In addition to other grounds of criminal responsibility under this Statute for crimes within the jurisdiction of the Court:

1. A military commander or person effectively acting as a military commander shall be criminally responsible for crimes within the jurisdiction of the Court committed by forces under his or her effective command and control, or effective authority and control as the case may be, as a result of his or her failure to exercise control properly over such forces, where:

   1. That military commander or person either knew or, owing to the circumstances at the time, should have known that the forces were committing or about to commit such crimes; and

   2. That military commander or person failed to take all necessary and reasonable measures within his or her power to prevent or repress their commission or to submit the matter to the competent authorities for investigation and prosecution.9

In addition to the existing heated debates surrounding effective command or control and actual or constructive knowledge of the commander, this provision gives rise to a novel question—whether causation between a commander’s failure and the crimes committed by his subordinates is required as an element. The statutory language specifically states that a commander shall be criminally responsible “for crimes […] committed by forces under his or her effective command and control […] as a result of his or her failure to exercise control properly over such forces […].” Absent from the ICTY and ICTR Statutes,10 the phrase “as a result of” sparks deliberation as to whether causation is required as a new element in Article 28. To properly address this question, this comment first examines the nature and appropriate characterization of command responsibility.

II. Legal Character of Command Responsibility

The dichotomy is whether command responsibility is a separate offense or a form of liability. The commander can be punished for his own failure to intervene and control as well as for the crimes of their subordinates.11 Therefore, the doctrine seemingly creates both direct liability (for lack of supervision) and indirect liability (criminal acts of subordinates), and the theoretical differences bear practical importance in issues such as evidentiary standards,12 as well as on the inquiry of this comment.

A. Form of Liability—Accomplice and Endangerment

The view that command responsibility should be treated as a form of liability receives extensive scholarly support. It is important to note that during the evolution of the doctrine, the theory of strict liability has largely been rejected.13 One of the first and most famous (or notorious) cases for command responsibility post-WWII is the trial of General Yamashita, who was charged (and eventually executed) for “failing to discharge his duty as a commander to control the operation of the members of his command, permitting them to commit brutal atrocities and other high crimes.”14 The term “permitting” in the charge suggests that Yamashita was guilty simply because the atrocities took place, which resembles a form of strict liability where the commission of crimes constitutes per se proof that Yamashita’s failure to control.15 Though the Yamashita Court was widely considered to have adopted the strict liability standard, it did not fully do so, because it still devoted some attention to mens rea. The Court employed a presumption of knowledge by concluding that the accused “either knew or had the means of knowing” the commission of the crimes because of their widespread nature and notoriety of crimes.16

Wu and Kang explore the plausibility of command responsibility as a strict liability by analogizing to the RCO doctrine, which deals with U.S. prosecution of public welfare offenses under the Food, Drug, and Cosmetic Act.17 The RCO doctrine dispenses the conventional mens rea requirement in criminal conducts, and in the interest of public welfare, puts the burden of acting at hazard upon a person otherwise innocent but standing in responsible relation to a public danger.18 Similarly, the commander imposes risk on the unsuspecting public and deprive profits; it would therefore be fair to hold him liable for his omissions irrespective of mens rea, because he assumed the risks associated with his position.19 Nevertheless, the analogy is defective because international criminal law deals with gross moral wrongdoing instead of creating an incentive structure for neutral underlying acts, and the punishment is considerably higher.20

Even though not imputed or vicarious, the liability of the commander is still derivative in some sense.21 A popular interpretation of command responsibility is accomplice liability. In some United States jurisdictions and the Model Penal Code, the accomplice is punished to the same degree as the principal if the accomplice shares the mens rea as to the result element of the crime; other legal systems have treated accomplice liability as a separate crime and punished less severely.22 The deterrence effect of a lower mens rea effect is not unambiguously greater, because the scope of criminal law is less predictable from a subjective point of view, and deterrence is not effective when one cannot anticipate being held liable for their actions.23 As an attempt of compromise to reconcile the lack of enforceability of subjective mens rea standard and the over strictness of an objective one, the existing international law seems to adopt a reduced mens rea standard analogous to knowing facilitation—a knowledge that one’s behavior facilitates the criminal purpose.24 However, much ambiguity surrounds the “knew or should have known” standard.25

Bantekas regards command responsibility as a form of complicity by omission under the ICTY and ICTR Statutes, distinguished from positive and direct participation in the crime.26 Criminal complicity requires a willful and substantial contribution to the criminal offenses; command responsibility should have the same effect when it arises out of knowledge of the crimes of subordinates or gross negligence that contributed to their commission.27 Similarly, Judge Eboe-Osuji readily appreciates complicity as a basis of command responsibility in his concurring opinion in the Bemba appeal for accessory before the fact.28 The inculpation of the commander in his subordinates’ crimes stems from the fact that the commander helped to create the danger of the crimes, and voluntarily condoned the crimes when in a position to prevent them, so the personal nexus of the commander would be amply demonstrated.29

Robinson analyzes command responsibility as a mode of accessory liability.30 He believes that this is the correct understanding with overwhelming scholarly support and reflects the application over the history of international criminal law, as commanders have been charged and convicted as a party to the underlying crime of their subordinates.31 Robinson specifically rejects command responsibility as a separate crime of dereliction, arguing that this approach is not available under applicable international law based on the organization and structure of the instruments.32 The crimes are listed in Articles 2 through 5 of the ICTY Statute whereas command responsibility is listed in Article 7 which includes individual criminal responsibility, suggesting that the doctrine is a mode of liability.33 Article 28 of the Rome Statute explicitly designates command responsibility as a ground for liability for core crimes created by subordinates, because definitions of crimes are laid out in Part I of the Statute whereas Article 28 appears in Part III.34

On a similar ground, Judge Eboe-Osuji rejects the theory of separate offense of dereliction in his concurring opinion of the Bemba appeal.35 Comparing the text of the command responsibility provision in Article 86(2) of Protocol Additional I of Geneva Convention (Additional Protocol I),36 Judge Eboe-Osuji notes that Article 28 of the Rome Statute deviates in important ways—the Rome Statute speaks of “criminal responsibility for crimes” committed by the subordinates, and proactively imposes such criminal responsibility; whereas Additional Protocol I contemplates “penal or disciplinary responsibility” and does not in itself impose such responsibility.37 The impulse of dereliction of duty is rational in that it attempts to assuage the discomfort that the commander can be punished for the crimes of subordinates when they neither participated in the actus reus nor shared the primary mens rea of the crimes, shifting the emphasis to the commander’s own criminal liability.38 However, this framework is incompatible with Article 28 of the Rome Statute, which expressly requires that commander be held liable for crimes within the jurisdiction of the Court, and dereliction of duty is not among the crimes defined in Article 5 of the Rome Statute.39 Obviously, Judge Eboe-Osuji regards dereliction of duty as a separate crime rather than a theory to attribute blame in an adjective sense, because he argues that the then-hypothetical substantive crime either struggles to go beyond dereliction of duty itself or renders the theory meaningless.40 Judge Eboe-Osuji’s position can be read broadly as to reject the doctrine of command responsibility as a separate crime regardless of the theoretical rationale, by the same logic that accused can only be liable for “crimes within the jurisdiction of the Court.”

From Judge Eboe-Osuji’s perspective, the doctrine of command responsibility is ultimately anchored in endangerment liability.41 Concerns of criminal law extend beyond conducts that cause actual harm to those that create the threat or risk of harm.42 Endangerment liability is consistent with the mens rea requirement since one could only be held liable if one knowingly creates the risk of harm.43 The theory of endangerment strikes a balance between individual autonomy and common welfare, and is especially applicable to conduct with limited social utility and well-known risks.44 Warfare qualifies for such a situation with its notorious danger, and this angle of policy should guide the interpretation of Article 28.45 The Rome Statute imposes on the commander a duty to take care through proper supervision to ensure that the operations do not cause harm to innocent civilians.46 Again, the endangerment liability under Article 28 is attenuated and not strict, because it is result-regarding and fault-based.47

B. Separate Crime of Omission

In direct contrast with Judge Eboe-Osuji and Robinson, Kai Ambos argues against labeling command responsibility as a form of complicity liability. Instead, Article 28 of the Rome Statute can be characterized as a separate crime of omission, since it holds a commander liable for failure to control the subordinates on an objective level, but not for the crimes committed by the subordinates.48 The peculiar structure of the provision extends the commander’s mens rea beyond their own failure to supervise to the crimes of their subordinates.49 Conceptually, simultaneous liability for complicity and omission can only arise when the mental object is the same in respect to crimes committed by subordinates.50 This is usually not the case, since the mens rea in the case of command responsibility is the failure to control which is the main object of the offense.51 Conversely, in the case of complicity, the commander must share the intent of their subordinates, having mens rea in regards to the commander’s own contribution and to the main crime.52 This would render command responsibility subsidiary.53

Moreover, Ambos notes that Article 28 imposes a broad liability because the commander can be punished for a mere negligent (should have known) failure of supervision.54 Therefore, the commander cannot be considered as a mere accomplice, which requires intent.55 A negligent commander seems less blameworthy than were he to have actual knowledge and acted with intent, and a genuine question arises as to how can the intentional act of subordinates be imputed to the commander.56 In this scenario, Ambos considers mitigation of punishment,57 since it would be logical not to find the commander liable at all and the sole obstacle is the incompatibility with the statutory language in Article 28.58

Jackson advocates for a separate offense reading as well.59 He argues that the Bemba Court unnecessarily trapped itself to designate command responsibility as mode of liability in the starting point question, which leads to the unconvincing approaches to the issue of the causation requirement.60 To rebut Robinson’s argument, Jackson claims that nothing in the structure of the Rome Statute or the text of Article 28 establishes command responsibility as a mode of liability.61 While Article 28 appears in Part III of the Rome Statute (General Principles of Criminal Law) instead of Part II which sets out crimes within the jurisdiction of the Court, Part II is not an exhaustive list of those crimes.62 Part III contains substantive crime such as direct and public incitement to genocide.63 Similarly, Judge Eboe-Osuji’s reliance on the provision that the commander shall be criminally responsible for crimes within the jurisdiction of the Court is impeachable. Article 25(3) has the same language and it straightforwardly reads to mean responsibility for those crimes listed, such as incitement to genocide.64

Scholars that reject treating command responsibility as a separate crime nevertheless acknowledge some value in the approach, including evading problems associated with derivative liability.65 The commander can be liable for criminal neglect of superior’s duty, for which he is unambiguously guilty for as a principal instead of an accomplice.66 The separate crime approach would resolve the culpability issue, as the commander would not be held indirectly liable for crimes they did not contribute to.67 The lowering of the degree and the scope of mens rea needs to be balanced against the reduction of punishment imposed.68 But designating command responsibility as a separate crime is nevertheless a useful tool to the prosecutor, since they can capture failure that does not meet the requirement of accomplice liability and mens rea is usually the most contentious issue of the crime.69

Robinson does not outright disapprove of the separate offense approach and simply believes that this approach is not available under international criminal law. However, he suggests that national legislatures are free to adopt the doctrine of command responsibility as a separate offense.70 The opponents of the separate offense approach also have argued that dereliction would be insufficiently serious to warrant intervention of the ICC, which should focus on the most serious crimes of concern to the international community and those responsible for them.71 However, the commander’s putative failure should be sufficiently serious to warrant liability at the international level as a matter of policy choice, as they are best placed to ensure compliance with international law.72

III. Causation Requirement in Article 28

The concept of causation, at its barest generalities, describes the relationship between two events, in which the preceding event occasions a subsequent one.73 With unsettled philosophical questions, including whether the relationship is determinative (in the sense that the preceding event necessitates the subsequent one) or probabilistic, exclusive (in the sense that there are no other contributing factors) or merely conducive, the law rests on policy choices rather than applying a unitary system.74

A. Impact of Theoretical Differences on Causation

The theoretical lens through which command responsibility is examined has an important bearing on the question that this comment seeks to address—whether causation is required as an element for liability under Article 28 of the Rome Statute. While the dichotomy of separate offense and mode of liability does not present a conclusive answer, it is a starting point. Reasonings behind various theories facilitate the understanding of whether Article 28 has a causation requirement.

As a separate offense, the breach of command responsibility can be non-contributory in that the commander’s failure need not have contributed to the core crimes of the subordinates.75 Causal relationship is not required, or even that the core crimes occurred at all.76 While conceivable in national legislatures, international criminal law generally aims to punish gross wrongdoing and is usually result-based. Therefore, the non-contributory separate offense approach should still be under the premise that the core crimes have taken place. Alternatively, the separate offense can be construed as contributory. The underlying crimes of the subordinates serve as a point of reference of the commander’s failure of supervision, thus a causal relationship between the commander’s failure and the occurrence of the crimes must exist.77 The “but for” test should be inverted for omissions: the crimes would not have been committed had the commander properly supervised the subordinates, therefore, the omission “caused” the consequences.78 Despite a reluctance arising out of the principle of criminal law emphasizing the distinctive responsibility of one’s own actions, omissions can amount to causes when the defendant is under a specific duty not to cause harm in question derived from reasonable expectation or imposed by law.79

Under the framework of accomplice liability, causal contribution is required. The personal culpability principle dictates that one must contribute to a crime to be liable for it, and the requirement that the accused be causally linked to the crime itself is a fundamental principle of criminal law.80 The contribution of an accomplice can be indirect, such as influencing or assisting the voluntary acts of the principal.81 Requirement for accomplice contribution is more elastic than a strict but-for causation test, since the accomplice is held liable by virtue of his influence on the perpetrator and tracing causation through voluntary acts of others could be problematic.82 The extent of contribution is risk-aggravation—requiring only that commander’s omission increased the risk of commission of the crimes by the subordinates.83

Under endangerment liability, culpability does not always require that the conduct of the accused have caused actual harm, so long as it created a dangerous situation.84 The classic form of endangerment is non-consummate, as in actual harm need not have taken place.85 Modulated form of endangerment liability may be described as consummate,86 which is more compatible with international criminal law. Conceptually, in situations where the harm does occur, the theory of endangerment seems to suggest at least a contribution, if not a causal contribution between the defendant’s conduct and the actual harm.

B. Arguments in the Opinions of the Bemba Case and Appeal

The ICC issued its first ruling on command responsibility in 2016 in the case of Jean-Pierre Bemba,87 a Congolese politician and former rebel leader.88 Bemba was the president of Mouvement de Libération du Congo and commander of its military arm, and he sent troops to the aid of Ange-Félix Patassé, the democratically elected president of the Central African Republic, where his troops committed rapes, murders, and pillaging.89 Bemba was charged with crimes against humanity and war crime.90 The Pre-Trial Chamber offered some preliminary considerations of Article 28 in the confirmation of charges decision.91 Appearing to treat command responsibility as a mode of liability, the Chamber also noted that Article 28 expressly requires causal contribution—that the crimes be “a result of” a failure by the commander to exercise control properly.92 However, causation only applies in relation to “duty to prevent” and prospective crimes, and this is a partial denial of the causation requirement.93 As for the extent of the contribution, the Chamber took an risk-aggravation approach, noting that it is only required that the commander’s failure increased the risk of the commission of the crimes charged.94

On the novel issue of causation, the Trial Court requires some form of “personal nexus” between the individual and the crimes for which he is held liable as a core principle of criminal law.95 Similar to the Pre-Trial Chamber, the Trial Court also rejects the but-for test.96 Instead, it suggests that one standard that can satisfy the personal nexus is that:

[C]rimes would not have been committed, in the circumstances which they were, had the commander exercised control properly, or the commander exercising proper control would have prevented the crimes.97

The Court further explained that this standard is higher than what the law requires, but does not elaborate in light of the factual findings of the case.98 It seems that the Court’s standard focused on whether exercise of proper control could have made a difference, which is aligned with the risk-aggravation approach,99 and are more inclined to find causation in situations that the commander failed to prevent.

Bemba was acquitted of all charges in the appeal judgment in 2018,100 but the causation requirement remains unsettled as the judges are drastically divided on this issue. Even if causation is required as an element under Article 28, what is the proper standard? From a shared foundational understanding of command responsibility as a mode of liability, judges advanced arguments about the causation requirement on different grounds.101

The separate opinion of Judge Van den Wyngaert and Judge Morrison represents the first approach, claiming that Article 28 does not require causation at all.102 First and foremost, an argument raised by many is that the causation requirement does not stand from the standpoint of logic—it is simply not possible that an omission after a fact has occurred can cause this fact.103 Were causation a requirement under Article 28, a commander can never be responsible for a single crime or the first crime in a series of crimes.104 The separate opinion rejects the risk-aggravation approach, because the commander’s failure to take measures is a failure to reduce an existing risk and does not increase the risk—the risk simply stays on the same level.105

The logical incompatibility creates complications in failure to repress or submit the matter to competent authorities. Robinson cautions against treating the commander in these scenarios as an accessory after the fact.106 This early concept in criminal law stipulates that a person assisting a perpetrator to avoid arrest, trial, or conviction thereby becomes a party to the original felony.107 A parallel argument can be made for command responsibility, as failure to punish the subordinates or submit them to competent authorities facilitates their escape from justice.108 However, due to concerns of culpability, liberal systems have moved away from treating accessory after the fact as a mode of liability for the original crime but instead as a separate offense, such as obstruction of justice or harboring a fugitive.109 One cannot be held liable retroactively for an already completed crime that one never participated in or contributed to even in the broadest terms.110

The separate opinion seems to concede that there is no inherent incompatibility between the causation requirement and failure to punish, suggesting that it is theoretically possible that failure to punish an initial crime may increase the risk of subsequent crimes.111 However, the incompatibility is evident when applied to failure to submit the matter to competent authorities. They refuse the theory of culture of impunity which increases the risk of subsequent crimes,112 because not every failure to submit a subordinate suspected of criminal activity automatically induces future criminal conduct.113 It is necessary that the subsequent perpetrator knew about the initial crime and knew that the commander was also aware of the crime and deliberately failed to take any measures, and that this played a significant role in the perpetrator’s decision to commit the subsequent crime.114 Moreover, the time lag between initial crime and formal prosecutions by the competent authorities makes the commander’s failure to submit to have a causal effect in extremely few exceptions, rendering most command responsibility cases based on failure to submit under subparagraph (ii) out of scope of Article 28.115 This is an absurd result that contradicts the principle of effectiveness.116

In his concurring opinion, Judge Eboe-Osuji recognizes the intricacies of the causation requirement when applied to failure to refer the subordinates to competent authorities, a scenario that seemingly renders the separate offense approach more attractive.117 However, he believes that there may be instances that past failure to submit could result in subsequent commission of crimes, satisfying the causation requirement, and these instances cannot be so rare as to render subparagraph (ii) ineffective.118 The annals of warfare signifies the possibility that certain commanders would willfully refrain from having their subordinates persecuted, treating such violations as licenses to deserving soldiers, thus encouraging further crimes in that kind.119 The time lag between the initial and subsequent crimes and whether the commander is still in post should not matter in finding liability, since there is no statute of limitation for international crimes.120 All that matter is the phrase “as a result of,” which connects that subsequent crimes to the commander’s failure.121

The separate opinion’s reading of command responsibility as a mode of liability coupled with its total abandonment of the causation requirement may violate the fundamental principle of culpability in criminal law.122 This approach would mean that Article 28 entails the responsibility of a commander as a party of the crime absent any contribution, and it is especially problematic for the lower fault standard of “should have known” for military commanders.123 The statutory text indicates a negligence standard, and this approach risks transforming a negligent omission into intentional criminality of the most serious nature.124 The partial denial approach, adopted by the Pre-Trial Chamber that interprets causation to only apply to duty to prevent cases, is subject to the same contradiction with culpability.125

The dissenting opinion by Judge Monageng and Judge Hofmanski adopts a second approach, which Jackson refers to as the consequent crime approach.126 The dissent believes that causation is required in all scenarios, and attempts to resolve the tension by denying that failure to punish alone is sufficient to establish criminal liability—further crimes are required.127

First of all, the dissenting opinion raises a textual argument. The statutory language of Article 28 explicitly stipulates a nexus requirement of “as a result of,” which strongly indicates that commander’s failure caused the commission of crimes by his subordinates.128 ICTY’s rejection of causation does not pose a major obstacle in the interpretation of Article 28, because ICTY’s jurisprudence postdates the adoption of the Rome Statute and could not have influenced its drafting.129 As to the standard of causation, the dissenting opinion also rejects a strict but-for test, as it involves assessment of hypotheticals and cannot be determined with empirical precision.130 This is further complicated with unforeseen intervening acts and the end result being the intentional crime of subordinates, who are capable of making their own decisions, potentially against the orders of the commander.131 In the view of dissenting opinion, causation can be satisfied if there is a “demonstratable close link” between the commander’s omission and the crimes.132 Causation would be established if there is “a high probability” that had the commander discharged his duties, the crimes would have been prevented or would not have been committed in the manner that it was.133 The dissenting opinion seems to adopt the position of the Trial Court, drawing on and adapting from the language of personal nexus and the idea that the commander would have made a difference.

Judge Eboe-Osuji also agrees with the “consequent crime” approach, insisting on subsequent commission of crimes as the prerequisite for establishing a causal relationship.134 While not every failure can be regarded as having resulted in subordinates’ subsequent crimes, commander need not be held liable for every instance of the crime, but only when it can be shown beyond reasonable doubt that commander’s failure resulted in the commission of subsequent crimes.135 Robinson argues that the faulty logic incompatibility that failure to punish a crime cannot cause that same crime was a major reason early tribunals took a wrong turn with the doctrine by rejecting the causation requirement.136 For the extent of contribution, Judge Eboe-Osuji favors the standard that defendant’s causal contribution should be significant and cautions against confusing the term with substantial.137 Defendant’s contribution need only be more than negligible or not to be so minute that it will be ignored under the de minimis principle.138 If based on the characterization of command responsibility as a mode of liability and the assumption that risk aggravation test is applied stringently, the consequent crime approach may satisfy the principle of culpability.139

Where do the opinions in the Bemba appeal leave us with the ultimate question of whether causation should be required as an element under Article 28? In the starting question of the legal character of the doctrine, the Bemba Appeal Chamber took the unanimous position of treating command responsibility as a mode of liability.140 Two judges argue that causation should not be required while three judges believe that it should, adopting a consequent crime approach. As discussed in the theoretical approaches to the characterization of command responsibility, it seems that causation is always required if the doctrine is interpreted as a mode of liability because the issue of culpability cannot be circumvented. Therefore, the short answer is yes—causation is required, albeit with unsettled standards based on the Bemba appeal. The dissenting opinion argues for a high probability that the properly exercised control of the commander could have made a difference, while the concurring opinion adopts the risk-aggravation approach, namely the omission of the commander increased the risk of the commission of the crimes by the subordinates.

C. Alternative Reality: No Causation Requirement

That seems to conclude the story if the Bemba appeal judgment is the final chapter of command responsibility. However, the Bemba case is only the ICC’s first attempt to address command responsibility and will hardly be the last. There are still other possibilities, both to the legal character of the doctrine and to the causation requirement. This comment ponders the alternative reality and argues that causation should not be required under Article 28.

First, a revisit to the language of Article 28 suggests that there are multiple interpretations of the text.141 The proponents of the causation requirement claim that the phrase “as a result of his or her failure” explicitly requires that there be a causal relationship between the commander’s failure and the subordinate’s crimes. Based on this reading, the clause “as a result of” relates to the crimes.142 However, an alternative reading is also plausible—the clause relates to the commander’s responsibility, and simply states that the criminal responsibility results from the commander’s failure.143 This interpretation, put forward by Amnesty International as amicus curiae to the Pre-Trial Chamber in Bemba, has gathered some recognition including by the separate opinion of the Bemba appeal.144 Among the six authentic texts of the Rome Statute,145 four languages appear to allow either interpretation, while the Chinese and the French text favors the alternative reading that the clause relates to commander’s responsibility.146 The text of Article 28 itself does not conclusively demand a causation requirement. On the contrary, application of the rules on authentic language texts in Vienna Convention on the Law of Treaties suggests that causation is not required.147 Terms of the treaty are presumed to have the same meaning in each authentic text; when it is impossible to find a common meaning, the interpreter can resort to the best reconciliation.148 Here, a common meaning exists because all six texts can read as that the relevant clause relates to commander’s responsibility and suggest that no causation is required.

Second, the causation requirement poses some structural difficulties to the existing understanding of the doctrine. On appeal, Bemba argued that the Trial Court erroneously conflated the “measure” and the “causation” elements of Article 28.149 Does a causation analysis overlap with the assessment of whether the commander has taken all necessary and reasonable measures? As rebuttal, the dissenting opinion explains that in the context of omission, the causation analysis necessarily entails counterfactual conditions, as in what would or might have happened, had the commander retained proper control and taken the measures expected.150 The causation analysis is “intrinsically and inextricably” intertwined with the assessment of the adequacy of the measures that the commander took and the efficacy of measures that the commander failed to take.151 The dissenting opinion’s argument helps illustrate that causation analysis is encompassed in the measures assessment. More plainly put, the dissent argues that there is no erroneous conflation of the analysis because the two assessments are supposed to be conflated. The underlying logic of the measures assessment is whether the crimes would have been prevented or would not have been committed in the circumstances that they were had these measures been taken. This corresponds to the framework of causation analysis, regardless of whether the “high probability” approach or the risk aggravation approach is used. If a particular measure is unlikely to make a difference, it would fail the measure assessment as well as the causation analysis. The perceivable conflation of elements begs the question—what is the point of a separate causation requirement then, other than to satisfy the culpability principle?

The structural difficulties of the causation requirement extend to other provisions of the Rome Statute. With all the previous discussions on the framing of command responsibility as a mode of liability, one cannot help but notice the similarity between command responsibility and the liability for co-perpetrator. If causation was required, Article 28 could be redundant because any conduct that is punishable under Article 28 will almost always be more simply punishable under Article 25(3)(d)(i) as accomplice liability.152 The dissenting opinion rebuts that the mental elements differ, and mens rea required for Article 28 are significantly lower than for any form of liability under Article 25(3).153 While theoretically true, the lines are not so bright in practice, especially given the latest development in the Bemba appeal. The liability of the co-perpetrator requires objective elements of joint control, namely a common plan and the defendant’s essential contribution; as well as mens rea, which is the defendant’s awareness of the contribution and an intent that the physical perpetrator commits the crime.154 However, the ICC has watered down the intent element to the absolute minimum, which is satisfied if the defendant is aware of the risk that physical perpetrator will commit the crime and reconcile themselves to the risk or consent to it.155 On the other hand, the discussion on the knowledge element in the Bemba appeal suggests that mens rea required in command responsibility is potentially heightened. Knowledge under Article 28 has two standards: the “know” standard under which actual knowledge can be established through direct or circumstantial evidence and the “should have known” standard which entails constructive knowledge.156 The separate opinion of Wyngaert & Morrison claims that the two standards trigger different obligations on the part of the commander, and the question of what measures that the commander should have taken does not arise in the “should have known” standard.157 The separate opinion insists on the difference between actual knowledge and merely having information that put the commander on notice of the crimes,158 and the liability of the commander is further limited with the various requirements on qualifying information on its certainty and credibility. When the commander has genuine reasons to doubt the accuracy of the information, he may be absolved of command responsibility even if that information turns out to be true.159 Under this framing, it is generally harder to find that the commander has knowledge, and the knowledge element in command responsibility resembles the intent element in co-perpetrator liability. Even if they do not blend seamlessly, they do not present a stark contrast claimed by the dissenting opinion that the mens rea in the knowledge element is much lower.

Third, a dissection of Article 28 suggests that the logical inconsistency between retrospective duties and the causation requirement is hard to bypass. The text explicitly and unambiguously prescribes responsibility regarding crimes that have already been committed for failure to repress and failure to submit to competent authorities.160 Under Article 28, liability can arise for entirely retrospective failures, consistent with preceding international criminal instruments.161 None of the various approaches to causation requirement address this inconsistency in a satisfying manner. The Pre-Trial Chamber’s position that causation only applies to the duty to prevent directly tackles this inconsistency. However, it lacks textual support—why should it be the case that causation only applies to the duty to prevent, but not to the duty to repress and submit to competent authorities? Instead of trying to reconcile the causation requirement with retrospective duties, this position circumvents the conflict by claiming that the causation requirement does not apply to retrospective duties. Meanwhile, this position is a partial denial of the causation requirement, and it would run into the issue of culpability if under the mode of liability framework.

The consequent crime approach by the proponent of the causation requirement in the Bemba appeal also attempts to address this inconsistency, but it comes at great costs by essentially denying retrospective duties. Claiming that the causation requirement applies to all duties but can only be satisfied when the commander’s failure has contributed to subsequent crimes, this approach constitutes a castration of the doctrine by effectively carving out significant areas in which liability can arise. While both failure to repress and failure to submit to authorities technically can give rise to subsequent crimes and can be covered by this approach, there might be more scenarios that these failures are not followed by subsequent crimes.162 These scenarios would be left out of the Article 28, despite the fact that they were explicitly contemplated in the instruments of ICTY and ICTR.163 Also, this approach would leave out significant categories of international crimes such as isolated crimes and those committed towards the end of the conflict.164 Command responsibility will be seriously limited in terms of application and practice, and it would be surprising if this aligns with the drafters’ intentions.165 By eliminating retrospective duties, this approach also faces some structural difficulties. Failure to punish would collapse into failure to prevent, because conduct that can be characterized as failure to punish a preceding crime can nevertheless be captured by failure to prevent the subsequent crime.166

These approaches to the causation requirement either abandon causation in the context of retrospective duties or interpret retrospective duties out of Article 28. It is obvious at this point that retrospective duties and the causation requirement are simply irreconcilable. If only one notion can exist in the doctrine, it should be retrospective duties. This position is supported by textual interpretation, policy choice, and preceding legal history of the tribunal cases.

IV. Conclusion

The common meaning of the six authentic texts of the Rome Statute suggests that Article 28 does not require causation between the commander’s failure and the commission of crimes. Moreover, the causation requirement is unworkable because of the risk of conflation with other elements of Article 28 and other provisions of the Rome Statute and its incompatibility with the retrospective duties. An inclination that causation is not required under Article 28 warrants a revisit of the starting question of the legal character of command responsibility. While the issue of culpability cannot be avoided and causation must be required under the mode of liability framing, characterizing command responsibility as a separate offense brings new possibilities. As evident in the Bemba appeal, standards for both the knowledge element and the control element167 have been heightened, and it would be increasingly more difficult to hold the commander criminally liable under the doctrine in the future. Even if they were, the scenarios that can give rise to liability would be severely limited due to the causation requirement. To preserve some functionality and the essence of the doctrine, as well as to promote justice, the ICC should consider the alternative of analyzing command responsibility as a separate offense with reduced punishment. This alternative reality has significant value as it avoids the impossible balance between serious castration of command responsibility and the principle of culpability, and instead holds the commander liable solely for their own failures.

Endnotes — (click the footnote reference number, or ↩ symbol, to return to location in text).

1. 1.

   Rome Statute of the International Criminal Court, Adopted by the United Nations Diplomatic Conference of Plenipotentiaries on the Establishment of an International Criminal Court, Jul. 17, 1998, U.N. Doc. A/CONF.183/9, as amended [hereinafter Rome Statute], Art. 28, available online. ↩

2. 2.

   The Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08 A, Judgment on the appeal of Mr Jean-Pierre Bemba Gombo against Trial Chamber III’s “Judgment pursuant to Article 74 of the Statute” (ICC AC, Jun. 8, 2018) [hereinafter Bemba Appeals Chamber Judgment], available online. ↩

3. 3.

   E.g., Michael J. Sherman, Standards in Command Responsibility Prosecutions: How Strict, and Why?, 38 N. Ill. U. L. Rev. 298, 299 (Apr. 1, 2018), available online. ↩

4. 4.

   Id. at 301. ↩

5. 5.

   See Ilias Bantekas, The Contemporary Law of Superior Responsibility, 93 Am. J. Int’l L. 573, 573 (Jul. 1999), paywall, doi. ↩

6. 6.

   See Updated Statute of the International Criminal Tribunal for the Former Yugoslavia, S.C. Res. 827, Art. 7(3) (adopted May 25, 1993, as last amended Jul. 7, 2009) [hereinafter ICTY Statute], available online. ↩

7. 7.

   See Statute of the International Tribunal for Rwanda, S.C. Res. 955, Art. 6(3) (Nov. 8, 1994) [hereinafter ICTR Statute], available online. ↩

8. 8.

   Rome Statute, supra note 1, at Art. 28.

   (For the purpose and scope of this comment, Art. 28(b), which discusses the responsibility of “other superiors,” is left out of the discussion). ↩

9. 9.

   Id. ↩

10. 10.

    See ICTY Statute, supra note 6, at Art. 7(3).

    (“The fact that any of the acts referred to in Articles 2 to 5 of the present Statute was committed by a subordinate does not relieve his superior of criminal responsibility if he knew or had reason to know that the subordinate was about to commit such acts or had done so and the superior failed to take the necessary and reasonable measures to prevent such acts or to punish the perpetrators thereof.”).

    See ICTR Statute, supra note 7, at Art. 6(3).

    (“The fact that any of the acts referred to in Articles 2 to 4 of the present Statute was committed by a subordinate does not relieve his or her superior of criminal responsibility if he or she knew or had reason to know that the subordinate was about to commit such acts or had done so and the superior failed to take the necessary and reasonable measures to prevent such acts or to punish the perpetrators thereof.”). ↩

11. 11.

    Kai Ambos, Superior Responsibility, in The Rome Statute of the International Criminal Court: A Commentary 823, 824 (Antonio Cassese, Paola Gaeta, John R. W. D. Jones eds., 2002), available online, doi. ↩

12. 12.

    Id. ↩

13. 13.

    E.g., Bantekas, supra note 5, at 577. ↩

14. 14.

    Sherman, supra note 3, at 301. ↩

15. 15.

    Id. at 302. ↩

16. 16.

    Bantekas, supra note 5, at 588. ↩

17. 17.

    Timothy Wu & Yong-Sung Kang, Criminal Liability for the Actions of Subordinates: The Doctrine of Command Responsibility and its Analogues in United States Law, 38 Harv. Int’l L.J. 272, 279–80 (1997), available online. ↩

18. 18.

    Id. ↩

19. 19.

    Id. ↩

20. 20.

    Id. at 280–81. ↩

21. 21.

    Id. at 282. ↩

22. 22.

    Id. ↩

23. 23.

    Id. at 283. ↩

24. 24.

    Id. ↩

25. 25.

    E.g., id. at 284–85. ↩

26. 26.

    Bantekas, supra note 5, at 577. ↩

27. 27.

    Id. ↩

28. 28.

    The Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08-3636-Anx3, Concurring Separate Opinion of Judge Eboe-Osuji on the appeal of Mr. Jean-Pierre Bemba Gombo against Trial Chamber III’s “Judgment pursuant to Article 74 of the Statute,” ¶ 198 (ICC AC, Jun. 14, 2018) [hereinafter Concurring Opinion of Eboe-Osuji], available online. ↩

29. 29.

    Id. ¶¶ 198–200. ↩

30. 30.

    Darryl Robinson, How Command Responsibility Got So Complicated: A Culpability Contradiction, Its Obfuscation, and a Simple Solution, 13 Melb. J. Int’l L. 1, 8 (Oct. 28, 2011), available online. ↩

31. 31.

    Id. at 34. ↩

32. 32.

    Id. at 32. ↩

33. 33.

    Id.; ICTY Statute, supra note 6, at Arts. 2–5, 7. ↩

34. 34.

    Robinson, supra note 30, at 32. ↩

35. 35.

    Concurring Opinion of Eboe-Osuji, supra note 28, ¶¶ 188–197. ↩

36. 36.

    Protocol Additional to the Geneva Conventions of 12 August 1949, and Relating to the Protection of Victims of International Armed Conflicts, 1125 U.N.T.S. 3 (Jun. 8, 1977) [hereinafter Additional Protocol I], available online, archived. ↩

37. 37.

    Concurring Opinion of Eboe-Osuji, supra note 28, ¶¶ 189–190, quoting Additional Protocol I, supra note 36, at Art. 86(2)

    (“The fact that a breach of the Conventions or of this Protocol was committed by a subordinate does not absolve his superiors from penal or disciplinary responsibility, as the case may be, if they knew, or had information which should have enabled them to conclude in the circumstances at the time, that he was committing or was going to commit such a breach and if they did not take all feasible measures within their power to prevent or repress the breach.”) (emphasis added). ↩

38. 38.

    Id. ¶ 191. ↩

39. 39.

    Id. ¶ 195. ↩

40. 40.

    Id. ¶ 197. ↩

41. 41.

    Id. ¶ 232. ↩

42. 42.

    Id. ¶ 233. ↩

43. 43.

    Id. ↩

44. 44.

    Id. at 234. ↩

45. 45.

    Id. at 243–44. ↩

46. 46.

    Id. at 246. ↩

47. 47.

    Id. ↩

48. 48.

    Ambos, supra note 11, at 833. ↩

49. 49.

    Id. ↩

50. 50.

    Id. at 834–35. ↩

51. 51.

    Id. ↩

52. 52.

    Id. ↩

53. 53.

    Id. ↩

54. 54.

    Id. ↩

55. 55.

    Id. ↩

56. 56.

    Id. ↩

57. 57.

    See also Wu & Kang, supra note 17, at 289–90. ↩

58. 58.

    Ambos, supra note 11, at 835. ↩

59. 59.

    Miles Jackson, Causation and the Legal Character of Command Responsibility after Bemba at the International Criminal Court, 20 J. Int’l Crim. Just. 437, 448 (May 2022), available online, doi. ↩

60. 60.

    Id. ↩

61. 61.

    Id. at 453. ↩

62. 62.

    Id. at 452. ↩

63. 63.

    Id. ↩

64. 64.

    Id. ↩

65. 65.

    Wu & Kang, supra note 17, at 289–90; Robinson, supra note 30, at 31. ↩

66. 66.

    Wu & Kang, supra note 17, at 289. ↩

67. 67.

    Robinson, supra note 30, at 31. ↩

68. 68.

    Wu & Kang, supra note 17, at 290; Robinson, supra note 30, at 32. ↩

69. 69.

    Id. ↩

70. 70.

    Id. at 31–32. ↩

71. 71.

    Jackson, supra note 59, at 455. ↩

72. 72.

    Id. ↩

73. 73.

    Concurring Opinion of Eboe-Osuji, supra note 28, ¶ 157. ↩

74. 74.

    Id. ¶ 158–159. ↩

75. 75.

    See Robinson, supra note 30, at 31. ↩

76. 76.

    Id. ↩

77. 77.

    Ambos, supra note 11, at 833. ↩

78. 78.

    Id. at 842. ↩

79. 79.

    Concurring Opinion of Eboe-Osuji, supra note 28, ¶ 167–169; see also Robinson, supra note 30, at 43. ↩

80. 80.

    Robinson, supra note 30, at 12. ↩

81. 81.

    Id. at 13–14. ↩

82. 82.

    Id. ↩

83. 83.

    Id. at 44–45. ↩

84. 84.

    Concurring Opinion of Eboe-Osuji, supra note 28, ¶ 234. ↩

85. 85.

    Id. ¶ 237. ↩

86. 86.

    Id. ↩

87. 87.

    The Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08-3343, Judgment pursuant to Article 74 of the Statute (ICC TC III, Mar. 21, 2016) [hereinafter Bemba Trial Chamber Judgment], available online. ↩

88. 88.

    Sherman, supra note 3, at 325. ↩

89. 89.

    Id. ↩

90. 90.

    Robinson, supra note 30, at 53. ↩

91. 91.

    The Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08, Decision Pursuant to Article 61(7)(a) and (b) of the Rome Statute on the Charges of the Prosecutor Against Jean-Pierre Bemba Gombo (ICC PTC II, Jun. 15, 2009) [hereinafter Bemba Pre-Trial Decision], available online. ↩

92. 92.

    Robinson, supra note 30, at 54. ↩

93. 93.

    Jackson, supra note 59, at 444. ↩

94. 94.

    Id. ↩

95. 95.

    Bemba Trial Chamber Judgment, supra note 87, ¶ 211. ↩

96. 96.

    Id. ↩

97. 97.

    Sherman, supra note 3, at 325. ↩

98. 98.

    Bemba Trial Chamber Judgment, supra note 87, ¶ 213. ↩

99. 99.

    Robinson, supra note 30, at 44. ↩

100. 100.

     Bemba Appeals Chamber Judgment, supra note 2. ↩

101. 101.

     Jackson, supra note 59, at 438. ↩

102. 102.

     See id. at 433; The Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08-3636-Anx2, Separate Opinion of Judge Christine Van den Wyngaert and Judge Howard Morrison, ¶ 56 (ICC AC, Jun. 8, 2018) [hereinafter Separate Opinion of Wyngaert & Morrison], available online, archived. ↩

103. 103.

     See Separate Opinion of Wyngaert & Morrison, supra note 102, ¶ 52. ↩

104. 104.

     Id. ↩

105. 105.

     Id. ¶ 55. ↩

106. 106.

     Robinson, supra note 30, at 48. ↩

107. 107.

     Id. ↩

108. 108.

     Id. ↩

109. 109.

     Id. ↩

110. 110.

     Id. ↩

111. 111.

     See Separate Opinion of Wyngaert & Morrison, supra note 102, ¶ 58. ↩

112. 112.

     Id. ¶ 52. ↩

113. 113.

     Id. ¶ 53. ↩

114. 114.

     Id. ↩

115. 115.

     Id. ¶ 54. ↩

116. 116.

     Id. ↩

117. 117.

     Concurring Opinion of Eboe-Osuji, supra note 28, ¶ 203. ↩

118. 118.

     Id. ¶ 211. ↩

119. 119.

     Id. ↩

120. 120.

     Id. ¶¶ 213–214. ↩

121. 121.

     Id. ↩

122. 122.

     See Jackson, supra note 59, at 443. ↩

123. 123.

     Id. ↩

124. 124.

     Id. at 444. ↩

125. 125.

     Id. at 444–45. ↩

126. 126.

     See id. at 445; The Prosecutor v. Jean-Pierre Bemba Gombo, ICC-01/05-01/08-3636-Anx1-Red, Dissenting Opinion of Judge Sanji Mmasenono Monageng and Judge Piotr Hofmański, ¶ 20 (ICC AC, Jun. 8, 2018) [hereinafter Bemba Appeals Chamber Dissent], available online, archived. ↩

127. 127.

     See Jackson, supra note 59, at 445. ↩

128. 128.

     Bemba Appeals Chamber Dissent, supra note 126, ¶ 331. ↩

129. 129.

     Id. ¶ 335. ↩

130. 130.

     Id. ¶¶ 337–338. ↩

131. 131.

     Id. ↩

132. 132.

     Id. ¶ 339. ↩

133. 133.

     Id. ↩

134. 134.

     Concurring Opinion of Eboe-Osuji, supra note 28, ¶ 210. ↩

135. 135.

     Id. ↩

136. 136.

     Robinson, supra note 30, at 27. ↩

137. 137.

     Concurring Opinion of Eboe-Osuji, supra note 28, ¶ 166. ↩

138. 138.

     Id. ↩

139. 139.

     See Jackson, supra note 59, at 446. ↩

140. 140.

     Id. at 441. ↩

141. 141.

     Id. at 449. ↩

142. 142.

     Id. ↩

143. 143.

     Id.

     (“A military commander […] shall be criminally responsible for crimes […] committed by forces under his or her effective command and control […] as a result of his or her failure to exercise control properly over such forces […]”) (emphasis added). ↩

144. 144.

     Id.; see Separate Opinion of Wyngaert & Morrison, supra note 102, ¶ 51 n.40. ↩

145. 145.

     Rome Statute, supra note 1, at Art. 128.

     (“The original of this Statute, of which the Arabic, Chinese, English, French, Russian and Spanish texts are equally authentic, shall be deposited with the Secretary-General of the United Nations, who shall send certified copies thereof to all States.”). ↩

146. 146.

     See Jackson, supra note 59, at 449–50.

     (The relevant clause in the Chinese text “如果未对 […] 适当行使控制” translated to English, means “if he or she failed to exercise control properly over such forces”; in the French text, the relevant clause “lorsqu’il ou elle n’a pas exercé le contrôle” means “when he or she did not exercise control”). ↩

147. 147.

     Id. at 448; Vienna Convention on the Law of Treaties, Art. 33, May 23, 1969, 1155 U.N.T.S. 331, available online, archived. ↩

148. 148.

     Jackson, supra note 59 at 450. ↩

149. 149.

     Bemba Appeals Chamber Dissent, supra note 126, ¶ 341. ↩

150. 150.

     Id. ↩

151. 151.

     Id. ↩

152. 152.

     Rome Statute, supra note 1, at Art. 25(3)(d)(i).

     (“In any other way contributes to the commission or attempted commission of such a crime by a group of persons acting with a common purpose. Such contribution shall be intentional and shall either: (i) Be made with the aim of furthering the criminal activity or criminal purpose of the group, where such activity or purpose involves the commission of a crime within the jurisdiction of the Court […]”). ↩

153. 153.

     Bemba Appeals Chamber Dissent, supra note 126, ¶ 331. ↩

154. 154.

     See Jens David Ohlin, Joint Intentions to Commit International Crimes, Cornell L. Faculty Pub., Paper 169, 32 (Jan. 1, 2011), available online. ↩

155. 155.

     Id. ↩

156. 156.

     Bemba Appeals Chamber Dissent, supra note 126, ¶ 265. ↩

157. 157.

     See Separate Opinion of Wyngaert & Morrison, supra note 102, ¶ 40. ↩

158. 158.

     Id. ¶ 38. ↩

159. 159.

     Id. ¶ 43. ↩

160. 160.

     See Jackson, supra note 59, at 453. ↩

161. 161.

     Id. ↩

162. 162.

     See Section III(B) of this comment for elaboration: in the debate between the concurring opinion of Eboe-Osuji and the separate opinion of Wyngaert and Morrison. ↩

163. 163.

     compare Jackson, supra note 59, at 446. ↩

164. 164.

     Id. ↩

165. 165.

     Id. ↩

166. 166.

     Id. ↩

167. 167.

     Compare the notion of remote commander in the Bemba Appeals Chamber Judgment, supra note 2, ¶ 171. ↩