Over the past several years, the perception has become widespread that the world has entered a period of profound change. A main feature of this change has been some erosion of the principle of state sovereignty as a major structural feature of international relations. The new activism of the United Nations and the trend toward selective military intervention for humanitarian purposes and as a means of international crisis management have been the most prominent features of this development.
At the core of this historic reshaping of North-South relations lies the phenomenon of increasingly complex humanitarian emergencies. In relation to Africa, at least, the phenomenon of complex emergencies is related to three broad factors:
• problems associated with the formation and development of the post-colonial state;
• regionalization of the global economy and the consequent marginalization of Africa’s formal economy; and
• reassessment of the North’s political aid priorities with the end of the Cold War.
Famine has been one of the defining features of this phenomenon of complex humanitarian emergencies, and is increasingly understood as a consequence of internal wars; of environmental degradation stemming from economic marginalization; of household restructuring through migration; of ethnic competition; and, in extreme cases, of the use of physical force to seize subsistence assets. In Eritrea and in Tigray (northern Ethiopia), for instance, the Ethiopian government’s counterinsurgency strategy of directly targeting civilians was the central cause of widespread famine in the 1980s. In retrospect, the Western response to this crisis represents the first stirring of a Western agenda of humanitarian intervention in contemporary North-South relations. But the Western response was hesitant and uncertain, and developed only slowly. During much of this period, the international community was reluctant to challenge Ethiopian sovereignty.
The Emergency Relief Desk (ERD) was formed in 1981 as an ecumenical consortium of NGOs to facilitate cross-border humanitarian assistance from Sudan in to Eritrea and Tigray, where the majority of civilians in need lived in opposition-held territory. ERD was based on an agreement between the Sudan Council of Churches in Khartoum and Norwegian Church Aid. The agencies which subsequently joined the consortium were mostly European Protestant donor agencies, plus several from North America. ERD saw itself as a non-political “temporary instrument” to provide strictly humanitarian aid in-kind to the Eritrean Relief Association (ERA) and the Relief Society of Tigray (REST), and to verify its legitimate distribution in reports to the member agencies. Over the following decade, the ERD played a pivotal and often controversial role in channeling some $350 million of in-kind assistance to Eritrea and Tigray. The activity of the ERD spans the profound changes underway in North-South relations. Its history provides an opportunity to document this transition and to judge the present moment and direction of humanitarian politics. The best feature of ERD was its involvement with indigenous political and humanitarian structures and its steadfast commitment to the politically disadvantaged within the international arena. As such, its experience is at the forefront of the call for a new framework for humanitarian assistance in the coming era.
Sovereignty, Neutrality, Access
During the Cold War, the official deference accorded to state sovereignty was an important element in Western aid-related sponsorship of anti-communist alliances in the Third World. The effects of internal war were as real then as they are now, but donor states were inclined to disregard evidence of human rights abuses on the part of allies and potential client states while highlighting those of adversaries. The end of the Cold War has unraveled this web of relations, allowing the West to discern and criticize authoritarian practices, especially if, in a period of declining aid budgets, this enlightenment helps to rationalize a redefinition of strategic interests. Since the mid-1980s, the humanitarian consequences of internal war have thus been subjected to an unprecedented degree of official attention in the West, even though they have been a problem for decades.
The complex emergencies associated with internal war have been defined mainly in terms of problems of access to humanitarian relief. The acuteness of such problems in several African emergencies by the late 1980s helped precipitate increasingly critical appraisals of the principles of sovereignty and non-interference as enshrined in the UN Charter. These principles restrict humanitarian assistance to persons in areas under central government jurisdiction. In conditions of internal war or divided governance, this restriction made it difficult or impossible to assist persons in non-government areas.
Prior to the Gulf war in 1991, some NGOs had already begun to question the principle of nation-state sovereignty. The reaction of the US-led coalition to the situation in Iraqi Kurdistan, together with the subsequent US intervention in Somalia, has served to bring this debate fully into the public domain.
The growing involvement of NGOs in disaster relief since the mid-1980s has also had a profound effect on North-South relations. NGOs, by their nature and by their mandates, have the ability to form contracts with people rather than states. The enhanced role of NGOs in the South is a practical manifestation of the challenge to sovereignty which has emerged. NGO involvement in Africa’s internal wars has had the effect of bypassing local states and bringing the plight of civilians directly into the international arena. Many oppressed African groups are now better represented in Washington, London or Geneva than in Khartoum or Nairobi.
The ERD’s modus operandi was to work in partnership with indigenous relief agencies. During the 1970s, the Eritrean People’s Liberation Front (EPLF) and the Tigray People’s Liberation Front (TPLF) developed the political practice of linking mass mobilization with the provision of public welfare.  In the last analysis, the Sudan-Eritrea-Ethiopia cross-border operation could not have existed in the absence of the Fronts’ own welfare activities.
An important facilitating feature was the animosity between the governments of Sudan and Ethiopia, which accounted for Sudan’s tacit acceptance of the cross-border operation’s effective infringement of Ethiopian sovereignty. The several occasions when the border was temporarily closed indicates the extent to which Khartoum could have jeopardized the cross-border operation project had it seriously wished to do so.
The ERD mandate and its underlying assumptions provide a good example of the extent to which Western humanitarian politics revolve around a tension between neutrality — discreet, non-political and strictly humanitarian assistance — and involvement — a wider range of help including development assistance, institution-building and advocacy/lobbying. The sole goal of neutrality is to save lives, which often means working around indigenous organizations and requires establishing a distance from indigenous political relations. Involvement seeks to work through indigenous organizations on broader programs to reduce vulnerability. The weakening of sovereignty has seen the issue of involvement move nearer to the fore within NGO communities and to enter the realm of public debate.
During most of the 1980s, ERD strived for a neutral stance, and in terms of attracting reluctant donor support to a sensitive cause, its efforts must be judged a success. But ERD’s self-image demanded an ideological sleight of hand in its construction, for its neutrality was compromised by the one factor that accounted for its effectiveness: its reliance on the indigenous relief agencies, ERA and REST, which enjoyed the institutional and material support of the EPLF and the TPLF. ERD portrayed ERA and REST as neutral by emphasizing their efficiency and humanitarian credentials. ERD’s stress on neutrality helped shield cross-border operation donors from the consequences of acknowledging the plight of civilians in territory controlled by the Fronts. But the need to do so highlights a weakness of neutrality as a humanitarian approach, for it achieved donor support at the price of weak internal analysis. The debate over providing development assistance was never satisfactorily resolved by ERD.
Until the mid-1980s, the ERD defined itself in terms of meeting humanitarian needs “in parts of Ethiopia which are not accessible from areas controlled by the government of Ethiopia.” This formulation reflected the views on sovereignty prevailing at the time. The deference accorded sovereignty in all international institutions — it was simply not an issue — had a number of consequences. For one thing, it created conditions in which cross-border relief work was seen as “illegal,” giving a strong incentive for donor governments and agencies to take the lead from nationally based “legal” programs and institutions. This notion of cross-border operation illegality led those NGOs which did become involved to segregate the personnel involved within the institution. By accomplishing political distance through internal management means, the agencies greatly reduced internal discussion of the important issues raised by the cross-border work.
A significant turning point in ERD’s experience came after the international media attention to the Ethiopian famine in 1984. It was then that the US Agency for International Development and the European Community also began to support the cross-border operation. This, along with the increase in the number of NGOs involved in the cross-border operation, can be seen in hindsight as signaling the end of the Cold War in the Horn of Africa. By the end of the 1980s, disregard of Ethiopian sovereignty had become quite routine on the part of nearly all parties except, significantly, the UN.
Evolution of Donor Policy
Prior to 1983-1984, the great bulk of USAID food aid to the Horn was administered from Nairobi, most of it channeled to Catholic Relief Services and the World Food Program in Addis Ababa. No aid was provided to the CBO. As the scope of the famine in Ethiopia became known, political pressures grew in the US for greater assistance. Resolutions in the House and Senate supporting increased assistance, however, notably failed to recognize that significant territory in Tigray and Eritrea was inaccessible through Ethiopian government and UN relief channels.
Providing increased aid thus posed a thorny diplomatic problem for US policymakers, one compounded by the uncertain applicability of the so-called Reagan Doctrine, which called for support of rebel movements opposed to Soviet-supported regimes such as Afghanistan and Nicaragua. In the case of Ethiopia, however, the intent of the Reagan Doctrine confronted the communist agendas of the EPLF and the TPLF. Washington’s response to the growing Ethiopian crisis was therefore hedged with restraint, as the cross-border operation was not yet an acceptable alternative.
Washington’s reluctance to deal with the Ethiopian government, though, was at no point shared by the UN. In common with many Western bilateral donors, the major UN agencies (WFP, UNHCR, UNDP and UNICEF) increased their assistance during the late 1970s and early 1980s. But considerations of sovereignty kept UN humanitarian assistance from reaching areas not held by the government. Ethiopia’s rulers denied there were areas they did not control; the UN, by not challenging this, tacitly helped limit aid to such areas and created large disparities in the ratio of aid going into government as opposed to Front-controlled areas. It also placed the onus on ERD and other agencies engaged in the cross-border operation to legitimize their case. The UN, furthermore, made no attempt to incorporate information gathered by cross-border operation agencies with access to areas not controlled by the government. Since donor aid packages normally rely on UN assessments, this significantly distorted donor perceptions of need and access.
The events of 1984-1985 relaxed some of the caution surrounding cross-border work. The increasing involvement of strategic donors increased the diplomatic leverage of ERD member agencies, who themselves became more engaged in a media and political learning process. The term “cross-border” was increasingly used within ERD to describe its activities. The polarization of NGOs around this question also helped the ERD assume a more distinct identity: it became an international consortium working in Eritrea and Tigray, not, as before, one working in areas that could not be reached from the government side. With the decision to support the cross-border operation, the US became the largest donor in Ethiopia and in the non-government areas. In government-controlled Ethiopia, almost all US aid was channeled through NGOs, whereas other Western donors used the government’s Relief and Rehabilitation Commission. The US also launched an investigation of the role of government human rights violations in creating the famine, an issue which had been ignored and obscured by the UN and other large donors. (The 1985 investigation concluded that there was no “deliberate policy of starvation,” but failed to note the linkage between destroying the means of peasant subsistence through counterinsurgency campaigns and subsequently using food aid to lure people into government-held garrison towns.) ERD, through Lutheran World Relief, became the primary channel for US-supplied relief to non-government areas. EC aid, in addition to supporting ERD’s involvement in the cross-border operation, was also channeled through other cross-border operation agencies, such as Oxfam UK. From 1985 through 1990, USAID supplied on average more than half of all ERD’s relief assistance. Following the visit of then-Vice President George Bush to Khartoum in 1985, and his subsequent meeting with Ethiopia’s foreign minister, USAID assistance also went through Catholic Relief Services and World Vision to government-held towns, an initiative which was known as Food for the North.
The large increase in humanitarian assistance, combined with the violation of Ethiopian sovereignty, was not motivated primarily by considerations of geopolitical strategy, though such arguments were sometimes used to soften opposition to the aid. The complexity of the situation in Ethiopia and Eritrea is reflected in the conflicting agendas of US government institutions — the CIA, the National Security Council, the State Department and USAID. It does appear that public opinion helped push ethical imperatives to the fore, over ideological and geostrategic considerations. (One geostrategic consideration served by the policy was to staunch increased refugee flows into Sudan.) The US, and increasingly other Western donors, were ahead of the UN in assessing and responding to the humanitarian emergency in Ethiopia and Eritrea. Handcuffed by the issue of sovereignty, the UN Emergency Office for Africa, set up in 1984 to coordinate the burgeoning relief response, ignored the critical role of the cross-border operation and repeatedly concealed evidence about egregious and systematic human rights abuses by the Ethiopian government. It was not until 1990, and then with some difficulty, that the UN was drawn into aid negotiations involving non-government areas. There was no mechanism for communication between the ERD/cross-border operation and the UN-led operation on the government side until 1990, when the unraveling effects of the end of the Cold war were reflected in the reform of the EC’s emergency capacity with the establishment of the European Community Humanitarian Office (ECHO).
Internationalizing Public Welfare
These changing donor positions regarding the cross-border operation are related to a wider shift in relations linking donor and recipient governments and NGOs. In addition to helping ERD, USAID and the EC also assisted NGOs in Ethiopia on a scale not seen before. One reason is that donors regarded NGOs as giving them more control over the use and destination of their aid. This reflected a growing suspicion of indigenous political authority, stemming from a realization that the Ethiopian government was engaged in the widespread diversion and abuse of food aid.
The enhanced role of NGOs, in other words, has emerged out of a process of political struggle between donor governments and recipient states. International NGOs are mandated to contract with people, not states. The growing interest in human rights has been an important mechanism whereby NGOs have attempted to reflect popular interests at the global level. The donor NGO safety net systems which have developed over the last decade are, essentially, parallel constructions providing basic public welfare assistance with little or no recipient government control. This internationalization of public welfare denotes a weakening of sovereignty. It also undermines internal government capacity, and means that the international community has begun to replace indigenous material resources that had previously been directed to public welfare. In most instances, the donor NGO community also assumed managerial responsibility. This is why the ERD experience is significant: The public welfare programs remained under indigenous, non-governmental management. On a broader scale, donor policy was clearly contradictory after 1985, as the emergence of a new paradigm of humanitarian politics brought about a polarization among, and to some extent within, different agencies.
Recipient government attitudes toward donor NGO safety nets have understandably been ambivalent: they are an important source of resources, but they represent an erosion of authority. Ethiopia’s rulers quickly learned how to use NGOs within its political program. In Sudan, though, a more antagonistic relationship developed. In the political flux that followed the fall of Jaafar Numayri’s regime in 1985, some 80 NGOs established themselves in Sudan. By 1986, the NGO presence had become a major domestic issue, especially as some NGOs attempted to provide humanitarian relief in southern Sudan and to publicize the plight of civilian populations there. It was in southern Sudan that UN policy began to reflect the wider changes in humanitarian politics. In 1989, the UN began coordinating a large-scale cross-border operation from Kenya called Operation Lifeline, based on a free-passage agreement negotiated with the government in Khartoum and the main armed opposition body, the Sudanese People’s Liberation Army (SPLA).
It is important to note that USAID and EC aid to Eritrea and Tigray via the cross-border operation continued to respect publicly the formalities of sovereignty. The USAID program was subsumed under the heading of “Ethiopian Refugees” and administered from Khartoum. The EC also maintained cross-border operation support as part of its Sudan program until 1989. In reality, though, Ethiopian sovereignty was ignored, while the Fronts gained implicit recognition for their control of areas not reached through government channels.
The ERD experience demonstrates that effective relief in a complex emergency demands the full participation and cooperation of local political parties and institutions. This is an unremarkable observation, but NGOs and other donors appear to forget it time and again, perhaps because it invites them to examine the quality of the relationship between rulers and ruled in countries or situations in which they operate. For many, this is a step too far, yet it lies at the heart of the role of indigenous NGOs and international institution-building pretensions. In complex emergencies, the nature of internal political relations becomes paramount. Where they are of a predatory or problematic character, NGOs must come to terms with the task of developing new forms of involvement.
Internal War, External Aid
Distinct from earlier NGO-led operations, for example in Nigeria and Cambodia, officially backed cross-border aid to Eritrea and Tigray can be seen as the first international humanitarian intervention of the post-Cold war era. The West had begun its hesitant and confused journey to Kurdistan, Somalia and Bosnia.
It is no coincidence that sovereignty has become less sacrosanct at a time when increasing numbers of people have been affected by protracted, complex emergencies. The policy and resource implications of this trend have pitched the international relief system into turmoil. ERD, in both its internal and external relations, represented a unique effort to cope with the resulting contradictions.
One question that the ERD member organizations never fully resolved is the extent to which ERD’s relief assistance fueled or prolonged the war in the Horn. Any attempt to resolve this question must begin with an appreciation of the nature of internal war, a form of total war which targets the cultural, commercial and economic infrastructures of opposing populations. An intrinsic feature of such a conflict is a systematic and widespread violation of human rights. It is impossible to be neutral within the logic of internal war, a war whose destructive consequences are aimed precisely at disrupting the lives of the people whom humanitarian aid seeks to sustain.
If the bottom line is to protect lives and livelihoods, perhaps the question should be whether ERD’s channeling of assistance through local political structures maximized its effect, regardless of the motives of those local structures. The answer — unambiguously affirmative in the case of ERD — indicates how important the character of indigenous political structures is to the provision of effective relief. NGOs, by depoliticizing situations, often overlook this. This is clear from the attempt to replicate in southern Sudan an ERD-type of relationship with the Sudan Relief and Rehabilitation Association (SRRA), established in 1985 as the humanitarian wing of the SPLA. But the SRRA could not fulfill NGO expectations that it would be a fledgling ERA or REST. This is because the SPLA, lacking a public welfare mandate, has functioned in a predatory manner to transform the sociopolitical system in southern Sudan by increasing imbalances among ethnic and other social groups. The NGOs involved insisted on seeing the SRRA as capable of acting independently of its political environment, by transposing a construct of neutrality onto the SRRA. In southern Sudan, unlike in Eritrea and Tigray, the dominant political entity, the SPLA, could not support this.
In the last analysis, the success of the ERD and the cross-border operation depended on the political practice of the Fronts. ERD’s capacity to help appears to be more the result of historic association, or even accident, rather than an ability to analyze the situation correctly. Most complex emergencies are much less accommodating.
ERD’s experience was unique and cannot be narrowly reproduced. Effective NGO consortia such as ERD emerge only when the emergencies provide a minimal humanitarian platform on which agencies of markedly different philosophies and approaches can cooperate. Most successful NGO consortia have been relief-oriented, as opposed to development-oriented, precisely for that reason. In the past, the “illegality” of many relief situations demanded the comparative advantage that only NGO consortia were able to provide.
UN and other military interventions in support of humanitarian aims mark a new departure, notwithstanding their uneven and problematic nature. A more interventionist climate has altered the terms of NGO engagement. The high-profile relief operations in Kurdistan and Somalia, for instance, have worked against the formation of NGO consortia. While cases of dire international neglect will continue to exist, they will be cast alongside, and forced to compete with, large-scale operations in conflict zones that attempt to provide relief to all sides using individual NGOs as subcontractors.
The Horn of Africa provides the contrast between the decentralized and participatory welfare structures established in Eritrea and Tigray and the more typical centralized, non-participatory structures that were operative in Ethiopia. Why have the exemplary practices of the former not supplanted the more common practices? One reason may be that the latter serve as a bridgehead for the extension of foreign management into a society. Centralized and non-participatory systems, moreover, provide a means of maintaining political distance. Paradoxically, the erosion of sovereignty appears to have strengthened the need for political distance at the same time as humanitarianism has become more interventionist. In Kurdistan, Somalia and Bosnia, the international response has been to insert non-participatory and non-accountable institutions of conventional relief, along with armed protection. Political distance has been achieved by defining this form of humanitarianism as the only possible type of intervention. But military humanitarianism, distrustful of indigenous political relations and institutions, cannot provide a long-term solution.
The manner in which internal wars are fought dictates that emergency assistance cannot help but affect the dynamics of the conflict. Acceptance that all relief has an impact on internal wars would free an agency to make rational decisions based on analysis of the actual situation. The period ahead is going to require more, not less, collective action by NGOs. Apart from the many cases of neglected and abused peoples, there are the problems thrown up by ill-conceived and short-term interventionism, coupled with donor government aversion to long-term rehabilitation and reconstruction programs.
The emergence of military humanitarianism following the Gulf war fostered the notion that the world was entering a more humanitarian age. Now it appears that Somalia and Bosnia mark watersheds in any recourse to military humanitarianism as credible solutions. For one thing, such interventionism has been extremely uneven. Why Somalia and not southern Sudan, for instance? Secondly, intervention has run into the hard realities of internal war in Somalia and Bosnia. Western nations appear to have learned that intervention, if it goes too far, carries the danger of becoming responsible for long-term reconstruction. The difficulties of policing internal wars and the high cost of reconstruction are keeping the West out of all but the most unavoidable (or most expedient) emergencies, and even then to limit engagement to humanitarian relief, traditionally defined.
The evidence of Somalia and Bosnia, moreover, shows that Western states and the UN still respond to complex emergencies by developing “neutral” institutions of humanitarian intervention which are then protected, if necessary, by military force. Enforcement is a difficult and vexed question; what we can insist on is that this change of emphasis has highlighted the political and structural nature of contemporary emergencies. “Neutral” intervention avoids engagement with the political reality it confronts. It eschews the need for supporting participatory and accountable structures and institutions, and arguably makes matters worse.
 An important distinction between the two movements is that the EPLF saw itself as leading a war of independence or national liberation, while the TPLF was one of several regional opposition movements fighting for greater autonomy within a liberated and decentralized Ethiopia.