The main difficulty is that if NATO uses force without an authorisation of the UN Security Council, it breaches International Law as codified in the UN Charter. NATO actions in similar (i.e. intrastate) conflicts can place the organisation outside legality. This question was given extensive discussion on the occasion of NATO´s war against the FRY, which denounced the attack before the International Court of Justice.
The threat or use of force is governed under international law by the UN Charter. Article 2 (4) reads:
The UN Charter4 provides only for two exceptions from this prohibition, embodied in Art. 51 and Art.39. Art. 51 reads:
This Article constitutes the legal foundation of the Washington Treaty by which NATO was founded, and serves as a basis for its Art. 5.
The second exception that allows for the use of force can be found under Article 42. It reads:
The sole competence of the Security Council for the maintenance of international peace and security is laid down in Article 39:
Consequently, any threat or use of force that is neither justified as self-defence against an armed attack nor authorised by the Security Council constitutes undeniably a breach of the UN Charter.6
Those paragraphs of the UN Charter referring to regional arrangements should be mentioned as well. Although NATO has never applied for recognition as a "regional agency dealing with security issues" in accordance with in Chapter VIII of the Charter,7 these provisions would be applicable by way of analogy.
While Art. 528 allows for the existence of regional arrangements, Article 53 explicitly forbids military intervention by regional agencies without a Security Council mandate:
It should be noted that Charter regulations prevail over any other treaty or agreement, as stated in Article 103 of the Charter itself:
In addition, the pre-eminence of Article 2(4) of the Charter over any other source of international law is enshrined in Art. 30.1 of the 1969 Vienna Convention on the Law of the Treaties, according to which the ban of force of Article 2(4) is part of ius cogens, i.e., it is accepted and recognised by the international community as an unalterable norm.9 It is binding on states both individually and as members of international organisations, as well as on international organisations.
Moreover, one should note that all NATO members reiterated the obligations acquired under the UN Charter at a regional level in the 1975 Helsinki Act of the Conference on Security and Co-operation in Europe.10 Even though it does not have legal consistence and is geographically restricted to Europe, it includes some provisions repeating the above-mentioned principles:
Title II calls for refraining from the threat or use of force:
Title IV sets forth the principle of non-intervention in internal affairs:
Finally, NATO members once again confirmed their commitment when they signed the organisation's founding act, the Washington Treaty. Article 7 explicitly binds NATO countries to act within the UN Charter:
Despite statements by US officials,11 the conduct of
operations such as the Kosovo intervention are inconsistent with the text of the
Washington Treaty. It runs counter to the Alliances commitment to refrain from the
use of force set forth in Art. 1 as well as to the responsibility of the Security Council
for the maintenance of peace recognised in Art. 7.
The question raised is: can interventions such as the one over Kosovo be possibly justified under international law?
One should look at the justifications adduced by the organisation and the states concerned. NATO governments, left with the responsibility of presenting the wars objectives to its respective constituencies individually,13 gave three legally relevant justifications for the recourse to armed force:
First, they argued that the operation took place within the framework of UN Security Council Resolutions. When Secretary-General Solana explained why NATO had issued the activation order, he made explicit reference to UN Security Council Resolutions stating that the FRY had not yet complied with the urgent demands of the International Community, "despite UNSC Resolution 1160 of 31 March 1998 followed by UNSC Resolution 1199 of 23 September 1998, both acting under Chapter VII of the UN Charter". He further said that "the deterioration of the situation in Kosovo and its magnitude constitute a serious threat to peace and security in the region as explicitly referred to in the UNSC Resolution 1199". 14
It was argued that a number of Security Council resolutions on Kosovo repeatedly defied by the Yugoslav government had largely determined that the crisis in this region constituted a threat to international peace and security and therefore provided some legal basis for military action. The intervention was aimed at enforcing FRYs compliance with the demands issued by the Security Council in previous resolutions. According to US Secretary of State Albright:
Resolution 1160 called upon the Federal Republic of Yugoslavia "to immediately take the further necessary steps to achieve a political solution to the issue of Kosovo through dialogue".16
Resolution 1199 adopted an increasingly assertive tone and demanded "that all parties, groups and individuals immediately cease hostilities and maintain a cease-fire in Kosovo"17 and "that the Federal Republic of Yugoslavia...cease all action by the security forces affecting the civilian population and order the withdrawal of security units used for civilian repression".18 Furthermore, in the same paragraph Yugoslavia was demanded "to enable...international monitoring in Kosovo".
Finally, Resolution 1203 of October 24, 1998 demanded the implementation of the agreements Yugoslavia had signed with both NATO and OSCE concerning a Verification Mission and the public commitment of Yugoslavia to negotiate a political settlement.19
While all these resolutions were the premise for authorisation of the use of force, the act needed to actually resort to force never materialised within the Security Council. However, it is only the competence of the Security Council to authorise the enforcement of its resolutions, not of the individual states or groups of states; thus, legally speaking, enforcing a resolution without enforcement provisions is unlawful. In the words of White, "where there is no authorisation, there is no legal basis".20 At best, the strict adherence to UN Security Council resolutions offers political advantages as opposed to other unilateral actions.21
US Secretary of State Albright also argued that the behaviour of Serb forces in Kosovo was a breach of the Geneva Conventions, the Charter on Human Rights and, if the ethnic cleansing continued, of the Genocide Convention of 1948. She contended that these treaties provided an alternative source of legitimisation of NATO action.22
Most jurists agree that the Genocide Convention was not affected.23 In any case, these treaties do not provide for enforcement; therefore, they do not offer a legal basis for the use of force. The reference to the breaches could at best be seen as a means to justify the situation in Kosovo was amenable to an intervention on humanitarian grounds.
In view of these prohibitions, it is clear that the Kosovo intervention could not possibly be legitimised by UN Security Council Resolutions. There seems to be no way for NATO to circumvent the unlawfulness of crisis management operations without a UN mandate.
Some within NATO claimed that military intervention in another state can be justified in cases of overwhelming necessity. While all leaders clearly put the emphasis on the humanitarian distress, reference to the theory of humanitarian intervention was sometimes made more explicitly than others. NATO Secretary-General Solana repeatedly stressed that the campaigns objective was to "avoid a humanitarian disaster". In the above-mentioned letter explaining the Allies approval of the activation order, he mentioned "the very stringent report of the Secretary-General of the United Nations pursuant to both resolutions, (which) warned inter alia of the danger of a humanitarian disaster in Kosovo". He further referred to "the continuation of a humanitarian catastrophe, because no concrete measures towards a peaceful solution of the crisis have been taken by the FRY."24 The risk of the conflict spilling-over the Yugoslav borders took a secondary place:
Among NATO leaders, British Prime Minister Tony Blair most vigorously intended to conceptualise the moral rationale of the intervention, even formulating a doctrine of the international community:
While the validity of the two first mentioned justifications - UN Security Council resolutions and international treaties - was easily dismissed, the question of the humanitarian intervention is more complicated. Could the doctrine of humanitarian intervention confer some legitimacy to the Kosovo operation?
In order to give an answer to this question, the first issue to be clarified is whether the humanitarian intervention actually exists as an institution.
Humanitarian intervention "encompasses any use of armed force by a State against another State for the purpose of protecting the life and liberty of the citizens of the latter State unwilling or unable to do so itself."27 Although a majority of writers supported the idea of a lawful humanitarian intervention during the era previous to the signing of the UN Charter, it is debatable whether it was clearly established under the customary international law of that time. In any case, it is clear that even if that right had existed prior to 1945, it had not survived the adoption of the Charter; although no explicit reference is made in the Charter to humanitarian intervention, it is in conflict with the prohibition of the threat or used of force in Art. 2 (4).
Only a minority of jurists maintained in the period after 1945 that the institution of humanitarian intervention is permitted under customary international law, a recognised source of international law. They argued that the principles of non-violence and the protection of human rights must be balanced against each other in any particular case, with the result that in extreme situations involving seriously inhumane treatment, the latter principle should override the former. It was often adduced that humanitarian interventions were not precluded by Art. 2 (4) of the UN Charter since it only banned the use of force when directed "against the territorial integrity or political independence" of the target state, and not at the protection of its population.28 This argument is based on a wrong understanding of the terms in question, which were not added to the text in order to restrict its scope, but in the spirit of specifying the prohibition. Furthermore, it has been generally rejected as incompatible with the system of the Charter, since its framers had attributed a predominant role to individual non-violence. The recognition of humanitarian intervention as an exemption from this prohibition of Art. 2(4) was deliberately abnegated in order to prevent a confrontation of the two power blocks.29
Some of these specialists have voiced the need for reappraisal of this question in view of a number of events taking place after the end of the Cold War. They contend that notions of legitimacy are changing, with humanitarian reasons emerging as a third option for justified military intervention in a sovereign state.30
Two constitutive elements define the existence of an international custom: diuturnitas, i.e. a consistent international practice and opinio juris, i.e. acceptance of this custom by states.31 In order to determine the existence of the institution of humanitarian intervention, it should be demonstrated that both elements are present. Advocates of this notion point to recent UN and state practice and codified and non-codified signs of growing acceptance of this institution. In the following, these arguments will be examined.
Greenwood defends that states practice allows for the consideration of this institution as established under customary international law. He contends that enough precedents of humanitarian interventions justify the claim that most states accept them as legal.32 For him, the emergence of "at least a limited right of intervention" is best exemplified by the allied intervention in northern Iraq in 1991 on behalf of the Kurds and the imposition of a no-fly zone in southern Iraq on behalf of the Shiites some months later. As further examples, he cites the cases of Indias invasion of Bangladesh (1971), Tanzania s invasion of Uganda (1979) and ECOWAS intervention in Liberia (1990). In Greenwoods words,
Greenwood cites cases in which there was no state authority to request or allow outside intervention, or when in civil strives, foreign help met with the consent of all parties to the conflict. These examples cannot bear precedential character for cases when a state authority existed.34 This leaves the enforcement of a no-fly zone in Iraq as the only case of non-UN authorised state intervention that would qualify as a humanitarian intervention.35
The UK justified its participation on this basis, stating that it intervened"36, in exercise of the customary international principle of humanitarian intervention", thus invoking this institution by name. Former British Foreign Secretary Douglas Hurd declared:
Nevertheless, non-compliance with UN-imposed requirements does not authorise states to use force. An enforcement provision is still required. One should not be mislead by official statements: Interventions conducted outside the UN umbrella are illegal. The enforcement of a no-fly zone in Iraq falls within this category.
Very few interventions both in the pre-Charter period and after 1945
often cited as precedents can be described as humanitarian without serious difficulties.38
Greenwood himself acknowledges that the circumstances under which the intervention in Iraq
took place are rather unique and unlikely to be repeated.39 One author has
written that "the first respectful piece of practice supporting the doctrine of
humanitarian intervention is represented by the NATO air campaign against the FRY."40
Some commentators contend that the UN Security Council has established a practice legitimising humanitarian grounded interventions so that they can now be considered part of customary law. This argument suggests that increased UN involvement in civil war situations can be regarded as evidence that international obligations and problems of humanitarianism may rank alongside the defence of national interests.41 As precedents, reference is made to Resolution 699 of 5 April 1991 on Iraq, Resolution 770 of 13 August 1992 on Bosnia-Herzegovina, Resolution 794 of 3 December 1992 on Somalia, Resolution 929 of 22 June 1994 on Zaire and Rwanda, Resolution 940 of 31 July 1994 on Haiti and resolution 1101 of 28 March 1997 on Albania. According to this analysis, NATOs strikes on FRY would be legal.
Indeed, cited developments evidence a notable shift underway in terms of the principle and practice of humanitarian intervention over the past ten years, suggesting that the protection of human rights in situations of extreme deprivation and suffering is sometimes given pre-eminence over the sovereignty principle.42 In order to correctly assess the existence of an international custom, though, Security Council and non-authorised practice require a separate analysis. This is so because, as explained above, military action taken by the Security Council is one of the legal exceptions provided for in the Charter, while state practice without authorisation is not.
In recent years, the UN Security Council has clearly become more active and more assertive in the field of security. This quantitative evolution was accompanied by a qualitative one in two senses:
First, in the early nineties, the UN Security Council conducted a considerable number of forcible interventions, or "peacekeeping of third generation".
During the Cold War, peacekeeping interventions were governed by the strict principles of consent of the parties on the deployment of the forces, impartiality and non-use of force, with the peacekeeping tasks being restricted to the interposition between the contending parties and the monitoring of the cease-fires. Towards the end of the 1980's, peacekeeping began to turn multifunctional, including new tasks such as election monitoring, national reconciliation or de-mining. Namibia, El Salvador and Cambodia are examples of the so-called peacekeeping operations of second generation. In this period it became the practice of the Security Council to allow for military action to conduct humanitarian relief or rescue operations; most examples mentioned by Guicherd correspond to this tendency. Nevertheless, the basic principles mentioned above continued to be respected.
At a third stage, the UN increasingly authorised operations conducted by regional organisations such as OSCEs Kosovo Verification Mission, NATO-led IFOR and SFOR in Bosnia-Herzegovina, CIS in the Trans-Caucasian area and international coalitions such as the Italian-led coalition in Operation Alba. The UN also authorised an individual state, the US, to act in Haiti, as permitted under Art.52 of the UN Charter. Some of these operations, notably Somalia, Rwanda, and IFOR/SFOR in Bosnia have been characterised differently from previous peace-keeping operations. They present characteristics such as the emergence of a coercive dimension beyond the right to self-defence43 and the delegation of the use of force to national (or coalition) forces.
While some observers have denominated these interventions "peacekeeping of third generation", others prefer to call them peace-enforcement operations, as a type separated from peacekeeping. This type has been subject to more acute criticism than the peacekeeping of first and second generation.44
Second, the UN Security Council has shown an increasing willingness to conduct this kind of forcible interventions in cases of purely domestic humanitarian distress.
Beyond its more active role in humanitarian aid operations,45 the Security Council is increasingly willing to characterise the plight of the people of a state as a threat to international peace and security, and to authorise interventions resting merely on humanitarian grounds.46 This constitutes a substantial restriction of the scope of Art. 2(7) of the Charter, which reads:
This trend suggests that the rationale of non-intervention in internal affairs enshrined here is being replaced by the principle that massive or widespread violations of human rights or humanitarian law arising from governmental acts or internal conflicts and the magnitude of human suffering that they engender can constitute a threat to international peace and security. In those circumstances, the Security Council can take appropriate measures, including the use of force for the protection of humanitarian relief operations. This was especially evident in the case of Somalia, where the threat to neighbouring states was relatively small and arguably receding by the time Resolution 794 was adopted.
In this light, it becomes clear that important difficulties remain with the argument that the Security Council has been authorising a number of "humanitarian interventions".
First, there seems to be confusion in the understanding of the term 'humanitarian intervention'. Guicherds analysis overlooks an important fact: with the sole exception of Resolution 940, all cited Security Council Resolutions authorised the use of force to protect humanitarian relief and rescue operations, but not to conduct operations effectively intervening in the conflict and aimed at influencing its outcome. NATO's intervention in Kosovo would not fall into this category. Humanitarian operations are designed to provide food, shelter, and medicine directly to victims of conflicts, or to protect relief workers who furnish these goods and services, and are not meant to influence the political incentives of the actors in a conflict. They should be distinguished from peace enforcement operations, which can imply the use of force against one of the parties to enforce an end to hostilities.
The above-mentioned analysis also fails to recognise that when resolutions provided for forcible implementation without the consent of the government of the target country, they did so in situations when there was effectively no government that could authorise such an operation.47
Finally, a further problem remains. While the practice shows a lesser degree of respect towards the sovereignty principle than in the Cold-War era, it is debatable whether developments taking place within a codified legal system such as the UN Charter can serve as a basis for justifying actions outside the UN umbrella. 48
What about the opinio juris? Legal doctrine attaches more importance to the psychological element than to the material one, noting that practice needs to be accompanied by a manifestation by states bringing it home that they view their behaviour as legal. In fact, it is the conviction that the act is legal what differentiates law from mere courtesy.
Since the entry into force of the UN Charter, most states taking forcible action abroad have tended to reject the notion that their actions were humanitarian interventions. Even in cases where there appeared to be a clear humanitarian rationale such as in Indias intervention in East Bengal in 1971 or the Vietnamese intervention in Cambodia in 1978, states did not make any reference to humanitarianism, what is still interpreted as an indication that they were willing to uphold the non-interventionist regime.49
Do developments following the Kosovo events add something new to this state of affairs? Several authors contend that now a majority of states regard humanitarian interventions as legal, pointing out that the voting on two UN Security Council resolutions can be seen as granting post-facto legitimisation to the Kosovo operation:
A draft resolution tabled by Russia two days after the commencement of the NATO strikes against FRY calling for the cessation of the use of force against the FRY was defeated by twelve votes to three- those of Russia, China and Namibia.50 Further, resolution 1244 of 10 June 1999 authorising "Member States and relevant international organisations to establish the international security presence in Kosovo", adopted with the abstention of China, can be regarded as sanctioning NATO's presence in Yugoslav territory.51
It is actually striking that most Third World countries, which have traditionally displayed firm opposition to humanitarian interventions, were supportive of the resolution. However, it should be kept in mind that opposition by states is enough to prevent the formation of a custom. A "majority of states" would not be enough. With the consistent expressions of disapproval by a number of states, in particular coming from countries such as Russia and China, no international custom binding upon them could be set up.
As a further sign of a growing international acceptance of humanitarian-grounded interventions, some commentators adduce that a principle recognising that populations in danger of starvation, massacre or other forms of massive suffering have the right to receive assistance was set out by General Assembly Resolution 43/131 of 8 December 1988.52
However, it is important to make clear that the resolution merely
invites countries to contribute and to facilitate the provision of humanitarian assistance
in natural disasters and similar situations. GA resolutions fall short of
furnishing a legal basis, since they are not legally binding. At best, they can be
considered to reflect a change in general attitudes of the international community. Only a
wide interpretation of the resolutions allows for a recognition of the right of endangered
populations to humanitarian assistance.
The case against the existence of this institution, represented by the overwhelming legal opinion, is quite conclusive: Both constitutive elements of the international custom, diuturnitas and opinio juris are lacking. Neither is there a consistent state practice nor the general agreement of the community of states required to establish an international custom is given. Therefore, it is clear that the institution of humanitarian intervention does not exist under customary international law.54