When Does a Child ‘Participate Actively in Hostilities’ under the Rome Statute? Protecting Children from Use in Hostilities after Lubanga

This paper explores the scope of activities children may engage in for a defendant to be convicted for using them to participate in hostilities under the Statute of the International Criminal Court (ICC). It analyses the relevant international law provisions and the ICC’s decisions in the Lubanga matter. It finds that a broad scope of activities more effectively assists the protection of children from use in hostilities. It also identifies inconsistencies in the relevant international law provisions and proposes a number of factors future ICC Chambers can use to consistently characterise activities in future prosecutions.


I. Introduction
The use of children in warfare is disturbing. Around 300,000 children are currently involved in conflicts worldwide as combatants, messengers, porters, cooks, and otherwise. 1 Children have been recruited in the Central African Republic and Syria, 2 and in some cases have been used to execute prisoners. 3 The United Nations (UN) has attempted to prevent the use of children in hostilities. 4 However, a strong international law response against this practice is still needed. 5 One such response is the release of the International Criminal Court (ICC)'s first ever Appeals Chamber decision on 1 December 2014. 6 The Appeals Chamber upheld the decision of the ICC's Trial Chamber on 14 March 2012 to convict Mr Thomas Lubanga of two war crimes: (a) enlisting and conscripting children

A. What Do 'Active' and 'Direct' Mean at IHL?
1. IHL Generally IHL governs armed conflicts and only operates when an armed conflict exists. 21 It aims to protect the victims of armed conflict and regulate the conduct of hostilities based on the balance between military necessity and humanity. 22 At its heart is the need to uphold the principle of distinction between armed forces who conduct the hostilities on behalf of parties to an armed conflict and civilians who must be protected against the dangers arising from military operation. 23 IHL exists predominantly as treaty law and has traditionally been attributed to two main sources: the Hague Conventions of 1899 or 1907, and the Geneva Conventions of 1949 and the Additional Protocols to these Conventions. 24 Interpreting international treaties is complex. 25 There are traditionally three schools of interpretation. The textual school aims to give effect to the ordinary meaning of a text. The intentionalist school, meanwhile, interprets provisions according to their drafters' intentions. Lastly, the teleological school seeks an interpretation that best fulfils the object and purpose of a treaty. 26 Article 31(1) of the Vienna Conventions on the Law of Treaties (VCLT) appears to incorporate elements of all three schools, 27 but prioritises the ordinary meaning of the text. 28 It reads as follows: A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in light of its object and purpose. 29 There are difficulties in interpreting the relevant provisions according to Article 31(1) of the VCLT. Nevertheless, 'whenever appliers (. . .) interpret a treaty, they should consider (. . .) articles 31-33 [of the VCLT] as a starting point '. 30 When Does a Child 'Participate Actively in Hostilities' under the Rome Statute? Protecting Children from Use in Hostilities after Lubanga 72 In 2009, the International Committee of the Red Cross (ICRC) released an interpretive Guidance on direct participation in hostilities, based on the principle of distinction. 34 The Guidance is not legally binding, nor does it alter customary or treaty IHL. 35 Nevertheless, it provides useful criteria that an act must meet before it is characterised as direct participation in hostilities: 36 1. Threshold of harm: the act must be likely to adversely affect the military operations or military capacity of a party to an armed conflict or, alternatively, to inflict death, injury or destruction on persons or objects protected against direct attack. 2. Direct causation: there must be a direct causal link between the act and the harm likely to result either from that act, or from a coordinated military operation of which that act constitutes an integral part. 3. Belligerent nexus: the act must be specifically designed to directly cause the required threshold of harm in support of a party to the conflict and to the detriment of another. 37 The Guidance distinguishes between ' direct' and 'indirect' participation in hostilities but does not provide further guidance as to what the latter includes. 38 The Israeli Supreme Court also addressed direct participation in hostilities at IHL for the purposes of distinction between combatants and civilians in the Targeted Killings Judgment. 39 The Court assessed whether the State of Israel acted illegally by engaging in preventative strikes against alleged terrorists, strikes which 'at times also harm[ed] innocent civilians'. 40 The Court addressed direct participation in hostilities as grounds for losing the protection against military attack. 41 It interpreted this doctrine broadly, 42 holding that people performing the following actions directly participated in hostilities: collecting intelligence in the army, transporting unlawful combatants to or from the place where the hostilities were taking place, operating weapons used by unlawful combatants, supervising the operation and providing service to unlawful combatants. 43 However, persons selling food or medicine to an unlawful combatant, or aiding unlawful combatants with general strategic analysis or logistical support, would not be directly participating in the hostilities. 44 This judgment shows ' direct' participation can be interpreted at IHL to determine whether civilians are protected. 45 However, future courts must be cautious when applying its reasoning. Firstly, the judgment was handed down before the Guidance was published. Accordingly, it did not use the Guidance's constitutive elements. 46 Secondly, the desire to justify Israel's policy of targeted killings may have influenced an expansive interpretation of 'taking a direct part in hostilities' to ' expand (. . .) the temporal horizon for lawful 34 (2) of AP I and Article 38(2) of the CRC, which read as follows: The Parties to the conflict shall take all feasible measures in order that children who have not attained the age of fifteen years do not take a direct part in hostilities and, in particular, they shall refrain from recruiting them into their armed forces (. . .). 56 State Parties shall take all feasible measures to ensure that persons who have not attained the age of fifteen years do not take a direct part in hostilities. 57 These provisions must be interpreted in good faith in accordance with their ordinary meanings, in their contexts and in light of their objects and purposes. 58 Accordingly, ' direct' in Article 77(2) of AP I could be granted a broader interpretation than afforded by the principle of distinction, as the provision's immediate context reveals that children are 'the object[s] of special respect' and are to be 'protected against any form of indecent assault.' 59 Indeed, the Commentary to AP I notes that 'the intention of the drafters of the article was clearly to keep children under fifteen outside armed conflict'. 60 This indicates the ICRC's intention to apply a broader interpretation. 61 Moreover, the obligation to refrain from recruiting children under the age of fifteen suggests that both direct and indirect participation may be prohibited. 62 However, relevant rules of international law applicable to relations between relevant parties must also be taken into account when interpreting this provision. 63 Article 51(3) of AP I uses ' direct' in the context of principle of distinction, 64 which could be a relevant rule of international law. Accordingly, some argue that a similar, narrow interpretation should apply to ' direct' in Article 77(2) of AP I. 65 The Commentary on AP I itself bases its reasoning that the provision prohibits a broader range of activities on the fact that the drafters did not include ' direct'. 66 However, it does not explore why the term was eventually included.
Therefore, Article 77 (2) of AP I appears to have contradictory purposes. It aims to protect children but restricts the scope of activities parties are forbidden to use children for by including ' direct'. Whether 'indirect' participation was prohibited under this provision was not clear from the provision's text or its Commentary. As Happold notes, however, a narrow interpretation of ' direct' is unlikely to provide children under fifteen with ' effective protection from (. . .) an adverse party'. 67 b. Article 38(2) of the UN Convention on the Rights of the Child The CRC's drafters also included ' direct' in Article 38(2), despite several States indicating support for a blanket ban against the participation of children below the age of fifteen years in hostilities. 68 The CRC's preparatory works indicate that ' direct' in Article 38 (2) is based on the principle of distinction. 69 Accordingly, it appears to ' allow (. . .) children under the age of 15 years to take (. . .) an 'indirect part' in hostilities'. 70 Nevertheless, the preparatory works do not show a clear policy reason for inserting ' direct' into the provision, beyond the fact that consensus could not be reached on it. 71 Ang attempts to explain this decision in her commentary on the provision: Many delegations belonging to the Working Groups drafting this article had expressed to be in favour of the deletion of the word ' direct', but in the end, the adoption of the third paragraph was considered to render the prevention of 15-to 18-year olds taking any part in hostilities while they could be legitimately recruited, unrealistic. 72 The third paragraph prevents the recruitment of children under the age of 15 into the armed forces and also requires States to give priority to the oldest children between 15 and 18 years of age when recruiting them. 73 However, it is unclear how the unrealistic nature of preventing 15 to 18 year olds taking any part in hostilities necessitates the removal of a blanket ban on the participation of children under 15 years of age.
Moreover, it was felt that the provision could undermine IHL because it was inconsistent with the level of protection offered in Article 4(3)(c) of AP II. 74 This issue was not resolved. According to one observer, it was regrettable that paragraph 2 had been adopted in light of such ' extensive opposition.' 75 Another observer noted that a blanket ban would have 'improve[d] the protection of the child in armed conflicts, which was necessary if there was a will to provide special protection for children.' 76 I submit that ' direct' should not have been included in this provision because it allows children to be used to indirectly participate in hostilities.

c. Article 4(3)(c) of Additional Protocol II
Had Article 38(2) of the CRC not included ' direct', it would have mirrored Article 4(3)(c) of AP II, which also prohibits the use of children in military operations and reads as follows: 77 Children shall be provided with the care and aid they require, and in particular (. . .) children who have not attained the age of fifteen years shall neither be recruited in the armed forces or groups nor allowed to take part in hostilities. 78 AP II addresses non-international armed conflict and AP I addresses international armed conflict. Accordingly, there appears to be a broader range of activities in which children are not to participate in non-international armed conflict than in international armed conflict, 79 as Article 4(3)(c) imposes 'a near-absolute prohibition' and a 'blanket ban' on the participation of children in hostilities. 80 However, the degree of this difference is unclear given the lack of certainty of the scope of activities covered by 'direct' in Article 77(2) of AP I. 81 Schabas dismisses this as a drafting inconsistency rather than a normative dispute. 82 However, I show in Part II(B) that this inconsistency has significant effects on the interpretation of similar provisions in the Statute.

Conclusion
Traditionally, ' active' and ' direct' were interpreted synonymously at IHL according to the principle of distinction. Commentators recognise that persons will not lose their status as protected civilians if they 'indirectly' participate in hostilities.
However, the scope of activities constituting ' direct' participation in Article 77(2) of AP I is unclear, as the provision aims to prevent children from participating in hostilities both directly and indirectly. Article 38(2) of the CRC, which also uses ' direct' to qualify participation in hostilities, has a similar purpose, despite its apparent allowance for children to indirectly participate in hostilities. Therefore, there is a tension between a broad interpretation of the phrase and the more traditional, narrow interpretation. Article 4(3)(c) of AP II, moreover, comprehensively prohibits all use of children under fifteen years of age in non-international armed conflict. This differs from the prohibition on ' direct' participation in Article 77(2) of AP I and Article 38(2) of the CRC.

B. What Does 'Active' Mean in the Statute's Child Protection Provisions?
1. The relationship between ICL, IHL and the Statute ICL aims to deter and prohibit certain categories of conduct and impose criminal liability on individuals in retribution for such conduct. 83 It serves to create a 'universal legal consciousness' in relation to criminal acts at international level. 84 It is drawn from primary sources (e.g. treaties like the Statute and customary international law), secondary sources (e.g. Security Council Resolutions establishing an international criminal tribunal), and general principles of law. 85 Article 21(1) of the Statute codifies the order in which these sources may be used. 86 ICL also draws from IHL and IHRL. 87 The Statute criminalises war crimes, 88 which are serious violations of IHL that entail the individual criminal responsibility of the individuals breaching the rules under either customary international law or international treaty law. 89 When determining whether certain violations of IHL entailed individual criminal responsibility, the International Criminal Tribunal for the Former Yugoslavia (ICTY) upheld the International Military Tribunal's use of the following criteria: The clear and unequivocal recognition of the rules of warfare in international law and State practice indicating an intention to criminalise the prohibition, including statements by government officials and international organisations, as well as punishments of violations by national courts and military tribunals. 90 While AP II regulates conduct in non-international armed conflicts at IHL, it does not punish war crimes committed in such conflicts. 91 Accordingly, it was widely accepted prior to the 1990s that there was no international criminal responsibility for war crimes committed during internal armed conflicts. 92 In the 1990s, however, the Statute of the International Criminal Tribunal for Rwanda 93 and the jurisprudence of the ICTY 94 extended international criminal responsibility to these crimes. Eventually, Article 8 prohibited war crimes both in international and non-international armed conflict. 95

The Statute's Child Protection Provisions
The Statute's prohibitions against the use of children in hostilities read as follows: Conscripting or enlisting children under the age of fifteen years into armed forces or groups or using them to participate actively in hostilities [in a conflict of a non-international character]. 96 Conscripting or enlisting children under the age of fifteen years into the national armed forces or using them to participate actively in hostilities [in a conflict of an international character]. 97 a. Introduction to the Provisions Both of these provisions are based on the IHL provisions described above. 98 They provide ' compulsorily penal sanctions' for using children in hostilities, which neither the APs nor the CRC provide. 99 Moreover, 85 Cassesse, International Criminal Law (n 24) 14. 86 ibid 15. Article 21(1) reads as follows: The Court shall apply (a) In the first place, this Statute, Elements of Crimes and its Rules of Procedure and Evidence; (b) In the second place, where appropriate, applicable treaties and the principles and rules of international law, including the established principles of the international law of armed conflict; (c) Failing that, general principles of law derived by the Court from national laws of legal systems of the world including, as appropriate, the national laws of States that would normally exercise jurisdiction over the crime, provided that those principles are not inconsistent with this Statute and with international law and internationally recognised norms and standards. 87 Cassesse, International Criminal Law (n 24) 6. 88  the different types of conflicts (non-international or international) these provisions address do not mean that different levels of participation are prohibited 100 or that there are any differences in the constituent elements of the crimes. 101 Accordingly, under the Statute, children in non-international armed conflict enjoy a degree of protection 'largely comparable' to that experienced by individuals in international armed conflict. 102 This contrasts with the difference in the requisite level of participation between Article 77(2) of AP I and Article 4(3)(c) of AP II. 103 Article 8(2)(e)(vii) was derived from Article 4(3)(c) of AP II. 104 However, the former's use of the word ' actively' indicates a higher level of participation that is required for activities to fall within its scope, as opposed to the blanket prohibition on participation in the latter. 105 Accordingly, parties in noninternational armed conflicts could use children to participate in hostilities in a way that is prohibited by the blanket ban in Article 4(3)(c) of AP II but not by the prohibition of ' active' participation in Article 8(2)(e)(vii).
In light of this, ' active' should be removed from the Statute's child protection provisions. This would ensure that the widest range of child use in hostilities is punished under the Statute. For further consistency, Article 77(2) of AP I should be amended to remove the word ' direct', creating a consistent blanket ban on the use of children in hostilities at IHL and ICL for both international and non-international armed conflict. However, amending either the Statute or AP I would be a fairly complex process. 106 In light of such difficulties, I focus on how the Statute's child protection provisions have been and should be interpreted to protect children from use in hostilities.

b. Interpreting 'Active' and 'Direct' in the Provisions
There is some uncertainty as to how the Statute should be interpreted. 107 The Statute, as a treaty, is subject to the VCLT. 108 Accordingly, the Statute's provisions must be interpreted in good faith in accordance with their ordinary meanings, in their contexts and in the light of their objects and purposes. 109 However, Article 22(2) of the Statute adopts a stricter approach to interpretation. 110 How these approaches should be balanced is unclear. 111 However, 'the judges of the Court will have to resolve this [tension] without any substantial assistance from the Statute.' 112 Moreover, the exact scope of ' active' in the Statute's child protection provisions was unclear prior to the Lubanga Trial Decision. The Elements of Crimes of both Articles 8(2)(e)(vii) and 8(2)(b)(xxxvi) do not provide relevant guidance. 113  that 'the perpetrator conscripted or enlisted one or more persons into an armed force or group or used one or more persons to participate actively in hostilities.' The remaining elements deal with the age of the person (element 2), the mens rea of the perpetrator (element 3), the context of the conflict (element 4) and the perpetrator's awareness of the factual circumstances establishing the existence of the conflict (element 5). 114 Ang (n 54) 40. However, the Special Court for Sierra Leone (SCSL) provided relevant commentary in the AFRC judgment. 115 It concluded that active participation in hostilities in Article 4(c) of the SCSL Statute -which is extremely similar to the child protection provisions in the Rome Statute 116 -was 'not limited to participation in combat'. 117 Rather, it included ' any labour or support that gives effect to, or helps maintain, operations in a conflict (. . .) [like] carrying loads for the fighting faction, finding and/or acquiring food, ammunition or equipment, acting as decoys, carrying messages, making trails or finding routes, manning checkpoints or acting as human shields.' 118 The SCSL also concluded that ' an armed force requires logistical support to maintain its operations (. . .) any labour or support that gives effect to, or helps maintain, operations in a conflict constitutes active participation.' 119 These interpretations both exceeded the usual scope of active/ direct participation in hostilities according to the principle of distinction. 120 It remained to be seen whether the ICC in Lubanga would adopt a similar approach.

The Lubanga Trial Decision
The Trial Chamber found that the relevant conflict in Lubanga was non-international in nature. Accordingly, it considered charges against Mr Lubanga under Article 8(2)(e)(vii) of the Statute. 121 The Trial Chamber acknowledged the need to strictly construe the definitions of crimes under Article 22(2) of the Statute. 122 However, it then stated that the interpretation of the Statute was governed by the VCLT. 123 It did not comment further on how the two approaches related to one another.
The Trial Chamber juxtaposed 'participate actively in hostilities' in Article 8(2)(e)(vii) of the Statute with ' direct participation' in Article 77(2) of AP I. 124 According to the Trial Chamber, to participate actively in hostilities 'was clearly intended to import a wide interpretation to the activities and roles that are covered by the offence of using children under the age of 15 actively to participate in hostilities.' 125 Therefore, it found that children were used to participate actively in hostilities by the UPC/FPLC: 126 they had participated in combat, 127 worked as bodyguards and escorts of UPC/FPLC main staff and commanders, 128 and had been part of a special unit of approximately 45 child soldiers. 129 Girls were also assigned domestic household tasks like cooking in addition to their combat, patrol and bodyguard duties. 130

a. Criticism of the Lubanga Trial Chamber's Delineation between 'Active' and 'Direct' Participation
The Trial Chamber's decision was met with acclaim as the ICC's first guilty verdict. 131 However, it was criticised for delineating between ' active' and ' direct' participation in hostilities. 115 The Prosecutor v Brima, Kamara

i. The Linguistic Inconsistency of the Delineation
Firstly, some argue that ' active' in the Statute's child protection provisions should be interpreted synonymously with ' direct' at IHL because the words are viewed synonymously at IHL. 132 Two sub-arguments are used to support this notion: (a) the consistent use of the same phrase in the equally authentic French versions of Common Article 3 and other IHL instruments; and (b) the fact that Common Article 3 is codified in the Statute.
In relation to (a), the same French phrase (participent directement) is used in the equally authentic French versions of the Geneva Conventions and the Additional Protocols, despite the use of ' active' and ' direct' in the English versions. Because ' active' and ' direct' are interpreted primarily according to the principle of distinction at IHL, the argument is that ' active' in the Statute must be interpreted narrowly as well. 133 However, 'participent directement' is not used in the French version of the Statute. Rather, Article 8(2)(e) (vii) uses 'participer activement'. 134 If one can argue that the use of the same phrase for different English words in the IHL instruments means there is no difference in meaning between the words, one can also argue that the Statute's drafters intended for a meaning to be applied in the child protection provisions that is different from the meaning attributed to ' direct' at IHL according to the principle of distinction. Therefore, the use of the same phrase in the French IHL provisions does not conclusively demonstrate that the Statute's drafters intended the meaning of ' direct' to be transposed to the meaning of ' active' in the Statute.
In relation to (b), Urban argues that because Common Article 3 is codified in Article 8(2)(c) of the Statute, the interpretation of ' active' according to the principle of distinction must be applied when interpreting ' active' in other parts of the Statute like Article 8(2)(e)(vii). 135 Nevertheless, ' active' in the Statute's child protection provisions should be interpreted more broadly than ' direct' as defined by the principle of distinction, as these provisions aim to protect all children from use in hostilities. 136 Accordingly, the existence of Article 8(2)(c) does not preclude the broad interpretation of ' active' in the child protection provisions according to their purposes.

ii. The Consequences of the Delineation
The Trial Chamber's reasoning was also criticised for making child soldiers more targetable. This is because the scope of persons who may be legitimately targeted by an adverse party with impunity will increase as more people are categorised as participating directly in hostilities. 137 However, the use of different principles to interpret the different IHL and Statute provisions addresses these concerns. The principle of distinction should be used to interpret provisions like Common Article 3, and the purpose of protecting children used to interpret provisions like Article 8(2)(e)(vii). 138 Accordingly, a broader interpretation of ' active' in the Statute's child protection provisions protects the interests of children without affecting the protection of civilians under IHL provisions according to the principle of distinction. It would not involve 'giving with one hand while taking with the other' because the two standards of ' active' are located at different levels. 139 The Statute itself indicates that the interpretation of a provision in the Statute will not limit or prejudice existing or developing rules of international law for other purposes. 140 Therefore, the interpretation of ' active' for the purpose of protecting children is likely not to affect the interpretation of ' active' or ' direct' for the purpose of distinguishing between combatants and civilians.
A narrow interpretation of ' active' in the Statute's child protection provisions could also lead to individuals like Mr Lubanga being acquitted even though the evidence demonstrates that they have used children to participate actively in hostilities. A broad approach is necessary to cover as wide a range of activities as possible within the meaning of ' active', and in doing so remove any justification for using children in hostilities.
iii. The Effect of the Trial Chamber's Approach on the Interpretation of 'Direct' in Article 77(2) of AP I While the Trial Chamber appropriately interpreted ' active' in Article 8(2)(e)(vii) of the Statute, its delineation between ' active' in that provision and ' direct' in Article 77(2) of AP I further adds to the confusion surrounding the interpretation of ' direct' in the latter provision, given that it also aims to protect children from use in hostilities. 141 The Trial Chamber used Article 4(3)(c) of AP II and Article 38(2) of the CRC as support for the broad interpretation, noting that the main objective underlying these provisions was to 'protect children under the age of 15 from the risks (. . .) associated with armed conflict (. . .) [and to] secur[e] (. . .) their physical and psychological wellbeing.' 142 Nevertheless, this reasoning is troublesome. The Trial Chamber implied that the absence of the word ' direct' in Article 4(3)(c) of AP II indicated a broader standard of participation than in Article 77(2) of AP I. 143 However, it also used Article 38(2) of the CRC to support a broader standard of participation, not mentioning that this provision does include the word ' direct' and appears extremely similar in purpose and wording to Article 77(2) of AP I. 144 Accordingly, if Article 38(2) of the CRC was used to support a broad interpretation based on the purpose to protect children, the Trial Chamber should not have made the distinction between the levels of participation in Article 4(3)(c) of AP II and Article 77(2) of AP I.
The Trial Chamber's reasoning would also have affected the interpretation of ' active' in Article 8(2)(b) (xxxvi), which was derived from both Article 77(2) of AP I and Article 38(2) of the CRC: 145 if the latter's qualifier on participation (' direct') was interpreted narrowly, should the equivalent qualifier in the Statute (' active') in Article 8(2)(b)(xxxvi)) be interpreted narrowly as well? The Trial Chamber did not resolve this question, leaving the scope of activities prohibited by Article 8(2)(b)(xxxvi) in international armed conflict unclear. This highlights the need for consistency in the terminology between related IHL and ICL provisions. A mere drafting inconsistency at IHL, 146 while minor, can have extremely significant consequences in the application of ICL.

The Lubanga Appeals Decision
The Appeals Chamber affirmed Mr Lubanga's conviction more than two years after the Trial Decision. 147 In his appeal, Mr Lubanga argued that the Trial Chamber misinterpreted the scope of the crime in Article 8(2) (e)(vii) and that 'use to participate actively' should be interpreted according to the principle of distinction undergirding Common Article 3 because there was no difference between ' active' and ' direct' participation at IHL. 148 However, the Appeals Chamber disagreed with Mr Lubanga, finding that: (. . .) the interpretation given to Common Article 3 (. . .) in the context of the principle of distinction cannot simply be transposed to that of article 8(2)(e)(vii) of the Statute. Rather, the term 'participate actively in hostilities' must be given an interpretation that bears in mind that provision's purpose. 149 The Appeals Chamber also commented that Article 77(2) of AP I should be interpreted based on its purpose to protect children, rather than the principle of distinction between combatants and civilians. 150 Accordingly, 140 Art 10 of the Rome Statute reads as follows: '[n]othing in this Part shall be interpreted as limited or prejudicing in any way existing or developing rules of international law for purposes other than this Statute.' See also Graf (n 136)  I submit that ' direct' in both Article 77(2) of AP I and Article 38(2) of the CRC should be interpreted broadly according to the purpose to protect children from use in hostilities, without reference to the principle of distinction. However, this reasoning still did not resolve the inconsistency between the levels of participation prohibited in Article 77(2) of AP I and Article 4(3)(c) of AP II; ' direct' in the former, even if interpreted broadly, may still cover a narrower scope of activities than 'take part in hostilities' in the latter.
Moreover, the Appeals Chamber did not address whether ' active' and ' direct' denote different standards. Waschefort presents an argument that the language indicates different standards: ' direct would speak to the proximity of one's contribution to the conduct in question, whereas active would speak to the intensity of one's participation in the conduct in question.' 151 While this analysis appears to give voice to the meanings of ' active' and ' direct', they should not be delineated so arbitrarily; a child's proximity to a conflict is likely to have at least some correlation to the intensity of their involvement. For instance, children working as bodyguards in Lubanga -a high-intensity form of 'indirect' participation -were extremely close to the conflict. As a witness in the Trial Decision explained, they saw two child bodyguards appear to run from their positions because shells were falling very close by them. 152 Accordingly, ' direct', were it to be used in the Statute's child protection provisions, is likely to have the same meaning as ' active'. Nevertheless, the meaning of either ' direct' or ' active' participation in hostilities should be interpreted according to the purpose of the provision in which the phrase is located.

C. Conclusion and Further Developments since the Lubanga Decisions
The Statute's child protection provisions prohibit the use of children to 'participate actively in hostilities' and are based on Article 77(2) of AP I and Article 4(3)(c) of AP II. However, this prohibition differs from the blanket prohibition on all participation in hostilities in Article 4(3)(c) of AP II for non-international armed conflict. 'Active' and ' direct' in the Statute's child protection provisions and Article 77(2) of AP II should be amended to prohibit any participation of children whatsoever. In light of the difficulties in amending these provisions, I evaluated the interpretation of these provisions in Lubanga.
The debates since the Lubanga Trial Decision reveal concerns that the broad approach taken by the ICC deviates from the current interpretation of ' direct' at IHL, which aims to protect civilians from being targeted in armed conflicts. Many fear that broadening ' active' participation in hostilities under the child protection provisions will cause civilians who indirectly participate in hostilities to be legitimately targeted. The Appeals Chamber appeared to address this concern by interpreting ' active' in civilian protection provisions like Common Article 3 differently to ' active' in the Statute's child protection provisions, also commenting that IHL child protection provisions should be interpreted broadly according to their purposes.
I support this approach. Defendants who use children to participate in hostilities will be more successfully prosecuted if ' active' is interpreted to include a broader scope of activities in the child protection provisions. Moreover, if ' active' also connotes a narrow definition in the civilian protection provisions like Common Article 3, armed groups will not be justified in targeting civilians who do not meet the high threshold for direct participation outlined in the Guidance. This approach still does not address the different standards of participation prohibited at non-international armed conflict in Article 4(3)(c) of AP II and Article 8(2)(e)(vii) of the Statute. However, it is preferable when considering the potential difficulty of successfully amending the latter provision.
Nevertheless, future Chambers may not follow this reasoning. For instance, the ICC Pre-Trial Chamber in the Ntaganda Confirmation of Charges Decision noted that it needed to assess whether children were taking a direct/active part in hostilities at the time they were victims of acts of rape and/or sexual slavery in order to determine whether they were entitled to protection from these acts. 153 Despite the Lubanga Appeals Chamber's reasoning, the Ntaganda Pre-Trial Chamber used Article 4(3)(c) of AP II as reflected in Article 8(2) (e)(vii) of the Statute to assess the direct/active participation in hostilities of children in the facts before them and appeared to equate direct/active participation under these provisions with a loss of protection under IHL. 154 This is problematic on two levels. Firstly, the Pre-Trial Chamber did not recognise that Article 4(3)(c) of AP II does not introduce a ' direct' or ' active' qualifier to participation in hostilities: rather, it is a blanket 151   prohibition on such participation. Therefore, that provision can arguably only be used to determine what 'participation in hostilities' means, rather than what 'participate actively in hostilities' means. Secondly, the Pre-Trial Chamber appears to assume that the interpretation of a provision drafted to protect children from use in hostilities can affect the interpretation of the same word in a provision drafted to distinguish between civilians and combatants. If this is so, future ICC Chambers may be more hesitant to interpret ' active' and ' direct' in the child protection provisions broadly because of fears that the scope of protection available to civilians may be narrowed. However, I have so far demonstrated that the scope of protection is not likely to be narrowed. Accordingly, the approach of the Lubanga Trial and Appeals Chambers should be preferred to the Ntaganda Pre-Trial Chamber's interpretation in future ICC decisions, as the former gives voice to the purposes of both the child and civilian protection provisions in the Statute and at IHL.

III. Determining the Scope of 'Active' Participation in Hostilities in the Statute's Child Protection Provisions: Evaluating the Approaches of the Lubanga Trial and Appeals Chambers
I have shown that it is reasonable to interpret ' active' in the IHL and Statute child protection provisions more broadly according to the purposes of these provisions rather than according to the principle of distinction. However, a consistent method for future Chambers to determine what kinds of activities fall within this broader scope is needed. The Lubanga Trial and Appeals Chambers used different methods to make this determination.

The Success of the Trial Chamber's Approach
Because the Trial Chamber interpreted ' active' in Article 8(2)(e)(vii) of the Statute broadly, it needed to determine whether the relevant children in Lubanga performed activities falling within the scope of this broader interpretation. To this end, the Trial Chamber noted that a child's exposure to real danger as a potential target is decisive when determining whether they are actively participating in hostilities by performing an 'indirect' role. 155 The SCSL has similarly concluded that children guarding mines met the threshold under the SCSL Statute due to the constant risk of attack they faced. 156 This approach prioritises the safety and protection of children while ensuring that defendants are prosecuted fully for putting children in danger by using them in hostilities. 157 Moreover, it grants the ICC 'the necessary flexibility when ruling on a specific case', 158 rather than being constrained by an excessively structured definition of active participation.

Deficiencies in the Trial Chamber's Approach
Nevertheless, this approach extends 'participate actively' beyond its scope. While the Trial Chamber correctly used the purposive approach of interpretation, the purpose of the Statute should not be extended too far. 159 The 'strict construction' approach in Article 22(2) of the Statute still applies, and acts as a safeguard to prevent an interpretation that 'yield[s] a new crime not contemplated by States Parties [to the Statute].' 160 According to the consequential risk analysis, a child within an army base may be continually at risk of attack by an enemy, but may only perform mundane cleaning chores, not doing anything to suggest that they are actively involved in the conflict. 161 A defendant charged under the Statute's child protection provisions may be found to have used such a child to participate actively in hostilities, when in fact the phrase 'participate actively in hostilities', on an ordinary construction, 162 implies a higher degree of participation. Accordingly, the threshold may be lowered too far if risk is the main factor in characterising activities under the Statute's child protection provisions, and the defendant may be convicted of a crime for which he is not liable. 163 Any interpretation of the phrase must consider both the purpose of the provision and the actual meaning of the word. For example, Judge Odio Benito, in her dissenting opinion in the Trial Chamber's judgment, argued that children actively participated in hostilities if they experienced sexual violence at the hands of members of armed groups that enlisted them. 164 However, this analysis goes ' clearly beyond the ordinary meaning of the wording' 165 and violates the strict construction requirement in Article 22(2) of the Statute. 166 Accordingly, being a victim of sexual violence cannot be viewed as ' actively' participation in the hostilities. 167 This does not mean that the provisions cannot be interpreted according to their purposes. The purposive approach taken by the ICC in Lubanga allows for broad interpretations within the reasonable parameters of a word's ordinary meaning, rather than extending a word or phrase beyond this meaning.
These deficiencies make consequential risk an unsuitable test by itself for future Chambers to use to determine whether activities fall within the scope of ' active' participation in the Statute's child protection provisions. Nevertheless, I argue in Part IV that consequential risk retains value as one of the factors to consider in making this determination.

B. The Appeals Chamber's Link-Based Approach to 'Active' Participation in Hostilities under Article 8(2)(e)(vii) of the Statute
In the Appeals Chamber, Mr Lubanga challenged the Trial Chamber's consequential risk analysis as 'wholly unfounded in international law or internationally recognised principles and rules.' 168 The Appeals Chamber agreed with Mr Lubanga; neither the ICL provisions nor their IHL equivalents referred to risk as a criterion to determine active participation in the Statute's child protection provisions. 169 Instead, a link between the hostilities and the activity in which the child was engaged was necessary for this determination. 170 While the extent to which a child was exposed to risk because of the activity could indicate a sufficiently close relationship between the activity and the hostilities, this risk could not replace an assessment of the relationship itself. 171 The Appeals Chamber did not provide further guidance on the parameters of active participation in hostilities under Article 8(2)(e)(vii) of the Statute, 172 beyond lists of activities in the Commentary on the Additional Protocols 173 and the Report of the Preparatory Committee on the Establishment of the ICC. 174 Rather, any determination as to whether a particular activity fell within the scope of the crime should be made on a caseby-case basis. 175 This was due to 'the complex and unforeseeable scenarios presented by the rapidly changing face of warfare in the modern world.' 176 The lists referenced by the Appeals Chamber read as follows: [Preparatory Committee's Report:] Active participation in military activities linked to combat (. . .) [includes] scouting, spying, sabotage and the use of children as decoys, couriers or at military checkpoints. It would not cover activities clearly unrelated to the hostilities such as food deliveries to an airbase or the use of domestic staff in an officer's married accommodation. However, use of children in a direct support function such as acting as bearer to take supplies to the front line, or activities on the front line itself, would be included within the terminology. 179 On the whole, the Appeals Chamber took a more faithful interpretation of the text in Article 8(2)(e)(vii)and by extension Article 8(2)(b)(xxxvi) -by avoiding the use of 'risk' as the central factor determining whether a child had actively participated in hostilities. Using the examples in the above lists to determine whether an activity constitutes ' active' participation under the Statute's child protection provisions allows judges to flexibly analyse the facts before them in various prosecutions for these crimes, ' as well as accounting for the great disparity between conflicts and the differing nature of roles within them.' 180 Nevertheless, these lists are deficient because they use undefined, inconsistent terminology and standards. For instance, the Commentary on Article 4(3)(c) of AP II prohibits a child's participation in 'military operations' but the Commentary on Article 77(2) of AP I appears to prohibit 'indirect' participation in hostilities and the Preparatory Committee's Report covers active participation in 'military activities linked to combat'. Furthermore, ' direct support function' in the Preparatory Committee's Report appears to require a child to be proximate to the front line, but does not specify the object of the child's support (whether an individual like Mr Lubanga or the conflicting party in general). These phrases are undefined and there is no listed example of these activities that crosses all three lists, which would allow the common ground between them to be determined.
Additionally, the examples in the Commentary to Article 4(3)(c) of AP II likely cannot be accurately applied to Article 8(2)(e)(vii). This is because the Commentary to Article 4(3)(c) gives examples of taking 'part' in hostilities but Article 8(2)(e)(vii) prohibits taking an ' active' part in hostilities. However, the Appeals Chamber did not consider this difference and applied the Commentary on both Additional Protocols equally as guidance. Therefore, the degree to which the Commentary on Article 4(3)(c) can be relied upon to interpret ' active' participation in Article 8(2)(e)(vii) is uncertain.
These lists do not appear to form a clear threshold for an activity to qualify as ' active' participation in hostilities. Accordingly, future ICC Chambers may apply them inconsistently to the various situations that come before them. As McBride notes, 'the prohibition on child recruitment will not be effective if its basic concepts are unclear or contradictory', 181 and ' active' participation is a basic concept of the prohibition. Therefore, determining the scope of activities it covers should be a clear and consistent process.

C. Conclusion
The Appeals Chamber's link-based approach to determining whether a child has actively participated in hostilities under the Statute's child protection provisions is a more effective, faithful approach to the text of these provisions than the Trial Chamber's risk-based approach. However, the Appeals Chamber did not provide sufficient, consistent guidance for future Chambers to determine this link. Accordingly, the case-by-case determination of activities as falling within the scope of active participation in the child protection provisions is likely to produce inconsistent jurisprudence on certain activities. Further clarification is needed to ensure consistent prosecution of those who use children in hostilities under these provisions.

IV. An Update to the Guidance: Assistance for the Interpretation of 'Active' and 'Direct' Participation in Hostilities under the Child Protection Provisions
The child protection provisions in the Statute and AP II should be amended to prohibit all participation in hostilities. 182 In light of the difficulties of amending these instruments, I propose that the Guidance should be updated with factors that future ICC Chambers may consider in determining whether children actively participated in hostilities under the IHL and Statute child protection provisions.

A. Benefits of an Updated Guidance
An updated Guidance provides a number of benefits. Firstly, it would clarify the nature of the crime under the Statute. It would clearly delineate between the different purposes of the relevant provisions while ensuring that parties are aware of their rights and obligations at both IHL and ICL in relation to the use of children in hostilities.
Secondly, it would give the ICC a more comprehensive outline of active participation in the child protection provisions, ensuring that its decision-making is consistent when assessing the link between the activity and the hostilities. This link would be the main factor in attributing liability to an individual under these provisions. 183 As the ICC is not bound by its previous decisions, 184 future ICC Chambers will be able to depart from the lists the Appeals Chamber used to assess the link between the activity and the hostilities. There are two factors to determine: (a) the type of hostilities that exist; and (b) whether the activity a child performs is sufficiently linked to these hostilities.

The Type of Hostilities that Exist
Because the Statute's child protection provisions require the existence of a link between the activity and the hostilities, the existence of hostilities must be confirmed. If there is doubt about whether hostilities exist, defendants could argue that their use of children for activities does not make them liable under these provisions.
The Elements of Crimes of the Statute's child protection provisions require that the conduct took place in the context of and was associated with an armed conflict. 185 An armed conflict exists: (. . .) whenever there is a resort to armed force between States or protracted violence between governmental authorities and organised armed groups or between such groups within a State (. . .) until a general conclusion of peace is reached (. . .) or, in the case of internal armed conflicts, a peaceful settlement is achieved. 186 The violence of any armed conflict must be more than sporadic in the case of a non-international armed conflict. 187 The Lubanga Trial Chamber endorsed the ICTY's approach to determining whether a conflict was an armed conflict or not: 188 the intensity of the conflict should be used 'solely as a way to distinguish an armed conflict from banditry, unorganised and short-lived insurrections, or terrorist activities, which are not subject to international humanitarian law.' 189 The intensity of an armed conflict is determined by the following: (. . .) the seriousness of attacks and potential increase in armed clashes, their spread over territory and over a period of time, the increase in the number of government forces, the mobilisation and the distribution of weapons among both parties to the conflict, as well as whether the conflict has attracted the attention of the UN Security Council, and if so, whether any resolutions on the matter have been passed. 190 These principles should be adopted by future Chambers when assessing whether hostilities existed at the time a defendant is alleged to have used children to participate actively in them.
1. The risk or danger to the child as a result of performing the activities; 2. The impact of the activities on the hostilities; 3. The degree to which the activities will prepare children to participate in the hostilities.
a. Risk or Danger to the Child as a Result of Performing the Activities Consequential risk is a useful factor to determine the link between the activity and the hostilities. 191 Including it in the analysis keeps future Chambers focused on the child's safety and liberty. As Ang comments, 'there is a real risk that any participation of some children in hostilities casts suspicion on and creates danger for all children.' 192 In Syria, Uganda, India and Nigeria, children have reportedly been used as human shields. 193 Armed forces in South Sudan have also used children to guard high-ranking officials, 194 and insurgent groups in Iraq have even used children plant explosive devices. 195 In 2013, moreover, children in the Philippines were reported to be 'injured and killed in hostilities as an immediate result of their association with [the military wing of the New People's Army]'. 196 The number of verified cases of child recruitment for conflict in the Philippines has decreased since 2013, but armed groups have continued to successfully recruit children. 197 These situations show the significant risk a child is placed in as a result of the activities they perform in support of combatants. Accordingly, this risk should be used as a factor to determine whether the child's activities are sufficiently linked to the hostilities under the child protection provisions.

b. The Impact of the Activities on the Hostilities
The more significant the impact of a child's actions on a conflict, the likelier it is that they will be participating actively in hostilities under the child protection provisions. In some instances, a child's actions may clearly constitute active participation in hostilities and have a significant impact on a conflict. For example, the Tamil Tigers in Sri Lanka used Thenmozhi Rajaratnam, believed to have been under 18 years old at the time, to assassinate former Indian Prime Minister Rajiv Gandhi. 198 However, a child's actions may still have a sizeable impact on the conflict without being as obvious as in Ms Rajaratnam's case. Accordingly, this impact can help courts determine whether the child was actively participating in the hostilities under the child protection provisions. An example of indirect yet active participation in hostilities is the use of children by insurgent groups in Iraq to videotape attacks for propaganda purposes: 199 while they may not be directly participating in the conflict, the effect of the propaganda on the conflict may be extremely significant. 200 A distinction must be made, however, between the use of children to participate in hostilities and other purposes. The effect of a child's actions on the hostilities is useful when making this distinction, because a child may not be actively participating in hostilities if their actions have no effect on the hostilities themselves. For instance, children in Mali enforced the Islamic dress code for women and conducted inspections of contraband items. 201 While such use may be suspect, there does not appear to be a link between these activities and the hostilities occurring.
c. The Degree to which the Activities Will Prepare the Children to Participate Actively in Hostilities Many children are indoctrinated and equipped to participate in hostilities. 202 Such training does not fall within the current definition of active participation in the child protection provisions. 203 As Okebukola argues, 'use' and 'participate' are present-tense verbs, and preparation for any such activities does not constitute the performance of the activities themselves. 204 Nevertheless, if the child protection provisions aim to prevent children from being used in hostilities, any work preparing them to be used in hostilities should be prevented as well. 205 A distinction must therefore be made between general training of children and training of children for specific hostilities. 206 Future defendants on trial for these crimes may argue that they were merely training children generally, rather than for use in a specific armed conflict. The criteria provided by the ICTY, endorsed by the ICC's Trial Chamber, 207 are reliable for determining whether an armed conflict existed. From that determination a Chamber could decide whether or not the training was for use in such a conflict. The Chamber could consider the following factors, among others: (a) had children been sent into this conflict before? (b) what kind of activities were the children being trained to do? (c) how close was the conflict to the training? (d) is there documentary evidence detailing plans to send children into this conflict?
Only the training of children for specific hostilities would fall under ' active' participation in hostilities in the child protection provisions, as there would be a link between the training and the hostilities. 208 In Somalia, for example, children have been trained both in basic arms techniques and assassination, intelligence collection, the use of improvised explosive devices and suicide missions. 209 The forced recruitment of children has been connected with the upsurge in fighting in the Somali civil war. 210 Accordingly, it would be open to future ICC Chambers to find that the training was sufficiently linked to these hostilities to fall under the Statute's child protection provisions.

V. Conclusion
The Lubanga decisions show that the ICC is ready, willing and able to prosecute the use of children in hostilities. In convicting Mr Lubanga and upholding this decision on appeal, the ICC demonstrated that it has the potential to be an effective weapon against this practice as a deterrent against potential offenders by justly and rigorously punishing those who do engage in it. In Lubanga, the ICC interpreted ' active' in the Statute's child protection provisions more broadly than ' active' in the civilian protection provisions. This approach best gives voice to the protective purpose of the child protection provisions by criminalising a wider range of conduct that children could perform, deterring individuals and groups from using them in hostilities. It also ensures that the interpretation of ' active' and ' direct' in the child protection provisions does not affect the protection of civilians from becoming legitimate targets at IHL. Moreover, while the Trial Chamber used 'risk' as the main element to determine whether a child participates actively in hostilities, the Appeals Chamber concluded that a link between the activity and the hostilities is the central requirement for this determination. The Appeals Chamber's approach is more faithful to the text of the Statute's child protection provisions. Accordingly, it should be adopted in future child soldier matters before the ICC.
These judgments raise a number of unresolved issues, however. Firstly, Article 4(3)(c) of AP II imposes a blanket prohibition on the use of children in non-international armed conflict at IHL, which is not reflected by either Article 77(2) of AP I or the Statute's child protection provisions. The Statute's child protection provisions and Article 77(2) of AP I should be amended to reflect this prohibition, ensuring that any use of children in hostilities is both prohibited at IHL and criminalised at ICL. However, amending these instruments is difficult.
Secondly, the ICC Pre-Trial Chamber in Ntaganda has not adopted the purposive approach taken in Lubanga. Instead, it appeared to conflate ' active' participation in child protection provisions with the loss of immunity from being legitimately targeted at IHL. If future Chambers follow the Pre-Trial Chamber's reasoning, they may not interpret ' active' in the child protection provisions broadly according to their purposes, for fear of narrowing the scope of protection available to civilians. Accordingly, the ICC may not effectively prosecute those who do use children to participate in roles that do not qualify as ' active' or ' direct' participation in the narrow sense. I argued that the ICC's reasoning in Lubanga should be followed in future decisions to give voice to the protective purposes of the Statute's child protection provisions.
Thirdly, the lists the Appeals Chamber provided to assess the link between a child's activity and the hostilities under the child protection provisions are insufficient, inconsistent with one another and contain undefined phrases. This may lead to inconsistent prosecution of individuals for using children in hostilities. Accordingly, the provisions will not fulfil their purposes to protect children from use in hostilities.
To help future courts address the inconsistencies in these lists and ensure that their own approaches to matters involving the use of children in hostilities, I suggested that the ICRC's Guidance should be updated, providing information to help future Chambers and parties determine whether, in any given situation, a conflict existed, and whether a child's activities were sufficiently linked to that conflict. I proposed three criteria to be included in the updated Guidance to assist Chambers in assessing this link: (a) the risk a child faces as a result of the activity; (b) the impact of the child's activity on the hostilities; and (c) the degree to which the activity prepares the child for participation in the hostilities.
Future Chambers are likelier to interpret the relevant provisions consistently if they use these criteria to assess the link between a child's activity and the relevant hostilities. Over time, it is hoped that the increased consistency in the ICC's jurisprudence will work to deter individuals and groups who would otherwise use children in hostilities. As massive numbers of children are being recruited to participate in hostilities around the world, the ICC's practice must be geared towards the effective prosecution of individuals and groups who commit these crimes. Clear, consistent guidelines and properly defined terms will help the ICC to streamline its decision-making processes, increasing the threat of successful prosecution facing parties who recruit children for use in hostilities.
The Prosecutor v Bosco Ntaganda (Decision Pursuant to Article 61 (7)

Israeli Supreme Court
The Public Committee against Torture in Israel and Palestinian Society for the Protection of Human Rights and the Environment v The Government of Israel and Ors (2006) HCJ 769/02.