1.

The Chamber interprets the term ‘intentional’ in the crime against humanity of torture (Art. 7(2)(e)) as excluding ‘knowledge’ in the sense of Article 30(3).Footnote ⁶ It further finds it unnecessary to demonstrate the perpetrator's awareness as to the severity of the harm inflicted.Footnote ⁷ It considers that this interpretation ‘is consistent’ with paragraph 4 of the General Introduction to the Elements of Crimes,Footnote ⁸ which states that with regard to ‘elements involving value judgement’ it is not necessary that the perpetrator him- or herself completes the respective value judgement.Footnote ⁹ With these few phrases the Chamber, maybe unconsciously, opens up a Pandora's box. The first question is what the Chamber means if it states that the term ‘intentional’ excludes knowledge in the sense of Article 30(3). Does it understand ‘intentional’ in a purely volitional sense as will, desire or purpose? Or does it only want to exclude awareness with regard to a circumstance required by the offence's definition? The former interpretation conflicts with Article 30(2)(b) second alternative, according to which intent also encompasses awareness that a consequence ‘will occur in the ordinary course of events’. If the Chamber had wanted to exclude the cognitive intent element it would have had to exclude not only paragraph 3 but also paragraph 2(b) second alternative of Article 30. In addition, the awareness with regard to a consequence as embodied in paragraph 2(b) is, in an identical fashion, also contained in paragraph 3, and this makes the Chamber's statement even more confusing. For if the Chamber had wanted to exclude only awareness regarding circumstances it should have said so explicitly. If, on the other hand, it had wanted to exclude awareness (the cognitive side of intent) in general, it should have excluded also the second alternative of paragraph 2(b). Be that as it may, the Chamber's following statement regarding the absence of knowledge as to the severity of the harm seems to indicate that what it really wanted was to exclude the knowledge requirement as to a circumstance, in concreto as to the normative element (circumstance) of ‘severe’. The Chamber's reference to the Elements in this regard creates, however, the impression that it mixes up the general knowledge requirement (cognitive element of intent) and the specific problem of knowledge with regard to normative elements of the offence (called ‘elements involving value judgement’ in the Elements). The Chamber seems to overlook that it is one thing to require, as Article 30(3) inter alia does, awareness with regard to circumstances but quite another to redefine this awareness requirement with regard to normative elements. In other words, the specific problem of knowledge with regard to these elements cannot justify the general exclusion of the knowledge requirement with regard to all (including descriptive) elements.

If the Chamber indeed wanted to interpret the term ‘intentional’ in a purely volitional sense on the basis of its literal meaning, it needs to be said that this meaning is, to say the least, ambiguous.Footnote ¹⁰ While traditional common law knows specific intent crimes implying aim and purpose, for example burglary,Footnote ¹¹ the concepts of intent or intention were always understood in both a volitional and a cognitive sense.Footnote ¹² Modern English law still includes in the definition of intention, apart from purpose, ‘foresight of virtual certainty’; at best, the core meaning of intent or intention is reserved to desire, purpose, and so on.Footnote ¹³ Also, the US Model Penal Code which served as a reference for the ICC Statute in many regards, albeit distinguishing between ‘purpose’ and ‘knowledge’ (s. 2.02(a)), defines the former in a cognitive sense by referring to the perpetrator's ‘conscious object’ with regard to conduct and result.Footnote ¹⁴ Similarly in civil law jurisdictions the distinction between purpose and knowledge and thus the meaning of ‘intention’ is not always clear-cut. In French law, the expression ‘intention criminelle’ was never defined in the Code Pénal. In the scholarly literature, ‘intention’ is defined in both a volitional senseFootnote ¹⁵ and a cognitive sense.Footnote ¹⁶ On this basis, a distinction between the volitional dolus directus and the cognitive dolus indirectus is drawn.Footnote ¹⁷ In German and Spanish law the apparently clear-cut volitional tendency of dolus directus of first degree (‘dolus specialis’, ‘intención’, ‘Absicht’)Footnote ¹⁸ is by no means uncontroversial.Footnote ¹⁹

2.

Similar conceptual inconsistencies or imprecisions can be found in the Chamber's discussion of the mental element according to Article 30.Footnote ²⁰ First of all, the Chamber's understanding of mens rea as ‘a certain state of guilty mind’ or ‘the subjective elements’Footnote ²¹ is a purely naturalistic one limiting the mens rea to the psychological state of mind of the perpetrator at the moment of commission. There seems to be no room for normative considerations linked to the blameworthiness (in the sense of Vorwerfbarkeit or reprochabilidad) and, on this basis, culpability (in the sense of Schuld or culpabilidad) of the perpetrator's conduct. In fact, the Chamber seems to employ the term ‘culpability’ at a later stage in a purely psychological-naturalistic sense.Footnote ²² Thus it seems as if the Chamber is unaware of the theoretical distinction between a psychological and normative concept of culpability (guilt) which is so essential for a modern and fair theory of criminal law.Footnote ²³ Second, the Chamber's reference to the concept of dolus in connection with Article 30Footnote ²⁴ is, to say the least, unfortunate, since this concept includes, as correctly acknowledged by the Chamber,Footnote ²⁵dolus eventualis, and this very standard is later (together with recklessness and other lower standards) rejected.Footnote ²⁶

This indeed is the essence of the Chamber's discussion of Article 30. While I concur with the exclusion of dolus eventualis (contrary to the Lubanga Pre-Trial Chamber)Footnote ²⁷ and recklessness from Article 30 in the result – in fact, I have argued earlier that in case of dolus eventualis a perpetrator is not aware, as required by Article 30(2)(b), that a certain consequence will occur in the ordinary course of eventsFootnote ²⁸ and this applies a fortiori to recklessness, which also requires awareness as to the risk of a negative consequenceFootnote ²⁹ – some clarifications are needed. While the travaux confirm a restrictive approach as to Article 30,Footnote ³⁰ they are only a ‘supplementary means of interpretation’ (cf. Article 32 Vienna Convention on the Law of Treaties) and thus not decisive in the light of a clear or different literal interpretation. The literal interpretation, in turn, is predicated on the conceptual understanding of dolus eventualis. In this regard one must not overlook the fact that the ‘commonly agreed’ standard invoked by the ChamberFootnote ³¹ is by no means the only one. In fact, there are other, more cognitive concepts of dolus eventualis (requiring awareness or certainty as to a consequence)Footnote ³² and these may indeed be included in Article 30. For this reason it is a petitio principii if the Chamber justifies its restrictive reading of Article 30 with the lex stricta rule of Article 22(2)Footnote ³³ because the invocation of this rule against a broad reading of Article 30 presupposes what has to be demonstrated in the first place, namely that Article 30 has to be understood in a narrow (and for that purpose strict) sense.

In contrast, in the case of recklessness it is generally agreed that it stands between dolus eventualis and conscious negligence (bewußte Fahrlässigkeit);Footnote ³⁴ in cognitive terms, the reckless perpetrator possesses at most a ‘risk consciousness’ which, on the one hand, clearly distinguishes recklessness from negligenceFootnote ³⁵ but, on the other, must not be confused with the awareness requirement in the cognitive theories of dolus eventualis. For this reason it is not correct if the Chamber equates the so-called subjective recklessness (Cunningham recklessness in English law)Footnote ³⁶ with dolus eventualis;Footnote ³⁷ it may be closer to it than to conscious negligence, but it cannot be put on an equal footing with it.

3.

The other, maybe most important, issue of the decision concerns the correct mode of liability to be applied. The Chamber's substitution of the Prosecutor's co-perpetration by the command responsibility mode is convincing in the result but not completely satisfactory in its reasoning. As to co-perpetration the Chamber follows the ‘control over the crime’ theory,Footnote ³⁸ but then, quite surprisingly, only focuses on the subjective requirements, arguing that these have not been satisfied and therefore it is not necessary to examine the objective requirements.Footnote ³⁹ Such a bypassing of the objective requirements of a mode of liability (or, mutatis mutandis, of an offence definition) is only justified if the subjective requirements are beyond any doubt missing. This is clearly not the case here; in fact, as will be seen in the following, the subjective requirements of co-perpetration are quite controversial. The Chamber distinguishes between three subjective requirements which must be fulfilled cumulatively – that is, (i) the co-perpetrator's intent and knowledge as to the committed crimes, (ii) his awareness and acceptance as to the fulfilment of the material elements by the implementation of the common plan, and (iii) his awareness as to the factual circumstances of the joint control.Footnote ⁴⁰ While the first requirement is uncontroversial – clearly, the co-perpetrator, being a perpetrator, must himself act with intent and knowledge within the meaning of Article 30 as to the underlying crimes – the two other requirements deserve closer attention. While the second requirement – as the first – stems from both the Lubanga and Katanga/Chui confirmation decisions,Footnote ⁴¹ the third requirement is recognized unambiguously only in Lubanga,Footnote ⁴² and not in Katanga/Chui.Footnote ⁴³ In this decision this requirement is discussed only in relation to co-perpetration through another person – that is, indirect perpetration or perpetration by means – while Pre-Trial Chamber I clearly excludes it for ‘plain co-perpetration’.Footnote ⁴⁴ This is the correct view, since this (third) requirement demands too much from the co-perpetrator whose form of control is structurally different from the one of the indirect perpetrator: while the latter exercises control over the physical perpetrators and must be aware of this powerful position, the co-perpetrator only exercises control over the crime jointly with the other co-perpetrator(s). Thus, in the former case there exists a vertical relationship between the indirect perpetrator and the physical perpetrator, while in the latter this relationship is horizontal. There may be a combination of a joint control over physical perpetrators – that is, a mixed horizontal–vertical relationship – in cases where two or more superiors exercise joint control over physical perpetrators. This, in fact, has been the situation of the Katanga/Chui decision (Katanga and Chui as co-perpetrators through physical perpetrators) and it is also, structurally, the situation in Bemba, for one may consider Bemba and Patassé as co-perpetrators jointly acting through their subordinates as direct perpetrators.Footnote ⁴⁵

In casu, the Chamber finds that Bemba does not even fulfil the first subjective requirement – that is, was not aware that the crimes would be committed.Footnote ⁴⁶ While this may be the right decision in the light of the evidentiary standard of Article 61(7) (‘substantial grounds to believe’) and the available facts, and, indeed, it has not been appealed by the Prosecutor,Footnote ⁴⁷ it is difficult to reconcile with the Chamber's subsequent finding as to command responsibility, namely that Bemba ‘knew’ or even ‘actually knew’ that his troops were committing or about to commit the respective crimes.Footnote ⁴⁸ One wonders how it can be that one and the same person acts, on the one hand (as co-perpetrator), without knowledge and, on the other, with knowledge (as commander) with regard to the very same crimes. The Chamber does not ignore this logical impasse but its attempted way out by proposing a different knowledge standard for co-perpetration and command responsibility, arguing that the cognitive element as defined in Article 30 only applies to Article 25,Footnote ⁴⁹ is plainly incorrect and does not find any support in any (judicial or scholarly) authority. Indeed, Article 30 provides a general definition of the volitional and cognitive side of intent which is applicable to the ICC Statute as a whole ‘unless otherwise provided’ – that is, unless there is no different (lower) standard provided for (as, for example, the ‘should have known’ standard of Article 28).Footnote ⁵⁰ The fact that the subjective side of the superior's liability under Article 28 is different from the one under Article 25(3) in that it refers to the subordinates’ crimes (and not to his own crimes)Footnote ⁵¹ is irrelevant in this respect. Apart from that, the Chamber's affirmation in this context that in the case of Article 28 the superior does not participate in the commission of the crimeFootnote ⁵² is highly controversial and presupposes a profound analysis of the doctrinal structure of Article 28 which the Chamber fails to undertake.Footnote ⁵³

While the Chamber's third subjective requirement, as explained above, does not even find unequivocal support in the ICC's own jurisprudence and appears to set too high a standard for co-perpetration, the second requirement – awareness and acceptance as to the fulfilment of the material elements by the implementation of the common plan – is in fact a consequence of the mixed objective–subjective contents of the common plan requirement. While the ICC's case law so far has given this requirement a rather objective meaning,Footnote ⁵⁴ it also possesses clearly a subjective tendency, since the common plan is only the potential co-perpetrators’ decision (Tatentschluß) jointly to commit the crime. This decision only manifests itself in the external world with the implementation of the plan – that is, the joint (functional) commission of the envisaged crime. Before this joint commission, which constitutes in fact the second (objective) requirement of co-perpetration,Footnote ⁵⁵ the common plan belongs only to the inner world of the planners and thus constitutes an inchoate offence whose punishability presupposes a specific codification, for example as conspiracy or Verbrechensverabredung (§30(2) 3rd alternative German Strafgesetzbuch).

4.

As to the Chamber's on the whole convincing considerations on command responsibility I have only two minor objections. First, the Chamber convincingly affirms – in explicit contrast to the case law of the International Criminal Tribunal for the former Yugoslavia (ICTY)Footnote ⁵⁶ – that there must be some form of causality between the superior's failure of supervision and the subordinates’ underlying crimes.Footnote ⁵⁷ In the result, the Chamber follows the Risikoerhöhungstheorie (‘theory of risk aggravation or increase’, ‘théorie du risque aggravé’, ‘teoría del aumento del riesgo’, hereinafter ‘risk theory’ or ‘risk approach’), according to which it suffices that the commander's non-intervention increased the risk of the commission of the subordinates’ crimes.Footnote ⁵⁸ The Chamber comes to this solution after acknowledging the problems of causality in cases of omission.Footnote ⁵⁹ Indeed, it seems to see the risk theory as the – apparently logical (‘[T]herefore’) – solution to these problems. It also seems to view the risk approach as something completely different from a causality theory.Footnote ⁶⁰ I may misread the Chamber's considerations on this point, but I think it would have been less confusing if it had more clearly separated its (brief) considerations of the causality demonstration in cases of omission from the solution it ultimately adopts. For the fact that omissions cannot display ‘causal energy’ and therefore must be determined on the basis of a normative (instead of naturalistic) concept of causationFootnote ⁶¹ does not predetermine the concept of causation ultimately applied in these cases. Indeed, it is logically possible to apply an inverted conditio sine qua non or ‘but for test’; in fact, the Chamber formulates such a test but seems to consider it inapplicable in these cases.Footnote ⁶² Yet it is ultimately a policy question if one prefers the risk approach or a stricter causality test.Footnote ⁶³ In any case, the risk approach also constitutes a causality test in the sense that it implies that the increased risk was at least one of the causes of the harmful result.Footnote ⁶⁴ Clearly, though, it is easier to demonstrate that certain conduct – be it an act or an omission – increased the risk with a view to the production of a harmful result than that it directly caused such a result.

My second query refers to the Chamber's obiter-like statement that the ‘had reason to know’ criterion embodied in the Statutes of the ICTY, the International Criminal Tribunal for Rwanda (ICTR) and the Special Court for Sierra Leone (SCSL) is different from the ‘should have known’ standard under Article 28.Footnote ⁶⁵ In fact, it quite clearly follows from the travaux of the command responsibility provisions since the 1976 Additional Protocol I to the Four Geneva Conventions of 1949 that both formulations essentially constitute negligence standards.Footnote ⁶⁶ While, for example, the UN Secretary-General's Report about the establishment of the ICTY describes the ‘had reason to know’ standard as ‘criminal negligence’,Footnote ⁶⁷ the US Model Penal Code refers to ‘should have known’ in the context of negligence.Footnote ⁶⁸ If one really wants to read a difference in these two standards considering that the ‘should have known’ standard ‘goes one step below’ the ‘had reason to know’ standard,Footnote ⁶⁹ it would be the ICC's task to employ a restrictive interpretation which brings the former standard in line with the latter.Footnote ⁷⁰

5.

As to the practice of cumulative chargingFootnote ⁷¹ – that is, the multiple charging of the same conduct – the Chamber deems this only possible, following the ICTY's Celibici-test,Footnote ⁷² with regard to ‘distinct crimes’ – that is, if ‘each’ of the offences in question ‘requires at least one additional material element not contained in the other’.Footnote ⁷³ In turn, in terms of the doctrine of concurrence of offences (concursus delictorum,Footnote ⁷⁴concours de qualifications/d'infractions, concurso de leyes/delitos, Konkurrenzen),Footnote ⁷⁵ this means that cumulative charging is only admissible if there exists a ‘real’ or ‘true’ concurrence (concours idéal, concurso ideal, Idealkonkurrenz) – that is, a situation where the same conduct fulfils various distinct offences;Footnote ⁷⁶ this is to be distinguished from a ‘merger’ or an apparent or ‘false’ concurrence (concours apparent, concurso apparente, Gesetzeskonkurrenz/-einheit), where one offence (the ‘smaller’ offence) is completely contained in another (the ‘larger’ offence) – that is, it is subsumed by the larger offence or this offence ‘consumes’ the smaller offence (consumption or inclusion/speciality).Footnote ⁷⁷ This parallel to the doctrine of concours has only been drawn implicitly by the Chamber when considering that torture (as a crime against humanity) and outrages against personal dignity (as a war crime) are ‘fully subsumed’ by rape (as a crime against humanity),Footnote ⁷⁸ since this act requires, compared with torture, only one additional element, namely the act of penetration,Footnote ⁷⁹ and, compared with outrages, ‘in essence the constitutive elements of force or coercion’.Footnote ⁸⁰ While this is, in principle, correct,Footnote ⁸¹ the Chamber's considerations would have been understood better if it had explained that the theory of concours is the other side of the coin of cumulative charging.Footnote ⁸² In any case, while rape may be ‘the most appropriate legal characterization’Footnote ⁸³ in cases of torture with an (additional) act of penetration, the Prosecutor must charge torture for those acts where the element of penetration is lacking.Footnote ⁸⁴

The Chamber's restrictive approach to cumulative charging is to be welcomed. This practice, which is one of the common-law legacies of the ad hoc tribunals,Footnote ⁸⁵ blows up the prosecution case unnecessarily and creates a difficult situation for the defence. In fact, this practice is incompatible with the information and delimitation functions of charging rules,Footnote ⁸⁶ since it entails an imprecise ‘overcharging’ which makes it difficult, if not impossible, for the defence to prepare adequately a defence case (which?). While cumulative charging may be considered indispensable from a prosecution perspective if it runs the risk of ‘losing’ offences it has not (properly) charged in the first place, such a risk does not exist if the judge has the ultimate word on the ‘correct’ legal classification anyway – that is, in a system governed by the iura novit curia principle. This is the case of the ICC procedure, as will be explained in the following section.

6.

As to the Chamber's change of the mode of liability two questions regarding authority under Article 61(7)(c)(ii) arise. First, can the Pre-Trial Chamber proprio motu amend a charge of the Prosecutor at all? Second, in the affirmative, does such a power also extend to the change of a mode of liability as opposed to a change of a crime? Unfortunately, the Chamber ignores these questions and amends the mode of liability by judicial fiat. Interestingly, Pre-Trial Chamber III (sitting with two of the same judges as this Pre-Trial Chamber II, namely Judges Trendafilova and Kaul) correctly acknowledged earlier that Article 61(7)(c)(ii) – which authorizes the PTC only to ‘request the Prosecutor to consider’ amending a charge – ‘is formulated in a discretionary fashion, leaving it for the Prosecutor to decide whether to amend the relevant charge’ (subpara. (ii)).Footnote ⁸⁷ Further, the Chamber made it clear that ‘it does not purport to impinge upon the Prosecutor's functions as regards the formulation of the appropriate charges or to advise the Prosecutor on how best to prepare the document containing the charges’.Footnote ⁸⁸ While these are clear words, it is not clear whether Pre-Trial Chamber III refers exclusively to the factual basis of the charges or also to the legal characterization of these facts. If the latter were the case it could not have amended the mode of liability as it did in the decision under analysis. It would be too formalistic to argue that the Chamber did in fact not amend the charges in the sense of Article 61(7)(c) but only declined to confirm (subpara. (b)) the co-perpetration charge and instead confirmed (subpara. (a)) the command responsibility charge. Ultimately, with this double operation the Chamber changed the mode of liability and thus amended the charge in this respect.

Be that as it may, the authority of the Pre-Trial Chamber under Article 61(7) must be analysed from the perspective of the relationship between the Prosecutor and the Chambers under the ICC procedural system. I have argued elsewhere that the iura novit curia principle reflected in Regulation 55 of the Regulations of the Court, providing the Trial Chamber with a wide ‘modification competence’Footnote ⁸⁹ during the trial phase, should also be applied to the pre-trial phase.Footnote ⁹⁰ The defence rights could be secured by a new Rule 128 (4) of the Rules of Procedure and Evidence, providing for a sufficient notice for the defence and a possible adjournment of the hearing to give it sufficient time to prepare the case.Footnote ⁹¹ The reference to ‘sufficient evidence’ in various parts of paragraph 7 does not – contrary to the Prosecutor's viewFootnote ⁹² – speak against such a judicial modification competence, since this reference does not stand alone but must be read together with the references to the ‘crimes charged’ and the ‘charges’ encompassing the facts (evidence) and their legal qualification. Clearly, the Pre-Trial Chamber cannot assess the evidence on its own or in abstracto but only with a view to a certain legal qualification (offences, modes of liability). Indeed, subparagraph (c)(ii) of Article 61(7) links the evidence to ‘a different crime’.

If one accepts such a judicial modification competence, the second question – as to its scope – must be answered accordingly in favour of a broad authority of the Pre-Trial Chamber. While again the letter of the provision (Article 61(7)(ii): ‘different crime’) seems to call for a restrictive interpretation excluding any amendment going beyond a change of the respective criminal offences, the travaux are silent on the point,Footnote ⁹³ but a systematic and teleological interpretation suggests a broader approach. For if it is argued that the jura novit curiae principle and the corresponding Regulation 55 should also be applied at the Pre-Trial stage, it follows – in fact from the very text of Regulation 55 – that the Pre-Trial Chamber may also amend the legal classification as to ‘the form of participation of the accused under Articles 25 and 28’. Interestingly, Pre-Trial Chamber III had adopted the same view earlier, arguing that the mode of liability has ‘a bearing on the structure of the crime’ and that both ‘correlate to each other’.Footnote ⁹⁴ Yet while it is obvious that there is a relationship between the modes of liability (as part of the ‘General Part’) and the crimes (the ‘Special Part’) and that the former cannot be interpreted without the latter,Footnote ⁹⁵ this does not necessarily allow for stretching a procedural rule like Article 61(7)(ii) beyond its wording. Such a broad interpretation can, in my view, only be based on a procedural interpretation as set out above. Ultimately, it rests on the acceptance of the jura novit curiae principle for the ICC, and this should have been discussed by the Chamber.