• Send page
  • Print page

The role of humanitarian issues in international politics in the 1990s

31-03-1999 Article, International Review of the Red Cross, No. 833, by Adam Roberts

 Adam Roberts is Montague Burton Professor of International Relations at Oxford University, and a Fellow of Balliol College, Oxford, UK. This article is based on a paper presented at the 3rd International Security Forum (Zurich, 19 - 21 October 1998).  


In the 1990s, humanitarian issues have played a historically unprecedented role in international politics. They have been prominent mainly in relation to armed conflicts and the use of armed force, rather than other types of disaster whether natural or man-made. They have been frequently cited by States and international organizations as a basis for action regarding both civil and international wars. The experience of this expanded and largely unforeseen role raises questions for theorists and practitioners alike. It poses an intellectual challenge for those who believe that international politics is essentially about States’ pursuit of power on the basis of interest; it also raises questions for those who believe that international humanitarian law and action can have a major impact, not just on particular conflicts and their effects, but on international politics generally.

There is nothing new about humanitarian issues featuring significantly in the conduct of international politics. In earlier decades of this century they played an important part in shaping popular and governmental reactions in many countries to such events as the two World Wars, the Spanish Civil War, and the Vietnam War. However, the way in which , in the 1990s, they have repeatedly had a central role, in relation to a rapid succession of crises from the Kurdish areas of Iraq in 1991 to Kosovo in 1998, has no obvious parallel in any previous period. Humanitarian issues may now be at the beginning of a phase of retrenchment, but they are not likely to disappear.

This paper does not attempt to provide an overall balance sheet of the considerable achievements and many failures of humanitarian approaches in the 1990s. Its purposes are more limited. The first section briefly discusses the changed role of international humanitarian law and action since 1989. The second explores the factors, some of which are rooted in considerations of power and interest, that have led to this increased role. The third goes on to outline six key areas in which the emphasis on humanitarian issues has resulted in innovations, all involving the United Nations. The final section suggests conclusions arising from experiences of the 1990s.

 The changed role of humanitarian issues since 1989  

The events of the 1990s have put into prominence both of the interconnected streams of international humanitarianism that are applicable to the problems posed by armed force: (1) the law of war, especially those aspects which seek to protect the victims of armed conflict (international humanitarian law); and (2) humanitarian action in the form of assistance to people in need during an emergency situation. Both these streams are related to, but in some ways distinct from, law and action in the field of human rights, not directly addressed here.

 Law of war  

Since 1990 there has been a heightened interest worldwide in the law of armed confli ct, also known as international humanitarian law. A number of factors have contributed to this development: (1) The law of war was seen to have played a positive role in the 1991 Gulf War. Although its application was not free from controversy, it did provide useful guidelines for the conduct of coalition military action. (2) Several conflicts, and the reporting of them, have stimulated widespread public awareness of war crimes. (3) There has been growing concern about the huge extent of pointless destruction caused by anti-personnel mines.

The considerable attention paid to law of war issues by the anti-Iraq coalition in the conflict over Kuwait in 1990-91 was evidence not only of the salience of these issues, but also of the compatibility of humanitarian law with military effectiveness. Indeed, attention to aspects of the law governing armed conflicts probably made a positive overall contribution to the coalition’s choice of methods and tactics, and contributed to the revival of serious interest in this law among the armed forces of several coalition States.

At the end of the 1990s the overall picture of State participation in existing treaties on the law of war is far from bleak. Adherence to the Protocol I of 1977 (on international armed conflicts) additional to the 1949 Geneva Conventions illustrates the point. On 1 July 1988 there were 76 States Parties. At the end of 1998 there were 152, including all NATO members except France (which is considering ratification), Turkey and the USA.

In addition, between 1994 and 1997 four new agreements relating to the law of war were concluded which, having passed the necessary threshold number of parties, have entered into force. The first, not exactly part of the law of war as such but with an important bearing on the conduct of hostilities, is the 1994 UN Convention on the Safety of United Nations and Associated Personnel. This treaty was concluded in haste by the UN because of t he lack of respect for peace-keepers and other international workers in a number of conflicts in the immediately preceding years, including in the former Yugoslavia. The very need for this treaty thus casts a sombre light on the role of humanitarian issues in the 1990s. Two new Protocols annexed to the 1980 Convention on Certain Conventional Weapons were concluded, addressing the use of blinding laser weapons (Protocol IV, 1995), and mines, booby-traps and other devices (Protocol II as amended, 1996). The latter was rapidly followed by the fourth of the new treaties, the 1997 Ottawa Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on Their Destruction.

On some issues, the 1990s have seen remarkable developments in the form, not of new treaties, but rather of authoritative expositions of existing law. Two striking examples, both drawing on a wide range of sources and relating them to today’s circumstances, are the San Remo Manual on International Law Applicable to Armed Conflicts at Sea (1994), and the ICRC/UN General Assembly Guidelines for Military Manuals and Instructions on the Protection of the Environment in Time of Armed Conflict (1994). In several areas, drawing up authoritative expositions of this type is more useful than trying to negotiate completely new agreements.

However, the main developments of the 1990s have been the result of a new international emphasis on implementation: on actions to ensure that the law is observed and violations are punished. In the 1990s, as in earlier periods, implementation has been different in practice from what is envisaged in the conventions. Trials for “grave breaches” under the 1949 Geneva Conventions have remained a rarity, and the International Fact-Finding Commission set up in 1991/92 under the terms of Additional Protocol I has not been asked to do anything. On the other hand, the UN Security Council has established the International Criminal Trib unals for the former Yugoslavia (1993) and Rwanda (1994). Although they have encountered many difficulties, especially in arresting suspects and in getting witnesses to testify, they are evidence of a growing seriousness in the international community about implementing the law of war. Finally, 1998 saw the adoption of the Statute for the International Criminal Court.

An issue of particular concern is the policy of the United States. Despite the experiences of 1990-91 and the care with which it has approached such tasks as the writing of military manuals and rules of engagement, the USA is not formally a participant in most of the main law of war agreements of recent decades. In particular, the USA is not a party to Additional Protocol I; the 1995 Protocol on laser weapons; the 1996 Protocol on anti-personnel mines; the 1997 Ottawa Convention on anti-personnel landmines; and the 1998 Statute of the International Criminal Court. Although a non-party, it takes at least some of these accords more seriously than some States that are parties. The reasons for US non-participation go far beyond the obduracy of one single aged Senator, and call for careful analysis rather than uncomprehending condemnation.

 Humanitarian action  

The striking increase in the work performed by humanitarian bodies in relation to armed conflicts has been reflected in the increases in the budgets and field activities of the Office of the UN High Commissioner for Refugees (UNHCR), the World Food Programme (WFP), the European Community Humanitarian Office (ECHO), the International Committee of the Red Cross (ICRC), and countless NGOs; and in the increased involvement of UN and other peace-keeping forces in humanitarian action.

The expansion of UNHCR expenditure from the mid-1970s illustrates the point. The agency’s budget, which depends primarily on voluntary donatio ns by States, grew from 69 million US dollars in 1975 to $570 m. in 1989, $1,307 m. in 1993, to a peak of $1,430 m. in 1996. Since then there has been significant retrenchment: expenditure in 1997 was $1,220 m., and the figures for 1998 and 1999 seem certain to be under $1,000 m. Large staff cuts are under way. Nonetheless, the huge expansion of the budget up to 1996, and its maintenance thereafter at a level far higher than in earlier decades, are evidence that States have been willing to put serious resources into taking action about at least some refugee problems.

 Factors contributing to the centraly of humanitarian issues  

What exactly has caused the remarkable increase in the centrality of humanitarian issues in international diplomacy in the 1990s? There are four main levels of explanation. They relate respectively to political belief systems, the role of the media, the interests of States, and the particular pressures of multilateral decision-making.

 Humanitarianism and political beliefs  

The end of the Cold War meant that the struggle between Western-style liberalism and Soviet-style communism, and all the variations and permutations of that struggle, ceased to have their old hold (which was in any case far from complete) on the minds of men and women around the world. The battle had been largely won by Western liberal and democratic ideas, but these did not themselves offer a guide to how to respond to many of the wars and civil wars of the post-Cold-War era. Humanitarianism could supply one possible basis for a response: a supplement to, if not a substitute for, a political ideology of liberal democracy.

Humanitarianism gained particular importance as a response to the many conflicts in which religious or ethnic identity was the app arent basis for murderous warfare. To proclaim the irrelevance of the minor differences between Serb and Croat, or Catholic and Protestant, or Muslim and Jew, was itself to say something important about the necessity for tolerance as a basis for human relations. This still left open, and unanswered, the hard question of where such humanitarianism leads: it is extremely difficult to make people live together when they have deep mutual antipathy, and security fears that are perfectly genuine. The problems of implementing the 1995 Dayton provisions for the return of refugees are evidence of the limits of efforts to treat all people as essentially equal on a largely humanitarian basis.

 Media reporting of humanitarian crises  

The capacity of the media (newspapers, radio and television) to report disasters as they happen has had profound effects on the actions of individuals and States. This is not just because the media can provide vivid information, but also because the very immediacy with which they do so serves as a reminder that nowhere on earth is beyond the reach of a response. In an age of instant mass communication, when some response to tragedy is often expected by informed and at the same time helpless citizens of States not directly involved, humanitarian action has become a natural part of the response.

Although media reporting is most often connected with the provision of humanitarian assistance, it has also contributed to the developing role of the law of war, or international humanitarian law. A fundamental reason for the emphasis in the 1990s on this body of law has been the scale, frequency, and vivid reporting of hideous war crimes. What is the point of having rules of restraint if they can be ignored with apparent impunity? Indeed, what is the point of calling ourselves civilized if we do nothing about cruelty and barbarism even when it is on our do orstep?

 Humanitarian approaches as an easy way out for States  

A third reason for the growing role of humanitarian issues is more sinister. When governments around the world are confronted with demands to take action about situations of extreme horror, humanitarian approaches are often the easiest way out. For example, faced with the prospect of massive refugee flows, taking some humanitarian action in or close to the country of origin (for example, by assisting in the maintenance of refugee camps) may be politically preferable to permitting refugees to settle permanently in one’s own country. Supporting action by an NGO or an international organization may help to associate a donor government with good works, while reducing the risk to itself should things go wrong. Setting up an international criminal procedure may be less costly than sending troops in to stop the atrocities.

 Tendency of institutions to agree on lowest common denominator  

A fourth reason for the emphasis on humanitarian issues is that this is an age marked by a strong tendency of States to act multilaterally through international institutions. When States with different perspectives, interests, fears and capabilities meet to discuss a particular crisis, it is easier for them to agree on an impartial humanitarian approach than to decide on a substantive policy to resolve the conflict. Time and again – over Somalia, the former Yugoslavia, Rwanda, and other crises – the UN Security Council could agree on humanitarian action even when it had no proposal as to how to end the troubles that had made such action necessary. Further, it could agree on a relatively low-risk strategy, when it would not have been able to agree on the much riskier course of military intervention to bring about a particular outcome.

 Conclusion on the four factors  

Major events have multiple and complex causes. There is no case for being reductionist and assuming that it is necessary to choose between the four types of causes (briefly outlined above) for the increased interest in humanitarian law and action. The last two factors in particular cast doubt on the idea that there is a fundamental departure from the system of sovereign States and power politics. None of the factors identified above as contributing to the increased interest in humanitarian law and action looks merely temporary. Even if the tendency towards humanitarian approaches to crises declines (and there is evidence that it is doing so), it will not disappear.

 Six UN-related developments  

The emphasis on humanitarian action in the 1990s has contained some new elements. Six are singled out here. Remarkably, the UN system, including some of the specialized agencies, has played a critically important role in respect of all of them.

The single greatest change in the international handling of humanitarian issues in the 1980s and 1990s has been the extensive involvement of UN bodies. Such involvement is not wholly new. The League of Nations Covenant stated in Article 25 that the members would encourage Red Cross activities. The UN Department of Humanitarian Affairs (set up in March 1992) and its successor have taken much further the process of encompassing humanitarian action within the ambit of the major international organization.

Humanitarian issues have featured especially frequently in resolutions of the UN Security Council. The word “humanitarian”, used only very seldom in such resolutions in earlier decades, has occurred with unprecedented frequency in resolutions from 1990 on wards. These resolutions have addressed both humanitarian law (calling for its observance in ongoing wars, condemning violations, and authorizing action intended to stop violations) and humanitarian action (providing for emergency relief during wars, and for various humanitarian services).

A particular problem of UN involvement in humanitarian action is that the UN may have some degree of responsibility for the creation of humanitarian problems in the first place. This is most clearly so in the case of international economic sanctions, the use of which has expanded hugely in the 1990s. Affecting ordinary inhabitants (including children) more seriously than governments and armed forces, economic sanctions have turned the principles of the law of war on their head. They have also necessitated massive UN programmes (most notably in Iraq) to try to alleviate their cruellest and most pointless consequences.

The UN involvement in humanitarian issues has led to major failures as well as some impressive achievements. Not all of its many distinct elements have been coordinated with each other and, even when this has been the case, they have not necessarily worked harmoniously together. They are examined separately below.

 Facing the challenge of civil wars  

Civil wars, including internationalized ones, have long been a principal form of conflict. In the past there were many efforts to bring humanitarian law and action to bear on such wars. This was, for example, a major part of the international response to the Spanish Civil War (1936-9). However, the extent of the international community’s humanitarian involvement in the civil wars of the 1990s has been unprecedented.

In the past, the law of war was conceived as applying, according to its terms, primarily to international armed conflicts between recognized States . Civil wars, especially those involving an ethnic or inter-communal dimension, have historically been the most difficult type of war to bring under legal or moral restraint. The very aims of parties in civil wars of this type often include trying to get part of the population out of the territory, and thus involve violations of fundamental principles of the law of war. In such circumstances, the crucial distinction between soldier and civilian is ignored; and the parties are particularly suspicious of humanitarian assistance activities.

Such civil wars call into question the idea, associated with the name of Jean-Jacques Rousseau, that war is only and exclusively a conflict between governments and their armies, and not at all a conflict between peoples. This idea has been one important philosophical basis of the development of humanitarian limitations in war. Where there is no acceptance of the fundamental principle of respect for civilians and non-combatants, it cannot be expected that there will be respect for humanitarian law, efforts and agencies.

In the 1990s, as a result of the UN’s preoccupation with tackling civil wars and other cases of internal violence, peace-keeping forces have been dispatched, safety zones and refugee camps established, and sanctions and embargoes instituted. Certain new legal developments (especially the establishment of the Yugoslavia and Rwanda tribunals, and the 1997 anti-personnel landmines treaty) bring conflicts with a strong element of civil war within the formal ambit of a wide range of rules and implementation procedures. This emphasis on humanitarian responses to civil wars has given rise to some acute problems. Among the most serious have been problems of providing protection for: (1) civilians, prisoners and other potential victims; (2) humanitarian workers (expatriate and local); and (3) humanitarian activities. These problems, which have many dimensions, are explored at several points below.

There has also been concern that, at least in some civil wars, the availability of humanitarian assistance might actually prolong the hostilities. On rare occasions this might be a desirable outcome, if for example it enabled an attacked party to gain time to build up its forces. More often, any prolongation of a war is far from being an intended or acceptable consequence of humanitarian aid. Yet wherever there is a situation of near-anarchy and a large quantity of aid, it is inevitable that a proportion of that aid will fall into the hands of belligerents or feed their supporters. As a result, they may be saved from realizing the true cost and pain of the wars to which they are parties. Worse, many aid organizations will seek to conceal these embarrassing facts from their donors and political supporters at home. This issue has arisen with particular force in relation to some African conflicts, and has contributed to the retrenchment that has characterized approaches to humanitarian action since the high point of around 1995.

 Establishing international tribunals as a response to war crimes  

The establishment of the Yugoslavia and Rwanda tribunals was widely, and rightly, hailed as a landmark in the history of the law of armed conflict. Yet in these instances the UN Security Council reached its decisions partly on the basis of the lowest common denominator. Both these tribunals were created at times when ongoing violence and atrocities, widely reported in the media, had caused public concern and demands for action; when the international community had failed to agree on any strong measures, or indeed any definite policy with regard to the crisis; and when there was notable reluctance in many countries to risk their soldiers’ lives in a cause in which they were not directly involved. The tribunals were in part a substitute for action.

The moves of the 1990s towards more systematic implementation of the law of war reached their logical conclusion with the opening for signature of the Rome Statute of the International Criminal Court on 17 July 1998. This instrument, which in one form or another had been under discussion at the UN for 50 years, appears to provide for a new system of international justice which could overcome some of the obvious weaknesses of relying on national courts or on ad hoc international tribunals. Further, the Statute breaks new ground in its definition of crimes, and in its application to civil wars. The ICC is envisaged as having jurisdiction over the crime of genocide, crimes against humanity, war crimes, and potentially also the yet-to-be-defined crime of aggression. However, early entry into force of the Rome Statute is improbable, because no fewer than 60 States Parties (an unusually high number) are required to bring it into effect; and many States are nervous because of the Statute’s potential effects on their political or military leaders. At the Rome negotiations in June-July 1998 many States (and not just the USA) were concerned to limit the Court’s powers of investigation, arrest and prosecution, in order to save their own military personnel or political leaders from being exposed to the risk of trial.

In addition to the international tribunals which have received so much attention, other methods of ensuring respect for the law of war have continued to be important in the 1990s. These include: (1) trials and court martials by national courts, which remain a principal form of legal implementation; (2) administrative action and internal inquiries by governments, a case in point being the episode in which the Canadian Airborne Regiment, whose members had committed crimes in Somalia during UN operations, was disbanded on 5 March 1995; (3) military action by way of response to violations, as for example when crimes committed by Serb forces in Srebrenica and Sarajevo in 1995 led to NATO’s Operation Deli berate Force bombing campaign against Serb military targets in Bosnia; and (4) refusal by third countries to offer a government any support, even when it has run into trouble, because it is perceived to have been involved in outrages.

The international tribunals mark an important advance in the implementation of international norms, but they face serious problems: in particular, there remains an unavoidable element of chance as to who is tried and who is not; and the Rwanda tribunal in particular has suffered from interference with witnesses and poor administration. So far, the only operative international tribunals have been set up in respect of countries within which there were serious grounds for doubt about the possibilities of fair trial. It is too soon to say whether the old and inadequate forms of implementation of the law of war have been decisively supplemented by something new and more effective.

 Transformation of the refugee regime  

A third key development of recent years has been the tendency to tackle refugee-producing situations at or near the source. Developments of the 1980s and 1990s, especially the increase in refugee numbers and the actions by States to prevent inflows of immigrants, have led to significant changes in the international handling of refugee issues. The roles of UNHCR and the UN Security Council in relation to refugees have included many changes and innovations; and questions relating to refugees have been of great importance generally in international relations in the 1990s.

In the past, many States recognized an obligation, derived from their own experiences and buttressed by international agreements, to accept refugees. Western States also had a political-strategic interest in those refugees whose presence was proof of the failures of communist systems. In the 1990s, the pattern is far from uniform , and some poorer countries have been willing to accept remarkable numbers of refugees. However, States have not generally reacted to crises, as they sometimes did in the past, by accepting large numbers of refugees.

The hardening of attitudes towards refugee influxes, coupled with a growing media-induced awareness of refugee-causing situations, has had major political and military consequences. It causes feelings of guilt (especially in countries with traditionally liberal immigration policies), and strengthens the desire to take some other form of action to help. It contributes to a willingness in many countries to finance humanitarian action. It adds to political pressures to tackle refugee issues in or near the country of origin, for example by creating safe areas and semi-permanent camps, and providing for the return of refugees as part of a peace settlement. All these developments have pointed to the need for international action, not just through UNHCR and other agencies, but also through the UN Security Council.

Statistics on the number of refugees, and the extent of international activity in respect of them, do not support the simple picture of an inexorably rising tide of refugees in the post-Cold-War era. Two qualifications must be made to such a generalization: (1) The steady increase in the numbers of refugees actually began in the mid-1970s. While it intensified in 1989-93, it is by no means purely a post-Cold-War phenomenon. (2) Since 1993 there has been a significant decline in overall refugee numbers – caused by large-scale repatriations and by deterrent measures imposed by countries of destination. The idea that there is a single “global migration crisis” is overstated. Some causes of mass migrations in the 1980s were peculiar to the Cold War (e.g. from Afghanistan); and some causes of mass migrations in the 1990s were peculiar to the circumstances surrounding the end of the Cold War (e.g. flows resulting from the break-up of former com munist States, especially the Soviet Union and Yugoslavia). However, other causes may be expected to persist even after the crises attendant upon the end of the Cold War subside. Many post-colonial and post-communist States have inherently fragile national identities, frontiers and institutions.

The changes in the handling of refugee issues, mainly in the 1990s, have included an increased focus by many agencies on preventive action, even within countries at war, to reduce the likelihood of massive refugee flows across borders; the extension by UNHCR of the categories of people it assists, principally to encompass the “internally displaced” i.e. those displaced within countries at war; the creation of safety zones, mainly under UN Security Council authorization, to reduce the likelihood of refugee outflows and to encourage return; the establishment of camps on the edges of countries at war; the authorization of military intervention, including by the UN Security Council, in refugee-producing situations; the granting by States of temporary protection to refugees rather than permanent asylum; the growth of the practice of assisted, or in some cases forced, repatriation; and the concern with monitoring conditions following resettlement.

Taken together, the changes of the past two decades amount to the emergence of a radically transformed refugee regime aimed at preventing refugee flows and ensuring refugee return. The simple humanitarianism of granting asylum to refugees has given way to a complicated pattern of action containing many elements of uncertainty and insecurity. Some of the transformation has been open, deliberate, and formally accepted, including by the UNHCR Executive Committee and the UN General Assembly. However, none of these changes is wholly satisfactory or definitive.

 Safety zones  

“Safety zone” is an unofficial gener ic term used to cover a wide variety of attempts to declare certain areas off limits so far as military targeting is concerned. In contrast to “undefended towns” (a long-established term in the law of war), there is no assumption that safety zones are open for occupation by the adversary. The establishment of such zones in war-torn areas involves a combination of humanitarian law and action, and has been one of the central features of international responses to conflicts in the 1990s. Safety zones have included the “open relief centres” established on the basis of consent in Sri Lanka (1990-1996); the coalition-enforced “safe haven” in northern Iraq (in existence since 1991); the six UN-declared “safe areas” in Bosnia and Herzegovina (1993-95); and the “secure humanitarian areas” that the UN-authorized and French-led Operation Turquoise set out to create in western Rwanda (summer 1994).

The idea that certain areas should enjoy special protection, even in the midst of ongoing conflict, has long been reflected in the law of war. Article 23 of the 1949 First Geneva Convention (on the wounded and sick) provides for the establishment of “hospital zones and localities”, normally by agreement between the belligerents. Articles 14 and 15 of the Fourth Convention (on civilians) develop these arrangements and provide for the establishment, by agreement between the belligerents, of “neutralized zones”, to shelter wounded and sick combatants or non-combatants, and civilian persons who take no part in hostilities. Article 60 of Additional Protocol I adds a further provision for “demilitarized zones”, also to be established by agreement between the belligerents.

These treaty-proposed arrangements, used only occasionally, have limitations. They are based on the assumptions that the security zone is of limited area; that all combatants and mobile military equipment have been withdrawn; and that no acts of hostility will be committed there by the authorities or the popul ation. They require consent, in most cases formal, between belligerents; they depend on complete demilitarization of the area, seldom achieved in practice; and they do not specify any arrangements for defending the areas or deterring attacks on them.

In the conflicts of the post-Cold-War period, there have been innovative attempts to create areas of special protection for victims and humanitarian bodies assisting them. Such areas have been variously called “corridors of tranquillity”, “humanitarian corridors”, “neutral zones”, “protected areas”, “safe areas”, “safe havens”, “secure humanitarian areas”, “security corridors”, and “security zones”. The UN Security Council has been active in promoting such zones, and has itself used at least five of the above terms. UNHCR has promoted these concepts, and established two “open relief centres” in Sri Lanka, where local inhabitants could take refuge when they felt threatened by the conflict between government forces and Tamil rebels. The variety of the terminology reflects the wide range of forms that such areas can assume, and the absence of a standard legal concept.

Five features have characterized most of the areas of special protection actually established in the post-Cold-War period: (1) different nomenclature from that specified in the conventions has been used to describe them; (2) they have generally been proclaimed or supported by outside States and international bodies, especially the UN Security Council, rather than by the belligerents; (3) outside military forces have in most cases been responsible for protecting them; (4) a central concern has been with the safety of refugees, and the prevention of massive new population movements; and (5) military activity of one kind or another by local belligerents has generally continued within the areas.

The first major area of special protection in the post-Cold-War era was northern Iraq in the immediate aftermath of the 1991 Gul f War. Following a failed uprising which had been encouraged by the Western powers, huge numbers of people, mainly in northern and southern Iraq, had fled to the Iranian and Turkish borders. Starting on 17 April 1991, a military operation by American, British and French forces helped to establish a “safe haven” in Iraq north of the 36th parallel, enabling some 400,000 Kurdish refugees who had fled to the Turkish border to return. The administration of assistance in the area was subsequently taken over by UN agencies. This “safe haven” has throughout its history been subject to military incursions, mainly by Turkish forces but also by the Iraqi army. Despite this, it has provided some modest degree of security.

The six “safe areas” in Bosnia and Herzegovina (1993-95) were established by the UN Security Council during an ongoing war, with the consent of the host government. They were supposed to protect the inhabitants of six towns from Bosnian Serb forces besieging them. The initial resolution (Security Council Resolution 819 of 16 April 1993) demanded “that all parties and others concerned treat Srebrenica and its surroundings as a safe area which should be free from any armed attack or any other hostile act”. A second resolution (SC Res. 824, 6 May 1993) extended this attempt at protection to five additional threatened areas: Sarajevo, Tuzla, Zepa, Gorazde and Bihac. The geographical limits of most of the Bosnian safe areas were never defined. A further resolution (SC Res. 836, 4 June 1993) provided a notoriously complex and ambiguous framework for the crucial issue of the use of force in defence of the safe areas. For the UN and NATO, the problem of protecting the six “safe areas” was compounded by the fact that these were no neutral zones, but areas in which, and from which, Bosnian forces operated. The Serbs complained that the so-called “safe areas” were being used by the Bosnian Muslims to launch attacks against them, as in the case of the Bosnian offensive launched from Bihac in November 1994. Yet there was no serious possibility that the six areas could have been neutralized: neither the Bosnian government, nor the inhabitants, would have been willing to entrust their security to international forces. In July 1995 two of the “safe areas”, Srebrenica and Zepa, were conquered by Bosnian Serb forces. The appalling atrocities committed in Srebrenica were an affront to the credibility of the UN and NATO, which subsequently took military action against the Serbs. NATO’s Operation Deliberate Force bombing campaign, effective in other respects, was too late to save the thousands who had died in Srebrenica.

The mass killings in Rwanda in 1994 led to some belated and largely unsuccessful attempts to create safety zones. During the worst of the Hutu slaughter of Tutsis in April 1994 there was a failure even to present an option to the UN Security Council for dealing with the war on civilians. Eventually a resolution (SC Res. 918, 17 May 1994) decided on the establishment of “secure humanitarian areas”, all within the framework of the existing peace-keeping operation undertaken by the UN Assistance Mission for Rwanda (UNAMIR). No country provided troops. Then the Security Council (SC Res. 929, 22 June 1994) authorized France to establish the temporary Operation Turquoise in western Rwanda with the right to use force. In the event, the zone established by the French in the second half of 1994 was of more assistance to Hutus fleeing the new government in Kigali than to the Tutsi survivors. This only reinforced criticism of the whole UN role: no sanctuary areas for the Tutsis were established in Rwanda when needed.

Overall, the experience of safety zones suggests that, while ad hoc arrangements have advantages over treaty-based ones, establishing such zones, preventing military activity in them, and protecting them from external assault, is difficult and demanding. While many lives have been saved by the establishment of such areas, especi ally in Sarajevo through three harsh winters, they have only rarely provided a secure and enduring haven from the horrors of war.

 Combining peace-keeping with the protection of humanitarian action  

In several conflicts in the 1990s, peace-keeping forces have been mandated to offer protection to humanitarian workers and activities. UN peace-keeping operations and humanitarian operations might seem to be natural partners. Both take place in war-torn countries; both are committed to the principles of impartiality and neutrality; and both tend to be based on a deep-seated opposition to the use of force wherever possible. Many peace-keeping operations in the 1990s have had as central tasks assistance to, and protection of, humanitarian action. Yet combining peace-keeping and humanitarian action has proved extremely difficult. Peace-keeping operations are not well suited to the protection of humanitarian action.

In at least three crises of the 1990s, the Security Council temporarily abandoned attempts at combining peace-keeping and humanitarian action, and authorized more forceful military intervention. It did so in Somalia with the launching of the Unified Task Force (UNITAF) in December 1992, in Rwanda with the launching of Operation Turquoise in June 1994, and in Bosnia with Operation Deliberate Force in August-September 1995. It is no accident that the same pattern was repeated in three very different situations. Protecting humanitarian action, and those it is designed to assist, generally requires a different mandate and force configuration from those used in peace-keeping operations. In short, it involves something more akin to war.

However, the logic of war conflicts at many points with the logic of humanitarian activities. War requires the concentration of forces, and avoiding exposed positions. Humanitarian action requires, in man y cases, dispersal of workers and activities. The clash was evident in Bosnia in 1995: much humanitarian work had to be halted, and humanitarian workers (as well as peace-keepers) had to be withdrawn from exposed positions where they might be taken hostage, before NATO could take any systematic military action against Bosnian Serb targets.

A key issue is the metamorphosis of peace-keeping operations: how they can change into enforcement mode when circumstances so require; or, alternatively, how some other intermediary force can operate in situations where humanitarian activities and those being assisted require physical protection.

 Humanitarian intervention  

The 1990s have seen many cases of “humanitarian intervention” in its classical meaning of military intervention in a State, without the approval of the authorities, and with the purpose of preventing widespread suffering or death among the inhabitants. It is possible, for example, to view certain actions of outside powers in northern Iraq, Somalia, Rwanda and Haiti in this light. These cases were widely seen as evidence that the sovereign State may have to yield to priorities of humanitarian law and action. However, it is far from clear that the sovereign State is on its way out. Indeed, perhaps the main questions raised by all these cases are: (1) Having intervened, what exactly are outside military forces supposed to do? (2) Do the outside powers involved have sufficient interest and commitment to stay for the time it takes to secure fundamental changes in the political, economic and social order?

 Conclusions from experiences of the 1990s  

In the 1990s, a remarkable number of governments, organizations and individuals have climbed on to the humanitarian bandwagon. This is a tribute to i ts solid merits and political appeal. However, the bandwagon is in danger of collapsing under the weight. Some difficult issues highlighted by the events of the 1990s are discussed below.

 The limits of humanitarian approaches  

In the 1990s, humanitarian approaches to conflicts have been exceptional in their prominence, and in the multiplicity and ingenuity of the forms they have assumed. Yet the experience of these years indicates that such approaches have strict limits. Many institutions involved recognize this point. The ICRC, actually more tough-minded than its public image might suggest, is a case in point. The truth is that in some circumstances humanitarian action may even prolong wars, and may fail to provide urgently needed protection, most obviously in cases of genocide. Even in conflicts where humanitarian organizations can usefully act, they must not allow their activities to become an excuse for governments to avoid hard policy decisions. Further, they know that even when they do act, their actions have inherent limits. A good example is the paradoxical fact that delegates of the ICRC, which is the body above all concerned with the development and dissemination of the law of armed conflict, cannot give evidence to courts about atrocities that they witness, as this would compromise the reputation for confidentiality and neutrality on which their right of access crucially depends.

 Need for accountability  

A particular weakness in much humanitarian work is the lack of full accountability. This is not simply a matter of ensuring that there are accurate and properly audited financial accounts (though that is in itself important), but also of ensuring that there is a full and honest assessment of the purposes and effects of humanitarian action, including with respec t to such delicate questions as whether local forces were paid to provide protection, what proportion of deliveries actually reached the intended recipients, whether the assistance had any adverse effects, and whether in fact refugees have been able to resettle in the manner intended by outside

powers and agencies.

 Impartiality and neutrality in humanitarian action  

Many humanitarian organizations working in conflict areas see the value of a reputation for neutrality and impartiality, but have experienced difficulties in maintaining it. The difficulties have arisen from objective problems that are not likely to disappear. It is not easy to maintain a neutral and impartial stance in conflicts in which one party may have strong claims (including legal ones) to international sympathy and support. There may be practical obstacles to giving assistance impartially, i.e. on the basis of need alone. Any element of armed protection or intelligence cooperation further complicates the picture. Finally, a neutral and impartial stance is peculiarly hard to maintain in a situation in which humanitarian assistance is given by literally hundreds of organizations, many subject to political pressures and all encouraged to work out a division of labour among themselves.

The substantial rise in humanitarian activities in armed conflicts poses problems as well as opportunities for the various branches of the International Red Cross and Red Crescent Movement, including the ICRC. The very success of the central Red Cross idea of trying to offer humanitarian assistance to those in desperate need means that there are many bodies active in this field, some of which operate on a basis quite different from that of the ICRC. Whereas the ICRC stresses the absolute nature of the principles of impartiality and neutrality as the basis of its action, and has thereby earn ed access and trust around the world, other bodies put a different spin on these principles.

UN-based bodies in particular cannot take quite such a committed view of the principles of neutrality as between belligerents, and impartiality in the delivery of aid. This is not only because the UN Security Council may be sanctioning one party or supporting another, but also because within the UN there are pressures for humanitarian activities to fall more firmly under political direction, at least in the sense of complying with overall political priorities aimed at fostering international peace and security. Similarly, some NGOs involved in humanitarian aid may openly believe in commitment: in supporting one side in a conflict, especially if they consider that side to be a victim of aggression or to represent the best hope for the recognition of human rights. Some National Red Cross Societies are still, in the words of the 1863 Geneva Conference resolution, concerned above all to “supply relief to their respective armies”. In some conflicts the ICRC may have difficulty in maintaining a strong reputation for impartiality and neutrality when it is associated in many minds with other bodies which have a different approach.

The adequacy or otherwise of the Red Cross principles of neutrality and impartiality is not a new issue. There has been revived interest in this matter in relation to the ICRC’s role in past wars and crises. Books such as Dominique Junod’s The imperiled Red Cross and the Palestine-Eretz-Yisrael Conflict 1945-1952 (1996) and Caroline Moorehead’s Dunant’s dream: War, Switzerland and the history of the Red Cross (1998) have suggested that the Red Cross principles have necessarily involved in past practice some awkward or even shameful moral compromises. The matter has simply arisen in new forms because of the events of the 1990s, especially in the former Yugoslavia and Rwanda. In respect of both crises there was a strong body of opinion that, whether or not these principles were right for the ICRC, they were not good enough as a guide to the overall response of the international community.

The ICRC has been rethinking some key aspects of its approach. In the 1990s, it has been more willing than before to make public statements on certain unacceptable situations; and it has also engaged in public campaigns, for example on anti-personnel landmines. At the same time, it has sought to maintain its distinct identity, not least in its repeated statements that the process of coordination with other bodies has to have some limits. This caution is justified. The implication of this approach is that bodies such as the ICRC need to recognize, much more openly than they have done so far, that the principles they follow, including those of impartiality and neutrality, are not moral absolutes of universal value. Rather they represent one moral position among several possible moral positions, each of which has its own merits and defects.

 The need for discrimination in humanitarian action  

A key issue raised by recent disasters is discrimination. Should humanitarian organizations recognize openly that there are some conflicts in which they cannot usefully get involved – whether because their activities prolong the war, or because their workers are threatened and killed? In the history of humanitarian action, the idea of a universal obligation, of a duty to get involved in all conflicts, is relatively new. It poses problems which have in turn led to such decisions as that taken in Liberia in 1996 to scale down the involvement of humanitarian organizations in a conflict where they could achieve little. This decision was right, and indeed there was an argument for an even more radical withdrawal. The problem is that such de cisions must always be extremely difficult and contentious. The procedure by which they are reached needs to be defensible.

 Physical protection  

The involvement of humanitarian workers in civil wars in the 1990s has repeatedly raised the thorny issue of physical protection of such workers, of their activities, and of those whom they assist. The lack of physical protection, especially of those being assisted, was a critically important issue in Rwanda in 1994, in the camps in the Great Lakes region ever since then, in Srebrenica in 1995, and in various other conflicts, especially in Liberia and Sierra Leone. The killing of Red Cross workers in Chechnya on 17 December 1996 was a particularly shocking episode illustrating both the urgency and the difficulty of the whole question of protection.

For understandable reasons, individuals and organizations involved in humanitarian work have been reluctant to discuss the problem of protection. To get involved in physical protection risks running contrary to the three most basic principles of humanitarian action in armed conflict: neutrality, impartiality and humanity. Yet the problem will not go away. Discussion is not assisted by the fact that in the world of humanitarian organizations, the term “protection” refers to two very different things: (1) legal protection in the sense of a special status and certain rights provided for under the 1949 Geneva Conventions and their Additional Protocols (hence the key term “protected persons”); and (2) physical protection from attack.

In the past, humanitarian relief activities in armed conflict often took place against the background of an unspoken assumption that the States engaged in conflict would at the same time provide a degree of stability and security. No s uch comfortable assumption can be made in cases in which State authority has collapsed. Thus the problem of physical protection of humanitarian workers and activities is by nature extremely complex. There is not any single answer for all organizations and contingencies. The proposition that protection should be provided automatically, and should not be a matter for political decision in each case, is unrealistic. Even if such an approach were adopted by humanitarian organizations, it would be viewed with great scepticism both among countries expected to provide forces and equipment, and among Third World countries which are understandably nervous about any doctrine which might seem to undermine their sovereignty. The reality is that different humanitarian organizations are bound to have different approaches to the matter of physical protection. This is mainly because of the different nature of their responsibilities.

For example, the ICRC is nervous about associating humanitarian activity with military protection and intervention. As Peter Küng, speaking for the ICRC, said in the Security Council discussion on protection for humanitarian assistance held in New York on 21 May 1997, armed action of various kinds might be needed, but armed protection of humanitarian activities could jeopardize their impartial status. This cautious attitude is particularly understandable in the light of the ICRC’s core functions, which include visiting people in places of detention on all sides in a conflict. That clearly requires a high degree of consent, and respect for the Red Cross principles. This is not to suggest that the ICRC is dogmatic on the point. At times it has had to employ armed escorts; it is bound to rethink its position periodically in the light of disasters such as the one in Chechnya; and it recognizes that some other humanitarian bodies may take different p ositions from those of the ICRC.

UNHCR has different responsibilities, which include in many cases the establishment of refugee camps and the maintenance of order within them. These tasks do at times require the provision of physical protection, whether by locally recruited forces or those from distant lands. Thus it is not at all surprising that UNHCR’s approach to the question of physical protection has differed from that of the ICRC. For different reasons, bodies such as Oxfam and Médecins sans frontières less attached than the ICRC to absolute principles of impartiality and neutrality, have also on occasion called for military protection of both victims and humanitarian activities.

Some have suggested the necessity of a new legal agreement affording protection to ICRC and other humanitarian workers. It is doubtful whether this would be useful. Such workers already enjoy a considerable degree of legal protection under the 1949 Geneva Conventions and their Additional Protocols. Some humanitarian workers as well as peace-keepers also get a modest degree of formal legal protection from the 1994 Convention on the Safety of United Nations and Associated Personnel. The problem is not an absence of rules, but an absence of respect for them, and of willingness to back them up, if necessary with force.

 Modes of advocacy of international humanitarian law  

The repeated violations of the most basic rules of the law of armed conflict must induce caution about the extent of any achievements of the 1990s. One difficulty in getting respect for the law may derive from the very mode by which it is advocated. Presenting a vision of the law as simply a creation of Geneva, imposed on a somewhat reluctant and sinful world, does not do justice to its ancient and diverse origins, nor to the fact that muc h of the law reflects the interests and ethics of States and their armed forces. Likewise, presenting the implementation of the law as resting entirely with the new UN criminal tribunals does not do justice to the variety of modes of implementation. The unfortunate way in which humanitarian law has been advocated, and the occasional lack of understanding of the positions and policies of States, may have contributed to the USA’s disenchantment with key contemporary developments in this law.

Indeed, the question has to be asked: should elements of restraint and decency in the conduct of war be seen as depending fundamentally on an international legal regime, or rather on a system of custom, honour, professional standards, and natural law? In recent years, many in the humanitarian field have been attracted by the idea of an expanding body of international law, armed at last with some teeth, enforced by supranational bodies, and gradually subduing the evil ways of selfish sovereign States and warlords. This vision is flawed. In particular, it may overestimate the willingness of States to submit their military activities to consideration by international bodies; and it may underestimate the importance of the State as one key means of implementing provisions of the law of war. Above all, it may neglect the extent to which the values underlying the law of war have to emerge from within States and armed forces, and can only to a very limited extent be imposed from without.

 Humanitarianism in context  

The events of the 1990s have challenged simple assumptions that the world is in the process of creating a more advanced international order. If there is, as some assert, an “international humanitarian order”, it is one in which States and institutions have repeatedly adopted the language of humanitarianism only to abandon victims of armed conflict, as in Bosnia and Rwanda, t o a dreadful fate.

Equally, events challenge any simple view that the world is transcending the system of States. In reality, in situations of conflict, States not only remain powerful, but are also often essential to securing a peaceful settlement. After a conflict a new local balance has to emerge, in which States and political forces in the area, including the belligerents themselves, are likely to play a key role. States should not necessarily be seen as operating on a lower moral plane than other actors. Governments seeking to address conflicts may make decisions on the basis of legitimate interests and moral principles which deserve respect even if they sometimes clash with humanitarian principles.

In the 1990s, humanitarian action and law have become more important parts of an international response to crises than ever before; and they are likely to remain so. Part of the reason why they have not had a major transformative effect on the conduct of international relations is that such approaches are in some measure a reflection of the reluctance of States to get involved more decisively in conflicts. Humanitarian approaches are not necessarily the opposite of those based on interest. Rather, they should be seen as one enduring part of international politics. To adapt Clausewitz, they can be a continuation of political intercourse with the admixture of other means.

All this is a plea, not for a single new approach by humanitarian bodies, but for complementing their technical and legal expertise with a heightened level of political and historical expertise, and a stronger culture of accountability. There have been too many elements of self-righteousness, self-delusion and ethnocentrism in some approaches to humanitarian advocacy and action. There is a need for more analysis of crises, drawing on political, strategic and area studies. There is also a need for a strong “institutional memory” and a culture of serious resear ch in the humanitarian field. This is both because some of the dilemmas and opportunities that are faced are historically new, and because some of them are timeless. Ignorance is no excuse for repeating old mistakes or making new ones.

Abstract in French