In the modern history of humanitarian action dating from civilian relief during the Second World War, never before has the legitimacy of the enterprise been so profoundly and publicly challenged, while at the same time never have the services of humanitarian organizations been more in demand. Many of the strongest critics are humanitarians themselves, or their longtime boosters in academia and journalism, who have witnessed the dirty little wars that spawn large-scale famine, massive human rights violations, and forced migration. 
That such a crisis of conscience is striking the humanitarian movement just now cannot readily be attributed to an unprecedented scale of suffering. Without minimizing recent horrors in Somalia, Bosnia, Rwanda and Sierra Leone, nothing in the last decade has exceeded the sheer scale of famine in China, genocide in Cambodia, or forced migration in Afghanistan during the Cold War. The faltering faith of global rescuers is rooted in other factors. First, as the Cold War wound down a string of successes including mediated resolution of wars in El Salvador and Namibia, and armed rescue of displaced Kurds in Iraq after the Gulf War raised expectations to unrealistically high levels. Unshackled by the Cold War, it was believed, the combined powers of humanitarianism, diplomacy, and multilateral military prowess would create a world that was not only more orderly, but also more fair to the weakest groups. When diplomatic and military efforts miscarried, as in Somalia and Bosnia, or were withheld, as in Rwanda, the s econd factor came into play. Confused Western governments threw humanitarian organizations into the front lines of these crises, where the agencies could bring no decisive solution, but where they did have greater access than ever before to witness and report atrocities in real time. Finally, some of the humanitarians began to discover and publicize how their own presence in the war zones became incorporated, and morally implicated, in warrior tactics of violence.
Each humanitarian organization has been chastened by unique experiences. In eastern Zaire during 1995, for example, camps for a million Rwandan refugees were controlled by the same ethnic militias who had carried out the 1994 genocide in Rwanda, and who were skimming international aid to rearm in preparation for returning to Rwanda to complete the genocide. Among the multitude of nongovernmental organizations (NGOs) providing aid in these camps, one agency attempted to evade the system of militia control by recruiting its own team of young refugee "scouts" to distribute food aid on the basis of need. The scheme seemed to work, until one scout had an altercation with a militia member. International NGO staff stood by helplessly as the scouts were assassinated one by one, and the aid program again fell under the sway of the militias.
Searing experiences such as this have provoked the humanitarian community to undertake an unprecedented project of self-scrutiny, and analysis of the military and political contexts in which they find themselves. With the major exception of former Yugoslavia, most of the cases of spectacular humanitarian failure have come out of Africa, where states are weakest, interests of the major powers are most peripheral, and perpetrators of violence appear to be least amenable to international leverage.
Largely missing from this humanitarian examination of conscience has been penetrating analysis of how strategies of warfare have shifted since 1990, particularly in Africa. The findings of such analysis may not be reassuring, however. When analyzed strategically, much humanitarian action appears not only to fuel particular wars, but also to help constitute the international architecture that institutionalizes the most irresponsible warrior strategies.
The Changing Face of War
For four decades, from independence until the late 1990s, the predominant form of war in Africa was internal conflict. Regardless of the purpose or ideology for which the war was fought, it most often took the form of citizens of a single country fighting each other on their own territory.  Only since 1996 have a significant number of states been drawn into interstate warfare--initiated, ironically, by some of America's best friends in the region, the "New Leaders" of Ethiopia, Eritrea, Rwanda, and Uganda.
War and mercy in Africa have been closely linked since independence. The predominant strategies of internal conflict have shifted from military coup in the 1960s, to protracted war in the 1970s and 1980s, to warlordism in the 1990s. In addition, the practices of humanitarians have evolved in tandem with these warrior strategies to reinforce them in hidden ways.
During the first decade of independent rule in the 1960s, African leaders were highly vulnerable to overthrow by riots in the streets or rebellion in the barracks. Between 1965 and 1970, there were 50 coups across the continent. Recognizing these threats, African leaders learned how to protect their personal safety by importing a security apparatus from a friendly major power, and how to control the capital city by employing some combination of co-optation and coercion: typically, subsidized food prices and ruthless policing.
Driven out of the cities, armed opposition groups desperately sought alternate strategies. Many turned to some form of Maoist insurgency--protracted war from a rural base. Using this strategy, a revolution could begin with only a small band of fighters cultivating political support among peasants in the countryside. The next step was to provoke the central state to overplay its advantage in the tools of coercion. By launching hit-and-run attacks against government military and economic targets, and then blending back into the peasantry, insurgent strategy was designed to call down a hail of government violence on the heads of its own peasant supporters. The net political effect, the insurgents hoped, would be to drive the peasants into their hands. Because they began from a position of weakness, insurgents intentionally protracted war over years or decades in order to buy time to build themselves up and wear the government down.
To a would-be revolutionary, the appeal of protracted war was its adaptability to a variety of ideological colors and political purposes. In Africa, it served successfully in anticolonial struggles against France in Algeria, against Portugal in Angola, Mozambique, and Guinea-Bissau, and against the white government of Rhodesia. Ethiopia, Sudan, and Somalia sponsored insurgencies against each other in the Horn of Africa. The African National Congress attempted to use insurgency against the apartheid regime of South Africa but failed to establish a territorial base within the country. Instead, South African Defense Forces turned the tables by sponsoring insurgencies against the front-line states of Angola and Mozambique that supported the ANG. Insurgency has always required an ideological component. In spite of the Maoist genealogy of the strategy, however, it need not be drawn from the political left. Nationalist, anticommunist, or ethnic ideologies would often do.
The strategic face-off between rebel insurgents and government counterinsurgents produced the distinctive features of war in Africa during the 1970s and 1980s: rural locus of violence, large-scale famine, and massive population...