Ongwen: the child soldier defense under international criminal law

Written by Juan Pablo Hernández Páez
The Treaty Examiner, vol. 2, issue 1
9 February 2021


Without exception, war always brings destruction and misery to all parties involved. Of all victims, the plight of children is, I believe, the most heartbreaking. War tears families and homes apart, leaving children vulnerable. As the future of their respective nations, the suffering of children during armed conflict leaves gaping wounds in national history and causes subsequent generations to bleed. However, children do not only suffer as victims. Conflicts have witnessed the horrific practice of some armed forces, whereby they recruit and conscript children to act as soldiers for their side. This heinous practice of conscription forces victims to become perpetrators: to take up arms and relinquish their life, physical integrity and innocence to protect the interests of their kidnappers. These child abductees often go on to commit international crimes, especially grave violations of international humanitarian law, under the command of their abductors. It is no wonder that enlisting children to participate in hostilities is prohibited by international humanitarian law and considered a war crime under Article 8 of the Rome Statute of the International Criminal Court. The 2012 conviction of Thomas Lubanga Dyilo by the ICC (¶1358) is an example of the horrors of conscription.

However, criminalizing the participation of children in armed conflict is only a half answer to the problem of child soldiers. What happens to the criminal liability of the child for the actions taken under the command of his or her kidnapper? In many cases, the child will be under the age of criminal responsibility in his or her national system. Yet, at the international level, customary international law mandates no particular age of criminal responsibility — State practice on this point is not very consistent. 1989 Convention on the Rights of the Child does not stipulate an age of criminal responsibility but obligates State parties to define such an age (Article 40(3)(a)). In humanitarian law, a proposal was made to exclude criminal liability under the age of sixteen in Article 77(5) of Additional Protocol I; however, the proposal was rejected and the issue of establishing an age of criminal responsibility for war crimes was left to national law (Happold, 2006). The Rome Statute of the ICC is silent on the age of criminal responsibility and avoids the issue by instead framing it in terms of personal jurisdiction in Article 26:

The Court shall have no jurisdiction over any person who was under the age of 18 at the time of the alleged commission of a crime.”

Whether the crime is imputable to a child perpetrator is, however, discussed less at the international stage (although sobering accounts describe the prosecution and chilling execution of child perpetrators at the national level; see Leveau, 2013, at pp. 55-59). The reluctance to charge children with war crimes before international criminal tribunals is evidenced, for instance, by a statement of the Prosecutor of the Special Court for Sierra Leone in 2002. The central problem arises when the child soldier reaches adult age and continues to commit international crimes, usually still under the command of his or her abductors. Can such a perpetrator raise the fact that he or she used to be a child soldier to disclaim responsibility for international crimes committed after reaching adult age? Participants of the 2017 ICC Moot Court Competition will remember the fictional story of Tarion Bannister of Valaria, child soldier conscripted at the age of eight, subjected to indoctrination, forcible use of narcotics and active participation in armed conflict, who later goes on allegedly to commit genocide at the age of 18. These haunting stories, unfortunately, are not reserved to the realm of fiction and thought experiments. The ICC had a unique opportunity to address the criminal liability of child soldiers in its recent Prosecutor v. Dominic Ongwen judgment, where it issued a conviction on war crimes and crimes against humanity against Sinia brigadier Dominic Ongwen. The purpose of this article is to examine the Ongwen case and the treatment of the child soldier defense under international criminal law.

Dominic Ongwen: victim to perpetrator

“It is a story of a child, like many in the LRA, forced to grow up in the image of their oppressors.”

LRA Crisis Tracker

It is believed that Dominic Ongwen was abducted from Northern Uganda by the Lord’s Resistance Army (LRA) at the age of nine or ten, around 1988. The name “Ongwen” (“born at the time of the white ant”) was an alias his parents gave him due to fear of abduction and retaliation by LRA leader Joseph Kony. It is reported that Ongwen attempted to escape from the LRA multiple times shortly after his abduction. All of these attempts were quashed and met with warnings and punishment. During the rest of his childhood and adolescence, he was subjected to cruel and intense indoctrination by LRA superiors in order to become a fighter for the rebel group. Under the tutelage of Vincent Otti, Ongwen grew in rank and, in his twenties, he became a commander for the LRA Sinia Brigade. His alias soon became his nom de guerre within LRA ranks.

In 2005, the ICC issued an arrest warrant against Otti, Kony and Ongwen. Otti is reported to have died circa 2007. Accounts of Ongwen’s own death were also circulated but later refuted by the ICC. In 2014, Ongwen managed to escape the LRA and flee to the Central African Republic (CAR). There, he turned himself over to CAR authorities and expressed his desire to surrender voluntarily to the ICC. On 16 January 2015, Ongwen was turned over to the ICC by the CAR and was assigned duty counsel by the Court.

On 23 March 2016, ICC Pre-Trial Chamber II confirmed the charges against Dominic Ongwen on 70 criminal counts including war crimes and crimes against humanity committed from 2002 to 2005 (pp. 71-104). Relevant to our inquiry here is the Chamber’s pronouncement on the grounds raised by the Defense to exclude Ongwen’s individual criminal responsibility. Counsel for Dominic Ongwen raised two main defenses to exclude the Defendant’s responsibility: firstly, a sui generis “child soldier” defense under Article 31(3) and the defense of duress under Article 31(1)(d) of the ICC Statute. The legal merits of these defenses will be analyzed in further detail in this piece. In its 2016 decision, the Pre-Trial Chamber found that the first sui generis defense invoking Ongwen’s former status as a child soldier was without legal basis (¶150). Moreover, the Chamber found that none of the elements of duress were established by the Defense to the applicable standard of proof (¶¶150-156). Thus, Ongwen was committed to trial.

Summary of the Trial Chamber’s findings

On 4 February 2021, Trial Chamber IX of the ICC found Dominic Ongwen guilty of a litany of charges, including war crimes and crimes against humanity, for his actions as a leading figure of the LRA. This conviction followed a renewed analysis by the Court of Ongwen’s individual criminal responsibility under Article 31 of the ICC Statute. The Defense raised, particularly, the grounds of insanity (Article 31(1)(a) of the Statute) and, once again, of duress (Article 31(1)(d) of the Statute). The Defense also raised an alibi defense for Ongwen’s participation in the Pajule attacks. The Chamber, however, rejected all defenses.

Firstly, on the alibi defense, the Court found that alibis are not, in technical terms, grounds for excluding criminal responsibility. Rather, their purpose is to prove the “physical impossibility of an accused’s guilt by placing him/her in a location other than the scene of the crime at the relevant time” (¶2449). The Chamber had arrived at the conclusion that Ongwen had committed the relevant acts beyond a reasonable doubt, and thus the alibi argument is not further entertained.

Secondly, the Chamber considered the insanity defense under Article 31(1)(a) of the ICC Statute. The Defense, particularly, had raised the argument that Ongwen acted under PTSD, dissociative disorder and suicidal ideation, as well as a dissociative amnesia and OCD, which according to expert evidence should exclude Ongwen’s responsibility for the relevant crimes (¶2450). The Court declined to consider the evidence of Drs. Akena and Ovuga, proposed by the Defense, as it ruled them to be unreliable (¶2574). On the basis of the remaining expert evidence, the Court ruled that it had not been established that Ongwen was under a mental disease or defect that destroyed his capacity to control his conduct or to appreciate the unlawfulness or nature of his conduct “at the time of [his] conduct” as required by Article 31(1)(a) (¶2580).

Finally, the Chamber considered the duress plea under Article 31(1)(d) of the Rome Statute. The Court ruled that the evidence did not support a finding that Ongwen was, at the time of the charges, under “a threat of imminent death or of continuing or imminent serious bodily harm.” While accepting the brutality of LRA disciplinary measures (¶2590), Ongwen was not a low-ranking LRA fighter. Rather, Kony relied on LRA commanders, like Ongwen, to enforce disciplinary measures. The Court ruled that Ongwen’s experience as a child soldier “is not central to the issue” as his position was different from that of a recent abductee (¶¶2591-2592). The Court also noted that Kony’s orders were not always implemented by LRA commanders (¶¶2593-2598). Ongwen’s experience, the Chamber noted, was no different: Ongwen would assess his orders and determine whether they were practical or feasible before implementing them (¶2599). Thus (¶2602),

…the relationship between Joseph Kony and Dominic Ongwen was not characterised by the complete dominance of the former and subjection of the latter. On the contrary, what results clearly from the above witness testimonies is that Dominic Ongwen was a self-confident commander who took his own decisions on the basis of what he thought right or wrong…

The killing of other LRA commanders, like Vincent Otti, was not indicative of a disciplinary measure to punish disobedience, but were rather a response to what the Court characterized as a political challenge to Kony’s power as exclusive leader of the LRA (¶2614). Moreover, escapes from the LRA were relatively common (¶2619), including by high-ranking LRA commanders (¶2621). All these factors, in the Chamber’s view, militated in favor of finding that Ongwen acted on his own accord and not under a threat as required by Article 31(1)(d) of the Statute (¶2668).

Availability of defenses under the ICC Statute and the relevant burden of proof

The relevant rule of law in the context of the child soldier defense in the ICC framework is Article 31 of the Statute (“Grounds for excluding criminal responsibility”), which provides as follows:

1. In addition to other grounds for excluding criminal responsibility provided for in this Statute, a person shall not be criminally responsible if, at the time of that person’s conduct:

(a) The person suffers from a mental disease or defect that destroys that person’s capacity to appreciate the unlawfulness or nature of his or her conduct, or capacity to control his or her conduct to conform to the requirements of law;

     (b) The person is in a state of intoxication that destroys that person’s capacity to appreciate the unlawfulness or nature of his or her conduct, or capacity to control his or her conduct to conform to the requirements of law, unless the person has become voluntarily intoxicated under such circumstances that the person knew, or disregarded the risk, that, as a result of the intoxication, he or she was likely to engage in conduct constituting a crime within the jurisdiction of the Court;

     (c) The person acts reasonably to defend himself or herself or another person or, in the case of war crimes, property which is essential for the survival of the person or another person or property which is essential for accomplishing a military mission, against an imminent and unlawful use of force in a manner proportionate to the degree of danger to the person or the other person or property protected. The fact that the person was involved in a defensive operation conducted by forces shall not in itself constitute a ground for excluding criminal responsibility under this subparagraph;

     (d) The conduct which is alleged to constitute a crime within the jurisdiction of the Court has been caused by duress resulting from a threat of imminent death or of continuing or imminent serious bodily harm against that person or another person, and the person acts necessarily and reasonably to avoid this threat, provided that the person does not intend to cause a greater harm than the one sought to be avoided. Such a threat may either be: (i) Made by other persons; or (ii) Constituted by other circumstances beyond that person’s control.

2. The Court shall determine the applicability of the grounds for excluding criminal responsibility provided for in this Statute to the case before it.

3. At trial, the Court may consider a ground for excluding criminal responsibility other than those referred to in paragraph 1 where such a ground is derived from applicable law as set forth in article 21. The procedures relating to the consideration of such a ground shall be provided for in the Rules of Procedure and Evidence.

The issue of the burden and standard of proof in connection with Article 31 was subject to some debate in Ongwen, in part due to the different procedural stages in which the provision has been addressed. Notably, in the confirmation of charges, counsel for Ongwen raised defenses under Article 31(1) and (3) of the Statute, to no success. Pre-Trial Chamber II, in rejecting the grounds for excluding criminal responsibility, considered the possibility of their invocation at the confirmation of charges and the issue of proof. As to invocation, the Court ruled that while Article 31(3) could only be invoked “[a]t trial” according to its ordinary language, no such limitation was imposed on Article 31(1) defenses, including duress (¶151). On the issue of proving duress at the confirmation of charges, the Chamber made an interesting finding of law (¶151):

There is no procedural rule precluding the Defence from raising duress at the stage of confirmation of charges, as also confirmed, a contrario, by the text of article 31(3) of the Statute. However, considering the nature and purpose of confirmation of charges proceedings, duress may only lead to the non-confirmation of charges when the evidence is so clear that it negates even the low evidentiary standard applicable. Otherwise, a trial is appropriate in order to resolve also this question.

The standard of proof at the confirmation of charges, in order to commit a defendant to trial, is that there must be “sufficient evidence to establish substantial grounds to believe that the person committed the crime charged” (Article 61(5) of the Statute). This standard, when compared to the “beyond reasonable doubt” requirement for trial (Article 66(3) of the Statute), is rather low. Following the Pre-Trial Chamber’s finding in the above-quoted paragraph, for a duress plea to succeed, the Defense would have to supply sufficient evidence to negate the confirmation of charges standard, i.e., to establish that there are no substantial grounds to believe that the person committed the crime charged, based on the relevant defense. This suggests that the standard of proof for defenses operates on a mirror-like fashion: where the Prosecutor has a lower evidentiary threshold to commit the accused to trial, the defense has a correspondingly high requirement to frustrate that low threshold. This also suggests that, at trial, the Defense would only be required to prove doubt (Article 66(3) a contrario).

Does this mean, however, that the onus of proving the defense was on counsel for Ongwen? Following a general theory of proof, the party raising an affirmative defense must provide the evidence that substantiates that plea. Thus, the Pre-Trial Chamber’s reasoning in the Ongwen confirmation of charges, read superficially, could suggest that the burden was on the Defense to prove the accuracy of the duress allegation. Yet, later in Ongwen, Trial Chamber IX found as follows:

The Chamber notes that there has been a considerable amount of litigation on the issue of burden and standard of proof. As already stated above, there is no specific provision in the Statute related to the burden and standard of proof as concerns grounds excluding criminal responsibility under Article 31, and for this reason, the general provisions of the Statute apply. Under Article 66(2) of the Statute, the onus is on the Prosecutor to prove the guilt of the accused, and, under Article 66(3), in order to convict the accused, the Court must be convinced of the guilt of the accused beyond reasonable doubt.

Did the Trial Chamber contradict the Pre-Trial Chamber’s finding at the confirmation stage? In truth, no. The reality is that Pre-Trial Chamber II never ruled on the burden of proof for defenses under Article 31(1) at the confirmation of charges. In other words, the Chamber’s finding does not address who was under the duty to prove the duress plea. The Pre-Trial Chamber, rather, ruled on the standard of proof —  to say it simply, on how much evidence was required to establish the defense.

Thus, the Chambers’ findings stand side to side on the issue: while the standard of proof changes throughout the proceedings depending on the relevant stage, as happens also to incriminating evidence, the burden of proof remains with the Prosecutor. Indeed, Article 66(2) of the Statute is clear that “[t]he onus is on the Prosecutor to prove the guilt of the accused” and the Statute prohibits any reversal of that burden under Article 67(1)(i), including “any onus of rebuttal”. Under Article 54(1)(a), the Prosecutor is under a statutory obligation to investigate the accused’s criminal responsibility, including exonerating evidence. This, of course, does not mean that it is not incumbent on defense counsel, in discharging his or her duties and protecting the interests of the accused, to supply the necessary evidence for an ICC Chamber to issue a finding excluding criminal responsibility.

Insanity (Article 31(1)(a))?

The first defense that can be considered in connection with a defendant’s status as a former child soldier is insanity, i.e., whether the defendant was under a “mental disease or defect” that destroyed his or her capacity to appreciate the unlawfulness of his or her conduct or that impeded him or her from controlling his or her conduct to comply with the requirements of the law. The Trial Chamber in Ongwen made a number of findings relevant to this defense. Firstly, the relevant disease or defect has to be established at the time of the conduct and not at the time of the proceedings, although the latter can, in principle, influence the former (¶2454). Secondly, either alternative (destruction of ability to appreciate unlawfulness or destruction of ability to control conduct) suffices to establish the defense (¶2452). From the Chamber’s considerations, then, two requirements can be discerned:

  1. There needs to be a mental disease or defect affecting the accused at the time of the imputed conduct.
  2. That mental disease or defect must have destroyed the accused’s ability to appreciate the unlawfulness of conduct or to control it to conform with law.

In Ongwen, however, the Trial Chamber only addressed the first requirement (the existence of the mental affliction). As noted above, the Chamber concluded that the evidence did not support a finding that Ongwen was under such an affliction at the time of the conduct. On this element, suffice it to say, it appears that specific scientific and expert evidence is required to establish the existence of the disease at the relevant time.

Conversely, the issue of the “destruction” of Ongwen’s capacity to appreciate unlawfulness or control his conduct was largely left unresolved, as logically follows from the absence of proof of mental disease. One should then proceed to interpret the relevant terms in light of the general rule of interpretation under public international law, as codified in Article 31 of the 1969 Vienna Convention on the Law of Treaties (VCLT). The term “destroy”, in its ordinary meaning, would require that the relevant capacities of the accused no longer exist or can no longer be used as a result of the disease or defect — thus, it would appear from a textual reading of Article 31(1)(a) that mere limitation of those capacities is not sufficient. This is confirmed by the context of that provision. Rule 145(2)(a) of the ICC Rules of Procedure and Evidence, which is part of the Court’s primary sources of law (Article 21(1)(a) of the Statute), indicates that diminished mental capacity does not suffice for exclusion of criminal responsibility but can be considered a mitigating circumstance for the sentence. As for object and purpose, the Statute was created to punish the gravest crimes known to the international community. Considering the gravity of the crimes within the Court’s material jurisdiction (genocide, crimes against humanity, war crimes and aggression), it follows that not all mental afflictions should lead to absolving the accused. While the particular circumstances of the case are of great relevance, it is submitted that the importance of the crimes in consideration, which are of utmost concern for all States, should require a similarly grave affectation of the accused’s mental capacity in order to lead to excuse. This applies a fortiori to such crimes under the Court’s jurisdiction that require a heightened mental element, like the crime of genocide (Article 6).

As a person that has only gone through legal education, I will not presume to know which diseases or defects, from a psychological perspective, would lead to acquitting a defendant under Article 31(1)(a) of the ICC Statute. However, it must be borne in mind that child soldiers are among the ones that suffer the most in armed conflict. The 2012 ICC trial of Thomas Lubanga has put in display the psychological effect that conscription has on children and the devastating consequences of forcing such children to participate actively in hostilities from such a young age. Relying on the expert report by Ms. Elisabeth Shauer, Trial Chamber I noted the following in Lubanga (¶30):

During her testimony, Ms Schauer stated that the trauma suffered by child soldiers has intellectual and cognitive consequences in the children’s minds. Children who have suffered trauma have problems with their memory and may have learning difficulties, particularly as regards reading and writing comprehension. She also affirmed that this trauma never goes away. The expert further stated that although persons with post-traumatic stress disorder may recall events that occurred in the past, their ability to answer and remember these events will depend on the way questions are asked, and if they are asked chronologically. She literally stated “you probably have a hard time just wanting to know – jumping and wanting to know little details here and there.”

From a legal perspective, it is reasonable to assume that the affectation to the child soldiers’ capacity to learn and retain memories as a result of the horrific trauma they experience in armed conflict could, in the right circumstances, lead to a finding that the soldier’s ability to appreciate the unlawfulness of his conduct or to control it was damaged beyond recognition. As is noted by Ms. Schauer in Lubanga, such trauma never goes away, even (or rather, especially) when the child soldier has spent years in rebel ranks and, thus, there is reason to believe that a former child soldier could be excluded from liability for acts committed in adulthood. Anything beyond those considerations, however, lies outside my area of expertise.

Duress (Article 31(1)(d))?

The second defense which is relevant for child soldiers is that of duress, established under Article 31(1)(d) of the ICC Statute. According to Trial Chamber IX’s findings in Ongwen, Article 31(1)(d) imposes the following requirements (¶¶2580-2584):

  1. The conduct must have been caused by duress resulting from a threat of imminent death or of continuing or imminent bodily harm against the accused or another person.
  2. The accused must act necessarily and reasonably to avoid the threat.
  3. The accused must not have intended to cause a greater harm than the one sought to be avoided.

On the facts, the Chamber found that Ongwen was not under a relevant threat and therefore did not address elements 2 and 3 of Article 31(1)(d). At the confirmation stage, the Pre-Trial Chamber ruled that Ongwen did not act in conformity with any of the requirements of duress. From the findings of both Chambers, the following conclusions can be drawn.

As to the first requirement, the existence of a threat is generally a matter of objective fact. The Chambers ruled that the mere existence of dangerous circumstances or an abstract threat does not suffice to trigger Article 31(1)(d). Moreover, as regards threats of death, the threat must be such that it will materialize “sufficiently soon”, unless the accused commits the crime. As regards threats of bodily harm, the threat can either be imminent or ongoing. The objectivity of this first requirement is justified, once again, by the gravity of the crimes against which the defense is intended to guard. Thus, it should not suffice that the person simply believes that he or she is under a threat, if that threat will not materialize.

However, this does not mean that the Court should disregard the subjective appreciation of the facts by the accused. A threat only exists insofar as the accused interprets it as such, so that it compels his or her conduct to commit the crime in question. Thus, even a subjective threat, not established in objective fact, can contribute to a finding that Article 31(1)(d) applies to the facts of the case. The possibility of a subjective threat (or a subjective contributing factor) under Article 31(1)(d) is especially relevant for cases involving child soldiers and was implicitly accepted by Trial Chamber IX in Ongwen, when it considered the effect of Kony’s alleged spiritual powers (¶¶2644, 2645, 2658):

During the trial, the Chamber heard a number of personal accounts by former LRA members who were questioned about the effect of LRA spiritualism on them. This evidence informs the Chamber’s view on whether spiritualism was in some way used to create or sustain a threat relevant under Article 31(1)(d) of the Statute.

Whereas there is evidence that some persons did believe in the spiritual powers of Joseph Kony, the Chamber observes that there is consistent evidence that for many persons who stayed in the LRA longer their belief followed a pattern: it was stronger in the young, new and impressionable abductees and then subsided and disappeared in those who stayed in the LRA longer.

(…)

All of this evidence leads the Chamber to the conclusion that LRA members with some experience in the organisation did not generally believe that Joseph Kony possessed spiritual powers. There is also no evidence indicating that the belief in Joseph Kony’s spiritual powers played a role for Dominic Ongwen, and in fact the evidence of Dominic Ongwen defying Joseph Kony, discussed above, speaks clearly against any such influence. The Chamber therefore does not discern in the issue of LRA spirituality a factor contributing to a threat relevant under Article 31(1)(d) of the Statute.

As to the second requirement, the Court established in Ongwen that it is not required for the accused to take all conceivable measures to avoid the threat. A requirement of feasibility informs Article 31(1)(d) so that the Court would need to assess the specific circumstances of the accused. The Court has found instructive, in particular, whether a person in a comparable circumstance had taken measures to avoid the threat successfully. In applying Article 31(1)(d), it is submitted that Trial Chamber IX’s approach is generally warranted: the ICC chamber should assess persons in comparable circumstances, particularly other child soldiers in relatively the same position as the accused. The Court cannot impose on a child soldier the standards of a person that has not gone through the same kind of indoctrination and torture. Circumstances vary even between child soldiers, as Trial Chamber IX noted in Ongwen’s case. Thus, for example, the Court could not take the conduct of the children abducted by Thomas Lubanga in the Democratic Republic of the Congo as a parameter of reasonableness or necessity for a child soldier abducted by the LRA in Uganda. The Court can also not take into account the actions of child soldiers in lower or higher ranks to establish the second requirement of duress. What is feasible for a child soldier, considering the continuous threats against life and limb, is to be assessed on a case-by-case basis.

The final requirement is one of proportionality. This is the only standard that is fully subjective in nature. This is accepted by the Trial Chamber in Ongwen when it referred to the intent not to cause greater harm. At the confirmation of charges, the Pre-Trial Chamber considered whether Ongwen’s actions were proportionate to the threat, concluding that his conduct did not evidence an intent to reduce the harm to the civilian victims of his crimes (¶¶150-155). To draw on a more general standard, it is submitted that the approach adopted by Judge Antonio Cassesse in his separate and dissenting opinion in the ICTY Erdemovic case is appropriate for cases involving child soldiers. Disagreeing with the majority on the invocation of duress in cases involving the killing of innocent civilians, Judge Cassesse drew a distinction between cases where the accused can and cannot prevent the commission of the crime by refusing to obey the order (¶44):

Thus the case-law seems to make an exception for those instances where — on the facts — it is highly probable, if not certain, that if the person acting under duress had refused to commit the crime, the crime would in any event have been carried out by persons other than the accused. The commonest example of such a case is where an execution squad has been assembled to kill the victims, and the accused participates, in some form, in the execution squad, either as an active member or as an organizer, albeit only under the threat of death. In this case, if an individual member of the execution squad first refuses to obey but has then to comply with the order as a result of duress, he may be excused: indeed, whether or not he is killed or instead takes part in the execution, the civilians, prisoners of war, etc., would be shot anyway. Were he to comply with his legal duty not to shoot innocent persons, he would forfeit his life for no benefit to anyone and no effect whatsoever apart from setting a heroic example for mankind (which the law cannot demand him to set): his sacrifice of his own life would be to no avail. In this case the evil threatened (the menace to his life and his subsequent death) would be greater than the remedy (his refraining from committing the crime, i.e., from participating in the execution).

The haunting reality of the child soldier problem is that abductors tend to treat them as replaceable. Accounts of child soldiers evidence that disobedience is usually met with death or severe punishment, and military troops can be rapidly replenished by further abduction of children (see, for instance, Grant, 2016). Child soldiers are indoctrinated to believe that they can be easily replaced by their leaders and that death can come at any moment.

So far, according to the ICC, the challenge when assessing the proportionality of a child soldier’s conduct is to determine whether the act was motivated by a genuine adoption of the abductors’ ideology (as was allegedly the case in Ongwen) or by a sense of self-preservation (as indicated by Judge Cassesse’s opinion in Erdemovic). It is submitted that this is a false dichotomy.  Abductors do not only punish non-compliance, but also any apparent hesitation by the child soldier (see, again, Grant, 2016). Sometimes, if the abductor thinks that the child is considering escape or disobedience, or is entertaining a similar desire, punishment can follow in order to subjugate the child’s spirit and disincentivize the mere thought of misconduct. Thus, in practice, external acceptance of the abductors’ ideology is necessary for the child’s preservation. As a result, whether the child soldier appears to have adopted the captor’s ideology is not dispositive; and whether self-preservation bleeds into true ideological conformity becomes almost impossible to determine abstractly and objectively.

In reality, when considering a defense under Article 31(1)(d) as applied to child soldiers, the question of proportionality has to be informed by Article 31(1)(a), to the extent that an intent to preserve oneself is necessarily affected by the accused’s mental capacity, especially when such capacity has been impaired by mistreatment and a constant sense of lethal danger (see below on concurrent defenses). Otherwise, the Court will artificially reject defenses of duress drawing on the child soldier’s apparent adoption of the captors’ violent ideology. It is only when the Court is able to consider the true mental state of the child soldier that the Court can consider whether his or her intent was to cause more harm that would have been caused if he or she refused to commit the crime.

Insanity and duress? Concurrent defenses

Counsel for Ongwen also raised a concurrent defense under Articles 31(1)(a) and (d), combining insanity and duress. In counsel’s words in the Closing Submissions (¶¶729, 730):

The Defence met its obligation to adduce some evidence to raise the defences of mental illness and duress. As discussed supra, the Prosecution has failed to disprove the defences beyond reasonable doubt. Although the two defences are distinct, in this case there is also a cumulative effect from the combination of lack of capacity to appreciate the unlawfulness of the conduct or to conform the conduct to the law and the extensive threats and coercion under which Mr Ongwen lived and acted.

Affirmative defences excuse or justify what is otherwise considered criminal conduct. The Defence submits that this case presents a compelling situation in which Mr Ongwen’s combined mental illness and duress should exclude him from criminal responsibility.

Trial Chamber IX, however, was not persuaded (¶2671):

…it may be observed that the two grounds for excluding criminal responsibility cannot coexist even in the abstract, given that one is premised on a destruction of the person’s capacity to appreciate the unlawfulness or nature of his or her conduct, or capacity to control his or her conduct to conform to the requirements of the law, and the other on a conscious choice to engage in conduct which constitutes a crime based on an evaluation of the harm that is caused.

This finding, in my view, is a glaring over-simplification of the issue. To analyze the compatibility of the defenses “in the abstract” betrays the very notion of insanity and duress as those terms are understood under Article 31(1): incapacity to appreciate unlawfulness or to control conduct cannot exist but in the specific circumstances of the accused, and so cannot the threat against the defendant and the fulfilment of the requirements of necessity, reasonableness and proportionality. Thus, to proclaim the incompatibility of the defenses in the abstract is contrary to the policy concern behind Article 31(1), which is that the accused’s conduct is influenced and ultimately determined by circumstances that escape his or her control. Those circumstances cannot be dismissed abstractly.

Moreover, while it is true that one defense is premised on the incapacity to assess lawfulness or control conduct (Article 31(1)(a)) and the other is premised on a voluntary choice that to some extent presupposes capacity (Article 31(1)(d)), the intersection between the two defenses is an analysis of the mental state of the accused at the time of the conduct. Particularly, as explained above, duress requires analysis of whether the defendant intended to cause a greater damage than the one sought to be avoided. Thus, even in the abstract, the defenses can reinforce each other. An accused that lacks the mental ability to adjust his conduct to the requirements of the law or to appreciate unlawfulness will be in a vastly different position to assess whether his acts create greater harm than the one sought to be avoided. Thus, the “cumulative effect” (in the words of counsel for Ongwen) of the two defenses can lead to an exclusion of criminal liability in the form of a concurrent defense. The factual similarities of the defenses, in this connection, were recognized by the Trial Chamber itself (¶2671).

It is submitted that the connective tissue between these two defenses exists, specifically, in the case of a child soldier that has been indoctrinated to follow orders blindly. The intent to cause harm under Article 31(1)(d) must be informed by whether the individual in question was in the position to make such a determination on the facts, as well as the circumstances of the choice. In the case of child soldiers, then, the existence and effect of a mental affliction caused by the horrific conduct of the abductors cannot be artificially separated from the child soldier’s intent to case greater harm than the one sought to be avoided. While Trial Chamber IX generally arrived at the correct solution (to convict Ongwen), the finding that duress and insanity under Article 31(1) are per se incompatible can have far-reaching effects for those under the influence of conscription.

How would a concurrent defense of duress and mental disease or defect materialize in practice? It is submitted that the defense is to be raised specifically under Article 31(1)(d), advocating for the use of subparagraph (a) as relevant context for interpretation (Article 31(1) of the VCLT). In applying Article 31(1)(d) to a child soldier, the incumbent Chamber must assess the mental state of the defendant and determine whether and how Article 31(1)(a) influences its determination on the existence of a threat (see above on subjective contributing factors) as well as the proportionality of the defendant’s conduct. The Court is free to make such a connection under Article 31(2) of the Statute.

Alternative defenses deriving from international law (Article 31(3))

At the confirmation of charges, counsel for Ongwen raised the argument that former child soldier status, in itself, excludes responsibility under international law. This defense was met with two hurdles. The first is that, as noted above, Article 31(3), which permits deriving defenses from the Court’s applicable law, only applies at trial. The second is that counsel for Ongwen did not supply sufficient legal basis to substantiate the defense (¶150 of the Pre-Trial Chamber’s decision):

The Defence has raised several times an argument that circumstances exist that exclude Dominic Ongwen’s individual criminal responsibility for the crimes that he may otherwise have committed. One side of this argument is that Dominic Ongwen, who was abducted into the LRA in 1987 at a young age and made a child soldier, should benefit from the international legal protection as child soldier up to the moment of his leaving of the LRA in January 2015, almost 30 years after his abduction, and that such protection should include, as a matter of law, an exclusion of individual criminal responsibility for the crimes under the Statute that he may have committed (…). However, this argument is entirely without legal basis, and the Chamber will not entertain it further.

Counsel for Ongwen had a second opportunity to raise this defense at trial (¶¶494-496 of the Closing Brief, above):

Article 4(3) of Additional Protocol (II) to the Geneva Conventions of August 12, 1949 provides that Children shall be provided with the care and aid they require and inter alia, those 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. Articles 39 and 6 of the Convention on the Rights of the Child and the Optional Protocol to the Convention on the Rights of the Child on the involvement of children in armed conflict respectively oblige the government to assist former child soldiers by providing “all appropriate assistance for their physical and psychological recovery and social reintegration.”

Counsel’s argument appears to be based on a tu quoque or “clean hands” reasoning: if Ongwen was not protected from conscription, his own acts of conscription should be not be considered. And if Uganda did not protect Ongwen from conscription, it should not be now allowed to assist in bringing criminal charges against him for the same crime. However, as the Chamber noted (¶2672), this argument continues to be without legal basis. No rule under international human rights law insulates a perpetrator from liability on account of the fact that he or she was a victim of that same crime.

Moreover, there are indeed good policy reasons not to adopt such an understanding. In practice, human rights violations breed more human rights violations. The only way to break the cycle is to assign criminal liability where it belongs. While the specific circumstances of a child soldier can be taken into account to exclude (Article 31 of the Statute) or mitigate (Rule 145(2)(a) of the Rules of Procedure and Evidence) responsibility, it should not be the position of any international tribunal that being a victim of a crime gives the victim a carte blanche to engage in the same criminal conduct. The purpose of international criminal law is not to claim revenge for gross human rights violations, but to put an end to them and allow for transitional justice, reconciliation and the reinstatement of the rule of law.

Closing thoughts

All in all, the Ongwen case leaves open many of the issues concerning the child soldier problem. The truth is that Dominic Ongwen is an imperfect vessel for this discussion. The evidence, as assessed by Trial Chamber IX, does indicate that Ongwen’s free will played a major role in engaging in the criminal conduct of which he was later convicted.

The bottom line is that the Court’s findings in The Prosecutor v. Dominic Ongwen, whether at the confirmation of charges or at trial, should not be taken as an all-out rejection of the diminished responsibility of child soldiers. International criminal law needs to assign the greatest responsibility with the persons that laid the groundwork for the underlying criminal acts: the persons, like Thomas Lubanga, who violated humanitarian law by abducting children and turning them into soldiers. The way of soundly assigning responsibility is by understanding the historical context of the armed conflict, the specific circumstances of the child soldier, and the overarching influence of the abductors, as well as the harm suffered by the victims of the former child soldier’s crimes. This can be achieved by applying Articles 31, 32 and 33 of the Rome Statute conscientiously, sensibly and realistically. If, all circumstances taken together, it appears that the former child soldier needs to bear a degree of responsibility (as was the case for Ongwen), then the criminal law needs to give effect to that distribution.

PIE DE IMPRENTA: Juan Pablo Hernández (editor-in-chief), Guatemala, 9 February 2021


Juan Pablo Hernández Páez
Juan Pablo Hernández Páez
Editor-in-Chief at The Treaty Examiner | juanhernandez@ufm.edu | Website | + posts

Juan Pablo is an international law enthusiast from Guatemala. He is a Moot Coach at Universidad Francisco Marroquín and the founder of The Treaty Examiner.