Military involvement in refugee relief operations has undergone a remarkable evolution during the past decade, from providing logistical support to aid organisations in Kurdistan in 1991 to leading relief efforts for Kosovan refugees in 1999. Some aid organisations have welcomed this development, and increasing attention is being paid to issues of civil-military cooperation.

However, although few would contest that military forces possess logistical capacities unmatched in the aid community, important questions remain as to the appropriateness of an increased military presence beside humanitarian organisations in the field.

First, the motivation of the military is different from that of humanitarian organisations, even if the intervention is couched in "humanitarian" terms. Humanitarian action is premised on the equal worth of all human beings, yet military interventions since Somalia have been selectively undertaken by governments with direct national interests: the French in Rwanda, the USA in Haiti, the Russians in Georgia, the Australians in East Timor, NATO governments in Kosovo, the Nigerians in Liberia and the British in Sierra Leone. Conflicts that pose no threat to powerful nations, either through security concerns, lost investments, or potential refugee flows, are largely overlooked, despite the human misery they generate.

The massive offensive undertaken in defence of Kosovar refugees contrasts starkly with the cynical indifference shown towards Sierra Leonean and Liberian refugees under siege from rebel forces in Guinea today. Can we accept that the lives of some human beings are worth more than the lives of others? Second, outside military forces are rarely perceived as impartial in conflicts, compromising the image, and hence the effectiveness, of aid organisations that associate with them. Few aid organisations will accept an escort from the UN peacekeeping force in Sierra Leone since its belligerent stance against the Revolutionary United Front (RUF) hinders access to civilians in RUF-held areas.

Moreover, civilian lives are put at risk through mixing humanitarian and military actions. The presence of NATO troops in Kosovan refugee camps undermined the civilian and humanitarian character of the camps, and those in northern Albania were shelled by the Yugoslav forces as a consequence.

Third, the military lacks the technical competence to respond to the needs of refugee populations. Military forces are trained and equipped to provide medical care and facilities to a predominately male, adult, healthy population. Many of the essential medicines used in emergency settings, such as oral rehydration salts and vaccines, are lacking in sufficient quantity in military supplies, and facilities are not adapted to the needs of refugees.

The French army hospital in Goma in 1994, for example, provided excellent care to some refugees, but given the scale of the cholera epidemic that began soon after their arrival (some 50 000 deaths in a matter of weeks), it was an inappropriate use of resources. Instead, the allocation of one helicopter to transport potable water could have alleviated the supply problem caused by the congestion of roads with refugees. But the most serious shortcoming of military involvements in relief operations of the past decade does not concern what they do, but what they do not do.

Protection from violence is the most vital need of refugee and displaced populations today, and is a task that humanitarian organisations are unable to assume. Yet most military forces have been deployed with a humanitarian mandate aimed at providing or protecting relief supplies.

This mandate gives governments an image of doing something, to appease public outcry, while avoiding engagement in potentially dangerous or protracted conflicts. In Goma, the military fought Vibrio cholerae while the Rwandan leaders and army responsible for the 1994 genocide installed themselves in the refugee camps in full view of the military contingents present.

As a consequence, the refugee camps were attacked by Rwandan government and rebel forces 2 years later, and 200 000 refugees remain missing to this day. In Somalia and Bosnia, the military were tasked with protecting aid convoys. But the provision of humanitarian aid is a means to an end, the end being the preservation of life and dignity.

Although insecurity can prevent aid reaching vulnerable populations, the deployment of military forces to protect the means in isolation of the ends is a dangerous travesty. A full belly does not provide civilians with protection. What is the point of protecting the aid supplies when the civilians it is intended to assist are in greater danger of losing their lives to violence? The most appalling consequence of the limited mandate is the false sense of security it provides to civilian populations.

In Kigali, Kibeho, and Srebrenica, troops stood by helplessly and witnessed the slaughter of civilians because their mandate did not extend to such a role. Aid organisations have called for military intervention in the past and no doubt will do so again in the future. But such calls are for political, not humanitarian action. This is the area in which the military can complement humanitarian activites, if the political will can be mustered to assume such a role.