The allegation that arms manufacturers in the Global North (who produce and export weapons ‘lawfully’ in accordance with national export regulations) are producing weapons that are used (unlawfully) against civilian populations in the Global South has come under elevated scrutiny in recent years. Some sectors of civil society have become ever more vocal about arms exports from the EU, the US, and other major manufacturing States that appear to facilitate atrocity, despite complying with their home state’s national export controls. How can a society claim respect for human rights at home, while producing weapons that fuel conflict and criminality abroad?
This old question is being re-examined in new legal contexts. The limitations of existing arms regulation under the Arms Trade Treaty, national export controls, and arms embargoes of the UN Security Council and EU in preventing weapons from reaching human rights abusers, has stimulated interest in other legal avenues for secondary responsibility under criminal law, State responsibility and tort law. Old assumptions that major manufacturers (BAE systems, Thales, Lockheed Martin, and others) are too powerful and too closely entwined with their State’s foreign policy interests to be genuinely regulated (satirised here by the British TV series Yes Minister in 1982) have been put under the spotlight. For the first time, those who doubt that an industry as lucrative as the arms trade can really be regulated are faced with the concrete reality of legal challenges to arms exports that have reached the appellate courts of manufacturing states, as well as unprecedented (albeit belated) shifts in arms export policies on humanitarian grounds. In this increasingly pluralistic field of global arms trade law, is there a potential role for the International Criminal Court (ICC) to hold an arms trader responsible under the Rome Statute’s modes of accomplice liability?
In these two blog posts for Rethinking SLIC*, I introduce my understanding of the legal basis for Rome Statute liability in a possible case before the ICC against an executive officer of an arms company on charges of complicity in genocide, war crimes, crimes against humanity or aggression. The other important legal issues that would likely arise in such a case are: (i) the attribution of responsibility to an individual executive for the corporation’s decision to carry out an arms export; (ii) challenges to the jurisdiction of the ICC where an arms export has been approved under national regulatory laws; and (iii) potentially relevant defences (grounds for exclusion of responsibility under Article 31 and 32 of the Rome Statute). I will return in later blog posts to these questions, noting that the limited commentary on the ICC and the arms trade has tended to focus more on questions of liability.
With a view to contributing to this existing discussion on liability, I argue that, in an arms trade case, the implications of Article 25(3)(d) of the Rome Statute still remain under-represented, or under-recognised, in debates about Rome Statute liability, both inside and outside the ICC.
Firstly, inside the Court, sophisticated attempts to read-down the requirement of ‘purpose’ in Article 25(3)(c) may have limited practical significance for actual proceedings in concrete cases. Secondly, in fora outside the ICC, the ‘for the purpose of’ requirement of aiding or abetting under Article 25(3)(c) is often wrongly treated as if it were the sole mens rea for complicity under the Rome Statute, perhaps due to a lack of familiarity with the ICC’s sui generis legal framework. Many such interpretations have failed to recognise, as Judge Pregerson did in the Alien Tort Statute case of Rio Tinto, that ‘even if the Rome Statute were an appropriate source for determining the mens rea standard for aiding and abetting under customary international law, Article 25(3)(d)’s knowledge standard would apply’.
Two debates on Rome Statute complicit assistance
We could say there are two parallel but inter-related debates going on in which the Rome Statute’s complicity provisions are being interpreted. The first asks, inside the ICC, what is required to be convicted as an accomplice under the Rome Statute. The second debate assesses the value of Rome Statute requirements as sources of law for external questions of interpretation, such as in Alien Tort Statute proceedings. The importance of the Article 25(3)(c) ‘purpose’ requirement may be over-stated in both of these debates, due to the presence of the other mode of complicit assistance in Article 25(3)(d). Shifting the focus away from the ‘purpose’ requirement would shed more light on how the ICC’s complicity provisions interact with the rules of complicity in other legal jurisdictions.
Instead of expending interpretive effort on contorting the ordinary meaning of the ‘purpose’ requirement of Article 25(3)(c), ICC jurisprudence and external assessments of the Rome Statute could interpret the modes of liability in Articles 25(3)(c), (d)(i) and (d)(ii) as providing three different labels for complicit assistance in the ICC’s sui generis legal system. The lowest mens rea requirement of complicit assistance in the Rome Statute would therefore be understood as simple knowledge of the intention of the group to commit the crime (under (d)(ii)). Inside the ICC, this understanding of the Statute would acknowledge that the Prosecutor, practically speaking, is always likely to rely on the lower requirements of this provision in their cases against accomplices charged with providing assistance. Outside the ICC, there would be no further controversy as to whether complicit assistance under the Rome Statute imposes a higher ‘purposive’ mens rea than aiding and abetting under customary international law. Both of these sources of law would be understood as imposing a mens rea for general complicity of simple knowledge of the intention of the group to commit the crime.
Adopting a concept of Rome Statute complicit assistance that recognises the ordinary meaning of the terms in Articles 25(3)(c) and (d), would echo some of the theoretical underpinnings of a unitary theory of perpetration, without imposing such a theory, which as Miles Jackson has explained, would involve the undesirable abolition of the distinction between principals and accomplices.
Is there a legal basis for the ICC to indict executives of major arms manufacturing corporations?
The first question of substantive law for an ICC case against an executive of a major arms manufacturing corporation is whether the Rome Statute, as the primary applicable source of law for ICC proceedings, provides a legal basis to hold an individual corporate official responsible for their involvement in the exportation of weapons that contribute to international crimes. What are the outer limits of secondary liability in Article 25 of the Statute, and what types of conduct typically carried out by the executive officers of a major arms manufacturers would fall within those limits, and therefore be deemed criminal?
Firstly, the Rome Statute does not provide for liability of the legal person of a corporation that manufactures and distributes weapons. Article 25(1) of the Statute provides ‘the Court shall have jurisdiction over natural persons pursuant to this Statute’. As such, an ICC case would be restricted to charging the individual directors of a company and not the legal person of the corporation itself. The fact that the ICC Prosecutor cannot investigate cases against a corporation, differentiates the potential of ICC proceedings from that of national criminal prosecutions such as the Lafarge case in France, wherein not only the Lafarge corporation and its subsidiary were indicted, but also eight former executives. In Lafarge, both juridical and natural persons were indicted for their secondary responsibility for Islamic State’s crimes in Syria, on the basis that the company had sought to continue its cement production by, inter alia, making significant financial payments to members of the Islamic State.
In contrast to corporate liability in domestic jurisdictions such as France, a significant feature of an ICC case against a business actor would be the Court’s inability to make reparations orders against the corporation itself. It has been suggested that Rules 94 and 95 of the ICC’s Rules of Procedure and Evidence could be interpreted to permit a victim to request that a corporate person be called at the reparations stage of ICC as an ‘impleader’, although reparations orders of this nature would likely require statutory amendment to the Rome Statute in order to be effective. A further possibility would be that an individual conviction at the ICC of a corporate executive might trigger satellite criminal and/or civil proceedings at the national level to obtain reparations or damages from the corporation itself.
Putting aside the lack of legal basis for corporate liability at the ICC, it is widely recognised that Article 25 of the Rome Statute provides a legal basis for the Prosecutor to investigate and prosecute a corporate official in their individual capacity, and to attribute to them business activities of the corporation. The notion societas delinquere non potest (‘companies cannot commit an offence’) fits well with the foundational premise of international criminal law: individuals have direct international law obligations. The legal principle that individuals are responsible for their actions, whether they take place in a commercial context or not, has been applied since the post-World War II cases of Flick, Krupp, and IG Farben as well as the ‘Zyklon B’ case of Tesch.
As such, senior corporate officers – CEOs, Presidents, COEs, COOs, CIOs, CTOs, CMOs, CROs and CFOs – any individual in an executive position, may attract individual liability for the broader activities of the corporation in which they play a role. An arms trade executive could, in principle, be held responsible for their own actions and omissions in the corporate organisational hierarchies in which they operate, and the corporation does not, per se, provide a shield to individual responsibility. As mentioned above, this does not bear on whether a State’s authorisation of the corporation’s conduct under national regulatory laws provides its executives with a shield from international criminal responsibility, either as a matter of jurisdiction or a ground for exclusion of responsibility (a defence).
Executives of Arms Manufacturers under the Rome Statute for ‘Aiding or Abetting’
There is legal uncertainty over the substantive requirements of ICC accomplice liability that would apply to an arms trade case. To recall, the sui generis Rome Statute legal system provides, in addition to its modes of principal liability, modes of accomplice liability of ‘aiding or abetting’ under subsection (c) and ‘knowingly contributing’ under subsection (d).
In full, Article 25(3)(c) provides that a person shall be criminally responsible and liable for punishment for a crime within the jurisdiction of the Court if that person:
(c) For the purpose of facilitating the commission of such a crime, aids, abets
or otherwise assists in its commission or its attempted commission,
including providing the means for its commission;
This form of complicit assistance is a potential avenue for the liability of an arms trader who, for the purpose of facilitating the commission of a crime, aids, abets or otherwise assists in its commission or its attempted commission.
The clause that reads ‘providing the means for its commission’ might appear, at first glance, as a particularly appropriate descriptor for allegations of arms traders’ complicity. However, this is merely an illustrative clause and does not add anything to the legal requirements (in my view, this clause is not strictly necessary from a legal drafting perspective since it is subsumed by the expansive ‘otherwise assists’ requirement). In any case, this clause is, like the rest of the provision, subject to the ‘purpose’ requirement.
The mens rea requirement ‘for the purpose of facilitating’ is a ‘game-changer’ for the ICC Prosecutor and, as Elies van Sliedregt and Alex Popova recognise, the requirement cannot be ‘interpreted away, into non-existence.’ It imposes a ‘specific subjective requirement stricter than mere knowledge’ as Kai Ambos puts it. While the ordinary meaning of ‘purpose’ has often been viewed as a troubling aspect of the Rome Statute, the elevated mens rea makes more sense when read together with the relatively decreased actus reus, which does not require a ‘substantial contribution’ to the crime, as was required for aiding and abetting at the ad hoc tribunals.
Most ICC case law has followed the ordinary meaning of the Statute. The Mbarushimana Pre-Trial Chamber held that ‘unlike the jurisprudence of the ad hoc tribunals, Article 25(3)(c) of the Statute requires that the person act with the purpose to facilitate the crime; knowledge is not enough for responsibility under this Article.’ Subsequent Chambers of the Court followed this literal interpretation of the ordinary meaning of the Statute or did not see a need to comment on it (Bemba et al Confirmation, Blé Goudé Confirmation, Bemba et al Appeals, Ongwen Confirmation, Al Mahdi Confirmation).
In Bemba and others, Trial Chamber VII (Judge Schmitt, Judge Perrin de Brichambaut and Judge Pangalangan) took a divergent view (paragraph 97 of the Trial Judgment). Trial Chamber VII found that the ‘purpose’ in Article 25(3)(c) ‘for the purpose of facilitating the commission of such crime’ does not relate to the facilitation of the commission of the crime; it relates only to the facilitation itself. Trial Chamber VII gave no legal reason for departing from the ordinary meaning of these words:
Unlike other international instruments, Article 25(3)(c) of the Statute expressly sets forth a specific ‘purpose’ requirement according to which the assistant must act (‘for the purpose of facilitating the commission of such crime’). This wording introduces a higher subjective mental element and means that the accessory must have lent his or her assistance with the aim of facilitating the offence. It is not sufficient that the accessory merely knows that his or her conduct will assist the principal perpetrator in the commission of the offence. Mindful of the twofold intent of the accessory (viz. firstly, the principal offence and, secondly, the accessory’s own conduct), the Chamber clarifies that this elevated subjective standard relates to the accessory’s facilitation, not the principal offence.
The judicial finding of Trial Chamber VII in this single sentence (in a 458 page judgment) contains no reference to legal sources (the enigmatic reference to ‘other international instruments’ earlier in the paragraph does not assist, since it footnotes the Mbarushimana Confirmation Decision (paragraphs 274 and 281) which cites the jurisprudence of the ad hoc tribunals. We are left to assume that Trial Chamber VII had in mind the ICTY and ICTR Statutes as ‘international instruments’).
On this limited legal basis, Trial Chamber VII departed from the existing case law on a central requirement of the ICC’s legal framework on complicity. Since Article 31 of the Vienna Convention on the Law of Treaties requires that statutory wording is given its “ordinary meaning” in light of the treaty’s object and purpose, Trial Chamber VII’s finding cried out for proper legal reasoning to justify its departure from the ordinary meaning of ‘[providing assistance] for the purpose of facilitating the commission of the crime’. A natural rewording of this phrase would be: ‘to help someone else for the reason that doing so will make it easier for them to carry out the crime’. The accomplice must intentionally provide whatever it is that facilitates the principal perpetrator’s offence, intending that what he provides will facilitate the principal perpetrator’s commission of the offence.
Application of Article 30 mental requirements
A further piece of challenging legal interpretation in Bemba and others arises from the next paragraph (paragraph 98 of the Trial Judgment). Breaking with the majority of judicial and academic interpretations of Article 25(3)(c), Trial Chamber VII usurps the mens rea ‘for the purpose of facilitating’ in relation to the criminal offence and replaces it with the default Article 30 mental requirement of mere knowledge (awareness that the crime ‘will occur in the ordinary course of events’). This is controversial because Article 30 only applies to parts of the Rome Statute that do not provide their own mental element (it is applicable ‘unless otherwise provided’). Clearly, however you interpret the mental element of the purposive language in Article 25(3)(c), the provision does provide some form of mental element that relates to the purpose of facilitating the commission of the crime. The default mens rea in Article 30 is therefore displaced. Trial Chamber VII’s finding, contrario, is that the purposive language relates only to the accessory’s facilitation, such that the Article 30 mens rea, rather than the Article 25(3)(c) mens rea, must be applied to the mental element of intent ‘with regard to the principal offence’.
Whether you accept Trial Chamber VII’s finding on the applicability of Article 30 to Article 25(3)(c) therefore depends to a large extent on whether you find the previous paragraph’s reasoning convincing. Can the purposive language be read in any logical sense as relating only to the accomplice’s own conduct? If it can, is that reading in accordance with the ordinary meaning of the text? Or, is this a teleological construction that stretches the words beyond their ordinary meaning to achieve a certain interpretative outcome? In my view, there is no coherent way for a lawyer to read the words ‘for the purpose of facilitating the commission of the crime’ other than in relation to the purpose of bringing about a criminal outcome. Although an enormous amount of scholarly effort has been devoted to unpicking the wording and providing alternative interpretations, these are really normative arguments that could be directed at the Assembly of States Parties to encourage them to amend Article 25(3)(c) by removing the purposive requirement. Lastly on this point, Trial Chamber VII’s legal finding was not subject to appellate scrutiny, leaving doubts as to whether it would pass muster with an ICC Appeals Chamber.
‘Interpreting away, into non-existence’?
The lack of sound legal reasoning in Trial Chamber VII’s finding is damaging for the ICC because it leaves the Bemba and others Trial judgment open to the criticism that Trial Chamber VII sought to do exactly what Van Sliedregt and Popova warned against: to ‘interpret away, into non-existence’ the ordinary meaning of the Rome Statute. Although ostensibly asserting a different rendering of ‘purpose’, in actuality, Trial Chamber VII reads it down into nothingness. In the context of Article 25(3)(c), relating the ‘purpose’ requirement solely to the assistant’s facilitation would render the requirement obsolete, since there is no question that the facilitative act must be carried out intentionally (i.e. not accidentally). Under Trial Chamber VII’s view, it is hard to see why the drafters would have included the ‘purpose’ wording. To the contrary, the drafting history suggests that ‘purpose’ was deliberately included and the delegations did not appear to have in mind an interpretation akin to Trial Chamber VII’s.
The discourse on the ‘purpose’ requirement in Article 25(3)(c) proves there is no end to the linguistic gymnastics that could be performed around the Rome Statute’s text. But this may be one of those questions of legal interpretation that is not as complicated as is sometimes made out. Although it has been argued that a literal interpretation in accordance Article 31 of the Vienna Convention is insufficient (thus warranting recourse to Article 32 of the Convention supplementary means of interpretation) it is hard to understand why ‘purpose’ is so controversial as a matter of legal construction (as developed below, it may be easier to understand why the specificity of ICC aiding and abetting mens rea is often perceived as controversial from a normative perspective). In short, the justifications for resorting to analyses of the full structure of the US Model Penal Code and the negotiating history of the ICC standard are not clear.
By artificially linking the elevated subjective standard to the accessory’s facilitation only, and not the principal offence, the Chamber took a major interpretive step without providing any legal reasoning. Not for the first time, commentators on the ICC might be disappointed that, in seeking to reinterpret the Statute, the judges did not engage with the vast trove of existing research and analysis accumulated over the last two decades, which has considered the interpretation of this provision in depth, including the two-fold intent requirement, in fact providing arguments that could have supported Trial Chamber VII in explaining and reasoning the finding. James G Stewart’s blog symposia on the topic and the comments thereon set out the opposing views and responses, including on two-fold intent, and engagement with these arguments could have legitimated Trial Chamber VII’s decision and enriched the communication of its legal reasoning to the public.
‘Two-fold intent’ in aiding or abetting
As well as lacking legal reasoning, it is not clear what the judicial bench had in mind (recall that Trial Chamber VII says that the elevated mens rea ‘for the purpose of facilitating’ relates to the accessory’s facilitation as such, not the principal offence).
Relating the elevated subjective standard to the accessory’s own act of facilitation, not the principal offence, could mean either or both of two things. It could mean, firstly, that the defendant is required to intend to perform the act capable of assisting (i.e. it is not accidental). This requirement would not be at issue for an executive of an arms manufacturer – there is no possibility that a shipment of weapons was accidental. Secondly, Trial Chamber VII could have been referring to a requirement that the defendant intended the act to be of assistance. This would mean that the ‘purpose’ requirement does not need to relate to any criminality, and it would be sufficient for the defendant to intend that their assistance generally assists in some way. In the arms trade context, this would mean that a corporate executive would often have sufficient mens rea – irrespective of whether they know of or intend criminal activities, it is always the case that, by selling weapons, they intend to provide objects (weapons) that will be useful to the recipient.
Two-fold intent, or dual intent, thus recognises that in addition to the primary aspect of mens rea (when you passed him the knife, did you mean the crime to happen, or just know that it was going to happen?), there will always be a secondary aspect of mens rea that concerns whether the accomplice intended their own actions (did you mean to pass the knife, or did you merely leave it on a table where you knew he could pick it up? If you did mean to give him the knife, did you intend that it would help him in some way?).
However, the problem with Trial Chamber VII’s reading of two-fold intent is that the plain text of Article 25(3)(c) connects the ‘purpose’ to ‘the commission of the crime’. It is not sustainable to read these words as if they relate only to the facilitation of lawful/non-criminal/innocent activities.
Two-fold intent is an understandable distinction, which quickly falls away when one tries to unravel Trial Chamber VII’s elision of the accessory’s intentions as to their own conduct with the requirement ‘for the purpose of facilitating the commission of the crime’. It is unclear how, in Trial Chamber VII’s understanding, the ordinary wording of Article 25(3)(c) can be contorted to apply only to the ‘accessory’s facilitation, not the principal offence’. It has been suggested that ‘the literal interpretation of the ICC Statute’s complicity provision is ambiguous’, but it is hard to understand why such legal construction of this phrase is thought to be necessary in going beyond the plain meaning of the phrase. The requirement that the aider and abettor must act for the purpose of facilitating the commission of the crime appears to inevitably relate the mens rea of the aider or abettor to the commission of the crime.
The relevance of dual-intent in this context has been illuminated by Thomas Weigend as follows: ‘The Statute speaks of “the purpose of facilitating the commission of such a crime”; the assistant’s purpose thus is not the crime but the facilitation. This means that the assistant’s objective must be to facilitate the act of the main perpetrator; but her will need not encompass the result of the perpetrator’s conduct.’ Although Weigend might initially appear to sever the connection between the assistant’s mens rea and the crime, his work should be understood as differentiating the purpose requirement’s relationship to the crime in terms of conduct and results. Weigend is saying that the accomplice must intend to facilitate the principal perpetrator’s criminal conduct (‘assist/enable/allow/enhance/make easier’ their conduct) but not necessarily the criminal result.
Weigend continues with the example of an arms trader: ‘if an arms trader sells weapons to a dictator, he will be punishable only if he does so with the purpose of facilitating the dictator’s use of armed force; but the fact that the armed force will be used against unarmed civilians and will therefore constitute a crime against humanity need not be the arms dealer’s “purpose” (although he needs to know about that particular use in order to be liable as an assistant under Art. 30 of the ICC Statute).’ If we accept the literal formulation that, under (c), facilitating the commission of the crime inevitably relates the mens rea of the aider or abettor to the crime, Weigend encourages us to consider whether the requirement relates to the conduct of the crime or the result of the crime. However, in both cases this requires from the accomplice some degree of mental commitment to the crime.
In this understanding of ‘dual-intent’, although there is a distinction between the accomplice’s purposes as to the criminal result, and their purposes as to their own facilitation of the criminal act, the literal interpretation of Article 25(3)(c) remains the same: the accomplice must intend that what she provides will facilitate the principal perpetrator’s commission of the crime. This is not, of course, the same thing as the provider of assistance simply knowing that the perpetrator will commit the offence in the ordinary course of events.
Further gymnastic attempts have been made to linguistically deconstruct the word ‘purpose’ but this seems a bridge too far. The ordinary meaning in the OED of an individual having purpose is to ‘intend, plan, resolve’ / ‘To have a purpose, plan, or design‘/ ‘That which a person sets out to do or attain; an object in view; a determined intention or aim’ / ‘The reason for which something is done or made, or for which it exists; the result or effect intended or sought; the end to which an object or action is directed; aim’ / ‘Resolution, determination, intention.’ These definitions all involve some form of wilful intention to achieve an end objective. Even if this objective is restricted to the perpetrator’s criminal conduct, this is no less a part of the commission of the crime, therefore relating the accomplice’s intentions to the crime’s commission.
Why contort the ordinary wording?
Although some academic support has been lent to Trial Chamber VII’s view on purposive aiding or abetting, it is difficult to see what this is based on, given the lack of reasoning in the judgment and the ensuing lack of legal certainty. One commentator has lent support to Trial Chamber VII’s interpretation, by suggesting that:
seeing Article 25(3)(c) in the manner proposed by the Bemba et al. Trial Judgment also has its advantages. It bypasses, in one clean stroke, the entire purpose v. knowledge debate – at least in its current iteration – that has engulfed US Alien Tort Statute jurisprudence and which Defence counsel have periodically tried to import into the ad hoc and hybrid Tribunals
This probably says more about Trial Chamber VII’s finding than was expressly intended. The view that ‘the elevated subjective standard relates to the accessory’s facilitation, not the principal offence’ may be much more to do with finding a pragmatic fix (aligning the Rome Statute with customary international law on aiding and abetting) than it is to do with statutory interpretation of the Rome Statute.
Desirable as some may perceive it to be, there is no ‘interpreting away, out of existence’ the purpose requirement as pointed out by Van Sliedregt and Popova. But after Van Sliedregt and Popova recognise this need for adherence to the statutory text, they conclude that the purposive requirement is indistinguishable from knowledge: ‘it is artificial to distinguish a person who knows that a certain consequence will follow his act and does it anyway, from one who intends the consequence. Knowledge thus equals intention.’ This seems to challenge or erode fundamental notions in criminal mens rea of knowledge and intention. Knowing that you are contributing to someone doing something is a different thing to intending the consequence. Van Sliedregt and Popova’s point may relate to a factual, not legal issue. As a matter of law, knowledge does not equal intention.
Understood from this perspective, the ‘purpose v knowledge debate’ has indeed become a debate rather than a matter of statutory interpretation of the Rome Statute based on legal construction in accordance with Article 21 of the Statute. Perhaps what Trial Chamber VII highlighted, in reiterating that Article 25(3)(c) is ‘unlike other international instruments’, if anything, is a perceived need for the Assembly of States Parties to consider amendment of the Rome Statute in order to align ‘aiding or abetting’ with customary international law.
Are we, however, missing the wood for the trees? There is a broader question here about whether the elevated purposive requirement of aiding or abetting in the Rome Statute actually makes much practical difference inside the ICC in terms of prosecution charging strategy, or outside the ICC for the purposes of assessments of the Rome Statute as a source of law. If we reflect on the requirements of Article 25(3)(d) we start to see that that the real ‘outer limits’ of complicit assistance at the ICC are not determined by the ‘purpose’ requirement in (c). The next blog post will therefore turn to look in more depth at arms trade complicity in terms of ‘knowingly contributing’ to an international crime under Article 25(3)(d).