective option, diplomatic pressure (essentially statements of concern, displeasure or outrage by governments and international organisations) and economic sanctions or donations of cash and surplus food remain the most common responses.

While voters may be primarily concerned about a specific set of risks and costs—the risk of intervention forces becoming entangled in a civil war from which withdrawal is politically and strategically difficult, and the cost in lives of the members of the intervention force—there are additional factors that need to be and increasingly are considered by national and UN leaders. These factors involve the practical aspects of military humanitarian interventions.

Early Intervention

After the 1994 genocide in Rwanda, the “intervention community”¹ concluded that intervention would be more effective and less risky/costly if it were undertaken in the early stages of a crisis. Very simplistically, the theory was that an ounce of pre-emption in the form of early military intervention would not only nip an emerging crisis in the bud, it would be less risky and costly than the pound of intervention that might be necessary later. Thus, one way to minimise the practical complexities of intervention and in so doing sharply reduce its risks and costs would be to intervene before violence becomes too widespread. This is, of course, the same logic that underpins the Bush administration’s much criticised doctrine of military pre-emption, and it may be that criticism of the doctrine will have the side effect of dampening whatever enthusiasm remains for pre-emptive military humanitarian intervention.

The “early warning, early action” lesson from Rwanda in 1994 seems all the more compelling in that there had actually been a specific moment early in the crisis when the relevant UN military commander saw an opportunity to intervene and requested UN approval for action. The commander, Major-General Romeo Dallaire of Canada, was the head of the UN Assistance Mission for Rwanda (UNAMIR).

UNAMIR had been in Rwanda for six months, monitoring compliance with agreements that ended an earlier period of strife, when it became aware that there was an opportunity to disrupt preparations being made for violence against the country’s minority Tutsi population. Dallaire believed he would need only five thousand soldiers to complete the mission, but as it happened, the UN leadership took no action on his request and the plans of the conspirators were not disrupted. Shortly thereafter, a massive campaign of violence against the Tutsis was launched and hundreds of thousands of people were brutally killed. To many in the intervention community, the genocide that followed the United Nations’ inaction is clear evidence that UNAMIR should have been given the reinforcements and authorisation it had requested.

However, it is a far cry from regretting an apparently missed opportunity to concluding that a reinforced UNAMIR would have been able to roll back permanently the tide of ethnic violence. While it is likely that a surgical strike against a group of conspirators could have been mounted and that this might have made a difference in the short term, it is unrealistic to think it would have done more than merely postpone the outbreak of violence.

It is now conceded that the genocidal violence in Rwanda was not the result of agitation by a small number of firebrands or street gangs. It was instead the outgrowth of an organised political movement, “Hutu Power”, which had connections to the government and had been investing for years in the mobilisation of anti-Tutsi sentiment through “hate radio” broadcasts, intemperate newspapers and agit-prop statements by community leaders. A UN force of 5,000–6,000 men would surely have been able to capture some of the Hutu Power conspirators, but many of the most important ringleaders and agitators were local officials or enjoyed government protection and thus were in practical terms immune from UN detainment.

The small UN force would also presumably have succeeded in two other tasks that UNAMIR leaders envisaged for it: confiscation of hit lists of designated victims that conspirators had prepared and seizure of some weapons caches. These actions would not, however, have permanently derailed the Hutu Power movement, nor would they have undone the effects of its anti-Tutsi propaganda campaign. Hit lists are easy to recreate and weapon caches can be quickly refilled. Indeed, unless a reinforced UNAMIR had stayed in-country for an extended period of time and assumed law-enforcement and border-control functions, the hypothetical intervening force would have done nothing to prevent new ringleaders from emerging, new hit lists from being written, and new weapons from being acquired and stored in caches.

Controlling Weapons

The weapons that figured most heavily in Rwanda’s genocide—machetes, pistols and torches—were low-tech, inexpensive and already widely available in-country. Weapons such as these can be easily replaced unless there are vigorous and sustained controls over imports. Rwanda is a small country, but control over cross-border weapons traffic would certainly have been beyond the combined capacity of the Rwandan government in 1994 and a UN force of five thousand with other, peacekeeping (as opposed to peacemaking), responsibilities in the interior. Moreover, given the influence of the Hutu Power movement on the government, it is hard to image that the latter would have co-operated in such a forfeiture of sovereignty. Even if it had, its officials would not have aggressively implemented measures designed by the United Nations to prevent small arms from entering the country and finding their way into the hands of extremist Hutu groups.

Sealing the borders and seizing a weapons cache do not, in any event, reduce the number of weapons that are already in circulation. In Rwanda, the weaponry was low-tech and widely available. This is not unusual. In many countries across the world there are already more than enough weapons in the hands of warlord militias, street gangs, and individuals to stoke the flames of violence. The Small Arms Survey, a Geneva-based independent research group, estimated in its 2003 report that there are approximately thirty million firearms in sub-Saharan Africa, and that more than two-thirds of those weapons are in non-government hands. There may be no cause-and-effect relationship between private-sector gun ownership and humanitarian crises: guns are apparently more widely available in some stable countries such as the United States than in the volatile sub-Saharan region. But the presence of large numbers of weapons in a polarised or anarchic society such as that of Rwanda, Somalia, Sudan, and DR Congo is obviously a major contributor to wide-scale violence. It should also be a major contributor to concerns about the risks that intervening peacekeepers may face.

There are two ways of reducing the number of weapons circulating in any society. The weapons can be either turned in voluntarily (for example through buy-back programmes) or confiscated. Both approaches were tried in Somalia by international peacekeepers, with very limited success.

Buy-back programmes often yield only damaged or obsolete weapons that armed groups might not have used anyway. With respect to involuntarily disarming individuals and groups, one of the lessons learned by peacekeepers in Somalia is that attempts to confiscate sidearms are resisted, sometimes forcefully, and for practical reasons. In a climate of civil disorder or intra-state conflict, such as existed in Somalia and exists today in Iraq, Afghanistan, the West Bank and Gaza, and DR Congo, individuals and armed groups know that as soon as their weapons are gone, they will be more vulnerable to violence by other individuals and factions. Gun-control and weapon-confiscation programmes under such conditions alter the local balance of power and necessarily have the effect of picking sides in a conflict even when the intervening force has a policy of neutrality.

When there is a functioning government that is complicit in the violence, controlling and confiscating weapons is an even less practical option. Complicit governments can hardly be expected to administer weapon-control programmes on an even-handed basis. Nor should they be expected to do anything other than subvert efforts to disarm their client groups. Opposition groups understand these dynamics and will, therefore, resist weapons confiscation. Moreover, dysfunctional as the Rwandan government in 1994 was, it was nevertheless still recognised as legitimate by the United Nations. Overriding Rwanda’s right of sovereignty in early 1994—before the genocide—to enable the United Nations to impose and administer gun-control and disarmament regimes would have been politically unrealistic and practically impossible, despite Rwanda’s small size and compact geography.

Geographical Constraints

If an intervention force employed pre-emptively would ultimately have made little difference in a small country like Rwanda, the prospects for enduring success must be even dimmer in humanitarian emergencies that occur in larger countries. Indeed, the operational challenges in large countries are so much more demanding that they may explain why the United Nations and its leading members have repeatedly declined to take military action on the basis of reliable early warnings of a crisis ever since they claimed to have learned the “early warning, early action” lesson. Sudan and DR Congo are two recent examples.

After many months of indisputable evidence of massive human rights violations and widespread violence against civilians in Sudan, the only actions that have been taken are low-risk and low-cost: expressions of concern by diplomats (although not all diplomats—Arab League states have been noticeably reticent about criticising fellow league member Sudan) and the donation of relief supplies. As of mid-August 2004, the only “military” intervention had been by the African Union, which sent a contingent of soldiers into the troubled Darfur region to protect the organisation’s ceasefire monitors from the continuing violence there.

Early warning information about repeated crises in DR Congo (formerly Zaire) in 1995, 1996, 1997, 2002 and 2003 was also widely available, but early action was not taken. The United Nations finally commissioned a limited military intervention in DR Congo’s easternmost provinces on 30 May 2003 after at least two years of Security Council resolutions that expressed alarm and concern over continuing instability and violence. There had been a UN peacekeeping force (MONUC) in DR Congo since December 1999, but its mission, like UNAMIR’s, was to help promote resolution of a prior conflict and it was neither authorised nor equipped to take military action in defence of the civilian population.

DR Congo and Sudan illustrate important points about military intervention in countries that are not as compact (few are) as Rwanda. DR Congo and Sudan are each approximately one hundred times as large as Rwanda. Sudan’s Darfur region alone is twenty times bigger than the whole of Rwanda. Other states afflicted by humanitarian emergencies also greatly exceed Rwanda in terms of geographic extent. Afghanistan contains more than twenty-five times as much territory as Rwanda. Even the relatively small west African states of Liberia and Sierra Leone are respectively four and three times as large as Rwanda.

With regard to suppressing violence and establishing civil order, the size of the territory matters a great deal. Conducting and maintaining military operations over large geographic areas and for the extended periods of time required to make a lasting difference on the ground is logistically daunting, particularly in parts of the world with inadequate infrastructure. It is, of course, an unfortunate fact of life that humanitarian emergencies often erupt in countries with underdeveloped infrastructures—limited or decrepit rail, road, and air transportation facilities, supply depots, and communications networks.

As a result, intervening military forces have to “come in heavy”, loaded with their own logistics and transportation support. The bigger the force and the wider the expanse of territory over which it must operate, the more numerous and dispersed its logistics sites must be. These sites include more than improvised motor pools and supply depots. They include military headquarters and sub-headquarters, sleeping facilities for the soldiers, field kitchens, and field hospitals. There are in addition quasi-headquarters where the military and relief agencies gather to co-ordinate operations and share information. For example, in Iraq there are twenty-one such “civil–military operation centres”. Extending protection to such a large number of widely dispersed headquarters and logistics sites obviously adds considerably to the cost and complexity of maintaining a military intervention force in a country. It also highlights the costs and risks that the media, voters and legislatures think of when they consider whether the response to an overseas humanitarian emergency should include military intervention.

Ethnicity

In a sense, DR Congo and Sudan also present more of an analytical challenge than did Rwanda. There were only two main ethnic groups in Rwanda and together they accounted for almost 99 per cent of the population. Almost 85 per cent were Hutu and almost 15 per cent were Tutsi. Moreover, about 95 per cent of the population professed Christianity as their religion. Quite obviously, this relative homogeneity did not mitigate the bitterness between Hutus and Tutsis. But in comparison to the heterogeneous ethnic and cultural composition of some countries, the situation in Rwanda was relatively uncomplicated. There are, for example, four hundred different ethnic groups in DR Congo. Not all of these were directly involved on a continuing basis in the violence in eastern DR Congo, but all were in one way or another bound to be affected by its outcome and many could have been directly affected by a foreign military intervention. Conducting military and political operations in such a complex setting with due respect for the rights and sensitivities of the various groups is a very challenging endeavour and, if mishandled, could inadvertently provoke groups that are not directly involved in the fighting.

Moreover, the ethnic affiliations and sensitivities of the various groups that might be affected by a humanitarian intervention often have international implications. This has been a particular concern in humanitarian crisis areas in Central Asia and sub-Saharan Africa where colonial boundaries do not follow traditional tribal lines, where single ethnic groups occupy both sides of some borders, and where other borders have substantial refugee camps and expatriate communities adjacent to them. The result is that humanitarian emergencies sometimes evolve into international disputes with diplomatic and occasionally military ramifications.

For example, Tutsi expatriates living in Uganda launched an organised invasion of Rwanda after the genocide started in the summer of 1994. Afterwards, large numbers of Hutu refugees assembled in camps in eastern DR Congo (then Zaire), instigating conflict among ethnically related Congolese groups and also staging cross-border reprisal raids into Rwanda. Eventually, this agitation blended with broader currents of discontent with the Zaire national government and a civil war ensued. Internal chaos, human rights abuses, cross-border raids by Rwandan refugee groups and Zairian factions into neighbouring countries, and the lure of Zaire’s immense mineral wealth led to incursions by the armed forces of Rwanda, Uganda, and eventually Chad, Tanzania, Angola, Namibia, and Zimbabwe. Obviously, military humanitarian intervention in a regional war or during a period when regional war appears likely to occur would be extremely high risk, politically and militarily.

Outsiders and Loot

A common cross-border aspect of humanitarian crises is the sponsorship of local subversive or opposition groups by neighbouring states or factions. Rwanda and Uganda both sponsored violence by groups in DR Congo—often groups with which they had ethnic ties—as a way of destabilising and weakening the Congolese regime. Military humanitarian intervention in DR Congo while indigenous factions were being actively supported by Rwanda and Uganda could very well have put peacekeepers into politically controversial no-win situations. Peacekeepers could have been drawn into conflict with violent factions made intransigent by Rwandan or Ugandan support. Support from neighbouring governments enables rebel groups to sustain operations for longer than they would be able to do unaided. The groups may also be able to use the territory of a neighbouring state as a base of operations or as a refuge. Either way, the result is to complicate greatly the task of military interveners.

DR Congo is far from being the only humanitarian emergency that has been exacerbated by third-party intervention. Other examples in recent years include Afghanistan, Angola, Bosnia, Iraq, Liberia, Chechnya, Kosovo, Sierra Leone, and the West Bank and Gaza.

The presence of valuable natural resources can be another complicating factor in humanitarian emergency situations. In some countries, armed groups have been able to take control of easily exploited natural resources such as hardwood timber, alluvial diamonds that can be dug up with hand shovels, gold nuggets that can be picked by hand from river gravel, opium poppies, and coca crops. Resources such as these were not a factor in Rwanda, but they have figured prominently in a number of humanitarian emergencies in Africa, South America and Central Asia. So-called blood diamonds played such a major role in the financing of Sierra Leone’s time of troubles that international controls have been placed (with limited effect) on the export of diamonds from the country. Diamonds, timber and coltan (a mineral used in computer chips and cell phones) were also a source of revenue for warring internal and external groups in DR Congo. Opium poppies and coca plants have funded warlord violence in Afghanistan and Colombia respectively.

When exploitable natural resources such as these are controlled by armed groups, the challenge of military humanitarian intervention is complicated in several ways. First, earnings from the export or smuggling of diamonds, gold, timber and drugs provide funds that can be used by violent groups to arm themselves, recruit adherents or hire security in the form of subsidiary gangs. This obviously increases the risk of casualties for an intervening military force. Second, participation in criminal enterprises inevitably leads to alliances with criminal organisations. The alliances between political rebels and drug cartels in Colombia are an example. Such alliances can have the same types of effect as military alliances among states. They can increase the size and geographic distribution of the groups that need to be brought under control in order to resolve the humanitarian emergency. Third, actually bringing violent groups under control may become substantially more difficult when the profit motive has been added to their reasons for fighting. Humanitarian emergencies are usually characterised by severe poverty and a lack of economic opportunity. Control over natural resources gives violent groups economic power they may not be willing to relinquish even if the political or ethnic grievances that initially sparked the crisis were to be resolved.

Planning and Preparation

Like any military operation, military humanitarian intervention into violent situations requires careful planning. In some humanitarian emergencies, the planning effort can be informed and accelerated by foreign military professionals already in the country concerned—for example UNAMIR in Rwanda and MONUC in DR Congo. UN peacekeepers can collect and analyse a variety of militarily useful information. They can assess features of the urban and rural terrain, the tactical situation at potential points of entry for military interveners, the organisational discipline, strategies, and cross-border relationships of the various armed factions, and the identity of faction leaders and their relationship to each other and to the government. When, as often, there is no advance element to help address such issues, planning may have to start from scratch and this will necessarily entail some degree of intellectual fiddling by military and political leaders while the humanitarian fires burn. During this planning phase the hypothetical opportunity for early action will probably slip by, but the complexity of the operations is such that it would be militarily foolish and politically reckless to intervene without having first examined the facts on the ground.

In August 2000, the United Nations commissioned a Report of the Panel on United Nations Peace Operations, also known as the Brahimi report, to identify things the world body could do to prepare better for military humanitarian intervention and other peace operations. The report’s findings constitute a call for greater political and financial support for the United Nations and its military humanitarian intervention missions. Indeed, the practical improvements that the Brahimi report recommended would be literally impossible without a substantial increase in such support from the United Nations’ leading member states.

The recommendations of the Brahimi report include expanding the size and increasing the expertise of UN military planning staffs; recruiting civilian experts in law enforcement and other functions essential to successful intervention; adequate Security Council resourcing for authorised peace operations; more realistic rules of engagement for intervening military forces; greater investment in the training and equipping of peace operation forces; negotiating standby agreements with national military units that might be on call for peace operations; improving the specificity of Security Council resolutions; and reforming the overall UN personnel system. The breadth of the report’s recommendations is testimony to the complexity, risks, and costs of military humanitarian intervention.

If the United Nations were actually to implement the Brahimi panel’s recommendations it would indeed be better prepared to deal with the practical aspects of intervention described above. However, problem recognition is not always the first step towards a comprehensive solution. Sometimes the problems can seem so complex and their solutions so costly that recognition is demoralising and can lead to inaction. So far, inaction seems to have been the response to recognition of the complexity of military humanitarian intervention. The United Nations’ leading members often have too few national interests at stake in humanitarian emergencies to enable them to persuade domestic constituencies that the costs and risks of intervention are worth taking, or that the investments in upgrading the world body’s capabilities are more important than other demands on national resources.

Endnotes

1. “Intervention community” is used here as a shorthand term for the United Nations, the UN member states that typically participate in humanitarian interventions, and the non-governmental organisations that typically advocate intervention and/or deliver relief services.