A.  Concept

1  Collective security has been referred to as ‘a system, regional or global, in which each state in the system accepts that the security of one is the concern of all, and agrees to join in a collective response to threats to, and breaches of, the peace’ (Lowe and others 13). In recent years it has come to have a wider meaning. In the 2005 UNGA Report ‘In Larger Freedom’ (at paras 77–78), the United Nations Secretary-General embraced a comprehensive concept of collective security as suggested by the High-level Panel on Threats, Challenges and Change which was convened by him in 2004. In accordance with this concept, any event or process that leads to large-scale death or lessening of life chances undermines States as the basic unit of the international system and poses a threat to international security. So defined, there are six inter-connected clusters of threats with which the world must be concerned. The first include economic and social threats, including poverty, infectious disease, and environmental degradation. The remaining five clusters concern traditional threats to State security, namely inter-State conflict, internal conflict, weapons of mass destruction, terrorism, and transnational organized crime (UN High-Level Panel on Threats, Challenges and Change 25).

2  The expression ‘collective security’ is not a term of art in international law. It belongs more to the discipline of international relations, where a ‘collective security system’ may be distinguished from military alliances, which are usually aimed at defence against third States on the one hand, and ‘world government’ which implies a much greater degree of integration on the other.

3  While the expression ‘collective security’ does not occur in the United Nations Charter (‘UN Charter’), it is often used to refer to the system for the maintenance of international peace and security under the UN Charter and the corresponding provisions of regional organizations. Some question whether the UN Charter provides for a true collective security system (Lowe and others 13–15), but United Nations organs regularly use the term, at least in an aspirational way (see for example, UNGA Res 60/1 ‘2005 World Summit Outcome’ [16 September 2005]; Statement by the President of the Security Council [19 November 2008]).

4  The system of collective security under the UN Charter is reflected principally in the provisions concerning the maintenance of international peace and security, especially those relating to the UN Security Council (United Nations, Security Council). These include Art. 2 (4) UN Charter, which contains a general prohibition on the threat or use of force (Use of Force, Prohibition; Use of Force, Prohibition of Threat), except in the case of self-defence, recognized by Art. 51 UN Charter, or the use of force by or authorized by the UN Security Council under Chapter VII UN Charter.

5  If it determines, in accordance with Art. 39 UN Charter, the existence of a threat to the peace (Peace, Threat to), breach of the peace (Peace, Breach of), or act of aggression, the UN Security Council may make recommendations, or decide what measures shall be taken by members of the UN to maintain or restore international peace and security. Such measures may be non-forcible, such as economic sanctions under Art. 41 UN Charter, or may involve the use of force as provided in Art. 42 UN Charter. In this manner, the UN Charter attempts to realize the first purpose of the UN set forth in Art. 1 (1) UN Charter, namely ‘[t]o maintain international peace and security, and to that end: to take effective collective measures for the prevention and removal of threats to the peace, and for the suppression of acts of aggression or other breaches of the peace…’

B.  Historical Evolution

6  The League of Nations had been preceded in the 19^th century by the arrangements known as Concert of Europe (History of International Law, 1815 to World War I) and some theoretical schemes for perpetual peace from earlier periods (W Penn An Essay towards the Present and Future Peace of Europe: By the Establishment of an European Diet, Parliament, or Estates [Taylor London 1693, reprinted by Georg Olms Verlag Hildesheim 1983]; CI Castel de Saint-Pierre Projet pour rendre la paix perpétuelle en Europe [Schouten Utrecht 1713, reprinted by Fayard Paris 1986]; I Kant Zum Ewigen Frieden [Nicolovius Königsberg 1795, reprinted by Fischer Frankfurt 2008]). The establishment of the League in 1919 marked the first serious attempt to institutionalize collective security in modern times. The League’s goals included disarmament, the prevention of war through collective security (Conflict Prevention), the settling of disputes between countries through negotiation, diplomacy, arbitration, and judicial settlement of international disputes, and the improvement of global welfare. However, the League Covenant did not seek to outlaw the unilateral use of force and lacked a system for central decision-making and the enforcement of sanctions. The members of the League undertook to respect and preserve as against external aggression the territorial integrity and political independence of all members of the League, but the Council was only empowered to ‘advise upon the means by which this obligation shall be fulfilled’ (Art. 10 League Covenant). In the central provision of the League Covenant on the use of force the members agreed to submit a matter that might lead to a rupture to arbitration or judicial settlement or to inquiry by the Council, but thereafter merely agreed ‘in no case to resort to war until three months after the award by the arbitrators or the judicial decision, or the report of the Council’ (Art. 12). These provisions were supplemented by the Kellogg-Briand Pact (1928), which sought to outlaw war as an instrument of national policy. In practice, the League proved to be ineffectual in the face of Italy’s conquest of Abyssinia (1936) and the aggressive acts of Germany and the Soviet Union. The League found itself unable to agree on collective sanctions, let alone to take military action. One of the lessons drawn was that in order for any collective security system to be effective, the existence of a centralized decision-making procedure for determining acts of aggression and imposing enforcement measures was essential.

7  The notion of collective security developed against the background of inter-State military conflict and the desire to prevent the outbreak of such conflict (Armed Conflict, International). This was also the case for the collective security system under the UN Charter elaborated against the backdrop of World War II. As a result, the enforcement system contained in Chapter VII UN Charter was not designed to deal with intrastate conflicts (Armed Conflict, Non-International). Nonetheless, since the adoption of the UN Charter, and in particular since the end of the Cold War (1947–91), the large number of internal armed conflicts around the globe has resulted in the deployment of the enforcement mechanisms foreseen under Chapter VII UN Charter. These included measures ranging from sanctions to military measures in relation to, inter alia, the former Yugoslavia (Croatia [eg UNSC Res 1037 (1996) (15 January 1996) SCOR 51^st Year 25]; Bosnia-Herzegovina [eg UNSC Res 824 (1993) (6 May 1993) SCOR 48^th Year 11]; and Kosovo [eg UNSC Res 1244 (1999) (10 June 1999) SCOR 54^th Year 32]), Somalia (eg UNSC Res 794 [1992] [3 December 1992] SCOR 47^th Year 63; Somalia, Conflict), Haiti (eg UNSC Res 841 [1993] [16 June 1993] SCOR 48^th Year 119; UNSC Res 940 [1994] [31 July 1994] SCOR 49^th Year 51; Haiti, Conflict), Angola (eg UNSC Res 864 [1993] [15 September 1993] SCOR 48^th Year 59), Rwanda (eg UNSC Res 929 [1994] [22 June 1994] SCOR 49^th Year 10), Congo, Democratic Republic of the (eg UNSC Res 1078 [1996] [9 November 1996] SCOR 51^st Year 115), Liberia (UNSC Res 788 [1992] [19 November 1992] SCOR 47^th Year 99), Sierra Leone (eg UNSC Res 1132 [1997] [8 October 1997] SCOR 52^nd Year 83); Côte d’Ivoire (eg UNSC Res 1464 [2003] [4 February 2003] SCOR [1 August 2002–31 July 2003] 176), Sudan (eg UNSC Res 1769 [2007] [31 July 2007] SCOR [1 August 2006–31 July 2007] 234), Libya (eg UNSC Res 1973 [2011] [17 March 2011] SCOR [1 August 2010–31 July 2011] 390), and Mali (UN Doc S/RES/2100 [2013] [25 April 2013]).

C.  UN Charter Provisions on Collective Security

8  The main provisions of the UN Charter on collective security are Art. 2 (4) (prohibition of the threat or use of force); Art. 51 (inherent right of individual or collective self-defence); and Arts 39–42, concerning respectively the determination of a threat to the peace, breach of the peace, or act of aggression (Art. 39); provisional measures (Art. 40); measures not involving the use of armed force (Art. 41); and measures involving the use of armed force (Art. 42).

9  Before the Security Council can take any measures in order to enforce collective security, it first has to cross the threshold that triggers the possibility of collective enforcement action under Chapter VII UN Charter by determining the existence of a threat to the peace, breach of the peace, or act of aggression in accordance with Art. 39 UN Charter.

10  Non-forcible measures under Art. 41 have in the past included broad economic sanctions regimes that affected large sections of the economy of the affected State, such as the sanctions regimes against Iraq (UNSC Res 661 [1990] [6 August 1990] SCOR 45^th Year 19; Iraq-Kuwait War [1990–91]) and Serbia (UNSC Res 757 [1992] [30 May 1992] SCOR 47^th Year 13). Since the 1990s the Security Council has shown a preference for sanctions that are more narrowly targeted, aimed at a specific sector, or at a government, rebel group or other entity, or individuals. Examples include oil embargoes, arms embargoes (including nuclear-related materials), travel bans, assets freezing, or any combination of the aforementioned (eg Sierra Leone [eg UNSC Res 1132 (1997) (8 October 1997) SCOR 52^nd Year 83]; Taliban [eg UNSC Res 1267 (1999) (15 October 1999) SCOR 54^th Year 148]; Sudan [eg UNSC Res 1556 (2004) (30 July 2004) SCOR (1 August 2003–31 July 2004) 150]; North Korea [eg UNSC Res 1718 (2006) (14 October 2006) SCOR (1 August 2006–31 July 2007) 279]; Iran [eg UNSC Res 1737 (2006) (23 December 2006) SCOR (1 August 2006–31 July 2007) 259]; Democratic Republic of the Congo [eg UNSC Res 1807 (2008) (31 March 2008) SCOR (1 August 2007–31 July 2008) 126]); Libya [eg UNSC Res 1973 (2011) (17 March 2011) SCOR (1 August 2010–31 July 2011) 390; Guinea Bissau [eg UNSC Res 2048 (2012) (18 May 2012)]).

11  As far as measures involving the use of force are concerned, Art. 43 (1) UN Charter directed Member States to make armed forces available to the Security Council at its request. Under Arts 46 and 47, the Military Staff Committee, consisting of the Chiefs of Staff of the five permanent members of the Security Council or their representatives, was to advise and assist the Security Council. But these provisions have remained largely a dead letter. While the Military Staff Committee has continued to meet, it has done nothing of substance since the late 1940s (though it was mentioned in UNSC Res 665 [1990] [25 August 1990] [SCOR 45^th Year 21] in connection with the Iraq-Kuwait war). Numerous proposals have been made over the years, particularly in the 1990s, for UN standing forces of one sort or another (see generally Roberts).

12  However, as the Art. 43 agreements never materialized, the UN was forced to look for alternatives. It found a solution in the authorization of ‘able and willing’ States and/or regional organizations to carry out military measures on its behalf (Blokker 542). The authorization model was used for the first time during the Korean War (1950–53) (UNSC Res 82 [1950] [25 June 1950]). The return of the Soviet Union to the Security Council prevented further use of this instrument during the Cold War era (Blokker 542–43). The only exception was UNSC Resolution 221 (1966) of 9 April 1966 which was adopted to enforce the oil embargo against Southern Rhodesia (Rhodesia/Zimbabwe). However, since 1990 this model has been revived, beginning with UNSC Resolution 678 (1990) of 29 November 1990, which was adopted after the Iraqi invasion of Kuwait. The concept of authorization is nowadays well established in the practice of the Security Council and was also recognized by the European Court of Human Rights (ECtHR) in Behrami and Behrami v France (at para. 133).

13  The most obvious basis for authorizing States to engage in military operations on its behalf can be found in Art. 42 UN Charter. It is now clear that the existence of an Article 43 agreement is not a prerequisite for the Security Council to resort to Art. 42 UN Charter, as this would severely limit the powers of the Security Council under this article. Consequently, Chapter VII UN Charter has been interpreted to mean that the Security Council can authorize Member States to undertake military measures for the restoration or maintenance of international peace and security, where they are willing to do so. This argument gains strength if one reads Art. 42 in conjunction with Art. 48 (1) UN Charter. The latter concretizes States’ obligation to carry out binding decisions of the Security Council, in that it provides the Security Council with the power to determine who will participate in enforcement action. If one reads Art. 48 (1) together with Art. 25, it provides the Security Council with the power to determine that the action required for the execution of Security Council decisions is undertaken by all or only some UN Members. This complements the material basis for authorizing Member States to undertake military measures on behalf of the Security Council, provided in Art. 42 (Krisch 1337).

14  Another possible basis for authorizing States to use force would be the right of self-defence recognized in Art. 51 UN Charter (Gill 92). In fact, it has been argued that the authorization of the use of force against Iraq in UNSC Resolution 678 (1990) of 29 November 1990 was an exercise of the right of self-defence recognized in Art. 51, rather than a collective security measure. However, the question does arise whether Art. 51 is intended to be used by the Security Council itself. It seems that this Article would instead provide a basis for action by States, either individually or collectively, if an armed attack occurs, pending Security Council action. Once the Security Council itself authorizes States the use of force on its behalf, it does so on the basis of Arts 42 and 48 (1) (Krisch 1337; Weston 520).

15  When authorizing States to use force on its behalf, the Security Council normally does not specify the Charter Article(s) on which it relies. Instead it authorizes the use of ‘all necessary means’ or ‘all necessary measures’ after reference to the fact that it is ‘acting under Chapter VII’, but without indicating the exact Charter Article. Examples include UNSC Resolution 678 (1990) of 29 November 1990 (para. 2; Iraq); UNSC Resolution 929 (1994) of 22 June 1994 (para. 3; Rwanda); UNSC Resolution 1080 (1996) of 15 November 1996 (SCOR 51^st Year 117; para. 3; Democratic Republic of the Congo); UNSC Resolution 1272 (1999) of 25 October 1999 (SCOR 54^th Year 130; para. 4; East Timor); UNSC Resolution 1386 (2001) of 20 December 2001 (SCOR [1 January 2001–31 July 2002] 272; para. 3; Afghanistan, Conflict); UNSC Resolution 1546 (2004) of 8 June 2004 (SCOR [1 August 2003–31 July 2004] 60; para. 10; Iraq, Invasion of [2003]; UNSC Resolution 1973 (2011) of 17 March 2011 (SCOR [1 August 2010–31 July 2011] 390, para. 4; Libya); and UNSC Resolution 2100 (2013) of 25 April 2013 (para. 18; Mali).).

16  In recent times pressure has developed for so-called humanitarian intervention. This is not addressed in the Charter, yet States and groups of States have sometimes asserted such a right without Security Council authorization. The classic example is the North Atlantic Treaty Organization (NATO) intervention over Kosovo in 1999. This was followed by efforts to develop the concept of responsibility to protect (‘R2P’),under which as a last resort the Security Council could, under certain conditions, authorize the use of force to protect populations from genocide, war crimes, ethnic cleansing, and crimes against humanity (2005 World Summit Outcome paras 138–39, reaffirmed by the Security Council in Resolution 1674 [2006] [28 April 2006] SCOR [1 August 2005–31 July 2006] 228). The Security Council authorized military interventions in Libya (UNSC Res 1373 [17 March 2011] para. 4) and Côte d’Ivoire (UNSC Res 1375 [30 March 2011] para. 6) have been hailed as concretizations of R2P, given their explicit aim of protecting civilians under (immanent) threat of attack. However, some, including the BRICS countries (Brazil, Russia, India, China, and South Africa) criticized NATO for interpreting the mandate to use force in Libya as an authorization to effect regime change, which was an abuse of the mandate. This disagreement about the scope of the military mandate in Libya has resulted in a backlash against the R2P concept within the Security Council. For example, attempts to adopt even weakly formulated measures in relation to the civil war in Syria (see France, Germany, Portugal, and the United States: Draft Resolution [4 October 2011]) have been blocked by China and Russia (Berman and Michaelsen, 350–57).

D.  Collective Security and Regional Organizations

17  Apart from authorizing individual States to enforce military measures on its behalf, the Security Council can also utilize regional organizations for this purpose. This follows from Art. 53 (1) UN Charter, which determines that the Security Council shall, where appropriate, utilize regional arrangements or agencies for enforcement action under its authority.

18  Since the 1990s the central role of the Security Council in relation to collective enforcement has been challenged by the actions of some regional organizations, which have engaged in military action in the absence of an explicit Security Council authorization. This was (in part) motivated by the widespread and systematic human rights atrocities committed in the States in which these organizations intervened.

19  In accordance with one line of argument, Art. 53 UN Charter constitutes a ‘right of emergency’ for regional organizations (Walter 261). Just as States can rely on the right to self-defence recognized in Art. 51 UN Charter in a case of an armed attack unless or until the Security Council takes action, regional organizations would have the power to intervene where the Security Council remains inactive in situations of gross and systematic human rights violations. This argument is underpinned by the rationale that the chances for abuse of the military mandate by a regional organization is unlikely, due to the institutional and collective control provided within the regional body, as well as by the higher degree of disinterest and objectivity within an organization composed of mutually independent States (Walter 262; Franck [2007] 25).

20  The line of argument favouring a ‘right of emergency’ for regional organizations contradicts the wording of the second sentence of Art. 53 (1) UN Charter, according to which no enforcement action shall be taken by regional organizations without authorization by the Security Council. The suggestion that the Security Council could prevent the regional organization from intervening by adopting a Chapter VII resolution turns the Charter system on its head, as it forces the Security Council to justify why it is not adopting military measures. In this way the Security Council is required to do the opposite of what is envisaged by the Charter system which is, in accordance with Art. 27 (3) UN Charter, based on an ‘opt-in procedure’ in the case of enforcement action, as opposed to an ‘opt-out’ procedure (De Wet 296). Second, any Chapter VII resolution intended to terminate the regional organization’s military action could be frustrated by a veto of a permanent member which is silently condoning the illegal military operation. A case in point was the attempt of Russia to terminate the air campaign in Kosovo. On 26 March 1999 Russia submitted a draft resolution that would have condemned the NATO military action as a breach of international law (see Belarus, India, and Russian Federation: Draft Resolution [26 March 1999]). However, this draft resolution, which was supported only by China and Namibia, failed to receive more than three votes.

21  Moreover, although the structure of the regional organizations and their collective decision-making procedures may lessen the possibility of abuse, the possibility remains. For example, it is possible that in situations where some members constitute a dominant economic and military force within a regional organization, they would be able to engineer a military intervention in accordance with their strategic interests. Questions have, for example, in the past been raised about the dominant role of Nigeria in the Economic Community of West African States (‘ECOWAS’) during its interventions in Liberia and Sierra Leone in the 1990s.

22  Despite these objections the African Union (AU) has formally claimed for itself the right to intervene in Member States in instances of gross human rights violations. In accordance with Art. 4 (h) Constitutive Act of the African Union, the organization may intervene in a Member State pursuant to a decision of the Assembly of Heads of State and Government in respect of grave circumstances, namely: war crimes, genocide, and crimes against humanity (Abass and Baderin 15). In accordance with Art. 7 (1) Constitutive Act of the African Union, the Assembly may take such a decision on the basis of a two-thirds majority. Apart from arguing that the AU would be claiming a ‘right of emergency’ for itself, it is also arguable that Art. 4 (h) constitutes a collective, ex ante form of intervention by invitation. Since the Member States of the AU have given their express consent to military intervention under certain conditions, the use of force would fall outside the scope of the prohibition in Art. 2 (4) UN Charter and not be in violation of the UN Charter. However, this argument in turn raises the question whether such an invitation can be extended for an open-ended period of time, or whether it has to be limited to a particular conflict (Abass and Baderin 17 and 19). Moreover intervention by invitation is already explicitly covered in Art. 4 (j) Constitutive Active of the African Union, which provides the right of ‘Member States to request intervention from the Union in order to restore peace and security’. It is therefore unlikely that Art. 4 (h) would also relate to intervention by invitation, as Art. 4 (h) would then amount to a mere repetition of Art. 4 (j) Constitutive Act. For the time being, however, the debate pertaining to Art. 4 (h) remains academic as it is yet to be relied on in practice. Thus far, all military interventions under the authority of the African Union have been based on the explicit invitation of the African government concerned and often accompanied by a Chapter VII United Nations Security Council resolution (De Wet [2013] forthcoming).

23  Throughout the 1990s, regional organizations have occasionally sought to justify military interventions on the basis of so-called ex post facto authorizations by the Security Council (Franck [2007] 25). Even though this does not find any textual basis in the UN Charter, it cannot be excluded that the Security Council could develop a practice of retroactive authorization. However, in order for such an authorization to be convincing, it would have to be given in unambiguous terms. Otherwise regional organizations or States could attempt to justify unauthorized, unilateral interventions on the basis of obscure language in subsequent Security Council resolutions which were not intended for that purpose. Examples of (albeit ambiguous) attempts to justify military interventions by regional organizations on the basis of ex post facto Security Council authorization include the interventions in the 1990s by ECOWAS in Liberia and Sierra Leone.

E.  Conclusion

24  The UN Charter system of collective security has undergone significant evolution since 1945. Especially since the end of the Cold War, the authorization of willing and able States or regional organizations to use military force on behalf of the UN has become a well-established substitute for the Art. 43 agreements foreseen in the UN Charter. In the post-Cold War era it provides a pragmatic way to facilitate military operations in a fashion that also takes account of the military complexities surrounding a military intervention, such as the need for unified command and control. Some have questioned whether the international law on the use of force as set out in the UN Charter is still relevant today, in the face of modern threats. Yet in September 2005, the Heads of State and Government meeting in the UN General Assembly solemnly declared ‘that the relevant provisions of the Charter are sufficient to address the full range of threats to international peace and security’ (World Summit Outcome para. 79; Wood 4).

25  At the same time, the central role of the Security Council in maintaining international peace and security is facing significant challenges, not least because of the persistent perception of the abuse of the veto by the permanent members. As a result, there is a tendency amongst economically and militarily strong States and regional organizations to circumvent the perceived ineffectiveness of the Security Council through various mechanisms. These range from the expansive interpretation of open-ended military mandates and the scope of the right to self-defence, to attempts to develop new customary exceptions to Art. 2 (4) UN Charter, such as a residual right of intervention for regional organizations and the right of States to exercise a ‘responsibility to protect’ in instances of genocide, crimes against humanity and war crimes. Although most of these developments are highly controversial, their presence in the discourse may lead to a weakening of the central authority of the Security Council in the maintenance of international peace and security.

26  Finally, it is worth noting that there has been a renewed academic debate as to whether there could be a role for the General Assembly in instances where the Security Council fails to authorize the use of force in the face of widespread human rights atrocities. Such a role would seem to contravene Art. 11 (2) UN Charter, which reserves (all aspects of) enforcement action exclusively to the Security Council (Certain Expenses of the United Nations [Advisory Opinion] 164–65), and to be beyond the powers of the General Assembly which are only recommendatory. Unlike the Security Council, the General Assembly has no power of authorization. One could, however, attempt to justify such a role with the argument that the General Assembly would be exercising an emergency role for which it had already set a precedent of a sort with the adoption of the Uniting for Peace Resolution (1950) at the time of the Korean War (Österdahl [1999] 133 et seq). The United Nations Secretary-General has, however, cautioned against measures that strayed from paras 138–39 2005 World Summit Outcome. These paragraphs underscored the central role of the United Nations Security Council in protecting the civilian population where the respective State failed to do so (UNGA Report of the Secretary-General [12 January 2009] ‘Implementing the Responsibility to Protect’ UN Doc A/63/377, para. 67).