Jump to Content Jump to Main Navigation
The Convention on Cluster Munitions - A Commentary edited by Nystuen, Gro; Casey-Maslen, Stuart (21st October 2010)

Commentary, Art.2 Definitions

Bonnie Docherty, Lou Maresca, Richard Moyes, Markus Reiterer

From: The Convention on Cluster Munitions: A Commentary

Edited By: Gro Nystuen, Stuart Casey-Maslen

Subject(s):
Weapons, conventional

(p. 144) Article 2.  Definitions

Article 2— Definitions

For the purposes of this Convention:

  1. 1.  ‘Cluster munition victims’ means all persons who have been killed or suffered physical or psychological injury, economic loss, social marginalisation or substantial impairment of the realisation of their rights caused by the use of cluster munitions. They include those persons directly impacted by cluster munitions as well as their affected families and communities;

  2. 2.  ‘Cluster munition’ means a conventional munition that is designed to disperse or release explosive submunitions each weighing less than 20 kilograms, and includes those explosive submunitions. It does not mean the following:

    1. (a)  A munition or submunition designed to dispense flares, smoke, pyrotechnics or chaff; or a munition designed exclusively for an air defence role;

    2. (b)  A munition or submunition designed to produce electrical or electronic effects;

    3. (c)  A munition that, in order to avoid indiscriminate area effects and the risks posed by unexploded submunitions, has all of the following characteristics:

      1. (i)  Each munition contains fewer than ten explosive submunitions;

      2. (ii)  Each explosive submunition weighs more than four kilograms;

      3. (iii)  Each explosive submunition is designed to detect and engage a single target object;

      4. (iv)  Each explosive submunition is equipped with an electronic self-destruction mechanism;

      5. (v)  Each explosive submunition is equipped with an electronic self-deactivating feature;

  3. (p. 145) 3.  ‘Explosive submunition’ means a conventional munition that in order to perform its task is dispersed or released by a cluster munition and is designed to function by detonating an explosive charge prior to, on or after impact;

  4. 4.  ‘Failed cluster munition’ means a cluster munition that has been fired, dropped, launched, projected or otherwise delivered and which should have dispersed or released its explosive submunitions but failed to do so;

  5. 5.  ‘Unexploded submunition’ means an explosive submunition that has been dispersed or released by, or otherwise separated from, a cluster munition and has failed to explode as intended;

  6. 6.  ‘Abandoned cluster munitions’ means cluster munitions or explosive submunitions that have not been used and that have been left behind or dumped, and that are no longer under the control of the party that left them behind or dumped them. They may or may not have been prepared for use;

  7. 7.  ‘Cluster munition remnants’ means failed cluster munitions, abandoned cluster munitions, unexploded submunitions and unexploded bomblets;

  8. 8.  ‘Transfer’ involves, in addition to the physical movement of cluster munitions into or from national territory, the transfer of title to and control over cluster munitions, but does not involve the transfer of territory containing cluster munition remnants;

  9. 9.  ‘Self-destruction mechanism’ means an incorporated automatically-functioning mechanism which is in addition to the primary initiating mechanism of the munition and which secures the destruction of the munition into which it is incorporated;

  10. 10.  ‘Self-deactivating’ means automatically rendering a munition inoperable by means of the irreversible exhaustion of a component, for example a battery, that is essential to the operation of the munition;

  11. 11.  ‘Cluster munition contaminated area’ means an area known or suspected to contain cluster munition remnants;

  12. 12.  ‘Mine’ means a munition designed to be placed under, on or near the ground or other surface area and to be exploded by the presence, proximity or contact of a person or a vehicle;

  13. 13.  ‘Explosive bomblet’ means a conventional munition, weighing less than 20 kilograms, which is not self-propelled and which, in order to perform its task, is dispersed or released by a dispenser, and is designed to function by detonating an explosive charge prior to, on or after impact;

  14. (p. 146) 14.  ‘Dispenser’ means a container that is designed to disperse or release explosive bomblets and which is affixed to an aircraft at the time of dispersal or release;

  15. 15.  ‘Unexploded bomblet’ means an explosive bomblet that has been dispersed, released or otherwise separated from a dispenser and has failed to explode as intended.

(p. 148) Overview of the Article

2.1  Article 2 sets out a number of definitions ‘for the purposes of this Convention’.2 Two of these definitions were derived either verbatim or with slight modification from the 1997 Anti-Personnel Mine Ban Convention:3 specifically ‘mine’ and ‘transfer’.4 Others were adapted from related definitions in 1996 Amended Protocol II5 and 2003 Protocol V6 annexed to the Convention on Certain Conventional Weapons (CCW).7 However, the definitions of cluster munition victims, and of cluster munition, explosive submunition, explosive bomblet, and dispenser were largely elaborated within the negotiation of the Convention on Cluster Munitions itself.

Preparatory discussions and negotiations

2.2  Preparatory discussions and the negotiations of Article 2 were generally divided into three ‘streams’: one focusing on the definition of cluster munition victims (paragraph 1),8 the second on the definition of cluster munition (paragraphs 2 and 3),9 and the third on ‘other’ definitions (the remaining (p. 149) definitions, i.e. paragraphs 4–15).10 The definition of ‘cluster munition’, and certain related definitions such as that of ‘explosive submunition’ were closely linked to the General Obligations and Scope of the Convention as articulated in Article 1.

2.3  There remain other terms within the Convention, including those used within certain definitions themselves, which would have warranted further clarification. These include terms such as ‘assist’, ‘in a position to do so’, and ‘air defence role’.11 Given differing interpretations among States, however, negotiating a definition of some of these terms might have proved contentious.

Cluster munition victims

Paragraph 1

‘Cluster munition victims’ means all persons who have been killed or suffered physical or psychological injury, economic loss, social marginalisation or substantial impairment of the realisation of their rights caused by the use of cluster munitions. They include those persons directly impacted by cluster munitions as well as their affected families and communities;

Overview

2.4  The definition of ‘cluster munition victim’ sets the stage for the Convention’s victim assistance provisions in Article 5. On a technical level it determines the group of people to whom States are obliged to provide victim assistance. More broadly, it informs the essence of victim assistance, for assistance to individuals killed or physically injured by the weapon will be different from assistance to others in the family or community who have not been physically injured by the weapons, but who are nonetheless significantly affected by them. The definition is broad, enumerating harm beyond physical (p. 150) injury to include psychological suffering, economic loss, social marginalization, and impairment of the realization of rights. As is seen in the commentary on Article 5, the substantive obligation to provide victim assistance has to respond to all those different types of harm.

Preparatory discussions

2.5  As noted above, the definition of cluster munition victims was generally discussed as an integral part of the victim assistance package which encompassed Article 5 itself.12 In the course of the conferences in Lima, Belgrade, Brussels, and Vienna13 there were regular calls for a definition of the term cluster munition victims to be included. This was done for the first time in the Draft Cluster Munitions Convention presented to the Wellington Conference. Some elements of the definition, however, had been introduced into the Oslo Process at a fairly early stage, as discussed below.

2.6  The Oslo Declaration committed the participating States to the establishment of ‘a framework for cooperation and assistance that ensures adequate provision of care and rehabilitation to survivors and their communities’.14 The drafters of the Vienna Discussion Text—while not introducing a definition of cluster munition victims proper—wished to highlight a broad understanding of victims by including a preambular paragraph referring to ‘victims of cluster munitions, which inter alia include the persons directly affected, their families and communities’.15 This formulation reflected discussions on victim assistance that had previously taken place in Lima, Belgrade, and Brussels.

2.7  At the end of the European Conference on Cluster Munitions held in Brussels on 31 October 2007, the rapporteur of the victim assistance session was able to report that:

there was a general understanding that victim assistance is a broad and comprehensive concept, which should use as a starting point, but also as a constant reference point the needs and rights of victims. [ … ] Already the term victim as such is to be understood broadly, i.e. encompassing the survivor as such—the victim of the direct impact, but also other victimized persons, including family and affected communities.16

(p. 151) This principled approach was essentially supported during the Vienna Conference and formed part of what the President of the Vienna Conference, Ambassador Wolfgang Petritsch, called the Vienna Consensus on victim assistance, which reinforced the endeavours to establish victim assistance as ‘a key legal obligation of the same quality as the other main building blocks of the future treaty’17 and hence work towards a definition of the term ‘cluster munition victims’.

2.8  The Draft Cluster Munitions Convention for the first time contained a draft definition of cluster munition victims, which was placed as the first definition of Article 2 thereby highlighting the importance attached to the humanitarian cause pursued by the Oslo Process.18 The definition contained in the draft Convention read as follows:

‘Cluster munition victims’ means persons who have suffered physical or psychological injury, economic loss, social marginalisation or substantial impairment of the realisation of their rights caused by the use of cluster munitions. They include those persons directly impacted by cluster munitions as well as their families and communities.

The draft essentially followed the assumption that any victim assistance provision would have to use as a starting and constant reference point the needs and rights of the victims. While the definition highlights the injury or loss incurred by a victim of cluster munitions, Article 5 concentrates on how best to respond to the needs arising from such injury or loss. The draft definition sought to paint a realistic picture of victimization by portraying the defining factors that lead to the assumption that a particular person in fact has become a victim of the weapon.

2.9  The draft definition built on the discussions outlined above and the description of the term ‘mine victim’ as contained in Paragraph 64 of the so-called Review Document adopted at the 2004 Nairobi Summit on a Mine- Free World, which states:

It is now generally accepted that victims include those who either individually or collectively have suffered physical or psychological injury, economic loss or substantial (p. 152) impairment of their fundamental rights through acts or omissions related to mine utilization.19

The definition in the Draft Cluster Munitions Convention also drew on the definition of ‘victims of gross violations of international human rights law and serious violations of international humanitarian law’, contained in the Basic Principles and Guidelines on the Right to a Remedy and Reparation for Victims of Gross Violations of International Human Rights Law and Serious Violations of International Humanitarian Law.20 That definition reads as follows:

For purposes of the present document, victims are persons who individually or collectively suffered harm, including physical or mental injury, emotional suffering, economic loss or substantial impairment of their fundamental rights, through acts or omissions that constitute gross violations of international human rights law, or serious violations of international humanitarian law. Where appropriate, and in accordance with domestic law, the term ‘victim’ also includes the immediate family or dependants of the direct victim and persons who have suffered harm in intervening to assist victims in distress or to prevent victimization.21

2.10  The Draft Cluster Munitions Convention signified two major developments as compared to the texts mentioned above. First, in addition to the concepts of physical and psychological injury, and economic loss or substantial impairment of the realization of rights, the draft text also included the concept of social marginalization as a form of victimization of a person. Second, the draft text developed the idea contained in the phrase ‘either individually or collectively have suffered’ by introducing a specific reference to the families and communities of persons directly impacted by cluster munitions.

2.11  This draft definition did not remain without criticism. During the Wellington Conference Switzerland proposed to delete the entire definition, expressing concern that the inclusion of a definition might facilitate discrimination against victims of other weapons or explosive remnants of war.

(p. 153) Switzerland argued that:

the present wording of Article 5 and 2 inappropriately reinforces the notion that ‘cluster munition victims’ should be considered a separate group. This contradicts one of the essential principles of victim assistance, namely the principle of non-discrimination.22

The delegation of the United Kingdom (UK) saw ‘a certain logic’ in the Swiss proposal not to include a definition of the term in the treaty text, but concluded that ‘it might be legally more prudent’ to have one. The main concerns expressed by the UK, however, related to the broadness of some elements of the definition, i.e. the reference to families and communities as well as social marginalization and the substantial impairment of the realization of the victims’ rights. In addition, the UK raised the question whether the definition should cover both civilians and military personnel.23

2.12  The subsequent discussions concentrated primarily on the question of including families and communities under the definition of cluster munition victims. Guatemala rightly pointed out that already the Oslo Declaration referred to the ‘survivors and their communities’; it further argued that the text of the treaty would fall short of the Oslo Declaration, if such a reference were not included.24 Croatia supported the inclusion of families and communities as this ‘reflects the reality on the ground’.25 Similarly, the delegations of the Lao People’s Democratic Republic (Laos), Morocco, Norway, Canada, and the Cluster Munition Coalition (CMC) expressly supported the reference to families and communities.

2.13  While the Wellington Conference did not make any changes to the draft text as such, two proposals relating to the definition of cluster munition victims were included in the Compendium: the Swiss proposal to delete the definition, and a proposal by the UK reflecting its concerns as described above.26 The UK proposal did not mention either social marginalization or the impairment of the realization of rights, and endeavoured to avoid the concern expressed as to families and communities by replacing the definition’s second sentence by the following formulation: ‘cluster munition victims include such (p. 154) persons directly impacted by cluster munitions’.27 By stating that victims include persons directly impacted, however, the UK signalled that also other entities than those individuals could be regarded as victims of cluster munitions.

Negotiations

2.14  While the preparatory work as well as the negotiations in Dublin focused mostly on the text of Article 5 of the draft Convention, the definition was the subject of a number of informal discussions undertaken at the Diplomatic Conference by the Friend of the President on Victim Assistance with interested delegations. The Swiss delegation signalled its readiness to lift its objection against the definition as a whole, but wished to see the inclusion of clear language aimed at preventing discrimination among victims of different weapons and explosive remnants of war. Switzerland’s concern was essentially solved by introducing a provision that later became Article 5, paragraph 2(e), and was supported by a significant number of delegations as well as the International Committee of the Red Cross (ICRC).

Families and Communities

2.15  The question of the reference to families and communities was also discussed in informal consultations held by the Friend of the President on Victim Assistance and two open-ended consultations convened during the first week of the Diplomatic Conference. Still the broadness of the concept remained at the centre of discussions, which were held in a constructive and cooperative spirit by all participating delegations.

2.16  At the outset of the consultations the UK stated that it was ‘broadly happy’ with the draft definition, but still expressed concerns about the broadness of the concepts of family and community. The UK therefore suggested additional wording at the end of the second sentence to limit the scope of the families and communities to those in areas affected by cluster munitions. The UK proposal read as follows:

They include those persons directly impacted by cluster munitions, as well as their families and communities in areas affected by cluster munitions.28

(p. 155) 2.17  In reaction, a number of delegations expressed scepticism about the introduction of a territorial approach to the victim definition, noting that victims often have to move away from affected areas29 and that ‘persons, families and communities that are really affected need to receive assistance regardless of territory’.30 This point was also reiterated by Croatia, which stated:

What is clear is that people, who have been victimized, whether they remain or leave an area or are forced to move about, must receive assistance.31

2.18  While the Friend of the President acknowledged the concerns expressed by the UK and others concerning the broadness of the concepts, he also reiterated the problems relating to a geographical limitation of the victim definition. To avoid these problems he suggested formulating a material rather than geographic limitation. The Canadian delegation during the first meeting of the Committee of the Whole on victim assistance had already proposed to solve this problem by adding after the terms ‘families and communities’ at the end of the definition’s second sentence the phrase ‘that have been materially and demonstrably affected by cluster munitions’.32 While this proposal provided a general direction in which a solution to the problem was eventually found, the formulation itself was criticized as being highly limited and overly subjective.33 Finally, Australia supported by Argentina, Belgium, Canada, France, Germany, and others proposed simply to include the term ‘affected’ before ‘families and communities’; this suggestion was subsequently accepted so that the final version of the last sentence reads:

They include those persons directly impacted by cluster munitions as well as their affected families and communities.

Persons killed; irrelevance of the victims’ status

2.19  The Philippines raised two further issues relating to the definition. First, it proposed a clarification whereby the death of persons as a result of the use of cluster munitions would also be covered by the definition. (p. 156) Second, it suggested that it be expressly clarified that ‘migrants under the jurisdiction and control of an affected State’ would fall under the ambit of the definition.34

2.20  With respect to the first suggestion, the Philippines argued that specific rules of international humanitarian law are relevant in cases of death as a result of weapon use and therefore this element would need to be reflected in the definition. Although it was argued that death caused by cluster munitions could be regarded as the most severe form of physical injury and would therefore already be covered by the definition,35 a short discussion revealed broad support for this proposal so that the phrase ‘who have been killed’ was included in the draft definition after the term ‘persons’. The inclusion of this phrase also shows the significance of the broadening of the definition to include families and communities: while assistance to persons who have been killed would be devoid of meaning were the definition only to take into consideration individuals, the inclusion also of families and communities in the definition makes it clear that the victim assistance obligation is also relevant for those whose victimization is the result of the demise of one of its members.

2.21  The second proposal by the Philippines relating to migrants was initially discussed in a series of informal bilateral consultations conducted by the Friend of the President and settled during the second open-ended consultations on victim assistance that took place on 22 May 2008. It was argued that this proposal created the risk of establishing a shopping list of a potentially unlimited number of groupings such as migrants, refugees, internally displaced persons, nationals of a State Party, nationals of States not party, etc. The definition, however, from the outset was intended to be all-embracing and non-discriminatory.36 Finally, the Philippines supported by Croatia and Serbia proposed to solve this issue by inserting the term ‘all’ before ‘persons’ in the definition’s first sentence.

(p. 157) 2.22  With these amendments the informal consultations reached a common understanding on the following definition, which subsequently was formally adopted by the Conference:

‘Cluster munition victims’ means all persons who have been killed or suffered physical or psychological injury, economic loss, social marginalisation or substantial impairment of the realisation of their rights caused by the use of cluster munitions. They include those persons directly impacted by cluster munitions as well as their affected families and communities.

In introducing the Non-Paper containing this definition to the Committee of the Whole the Friend of the President specifically referred to the formulation ‘all persons’ as clarifying that the definition covers all persons ‘who had suffered harm regardless of their status as migrants, refugees, Internally Displaced Persons, etc’.37

Commentary

2.23  The definition of cluster munition victims contains the defining elements of victimization caused by the use of cluster munitions and in doing so paints a realistic picture of the effects the use of such munitions may have on individuals as well as families and communities. The history of the negotiations as described above reveals a number of principles which were followed during the negotiations: pursuit of a broad approach by including the various ‘circles’ of victims38 in the definition; reflecting as accurately as possible the reality of victimization; and irrelevance of status of the victim.

‘all persons’

2.24  The use of the term all persons in the definition makes clear that any person irrespective of his/her status has to be regarded as a cluster munition victim if the defining elements contained in the definition’s first sentence apply, i.e. that he/she has been killed or suffered physical or psychological injury, economic loss, social marginalization or substantial impairment of the realization of their rights caused by the use of cluster munitions. This interpretation, which already derives from the ordinary meaning of the term, is also confirmed by the negotiating history. As described above, the word ‘all’ had been inserted to ensure that all persons ‘who had suffered harm regardless of their status as migrants, refugees, Internally Displaced Persons, etc.’ are to be (p. 158) regarded as cluster munition victims. The definition also makes it clear that there is no difference as to whether a person was a civilian or a combatant at the time he/she became a cluster munition victim.

‘killed or suffered physical or psychological injury, economic loss, social marginalisation or substantial impairment of the realisation of their rights’

2.25  This formulation lists the defining elements of victimization: in addition to the more obvious elements of death or physical or psychological injury that lead to the qualification as victims, these defining elements also include other less obvious, yet serious elements: economic loss, social marginalization, or substantial impairment of the realization of rights. The use of the term ‘or’ between the various defining elements makes clear that it is sufficient that only one of those elements needs to apply in a given case in order for a person to fall under the scope of the definition. It is not necessary therefore to have suffered physical or psychological harm plus economic loss, social marginalization, or substantial impairment of the realization of rights: either of these elements is sufficient as long as its occurrence is caused by the use of cluster munitions. Economic loss and social marginalization are common factors of victimization. For example, suffering from a disability in many societies leads to social stigmatization of the individual in question as well as his/her family or often larger community.

2.26  The formulation ‘substantial impairment of the realization of their rights’ in essence is a further development of the formulations used in both the Nairobi Review Document and the Guidelines adopted by the UN General Assembly: substantial impairment of their fundamental rights. There are two main distinctions, as discussed below.

2.27  First, Article 2, paragraph 1 does not refer to an impairment of rights, but an impairment of the realization of rights. In doing so, the text defines more precisely the subject of impairment by acknowledging that the main problem for victims is not so much an impairment of the rights as such— which regularly continue to be the same before and after the use of cluster munitions—but the impediment to actually realize those rights. For example, while in a democratic society victims and ‘non-victims’ have a right to vote in elections, it might be much harder for a cluster munition victim actually to participate in the vote which may be, for instance, the result of a physiological barrier (e.g. victim has lost sight through cluster munition use) or the land denial caused by cluster munition use makes it extremely difficult to have access to a voting station.

(p. 159) 2.28  Second, Article 2, paragraph 1 does not restrict impairment to only fundamental rights, but covers rights in general. This provision is, therefore, broader than merely covering those rights which might be deemed fundamental, such as the right to life, the prohibition of torture, or the right to freedom of movement. Any subjective right, not just fundamental rights, of the person in question may be the object of the substantial impairment.

‘caused by the use of cluster munitions’

2.29  This formulation clarifies that the detrimental effects enumerated in the definition need to be caused by the use of cluster munitions. In other words there needs to be a causal link between the use of the weapon and the occurrence of the detrimental effect to establish victimisation in the meaning of Article 2, paragraph 1. Not all detrimental effects related to the use of cluster munitions qualify for victimization. This formulation essentially functions as a crucial qualifier and puts necessary limits to the broadness of the definition.

‘affected families and communities’

2.30  According to the second sentence of the definition not only the individual human being falls under the ambit of the definition of cluster munition victim, but also affected families and communities, thus going beyond the scope of application of the Convention on the Rights of Persons with Disabilities which does not apply to persons other than those with disabilities themselves.39 The inclusion of these concepts in the definition was at the centre of negotiations on the definition, as some delegations expressed the concern that these concepts were overly broad.

2.31  The formulation adopted clarifies that in order to fall under the definition of a cluster munition victim a family or community needs to be affected in reality by the use of cluster munitions. As the second sentence of the definition commences with the phrase ‘They include’, it clearly refers back to the term ‘persons’ contained in the definition’s first sentence. Hence, families and communities have to undergo any of the adverse effects enumerated in the definition to qualify as a victim: in other words, in order for a family or community to qualify as victim it needs to have suffered from either psychological injury, economic loss, social marginalization, or substantial impairment of the realization of their rights caused by the use of cluster munitions. Death or physical injury as such might seem not to be directly applicable when victimization relates to families or communities, however the death or physical (p. 160) injury of a family or community member may result in psychological injury, economic loss, social marginalization, or substantial impairment of the realization of their rights caused by the use of cluster munitions and therefore lead to victimization of the family or community.

2.32  In the absence of an internationally agreed definition of the terms family or community, it is to be assumed that in fulfilling its obligation vis-àvis families and communities as cluster munition victims the respective State Party will have a rather wide margin of appreciation and that the question who belongs to a certain family or community will eventually have to be clarified through the internal legal order of the respective State Party. The situation is similar to, for example, the use of the term ‘family member’ in Article 37 of the 1961 Vienna Convention on Diplomatic Relations, which also contains no definition of the term and hence leaves a wide margin of appreciation to each State Party.

The definition of a cluster munition

Overview

2.33  The definition of cluster munition delineates the technological category that is subject to the prohibitions contained in Article 1 and subject to obligations and provisions throughout the treaty. In general terms, a cluster munition means a conventional munition designed to disperse or release explosive submunitions—conventional munitions dispersed and released by cluster munitions and which are designed to detonate an explosive charge before, upon, or following impact, typically with the target object or the ground40—each weighing less than 20 kilograms. However, a series of exclusions set out in sub-paragraphs (a) to (c) restricts the weapons that are captured by the general definition in the chapeau of Article 2, paragraph 2. The scope of the prohibition is also limited by Article 1, paragraph 3, which stipulates that the Convention does not apply to ‘mines’. In contrast, at least the core prohibitions of the Convention are applied mutatis mutandis to explosive bomblets, under Article 1, paragraph 2.41

(p. 161) 2.34  The Convention would appear to prohibit all conventional weapons with explosive submunitions that had been evidenced, at the time of the negotiations, as resulting in civilian casualties either at the time of their use or subsequently.42 For certain types of submunition-based weapons, the definition has both effects-based and technical requirements to determine their permissibility. Where numbers are used in the definition they arguably facilitate a clear determination of what falls inside or outside the definition.43

2.35  The main definition and sub-paragraph (a) both exclude from prohibition certain systems that employ submunitions and that are weapons in the common usage of the term, or which might be considered or used as weapons (e.g. weapons with submunitions weighing more than 20 kilograms and air-defence weapons.) No further technical requirements are demanded of weapons excluded through the main definition or at subparagraph (a). Sub-paragraphs (a) and (b) also exclude from prohibition other military systems that are not weapons in the common usage of the term but that may be used to create illuminating, obscuring, or diversionary effects. As described above, however, the greatest focus in negotiations was on sub-paragraph (c).

2.36  Sub-paragraph (c) provides a statement of the problematic effects that should be avoided in order to warrant exclusion from prohibition: ‘indiscriminate area effects and the risks posed by unexploded submunitions’. This formulation should be read in conjunction with the preamble wherein the intent of the Convention is said to be, inter alia, ‘to put an end for all time to the suffering and casualties caused by cluster munitions at the time of their use, when they fail to function as intended or when they are abandoned’. Thus the problematic humanitarian effects of cluster munitions are understood to be both ‘indiscriminate area effects’ at time of use as well as ‘risks posed by unexploded submunitions’ if they ‘fail to function as intended’ as well as the general risk posed by all explosive weapons if they are left abandoned in the post-conflict environment.

(p. 162) 2.37  In order to avoid both ‘indiscriminate area effects and the risks posed by unexploded submunitions’ the munition must have five technical characteristics relating to: the number of submunitions, the minimum weight of the submunitions, the capacity of individual submunitions to detect and engage targets, and the presence of both electronic self-destruct and self-deactivation features. Taken together these characteristics should reduce sufficiently the likelihood of these munitions creating a significant humanitarian problem.

2.38  The technical characteristics listed in paragraph 2(c) (i) to (v) can exclude a weapon from being defined as a cluster munition and thereby from prohibition. However, weapons that meet the technical characteristics but that when used are found to generate ‘indiscriminate area effects or risks posed by unexploded submunitions’ because the technical characteristics do not function as intended could be examined within the context of the Convention, for example at the Meeting of States Parties or a Review Conference. The explicitly stated intent of the exclusion in the chapeau of Article 2, paragraph 2(c) necessitates the effective functioning of the technical characteristics. If, for example, a weapon excluded under paragraph 2(c) turned out to have self-destruct mechanisms that did not work, it may no longer be justified to exclude it from the prohibition. Such an interpretation is supported by statements by Austria, Lebanon, Malta, and the CMC during the negotiation of the Convention44 and implicitly supported in other proposals.45

Preparatory discussions

2.39  Article 2, paragraph 2 was a focus of extensive discussions during all of the preparatory meetings for the Dublin Diplomatic Conference, and during the final negotiations. The February 2007 Oslo Conference brought together States most of which had previously indicated support for prohibitions on at least certain types of cluster munitions. During this conference a number of States called for a ban on ‘all cluster munitions’, while others expressed a clear (p. 163) sense that only certain types might warrant prohibition. Other participants noted that what might count as ‘all cluster munitions’ would be dependent upon how a future treaty would define this term.46

2.40  The outcome document of this meeting, the Oslo Declaration, committed States to conclude by 2008 a legally binding international instrument that would ‘prohibit the use, production, transfer and stockpiling of cluster munitions that cause unacceptable harm to civilians’.47 However, there was no commonly accepted definition of a cluster munition and no common agreement as to how the categories of ‘cluster munition’ and ‘cluster munitions that cause unacceptable harm to civilians’ related to each other. The lack of a commonly agreed or accepted definition at the outset of the Oslo Process led to some tough negotiations.

2.41  The diversity of possible definitions was reflected in a paper prepared by the Geneva International Centre for Humanitarian Demining (GICHD) in June 2007,48 which listed definitions and descriptions of cluster munitions and related terms that had been adopted or proposed by different bodies. Such definitions and descriptions varied in both basic terms and structures. Some linked the term cluster munitions to the container whereas others applied the term to the container and its submunitions as a system. Some made explicit statements about the delivery system for the weapons and others described how the submunitions would create an effect within the target area. The GICHD paper illustrated the numerous possible approaches to defining what was—and what was not—a cluster munition for the purpose of the future Convention.

2.42  A persistent discussion within the preparatory meetings leading to the Dublin Diplomatic Conference was the extent to which the clause ‘that cause unacceptable harm to civilians’ necessitated a category distinct from ‘cluster munitions’. A number of States suggested a distinction between ‘cluster munitions’ and ‘cluster munitions that cause unacceptable harm to civilians’.49 Yet, despite the importance of the latter term to the structure of the negotiations, (p. 164) no State chose to delineate what might be considered acceptable harm.50 The subsequent debate also reflected a number of themes that were to be discussed in later meetings of the Oslo Process, notably:

  • •  the possibility of limiting the number of submunitions;

  • •  the need to establish ‘reliability’ and ‘accuracy’ criteria for any weapon to be exempted from the ambit of the future convention;

  • •  the potential utility of self-destruct mechanisms and self-neutralization mechanisms;

  • •  the questionable utility of testing reliability rates as an indicator of combat performance;

  • •  the potential prohibition on use in populated areas of any cluster munitions not prohibited by the convention; and

  • •  that the burden of proof was on those seeking exclusions from prohibition to justify their claims.51

Although certain States endorsed particular technical characteristics as providing grounds for confidence that humanitarian problems would be avoided or sufficiently reduced in the future, there was little done to interrogate the relationship between certain technologies and the actual or likely outcomes for civilians during and after conflict. A significant exception to this was an analysis undertaken by Norwegian People’s Aid, the Norwegian Defence Research Establishment, and Colin King Associates into the reliability of M85 submunitions both in testing and operational conditions.52

2.43  The Lima Discussion Text already included a categorical prohibition on the use, production, stockpiling, and transfer of cluster munitions under Article 1. Article 2 of that text stated that ‘the following weapons systems shall be considered prohibited cluster munitions under this treaty’:

Air carried dispersal systems or air delivered, surface or sub-surface launched containers, that are designed to disperse explosive sub-munitions intended to detonate following separation from the container or dispenser, unless they are designed to, manually or automatically, aim, detect and engage point targets, or are meant for smoke or flaring, or unless their use is regulated or prohibited under other treaties.

(p. 165) 2.44  The definition of cluster munition presented in the Lima Discussion Text provided an initial indication of how the Oslo Declaration language of ‘cluster munitions that cause unacceptable harm to civilians’ might be interpreted. In Article 1, it adopted prohibitions on the use, production, and stockpiling of cluster munitions as a defined category of weapons. It then presented a broad definition of cluster munitions, including air-carried ‘dispensers’ as well as ‘containers’, but recognized that certain items that might fall under the principal definition should not be subject to prohibition. This definition excluded from prohibition systems that produce smoke or flare effects. It also excluded systems with submunitions designed to ‘aim, detect and engage point targets’. With respect to such systems, however, the definition applied no further controls that might be expected to limit their propensity to become unexploded ordnance (UXO). It is also important to note that the definition in the Lima text considered cluster munitions to be the ‘dispersal system’, i.e. the dispenser or container, rather than the combination of both container and submunitions that was adopted in subsequent definitions.53

2.45  The definition also excluded from prohibition weapons subject to regulation or prohibition under other treaties. Such a formulation was intended to avoid problems caused by the category of cluster munitions overlapping with other weapon categories, such as remotely delivered landmines. It was also noted, however, that the general rules of international humanitarian law regulate all weapons.54 At a more specific level this approach would have created the potential for weapons initially prohibited under the Convention on Cluster Munitions to later fall outside the scope of the prohibition if subject to regulation under another legal instrument. Subsequent draft texts excluded mines explicitly from the scope of the treaty rather than as a component of the definition of cluster munition.

2.46  Discussions of the definition at the May 2007 Lima Conference were chaired by Ambassador Don MacKay of New Zealand, who would subsequently chair discussions on definitions at the Vienna Conference, serve as President of the Wellington Conference on Cluster Munitions, and act as Friend of the President on definitions at the Dublin Diplomatic Conference. In introducing the definition in the Lima Discussion Text, MacKay highlighted to participants that while the definition related to (p. 166) cluster munitions as a whole category, certain weapons with submunitions were not prohibited.55

2.47  Some States explicitly argued that the Lima definition was too broad and would prohibit weapons without justification, while others argued that the definition was not wide enough or pressed for a ‘total ban’.56 The definition proposed in the Lima Discussion Text was similar in its implications for current weapon systems to the one ultimately adopted in the Convention. In many respects the definition was reinforced during the course of the Oslo Process, in particular in the way that the exclusion for weapons that detect and engage so-called ‘point targets’ was drafted. However, in some other areas the definition was to become less comprehensive, by providing exemptions for submunitions over a certain weight and for so-called ‘air-defence’ systems. It has been argued that the strengthening of the definition was facilitated in part by the influence of the revised structure of a definition, presented at the December 2007 Vienna Conference, on the subsequent discussions and negotiations.57

2.48  Thus, the Vienna Discussion Text58 presented a wholesale restructuring of the definition into the form that was to shape the subsequent negotiations. Rather than stipulating the precise boundaries of what was and was not a cluster munition, the Vienna text introduced a broad definition but left open three serial lines for systems that should not be subject to prohibition (see below). Structured in this way, the text invited States to argue into the text systems that should be considered ‘acceptable’ against a categorical presumption that such systems cause ‘unacceptable harm’:

‘Cluster munition’ means a munition that is designed to disperse or release explosive sub-munitions, and includes those explosive sub-munitions. It does not mean the following:

  1. (a) 

  2. (b) 

  3. (c) 

2.49  Significantly, cluster munitions were now defined as ‘munitions’ that comprised both a container and the submunitions it contained. This served (p. 167) to clarify certain elements of the text by ensuring that any prohibitions that applied to ‘cluster munitions’ applied to both the containers and the submunition contents of those containers. However, this approach also produced challenges, notably its apparent exclusion of systems that used dispensers affixed to aircraft (which could not readily be considered ‘munitions’),59 and risks of circularity in the definition of cluster munition and explosive submunition. The definition also now used the phrase ‘dispersed or released’, a phrase previously used in a definition of cluster munitions contained in a paper circulated by UN agencies for discussions in the context of the CCW,60 and that describes both active and passive mechanisms by which the submunitions may separate from the container.

2.50  In addition to the definition of cluster munition in the Vienna Discussion Text there was a definition of the term ‘explosive sub-munitions’:

‘Explosive sub-munitions’ means munitions that in order to perform their task separate from a parent munition and are designed to function by detonating an explosive charge prior to, on or immediately after impact.

This definition drew on a NATO definition of a submunition as ‘any munition that, to perform its task, separates from a parent munition’,61 a definition also followed in the International Mine Action Standards (IMAS) glossary of mine action terms, definitions, and abbreviations.62 The second part of the definition drew upon elements of definition developed in the context of the CCW. The Declaration on Cluster Munitions endorsed by 25 States at the Third Review Conference of the CCW (17 November 2006) referred to cluster munitions as being ‘designed to eject sub-munitions containing explosives designed to detonate on, prior to, or immediately after impact …’ A similar formulation had been used by the UK in a statement to the CCW on 13 November 2006.63

2.51  An ‘Explanatory Annex’ to the Vienna Discussion Text presented a summary of different elements that had been proposed during the Lima (p. 168) Conference as offering possible grounds for systems being excluded from prohibition:

Article 2: Definitions

At the Lima Meeting there appeared to be broad agreement that landmines would be excluded from the definition of ‘cluster munition’ since they are already covered by other treaties. Some States also proposed that one or more of the following should be excluded: flare, smoke and chaff munitions and sub-munitions that are inert post impact. There were other proposals to exclude sub-munitions that aim, detect and engage point targets, and some States proposed to exclude cluster munitions which contain fewer than a specified number of explosive sub-munitions, sub-munitions with self destruct or self deactivation or other failsafe mechanisms, explosive sub-munitions with a tested failure rate of less than a specified percentage, and that the age of the sub-munition should be relevant. Some other States opposed some or all of these elements, with some proposing a comprehensive prohibition on all cluster munitions.

2.52  During the Vienna Conference, the Friend of the Chair on Definitions adopted a process of work to consider in turn the different characteristics proposed as a basis for exclusion from prohibition. In line with the order in which these characteristics were listed in the Explanatory Annex, this process of consideration would start with characteristics that were considered to be the least contentious; that is to say, it would start with those characteristics that might be expected to have the lowest risk of allowing submunition-based weapons with ‘unacceptable’ humanitarian effects to escape prohibition. Yet, despite the structure proposed by the Chair, the discussion of definitions in Vienna was again wide-ranging. Only a few States spoke explicitly in favour of excluding systems that dispensed flares, smoke, or chaff; landmines; or systems that deployed electrical devices.64

2.53  As had occurred at the Lima Conference, a number of States raised concerns that the approach being taken to defining a cluster munition was not in accordance with their understanding of the Oslo Declaration, while a few made reference to a perception that discussing the technological extent of the prohibition under the definition of a ‘cluster munition’ seemed to prejudge the process leading to a categorical prohibition.65 By contrast, other States, including a number affected by cluster munition remnants, argued broadly (p. 169) in favour of the categorical approach to prohibition and urged against further exclusions.66

2.54  Germany provided a detailed presentation of a ‘step-by-step’ proposal it had introduced in the context of the CCW in May 2007 as a draft Protocol VI to that treaty. This proposal took a three-stage approach: a prohibition on so-called inaccurate and unreliable cluster munitions; a phase-out period for so-called accurate and reliable cluster munitions; and an eventual replacement of cluster munitions with ‘alternative munitions’.67

2.55  The ICRC noted that the terms ‘inaccurate and unreliable’, which it had itself used to articulate the problems of cluster munitions, were being widely used by States and offered some points of clarification on these terms. It noted that while the language was neither legal nor technical, it considered that the terms applied to ‘all cluster munitions used to date’. The ICRC went on to suggest that of the criteria proposed by States, so-called sensor fuzing, increased reliability measures, and limitations on numbers, if adopted cumulatively, seemed to offer a better degree of civilian protection than most other proposals. However, the ICRC also noted that the burden of proof was on those calling for such exclusions, and that this should relate to how munitions functioned in reality not just under ideal test conditions.68

2.56  The Vienna Conference also saw the publication of an influential report and a detailed presentation of research on the performance of M85 submunitions in both test conditions and operational use.69 A DPICM-type submunition70 with a mechanical self-destruct mechanism, the M85 (or variants) was in the arsenals of a number of States taking part in the negotiations, including Austria, Norway, Switzerland and the UK. While Austria and Norway had both imposed a moratorium, several States had already resolved internally to exempt such weapons from prohibition. The grounds offered for such an exemption were either the presence of a self-destruct mechanism (or more generally a ‘fail-safe’ mechanism) or a stipulation that submunitions (p. 170) meeting certain reliability criteria should not be prohibited. The latter was generally expressed in terms of the proportion of submunitions failing to detonate as designed during test firings.

2.57  The research presented during the Vienna Conference showed not only that M85 submunitions had fallen below specified reliability benchmarks in test conditions, but that in operational use there were indications of significantly higher failure rates. The significance of this was two-fold: on the one hand there was evidence that munitions with self-destruct mechanisms could still present a post-conflict humanitarian threat, and on the other hand it called into question the possibility of using testing as a basis for meaningful regulation.

2.58  The report also rejected the concept of ‘non-dangerous duds’, by which certain States had argued that any submunitions left unexploded that had not ‘armed’ should not be considered within calculations of reliability because they did not present a significant civilian risk.71

2.59  In addition to challenging a number of specific technical arguments that were of central importance to discussion of what should, or should not, be prohibited, the M85 report set a high standard for evidence and analysis linked to actual testing and combat data.

2.60  The discussions at the Vienna Conference allowed the definition of cluster munitions to be further developed in the Draft Cluster Munition Convention prepared in advance of the Wellington Conference, while maintaining the same overall structure.72 This text excluded ‘mines’ from prohibition, not by amending the definition of a cluster munition but through an amendment to the scope of the Convention as expressed in Article 1, paragraph 2:

This Convention does not apply to ‘mines’ as defined by the Protocol on Prohibitions or Restrictions on the Use of Mines, Booby-Traps and Other Devices, as amended on 3 May 1996, annexed to the Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons which may be Deemed to be Excessively Injurious or to have Indiscriminate Effects.73

(p. 171) 2.61  Under the definition of cluster munition in Article 2, paragraph 2, there were now exclusions accepted for systems ‘designed to dispense flares, smoke, pyrotechnics or chaff’, and for systems ‘designed to produce electrical or electronic effects’:

‘Cluster munition’ means a munition that is designed to disperse or release explosive sub-munitions, and includes those explosive sub-munitions. It does not mean the following:

  1. (a)  a munition or sub-munition designed to dispense flares, smoke, pyrotechnics or chaff;

  2. (b)  a munition or sub-munition designed to produce electrical or electronic effects;

  3. (c) 

This left sub-paragraph (c) open for the possible insertion of further characteristics that might warrant exclusion from prohibition.

2.62  The definition of ‘explosive sub-munitions’ in the Draft Cluster Munition Convention remained unchanged from the Vienna text:

‘Explosive sub-munitions’ means munitions that in order to perform their task separate from a parent munition and are designed to function by detonating an explosive charge prior to, on or after impact.

2.63  Discussions at the Wellington Conference continued the process of work established in Vienna of considering certain characteristics as a basis for exclusion, proceeding from those that might be considered least contentious. Having excluded essentially ‘non-weapon’ systems after the discussion in Vienna, consideration now came to systems that employed submunitions designed to detect and engage specific targets.74

2.64  A number of States taking part in negotiations had sensor-fuzed munitions in their arsenals or were in the process of procuring such munitions. These weapons use sensors and electronic micro-processors to identify and then attack targets, primarily vehicles, based on their heat-signature and profile. Detonation of the munition is initiated by an electrical current from the depletion of a battery integral to the submunition. Such systems generally create an explosively formed projectile that is fired down onto the target from a significant stand-off distance, without creating the lateral fragmentation at ground level that has typically been used in submunitions in order to create an anti-personnel effect. Through this combination of targeting specific objects (p. 172) and limited use of lateral fragmentation, these systems differed from systems that scattered submunitions randomly. While the sensors of the submunitions search for targets within a ‘footprint’ area on the ground, the application of physical force within that footprint is much more narrowly defined than that created by the broad scattering of traditional explosive submunitions.75

2.65  The munitions of this type in service with negotiating States also contained electronic self-destruct and self-deactivation systems. Electronic self-destruct mechanisms were different to the mechanical self-destruct mechanisms employed on submunitions such as the M85 because they were not reliant on the functioning of moving parts after the submunition had armed. Furthermore, electronic self-deactivation mechanisms depleted the battery normally required to initiate detonation of the main explosive charge and were therefore held to render any unexploded submunitions less susceptible to detonation as a result of physical contact.

2.66  It was argued that such systems were more reliable than mechanical mechanisms because circuitry could be tested during production and munition storage providing an elevated level of quality control.76 In addition, these self-destruct mechanisms were supposed to function in the air before the submunition struck the ground, which might further limit unpredictability resulting from the effects of ground conditions on the mechanical systems designed to operate after impact. Whereas the self-destruct mechanism on impact-fuzed systems such as the M85 were ‘back-up’ mechanisms designed to operate if the submunition failed to function as designed, in sensor-fuzed systems the self-destruct mechanism operates where no valid target is identified by the sensor system. This is not necessarily a result of a failure in the system but may simply be because no valid targets are present in the area being (p. 173) searched by the sensors. These considerations are among a number of others that differentiate such systems from cluster munitions of the DPICM type and are likely to affect system reliability.77

2.67  In addition to these features of electronic fuzing, the sensor-fuzed systems in the arsenals of negotiating States at the time were all limited to two relatively large submunitions in each container munition (by contrast the US ‘Sensor Fuzed Weapon’78 contained a total of 40 individual explosive submunitions, often described as ‘skeets’). The limitation on numbers has a bearing on both the area-effect of the munition at the time of use and the likely level of UXO contamination that would result from each munition being fired. These munitions contained in combination—and in many cases in the most reliable available form—many features that States had proposed independently as possible grounds for exclusion from prohibition. Thus, if such features in combination were not sufficient grounds to avoid prohibition, it was difficult to see how they could be adopted individually as a basis for exclusion.

2.68  The discussion of sensor-fuzed munitions at the Wellington Conference began with a statement by one State in which it argued that these were not ‘area weapons’ and also that they had a number of cumulative safeguards, including self-destruct and self-neutralization features, which meant they were sufficiently different from cluster munitions as commonly understood not to warrant prohibition.79 During the course of the subsequent discussion a number of other States spoke positively about a possible exclusion for sensor-fuzed munitions.80 However, a significant body of States raised concerns about such an approach. Two States stated that the cost of these systems would mean the Convention would penalize those that could not afford them. Another State noted difficulties in defining ‘point’ and ‘area’ targets, and expressed scepticism about safeguards such as self-destruct and self- neutralization mechanisms in practice, while a fourth State questioned reliance on ‘unproven technical safeguards’.81 The CMC raised concerns about the wide area covered by some of these munitions and emphasized that the burden of proof was on the advocates of these weapons to demonstrate (p. 174) sufficient safeguards for humanitarian protection.82 The ICRC raised concerns that the devolution of certain targeting decisions to automated systems raised other legal concerns.83 Statements opposing any further exclusions from the definition of a cluster munition were made by some 20 States.84

2.69  While the plenary discussions in Wellington were effectively locked between those in favour of an exclusion (at the very least) for submunitions that detect and engage individual targets and those that were not, certain States continued to argue for exemptions for weapons on other grounds. The formal proposals submitted during the course of the Wellington Conference provide an indication of the range of exclusions still being sought.

2.70  The discussion of the definition in Wellington saw a number of formal proposals put forward by States. These were incorporated into the ‘Compendium’ attached to the Wellington Declaration, and became official documents during the Diplomatic Conference. While some of these papers proposed actual changes to the treaty text, others were more discursive. For example, a joint proposal by Australia, Canada, Denmark, Finland, Germany, Italy, Japan, the Netherlands, and the UK (which later became Conference document CCM/17 in Dublin) listed characteristics, which ‘either individually or in some combination’ were put forward as being potential descriptors of those cluster munitions that do not cause unacceptable harm to civilians. It went on to list ‘sensor fuzing’, ‘fail-safe systems’ (such as mechanical or electronic self-destruct, self-neutralization and self-deactivation mechanisms), ‘restrictions on the numbers of submunitions’, ‘delivery by direct fire’, ‘failure rates’, and ‘accuracy’ (in terms of delivery of the cluster munition to the target area). Further, this paper linked the scope of the definition to the possible need for transition periods (either with respect to Article 1 as a whole or to Article 1 as it would relate to certain subsets of prohibited cluster munitions.)

2.71  While that paper sought to maintain a broad menu of options for exclusion, individual States argued for different specific limitations on the scope of technology that would be prohibited. Japan submitted a proposal that weapons with fewer than 10 submunitions would fall outside the scope of the treaty and that reliable ‘or’ accurate cluster munitions should not be prohibited.85 Germany proposed a more comprehensive definition that would (p. 175) exclude from prohibition only weapons with fewer than 10 submunitions each designed ‘to engage point targets within a predefined area’ and also incorporating self-destruct and self-deactivation mechanisms. However, their proposal also foresaw a sub-paragraph (d), implying that this might not be the extent of the exclusions from prohibition.86

2.72  France proposed that the term ‘cluster munition’ should refer specifically to the ‘container’ or ‘carrier’ rather than to the weapon system inclusive of its submunitions. It suggested that weapons with fewer than 10 submunitions should be excluded from prohibition. It also proposed different options for an additional exclusion at a sub-paragraph (d) for a munition ‘designed to engage within a pre-defined area in a reliable and accurate manner’ or ‘that fulfils a combination of precise criteria regarding reliability and accuracy’. Beyond these exclusions, the French text proposed an additional definition of a ‘carrier-container’ which specifically listed artillery shells, air bombs, and guided or unguided missiles (potentially leaving open questions as to whether submunitions delivered by rockets or mortars would fall within the scope of the treaty.) The proposed definition of carrier-container also included dispensers that would remain attached to an aircraft during use unless these were for the delivery of ‘direct fire munitions’, this latter exclusion so as to avoid prohibiting ‘rocket pods’. The French proposal made reference to a review mechanism that would allow the parameters of the definition to be considered again by States Parties ‘no later than 5 years after entry into force’ in line with the treaty’s draft provisions regarding review conferences.87

2.73  Switzerland also proposed that ‘cluster munition’ should refer to the ‘container’ or ‘carrier’ rather than to the weapon system inclusive of its submunitions. It proposed excluding under Article 2, paragraph 2(c) only weapons that were ‘designed to engage a point target within a pre-defined area’ and which were also equipped with self-destruct, self-neutralization, ‘or’ self-deactivation mechanisms. The Swiss proposal followed the French definition of a ‘carrier container’ but proposed an alternative option for excluding rocket pods by requiring that dispensers be ‘designed to dispense submunitions in a single act’.88

2.74  The UK proposed that the term cluster munition refer to a ‘carrier container’ and argued that weapons with less than a certain—undefined—number (p. 176) of submunitions should fall outside the scope of the treaty. The UK proposal emphasized that this referred to ‘conventional’ munitions out of apparent concern that the process might inadvertently prohibit nuclear weapons. The UK sought to define cluster munitions in general as weapons ‘designed to dispense conventional explosive submunitions over a pre-defined area’, which seemed to be at odds with other proposals that used similar language to articulate specific exclusions from definition as a cluster munition. Indeed, the same UK document proposed an exclusion from prohibition for weapons ‘designed to deliver effects within a pre-defined area’. Further to this they proposed exclusions for ‘munitions that incorporate a failsafe system’, for ‘direct fire weapons’ or those ‘designed to deliver effects … on a point target’.89

2.75  The UK was one of a number of States that called for a blanket exclusion from prohibition for so-called ‘direct fire’ weapons (such as cluster munition variants of the Hydra and CRV-7 rockets used by attack helicopters). Such weapons were argued to be more accurate than other cluster munitions because the operator had a ‘line of sight’ to the target. However, questions were raised by the CMC and others regarding the relevance of such a characteristic to the area-effect of the cluster munition or of the likely risk of UXO being generated. Discussion of an exclusion for ‘direct fire’ munitions persisted into the negotiations in Dublin but gained little traction.

2.76  In a more discursive proposal, Sweden noted that one ‘essential feature’ in exclusions from prohibition should be ‘the existence of an electrical fail safe system which must embrace both self destruct (SD) and self-deactivation (SDA) mechanisms’. They noted that ‘the rationale for electrical systems is that batteries always discharge and render the munitions inoperable in the self-deactivating phase’. Sweden also suggested that these features should be cumulative with an ‘internal guidance system—including sensors—to aid accuracy’.90

2.77  Peru proposed a general exclusion in Article 2, paragraph 2(c), to exclude from prohibition munitions or submunitions that have ‘technical characteristics that allow to limit [sic] the area affected and reduce the risk of UXO contamination’.91 Although this proposal did not delineate the characteristics that would be required, the idea of linking specific technical characteristics to (p. 177) a statement of the effects being sought was picked up again during the negotiations at the Diplomatic Conference with the inclusion of an ‘effects-based’ chapeau statement in Article 2, paragraph 2(c).

2.78  Ireland92 proposed an amendment to the definition of ‘explosive submunition’ so as to link these directly to the main definition of a cluster munition as opposed to a ‘parent munition’. Such a proposal did away with the apparent need for a definition of a ‘carrier container’ as distinct from a ‘cluster munition’. This proposal was picked up, in a modified form, in the development of the definition of explosive submunition in the negotiations in Dublin. Ireland also proposed a definition of ‘explosive bomblet’ as ‘a munition which in order to perform its task is dispersed or separated from a dispenser, affixed to an aerial platform, and is designed to function by detonating an explosive charge prior to, on or after impact’.

2.79  Similar to Ireland’s proposal, the ICRC argued that the definition of ‘explosive submunition’ be amended to refer to those being ‘dispersed or released from a cluster munition’.93 This ensured that, in the terms ‘dispersed or released’, the relationship between a cluster munition and the explosive submunitions was expressed in the same terms in both parts of the definitions.

Negotiations

2.80  The document that carried the Draft Cluster Munitions Convention from the Wellington Conference to the Diplomatic Conference in Dublin was the Wellington Declaration, which States were required to endorse in order to be full participants in the negotiations. The Wellington Declaration used the language of the Oslo Declaration regarding a prohibition on ‘cluster munitions that cause unacceptable harm to civilians’:

States met in Wellington from February 18 to 22, 2008, to pursue an enduring solution to the grave humanitarian consequences caused by the use of cluster munitions. They are convinced that this solution must include the conclusion in 2008 of a legally (p. 178) binding international instrument prohibiting cluster munitions that cause unacceptable harm to civilians. In that spirit they affirm that the essential elements of such an instrument should include:

  • •  A prohibition on the use, production, transfer and stockpiling of cluster munitions that cause unacceptable harm to civilians …94

2.81  The text of the Draft Cluster Munitions Convention was submitted unchanged by the Wellington Conference to the Diplomatic Conference to serve as the basis for negotiations. Although there were to be changes made in almost all sections of the definition of cluster munition, discussion focused on paragraph 2(c) of Article 2. This discussion was broadly polarized between those States arguing for language to be inserted that would allow certain weapons to be excluded from prohibition and others arguing for the deletion of this sub-paragraph.

2.82  A significant number of those States seeking to exclude or exempt95 particular weapon types from prohibition continued to refer to the phrase ‘unacceptable harm’ as used in the Oslo Declaration and the Wellington Declaration as a basis for expecting that certain weapons with submunitions would not be prohibited.96 Some of these States sought initially to promote the discursive listing of a wide range of technologies as a possible basis for exclusions or exemptions contained in Conference document CCM/17.97 However as the negotiations progressed, the emphasis shifted towards those endorsing a set of cumulative technical provisions. By contrast, those arguing for the deletion of Article 2, paragraph 2(c) often expressed scepticism that technological features could be relied upon to avoid the humanitarian problems associated with cluster munitions.98

2.83  With divergent views expressed in the opening session of the Committee of the Whole, the President of the Conference, Ambassador O’Ceallaigh appointed Ambassador Don MacKay of New Zealand to act as Friend of the President, to convene informal discussions on Article 2 and if (p. 179) a text was not informally agreed to ‘submit the proposal that he considered best’.99 The informal consultations were structured first to consider whether certain technical characteristics might individually be considered sufficient to warrant weapons falling under the main definition being excluded from prohibition. The Friend of the President circulated an informal paper containing ‘Elements for discussion in relation to (c)’, which listed the following:

  • •  A munition or sub-munition designed to locate and engage a point target within a pre-defined area (i.e. sensor fusing, multiple or single)

  • •  A munition or sub-munition that is delivered by direct fire (i.e. line-of-sight fire)

  • •  A munition or sub-munition that otherwise meets accuracy requirements, in terms of delivery of the munition and sub-munition to the target area

  • •  A munition that contains less than … sub-munitions (i.e. limited numbers of sub-munitions per cluster munition)

  • •  A sub-munition that contains electronic fail-safe mechanisms (self destruct and/or self-deactivation)

  • •  A sub-munition that contains mechanical fail-safe mechanisms (self destruct and/or self-neutralization)

  • •  A sub-munition that meets prescribed reliability/failure rates

  • •  A sub-munition that otherwise meets reliability requirements, in terms of not leaving unacceptable quantities of hazardous unexploded ordnance (UXO)

  • •  Any other elements that States may wish to propose.

2.84  The paper also noted the need to ‘future-proof ’ the definition, which was taken to mean that any exclusion at paragraph 2(c) should not allow the same problematic effects experienced in the past to be repeated with other similar weapons in the future. This paper noted that these elements might be discussed in combination (i.e. cumulatively) as well as individually. This list did not explicitly make reference to the weight of submunitions or to air-defence systems, both of which were addressed during subsequent negotiations and included in the definition ultimately adopted.

2.85  With many States expressing scepticism of technical characteristics individually, discussion proceeded to consider whether characteristics taken cumulatively might provide grounds for confidence that ‘unacceptable harm’ (p. 180) would not result from the use of these weapons.100 In order to structure discussion of these cumulative options the Friend of the President circulated an informal paper of ‘Proposals that have been made on cumulative lists of criteria’. This contained eight options, labelled A to H, of cumulative technical characteristics for a possible exclusion at paragraph 2(c). Each option contained between two and five bullet points drawn in the main from the specific elements that had been listed individually in the first informal paper. Throughout the informal discussions, the technical characteristics, both individually and in combinations, were considered in relation to their relevance to the negative humanitarian effects that were associated with cluster munitions.101

2.86  Formal proposals made by States during the Wellington Conference and compiled into the Compendium were also circulated individually as formal documents of the Diplomatic Conference. Furthermore, a number of formal proposals on the definition of cluster munition and explosive submunition were circulated at the very start of the Diplomatic Conference. Slovakia argued that sub-paragraph (c) should exclude submunitions with a failure rate of ‘not more than 1%’ and which also had some sort of self-destruct, self-deactivation, or self-neutralization feature.102

2.87  Spain followed a number of the proposals submitted in Wellington in arguing that cluster munition should refer solely to the ‘carrier container’.103 The Spanish proposal advocated an exclusion from prohibition in sub-paragraph (c) for weapons with submunitions equipped with ‘a self-safe mechanism that, combined with the normal function mechanism, guarantees that the number of remaining dangerous duds that can cause harm to non-combatants is in practice equal to zero’. By using the term ‘dangerous duds’, however, this proposal returned to criteria that had been widely criticized during previous meetings. Other terms such as ‘guarantee … in practice’ raised significant concerns about how such a definition would be implemented. In addition to these technical characteristics, the Spanish proposal argued that any submunitions (p. 181) excluded in sub-paragraph (c) should be ‘painted and marked in order to distinguish it from the terrain and to warn about their dangerousness’.

2.88  The Czech Republic proposed excluding ‘landmines’ from the Convention in Article 2, paragraph 2(c) (rather than through Article 1) and also proposed an additional sub-paragraph (d) that would exclude from prohibition weapons with fewer than ten submunitions each equipped with self-destruct ‘and/or’ self-deactivation mechanisms.104 In addition, the Czech Republic proposed stipulating that the term ‘explosive submunitions’ refer only to submunitions that would explode ‘immediately’ after impact, a stipulation that seemed also to be concerned to exclude remotely delivered landmines from prohibition.

2.89  On 21 May, Norway circulated a proposal that suggested excluding at sub-paragraph (c) munitions with submunitions that met a number of cumulative criteria. In these criteria, and also in the main definition, the Norwegian proposal used thresholds based on weight to further define a cluster munition:

‘Cluster munition’ means a munition that is designed to disperse or release explosive sub-munitions each weighing less than 20 kilograms, and includes those explosive sub-munitions. It does not mean the following:

  1. a)  a munition or sub-munition designed to dispense flares, smoke, pyrotechnic or chaff effects.

  2. b)  a munition or sub-munition designed to produce electrical or electronic effects;

  3. c)  a munition with sub-munitions each weighing more than five kilograms, designed to seek, detect and engage point targets, and equipped with electronic self destruct and self deactivation mechanisms.

The introduction of the upper submunition weight threshold of 20 kilograms in the main definition went on to be adopted in the final definition of a cluster munition. The cumulative list of criteria presented at sub-paragraph (c) is similar to that contained in the final definition, though there were to be further developments in drafting, the addition of a limitation on submunition numbers and, very importantly, the addition of a chapeau at sub-paragraph (c) that links the technical characteristics to the effects of humanitarian concern. The lower weight limit of a minimum of five kilograms for permitted explosive submunitions was also to be lowered to four kilograms in the final stages of negotiation.

(p. 182) 2.90  But while the Norwegian proposal began to establish some support for a set of cumulative criteria at sub-paragraph (c), the only alternative position commanding support from a range of countries was for the removal of the sub-paragraph altogether. In a formal proposal tabled jointly, Argentina, Chile, Costa Rica, the Dominican Republic, Ecuador, Guatemala, Guinea, Honduras, Mexico, Nicaragua, Panama, Peru, Uruguay, and Zambia all called for the deletion of Article 2, paragraph 2(c).105 In a subsequent session of the Committee of the Whole, on 26 May, Austria, Benin, Botswana, Burkina Faso, Burundi, the Democratic Republic of Congo, Ghana, Guinea, Laos, Nigeria, Panama, Senegal, Sierra Leone, and Venezuela, all expressed support for the deletion of the sub-paragraph.106

2.91  Against this backdrop, on 22 May, a Friend of the President’s ‘Paper on Definition of a “Cluster Munition”’ was circulated. This effectively summarized the position as a choice between the deletion of 2(c) and a set of cumulative criteria at 2(c) which were linked to a statement of the negative effects that were the basis for considering cluster munitions unacceptable. This summary paper did not, however, adopt the weight-based criteria proposed by Norway.

‘Cluster munition’ means a munition that is designed to disperse or release explosive sub-munitions, and includes those explosive sub-munitions. It does not mean the following:

  1. a)  a munition or sub-munition designed to dispense flares, smoke, pyrotechnics or chaff or air defence systems;

  2. b)  a munition or sub-munition designed to produce electrical or electronic effects;

  3. c)  a munition that has all of the following characteristics to minimise its area effect and the risk of unexploded ordnance contamination from its use:

    1. a)  each munition contains fewer than 10 sub-munitions;

    2. b)  each sub-munition is designed to locate and engage a point-target within a pre-defined area;

    3. c)  each sub-munition is equipped with an electronic self-destruction mechanism;

    4. d)  each submunition is equipped with an electronic self-deactivating feature.

The Friend of the President’s Paper introduced for the first time a chapeau at 2(c). This provided a mechanism for linking the technical (p. 183) characteristics that might be held to justify an exclusion from prohibition with the problematic effects that needed to be avoided. This chapeau would be further developed during the course of the negotiations to establish a new precedent for defining weapons in relation to their problematic characteristics.

2.92  Also on 22 May, Canada circulated a formal proposal supporting the inclusion of effects-based language in the chapeau of paragraph 2(c) but which sought to modify the specific text of that chapeau. The Canadian proposal also called for the removal of the proposed limitation based on the number of submunitions.107 Canada had expressed on several occasions a preference to avoid numbers that must necessarily be arbitrary.

2.93  On 23 May, the UK circulated a proposal that continued to argue for wide exemptions for systems such as the M85.108 Elements of this proposal were now substantially at odds with the broad direction of the definition debate. The paper proposed blanket exemptions from prohibition for ‘direct fire’ munitions, for any munition with fewer than 10 submunitions, and for any submunitions equipped with a self-destruct mechanism. There were no statements in support of the UK proposal.

2.94  On 23 May, a revised ‘Chair’s Discussion Paper’ was circulated which further developed the paper of the previous day based on the subsequent discussions.109 It incorporated the weight-based criteria proposed by Norway, modified the language of the chapeau at paragraph 2(c) and amended the terminology for self-destruction and self-deactivation features:

‘Cluster munition’ means a munition that is designed to disperse or release explosive sub-munitions each weighing less than 20 kilograms, and includes those explosive sub-munitions. It does not mean the following:

  1. a)  a munition or sub-munition designed to dispense flares, smoke, pyrotechnics or chaff; or air defence systems;

  2. b)  a munition or sub-munition designed to produce electrical or electronic effects;

  3. c)  a munition that has the following characteristics to limit its area effect and reduce the risk of unexploded ordnance contamination from its use:

    1. a.)  each munition contains fewer than 10 sub-munitions;

    2. b.)  each sub-munition weighs more than five kilograms;

    3. (p. 184) c.)  each sub-munition is designed to locate and engage a point-target within a pre-defined area;

    4. d.)  each sub-munition is equipped with an electronic self-destruct mechanism;

    5. e.)  each submunition is equipped with an electronic self-deactivating mechanism.

In presenting this paper to the Committee of the Whole, Ambassador Don MacKay ‘emphasised that the discussion paper was not an agreed text, but represented the Friend of the President’s own assessment of possible language for Article 2(c), if it is to exist. Fundamental differences remained on an Article 2(c) and whether or not it should be included. A formal proposal had been made for its deletion.’110

2.95  At the beginning of the second week of negotiations, States remained broadly polarized between those seeking certain exclusions at Article 2, paragraph 2(c) and those favouring the deletion of the sub-paragraph. However, the discussion paper circulated by the Chair was receiving some support from both sides of this divide. Among those favouring additional provisions at paragraph 2(c), Australia, Canada, France, Germany, the Netherlands, Norway, Sweden, and Switzerland were all broadly supportive of the direction of the Chair’s paper. Among those calling for deletion of the sub-paragraph, many indicated some degree of support for the Chair’s text or an openness to continue discussions along these lines.111 A significant number of States spoke critically of the inclusion of an exemption for ‘air defence systems’ proposed by the UK to be included in paragraph 2(a),112 including Argentina, Botswana, Costa Rica, Norway, and Panama.113

2.96  A number of States also sought further evidence or proof to justify claims that weapons excluded at paragraph 2(c) would not cause ‘unacceptable harm’ to civilians.114 Austria and Lebanon, supported by Malta, argued that any weapons excluded under the sub-paragraph should be subject to scrutiny and their status in relation to the Convention reviewed in the future. Canada also implicitly supported such a position in arguing that, ‘it is incumbent on States to monitor the use and results of weapon development’ (p. 185) and reminded States ‘of the possibility of amending the Convention by a two-thirds majority. States will have the opportunity in the annual meetings of States Parties and in the Review Conference to adjust to future developments.’115

2.97  On 26 May, as a supplement to their earlier proposal, Norway circulated an ‘Explanatory note’ providing further analysis of the thinking behind using weight as one of a number of cumulative criteria.116 It summarized these in the following terms:

The proposed weight limit will in practice eliminate up to 99% of all cluster munitions currently in stocks, and all existing cluster munitions that have ever been used in war.

A weight based criteria is robust. It is easy to understand, easy to verify and in practice almost impossible to circumvent.

The proposed weight limit will allow for weapons that do not cause the problem we aim to address, and will prohibit those that inherently cause such problems.

A weight limit for point target sub-munitions would automatically reduce the total numbers of sub-munitions that could physically fit into any parent munition. This will substantially reduce the problem of unexploded ordnance.

A weight limit for such point target sub-munition would safeguard against a future miniaturization (and resulting increase in numbers). This will prevent a repetition of the very problems we seek to address with this treaty.

Larger sub-munitions are easier to detect, less sensitive to handle and thus easier to clear. This will reduce the risk for civilian accidents due to inadvertent contact with unexploded ordnance.

2.98  As a summary of the implications of the weight criteria, the ‘Explanatory note’ stated:

  1. 1.  More than 20 kilograms per submunitions—excluded from the scope of the convention

  2. 2.  Between 5 and 20 kilograms per sub-munition—excluded under the condition that they meet strict requirements of single target, self destruct and self- deactivation

  3. 3.  Less than 5 kilograms per submunition—categorically prohibited.

In addition to these explanations based on perceived humanitarian considerations, the upper weight limit proposed by Norway had the effect of excluding France’s KRISS anti-runway submunitions from the scope of the Convention. (p. 186) During discussions of the definition on 26 May in the Committee of the Whole, France expressed ‘full support’ for the weight criterion proposed by Norway.117

2.99  A proposal by Spain on 27 May was the last formal proposal submitted on the main (p. 187) definition of cluster munition during the negotiations.118 It adopted the effects-based chapeau introducing sub-paragraph (c) but proposed vaguely defined exclusion for munitions with ‘sub-munitions which only address the area encompassed by the intended military objective’119 in addition to requiring the presence of electronic self-destruction and self-deactivation mechanisms.

2.100  In addition to the discussion of the main definition of ‘cluster munition’ were parallel, ongoing discussions of terms integral to this definition. These included the definition of ‘explosive submunition’ and such terms as ‘point target’, ‘self-destruct devices’, ‘self-deactivation’, and ‘self-neutralisation’.

2.101  On 23 May a revised definition of an explosive submunition was circulated by the Friend of the President on ‘other’ definitions.120 Very similar to that in the Draft Convention, this definition was for ‘explosive submunition’ in the singular rather than in the plural. It defined an explosive submunition as separating ‘from’ a cluster munition as opposed to from a ‘parent munition’.121 Three days later, a further revised definition was circulated by the Friend of the President on ‘other’ definitions.122 This version inserted explicit reference to a ‘conventional’ munition and changed the term ‘separates from’ to ‘dispersed or released by’ to describe the relationship between explosive submunitions and cluster munitions as had been proposed in Wellington by the ICRC. This latter change created coherence of language between the definition of explosive submunition and that of cluster munition. The term dispersed and released ‘by’ was preferred to ‘from’ because the latter emphasized more strongly a distinction between ‘cluster munition’ and ‘explosive submunitions’ whereas the definition of cluster munition explicitly included the explosive submunitions. This definition of explosive submunition was adopted in the final Convention text:

‘Explosive submunition’ means a conventional munition that in order to perform its task is dispersed or released by a cluster munition and is designed to function by detonating an explosive charge prior to, on or after impact.

By defining the term ‘explosive submunition’ in relation to a ‘cluster munition’, space was also created for the existence of ‘submunitions’ that were used in other systems not defined as cluster munitions (for example, those systems that might be excluded from prohibition either due to the weight criteria in the main definition, or through the characteristics described at 2(a)–(c)).

2.102  With proposed definitions of a cluster munition making reference to weapons that would ‘engage a point target’, on 26 May an informal paper circulated by the Friend of the President on ‘other definitions’123 noted the NATO definition of a ‘point target’ as a ‘target that requires the accurate placement of bomb or fire’. Later in the day a revised paper summarized the state of the debate over the definition of this term:124

Two approaches were discussed. One was to define ‘point target’. The other was to use alternative language in Article 2c. that would avoid the need to define ‘point target’.

  1. 1.  ‘Point target’ means a target that requires the accurate placement of bomb or fire on a single object.125

  2. 2.  2.c.b. Each submunition is designed to detect and engage a single target object.

The second of these options was adopted in the final text of the Convention.

2.103  On 28 May 2008, the ‘Presidency Paper, draft Convention on Cluster Munitions’ (Conference document CCM/PT/15) was circulated. The definitions in this paper were to be adopted without further change. The definition of explosive submunition was unchanged from that circulated by the Friend of the President on 26 May. However, the definition of cluster munition had been subject to a number of adjustments from the ‘Chair’s Discussion Paper’ of 23 May.

(p. 188) 2.104  It was now clarified that ‘cluster munition’ meant a ‘conventional munition’.126 While the lack of a definition for ‘air defence’ remained a weakness, the wording of the exclusion for air-defence systems had been improved through a stipulation that such systems should be ‘designed exclusively’ for such a role. The minimum weight limit at sub-paragraph (c) (ii) had been reduced from five kilograms to four kilograms, a change that went in the direction of widening the scope of the exclusion although no open discussions explained the implications or reasoning behind this. The phrase ‘point-target within a pre-defined area’ was replaced with ‘a single target object’, reflecting the discussions that had taken place on this. The term ‘self-deactivating mechanism’ was replaced with the term ‘self-deactivating feature’.

2.105  The chapeau at sub-paragraph (c) had been significantly reworded. Whereas the discussion text of 23 May stipulated that the technical characteristics were ‘to limit’ area effect and ‘reduce’ the risk of UXO, the new chapeau brought in more absolute language and, by introducing the term ‘indiscriminate area effects’, linked the chapeau to terminology in other areas of international humanitarian law. Following a suggestion by Germany, the original intent of the chapeau was clarified that a munition must have ‘all’ of the technical characteristics at (c)(i) to (v) in order not to be considered a cluster munition.127

2.106  A final change of note was that all of the serial lines under subparagraph (c) now referred to ‘explosive submunitions’ where previously they had referred simply to submunitions. This is problematic because it creates a logical flaw in the definition. Serials (i) to (v) refer to the characteristics of weapons that are not cluster munitions due to the characteristics of their ‘explosive submunitions’, yet ‘explosive submunitions’ are subsequently defined as components of a ‘cluster munition’. It is thus not logically coherent that these weapons are not cluster munitions but also have explosive submunitions. This problem did not exist in the Chair’s text of 23 May because the serial items under sub-paragraph (c) referred simply to ‘submunitions’. The Summary Record of the Committee of the Whole for 26 May 2008 attributes the proposal to use the term ‘explosive submunitions’ in the sub-paragraph to Germany.128 Notwithstanding this technical inconsistency, it is clear that it does not alter the meaning of the definition, and that cluster munitions including their explosive submunitions are prohibited unless they fulfil the five technical criteria in Article 2, paragraph 2(c)(i)–(v). The issue was identified and (p. 189) raised by the CMC prior to adoption of the Convention, but it was too late to reopen the text for anything but editorial changes.

2.107  In presenting the text, the President of the Conference, Ambassador O’Ceallaigh, noted that ‘a restrictive effects-based definition had been used which would prohibit the vast majority of submunitionbased weapons systems existing in the world today, and all of those which have been used’.129 On recommending the text for adoption the President described ‘an extremely ambitious Convention text’ and noted that that ‘the headline definition of a “cluster munition” will lead to the prohibition of all cluster munitions that cause unacceptable harm to civilians and that it prohibits all cluster munitions ever used in armed conflict. For many states represented this will involve the removal of all cluster munitions from national stocks.’130

2.108  The far-reaching scope of the definition was widely recognized by States as a very good result, although a number of States would have preferred a definition with no exclusions whatsoever.131 Civil society also recognized the strength of the definition. ICRC welcomed the definitions as ‘comprehensive’.132 The CMC stated that ‘the draft Convention outcome far exceeded the expectations of nearly everyone. The prohibition contained therein was more comprehensive than that of the Mine Ban Treaty: not just some cluster munitions but all were prohibited; no distinction was made between good and bad cluster munitions.’ It noted further that: ‘The exclusion in Article 2(c) applied to munitions that do not have the same effects as cluster munitions, that is, that do not have wide area and excessive unexploded ordnance effects.’133

Paragraph 2

‘Cluster munition’ means a conventional munition that is designed to disperse or release explosive submunitions each weighing less than (p. 190) 20 kilograms, and includes those explosive submunitions. It does not mean the following:

  1. (a)  A munition or submunition designed to dispense flares, smoke, pyrotechnics or chaff; or a munition designed exclusively for an air defence role;

  2. (b)  A munition or submunition designed to produce electrical or electronic effects;

  3. (c)  A munition that, in order to avoid indiscriminate area effects and the risks posed by unexploded submunitions, has all of the following characteristics:

    1. (i)  Each munition contains fewer than ten explosive submunitions;

    2. (ii)  Each explosive submunition weighs more than four kilograms;

    3. (iii)  Each explosive submunition is designed to detect and engage a single target object;

    4. (iv)  Each explosive submunition is equipped with an electronic self-destruction mechanism;

    5. (v)  Each explosive submunition is equipped with an electronic self-deactivating feature;

Commentary

Conventional munition

2.109  Cluster munitions are defined as ‘conventional’ munitions, thus clarifying that the Convention is not concerned with nuclear, chemical, or biological weapons that may use submunitions. Without such clarification, certain nuclear weapons that have multiple submunition warheads might have fallen under the definition of cluster munitions.

2.110  Defining a cluster munition as a ‘munition’ distinguishes it from the proposals made during the negotiations to define a cluster munition as a ‘carrier container’. This formulation, introduced in the Vienna Discussion Text, makes it clear that the term cluster munition refers to the whole weapon system, not simply a component or delivery mechanism. However, by using the term ‘munition’ it opened up the requirement for a separate definition in order to capture those weapons that used a dispenser that remained affixed to an aircraft. For more discussion, see the commentary below on the phrase ‘and includes those submunitions’134 as well as those discussing the definitions of ‘explosive bomblet’135 and ‘dispenser’.136

(p. 191) Disperse or release

2.111  The phrase ‘disperse or release’ acknowledges that the mechanism by which explosive submunitions (or explosive bomblets) are separated from the container or dispenser can be either active or passive. That is to say, that the submunitions may be actively propelled from the container or may simply fall from the container due to the removal of some barrier to their movement.

Explosive submunition

2.112  Explosive submunitions are defined at Article 2, paragraph 3 as follows:

3.  ‘Explosive submunition’ means a conventional munition that in order to perform its task is dispersed or released by a cluster munition and is designed to function by detonating an explosive charge prior to, on or after impact;

This definition is discussed further below.

Weighing less than 20 kilograms

2.113  The upper weight limit of 20 kilograms in the first part of the definition provides a broad basis for weapons with submunitions not to be considered cluster munitions. The Convention does not therefore apply to any weapons with submunitions weighing 20 kilograms or more.

2.114  The CMC noted during the Diplomatic Conference that a high upper weight limit might be necessary to distinguish submunitions from other aircraft bombs deployed from bomb racks.137 While most weapons commonly considered cluster munitions have submunitions weighing significantly less than 20 kilograms it was significant in the context of the negotiations that this component of the definition served to exclude from prohibition the French Apache,138 an anti-runway system that employs ten KRISS submunitions, each weighing approximately 50 kilograms.139

(p. 192) And includes those explosive submunitions

2.115  This phrase was introduced in the Vienna Discussion Text and serves to define cluster munitions as both the container and its submunitions. This was in contrast to the Lima Discussion Text and a number of proposals that sought to define a cluster munition as a ‘container’, and thus by implication a component of a weapon system rather than the weapon system itself.

A munition or submunition designed to produce flares, smoke, pyrotechnics, or chaff

2.116  Systems that use explosive submunitions but which are designed to produce flares, smoke, pyrotechnics, or chaff are excluded from prohibition. Systems designed for these purposes are not primarily weapons designed to kill or wound people or to damage military objects but rather are intended to create illuminating, communicative, obscuring, or diversionary effects. They may be used to ‘mark’ targets on the ground or may serve as countermeasure systems to protect vehicles from attack. Submunitions employed in such systems may use an explosive charge in order to create these effects and so, without this clarification, these systems might be held to fall under the definition of a cluster munition and so be prohibited. While such systems may not primarily be intended for use against personnel, the border line between these functions can be blurred. For example, white phosphorous can be used both to create smoke and as an incendiary weapon.

A munition designed exclusively for an air-defence role

2.117  Unlike systems for producing ‘flares, smoke, pyrotechnics or chaff’, munitions ‘designed exclusively for an air-defence role’ are primarily weapons for use against aircraft. It is a potential weakness of the definition in the Convention that this phrase is not subject to further specific definition.140 During the negotiations in Dublin, the UK raised the concern that certain airdefence weapons might inadvertently be prohibited, despite being quite different from cluster munitions as generally understood.141 The specific weapons of concern to the UK were understood to be its ‘Starstreak’ missiles, generally (p. 193) considered an air-to-air or ground-to-air weapon that uses three explosive submunitions.142

2.118  The UK noted on a number of occasions, and it was clarified by the chair of the informal discussions, that the term ‘air defence’ referred to ‘ground-to-air’ weapons that resulted in ‘effects in the air’ rather than on the ground.143 The term ‘air-defence role’ should not therefore be interpreted as allowing so-called ‘anti-runway’ munitions that, it might be argued, are used to defend pre-emptively against aerial attack. The interpretation presented here is further supported by the fact that a Chair’s discussion text produced in the context of the CCW presented munitions ‘designed for an air-defence role’ and munitions ‘designed primarily or exclusively as anti-runway munitions’ as two distinct items on the same list.144 It is notable that the exclusion from prohibition only applies to systems ‘designed exclusively’ for this role, and therefore any systems that were developed to provide both ground-to-air and ground-to-ground capability, and falling under the broader definition of a cluster munition, would be prohibited.

Submunitions designed to produce electrical or electronic effects

2.119  Systems with submunitions that produce electrical or electronic effects may be used as radio-jammers to disrupt communications equipment or in other ‘counter-measure’ functions. Such systems are not primarily weapons designed to kill or wound people or to damage military objects.

(p. 194) To avoid indiscriminate area effects and to avoid the risk posed by unexploded submunitions

2.120  The chapeau clause introducing sub-paragraph (c) of Article 2, paragraph 2 articulates the problematic humanitarian effects that are considered defining features of cluster munitions and which must be avoided—for any weapons with explosive submunitions weighing less than 20 kilograms and that are not designed exclusively for air defence—to be considered permissible. The chapeau provides a statement of the problematic effects that the subsequent technical characteristics are intended, cumulatively, to avoid. It therefore provides both a justification for the exclusion of weapons that meet these technical criteria and also a potential mechanism for determining if these technical criteria function as intended.145 By linking the definition of what is prohibited to the humanitarian effects that are the basis for prohibition, this chapeau is an important legal innovation. While cluster munitions are not defined as being ‘designed’ to create these problematic effects, paragraph 2(c) stipulates that munitions must be designed positively to avoid these effects if they are not to be prohibited.

2.121  The phrase ‘to avoid’ is made problematic by being applied equally to both ‘indiscriminate area effects’ and ‘risks posed by unexploded submunitions’. As explained in further detail below, some risk from UXO cannot be wholly avoided in explosive munitions. Absent any qualification on the level of ‘risks posed by unexploded submunitions’ that must be avoided, it is difficult to read ‘avoid’ in that context as being absolute; rather it needs to be understood as more akin to ‘greatly limit’. By contrast, ‘indiscriminate’ attacks are subject to absolute prohibition under 1977 Additional Protocol I146 and hence it is reasonable here to assume that the term ‘to avoid’ should be read in an absolute sense (i.e. indiscriminate area effects are not allowed).

Indiscriminate area effects

2.122  ‘Indiscriminate area effects’ can be understood to represent a problematic characteristic of cluster munitions ‘at the time of their use’147 and are distinct from the pattern of UXO contamination that may be found after (p. 195) use.148 ‘Indiscriminate area effect’ refers to the so-called ‘footprint’ within which cluster munitions scatter individual explosive munitions randomly. The term ‘indiscriminate area effects’ is not defined in the Convention, but can be further understood both by reference to the specific technical characteristics held as necessary to avoid such effects and by consideration of other legal instruments that make use of similar terms. Of particular relevance would be Article 51, paragraphs 4 and 5, of 1977 Additional Protocol I which prohibits indiscriminate attacks and describe such attacks. The description of indiscriminate attacks as those that ‘are of a nature to strike military objectives and civilians or civilian objects without distinction’149 and the specific identification as indiscriminate of attacks ‘by bombardment by any methods or means which treats as a single military objective a number of clearly separated and distinct military objectives located in a city, town, village or other area containing a similar concentration of civilians or civilian objects’150 have particular bearing on the concept of ‘indiscriminate area effects’. The detailed definition of what technologies are prohibited under the Convention on Cluster Munitions should serve to influence interpretation of these general rules as they relate to the distribution of explosive force and fragmentation across areas.

2.123  Of the cumulative characteristics that are necessary to avoid the humanitarian problems indicated under the paragraph 2(c) chapeau, points (i) to (iii) are related, inter alia, to the aim to avoid ‘indiscriminate area effects’. These technical characteristics serve to calibrate how such effects are to be understood. Paragraph 2(c) (iii) requires the individual submunitions to detect and engage a ‘single target object’. Such a ‘target object’ should be a vehicle, artillery piece, or other such distinct item. It is prohibited, therefore, to use weapons where the explosive submunitions are scattered and distribute explosive force and fragmentation randomly across an area. This represents a substantial functional distinction between weapons excluded from prohibition at paragraph 2(c) and cluster munitions. This obligation not to distribute explosive force and fragmentation randomly across an area is further strengthened by sub-paragraph (c), points (i) and (ii).151

(p. 196) 2.124  None of the characteristics intended to avoid indiscriminate area effects relates to the accuracy of the container or dispenser. So ‘indiscriminate area effects’ are not considered to result primarily from the risk of ‘area effects’ from submunitions being deployed in the wrong place. Rather ‘indiscriminate area effects’ are wholly considered to result from the use of multiple explosive items to distribute explosive force and fragmentation randomly across a pre-defined area, which was a key purpose and salient design feature of cluster munitions. Article 2, paragraph 2(c)(i) and (ii) may further suggest that ‘indiscriminate area effects’ can result from submunitions that detect and engage single object targets if these weapons are used in large quantities.

2.125  Prior to the Oslo Process, concern regarding the ‘area effect’ of cluster munitions had led some States to call for a prohibition on certain types of cluster munitions, combined with a ‘prohibition on the use of cluster munitions in areas of civilian concentration’.152 This call was also echoed during the Oslo Process.153 The Convention on Cluster Munitions adopts a prohibition on a defined category of technology but has no rules or regulations governing technologies that are not subject to that prohibition. However, any munitions excluded under paragraph 2(c) must still be used in accordance with the rules of international humanitarian law governing attacks,154 and the mere fact that these weapons are not prohibited under the terms of paragraph 2(c) does not mean that they are necessarily appropriate for use in areas of civilian concentration. With respect to weapons designed to detect and engage individual targets, the CMC raised the concern during the Oslo Process that such weapons cannot themselves distinguish between military objectives on the one hand, and civilians and civilian objects on the other.155

(p. 197) To avoid the risk posed by unexploded submunitions

2.126  In addition to avoiding ‘indiscriminate area effects’, weapons that fall outside the definition of a cluster munition at paragraph 2 (c) must also avoid creating ‘risks posed by unexploded submunitions’. It is widely recognized that the use of any type of explosive ordnance will create some level of UXO risk. Efforts to minimize these risks focus on reducing the likelihood of individual items being left unexploded and on reducing the susceptibility of any such unexploded items to subsequent accidental detonation. Through cumulative technical criteria, paragraph 2(c) establishes the level of precaution that must be taken in order to reduce the risk of unexploded submunitions to an acceptable level.

2.127  Of the cumulative characteristics necessary to avoid the humanitarian problems indicated in the sub-paragraph (c) chapeau, points (i), (ii), (iv), and (v) relate inter alia to avoiding ‘risks posed by unexploded submunitions’. The limitation on numbers serves to reduce the likelihood of items being left unexploded while the minimum weight requirement was argued by Norway as serving, among other things, to prevent miniaturization that would facilitate use in greater numbers in the future. The requirements of (iv) and (v) are both wholly concerned with reducing UXO risk. It should also be noted that current submunition-based weapons designed to detect and engage individual target objects are designed to detonate in the air, a feature that substantially removes variations in ground conditions from factors likely to influence reliability.

i.  Each munition contains fewer than ten explosive submunitions

2.128  Munitions with 10 or more explosive submunitions cannot be excluded from prohibition under paragraph 2(c). Limitations on the number of submunitions that could be delivered in a single container were discussed extensively as offering potential to reduce the humanitarian risk presented by cluster munitions. Certain States put forward proposals that the number of submunitions, on its own and usually set at ‘10 or more’, should determine whether or not a munition is considered a cluster munition.156 Such proposals were significantly more permissive that the final definition, which requires munitions with fewer than ten submunitions to meet a number of other criteria to enhance civilian protection.

(p. 198) 2.129  The number of submunitions does have a direct bearing on the likely level of UXO threat that will be created by a single cluster munition; the greater the number of submunitions, the greater the likelihood of one or more of the submunitions failing to explode as intended. This same relationship makes it logically problematic to regulate cluster munitions according to a submunition percentage failure rate without linking this percentage to the number of submunitions. Limiting the number of submunitions also has a bearing on the likelihood of a munition creating an ‘indiscriminate area effect’, even where each of those submunitions is ‘designed to detect and engage a single target object’.

2.130  There are, however, certain munitions where the relationship between the ‘cluster munition’ and its ‘submunitions’ is not straightforward. The CBU-97/CBU-105 ‘Sensor Fuzed Weapon’, which is prohibited under the Convention on Cluster Munitions, contains 10 BLU-108 units, each of which in turn dispenses four explosive devices (commonly called ‘skeets’ or ‘warheads’ by the manufacturers) that have the capacity to detect and engage single object targets. In this situation the individual so-called skeets are the ‘explosive submunitions’ under the terms of the Convention because they match the description of ‘explosive submunition’ whereas the BLU-108 taken as a whole does not.157 On the other hand, the BLU-108 should not itself be considered a ‘cluster munition’ where it is used as a component of the CBU-97/CBU-105 but rather to be part of the mechanism by which the cluster munition actively disperses the submunitions.

ii.  Each explosive submunition weighs more than four kilograms

2.131  As noted above, the minimum weight requirement for submunitions serves to prevent miniaturization of submunitions considered acceptable under other criteria and hence reduces the likelihood of such submunitions being deployed in such quantities that they might still create the problematic effects associated with cluster munitions in the past. It was also argued that smaller submunitions are more likely to be handled and engaged with by civilian populations and that larger munitions may be less sensitive and therefore less problematic to clear.158 The characteristic was derived from a Norwegian (p. 199) proposal,159 but in the final stages of the negotiations the weight limit was reduced from five kilograms to four kilograms.160 A proposal for decreasing the weight limit to three kilograms had been informally presented to the Irish Presidency, and a proposal for four kilograms was presented as a compromise.

iii.  Each explosive submunition is designed to detect and engage a single target object

2.132  The requirement that each explosive submunition be designed to detect and engage a single target object is the most critical distinguishing feature for weapons with submunitions that are not considered cluster munitions under paragraph 2(c). While the other technical criteria can all be held to limit the effects associated with cluster munitions, this requires munitions to function in a way that is arguably very different to cluster munitions as commonly understood. Submunitions generally held to fulfil this requirement to ‘detect and engage a single target object’ include the German SMArt 155 or the French/Swedish BONUS, and the ‘skeet’ warheads deployed via BLU-108 from the US CBU-97/CBU-105 Sensor Fuzed Weapon.161 The legal status of these submunitions is, however, not only dependent on these capacities, but also on the other cumulative characteristics relating to their weight, electronic self-destruct and self-neutralization mechanisms and the number deployed in each container. These submunitions all use combinations of infrared and other sensors to identify certain types of object and then engage this target while still in the air, by detonating to create an explosively formed projectile.162 As a result, the explosive force of the submunitions is generally directed downwards and narrowly onto the target object without the lateral anti-personnel effects at ground level associated with traditional impact-fuzed cluster munitions.

2.133  Weapons that detect and engage targets in this way are often referred to in general as sensor-fuzed weapons.163 Certain of these weapons are unitary munitions (i.e. they do not use submunitions) and so do not fall under the (p. 200) Convention on Cluster Munitions at all. Others, such as the US CBU-97/ CBU-105 with ‘skeet’ submunitions, do fall under the Convention and are prohibited due to the number of submunitions (40) and the weight of these submunitions (less than four kilograms). Thus, the Convention does not offer a blanket exclusion for sensor-fuzed munitions as a category but splits that category into prohibited and not-prohibited variants. The only such weapon system thought not to be prohibited by the Convention on Cluster Munitions and that had been used in combat at the time of the Convention’s negotiation was the US SADARM system.164 This system was used by US forces in Iraq in 2003. No civilian casualties had been presented as resulting from this specific system at the time of the Dublin negotiations though Human Rights Watch in an analysis of the conflict in Iraq noted that further research would be needed ‘to determine the weapon’s humanitarian effects’.165

2.134  Munitions such as SMArt 155 and BONUS, and the ‘skeet’ explosive submunitions of the BLU-108 do not distinguish between military objectives and civilian objects but rather are programmed to identify as targets objects that fit profiles of heat and shape associated with certain military objects (typically armoured fighting vehicles). It is not known what analysis has been done of how such profiles compare with the profiles of objects commonly in civilian use (such as different types of civilian vehicles, generators, and other heat sources) or how susceptible such submunitions are to identifying objects commonly in civilian use as targets. While certain types of (p. 201) weapons with submunitions that ‘detect and engage individual target objects’ are excluded from prohibition under the Convention, they are nonetheless still subject to the general rules of international humanitarian law that govern attacks. Any limitations of such systems in distinguishing between objects that may be in civilian use and objectives that may be in military use might limit circumstances in which they can lawfully be used. For example, the likelihood of an attack being lawful in a desert situation where a party is attacking an area full only of military vehicles is greater than for an attack in a populated area where military vehicles are interspersed with vehicles in civilian use.

2.135  The phrase ‘detect and engage’ is not defined further in the Convention. The Lima Discussion Text provided an exclusion for certain weapons with submunitions ‘designed to, manually or automatically, aim, detect and engage point targets’. In contrast, a number of proposals in Wellington referred only to weapons ‘designed to engage’ certain targets. The requirement that the submunitions ‘detect’ the target establishes a greater sense of autonomous targeting within the submunition itself and suggests that the force of the submunition is to be applied to the single target object directly.

2.136  This sense of the term engage is linked to the phrase ‘single target object’. During the negotiations this term was adopted in preference to the term ‘point target’, which was already defined by NATO as ‘a target that requires the accurate placement of bombs or fire’. This definition did not seem practical as a legal definition. Instead, by requiring a ‘target object’ the sense is given that what is being referred to are anti-vehicle or anti-matériel weapons, rather than anti-personnel weapons. Furthermore, by requiring each submunition to engage a ‘single’ target object the sense is further given that the explosive force of the submunition should be directed at and limited to the individual target object.

(p. 202) iv.  Each explosive submunition is equipped with an electronic self-destruction mechanism
2.137  In this context, a self-destruction mechanism is intended to automatically destroy the submunition in the event that it fails to detect a single target object or, having detected such an object, fails to ‘engage’ it in a way that results in the destruction of the submunition. Thus in terms of the analysis above of ‘risks posed by unexploded submunitions’, the self-destruction mechanism works to reduce the likelihood of a submunition being left unexploded after use. A self-destruction mechanism is defined at Article 2, paragraph 9 as:

‘Self-destruction mechanism’ means an incorporated automatically-functioning mechanism which is in addition to the primary initiating mechanism of the munition and which secures the destruction of the munition into which it is incorporated.

There is no specific definition of ‘an electronic self-destruction mechanism’. During the preparatory discussions and negotiation of Article 2 it was noted that electronic self-destruction mechanisms were preferable to mechanical self-destruction mechanisms because they were both more reliable and more amenable to in-service reliability testing.166

v.  Each explosive submunition is equipped with an electronic self-deactivating feature
2.138  There is no specific definition of an ‘electronic’ self-deactivating feature in the Convention. In this context, a self-deactivating feature is intended to render an unexploded submunition incapable of functioning.167 Thus, in terms of the analysis above of ‘risks posed by unexploded submunitions’, a self-deactivation feature works to reduce the threat posed by any submunition that has been left unexploded. A self-deactivating mechanism is defined in Article 2, paragraph 10 as follows:

‘Self-deactivating’ means automatically rendering a munition inoperable by means of the irreversible exhaustion of a component, for example a battery, that is essential to the operation of the munition.

Paragraph 3

‘Explosive submunition’ means a conventional munition that in order to perform its task is dispersed or released by a cluster munition and is designed to function by detonating an explosive charge prior to, on or after impact;

Commentary

2.139  Explosive submunitions are conventional munitions168 that are dispersed and released by cluster munitions and which are designed to detonate an explosive charge before, upon, or following impact, typically with the target object or the ground. Different types of cluster munition have contained from fewer than 10 up to several hundred explosive submunitions in a single (p. 203) container. When the cluster munition opens, the submunitions inside are generally scattered over the area below. Explosive submunitions have often consisted of a body formed of high explosive encased in metal, and a structure or mechanism for orienting and stabilizing the submunition in the air (sometimes also serving to arm the munition). Submunitions may have a wide range of further characteristics relating to the type of targets they are designed to attack, their methods of arming and detonation, and the presence or absence of further features intended to reduce the risks of their creating UXO contamination.169

2.140  Considerations of target type have seen submunitions with combinations of pre-formed fragmentation around the outer casing (to create an anti-personnel effect), a shaped charge at the forward end (to create an anti-armour effect), and an incendiary lining (to create a further anti-matériel effect). Methods of arming may use the spin of the submunition in the air or the inertia of certain components under acceleration. Among other methods, detonation may be initiated through the inertia of a firing pin under rapid deceleration of the submunition or stress applied to a piezo-electric170 crystal. Some more recent submunitions have been fitted with self-destruct mechanisms, so-called self-neutralization mechanisms, and self-deactivation mechanisms in an effort to reduce the likelihood of UXO being produced as well as to reduce the threat of any items that are left unexploded.171

‘Other definitions’ (paragraphs 4–15)

Overview

2.141  The remaining 12 definitions in Article 2—the so-called ‘other definitions’—were far less controversial during the Oslo Process than the first three, but they play an important role in clarifying the meaning of the Convention on Cluster Munitions. Several deal with terms used in Article 1 on General Obligations and Scope of Application. The definition of transfer (paragraph 8) spells out what the prohibition on transfer encompasses while the definition of mine (paragraph 12) describes a weapon not covered by the (p. 204) Convention. Paragraphs 13 and 14 work with Article 1, paragraph 2 to help close a loophole that would have meant the Convention did not apply to explosive bomblets launched from dispensers. The definitions of self-destruction mechanism and self-deactivating (paragraphs 9 and 10) relate to Article 2, paragraph 2, helping to elucidate the definition of cluster munition. Finally, six definitions articulate the parameters of the provisions related to clearance. They include the definitions of cluster munition remnants and its components (paragraphs 4–7 and 15) and of cluster munition contaminated area (paragraph 11).

Preparatory discussions

2.142  While Article 2, paragraphs 1 to 3 were the subject of more discussion during the Oslo Process, the other 12 definitions, ultimately articulated in paragraphs 4 to 15, developed alongside them. None of these definitions appeared in the Lima Discussion Text, which only provided a preliminary definition of cluster munition.172 At the May 2007 Lima Conference, one State noted the need to define other terms, such as explosive submunition, abandoned submunition, and cluster munition remnants existing at the time of entry into force.173 Another State called for a definition of transfer, an activity banned under Article 1.174 There was no formal discussion of the content of these proposed definitions, however.

2.143  The Vienna Discussion Text responded to the call to define terms related to cluster munitions that become threats as ERW. In that text, unexploded cluster munitions referred to cluster munitions that were prepared for use and used yet failed to explode as intended. Abandoned cluster munitions referred to unused cluster munitions that were ‘discarded or dumped’ by a party and were no longer under the control of that party. The text used the overarching term cluster munition remnants to encompass the two categories just mentioned.175 Cluster munition remnants replaced the Lima Discussion Text’s use of ‘unexploded ordnance from cluster munitions’ in the Articles on (p. 205) clearance, international cooperation and assistance, and transparency.176 All three definitions drew heavily from Article 2 of 2003 Protocol V.177

2.144  The Vienna Discussion Text also added a definition of transfer. The definition read:

‘Transfer’ means the physical movement of cluster munitions into or from national territory or the transfer of title to or control over cluster munitions, but does not include the transfer of territory containing cluster munition remnants.178

It included the main elements of the definition of transfer in the 1997 Anti-Personnel Mine Ban Convention but changed their presentation. The 1997 Anti-Personnel Mine Ban Convention states that: ‘“Transfer” involves, in addition to the physical movement … , the transfer of title … and control … ’.179 The Vienna Discussion Text was arguably clearer than its predecessor. It specified that transfer meant either physical movement or transfer of title and control. The 1997 Anti-Personnel Mine Ban Convention, by contrast, left open for debate whether transfer meant either physical movement or transfer of title and control or whether it meant there had to be both physical movement and transfer of title and control to be transfer.180 The structure of the definition of transfer would be discussed again during the May 2008 Diplomatic Conference in Dublin.

2.145  The Explanatory Annex to the Vienna Discussion Text made clear that the new Convention would not cover landmines because they were governed by other treaties.181 The Annex listed them as one of several exclusions to the definition of cluster munition. The text, however, did not define mine.

(p. 206) 2.146  At the December 2007 Vienna Conference, discussion on the other definitions was again limited but some participants made relevant statements. Regarding the Vienna Discussion Text’s definition of transfer, the CMC argued that it should make clear that transit of cluster munitions through national territory is explicitly prohibited.182 The ICRC, concerned about ambiguities in the definition of unexploded cluster munition, suggested the development of a definition for unexploded submunitions.183 One State called for a definition for contaminated field; the Vienna Discussion Text used a variety of undefined terms for that concept.184 Several States expressed support for the Explanatory Annex’s exclusion of mines although they did not address how to define mine.185 Finally, the GICHD noted a possible loophole in the Convention—submunitions released from a dispenser affixed to an aircraft, instead of from a cluster munition.186 States would seek to close this loophole at the Diplomatic Conference in part by adding definitions to Article 2.

2.147  The Draft Cluster Munitions Convention made only small changes to the other definitions in the Vienna Discussion Text. Most notably it clarified that ‘“[u]nexploded cluster munitions” includes both unexploded parent munitions and unexploded explosive sub-munitions’.187 It also, in Article 1, made the first effort to define mine, combining the definition with a statement of its exclusion from the scope of the Convention. It said the definition of mines would be the same as that in 1996 Amended Protocol II.188

2.148  At the Wellington Conference, Chair of the Conference Don MacKay of New Zealand set the other definitions aside to focus on the more controversial definition of cluster munition.189 There were, however, (p. 207) significant proposals for other definitions. The ICRC proposed a structure for Article 2 that included a definition for unexploded explosive submunitions and introduced the concept of failed cluster munitions.190 Several government statements on other definitions were also made in the main hall. While there was no dispute that the new Convention should not apply to mines, two States objected to defining the weapons with reference to 1996 Amended Protocol II, to which they were not a party.191 One State also suggested defining cluster munition areas as ‘areas which are dangerous due to the suspected presence of cluster munitions’.192

2.149  Despite the limited attention paid to these definitions during the Wellington Conference, the Wellington Compendium, which would be forwarded to the Diplomatic Conference in Dublin, included six related proposals. Several sought to clarify the definition of cluster munition remnants and its components. Following the ICRC proposal mentioned above, Ireland proposed deleting the definition of unexploded cluster munitions and replacing it with failed cluster munition and unexploded explosive submunition. Failed cluster munition referred to a cluster munition that suffered a catastrophic failure, in other words that was delivered but failed to disperse or release its submunitions. Unexploded explosive submunition meant an explosive submunition that had separated from a cluster munition but ‘failed to explode as intended’. Ireland’s proposal updated the definition of cluster munition remnants accordingly.193 The ICRC’s suggestion noted earlier was also recorded in the Compendium.194

(p. 208) 2.150  France and Germany also proposed revisions to the definitions related to cluster munition remnants but took theirs almost verbatim from 2003 Protocol V.195 Like Ireland and the ICRC, they dispensed with unexploded cluster munition and added a definition of unexploded submunition although they did not address the category of failed cluster munitions.196 They also proposed a revised definition of abandoned explosive cluster munition and a new one of existing explosive remnants of submunitions.197 They designed the latter to accompany a proposal they made for a provision on user State responsibility for clearance of unexploded submunitions that pre-dated entry into force of the Convention.198 The French and German proposal updated the definition of cluster munition remnants to cover unexploded submunitions and abandoned explosive cluster munitions.199

2.151  The Wellington Compendium included a second set of proposals that sought to address the loophole GICHD had pointed out in Vienna regarding munitions released from a dispenser. Ireland defined the term explosive bomblet as ‘a munition which in order to perform its task is dispersed or separated from a dispenser, affixed to an aerial platform, and is designed to function by detonating an explosive charge prior to, on or after impact’. These (p. 209) munitions were not covered by the definition of explosive submunitions. Explosive submunitions are released from cluster munitions, which, unlike dispensers, are a type of munition. Explosive bomblets pose the same threat to civilians as explosive submunitions, however, because they have an area effect and are prone to leaving behind UXO. Ireland included unexploded explosive bomblets under its revised definition of cluster munition remnants.200

2.152  France and Switzerland adopted a different approach to the dispenser problem in separate proposals in the Wellington Compendium. They listed dispensers as a type of cluster munition. France described a cluster munition as a type of carrier-container and defined carrier-container, in part, as ‘a dispenser, affixed to an aircraft, which is not designed to dispense direct-fire munitions’.201 Switzerland offered two options for the description of dispenser within the definition of carrier-container. Option one stated a dispenser was ‘affixed to an aircraft, which is designed to dispense multiple sub-munitions in a single act’. Option two was identical to France’s proposal.202

2.153  Finally, Indonesia formalized its proposal on cluster munition areas in the Wellington Compendium. It defined them as ‘areas which are dangerous due to the presence or suspected presence of cluster munitions’.203

Negotiations

2.154  While efforts to finalize the definition of cluster munition dominated negotiations of Article 2 at the Diplomatic Conference, Friend of the President on Definitions Don MacKay asked Lieutenant-Colonel Jim Burke of Ireland to convene informal meetings devoted to the other definitions. During these discussions, several definitions were added to those in the Draft Cluster Munitions Convention and others were revised. The meetings also (p. 210) addressed the definition of explosive submunition, the negotiation of which is discussed above.204

2.155  Participants in these meetings discussed the various proposals from the Wellington Compendium that had been resubmitted at the Diplomatic Conference. The Draft Cluster Munitions Convention had used the term unexploded cluster munitions to encompass both parent munitions and submunitions that were used but did not explode. The ICRC explained that that term was inappropriate because cluster munitions merely release submunitions and are not designed to explode themselves.205 Ireland and the ICRC both suggested as an alternative dividing that term into failed cluster munition and unexploded submunition.206 There was no public disagreement with this point. The Convention on Cluster Munitions did not reflect the French and German proposal based on 2003 Protocol V; the negotiating States preferred to include the Irish proposal almost verbatim.207

2.156  Abandoned cluster munition received similarly brief attention. The GICHD suggested changing ‘discarded’ to ‘left behind’ to encompass more passive abandonment.208 The final version of the definition used ‘left behind’. It also added the phrase ‘or explosive submunitions’ to ensure it encompassed caches of individual submunitions as well as the parent cluster munitions.209

2.157  Negotiators generally supported the Wellington Compendium proposals for the Convention on Cluster Munitions to cover explosive bomblets. Ireland called on States to cover dispensers with explosive bomblets under Article 1 and to define the terms in Article 2.210 Expressing support for the Irish proposal, one State said the French and Swiss alternatives relating to (p. 211) carrier-containers were too complicated.211 The Convention on Cluster Munitions ultimately included in the definition both dispenser and explosive bomblet. Dispenser, defined for the first time at the Diplomatic Conference, meant a container ‘affixed to an aircraft’ that released explosive bomblets.212 The definition of explosive bomblet, based on Ireland’s original proposal in the Wellington Compendium, was expanded to add a weight requirement of less than 20 kilograms and to specify that it was not self-propelled. In addition, the phrase ‘was separated’ was changed to ‘released’.213 The Convention also defined unexploded bomblet, making only minor revisions to the Irish proposal.214 To cover all of the different types of ERW now listed in the Convention, cluster munition remnants was redefined to include ‘failed cluster munitions, abandoned cluster munitions, unexploded submunitions and unexploded bomblets’.215

2.158  The final proposal for Article 2 in the Wellington Compendium was Indonesia’s proposal to define cluster munition area. Participants at the Diplomatic Conference generally supported adding this term in the list of definitions,216 but made several amendments to Indonesia’s original language in the Compendium. One State proposed changing cluster munitions to cluster munition remnants.217 The ICRC suggested removing the phrase ‘which are dangerous’, even though it came from the 1997 Anti-Personnel Mine Ban Convention, because it was too subjective.218 Ultimately, language suggested by the CMC served as the basis for the final definition, which read ‘an area known or suspected to contain cluster munition remnants’.219

(p. 212) 2.159  Although not discussed in Wellington, transfer was modified somewhat during the negotiations of Article 2 in Dublin.220 Some States objected to the fact that the definition in the Draft Cluster Munitions Convention differed from those in the 1997 Anti-Personnel Mine Ban Convention and 1996 Amended Protocol II.221 On 22 May, Norway therefore proposed a new definition almost identical to those in the earlier treaties.222 The ICRC expressed concern that the ‘in addition to’ formulation of past instruments would lead to confusion about whether transfer required both physical movement and transfer of title and control, or required only one or the other.223 At the Committee of the Whole on 26 May, Jim Burke, who was chairing the discussions on the other definitions, noted that it would be difficult to achieve consensus on this topic. Ultimately the Convention on Cluster Munitions adopted verbatim the language from the 1997 Anti-Personnel Mine Ban Convention and 1996 Amended Protocol II.

2.160  As the definition of cluster munition evolved during the Diplomatic Conference, it became clear that definitions of self-destruction mechanism and self-deactivating needed to be added to the text of the Convention.224 Spain made the first proposal, drawing language directly from 1996 Amended Protocol II and defining the terms as follows:

‘Self-destruction mechanism’ means an incorporated or externally attached automatically-functioning mechanism which secures the destruction of the mechanism into which it is incorporated or to which it is attached.

(p. 213) ‘Self-deactivating’ means automatically rendering a munition inoperable by means of the irreversible exhaustion of a component, for example a battery, that is essential to the operation of the munition.225

2.161  Three days later, Norway offered an alternative proposal that it said sought to strengthen the definitions in 1996 Amended Protocol II. Its proposal stated:

‘Self-destruct mechanism’ means a mechanism that physically destroys the warhead in the event that it does not function as intended and thus leav[es] no unexploded objects behind;

‘Self-deactivation mechanism’ means a mechanism that drains the sub-munition of the energy required to bring it to detonation and thus render[s] the remaining unexploded object safe to handle and safe in any incidental contact.226

Norway’s definition of self-destruct mechanism did not address whether the mechanism was incorporated into the munition and put more emphasis on effects, i.e., ‘leaving no unexploded objects behind’. Besides turning the adjective self-deactivating into the noun self-deactivation mechanism, Norway’s definition of the latter again focused on effects, i.e., ‘rendering the remaining unexploded object safe to handle’.

2.162  The Norwegian proposal attracted some support, but other participants in the negotiations objected to its approach. Several States said they were willing to adopt the Norwegian version although one suggested adding ‘is designed to’ before ‘destroy’ and ‘drain’ in order to cover mechanisms that functioned improperly.227 ICRC welcomed the effort to clarifying the language in 1996 Amended Protocol II; however, it questioned the use of the new term warhead and suggested using explosive ordnance from 2003 Protocol V instead.228

2.163  Concerns centred on the last phrase of self-deactivation mechanism. One State said civilians often find creative ways to use UXO so UXO never reaches the level of being entirely safe to handle.229 Another noted that it is (p. 214) problematic from a mine action perspective to describe UXO as safe to handle.230 At least four States said they preferred the existing version in 1996 Amended Protocol II.231

2.164  In the end the States rejected Norway’s proposals and adopted the approach of 1996 Amended Protocol II. The definition of self-deactivating was identical to that in the Protocol. The definition of self-destruction mechanism took out references to being ‘externally attached’ and clarified that such a mechanism was ‘in addition the primary initiating mechanism of the munition’.232

2.165  Finally, States discussed the insertion of a definition of mine. In the Draft Cluster Munitions Convention, mine had been defined in Article 1 by reference to the definition in 1996 Amended Protocol II.233 Indonesia reiterated its earlier objection to that approach in the first meeting of the Committee of the Whole and proposed replacing it with the language of the 1997 Anti-Personnel Mine Ban Convention.234 The final text of the Convention on Cluster Munitions adopted Indonesia’s suggestion. It moved the definition of mine to Article 2 and used the definition of the 1997 Anti-Personnel Mine Ban Convention.235

2.166  Article 2 changed dramatically over the course of the Oslo Process not only because of debate about the first three definitions but also because of the addition and evolution of the other 12. Some of the final definitions, notably transfer, self-deactivating, and mine, reproduced the language of earlier treaties verbatim. Others, such as the components of cluster munition remnants, self-destruction mechanism, and cluster munition contaminated area, followed the approach of earlier treaties but sharpened the precedent or tailored it to the needs of this new Convention. Still others, in particular explosive bomblet and dispenser, emerged to deal with areas previously unaddressed in international law. While the first three definitions of Article 2 determined the scope of some of the core obligations of the Convention, the other definitions made its provisions clearer and more precise.

(p. 215) Paragraph 4

‘Failed cluster munition’ means a cluster munition that has been fired, dropped, launched, projected or otherwise delivered and which should have dispersed or released its explosive submunitions but failed to do so;

Commentary

2.167  A failed cluster munition is one of four types of cluster munition remnants.236 Its definition covers the case where a cluster munition fails to open and release the explosive submunitions it contains. Such a situation is often referred to as a catastrophic failure. When a cluster munition falls to the ground intact, it may break open on impact scattering submunitions in the immediate area of the crash. These submunitions may not have armed themselves as they have not been released or dispersed. However, they remain a danger to civilians and could explode if they are disturbed, manipulated, or tampered with.

2.168  A failed cluster munition must first and foremost meet the requirements of Article 2, paragraph 2 to establish that it is a cluster munition for the purposes of the Convention. In addition, the weapon must have been fired, dropped, launched, projected, or otherwise delivered. In other words, the munition must have been used.237 These terms are not specifically defined in the Convention, but they generally describe the methods by which air- or ground-launch systems deliver cluster munitions. They are also key criteria for distinguishing a failed cluster munition from an abandoned cluster munition.

2.169  The main characteristic of a failed cluster munition is that it has been unable—for whatever reason—to release or disperse the explosive submunitions it is carrying. As a result, it threatens civilians and is included in the definition of cluster munition remnants. The clearance of failed cluster munitions is essential, and as a component of cluster munition remnants, it is governed by the obligations and deadlines laid out in Article 4 of the Convention. Articles 6 and 7 also establish obligations regarding failed cluster munitions as a type of cluster munition remnant. Article 6, paragraph 4 obliges each State Party in a position to do so to provide assistance for the clearance of cluster munition remnants. Article 7, paragraph 1(i) requires a State Party to report on the status and progress of programmes for the (p. 216) clearance and destruction of all types of cluster munition remnants, including failed cluster munitions.238

2.170  The term failed cluster munition and the other terms in the definition of cluster munition remnants were developed to ensure that the Convention, and particularly Article 4 on the clearance of cluster munition remnants, clearly applied to the different circumstances in which cluster munitions pose a danger to civilians. As explained below in more detail,239 ambiguities in a number of draft definitions raised concerns that important aspects of the cluster munitions problem would not be covered by the obligations being proposed in the drafts of the Convention.

2.171  The Lima Discussion Text did not specify if it applied to a cluster munition that failed to release or disperse its submunitions. It did not specifically include or define concepts such as failed cluster munitions and unexploded cluster munitions. Although the Lima Discussion Text required each State Party ‘to clear all unexploded ordnance from cluster munitions in areas under its jurisdiction and control … ’,240 the scope of this obligation was ambiguous. The requirement to clear UXO originating from cluster munitions would almost certainly apply to unexploded submunitions that had been released and failed to explode as intended (the most obvious form of UXO), but it was less clear to what extent it also encompassed cluster munitions that had failed to open and remained intact when hitting the ground.241

2.172  While the Vienna Discussion Text set out the first definitions for unexploded cluster munitions, abandoned cluster munitions, and cluster munition remnants, it did not completely clarify the situation on failed cluster munitions. At first glance, failed cluster munitions would appear to be integrated into the definition of unexploded cluster munitions.242 However, the definition stated that unexploded cluster munitions ‘should have exploded but failed to do so’. This sentence created confusion because it is the submunition, and not the cluster munition itself, that is expected to explode.243

(p. 217) 2.173  There was an attempt to clarify this matter in the Draft Cluster Munitions Convention. This text added a sentence to the definition in the Vienna Discussion Text to specify that the definition ‘includes both unexploded parent munitions [i.e., failed cluster munitions] and unexploded submunitions’.244 Concerns remained about the ambiguity of the definition, however, which led the ICRC to suggest a different structure for Article 2.245 It proposed that the definition of unexploded cluster munition be changed to failed cluster munition and that a definition of unexploded explosive submunition be added to Article 2.246 It also proposed that both of these concepts be included in the definition of cluster munition remnants to indicate that both were covered by the Convention.247 The CMC supported this restructuring.248

2.174  Ireland subsequently submitted a proposal on Article 2 for the Wellington Compendium that included the ICRC’s suggestions as well as additional elements to ensure that Article 2 was comprehensive.249 It resubmitted it at the May 2008 Diplomatic Conference in Dublin. The Irish proposal (p. 218) included the following definition of failed cluster munition:

‘Failed cluster munition’ means a cluster munition that has been fired, dropped, launched, projected or otherwise delivered and which should have dispersed or released its explosive submunitions but failed to do so.250

The language proposed in this definition received broad support in both the informal consultations on definitions and the plenary sessions of the Dublin negotiations. The final text of the Convention adopted it verbatim.

Paragraph 5

‘Unexploded submunition’ means an explosive submunition that has been dispersed or released by, or otherwise separated from, a cluster munition and has failed to explode as intended;

Commentary

2.175  Unexploded submunitions are generally regarded as a principal feature of the cluster munition problem. The large numbers of submunitions released when cluster munitions are used and the significant quantities that often fail to explode as intended can quickly contaminate wide swathes of land. Unexploded submunitions have been a high proportion, and sometimes the bulk, of the UXO threat in countries where cluster munitions have been used.

2.176  The definition of unexploded submunition is intended to ensure that the Convention’s obligations clearly apply to those submunitions that have separated from the cluster munition container but have not exploded. Unexploded submunition is a component of cluster munition remnants and as such its definition has particular relevance for Article 4 of the Convention as well as some provisions in Articles 6 and 7.251

2.177  Three elements must be present for a munition to qualify as an unexploded submunition. First, the munition in question must be an explosive submunition as defined in Article 2, paragraph 3 of the Convention. Second, the explosive submunition must have been dispersed, released, or otherwise separated from the cluster munition container. Finally, the explosive submunition must have failed to explode as intended.

(p. 219) 2.178  The Convention does not specifically define the terms dispersed, released, and otherwise separated. Thus, they are to be understood in accordance with their everyday meaning taking into account the context in which they are used. The main notion underlying these terms in this case is that the submunitions are no longer physically contained in but are now independent of the cluster munition.

2.179  The phrase ‘failed to explode as intended’ is meant to take into account that there are different ways through which the detonation of an explosive submunition can be initiated. Many models are designed to detonate following impact on a hard surface. Others explode after a time-delay. This distinction is highlighted in the definition of explosive submunition where it refers to munitions ‘designed to function by detonating an explosive charge prior to, on or after impact’. The main consequence of ‘failed to explode as intended’ is that submunitions designed to explode after a time delay would not be considered unexploded submunitions during the period that their detonation was intentionally delayed but would become an unexploded submunition once that time period had expired.252

2.180  Like failed cluster munition, the definition of unexploded submunition was a rather late addition to Article 2. The Lima and Vienna discussion texts and the Draft Cluster Munitions Convention did not contain a specific definition for unexploded submunition. Instead, the definition of unexploded cluster munitions appeared to encompass the concept. The ICRC noted a lack of clarity at the Vienna Conference, however:

We have some concern that several definitions in Article 2 may be ambiguous and cause confusion as States implement their obligations. For example, we feel that it is unclear whether the definition of ‘unexploded cluster munitions’ encompasses submunitions which have been dispersed and failed to explode as intended (unexploded submunitions). It might be read as describing only a cluster munition which had been delivered but failed to disperse its submunitions. This can, for example, have implications on how a State interprets the clearance obligations in subsequent articles.253

2.181  To overcome these concerns, the ICRC proposed that a separate definition of unexploded submunitions be added to the draft Convention.254 Such a definition was not, however, included in Article 2 of the Draft Cluster Munitions Convention which served as the basis for negotiations at the (p. 220) Diplomatic Conference. Nevertheless, it was a component of the Irish and ICRC proposals to restructure Article 2 that were annexed as a compendium to the Wellington Declaration and forwarded to the Diplomatic Conference, with support from the CMC.255 At the negotiations, France and Germany also submitted a proposal to amend Article 2 that included a definition of unexploded submunition based on 2003 Protocol V.256

2.182  Following informal discussions at the Diplomatic Conference, the Friend of the President on Article 2 circulated several papers suggesting possible language for definitions including that of unexploded explosive submunition. This language closely followed the Irish proposal. Overall, the elements of this definition were not contentious or the subject of extensive debate in the informal discussions or the Committee of the Whole. The most significant change to occur in the final text of the Convention was the deletion of the term ‘explosive’ in the main heading.257

Paragraph 6

‘Abandoned cluster munitions’ means cluster munitions or explosive submunitions that have not been used and that have been left behind or dumped, and that are no longer under the control of the party that left them behind or dumped them. They may or may not have been prepared for use;

Commentary

2.183  A central feature of failed cluster munitions and unexploded submunitions is that the weapons have been used. In the case of failed cluster (p. 221) munitions, they have been fired, launched, or otherwise delivered, and if they are unexploded submunitions, they have separated from the cluster munition container. As such, neither term covers the situation in which stockpiles of cluster munitions have been left behind on the battlefield or at a military base and not used. In addition, the terms may not be seen to include situations where the aircraft of a party to a conflict has dumped its cluster munitions prior to returning to its base for landing—as sometimes occurs when an aircraft returns to an aircraft carrier at sea.258 The definition of abandoned cluster munitions would fill this gap.

2.184  The concept of abandoned explosive ordnance is a rather new addition to international humanitarian law. It was first proposed and adopted in the context of 2003 Protocol V. As noted in an early paper proposing the framework for an ERW protocol, as no definition of abandoned explosive ordnance existed, one would need to be developed during the Protocol’s negotiation.259 As a result, the definition of abandoned cluster munitions in the Convention on Cluster Munitions borrows a number of concepts from the Protocol.260

2.185  To qualify as abandoned cluster munitions, weapons must first meet the definition of cluster munition or explosive submunition.261 As mentioned above, one of the main features of abandoned cluster munitions is that they have not been used, which helps clarify that the definition is not intended to encompass munitions that have been used and failed to explode (i.e., failed cluster munitions or unexploded submunitions). Rather, it is meant to cover cluster munition stockpiles or even individual cluster munitions that may have been left or discarded by a party to the conflict. Such weapons can be found on the battlefield, around military positions, and in or near military bases and depots.

(p. 222) 2.186  Another element of this definition is that the weapons have been left or dumped by a party to the conflict. The terms ‘left behind or dumped’ indicate that a cluster munition can become abandoned through the passive or active conduct of a State Party or a party to the conflict. In this respect, the definition of abandoned cluster munition would encompass instances where a party moves from an area or position and leaves behind a stock of cluster munitions as well as situations where the party actively discards the weapons. An additional element of this definition is that the weapons are no longer under the control of the party that left them behind or dumped them. This criterion is meant clearly to reflect that the munitions are indeed abandoned.

2.187  Both large and small stockpiles of cluster munitions are, in principle, covered by the definition as there is no requirement that the stocks be a particular size or that they contain a minimum number of munitions. Although the definition is phrased in the plural, it would be logical that individual cluster munitions are also covered even if they are not amassed in a stockpile. First, small numbers of munitions, even individual munitions, can be left at temporary military positions. Second, cluster munitions are likely to be discarded individually and not in large numbers. Finally, a different understanding on the scope of this definition would mean that an individual cluster munition that had been abandoned would not be covered by Article 4 as it would not meet any of the requirements to be a cluster munition remnant. This reading would be contrary to the object and purpose of the Convention.

2.188  The definition also explicitly includes explosive submunitions that have not been used but have been left behind or dumped by a party to the conflict. It would thus cover situations where a dumped cluster munition broke open exposing or spilling the explosive submunitions contained inside. It would also include instances where explosive submunitions spilled or otherwise separated from an abandoned stockpiled cluster munition due to corrosion, ageing, or other factors. Although these submunitions had separated from the cluster munition, it might be argued they did not meet the definition of an unexploded submunition because they had not been used or failed to explode as intended. The reference to explosive submunitions in this definition helps ensure that the Convention covers the variety of scenarios where cluster munitions and explosive submunitions could pose a danger to civilians.

2.189  Finally, the last sentence of the definition indicates that cluster munitions and explosive submunitions that may or may not have been prepared for use qualify as an abandoned cluster munition. It is meant to clarify that such weapons need not be armed, fuzed, or otherwise ready for use. A similar clarification is found in the definition of abandoned explosive ordnance in Article 2 (p. 223) of 2003 Protocol V. It was included in that instrument in response to discussions on whether it was necessary for the protocol to encompass weapons that had not yet been prepared for use, on the basis that they did not pose an immediate danger to civilians. Clearance organizations stressed that it was essential for the Protocol to include all explosive munitions, whether or not prepared for use, in light of the fact that such weapons would still need to be cleared as part of their operations.262 They also highlighted that such weapons are a danger as civilians will often tamper with them to extract explosives or metal.

2.190  Abandoned cluster munitions, like failed cluster munitions and unexploded submunitions, fall under the definition of cluster munition remnants. Therefore obligations related to clearance, clearance assistance, and clearance reporting in Articles 4, 6, and 7, respectively, apply to this type of munition.

2.191  The need for a definition of abandoned cluster munition was realized early in the Oslo Process. A formulation of this term appeared in Article 2 of the Vienna Discussion Text which defined abandoned cluster munitions as:

cluster munitions that have not been used and that have been discarded or dumped, and that are no longer under the control of the party that discarded or dumped them. Abandoned cluster munitions may or may not have been prepared for use.263

With the exception of one editorial adjustment, this definition was left unchanged in the Draft Cluster Munitions Convention.264 Overall, the definition was not contentious at this stage, and there was only one proposal for its amendment in the Wellington Compendium.265

2.192  At the Diplomatic Conference, the definition underwent several minor modifications in the informal discussions hosted by the Friend of the President on Definitions. The term ‘discarded’ was replaced by ‘left behind’ and ‘or explosive (p. 224) submunitions’ was inserted after ‘cluster munitions’ at the start of the first sentence. This latter change was intended to ensure that individual submunitions that escaped from a left or discarded cluster munition would fall under the definition.266 With these changes the definition moved to the President’s Text and was agreed to by the Committee of the Whole without further amendment.

Paragraph 7

‘Cluster munition remnants’ means failed cluster munitions, abandoned cluster munitions, unexploded submunitions and unexploded bomblets;

Commentary

2.193  The phrase cluster munition remnants was developed to provide an umbrella term encompassing the four main forms in which cluster munitions can generally be found in conflict-affected areas: failed cluster munitions, abandoned cluster munitions, unexploded submunitions, and unexploded bomblets.267 To qualify as a cluster munition remnant, an object must meet the requirements of one of these categories. While cluster munition remnants is not a term with technical or operational implications, its use in the Convention makes it easier to follow and understand the relevant obligations, particularly in Article 4 on Clearance and Destruction of Cluster Munition Remnants and Risk Reduction Education. It also has a bearing on the comprehension and application of the preamble of the Convention, the definitions of transfer and cluster munition contaminated area, Article 6 (international cooperation and assistance), Article 7 (transparency measures), and Article 11 (Meetings of States Parties). The term cluster munition remnants did not previously exist in any instrument but is influenced by the approach to the definition of ERW taken in 2003 Protocol V.268

2.194  The definition of cluster munition remnants evolved along with that of other definitions related to the post-conflict dangers of cluster munitions.269 (p. 225) The Vienna Discussion Text and Draft Cluster Munitions Convention defined the term as ‘unexploded cluster munitions and abandoned cluster munitions’.270 In the Wellington Compendium, Ireland proposed a definition that encompassed ‘failed cluster munitions, abandoned cluster munitions, unexploded explosive submunitions and unexploded explosive bomblets’.271 The ICRC submitted a similar proposal, but it did not include the concept of unexploded explosive bomblets.272

2.195  At the Diplomatic Conference, the Friend of the President’s first informal paper proposed a definition comparable to the ones submitted by Ireland and the ICRC in the compendium attached to the Wellington Declaration:

Cluster munition remnants means failed cluster munitions, abandoned cluster munitions and unexploded explosive submunitions.273

This wording changed slightly when it was decided to delete ‘explosive’ from the heading ‘unexploded explosive submunitions’. The amendment was more editorial than substantive as the term had also been removed from the title of the definition of unexploded submunition.274

2.196  A second change occurred once it was decided that the Convention would require definitions for explosive bomblet and unexploded bomblet.275 (p. 226) With agreement on these definitions, it was evident that the definition of cluster munition remnants needed to be broadened to reflect this development. Thus, in his final informal paper the Friend of the President proposed that unexploded bomblets be added to the definition of cluster munition remnants.276 With this amendment the definition went into the President’s Text and was adopted.

Paragraph 8

‘Transfer’ involves, in addition to the physical movement of cluster munitions into or from national territory, the transfer of title to and control over cluster munitions, but does not involve the transfer of territory containing cluster munition remnants;

Commentary

2.197  The definition of transfer addresses the relocation of cluster munitions across State borders and changes in their ownership. As a result, it clearly covers the import and export of cluster munitions by sale or through provision of military assistance. Because it links its initial phrases with ‘in addition to’ instead of merely ‘or’, however, the definition raises serious questions of interpretation. It could mean that either movement across borders or transfer of title and control is sufficient to constitute transfer. In that case, if a State Party allowed a State not party to deliver cluster munitions to a military base it is hosting on its territory, it would violate the prohibition on transfer, which is articulated in Article 1, paragraph 1(b). Many States have interpreted the virtually identical definition of transfer in the 1997 Anti-Personnel Mine Ban Convention in this way.277 Alternatively the definition could mean that there (p. 227) must be physical movement across borders and a handover of title and control for transfer to occur. This reading restricts the scope of the provision and would primarily cover sales and foreign aid. Regardless of the interpretation, a passing of title must always be accompanied by a passing of control to amount to transfer.

2.198  Transfer can also be understood to encompass transit, i.e., the movement of cluster munitions across a State’s territory whether by land, across territorial waters, or through airspace. During the Oslo Process, the CMC repeatedly argued that transfer should be read as encompassing transit. According to the Summary Records from the Diplomatic Conference, for example, the CMC stated that ‘it was the clear understanding of all States that the definition of “transfer” encompassed the transit of cluster munitions through a State Party’s territory’.278 Since the Convention opened for signature in December 2008, several States, including Bulgaria, Burkina Faso, Ecuador, Lebanon, Malta, Mexico, and South Africa, have publicly stated their view that transit is prohibited.279

2.199  The second half of the definition of transfer creates an exclusion. It states that transfer ‘does not involve the transfer of territory containing cluster munition remnants’.280 The prohibition on transfer in Article 1, paragraph 1(b) seeks to end the proliferation of weapons and the spread of use. A change in control of an area contaminated with cluster munition remnants does not interfere with that goal, and therefore the Convention did not have to include it under the prohibition on disarmament or humanitarian grounds. The exclusion does not give a State Party an incentive to acquire contaminated territories to obtain potentially usable abandoned cluster munitions because it would also assume the obligation to clear them. (p. 228) It allows, however, a State Party to take control of a territory that it wishes to help clear. Without the exclusion in the definition of transfer, the Convention would prohibit a State Party from acquiring for any reason new territory that happens to be affected by cluster munitions. Public international law, not the Convention on Cluster Munitions, should govern the legality of the acquisition of territory.

2.200  The term transfer appears in two places in the Convention on Cluster Munitions. Article 1, paragraph 1(b) makes the prohibition on transfer one of the core obligations of the Convention. That provision establishes the absolute nature of the ban by stating that transfer is prohibited ‘under any circumstances’ and ‘to anyone, directly or indirectly’.281 Transfer is the only activity prohibited by the Convention that is defined. The term also appears in Article 3, which permits ‘the transfer of cluster munitions to another State Party for the purpose of destruction’, for the ‘development of and training in cluster munition and explosive submunition detection, clearance or destruction techniques, or for the development of cluster munition counter-measures’.282 States Parties must report on the details of any such transfers.283 While allowing transfer of cluster munitions for destruction was largely uncontroversial, during the negotiating process many States challenged allowing transfer for training and development purposes.284

2.201  The definition of transfer received only a little attention before the Diplomatic Conference. At the Lima Conference, one State listed it among several terms that might need to be defined in the Convention,285 and a preliminary definition was added in the Vienna Discussion Text. That version read:

‘Transfer’ means the physical movement of cluster munitions into or from national territory or the transfer of title to or control over cluster munitions, but does not include the transfer of territory containing cluster munition remnants.286

(p. 229) It drew much of its language from the 1997 Anti-Personnel Mine Ban Convention and 1996 Amended Protocol II but organized it somewhat differently.287

2.202  During the Diplomatic Conference, several States expressed a desire to use the same language as other treaties. One State noted that this approach would be consistent with past documents and ‘avoid legal confusion’.288 In response, Norway introduced alternative language; its formal proposal structured the provision like the corresponding definitions in the 1997 Anti-Personnel Mine Ban Convention and 1996 Amended Protocol II, beginning it with ‘in addition to the physical movement of cluster munitions into or from national territory’.289 The ICRC expressed concern, saying that that formulation had led to ‘years of debate’ in the landmine context. It said that the wording did not make clear whether transfer required both physical movement and a change of title and control, or only one or the other.290 Despite the lack of clarity, the final version of the definition adopted Norway’s amendment. It also replaced ‘means’ and ‘includes’ with ‘involves’, thus making its language effectively the same as that in the 1997 Anti-Personnel Mine Ban Convention and 1996 Amended Protocol II.291

Paragraph 9

‘Self-destruction mechanism’ means an incorporated automatically-functioning mechanism which is in addition to the primary initiating (p. 230) mechanism of the munition and which secures the destruction of the munition into which it is incorporated;

Commentary

2.203  Article 2, paragraph 9 uses technical and objective language to define a self-destruction mechanism as a device intended to decrease the number of unexploded submunitions and thus the harm they cause. The definition states that it is ‘an incorporated automatically-functioning mechanism’, which means it must be a component of a munition and operate without human or other intervention. If the mechanism functions as designed, the munition destroys itself independently.

2.204  The definition adds two requirements to that basic criterion. First, the mechanism must be ‘in addition to the primary initiating mechanism of the munition’. This phrase clarifies that it serves as a secondary system. Second, it ‘secures the destruction of the munition into which it is incorporated’. In other words, it must achieve a certain goal, i.e., to eliminate an explosive submunition that failed to detonate on impact. The definition uses ‘secures the destruction’ rather than ‘is designed to secure’.292 Although presumably it was not intended, the definition literally suggests that a device is not a self-destruction mechanism if it fails to function, even if it was designed to destroy its munition and meets all the other criteria.293

2.205  The term self-destruction mechanism appears only once in the Convention. Article 2, paragraph 2(c)(iv) states: ‘Each explosive submunition is equipped with an electronic self-destruction mechanism.’294 That sub-paragraph presents one of five cumulative characteristics a weapon must have to fall under Article 2, paragraph 2(c)’s narrow exclusion to cluster munition. To reduce the ‘risks posed by unexploded submunitions’, the self-destruction mechanism backs up an explosive submunition’s primary initiating mechanism and is in turn backed up by an electronic self-deactivating feature.295 The provision further narrows the type of self-destruction mechanism by stating that it must be electronic, which, at least one expert has said, increases its chance of functioning properly.296

(p. 231) 2.206  States first highlighted the relevance of self-destruction mechanisms to the definition of cluster munition at the Lima Conference, and the discussion continued during each of the subsequent Oslo Process meetings. They saw the mechanisms as tools for reducing the ‘unacceptable’ harm of the weapons.297 Although some States initially argued that submunitions with self-destruction mechanisms should not be banned, other States and the CMC countered that the mechanisms were insufficient by themselves because they often failed in conflict.298 The latter view ultimately prevailed. States also rejected the proposition to exclude certain weapons from the definition of cluster munition based on their failure rate, which a self-destruction mechanism could lower, because it is impossible to determine an accurate rate for conflict situations in testing environments.299

2.207  Despite these lengthy discussions, States began considering language for the definition of self-destruction mechanism only during final negotiations at the Diplomatic Conference. Spain noted that the term was already well defined in international law and proposed borrowing verbatim the definition of 1996 Amended Protocol II, which states:

‘Self-destruction mechanism’ means an incorporated or externally attached automatically- functioning mechanism which secures the destruction of the munition into which it is incorporated or to which it is attached.300

(p. 232) 2.208  Norway said it sought to build on existing models and proposed defining the term as ‘a mechanism that physically destroys the warhead in the event that it does not function as intended and thus leav[es] no unexploded objects behind’.301 Norway’s version focused on the desired effects of the mechanism—destroying the warhead and leaving no unexploded submunitions— rather than the technical characteristics of the device. Several States objected to this proposal, stating that they preferred to use an already accepted definition.302 Ultimately the Convention’s final definition drew on that in 1996 Amended Protocol II and refined it. It removed references to ‘externally attached’ and clarified that a self-destruction mechanism was a secondary device ‘in addition to the primary initiating mechanism of the munition’.303

Paragraph 10

‘Self-deactivating’ means automatically rendering a munition inoperable by means of the irreversible exhaustion of a component, for example a battery, that is essential to the operation of the munition;

Commentary

2.209  The definition of self-deactivating adopts a technical and objective approach similar to that of the definition of self-destruction mechanism, and it too relates to efforts to decrease the harm of unexploded submunitions.304 To be self-deactivating, a feature must render a munition ‘inoperable’ so that it can no longer detonate even if disturbed, and it must do so ‘automatically’, i.e., independently and without outside intervention. It accomplishes this end ‘by means of the irreversible exhaustion of a component, for example a battery, that is essential to the operation of the munition’.305 The component must be (p. 233) essential to the workings of the munition in order to ensure that the munition cannot function after the component’s exhaustion when it is no longer appropriate for it to do so. While the definition allows for a range of components, including those yet to be invented, it singles out batteries as the most common components that can lose their charge and self-deactivate.

2.210  Irreversibility guarantees that the inoperability is permanent and the munition cannot become dangerous again. Like a self-destruction mechanism, a self-deactivating feature serves the purpose of removing the threat of an unexploded submunition. While the former works by detonating a weapon shortly after it has failed, however, the latter realizes its goal not by exploding the weapon but by making sure that, after a certain point, it is incapable of exploding.

2.211  The term self-deactivating is used only once in the Convention, in Article 2, paragraph 2(c)(v), which provides that: ‘Each explosive submunition is equipped with an electronic self-deactivating feature.’306 This subparagraph presents one of five cumulative criteria necessary to meet the narrow exclusion to the definition of cluster munition.307 It seeks to eliminate the danger of an unexploded submunition in a situation where both primary and secondary fuzing mechanisms have failed. That provision limits the feature even more than the definition of self-deactivating by requiring that it be electronic. Self-deactivating features generally are electronic because a mechanical mechanism can be neutralized but not exhausted.

2.212  States debated throughout the Oslo Process whether weapons excluded under the definition of cluster munition should be required to have a self-deactivating feature, but they did not discuss a formal definition of the term until the Diplomatic Conference. None of the discussion texts offered draft language.

2.213  During the Diplomatic Conference, Spain suggested adopting the definition used in 1996 Amended Protocol II,308 and Norway supported drafting a new one. Norway’s proposal stated that:

‘Self-deactivation mechanism’ means a mechanism that drains the sub-munition of the energy required to bring it to detonation and thus rendering the remaining unexploded object safe to handle and safe in any incidental contact.309

(p. 234) While Norway said it sought to ‘strengthen’ 1996 Amended Protocol II’s definition,310 the second part of its proposal describing the effects of the mechanism attracted criticism. One State, for example, expressed concern about the difficulty of meeting the standard of ‘rendering the … object … safe in any incidental contact’ because a self-deactivating mechanism could not be expected to protect against all unsafe acts, such as civilians taking submunitions apart for explosives.311 Another added that it was problematic to describe a piece of UXO as ‘safe to handle’;312 indeed, explosive ordnance disposal standards strongly oppose anything that could encourage an untrained civilian to handle ordnance. Some States said more generally that they preferred to use 1996 Amended Protocol II’s version because it had already been accepted internationally.313 In the end, States rejected Norway’s proposal and adopted Spain’s, which used verbatim the language of 1996 Amended Protocol II’s definition.314

Paragraph 11

‘Cluster munition contaminated area’ means an area known or suspected to contain cluster munition remnants;

Commentary

2.214  The definition of cluster munition contaminated area delineates what territory is unsafe and needs to be checked for contamination which, if any is found, must be cleared and destroyed in accordance with Article 4. To be considered a ‘cluster munition contaminated area’, it is sufficient that an area contain any cluster munition remnants, in other words failed cluster munitions, abandoned cluster munitions, unexploded submunitions, or unexploded bomblets.315

2.215  The presence of these remnants can be either ‘known or suspected’.316 This phrase ensures that the definition covers most possible contaminated areas. Immediately after a strike, it can be difficult to determine for certain if (p. 235) cluster munition remnants litter a field or not, especially if the submunitions have penetrated a soft surface. The Convention does not require an absolute determination because, given its humanitarian purpose, precautions must be taken with any area that is potentially contaminated in order to protect civilians.317

2.216  The term cluster munition contaminated area is used primarily in relation to clearance and risk reduction education. Article 4, paragraph 1 of the Convention obliges each State Party ‘to clear and destroy, or ensure the clearance and destruction of, cluster munition remnants located in cluster munition contaminated areas under its jurisdiction or control’.318 Article 4, paragraph 2 requires the State Party to ‘mak[e] every effort to identify all’ such areas, take ‘all feasible steps to ensure’ they are ‘perimeter-marked, monitored and protected’, and conduct risk reduction education for civilians ‘living in or around’ the areas.319 Article 7, which governs transparency measures, mandates related reports on ‘the size and location of all cluster munition contaminated areas under [a State Party’s] jurisdiction or control’, the size and location of areas cleared, and measures to warn civilians living in the areas.320 The definition clarifies the areas to which certain of the Convention’s obligations apply.

2.217  A definition of cluster munition contaminated area was first proposed only in the Wellington Compendium, but discussion of the need for such a term began at the Vienna Conference. At that meeting, one State suggested the phrase contaminated field, and the Vienna Discussion Text used a variety of terms to capture the concept, including contaminated areas.321 In the (p. 236) Wellington Compendium, Indonesia formally proposed a definition, drawn almost verbatim from the 1997 Anti-Personnel Mine Ban Convention, stating, ‘Cluster munitions areas mean areas which are dangerous due to the presence or suspected presence of cluster munitions.’322 The ICRC noted that use of the subjective word dangerous in the definition was problematic because it required a determination of what was dangerous.323 The final version removed that word and changed cluster munitions to cluster munition remnants,324 which meant that areas containing unexploded bomblets would be covered by the definition and areas containing stockpiles under destruction would not be. The Convention on Cluster Munitions also added the word contaminated to the term itself, characterizing such areas as dangerous and pointing to the need to clear them.325

Paragraph 12

‘Mine’ means a munition designed to be placed under, on or near the ground or other surface area and to be exploded by the presence, proximity or contact of a person or a vehicle;

(p. 237) Commentary

2.218  The definition of mine clarifies the differences between this weapon and an explosive submunition. Drawn completely from the 1997 Anti-Personnel Mine Ban Convention,326 it has two elements. The first relates to how a mine is used. It states that a mine is ‘designed to be placed under, on or near the ground or other surface area’. The second element describes how the munition can be detonated. It states that a mine is ‘designed to be … exploded by the presence, proximity or contact of a person or vehicle’. The term mine only appears once in the Convention on Cluster Munitions. Article 1, paragraph 3 states: ‘This Convention does not apply to mines.’327

2.219  Mines are distinguishable from explosive submunitions on both counts. The latter are not ‘placed’; they are ‘dispersed or released by a cluster munition’. Submunitions are also designed to explode ‘prior to, on or after impact’, rather than primarily on contact from a person or vehicle. While unexploded submunitions resemble de facto landmines because people can set them off by disturbing them after strikes, these weapons have failed and are not operating as intended.

2.220  While States clearly expressed their desire to exclude mines from the Convention from the beginning of the Oslo Process, the means of doing so gradually evolved. The Explanatory Annex to the Vienna Discussion Text noted, ‘At the Lima Conference there appeared to be broad agreement that landmines would be excluded from the definition of “cluster munition” since they are already covered by other treaties.’328 The Vienna Discussion Text itself did not include any mention of mines, but several States reiterated the positions expressed in Lima at the Vienna Conference.329

2.221  In its Article 1, paragraph 2, the Draft Cluster Munitions Convention stated that the Convention did not apply to mines as defined in 1996 Amended Protocol II.330 The text combined the scope of application and the definition in a provision outside of Article 2 on definitions.

(p. 238) 2.222  During the Wellington Conference and the Dublin Diplomatic Conference, Indonesia objected to the reference to 1996 Amended Protocol II because it was not a party to that instrument and proposed replacing it with language from the 1997 Anti-Personnel Mine Ban Convention.331 The final Convention split the provision on scope of application and definition, adopted the previously accepted 1997 Anti-Personnel Mine Ban Convention wording for the latter, and placed it in Article 2.

Paragraph 13

‘Explosive bomblet’ means a conventional munition, weighing less than 20 kilograms, which is not self-propelled and which, in order to perform its task, is dispersed or released by a dispenser, and is designed to function by detonating an explosive charge prior to, on or after impact;

Commentary

2.223  The definition of explosive bomblet captures weapons that resemble explosive submunitions but are delivered in a different way. The Convention uses similar language to describe the two weapons and their means of operating. Both are ‘conventional munitions’ and are ‘designed to function by detonating an explosive charge prior to, on or after impact’.332 The inclusion of the phrase ‘designed to’ is important because it ensures a munition that otherwise fits the definition is still an explosive submunition or bomblet even if it fails to detonate as intended. Although explosive submunitions do not by definition have a weight requirement, the definition of cluster munition refers to them as ‘weighing less than 20 kilograms’, and the definition of explosive bomblet adopts the same specification.333

2.224  Submunitions and bomblets are distinguishable because the latter are ‘dispersed or released by a dispenser’, which according to Article 2, (p. 239) paragraph 14, is a container affixed to an aircraft.334 Submunitions, by contrast, are dispersed or released by cluster munitions, typically bombs, artillery projectiles, rockets, or missiles.335 While no other treaty has defined explosive bomblet, in common usage the term had previously meant an air-dropped submunition, and sometimes any kind of submunition.336 In the context of the Convention on Cluster Munitions, it still refers to a weapon released in the air but has a much narrower meaning.

2.225  The term explosive bomblet appears in Articles 1 and 2. Article 1, paragraph 2 states, ‘Paragraph 1 of this Article applies, mutatis mutandis, to explosive bomblets that are specifically designed to be dispersed or released from dispensers affixed to aircraft.’337 It means that the core prohibitions of the Convention apply to explosive bomblets. Given that, as described above, explosive submunitions and explosive bomblets are designed to explode in the same way, they are prone to the same malfunctions and can cause comparable humanitarian harm. They also have similar area effects. Therefore, the Convention bans them both.338

2.226  The only other use of explosive bomblet is in Article 2. In that Article, the definitions of dispenser, which is ‘designed to disperse or release explosive bomblets’, and unexploded bomblet, which is a failed explosive bomblet, also use the term.339 The failure to mention explosive bomblets in other Articles of the Convention suggests that some obligations, such as (p. 240) those related to victim assistance and transparency, may not take them into account. The Commentary discusses the implications of this in greater depth elsewhere.340

2.227  Most of the discussions surrounding explosive bomblets occurred late in the Oslo Process.341 At the Vienna Conference, the GICHD warned that, if left unaddressed, munitions released from a dispenser might become a potential loophole,342 but States did not take up the issue in detail until the Diplomatic Conference. There they considered an Irish proposal, published first in the Wellington Compendium and resubmitted in Dublin, which defined explosive bomblet as ‘a munition, which in order to perform its task is dispersed or separated from a dispenser, affixed to an aerial platform, and is designed to function by detonating an explosive charge prior to, on or after impact’.343 An accompanying proposal for Article 1 stated that dispensers with explosive bomblets were ‘subject to the same provisions as cluster munitions’.344

2.228  The final version of the definition of explosive bomblet generally resembled the Irish proposal, but it made a few amendments. To parallel more closely the definition of explosive submunition, it modified munition with ‘conventional’, replaced ‘separated from’ with ‘released by’, and inserted a weight requirement. It also removed the phrase ‘affixed to an aerial platform’ because dispenser had been defined in another paragraph. Finally it clarified that an explosive bomblet is ‘not self-propelled’. Ireland’s related proposal for Article 1, paragraph 2 was also modified in the text of the Convention to the language discussed above.

Paragraph 14

‘Dispenser’ means a container that is designed to disperse or release explosive bomblets and which is affixed to an aircraft at the time of dispersal or release;

(p. 241) Commentary

2.229  The definition of dispenser clarifies the nature of an explosive bomblet’s delivery system. The first part of the definition explains the form and purpose of a dispenser. It is ‘a container that is designed to disperse or release explosive bomblets’.345 The phrase uses the words ‘is designed to’ because such a container is still a dispenser even if it fails to function. The second half of the definition describes the key feature that distinguishes a dispenser in the context of the Convention, namely it is ‘affixed to an aircraft’. Since it needs only to be affixed ‘at the time of dispersal or release’, later loss or jettisoning does not change its status.346

2.230  While a dispenser disperses or releases smaller munitions like a cluster munition does, it differs in three significant ways. It is considered a container not a munition, it carries explosive bomblets rather than explosive submunitions, and it is an attachment to an aircraft instead of an independent unit.347

2.231  Articles 1 and 2 both have references to dispensers. Article 1, paragraph 2 states, ‘Paragraph 1 of this Article applies, mutatis mutandis, to explosive bomblets that are specifically designed to be dispersed or released from dispensers affixed to aircraft.’348 The use of the term in that provision reinforces the primary distinction between explosive bomblets and explosive submunitions, i.e., their delivery system, but it is redundant. Explosive bomblets, as defined in Article 2, paragraph 13, are always carried in dispensers, and a dispenser is always affixed to an aircraft.

2.232  The term also appears in the definitions of explosive bomblet in Article 2, which is a munition that is ‘dispersed or released by a dispenser’, and unexploded bomblet, an explosive bomblet that has been delivered by a dispenser and failed.349 The word dispenser is unnecessary in the latter definition because an unexploded bomblet is a type of explosive bomblet and all explosive bomblets are dispersed from dispensers; however, its inclusion does not undermine the definition in any way.

(p. 242) 2.233  There were several proposed uses and definitions of the term dispenser during the Oslo Process. In its definition of cluster munition, the Lima Discussion Text referred to a ‘container or dispenser’ from which explosive submunitions separated. It did not, however, define dispenser so it is unclear if it was intended to mean a unit attached to an aircraft.350 The term did not appear in the Vienna Discussion Text, but at the Vienna Conference, as noted above, GICHD warned that munitions released from a dispenser might represent a potential loophole.351

2.234  The Wellington Compendium included three potential definitions of dispenser for Article 2. France proposed defining carrier-container, a term it had inserted in the definition of cluster munition, as:

a)  a conventional munition that may be [an] artillery shell, air bomb, guided or unguided missile or,

b)  a dispenser, affixed to an aircraft, which is not designed to dispense direct-fire munitions.352

Switzerland supported the French proposal as one option and proposed as a second: ‘a dispenser, affixed to an aircraft, which is not designed to dispense multiple sub-munitions in a single act’.353 In the same document, Ireland did not define dispenser separately but described it as being ‘affixed to an aerial platform’ in its definition of explosive bomblet.354

2.235  At the Diplomatic Conference, States considered these formulations. Spain put forward a new proposal for Article 2, which contained France’s definition of dispenser.355 Ireland resubmitted its Article 2 proposal and supplemented it with a proposal for Article 1, paragraph 2 that described but did not officially define the term:

Dispensers, affixed to an aerial platform and designed to disperse or release explosive bomblets, are subject to the same provisions as cluster munitions.356

2.236  Ultimately, the Convention adopted a definition of dispenser that is separate from that of cluster munition. It highlights the shared characteristic of these proposals (it is ‘affixed to an aircraft’) while drawing on the Irish idea that it must be ‘designed to disperse or release explosive bomblets’. It also clarified that a dispenser is a container and need only be affixed ‘at the time of dispersal or release’. The Convention changed the focus of Article 1, paragraph 2 to explosive bomblets because it is they, not the inert dispenser, which pose a threat to civilians. As mentioned above, however, it retained a mention of dispenser in that provision.357

Paragraph 15

‘Unexploded bomblet’ means an explosive bomblet that has been dispersed, released or otherwise separated from a dispenser and has failed to explode as intended.

Commentary

2.237  As discussed above,358 a definition of explosive bomblets was added to Article 2 at a late stage of the negotiations. Explosive bomblets differ from explosive submunitions in that they are delivered directly from a dispenser attached to an aircraft as opposed to being dispersed or released from a bomb, shell, rocket, or missile that has been dropped or fired. Despite the different mode of delivery, explosive bomblets raise the same concerns as explosive submunitions from a technical and humanitarian point of view.

(p. 244) 2.238  The concept of unexploded bomblet was added to Article 2 to ensure that the Convention applied to explosive bomblets when they fail to explode as intended and thus become a danger to those living in conflict-affected areas. Like failed cluster munitions, abandoned cluster munitions, and unexploded submunitions, unexploded bomblets are a constituent element of cluster munition remnants and are therefore covered by the clearance obligations of Article 4. They are also covered by clearance-related provisions in Articles 6 and 7.359

2.239  The definition of unexploded bomblet has direct links to several other Article 2 definitions, namely unexploded submunition, explosive bomblet, and dispenser. As a result, it shares a number of terms and elements and has few new or specific features. For example, like an unexploded submunition, an unexploded bomblet must have ‘failed to explode as intended’.360 In addition, it must meet the criteria for an explosive bomblet that has ‘separated from a dispenser’.361

2.240  Unexploded bomblets were not specifically discussed in the early stages of the Convention’s development. There was no definition of the term in the Lima Discussion Text, the Vienna Discussion Text, or the Draft Cluster Munitions Convention. Later in the Oslo Process, it was generally assumed that these bomblets would be covered by the definitions of explosive submunition and unexploded submunition due to the fact that several proposals for the definition of a cluster munition included the phrase carrier/container,362 which would have covered direct dispensing systems. As negotiations advanced on the definition of cluster munition and the concept of carrier/container fell aside, however, it became clear that explosive bomblets and unexploded bomblets would need distinct definitions.363

2.241  A definition for unexploded explosive bomblet was first proposed by Ireland in its submission to the Wellington Compendium. It suggested the following wording:

‘Unexploded explosive bomblet’ means an explosive bomblet which has been released dispersed or otherwise separated from a dispenser, affixed to an aerial platform, and has failed to explode as intended.364

(p. 245) Ireland formally resubmitted the proposal at the Diplomatic Conference.

2.242  The main changes to the Irish definition of unexploded bomblet in Dublin were the deletion of the phrase ‘affixed to an aerial platform’ and the removal of ‘explosive’ from the definition’s heading. The phrase ‘affixed to an aerial platform’ was deemed unnecessary in light of the inclusion of a definition for ‘dispenser’. The removal of ‘explosive’ from the heading was mainly editorial in light of the definition’s direct link to the definition of ‘explosive bomblet’. With these changes the definition was included in the President’s Text and subsequently adopted.

Footnotes:

1  Markus Reiterer wrote the commentary on paragraph 1, Richard Moyes wrote the commentary on paragraphs 2–3, Lou Maresca wrote the commentary on paragraphs 4–7 and 15, and Bonnie Docherty wrote the overview, preparatory discussions, and negotiating history of the ‘other’ definitions as well as the commentary on paragraphs 8–14.

2  This implies that they may not have a wider impact on international law, including the use of these terms in other treaties that may govern cluster munitions or even other weapons.

3  The formal title of this treaty is the Convention on the Prohibition of the Use, Stockpiling, Production and Transfer of Anti-Personnel Mines and on Their Destruction.

4  Although the definition of transfer in the 1997 Anti-Personnel Mine Ban Convention was itself taken from 1996 Amended Protocol II to the 1980 Convention on Certain Conventional Weapons.

5  The formal title of this instrument is the Protocol on Prohibitions or Restrictions on the Use of Mines, Booby-Traps and Other Devices as amended on 3 May 1996. The definitions of self-destruction mechanism and self-deactivating are taken from the Protocol.

6  2003 Protocol V. The definition of abandoned cluster munition is adapted from the definition of abandoned explosive ordnance in Article 2, paragraph 3 of the Protocol.

7  The Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons Which May Be Deemed to Be Excessively Injurious or to Have Indiscriminate Effects, as amended on 21 December 2001 (CCW).

8  The definition of ‘cluster munition victims’ was generally discussed as an integral part of the victim assistance package which encompassed Article 5 itself, the relevant parts of the preamble, as well as other pertinent parts of the Convention text. In the course of the conferences held in Lima, Belgrade, Brussels, and Vienna the call for the Convention to include a definition of the term was made regularly with the result that the Draft Cluster Munitions Convention presented to the Wellington Conference defined, for the first time, those covered by cluster munition victims. Some elements of the definition, however, had been introduced into the Oslo Process at a fairly early stage.

9  From the Lima Discussion Text, through the Vienna Discussion Text of November 2007, the February 2008 Wellington Conference, and the Dublin Diplomatic Conference, drafts of Article 1 presented prohibitions on the use, production, stockpiling, and transfer of ‘cluster munitions’ as a whole category. Discussions regarding the definition of ‘cluster munition’ therefore provided the primary mechanism for determining how wide-ranging or narrow these prohibitions would be in practice (i.e. what specific existing weapons and possible future weapons would be subject to the prohibitions in Article 1).

10  The ‘other’ definitions had implications for Article 1 (transfer, mine, dispenser, and explosive bomblet), the main definition of cluster munition in Article 2 (‘self-destruction mechanism’ and ‘self-deactivating’), and Article 4 regarding clearance of cluster munition remnants. Many of the drafts for these definitions drew initially on those contained in 1996 Amended Protocol II, the 1997 Anti-Personnel Mine Ban Convention, and 2003 Protocol V.

11  For a brief discussion of this provision see, infra, §§2.95, 2.104, and 2.117–2.118.

12  See, infra, the commentary on Article 5.

13  See, supra, the Introduction to this Commentary for a description of these conferences.

14  See Oslo Declaration, paragraph 1(ii).

15  See Vienna Discussion Text, first preambular paragraph.

16  Summary on the Victim Assistance discussion, European Regional Conference on Cluster Munitions Brussels, 31 October 2007, Rapporteur: Markus Reiterer, Austria, available at: <http://www.diplomatie.be/en/pdf/reiter.pdf> (accessed 8 March 2010).

17  Closing Remarks by Ambassador Wolfgang Petritsch at the end of the Vienna Conference. During the Conference, Spain—among others—stated that there was a need to define who is entitled to the rights outlined in the Article on victim assistance and the preamble in order to facilitate the provision of international cooperation and assistance and to provide a sense of direction for assistance. See Harrison, K., ‘Report of the Vienna Conference on Cluster Munitions, 5–7 December 2007’, Women’s International League for Peace and Freedom, January 2008, <http://www.wilpf.int.ch/disarmament/clustermunitions/ViennaConference/viennareport.htm> (accessed 8 March 2010).

18  The decision on the draft text and its placement of the definition was made during a meeting of the core group States in early January 2008 following an informal proposal by Austria.

19  See Nairobi Summit on a Mine-free World, Review of the Operation and Status of the Convention on the Prohibition of the Use, Stockpiling and Production and Transfer of Anti-Personnel Mines and on their Destruction: 1999–2004, contained in doc. APLC/CONF/2004/5, paragraph 64.

20  The Guidelines were adopted by the UN General Assembly in its Resolution 60/147 of 16 December 2005.

21  Basic Principles and Guidelines on the Right to a Remedy and Reparation for Victims of Gross Violations of International Human Rights Law and Serious Violations of International Humanitarian Law.

22  See Statement on Victim Assistance by Ambassador Beat Nobs, Wellington Conference, 20 February 2008, available at: <http://www.mfat.govt.nz/clustermunitionswellington/conference-documents/statements.php> (accessed 25 March 2010). Cf. also ‘Proposal by Switzerland for the amendment of Article 2’, Diplomatic Conference doc. CCM/21, 19 May 2008.

23  See Statement of the UK, Wellington Conference, 20 February 2008, available at ibid.

24  See Intervención de la delegación de Guatemala, Articulo 5: ‘Asistencia a las Victimas’, 20 February 2008, available at ibid.

25  See Statement of Croatia, available at ibid.

26  See, supra, §2.11.

27  See Wellington Compendium (included in Annex 6 to this Commentary), p. 13. The full text of the definition of cluster munition victims as proposed by the UK read as follows:

‘Cluster Munition Victims’ means any persons who have suffered physical or psychological injury or economic loss, caused by the use of cluster munitions; cluster munition victims include such persons directly impacted by cluster munitions.

28  Suggestion made orally by the UK delegate during informal consultations on victim assistance on 21 May 2008. The UK’s suggestion was supported, among others, by Madagascar.

29  Statement of Laos during informal consultations on victim assistance on 21 May 2008.

30  Statement of Norway during informal consultations on victim assistance on 21 May 2008, which was then supported by among others, Belgium, Chile, Croatia, Guatemala, New Zealand, and the CMC.

31  Statement of Croatia during informal consultations on victim assistance on 21 May 2008.

32  See ‘Summary Record of Second Session of the Committee of the Whole’, Diplomatic Conference doc. CCM/CW/SR/2, 18 June 2008.

33  Observations by the International Federation of Red Cross and Red Crescent Societies and the UN Mine Action Service during informal consultations on victim assistance on 21 May 2008.

34  See ‘Proposal by Philippines for the amendment of Article 2’, Diplomatic Conference doc. CCM/57, 19 May 2008. The proposal read as follows:

‘Cluster munition victims’ means persons who have suffered death, physical or psychological injury, economic loss, social marginalization or substantial impairment of the realization of their rights caused by the use of cluster munitions. They include those persons directly impacted by cluster munitions as well as their families and communities and also migrants under the jurisdiction and control of an affected State. (original emphasis.)

35  Remarks by the Friend of the President during informal consultations on victim assistance on 21 May 2008.

36  This view was also supported by the Office of the High Commissioner for Refugees (UNHCR). Statement of UNHCR during informal consultations on victim assistance on 22 May 2008.

37  See ‘Summary Record of Eighth Session of the Committee of the Whole’, Diplomatic Conference doc. CCM/CW/SR/8, 18 June 2008 (cf. discussion on Article 5).

38  i.e. the affected individual, family, and community.

39  Article 1, UN Convention on the Rights of Persons with Disabilities.

40  See, infra, the commentary on Article 2, paragraph 3 for detailed commentary on the definition of explosive submunition.

41  This means that the production, stockpiling, transfer, and use of explosive bomblets is prohibited even though they do not fall within the definition of a cluster munition set out in Article 2, paragraph 2.

42  e.g., all munitions listed on the Human Rights Watch chart ‘Overview of a Dirty Dozen Cluster Munitions’ are prohibited under the Convention (<http://www.hrw.org/sites/default/files/related_material/munitionChart0806.pdf>, accessed 8 March 2010). However, there have also been criticisms that the definition allows some types of weapons with submunitions to continue to be used and that the definition is based in part on ‘numbers’ that must to some extent be arbitrary. See, e.g., Statements of Chile and Spain, ‘Summary Record of Eleventh Session of the Committee of the Whole’, Diplomatic Conference doc. CCM/CW/SR/11, 18 June 2008.

43  It might, however, be noted that the general concept of a cluster munition, as a weapon with multiple submunitions, is based on the numerical distinction between ‘one’ and ‘more than one’ as much as on the subsidiary relationship of the submunitions to the container.

44 Summary Record of Eleventh Session of the Committee of the Whole’, Diplomatic Conference doc. CCM/CW/SR/11, 18 June 2008. Austria ‘considered that effects-based language should be adopted. This should be complemented by a reporting requirement on new weapons and their effects in Article 7.’ Lebanon stated that Article 2, paragraph 2(c) ‘must comply with the highest humanitarian standards, and contain safeguards for the review of exemptions at Review Conferences.’

45  See ‘Proposal by France for the amendment of Article 2’, Diplomatic Conference doc. CCM/20, 19 May 2008. As noted infra in §2.72, the French proposal made reference to a review mechanism that would allow the parameters of the definition to be considered again by States Parties ‘no later than 5 years after entry into force’ in line with the treaty’s provisions regarding Review Conferences.

46  Oslo Conference on Cluster Munitions, 22 February 2007 (Landmine Action notes; copy on file with author).

47  Declaration of the Oslo Conference on Cluster Munitions of 23 February 2007. The wording appeared to favour a more selective approach to the prohibition of weapons that might be considered cluster munitions.

48  See ‘Overview of Existing and Proposed Definitions, Submitted by the Geneva International Centre for Humanitarian Demining (GICHD)’, UN doc. CCW/GGE/2007/WP.5, Geneva, 12 June 2007.

49  Statements made at the Lima Conference, 24 May 2007 (Landmine Action notes; copy on file with author).

50  Indeed, on 19 February 2008 during the Wellington Conference, Zambia stated:

If any form of harm is acceptable, it would be interesting to see those of us advocating for a partial ban to volunteer themselves as examples for all of us to see just how acceptable harm is.

Available at: <http://www.mfat.govt.nz/clustermunitionswellington/conference-documents/Zambia-definitions.pdf>.

51  Landmine Action notes; copy on file with author.

52  King, C., Dullum, O., and Østern, G., M85—An Analysis of Reliability (Oslo: Norwegian People’s Aid, 2007), see also §2.56.

53  From the Vienna Discussion Text onwards, a cluster munition was deemed to comprise both the container and its submunitions.

54  Statements on definitions of the CMC and the ICRC, Lima Conference, 24 May 2007 (Landmine Action notes; copy on file with author).

55  Introduction by Ambassador Don MacKay, Lima Conference, 24 May 2007 (Landmine Action notes; copy on file with author).

56  Landmine Action notes; copy on file with author.

57  Rappert, B., and Moyes, R., ‘The Prohibition of Cluster Munitions: Setting International Precedents for Defining Inhumanity’, Nonproliferation Review, Vol. 16.2, July 2009.

58  The Vienna Discussion Text of 14 November 2007 is included in Annex 4 to this Commentary.

59  This implication of the Vienna Discussion Text definition was raised by the GICHD during the Vienna Conference on 6 December 2007 (Landmine Action notes; copy on file with author).

60  See UNMAS, UNDP and UNICEF, ‘Proposed definitions for cluster munitions and submunitions’, UN doc. CCW/GGE/X/WG.1/WP.3, 8 March 2005.

61  NATO Glossary of Terms and Definitions, AAP-6 (2009), 2-S-14, available at: <http://www.nato.int/docu/stanag/aap006/aap6.htm> (accessed 8 March 2010).

62  IMAS 04.10: ‘Glossary of mine action terms, definitions and abbreviations’, Second Edition, 1 January 2003, available at <http://www.mineactionstandards.org> (accessed 8 March 2010).

63  Statement by Ambassador John Duncan on Cluster Munitions, Third Review Conference of the CCW, Geneva, 13 November 2006.

64  Landmine Action notes; copy on file with author.

65  Ibid.

66  Ibid.

67  Draft CCW Protocol on Cluster Munitions submitted by Germany, 2007 Session of the CCW Group of Governmental Experts on Cluster Munitions, UN doc. CCW/GGE/2007/WP.1, Geneva, 1 May 2007.

68  Harrison, K., ‘Report of the Vienna Conference on Cluster Munitions, 5–7 December 2007’, Women’s International League for Peace and Freedom, January 2008, <http://www.wilpf.int.ch/disarmament/clustermunitions/ViennaConference/viennareport.htm> (accessed 8 March 2010).

69  King, C., Dullum, O., and Østern, G., M85: An Analysis of Reliability, op. cit.

70  Dual-purpose improved conventional munitions. See, e.g., GICHD, A Guide to Cluster Munitions, Second Edition, Geneva, 2009, pp. 10–11, 13, 17.

71  See the use of the term ‘dangerous duds’ in the proposal, ‘Draft CCW Protocol on Cluster Munitions submitted by Germany’, 2007 Session of the CCW Group of Governmental Experts on Cluster Munitions, CCW/GGE/2007/WP.1, Geneva, 1 May 2007.

72  Draft Cluster Munitions Convention of 21 January 2008.

73  On 6 December 2007, during the Vienna Conference, the CMC had questioned whether remotely delivered anti-vehicle mines dispersed from a container should be excluded from definition as cluster munitions and asked delegations to consider whether these weapons might not cause unacceptable harm to civilians.

74  Introduction by Ambassador Don MacKay, Wellington Conference, 19 February 2008 (Landmine Action notes; copy on file with author).

75  See, e.g., the entries ‘GIWS DM 702 SMArt 155 ammunition system’ and ‘Bofors/Giat Industries 155mm Bonus sensor-fuzed munition’ in Ness, L. and Williams, A. G. (eds.), Jane’s Ammunition Handbook 2006–2007 (UK: Jane’s Information Group, November 2006), pp. 657–658, and 649–650; see also Rheinmetall Defence press releases on SMArt 155 (01/07/2008), <http://www.rheinmetall-defence.com/index.php?fid=4504&qid=&qpage=0&lang=3&query=SMArt%20155>, and (10/03/2005), <http://www.rheinmetall-defence.com/index.php?lang=3&fid=3241>; and also UK Ministry of Defence, ‘New precision “search and destroy” anti-armour weapon’, News release, 19 November 2007; and Jane’s Information Group, ‘155mm Bofors Trajectory Correctable Munition (Sweden), Field artillery’, <http://www.janes.com/articles/Janes-Ammunition-Handbook/155-mm-Bofors-Trajectory-Correctable-Munition-Sweden.html> (all accessed 8 March 2010).

76  See, e.g., Presentation by Colin King to the Vienna Conference on Cluster Munitions, 6 December 2007; the discussion on definitions at the Wellington Conference, 19 February 2008 (Landmine Action notes; copy on file with author); and US Defense Science Board Task Force on Munitions System Reliability, September 2005, Munitions System Reliability, Office of the Under Secretary of Defense for Acquisition, Technology and Logistics, Washington, DC, 20301–3140.

77  The report M85: An analysis of reliability (op. cit.) provides an analysis of such factors at Section 4.4: ‘Failure Analysis’. While it does not compare the susceptibility of different systems to these factors, the typology it presents would provide a useful framework for analysis.

78  US CBU-97/CBU-105.

79  Statement of Germany, Wellington Conference, 19 February 2008 (Landmine Action notes; copy on file with author).

80  Discussions on definitions at the Wellington Conference, 19 February 2008 (Landmine Action notes; copy on file with author).

81  Landmine Action notes; copy on file with author.

82  Statement of the CMC, Wellington Conference, 19 February 2008 (Landmine Action notes; copy on file with author).

83  Statement of the ICRC, Wellington Conference, 19 February 2008 (Landmine Action notes; copy on file with author).

84  Discussions on definitions at the Wellington Conference, 19 February 2008 (Landmine Action notes; copy on file with author).

85  Became Diplomatic Conference doc. CCM/18.

86  Became Diplomatic Conference doc. CCM/19.

87  Became ‘Proposal by France for the amendment of Article 2’, Diplomatic Conference doc. CCM/20, 19 May 2008.

88  Became ‘Proposal by Switzerland for the amendment of Article 2’, Diplomatic Conference doc. CCM/21, 19 May 2008.

89  Became ‘Proposal by the United Kingdom for the amendment of Article 2’, Diplomatic Conference doc. CCM/23, 19 May 2008.

90  Became ‘Proposal by Sweden for the amendment of Article 2’, Diplomatic Conference doc. CCM/26, 19 May 2008.

91  Became ‘Proposal by Peru for the amendment of Article 2’, Diplomatic Conference doc. CCM/24, 19 May 2008.

92  Became ‘Proposal by Ireland for the amendment of Article 2’, Diplomatic Conference doc. CCM/25, 19 May 2008. Under the Irish proposal:

‘Explosive sub-munition’ means a munition that in order to perform its task separates from a cluster munition and is designed to function by detonating an explosive charge prior to, on or after impact.

93  According to the ICRC paper included in the Wellington Compendium:

‘Explosive sub-munitions’ means munitions that in order to perform their task are dispersed or released from a cluster munition and are designed to function by detonating an explosive charge prior to, on or after impact.

94  Declaration of the Wellington Conference on Cluster Munitions, 22 February 2008.

95  As these terms were used by the CMC, an ‘exempted’ technology would still be a cluster munition. This echoed debates over whether the treaty should distinguish between prohibited and allowed cluster munitions. By contrast, an ‘excluded’ technology was held to be similar to a cluster munition, but also sufficiently different to warrant falling outside the definition of that term. These distinctions are implicit in the CMC document, ‘Dublin Diplomatic Conference on Cluster Munitions 2008—Lobbying Guide’ (London: CMC, 2008).

96  Australia, Canada, France, the Netherlands, Norway, and the UK. See, e.g., ‘Summary Record of First Session of the Committee of the Whole’, Diplomatic Conference doc. CCM/CW/SR/1, 18 June 2008, p. 4.

97  Bulgaria, Denmark, Italy, Japan, South Africa, Sweden, and the UK. Ibid., pp. 4, 5, 6.

98  Indonesia supported by Guatemala, Jamaica, Mexico, and Venezuela. Ibid., p. 5.

9