Category Archives: International Criminal Justice

On the Independence of International Prosecutors

Richard J. Goldstone is a former Justice of the Constitutional Court of South Africa. He was the first Prosecutor of the United Nations International Criminal Tribunals for Rwanda and the former Yugoslavia.

David Bosco’s book Rough Justice contains an excellent survey of the first decade of the International Criminal Court (ICC) and, in particular, of the role played by its first Chief Prosecutor, Luis Moreno Ocampo. Ocampo’s sometimes active and sometimes passive role with regard to each of the nine situations presently before the Court are carefully and comprehensively described and analysed.

The central theme that runs throughout is the role of politics and especially major power politics with regard to the decisions taken by the prosecutor and its influence on the successes and failures of the Court. The development of that theme is set against the history of the international criminal tribunals that preceded the ICC.

In setting up the two UN ad hoc tribunals for the former Yugoslavia and Rwanda, the major Western powers, and especially the United States, played an indispensable role. As the cold war had ended and atrocities were again being perpetrated in Europe, in 1992 Russia and China were prepared to support an ad hoc war crimes tribunal under the auspices of the Security Council. When, soon after, Rwanda initiated a call for a similar tribunal in response to the genocide committed in its country in the middle of 1994, the Security Council could hardly refuse. Importantly, both of those tribunals were in no way inconsistent with the foreign policies of the P5 members of the Security Council.

The successes of the ad hoc tribunals and of the “hybrid” Special Court for Sierra Leone encouraged a number of less powerful nations, under the leadership of Canada, to call for a permanent international criminal court. They found it to be unacceptable that the final decision on whether to investigate atrocity crimes should be left to the Security Council subject to the veto power of the P5. The United States, China and Russia had some misgivings about such a court. They realised that it would operate outside their direct control. With the international courts established by the United Nations they were able to exercise a large measure of control over the jurisdiction, reach and powers of the court. They could not necessarily dictate policy to independent prosecutors and judges but they could certainly control their jurisdiction, resources and, to a large extent, the implementation of their orders and responses to their requests.

I feel more strongly than does Bosco about the extent to which international prosecutors have acted independently of the views of the major powers. He does refer to the actions of Ocampo in calling for an arrest warrant for President Omar al-Bashir of Sudan in the face of objections from all of the P5 members of the Security Council. However, he raises some doubts about the reasons for other decisions such as the decision by the ICTY prosecutor deciding not to investigate alleged NATO war crimes in Serbia during 2000; the ICTR prosecutor deciding not to investigate war crimes allegedly committed by the RPF during 1994 in Rwanda; and some of the investigations abandoned by Ocampo. I will respond briefly.

With regard to alleged war crimes committed by NATO during its bombing campaign in 2000, the Prosecutor (Carla del Ponte) accepted the advice given her by the ICTY’s chief international lawyer to the effect that the evidence available was not sufficient to justify a formal investigation. In particular he came to the conclusion that there was no basis upon which indictments could be issued against individual officials. The evidence was clear that the NATO leaders, political and military, were at pains to avoid, to the extent possible, targeting civilians. At worst, the allegations of civilian casualties were a consequence of negligence or errors of judgment. There was no evidence at all to suggest intentional targeting of civilians. In any event, war crimes that might nonetheless have been committed by NATO were substantially less grave than those that were being investigated by the ICTY against the Serb military. Serbia, under Slobodan Milosevic, had been conducting an egregious campaign of ethnic cleansing against the Albanian population of Kosovo. Even if, as Bosco, suggests, NATO was unwilling to furnish information to the prosecutor concerning its conduct, I would suggest that Del Ponte’s decision was a justifiable one.

The case of RPF crimes allegedly committed in Rwanda is a more complex and unhappy one. As Bosco points out, the allegations of crimes committed against civilians were serious and merited the attention of the prosecutor. While they were not committed with genocidal intent, some of them appear to have reached the level of crimes against humanity. At the time that the allegations emerged it must have been obvious to both Louise Arbour and Carla del Ponte that if an investigation had been launched, the Government of Rwanda would have severed its relationship with the ICTR. In that light, the choice would have been to proceed with the RPF investigation in the knowledge that the response from Rwanda would effectively have brought the life of the tribunal to a premature end. It could not have proceeded with trials without witnesses and evidence from Rwanda. The mission of the ICTR was to investigate the genocide committed in 1994. I would suggest that the prosecutor was justified in abandoning the RPF investigation in order to enable her to continue with the primary mission of the ICTR. That this was not stated openly is a matter for regret.

With regard to the record of the prosecutions initiated by Ocampo, Bosco’s conclusion reads as follows:

“There is no “smoking gun” evidence that the prosecutor has made these choices because of perceived major-power preferences or out of a desire to avoid entanglement with them. There are plausible nonpolitical arguments against investigations in each of these cases. Because the prosecutor has only infrequently explained a decision not to open an investigation, moreover, there is little documentary evidence to assess. But the overall pattern strongly suggests that the prosecutor’s office has, to this point, used its discretion on where to open investigations strategically.”

That prosecutors take into account the support that one or other investigation and prosecution will receive from relevant governments seems to me to be obvious. It would indeed be folly to leave that out of account. There are many issues and considerations that dictate whether this or that investigation is appropriate. They include the gravity of the alleged crimes, the evidence available or likely to become available, the official position of the alleged perpetrators and the time, effort and expense of the investigation and prosecution. There are others. One is certainly the prospect of cooperation from relevant governments. It is in this respect that the United States is of particular importance. The intelligence information that it furnished to the prosecutor of the ICTY is well known.

In conclusion, the success of any international court will depend upon its independence and especially from the great powers. It was primarily for that reason that the ICC was established. The selection of its judges and their actual and perceived independence are crucial and no less that of the prosecutor. It is in this context that the issues raised and objectively analysed by Bosco are so important.

Symposium: Whither the International Criminal Court?

The International Criminal Court (ICC) finds itself in an interesting predicament. On the one hand, it purports to function as an independent mechanism for holding those responsible for atrocities to account, regardless of their nationality, political allegiances, or geopolitical significance. On the other, the institution is embedded in international law first and foremost, which is itself part and parcel of an international legal order where sovereign equality is only formal.

David Bosco has written an excellent book on the ICC’s initial years navigating this tension. The substance of the book, called Rough Justice: The International Criminal Court in a World of Power Politics (OUP, 2014), is ably introduced by the various participants in this symposium, so I will resist the temptation to rehearse its full argument now. In short, Bosco assesses the ICC’s first years within a framework that questions the extent to which powerful states have marginalized, controlled or accepted the Court, pointing to an important degree of “mutual accommodation.”

There is much to commend about this excellent work, which will no doubt animate discussions about international criminal justice generally and the ICC specifically for some time to come. I hold my own applause for my substantive contribution later in the symposium, but I do want to mention at the outset that Bosco’s text has prompted me to add another line to my blogging manifesto, namely, a commitment to showcasing aesthetic excellence on this site. His book is beautifully written.

In terms of format, the symposium will involve a leading group of experts. In keeping with my commitment to promoting conversation between scholars, members of civil society and practitioners, I have invited a former Prosecutor of the ad hoc tribunals, others who have worked as senior practitioners, two very prominent members of civil society, and academics from leading institutions. The resulting group of experts come at these issues from different starting points and offer contrasting perspectives.

The result, I hope you’ll agree, is a truly fascinating set of reflections on this historic institution.

A New Instrument on “Gross” Violations? Enthusiasm and Apprehension

I join this fascinating discussion to offer reflections on Professor Ruggie’s interesting proposal for “a legal instrument addressing corporate involvement in the category of “gross” human rights violations.” As someone whose work focuses on the relationship between commerce, atrocity and international criminal law (“ICL”), I applaud Professor Ruggie’s consistent expressions of interest in this relationship, and his desire to play a proactive role in moving this type of accountability forward. His desire coincides with a range of new initiatives that share similar aspirations: in one recently launched by the International Corporate Accountability Roundtable (ICAR), of which I am a member, a group of experts plans to explore the sorts of problems (legal, investigative and practical) that impede prosecutions of these sorts. In another, recently announced by the United Nations Office of the High Commissioner for Human Rights, ICL will feature as one part of a wider and longer project investigating best practices in corporate accountability for “gross” human rights abuses.

Neither of these twin initiatives advocates for the promulgation of a new treaty; both contemplate building frameworks similar to the UN Guiding Principles, that work with pre-existing legal tools. By contrast, the idea of a “new instrument” attempts to break new ground, presumably in treaty form. A treaty would certainly offer a number of benefits. A single instrument addressing corporate responsibility for “gross” human rights violations could help in producing clear, uniform law that provides helpful guidance to businesses and human rights advocates alike. A treaty could identify and confront barriers to justice, including the cost of financing litigation, difficulties with investigative capacity or the absence of a regulatory level playing field globally. It could also be helpful in recommending divisions of labor between home and host countries, such that everything from evidence acquisition to conduct of trial and enforcement of sentences is better coordinated. All of these features are salutary, important, and worth pursuing.

This said, I want to express a series of countervailing dangers involved in codifying a new instrument on corporate responsibility for “gross” violations of human rights, in the hopes that attempts at generating a legal instrument like this are appraised of the possible pitfalls that await. In a way, my concerns are reminiscent of David Kennedy’s Dark Sides of Virtue—the idea that while human rights initiatives frequently bring about a great deal of good into the world, at a very minimum, they must make conscious and address (if possible) their potential downsides. In what follows, I expand on several of these, in ways that I hope act as a friendly caution to those involved in this laudable project.

The first concern stems from how we understand “gross” violations. I appreciate “gross violations of human rights” is something of a term of art in the field, and that the UN General Assembly and others have adopted definitions that equate “gross” violations with ICL to avoid the ambiguities of separating more fundamental human rights from less. Whether ICL and “gross” human rights overlap perfectly or just substantially, there is a sense that these two sisters of international law are again lifting one another up. If some (not Moyn) see Nuremberg as the genesis of both international human rights and ICL, perhaps modern initiatives focused on civilizing business, such as this new instrument, can replicate the catalytic effect between the two fields. Personally, I see this possibility in positive terms, but we should also pause to observe the potential downsides.

For one reason, ICL is a relatively poor vehicle for enforcing economic, social and cultural rights. In its early years, the ICTY flirted with including violations of economic, social and cultural rights in its understanding of persecution as a crime against humanity, but that approach has received a mixed welcome, and by and large, is not close to adequately protecting systemic violations of economic, social and cultural rights. The mismatch between ICL and “gross” violations of human rights would cut the other way too. It’s unclear for instance, whether pillage of natural resources (a primary mechanism for modern conflict financing) constitutes a “gross” human rights violation within the meaning this new instrument would adopt, even though it is unquestionably an international crime that has deleterious consequences for civilian populations in many corners of the world. From the foregoing, one is left wondering whether a focus on “gross” human rights violations will do full justice to human rights or ICL?

And how about national law? Over the summer, a colleague and I sat through the entire Blackwater trial in Washington D.C. (see initial commentary here and a presentation here), in part, because we saw it as a pivotal moment for the idea of home states holding their own corporate officers accountable for conduct that amounts to international crimes perpetrated in foreign war zones. I say “amounts to” because the Blackwater trial was most striking in one respect: it made not an iota of reference to international law at any point. This purely American criminal trial could have constituted a corporate war crime case if charged as such, but instead, the US Attorney’s preferred to employ different, local offenses in providing a judicial response to the gross (corporate) human rights violations that transpired in Baghdad that day.

Still, the Blackwater trial should still count as a judicial response to “gross” human rights violations by a corporation, no? The trial is a remarkable example of the accountability the business and human rights movement aspires to, absent only the reference to international law. Surely we aren’t so wedded to international law that we deprive it of this status. The question for the new instrument then becomes, how would a treaty governing business and “gross” violations of human rights address purely domestic trials like this, that make no mention of human rights of international crimes at all. Is there not a danger that the new category of “gross” violations obscures more than it clarifies?

Leaving the scope of this new treaty to one side, what of the implications for ICL of a new treaty governing “gross” violations of human rights? A new instrument could allow a wholesale departure from previous standards in ICL that already rightly implicate private actors. This anxiety isn’t purely academic—one of the reasons we do not see new treaties governing International Humanitarian Law presently is that the International Committee of the Red Cross (ICRC) knows full well that opening up the Geneva Conventions in a post-September 11 world will lead to a net diminution of humanitarian protections. Are we certain that a similar process will not transpire for “gross” corporate violations of human rights, in ways that push the two bodies of international law further underground rather than lifting them up?

One idea is that a new instrument governing corporate responsibility for “gross” violations of human rights could contain an entirely compartmentalized set of principles that apply to businesses and their representatives, leaving ICL entirely unaffected. Yet, this idea of a segregated regime could pose both symbolic and substantive problems. At the level of symbolism, why should there be a separate category for one set of actors, when they are already bound by pre-existing doctrine in ICL itself? Does this preferential treatment imply that business is normatively or morally privileged? Although I’m sometimes tempted by Jules Coleman’s argument that markets deserve special moral deference because they stabilize notions of the good that we cannot otherwise agree on, overall, I am reluctant to venerate businesspeople over and above politicians, military leaders or other groups capable of committing these crimes.

I suspect that part of the response to these symbolic concerns is that the new instrument will really just focus on harmonizing disparate standards particular to corporations. The problem with this idea is that ICL itself is disparate already, so one can’t harmonize some standards (like complicity) without cutting across pre-existing law. Consequently, if the concern is harmonization, perhaps the task is to harmonize ICL as a whole, or at least portions of it that most closely affect these debates. Over the past years, I have argued that we should adopt a single concept of blame attribution universally (including, but not limited to, complicity) to address some of these problems. Since then, I have set out a set of arguments (see here) for this type of global standardization. Although commerce was a major driver in my thinking, I consistently pitched this claim to the entire field of ICL. The idea of a new instrument to do or encourage this for just business cases is less ambitious, but it does fragment the discipline.

In addition, equating “gross” human rights with ICL brings business and human rights face to face with transitional justice. Up until this point, much of this discussion has assumed a very juridical response to corporate malfeasance. For various reasons I won’t labor here, I believe that judicial responses to this problem are critically important, especially given the immense culture of impunity presently in place. Nonetheless, a number of scholars are less enthusiastic about the fetishization of legal accountability that ICL has brought about. To repackage their concerns into the present context, a new instrument governing gross violations of human rights should not preclude a Truth and Reconciliation Commission instead of a criminal trial, in response, say, to corporate implication in Apartheid South Africa. This poses an interesting tension, however, since we are unequivocally calling for greater judicial-type accountability, including overcoming legal barriers that tend to inhibit it. Those negotiating a new instrument will have to confront this inherent tension.

This brings us to the dangers of “crowding out”. A focus on “gross” violations of human rights could undermine Professor Ruggie’s excellent work on corporations and human rights simpliciter. A new and exciting scholarship is emerging in ICL lamenting the extent to which ICL crowds out other agenda. The moral intensity of atrocity impedes our vision of political economy, colonial history, and human rights performance, all of which also play important causal roles in reproducing mass violence. We simply forget about these other contextual factors in our enthusiasm for sensationalized trials (which arguably do too little to deal with root causes). I have misgivings about this “crowding out” thesis as a critique of ICL (see here), but it is helpful in reminding us of the need to pursue solutions to the problem of business and human rights generally at the same time we develop new tools for the worst types of violations. In other words, our enthusiasm for a new instrument on corporate responsibility for “gross” human rights violations should not obscure the need for deeper structural change and our commitment to pursuing it.

Overall, with respect to “gross” violations at least, one wonders whether the better approach is just to focus on what we already have—the relationship between current ICL and commerce remains very poorly understood, not to mention very infrequently enforced. To be sure, there are upsides to the treaty approach that may outweigh the potential pitfalls I point to; my enthusiasm may win out over my apprehensions depending on the precise parameters of a draft treaty. But however this particular initiative plays out, greater emphasis on the relationship between extant ICL and business will illuminate the possibilities for accountability that already exist, without inviting States back to a negotiating table. In this respect, too, the possibility of a new instrument should not blind us to the work already at hand.

An Important New Orthodoxy on Complicity in the ICC Statute?

This post is exceptionally long by blogging standards, partly because my own views on aiding and abetting in the ICC Statute only crystallized during this symposium, but also because I wanted to offer a semi-comprehensive defense of this new position to close out the groundbreaking dialogue. I do not intend to post anything this long again for this bog, it just seemed important and timely in this instance. I’ve written this piece very quickly, without the time to seek input from the experts I sometimes speak for in this text. Accordingly, I have opened up the possibility for readers to write comments (click the ‘Leave a Comment’ button immediately below the title to this post or scroll to the end of it). I hope that the experts I cite, those I have unfortunately not been able to include in this debate, and interested readers from all backgrounds will improve my account by criticizing it.

Something very significant happened over the course of this symposium—a new, analytically compelling, and very consequential interpretation of the “purpose” standard of complicity in the ICC Statute may have emerged among a leading group of scholars. In this closing post, I offer a defense of this new definition, which I call orthodox now because I take it to be supported by the majority of the scholars that participated in this symposium and some who did not. Under the twelve headings that follow, I offer an argumentative synthesis of the debate, which begins with doctrine, addresses theory, then concludes with a set of residual points of disagreement that I hope will spark further research.

The ramifications of this new interpretation are significant.

I suspect that, like me, most judges, academics, and practitioners have entertained a doctrinally flawed and theoretically indefensible interpretation of “purpose” as a standard for accomplice liability in the ICC Statute for many years, which I hope this final post, together with the fine expert opinion upon which it is based, will help dispel. The new orthodox interpretation not only overturns reasonably firmly held scholarly and professional views to the contrary, it also countermands appellate decisions in US Alien Tort Statute cases that had drawn heavily on the ICC language, breaths new life into discussion about the role of complicity in business and human rights, and arguably adds fuel to the fire of those who believe that forms of responsibility in the ICC Statute are arranged hierarchically.

  1. The history of the Old Interpretation of the “purpose” standard

To recall, the English version of Article 25(3)(c) states that:

“In accordance with this Statute, 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”

The received wisdom (I call “the Old Interpretation” for the remainder of this blog), is that the “for the purpose of facilitating the commission of such a crime” denotes a volitional commitment to the consummated crime. An accomplice has to positively want the perpetrator to use her assistance to commit the crime. On this interpretation, cognizance of a criminal outcome that would certainly flow from one’s assistance is insufficient, with the consequence that indifference marks the dividing line between the ICC Statute’s “purpose” variant of complicity and the knowledge standard other international tribunals apply as a matter of course. In light of points made during this symposium, I now believe that this position is doctrinally inaccurate and theoretically indefensible.

Nonetheless, many (myself included) bought it hook, line and sinker. At the level of theory, we posited that the knowledge standard entailed a more communitarian notion of responsibility, whereas “purpose” was libertarian in construction. In practice, fever-pitch battles were fought between advocates of either side of a purpose/knowledge divide, culminating in a circuit split among US appellate courts on the topic within Alien Tort Statute cases and detailed discussion at various ad hoc tribunals. Although the ICC itself has not addressed the provision in great depth, it has indicated (somewhat confusingly) that “what is required for this form of responsibility is that the person provides assistance to the commission of a crime and that, in engaging in this conduct, he or she intends to facilitate the commission of the crime.” (see Goudé Confirmation Decision, para. 167). All the while, experts within the Business and Human Rights movement insisted on the knowledge standard of complicity in customary international law, watering down “purpose” as best they could.

I argue here that this assumed interpretation of “purpose” was incorrect, and that accordingly, bringing forth the more accurate (and far more defensible) meaning ushers in something of a Kuhnian paradigm shift for all these fields. In fact, if Markus Dubber is correct that the history of German criminal law is a history of “discoveries”, it strikes me that this collective undertaking has unearthed an interpretation of complicity in the ICC Statute that may also deserve that label.

  1. The important new orthodox interpretation of complicity in the ICC Statute

I start by setting out what I will describe as the new interpretation of aiding and abetting in Article 25(3)(c) of the ICC Statute that emerged most clearly over the course of this symposium (for convenience, I will call it the “New Interpretation” hereafter). According to this New Interpretation, the mental element of aiding and abetting in the ICC Statute should be interpreted as requiring a double test that is comprised of the following two elements:

  1. As for the fact of assistance, the accomplice must purposefully do that which facilitates the crime (or attempt to do that which would facilitate the crime) – The “purpose” requirement does not go to the consummated offense, it attaches to the act of facilitation. An accomplice cannot facilitate by negligence or recklessness, say by forgetfully leaving a gun on the kitchen table that someone else uses to murder a third party, but she is responsible for an international crime that requires intent (say deportation as a crime against humanity) if she purposefully supplies the weapon to the perpetrator, in the awareness that it will be used to forcibly displace civilians as part of a widespread and systematic attack in the ordinary course of events. For clarity, I use language in the heading above that deliberately steers clear of describing this requirement as “for the purpose of helping” or “for the purpose to assist”, because the words “help” and “assist” often (wrongly) imply some type of disposition towards to consummated crime when, as we will see below, this language is really just meant to reference the conduct that facilitates the crimes;


  1. As for the criminal result of the facilitation (whether attempted or completed), the accomplice must have whatever mental element is announced in the crime charged. Importantly, this second element arises from Art 30 of the Statute, which stipulates that mental elements require intention and knowledge “unless otherwise provided” elsewhere. Thus, because Art 25(3)(c) is silent as to the mental element for consequences of an aider and abettor’s assistance, we should use definitions contained in Article 30 to fill this void. After all, this is how we read all the other forms of participation in Articles 25(3)(a) through (d). Thus, because the vast majority of international crimes are silent as to the mental element, Article 30 stipulates that the accomplice is liable if “in relation to a consequence, that person means to cause that consequence or is aware that it will occur in the ordinary course of events.” A minority of crimes explicitly raise the mental element higher by demanding a special intent (think genocide, persecution, torture), whereas a select few drop it lower (think of the war crime of using, conscripting or enlisting children in Art 8(2)(b)(xxvi), which only requires that “[t]he perpetrator knew or should have known that such person or persons were under the age of 15 years.” This is negligence.) For these exceptional offenses, the mental element for the accomplice is “otherwise provided for” by the crime. For all others, the lowest standard of intention applies, meaning that an accomplice will be found guilty if he purposefully provides the assistance, “aware that it [the prohibited criminal result] will occur in the ordinary course of events.”

 In what follows, I defend this New Interpretation, first by aggregating and synthesizing selected arguments made by other experts in this symposium, then by taking issue with the idea that a literal interpretation of Article 25(3)(c) necessarily leads to any particular conclusion. I go on to show how experts in our symposium might justifiably reach this new reading of the provision based on a range of factors that include the full structure of the US Model Penal Code and the negotiating history to the ICC standard. Finally, I argue that the Old Interpretation is theoretically indefensible whereas the new is not, even if this leaves a set of residual questions for further debate.

  1. Through different routes, the majority of experts confirm that, doctrinally speaking, “purpose” means the New Interpretation

 Here, I simply want to highlight how and to what extent our various discussants in this symposium support the New Interpretation. As you will see, they endorse it with varying degrees of directness and commitment, to the point that some may wish to retort at the bottom of this post if I have misunderstood their position. Until then, I explain my reading of each of our discussants in order to transform the New Interpretation into the dominant orthodoxy on this issue—given that the symposium involves a significant cross-section of experts who have worked very extensively on these topics for a large number of years, I believe their shared opinion holds great weight in this regard.

  • Thomas WeigendThomas Weigend’s contribution is a masterpiece. Later, I show how one of his arguments is a genuine breakthrough for the theory of complicity, which cuts through hundreds of pages in the (Anglo-American) literature on the topic. Leaving theory to one side momentarily, doctrinally speaking, Weigend is a powerful and explicit advocate for the New Interpretation I highlight here. His paragraph on the topic is worth re-quoting in full:

“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. For example, 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).”

Although none of the other authors employ wording so closely attuned to the New Interpretation, I believe they all offer analyses that support it. Below, I synthesize portions of their thinking that I read as supporting Weigend’s interpretation in an attempt to at least partially substantiate my claim that this represents the new orthodoxy in the hermeneutics of this provision within the ICC Statute.

  • Flavio Noto – Noto concludes his excellent post by stating that “a volitional commitment requirement for aiding and abetting [is] redundant and inappropriate.” This conclusion comprises both normative and doctrinal components, but focusing on just the doctrinal limb for now, he is of the opinion that “there is merit in suggesting that proof of certain knowledge fulfills the mens rea required by Article 25(3)(c)”. For most international crimes, this position squares with the language of Article 30, which requires, as a minimum, that an accomplice is “aware that it [the perpetrator’s crime] will occur in the ordinary course of events.” This terminology is as close as one gets to “certain knowledge of future events” (Noto’s term), meaning that Article 30 provides a powerful doctrinal grounding for his argument. Personally, I would argue that the mental element for accomplices should also vary for the small number of international crimes that require more or less than intention, in order to stay true to the “unless otherwise provided” language in Article 30, but I see counterarguments, and this is perhaps a topic for further research. The upshot is that Noto rejects forcefully a strong “purpose” standard, and embraces an interpretation that very significantly overlaps with the New Interpretation I offer here.
  • Sarah Finnin & Nema MilaniniaThis joint contribution to our symposium adroitly places the “purpose” standard in context, reasoning that “an additional ‘“purpose”’ requirement is problematic for a number of reasons”. Although their contribution raises a number of very helpful points that feature elsewhere in this synthesis, they limit they argument about interpreting the “purpose” standard in the ICC by arguing that knowledge of a particular outcome will usually allow courts to infer “purpose” absent other compelling explanations, and that a “purpose” may be one of many rationale for the accomplice’s actions; it need not be the sole Because Finnin and Milaninia’s contribution is more directed to a wider context than technicalities of interpretation, one cannot find anything overtly supporting the New Interpretation in their helpful contextualization. Nonetheless, nothing they say is obviously inconsistent with the New Interpretation, and much of their reasoning supports it in spirit
  • Cassandra Steer – I am not entirely sure whether she would agree with me, but I read Cassandra Steer’s contribution as consistent with the new definition I argue for. Steer defends the so-called compensation theory, which is the traditional rationale for elevating the mental element for complicity to a strong notion of “purpose”, viz. a volitional commitment to the criminal outcome. The rationale for this compensatory move derives from the relative weakness of the accomplice’s physical contribution as compared with that of the perpetrator (I return to this argument later). However, I read her use of this argument as defending the idea that “purpose” should go to the act of facilitation (not the consummated offence), in part because Cassandra helpfully points to the possibility of “double intent”, but predominantly since she ultimately concludes that in interpreting aiding and abetting in the ICC statute, “it may be possible to include knowledge, willful blindness or dolus eventualis, especially since in civil law jurisdictions these all amount to gradations of intent.” Therefore, “purpose” must define facilitation, whereas intent goes to results. If this is a fair reading of her, her position coincides with the New Interpretation.
  • Adil Ahmad HaqueHaque’s post affirms the New Interpretation very directly, if we read him as endorsing one of the possibilities he raises, namely, that “the drafters [of the ICC Statute] intended to track the MPC.” In particular, he argues that “[a]t the first step, we apply 2.06(3) to determine whether the defendant is an accomplice to the perpetrator’s conduct, ie, if the defendant aided the perpetrator with the “purpose” of facilitating the perpetrator’s conduct. Only at the second step do we ask whether, in addition, the defendant had whatever mental state with respect to the results of that conduct is required for commission of the crime. So 2.06(4) adds to, and does not subtract from, the “purpose” requirement of 2.06(3).” On the assumption that States meant to incorporate the whole MPC scheme into the ICC standard and used Art 30 of the ICC Statute to do the work the MPC assigned to 2.06(4) (see below on legislative intentions and the relevance of the MPC), I take Adil as an explicit advocate of the New Interpretation.
  • Elies van Sliedregt and Alexandra Popova – In their contribution to this debate, these authors too begin by “agree[ing] with James Stewart’s initial intuition, and the conclusions reached by others in this series of posts, that interpreting Article 25(3)(c)’s reference to “purpose” as requiring that the accomplice share the principal’s intent would set too high a threshold for responsibility.” However, they also opine that “it is self-evident that [purpose’s] inclusion in Article 25(3)(c) has the effect of displacing the application of Article 30.” While I would agree with respect to the facilitation, I (and others who support the New Interpretation) consider that it does not do so with respect to prohibited results. They may share this view—they go on to advocate for a double intent that is analogous in form to that contained in the New Interpretation, and a clear rejection of the old dominant interpretation. van Sliedregt and Popova argue that “purpose presupposes knowledge of the principal’s intent coupled with voluntariness, or will, to be party thereto.” All that is required to merge this language with the New Interpretation is to understand their “will to be party” as a purpose to do that which facilitates and their “knowledge of the principal’s intent” as an intention to bring about the criminal result, relying on Article 30 of the ICC Statute to enunciate the meaning of intent (which, of course, includes “aware[ness] that [the criminal result] will occur in the ordinary course of events,” which their “knowledge of the principal’s intent” could help prove).
  1. Other leading academics support the New Interpretation

 I describe the orthodoxy I believe emerged over the course of this symposium as new, but it is really only its rise to prominence that is especially novel. In truth, a number of leading experts in the field of international criminal justice had already advanced this interpretation, or something close to it, well before these debates. I take this opportunity to summarize some of this pre-existing expert opinion. Usually, views on the question are relatively concise, so I content myself in citing them verbatim then offering minor explanation where necessary:

  • Albin Eser – Albin Eser is a leading theorists of international and comparative criminal law, who has served as the Director of the Max Planck Institute for Foreign and International Criminal Law in Freiburg and an ad litem judge at the ICTY. Well before this symposium, he argued for the New Interpretation. His argument is also worth quoting at length and requires no commentary on my part:

“As a general norm on the mental element, Article 30 of the ICC statute is not only applicable to the perpetrator, but other participants in terms of article 25(3)(a) – (e) of the ICC statute as well. This means that, in principle, the mental requirements for an accomplice are neither higher nor lower than those for the perpetrator, therefore a participant can in particular not be held responsible for mere recklessness or negligence either. Nevertheless, there are some particularities of complicity to be observed.

 In general, due to the accessorial nature of complicity, the accomplice must have a ‘double intent’, both with regard to his own conduct and with regard to the content and knowledge of the principal. In both relations the requirements of intent and knowledge of basically the same as with regard to a single perpetrator. This general line is not without exceptions, however, which in particular concern two groups: one being aiders and abettors who, beyond their general double intent, must act “for the “purpose” of facilitating the commission of [such] a crime” according to article 25(3)(c) of the ICC Statute.”

Albin Eser, Individual Criminal Responsibility, in The Rome Statute Commentary, pp. 933-934

  • Kia Ambos – Although Ambos does not argue for the New Interpretation quite as explicitly as his compatriot, I read him as supporting it implicitly. Ambos argues that:

“it is important to note that this higher subjective threshold (‘“purpose”’) only applies to the relation between the contribution and the execution of the crime (‘facilitation’). With regard to additional mens rea requirements, for example, the ‘intent to destroy’ in article 6, it suffices for the assistant to be aware of the perpetrator’s special intent, but he need not himself possess this intent.”

(See Treatise on International Criminal Law, p. 166).

I hope Professor Ambos will correct me if I misread him, but I take his reasoning as oblique support for the New Interpretation. If “purpose” only goes to the facilitation, then the mental element required for consequences of the criminal undertaking is derived from the crime itself. Ambos uses genocide as his example but I see no reason why the principle should not hold for crimes that do not have special intents. Also, I believe that awareness of the perpetrator’s intent could well be and often is an indicia of the accomplice’s awareness that a crime will follow from her purposeful assistance in the ordinary course of events.

* * *

I do not include other excellent authors here, many of whom have argued against interpreting “purpose” as entailing a volitional commitment to the consummated offence. This is partly due to a lack of space, but predominantly because they adopt a different interpretative strategy, at least in the scholarship I’ve seen. Nevertheless, I did want to acknowledge the exceptional work of Hans Vest and Doug Cassal in this regard. I suspect that these scholars may also support the New Interpretation, but here I have no basis to speak for them.

  1. The literal interpretation of the ICC Statute’s complicity provision is ambiguous

 My task now is to defend this new orthodoxy, in doctrinal terms and (very briefly) in theory. I start by attempting to defeat its main adversary in these debates: the argument that a literal interpretation cannot support any reading other than the Old Interpretation. Undoubtedly, the Old Interpretation represents a very plausible literal interpretation of Article 25(3)(c) of the ICC Statute, but I here suggest that there are at least four others, and that the language of the provision itself does little work in guiding our choice between the variants. To draw on Herbert Hart, the provision is more penumbra of doubt than core of settled meaning. So, given this literal ambiguity, I believe that the contextual factors I address in subsequent sections are most important in suggesting the New Interpretation as the most cogent interpretation of all the literal possibilities.

Taking this language at face value, one can certainly come to the conclusion that aiding and abetting in the ICC Statute requires that the accomplice positively want to facilitate the commission of the entire offense. This is the first and most common interpretation. Yet, it is far from inevitable. This Old Interpretation makes several assumptions that the text itself does not inevitably impose, namely that: (a) the term “purpose” attaches to “commission of such a crime”; (b) the English language version of the ICC Statute is the only version worth considering in these debates; (c) “purpose” relates to the accomplice’s subjective mental element; and (d) “purpose” signifies the overall objective, motivation, or rationale for the acts that gave rise to the accessorial liability. Each of these assumptions is contestable, and in a way, all of the experts in this symposium have rejected at least one of them.

So, the New Interpretation offers a plausible second literal reading by contesting assumption (a) above. Structurally speaking, Article 30 of the ICC statute creates a general provision that goes to all forms of responsibility (and indeed crimes) unless these forms of responsibility and crimes designate otherwise. This is evident from the beginning of Article 30 of the ICC statute, which starts with the famous words “unless otherwise provided for.” Mental elements for forms of responsibility are frequently “not provided for” in the ICC Statute, which means that Article 30 does all the work in generating the applicable mental elements. For example, article 25(3)(a) of the ICC Statute, which deals with perpetration rather than complicity, makes no mention of mental elements at all, since these are left to Article 30 in the wider ecology of the statute.

If “purpose” goes to the act of facilitation rather than the consummated criminal offense, Article 30 is binding in defining mental elements for results of this facilitation. Some may say that this effectively inserts the words “the conduct that led to” into the phrase “for the “purpose of facilitating the commission of such a crime” such that a new reworked provision actually reads “for the “purpose” of facilitating the conduct that led to commission such a crime”. One can certainly understand how critics might object that this insertion is inconsistent with the strictures of literal construction, but it is better to think of the additional language as a mere clarification of an inherent ambiguity, which is consistent with the origins of the provision, expert opinion, and basic principles in the theory of blame attribution. I say more about each of these below. For now, I merely want to highlight this second, imminently plausible literal reading of this provision.

It is too early to say, but some might offer third interpretations by reading the equivalent of “purpose” in other official languages of the ICC Statute. Over the course of this symposium, a translator from the ICTY contacted me inquiring about the French equivalent “en vue de,” especially when French is the ICC’s other working language. Robert Roth’s insightful remarks assimilated the phrase “en vue de” to the strongest form of intention, but regrettably, I failed to ask him to explore precisely how, why and when this takes place in Swiss criminal law. My failure is important, since it leaves open the possibility that, if translated as “with a view to,” the French might prioritize cognition where the English “purpose” seemingly implies volition. I include this question as one of a long list of issues that require further research (along with analysis of the equivalent terms in the equally authoritative Chinese, Russian, Arabic and Spanish versions of the Statute). For now, suffice it to say that linguistic variations undermine the thesis that a literal interpretation of “purpose” necessarily leads anywhere particular.

In a fourth possible reading, Thomas Weigend points to an interpretation that treats “purpose” not as a mental element at all, but as an objective characteristic of the facilitation. In effect, he points to scholars who contest (c) above. In describing the work of Antje Heyer and Katherine Gallagher, both of whom I respect as scholars, Weigend considers as “plausible” that “for the “purpose” of facilitating the commission can also be interpreted as an element of the actus reus of assisting: the assistant’s conduct must be specifically shaped in a way to be of use to the perpetrator.” I don’t want to rush to judgment on this idea and defer to Weigend’s much greater wisdom on what may or may not pass the plausibility threshold and certainly appreciate these scholars’ work, but at present, I do confess grave doubts about the coherence of this explanation. The point is, the text itself is entirely silent on the topic; it does not confirm or deny this reading. Thus, I include this interpretation here to undermine the thesis that a literal interpretation inexorably leads to the Old Interpretation of “purpose.” Analytically, that’s simply untrue.

Finally, what does “purpose” mean anyway? Even if the provision was clear about what “purpose” attaches to (facilitation itself or the consummated crime), whether the reference to “purpose” is a mens rea requirement or an objective characteristic of the facilitation offered, and how linguistic variations of the standard affect the concept’s meaning across different languages, we still have to come to some understanding about the interpretation we give the term. In this regard, Thomas Weigend brilliantly insists on a firm distinction between “purpose” and motive, downgrading common perceptions of “purpose” as requiring a singular, ultimate desire towards a defined end. In short, he contests assumption (d) above. Robert Roth, Elies van Sliegdredt and Alexander Popova join Weigend on this score. Some of them also employ the term “joint-intention,” which adds new valences to an interpretative smorgasbord that the language in Art 25(3)(c) does not restrain.

In my view, references to “intention” are a great source of confusion in the theory of complicity generally and its incarnation in the ICC Statute specifically. In the 1950’s, when the American Law Institute was developing the U.S. Model Penal Code under the direction of Herbert Wechsler, the leading American scholars involved in the project elected to abandon the term “intention” completely, because it lent itself to far too many meanings, many of which were more prone to spark profound and lasting dispute than produce nuanced standards to work with. If that was true within a single nation state, one can only begin to imagine how much worse the problem is internationally, especially when other nations understand the term differently and there is an attempt to insert it onto a provision governing complicity in the ICC Statute that makes no mention of intention at all. Again, however we resolve these ambiguities, the language of Art 23(3)(c) itself will not prove terribly helpful.

For all these reasons, literalism does not inevitably support the Old Interpretation, requiring us to look elsewhere for guidance in deciding between these options.

  1. The US Model Penal Code, from whence the ICC standard comes, confirms the New Interpretation

As I mentioned in my initial post that began this symposium, the US Model Penal Code (“MPC”) is widely regarded as the inspiration for Article 25(3)(c) of the ICC Statute. Despite this, a key provision within the MPC’s treatment of complicity has never featured in debates about the shape we give to aiding and abetting in the ICC context, despite the fact that it clearly militates in favor of the New Interpretation. I start this section by demonstrating the striking paralleled between complicity in the ICC Statute and the version in the MPC to substantiate the latter’s influence on the former. Then, I set out the missing provision in the MPC that has important but under-appreciated implications for our preference between the different literal interpretations of Article 25(3)(c) we just considered.

Two features of the provision governing aiding and abetting in the ICC Statute are dead giveaways of its provenance. The first, of course, is that the MPC speaks of “with the purpose of promoting or facilitating the commission of the offense…”, whereas the ICC Statute statute reads “[f]or the “purpose” of facilitating the commission of such a crime…” In a second dead giveaway of the MPC’s great influence, the ICC standard for complicity is triggered when an individual merely attempts complicity. Art 25(3)(c) of the ICC reads “aids, abets or otherwise assists in its commission or its attempted commission.” This is something of a scandal conceptually, but doctrinally, it is a very significant parallel with the MPC that has no equivalent elsewhere in international criminal justice and is very rare nationally. Like the ICC Statute, the MPC reads “aids or agrees or attempts to aid such other person in planning or committing it” (See § 2.06(3)(a)(ii) (emphasis added). So, both points of mimicry between the two instruments substantiate the received wisdom that the provision in the ICC Statute was largely a copy and paste.

And yet, there is one provision within the MPC definition that has not featured within these debates, despite the fact that it obviously favored the New Interpretation of the ICC Statute. As I set out in my original post, the very next provision in the MPC after the “purpose” reference on aiding and abetting reads that “[w]hen causing a particular result is an element of an offense, an accomplice in the conduct causing such result is an accomplice in the commission of that offense if he acts with the kind of culpability, if any, with respect to that result that is sufficient for the commission of the offense.” (see page 22 of the article). For several reasons, the import of this second missing provision is hard to overstate in the transition from the Old to the New Interpretation of complicity in the ICC Statute.

Most importantly, this missing provision supports the idea of reading “the conduct that led to” into the phrase “for the “purpose” of facilitating the commission of such a crime” such that the new reworked provision in total now reads “for the “purpose” of facilitating the conduct that led to commission of such a crime”. Tellingly, this is precisely the way one must read the MPC, too. Under the MPC’s definition, there is no way of making sense of the inclusion of this reference to the missing passage dealing with “causing a particular result” (§ 2.06(4)) without assuming that “with the “purpose” of promoting or facilitating the commission of the offense (§ 2.06(3)(a)(ii)) goes to the act of facilitation, not the criminal result. Adil Haque’s excellent post on the topic from an American perspective confirms exactly this reading (see in particular, his discussion of Riley v. State as a good illustration).

Let me deal with the retort that, “well, this is all very pleasant but these intricacies in the MPC don’t have much to do with the entirely separate international treaty that is the ICC Statute.” A number of my colleagues mentioned Article 31 of the Vienna Convention as requiring a plain meaning to these terms. As I argue above, to my mind, that argument does not advance the ball terribly much: the provision governing complicity in the ICC Statute is literally silent as to whether “purpose” goes to the facilitation alone or the consummated offense, some leading theorists think there is plausible ambiguity about whether “purpose” should be considered a mental element, linguistic discrepancies pose real challenges to literal interpretations, and “purpose” goes undefined in the Statute too. If Thomas Weigend considers this drafting “enigmatic,” literalism alone is unhelpful.

Therefore, Article 32 of the Vienna Convention is germane. To recall, Article 32 of the Vienna Convention refers to the “preparatory work of a treaty”, that can be employed to determine the meaning of a treaty provision when the literal interpretation “(a) leaves the meaning ambiguous or obscure; or (b) leads to a result which is manifestly absurd or unreasonable.” The very fact that the interpretation of complicity in the ICC Statute gives rise to so much debate, that so many interpretations are available from the text, that so many of us who have spent years working on this topic seem to have got the wrong end of the stick up until now, and that at least one of the world’s leading scholars views the language as “enigmatic” would tend to prove that this wording is “ambiguous or obscure”.

As we saw a moment ago, I also read all participants in this symposium as concluding that the Old Interpretation of complicity in the ICC Statute (requiring a volitional commitment to the consummated offense) is “manifestly absurd or unreasonable.” On either count, I believe that reading the ICC standard of complicity in light of its forebear in the MPC finds a firm mandate in international law.

In fact, ignoring this genesis risks fundamentally distorting the concept. In light of the fact that the ICC standard incontrovertibly heralds from the MPC, that recourse to the MPC offers a compelling explanation of how to read an inherent ambiguity in the ICC standard, and that this New Interpretation accords with the interpretation that the vast majority of leading experts in this symposium would support as a matter of both doctrine and theory, it would be unfortunate to maintain an old interpretation that is effectively disproved merely because of some artificially formal divide between the ICC Statute as a treaty and the MPC as a national code. This is all the more true when other factors also militate so powerfully in favor of the New Interpretation.

  1. Negotiators of the ICC Statute intended the New Interpretation, not a volitional commitment to the consummated crime

In the proceeding section, I argued that the MPC is an important source of interpretation for the ICC Statute’s complicity standard, but if the MPC is the ICC’s obvious progenitor on this topic, it remains to be seen how those responsible for negotiating the Rome Statute saw these matters. Here, we are confronted with a curious fact—they never mention the MPC. Nonetheless, they do interpret the “purpose” standard in ways that are perfectly consistent with the New Interpretation derived from the MPC, and their views cannot be reconciled with the Old Interpretation in any way, shape or form. I start by setting out two of the most cited comments from prominent experts who negotiated the provision in the ICC Statute, then show how they more or less directly endorse the New Interpretation.

In my opening post, I cite an abbreviated passage written by Donald Piragoff, Canada’s representative during the negotiations of the ICC Statute, who played a leading role in the negotiation of the aiding and abetting provision at Rome. I include the full citation below because it unequivocally confirms the New Interpretation:

“A question arises as to whether the conjunctive formulation [intent and knowledge] changes existing international jurisprudence that an accomplice (such as an aider or abettor) need not share the same mens rea of the principal, and that a knowing participation in the commission of an offence or awareness of the act of participation coupled with a conscious decision to participate is sufficient mental culpability for an accomplice. It is submitted that the conjunctive formulation has not altered this jurisprudence, but merely reflects the fact that aiding and abetting by an accused requires both knowledge of the crime being committed by the principal and some intentional conduct by the accused that constitutes the participation . . . . Article 30 para. 2(b) makes it clear that “intent” may be satisfied by an awareness that a consequence will occur in the ordinary course of events. This same type of awareness can also satisfy the mental element of “knowledge,” as defined in article 30, para. 3. Therefore, if both “intent” and “knowledge” are required on the part of an accomplice, these mental elements can be satisfied by such awareness.” (See page 355 of this article).

Pause momentarily to notice the structure of this explanation before we move to analyze its content. Piragoff speaks of two mental elements: a knowledge component that goes to the principal’s commission of the crime, and an intentional disposition towards the accomplice’s participation. In the passage just quoted, he explicitly refers to this as a “conjunctive formulation.” That there are two elements immediately discredits the Old Interpretation, which viewed “purpose” as the singular standard that required the accomplice to harbor a volitional commitment to the completed offense. That there are two mental elements immediately contradicts that reading, regardless of their content.

In terms of content, Piragoff’s expression is readily reconcilable with the New Interpretation. His first element—knowledge of the crime being committed by the principal—squares with the lower standard of intention in Article 30 of the ICC Statute, which includes awareness that a consequence will occur in the ordinary course of events. He acknowledges this explicitly. If we take his second element, which refers to “intention,” to envision the strongest sub-component of that amorphous term, then he is explaining that “purpose” goes to what he calls “conduct by the accused that constitutes the participation.” Admittedly, he does not reference “purpose” at all in this explanation, but there is no other non-bizarre way of mapping his account onto the language that actually exists in the Statute he negotiated.

This reading of his explanation is in perfect accord with the content of the New Interpretation, which to repeat, views “purpose” as attaching to the act of facilitation and awareness that a consequence will occur in the ordinary course of events as the lowest relevant mental element for most international crimes in the ICC Statute. (Again, for the sake of completeness, recall that some international crimes require more than intention while others require less. I suggest that the second mental element for complicity should shift in line with these definitions of crimes, so that awareness that a consequence will occur in the ordinary course of events will not be the applicable standard in all instances).

David Scheffer, the head of the U.S.’s delegation in Rome, agrees with this assessment in even clearer terms. He states that:

“the ‘purpose’ language stated the de minimus and obvious point, namely, that an aider or abettor “purposely acts in a manner that has the consequence of facilitating the commission of a crime, but one must look to Article 30(2)(b) for guidance on how to frame the intent of the aider or abettor with respect to that consequence.” (See page 355 of this article).

The explanations both these authors offer regarding the text coincide with its origins in the MPC, the new orthodoxy among participants in this symposium, and theoretical questions about complicity I turn to below. Moreover, there is nothing whatsoever in this history that supports the Old Interpretation, namely, that “purpose” requires a volitional commitment to the consummated offense. Accordingly, it is hard to resist the view that the negotiating history to the ICC’s provision governing complicity is another nail in the coffin of the old mistaken interpretation so many of us unwittingly assumed for so long. The negotiating history is especially potent given the literal ambiguities I point to.

  1. The majority of the few national legal systems that employ “purpose” as a complicity standard confirm the validity of New Interpretation of the ICC Statute

In their post contextualizing the “purpose” standard in the ICC Statute, Sarah Finnin & Nema Milaninia do a great job pointing out how “purpose” is only applied as a standard of complicity in a great paucity of criminal law systems. All other international courts and tribunals apply a knowledge standard (that boils down to recklessness in practice), which is largely drawn from an equivalent standard in Anglo-American systems. Generally speaking, systems inspired by continental models apply dolus eventualis (vaguely akin to recklessness) as the lowest standard for accomplice liability, and the unitary theory countries like Norway, Denmark, Italy, Austria, and Brazil pair the accomplice’s mental element to that required for perpetration. Moreover, even if “purpose” is a great outlier in comparative terms, the majority of the few examples of it in national legal systems are striking in that they confirm the New Interpretation.

After accepting that the old interpretation of “purpose” in the ICC Statute is indefensible and therefore undesirable, Elies van Sliedregt & Alex Popova argue that “nor can Article 25(3)(c)’s reference to “purpose” be interpreted away, into non existence.” I agree with this argument, and hope that the foregoing shows how the New Interpretation does not bring about an affront on literal interpretation; it continues to assign “purpose” an important role but limits this role to the act of facilitation, leaving Article 30 to govern consequences. Put differently, the New Interpretation respects the terminology set out in Article 25(3)(c), it just attaches it to the conduct of the accomplice not the criminal enterprise en gross. This much is repetition. What is distinct about the limited national experience with “purpose” as a standard of complicity, however, is the fact that major national systems do exactly what van Sliedregt & Popova say is impossible—in the majority of national systems where the legislature has adopted a “purpose” standard of complicity, courts do interpret it into non-existence.

In my opening post, I set out a series of examples from various national systems that adopt “purpose” standards. I will not repeat them all again here, but in summary, the US Supreme Court recently adopted a knowledge standard explicitly in a case called Rosemond v. United States, even though their earlier caselaw required the accomplice to have “a stake in” the resulting offense. Justice Alito observed in dissent, having reviewed the history of the knowledge and purpose debates up until then in the U.S., that the majority opinion confounds these two standards. Nonetheless, it is tremendously significant that the resulting standard for complicity is knowledge, and that the US Supreme Court is clear that “[t]he law does not, nor should it, care whether he participates with a happy heart or a sense of foreboding.” This is the country that is said to be at the origins of the “purpose” standard for accomplice liability.

As the citations in my earlier post show, both Canada and New Zealand follow a similar logic. Both contain “purpose” standards in legislation, but their Supreme Courts interpret them as requiring either knowledge or intention vis-à-vis the completed offense. If reducing “purpose” to intention seems strange, see John Finnis’s (one of English law’s most important figures) explanation of how most English jurisdictions extent intention downwards, whereas “Canadians select purpose as the term to be artificially extended.” (see this article, fn 74). By this, he means that English systems include standards lower than a volitional commitment as intention, which he views as terminologically inaccurate. This, of course, reflects the debate about whether dolus eventualis can be accurately described as an element of intention in civil law systems, or whether it requires its own autonomous existence as a basis for blame attribution. Following Finnis’ logic, the jurisdictions that view “purpose” as containing more than pure volition towards a completed crime are just mimicking a similar approach in all other jurisdictions, including the ICC. Importantly, however, purpose means knowledge in these countries and cannot, therefore, be used to bolster the Old Interpretation.

On the other hand, Isreali criminal law clearly adopts the New Interpretation. As I point out in my earlier post, Israel is also a “purpose” jurisdiction, but the leading case stipulates that “where the aider only foresees the possibility of the commission of the principal offense, the aider may be convicted if it is his or her desire that should the offense actually be committed, his or her act will facilitate its commission.” Itzhak Kugler, Israel, in The Handbook of Comparative Criminal Law 352, 370 (Kevin Jon Heller & Markus Dubber eds.) (citing the Israeli Supreme Court case of CA. 320/99 Plonit v. State of Israel 55(3) PD 22 [1999]. In commenting on the decision, Kugler explains that “[t]he requirement of the code that the actor act with the “purpose” of facilitating the crime relates only to the contribution of the aider; that is, it is required that he or she want his or her act to facilitate the commission of the offense… Thus, in the case where the aider was almost certain that his or her act would facilitate the commission of an offense, the aider may be convicted in spite of the fact that he or she did not desire to facilitate the commission of the offense.”

These nationals examples displace the old assumptions about “purpose” as a mental element for complicity, which turn out to be unsupported by so many different sources of authority, including national law.

  1. The New Interpretation of complicity in the ICC Statute minimizes the discrepancy with the standard in customary international law

Finnin and Milaninia assert that “there is scope for the [ICC] to interpret the ‘purpose’ requirement broadly, and in a manner that minimizes the divergence from customary international law.” This opinion coincides with that of David Scheffer, who writes that:

“The wording of article 25(3)(c) was uniquely crafted for the ICC, and when read in conjunction with the mens rea standards set forth in article 30 of the Rome statute, it leaves the judges of the ICC the task of determining precisely the proper criteria for accessorial liability. Nothing discourages or prevents them from looking to the growing jurisprudence of the international criminal tribunals for the former Yugoslavia and Rwanda, the Special Court for Sierra Leone, the Extraordinary Chambers in the Courts of Cambodia, to state practice, and the scholarly texts for guidance on this issue.” (See page 352 of this article).

There are a range of good reasons for taking this advice seriously. In a separate piece I wrote for Elies van Sliedregt and Sergey Vasiliev’s new edited volume, Pluralism in International Criminal Law
(OUP, 2014), I pointed to a range of problems that arise from disparate understandings of forms of attribution for international crimes, amongst international and domestic jurisdictions alike. I will not rehearse those arguments here, except to say that the difficulties with the fragmentation of international law are real, particularly relevant for complicity, and without obvious solution apart from asking judges to attempt harmonization wherever possible. I suggest that the New Interpretation offers them an opportunity to do just this.

The Old Interpretation of “purpose” in the ICC Statute creates an important cleavage between complicity in the ICC Statute and customary international law. The idea that “purpose” somehow denotes a volitional commitment to the outcome, a desire to bring about the completed offense, clashes with “knowledge” as applied by other international courts and tribunals who purport to draw on custom. The choice between these two standards has led to protracted litigation in the context of the Alien Tort Statute, appellate litigation in national criminal tribunals, and confounds the business and human rights discourse. Moreover, as I have attempted to show once or twice (see here, pp. 38-39 and here, pp. 30-31), the customary standard reduces to recklessness in practice, which is problematic when recklessness will not suffice for perpetration of the crime the accomplice will be held responsible for.

The New Interpretation of aiding and abetting brings the mental element for complicity much closer to this customary standard, and does so while simultaneously preventing against excesses the customary standard may occasion. I say more about the theoretical credentials of both the old and New Interpretations further below. For now, I simply want to add the need for greater harmony in this area of law to the catalog of arguments for the New Interpretation listed elsewhere in this post.

10. The Old Interpretation is not theoretically defensible

One could easily write a book many times longer than this post on the theory of accomplice liability (and many, including those who contributed to this symposium, have). I don’t want to delve into this theory too deeply here, in part because I have written about the topic at far greater length elsewhere. In previous work, I have set out a relatively neutral survey of the various theoretical options for constructing accomplice liability (see here), and offered a more opinionated set of arguments for adopting a unitary theory of perpetration as the best option for international crimes (see here). Most recently, I canvased the literature for and against a “purpose” standard for accomplice liability (see here, Section II.C Towards a Moral Theory of Accomplice Liability).

In each of these earlier pieces of work, I made various normative criticisms of the Old Interpretation of the “purpose” standard (i.e. one that requires the accomplice to positively want the completed offense). They range from a strong sense of “purpose” driving a stake between desert and responsibility, to a “purpose” standard failing to match popular notions of blame and guilt, thereby undermining the social function of international trials (see here, pp. 44-47). Instead of rehashing these various arguments here again, I want to pick up on Falvio Noto’s observation about how this Old Interpretation came into being internationally, then address the three strands of argumentation he rightly claims maintained the Old Interpretation as a received wisdom about complicity in the ICC Statute. Before proceeding, however, I do think it is important to note that no expert in this symposium defended the Old Interpretation in conceptual terms.

According to Noto:

“[t]hree lines of argumentation can be discerned: Some authors claim that the purposive motivation requirement balances the low objective threshold. That reasoning is difficult to uphold given that the Lubanga Trial Chamber interpreted Article 25(3)(c) as requiring substantial effect (even though it did so in an obiter dictum). Other commentators appear to view Article 25(3)(c) in the light of domestic doctrines restricting the scope of aiding and abetting by means of an elevated mens rea requirement. Lastly, a variety of scholars derive a dolus directus in the 1st degree threshold from their reading of Article 2.06 MPC, on which Article 25(3)(c), they claim, was based on.”

We have already addressed the second and third arguments, I hope convincingly. The majority of the very few national systems that use “purpose” as a standard for complicity do not support the Old Interpretation as we all suspected—they either dilute the term so that in means knowledge or adopt the New Interpretation that attaches “purpose” to the facilitation rather than the completed offense as a whole. Moreover, one can only think that the MPC supports the Old Interpretation by leaving out a key provision within that instrument—as we’ve seen, once this missing provision is reinserted into the interpretative frame, the MPC unmistakably favors the New Interpretation (see section 5, above). Finally, those who actually negotiated the ICC standard report that States intended the New Interpretation, trumping all arguments from national law anyway. The second and third arguments fall away, leaving just the first.

It is really Noto first argument that has served as the Old Interpretation’s theoretical anchor—we need to drive the mental element of “purpose” to the highest possible ground, goes the argument, in order to compensate for the weak physical contribution an accomplice makes relative to the perpetrator. On its face, this idea of compensation is appealing, and it looms large in the very few conceptual accounts of accomplice liability that are prepared to defend a strong notion of “purpose” as the appropriate mental state for accessorial liability. As I say, it appeared once or twice in the symposium too, although no one appeared to use it to defend the Old Interpretation explicitly.

In a very significant moment for the field, Thomas Weigend’s post dispatched this argument very convincingly. His dismantling of the compensation argument for “purpose” as a standard for aiding and abetting is one of the most exciting (and important) aspects of this symposium. To reiterate, the compensation argument, which features throughout the literature and once or twice in this series, suggests that elevating the mental element for aiding and abetting beyond intention to “purpose” (note the ambiguities of intention) is perfectly justifiable given that the accomplice makes a weaker or less direct causal contribution to the crime. The frailties of the physical contribution, goes the argument, are cured by amplifying the requisite mental requirement.

In a passage of critical importance Weigend masterfully dissects this position. In one portion of his samurai-like dispatch of the thesis, he argues that:

this calculus, to me, makes little sense.  If the assistant’s objective contribution is of lesser importance, then her sentence should reflect that fact. But the question whether the assistant desires the perpetration of the crime should have no influence on her punishment, because her “volition” does not increase the harm she causes or supports.”

Later, Robert Roth agreed, calling the compensation theory a “paralogism”, which to my mind, captures the thesis perfectly. Thus, all three rationale for a strong “purpose” standard are without merit. Again, none of these expert commentators defended it.

11. The New Interpretation is theoretically defensible

 A few years ago, I wrote a paper called The End of Modes of Liability for International Crimes (see here). If the somewhat unnecessarily provocative title suggests a nihilistic approach to blame attribution, it obscured the fact that the project was a very intellectually honest attempt at arriving at a concept of complicity I felt able to defend. As I entered into the project, I quickly found that the hallmarks of the “modes of liability” literature in ICL indicated that “modes of liability” should not extend beyond the contours of the crimes they couple with (for fear of violating principles of culpability and fair labeling). On this basis, I argued that the mental element for complicity should be exactly the same as it is for perpetration. In effect, this meant that the mental element for complicity had to be dynamic (because different crimes require different mental elements), not static like knowledge and “purpose” (which seemed to apply to the accomplice regardless of the mental element in the crime she was charged with).

In actual fact, I was wrong that the “purpose” standard for complicity in the ICC Statute is static; that position assumed the Old Interpretation, which has turned out to be false. The New Interpretation corrects for this problem. Notice how the missing provision in the MPC is dynamic in structure, inviting courts to determine, with respect to results of one’s assistance, whether the accomplice has the necessary mental element required for conviction of the crime she is charged with. This structure is mirrored in the ICC Statute to the extent that Article 30 functions in a dynamic manner, too. Because Article 30 of the Statute commences with the words “unless otherwise provided,” the definitions of intention and knowledge within it apply in instances where the Statute is silent (as is the case for complicity, on issues of result). If the Statute requires a stronger mental element (for genocide, which requires a special intent) or a weaker standard (for the war crime of using, conscripting or enlisting children, for which negligence suffices), the mental element required for complicity shifts, too.

In my opinion, this is entirely theoretically defensible—indeed, it is preferable to all other standards on offer in customary international law or national law. If “purpose” goes to assistance, then someone is not liable for negligently leaving their gun unlocked when someone else removes it for a crime spree, but they are responsible for an international crime that requires intent (say deportation as a crime against humanity) if they purposefully supply the weapon to the perpetrator, in the awareness that it will be used to forcibly displace civilians as part of a widespread and systematic attack in the ordinary course of events. The New Interpretation is sensitive to the crimes complicity couples with whereas both the knowledge standard and the Old Interpretation of “purpose” randomly skew the meaning of responsibility by making liability turn on chance couplings between mental element and the crime charged.

I resist the temptation to defend this theory again here. I am conscious that many excellent scholars disagree with me about the unitary theory of perpetration as a model for all forms of liability for international crimes (for an interesting critique, see Gerhard Werle and Boris Burghart’s article in this edited volume and Cassandra Steer’s great book Translating Guilt: Identifying Leadership Liability for Mass Atrocity (T.M.C Asser Press, 2015)). I confess that I am not entirely convinced by their thoughtful responses, but the interesting aspect for present purposes, is that the New Interpretation creates dynamism within the mental element for complicity without leading to the collapse of the differentiated system a number of theorists hold dear. Once cabined in this way, I suspect that the dynamism of the mental element I call for will seem considerably more palatable conceptually. Certainly, I hope I raised a number of arguments for it, and have seen none against. Perhaps this dialogue will begin that new debate. Whatever the case, I believe that there are strong conceptual arguments against the Old and for the New Interpretation.

12. Points of residual disagreement, areas for further research

There are numerous points of residual disagreement, which will hopefully stimulate a new wave of critical scholarship. First, what is the equivalent of the English “purpose” in all the other official language versions of the ICC Statute? Second, is this double intent standard normatively defensible? For myself, I wonder whether the first step (requiring “purpose” for the facilitation) is conceptually redundant—why not just consider whether the person who left their weapon out negligently had the mental element(s) necessary for being found guilt of the offense? In other words, I acknowledge that without amendment, the ICC Statute commits us to a two-step analysis, I just wonder whether this makes sense theoretically. Third, how specific do the two mental elements for accomplice liability have to be? There is interesting caselaw on these questions in England, France and Germany, which remains to be debated within international criminal justice. Fourth, what of attempted complicity in the ICC Statute? How does this change matters relative to customary international law? Fifth, is “shared intent” really the appropriate phrase to describe issues of complicity, given that there is no necessary solidarity between perpetrator and accomplice—there need be no agreement between them vis-à-vis the completed crime. These, and a host of other questions, are of utmost importance, not just for our understanding of international criminal justice in an interconnected world, but also for the scholarly disciplines that draw so heavily on it. In the end, I believe that this symposium broke new ground in displacing an old and ushering in a new interpretation of “purpose” in the ICC Statute. My kind thanks to all those experts who lent their knowledge, time and insight to the discussion.





“En vue de”: The Significance of the French Equivalent of “Purpose” in the ICC Statute’s Complicity Provision

Robert Roth is a Professor of Criminal Law at the University of Geneva and Director of the Geneva Academy of International Humanitarian Law and Human Rights (one of my alma mata, and a tremendous influence on my thinking). He was the Presiding Judge of the Special Court for Lebanon between 2011 and 2013.

Je remercie James Stewart de sa sollicitation et de son autorisation/permission d’intervenir en français. Il se justifie d’autant plus d’intervenir dans une langue « continentale » que l’approche est essentiellement différente et que les concepts, même les plus importants, se traduisent malaisément. James m’a demandé de m’exprimer à partir du droit suisse et de l’usage que ce droit, passerelle entre les mondes francophones et germanophones, fait de l’expression « en vue de » de l’article 25 al. 3 lit. c) du Statut de Rome.

Je ne vais pas répéter ce qui a été bien exposé par mes prédécesseurs, en particulier quant à la distinction essentielle entre motifs (ou mobiles) et intention in actu ou quant au fait que l’imputation d’une intention se fait sous forme de dol direct quand un événement (la mort de tous les passagers de l’avion) est la conséquence inéluctable d’un acte délibéré (tuer l’un des passagers en faisant exploser l’avion).

En bonne doctrine suisse – et allemande dont la première subit l’influence déterminante -, la présence de l’élément subjectif « en vue de » permet de catégoriser l’infraction en tant que délit de dessein (Absichtdelikt). Cela signifie que, pour que l’infraction soit consommée, il suffit que l’auteur ait commis les actes énumérés dans la norme (tuer, soustraire, dénoncer faussement) en ayant le dessein d’obtenir un résultat (l’enrichissement illégitime dans le cadre du vol, provoquer l’ouverture d’une poursuite pénale dans le cas de la dénonciation calomnieuse). En revanche, il n’est pas nécessaire que l’événement désiré se réalise pour que l’infraction soit consommée : celui qui soustrait un objet sans parvenir à s’enrichir est bien un voleur.

Sur quoi doit porter ici le dessein prévu par l’article 25 al. 3 lit. c)? Il vise la facilitation de la commission d’un crime (facilitating [the crime]), et donc pas directement le crime. S’agissant du crime lui-même, les dispositions ordinaires de l’article  30 du Statut suffisent (cf. l’interprétation systématique des rapports entre 25 et 30 proposée de van Sliedregt/Popova dans leur contribution à la présente discussion). Dès lors, celui qui apporte son aide sans véritablement faciliter la commission du crime (par exemple il y a suffisamment d’armes, et l’arme fournie par le « complice »  vient en surplus) serait-il punissable pour complicité consommée et non pour tentative de complicité, impunissable selon le Statut ? La question sera résolue au stade de la causalité objective : si les moyens fournis ne facilitent en aucune manière la commission du crime, le participant en sera resté au stade de la pure intention, fût-elle sous la forme qualifiée du dessein ; il est donc impunissable.

A quoi sert alors la qualification en tant que forme de participation à dessein ? Le dessein est une forme d’intention qualifiée. Le « complice » n’accepte pas simplement que son acte facilite la commission du crime ; il le veut au sens fort du terme. Le dessein se situe à l’opposé du dol éventuel, forme d’intention dans laquelle l’auteur s’accommode de la commission d’une infraction (il l’envisage et l’accepte), sans toutefois que son comportement tende à cette fin. La doctrine suisse et allemande s’est toujours interrogée sur la question de savoir si la forme du dessein éventuel était logiquement admissible. A mon sens, elle ne l’est pas, car il y contradiction entre les deux éléments (le dessein et son caractère éventuel). L’événement lui-même peut parfaitement avoir un caractère non inéluctable (exemple de l’ouverture d’une poursuite en cas de dénonciation calomnieuse) ; en revanche, l’auteur ne peut pas en même temps vouloir et ne pas vraiment vouloir.

Cela signifie que celui qui s’accommode simplement du fait que son assistance puisse faciliter la commission d’une infraction ne réalise pas à mon sens l’élément subjectif de 25 al. 3 lit. c). Ce point est important car il permet d’éliminer ce que la doctrine allemande appelle une simple Handlung mit Hilfetendenz (action tendant à prêter assistance) ; cf. Welz, Zum Verhältnis von Anstiftung und Beihilfe, Frankfurt am  M.  2010, p.45.

J’aimerais encore intervenir sur un point soulevé à diverses reprises dans ce débat. Sur le plan conceptuel, j’ai énormément de peine à accepter la théorie de la « compensation » : il serait nécessaire de compenser la plus faible implication (objective, actus reus) du « complice » par une exigence plus élevée en matière subjective (mens rea).  (On trouve un développement de cette thèse dans la contribution de Cassandra Steer). Cela me paraît un paralogisme : la thèse est d’abord discutable sur le plan de la légalité, puisque l’on donne à une norme d’imputation pénale une interprétation difficilement compatible avec son texte. Mais, surtout, les éléments objectifs et les éléments subjectifs ne relèvent pas de la même catégorie conceptuelle et on ne peut pas remédier à la relative légèreté des uns en invoquant la solidité des autres ; ce mélange des genres n’apporte guère de cohérence à un édifice déjà fragile. Le raisonnement me fait penser au raccourci, emprunté par certains législateurs, tendant à contourner les difficultés en matière de preuve par un assouplissement des éléments matériels de l’infraction ; là aussi, on mélange deux registres différents.

La démarche « compensatoire » est essentiellement justifiée par le fait que la commission du crime (article 25 al.1 lit. a)) et la « simple » participation (entre autres aiding and abetting) sont traitées de manière équivalente dans le Statut, en particulier au stade de la fixation de la peine. La justification ne me paraît pas convaincante : d’une part, le choix de ne pas distinguer dans le « texte législatif » (le Statut) n’interdit pas d’opérer une gradation au stade de la fixation effective de la peine ; d’autre part, dans les systèmes qui différentient entre action et  participation à titre principal d’une part et participation à titre accessoire d’autre part, cette  différentiation se fait généralement sur la base de critères purement objectifs (cf. Roth, « Responsabilité pénale individuelle pour délits collectifs : droit continental » in de Frouville ed., Punir les crimes de masse : entreprise criminelle commune ou co-action ? Bruxelles, 2012, pp. 55-71).

Reste posée la question délicate de la divergence entre l’infraction auquel le « complice » avait le dessein de prêter assistance et le crime réellement commis. A mon sens, le texte français, plus encore que le texte anglais, de l’article 25 al. 3 lit. c), impose une interprétation restrictive : au cas où le crime effectivement commis est différent du crime auquel le « complice » avait le dessein de contribuer, ce dernier ne pourra être condamné sur la base de l’article 25 al. 3 lit. c) (sur les diverses approches « continentales » en la matière, voir Roth, ibidem, avec en particulier l’approche beaucoup plus large du droit italien (article 116-117 CPI), selon lequel en substance si le délit commis est différent de celui qui avait été voulu par l’un des participants, celui-ci répond néanmoins du délit effectivement commis si ce dernier est la conséquence de son action ou de son omission, la doctrine commandant toutefois une interprétation restrictive selon laquelle la contribution au délit de celui qui est dépassé par la suite des événements est en elle-même fautive, en ce sens qu’un « homme raisonnable » aurait prévu la réalisation du délit, cf. G. Marinucci/E. Dolcini, Diritto penale, 2ème éd, Milano 2006, pp. 363).

Interpreting “for the purpose of facilitating” in Article 25(3)(c)?

Elies van Sliedregt is Professor of Criminal Law at VU University Amsterdam is the author of Individual Criminal Responsibility in International Law (OUP, 2012).

Alexandra Popova is a doctorate candidate at VU University Amsterdam, writing on aiding and abetting in international criminal justice  as part of the project ‘Dealing with Divergence: National Adjudication of International Crimes’.

The starting point in this discussion was whether the reference to “purpose” in Article 25(3)(c) requires a volitional commitment to the criminal outcome (to the consummated offence). This translated to the question: does the aider/abettor need to share the principal’s intent, or might something else suffice? We agree with James Stewart’s initial intuition, and the conclusions reached by others in this series of posts, that interpreting Article 25(3)(c)’s reference to “purpose” as requiring that the accomplice share the principal’s intent would set too high a threshold for responsibility, for the reasons that follow.

First, requiring a higher mens rea of accomplices than of principal perpetrators is not necessitated by the framework of Article 25(3) of the ICC Statute. By way of analogy: in U.S. law, there has been a long-standing split between a purpose-based approach (where the aider/abettor must share the intent of the principal to be liable) and a knowledge-based approach (where knowledge of the principal’s intent suffices). These differing mens rea standards can be traced back to the distinction between accessories before the fact and secondary principals, who were present at the scene and aiding in the commission of the offence. As all would be punished for the crime proper, a higher mens rea was required for accessories to balance their comparatively lesser physical involvement in the crime. As pointed out by Flavio Noto, requiring a higher mens rea standard (dolus directus in the 1st degree, or shared intent) for aiders and abettors might still be justified in jurisdictions where accessories receive no discount in punishment; similarly it makes sense where juries are barred from lowering sentence for minor involvement in a crime. In these sorts of circumstances, “balancing” an aider/abettor’s comparatively lesser physical involvement with a higher mens rea threshold ensures that only those possessing a sufficient degree of culpability face punishment for the crime. This line of reasoning does not apply at the ICC, where the convicted person’s degree of participation in the crime is taken into consideration, along with other factors, at the sentencing stage (see Article 78, ICC Statute; Rule 145(1)(c), RPE).

In the absence of necessity for a higher, balancing mens rea for aiders and abettors, the issue is subject to be resolved with reference to policy. This leads to a second point: as matter of policy, requiring that an accomplice possess a volitional commitment to the criminal outcome does not fit the nature of the crimes and would be contrary to the object and purpose of the ICC Statute. It is now commonplace to point out that international crimes are collective and systemic. We agree with Thomas Weigend in the post preceding this one that the commission of international crimes requires the coordination, cooperation and contributions of many actors, who may have vastly differing motives and goals. This broad division of tasks/contributions within, among and from the peripheries of organizations and hierarchies, means that many more participate than do so ardently; personal objectives are easily divorced from passions in organized murder. It would be contrary to the object and purpose of the ICC Statute to exempt from responsibility those who provide assistance knowing to a virtual certainty that they aid the commission of a crime, merely because they do not desire its commission but assist with some other objective in mind. Deterring international crimes or – to adopt the preferred phrase – fighting impunity, requires that all those who willingly participate are held responsible. Indeed, a parallel development in domestic law has seen a focus on the seriousness of the underlying crime coupled with policy concerns of crime prevention. It lead municipal courts and legislators to adopt knowledge-based approaches [See Westerfield, The Mens Rea Requirement of Accomplice Liability, at 183 referring to People v. Lauria, 251 Cal. App. 2d 471, 59 Cal. Rptr. 628 (1967) and 177; An illustration is People v. Germany 42 Cal. App. 3d 414, 116 Cal. Rptr. 841 (1974).]

On the other end of the scale, nor can Article 25(3)(c)’s reference to “purpose” be interpreted away, into non-existence. Primarily, this is because it is self-evident that its inclusion in Article 25(3)(c) has the effect of displacing the application of Article 30 (applicable “unless otherwise provided”) and that a standard higher than “knowledge” must be required; in other words, it would make little sense, and would have the effect of making that phrase of the Statute redundant, to displace Article 30 knowledge in favour of an identical Article 25(3)(c) knowledge. More generally, Article 31 of the Vienna Convention on the Law of Treaties requires that the words be given their “ordinary meaning” in light of the treaty’s object and purpose. This could lead one to argue that it is not open to the court to read down “purpose”.

So how to interpret “purpose” when looking at its wording? How to relate to section 2.06(3) of the Model Penal Code (MPC), from which Article 25(3)(c) – partly – takes its wording? The fact that Article 25(3)(c) reflects the MPC provision on “purpose” does not imply that it was the intention of the drafters of the Rome Statute to bring in the body of case law that interprets this provision, however, instructive this case law may be. Only part of the MPC provision was adopted. Moreover, as “insiders” have noted, it was the intention of drafters to accommodate different legal traditions. According to Scheffer, Article 25(3)(c) “was negotiated not to codify customary international law but to accommodate the numerous views of common law and civil law experts about how to describe the actions of an aider and abettor.” (p. 351, “The Five Levels of CSR Compliance”) Drawing on MPC-wording and inserting “purpose” seems to have been nothing more and nothing less than a copy-paste job, to use Cassandra Steer’s words.

In a similar vein, it does not seem appropriate to interpret Article 25(3)(c) in conformity with customary international law or general principles of law. Even assuming that this is possible – especially in the realm of modes of liability there is disparity in the law – several further problems arise, related to the strength of the ad hoc case law’s claim to actually reflect customary international law. It is difficult to maintain that customary international law of aiding and abetting is entirely settled, considering the very recent upheaval in relation to ‘specific direction’ – a debate which, incidentally, parallels many of the same concerns about appropriately establishing the culpability of temporally and geographically remote actors providing neutral (not “purposed”) assistance, as those that might be addressed by a standard of “purpose”. Seeking an interpretation of Article 25(3)(c) that is in keeping with a “knowledge” standard might be akin to trying to anchor to floating debris.

What then, might “purpose” mean? Does it necessarily entail shared intent, or might a looser interpretation be available? Purpose presupposes knowledge of the principal’s intent coupled with voluntariness, or will, to be party thereto. We agree with Thomas Weigend that the actor’s will flows from his conduct: 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. The level of knowledge seems key when interpreting “purpose”. An awareness ‘of a likelihood’ would be insufficient for “the purpose of facilitating the commission of such a crime”. As noted by Flavio Noto, citing Markus Dubber, proof of positive knowledge would fulfil the mens rea of Article 25(3)(c): the aider and abettor’s commitment to the criminal outcome can be derived from his certain knowledge about the facilitating effect his assistance has on the crime.

An aider’s knowledge – and his will to facilitate the act of the main perpetrator – can also be inferred from his provision of assistance that is tailored to the crimes (as opposed to neutral assistance): this refers to the example of providing weapons that can only be used to kill civilians. Indeed, this final example illustrates the parallels between specific direction and purposefulness, as well as the inevitable interplay between mens rea and actus reus, also considered by Weigend.

Finally, for the sake of completeness, it must briefly be noted that we do not share the concern that a mens rea of purposefulness would preclude the responsibility of those acting for cold-blooded profit maximisation, or indeed any other strategic or passionless motive. This is because, as Weigend notes, purpose or object is distinct from motive and goes not to the crime per se but to the facilitation. Consequently, it also seems unnecessary to distinguish between ‘primary’ and ‘secondary’ purposes, and argue that secondary purposes are sufficient as discussed by Flavio Noto, as well as Sarah Finnin and Nema Milennia.

How to Interpret Complicity in the ICC Statute

Thomas Weigend is a Professor of International, Comparative and German Criminal Law at the University of Cologne. At the risk of embarrassing him slightly, this is among the most insightful commentaries on complicity I have read in over four years of researching the topic.

My comment on the enigmatic words “for the purpose of facilitating the commission of such a crime” in Article 25 (3) (c) of the ICC Statute comes in two parts: First, is it good criminal policy for international criminal law (ICL) to require a higher degree of mens rea for convicting an assistant than for convicting a perpetrator of the same crime? Second, does Article 25 (3) (c) demand such a distinction by using the words “for the purpose”?

(1) As a general principle, it makes little sense to require a “thick” intention – however it is defined – for holding an assistant criminally responsible where a lesser degree of mens rea is sufficient for convicting the perpetrator of the same crime. The definition and sentence for any particular crime are devised with the perpetrator in mind. The allocation of responsibility to other persons, who have not “controlled” the commission of the offense or are otherwise further removed from the center of the criminal activity, is in some way accessorial to the perpetrator’s act. As the moon receives its light from the sun, an accomplice’s responsibility depends on an extension from the “natural”, primary responsibility of the perpetrator. Art. 25 (3) (c) clearly is based on this concept since this provision makes the commission or attempted commission of a crime by a perpetrator a prerequisite for convicting an assistant.  If that is so, it is counter-intuitive – and would require special justification – to add a specific mental requirement for convicting an assistant where the perpetrator will be held responsible on a less demanding standard.

This consideration is independent of whether one sees in Article 25 (3) (a) through (d) a “hierarchy” of modes of responsibility. Even if there is no such gradated system inherent in Art. 25 (3) and assisting is (or can be) of equal seriousness as perpetration, there is no good reason why an especially high degree of mens rea should need to be proved in order to convict an assistant. Some writers have proposed a “balancing” theory to justify this result: since the assistant does not singlehandedly complete the actus reus of the offense, they claim, her liability can be equal to that of the perpetrator only if the assistant’s mens rea is of a higher degree. But this calculus, to me, makes little sense.  If the assistant’s objective contribution is of lesser importance, then her sentence should reflect that fact. But the question whether the assistant desires the perpetration of the crime should have no influence on her punishment, because her “volition” does not increase the harm she causes or supports.

In ICL especially there is no good reason to require an “extra” degree of mens rea for convicting assistants. It makes little sense to exempt from responsibility those who know very well that the person whom they assist will make use of their contribution for committing a core crime but who have no direct personal interest in the perpetration of that crime. The commission of ICL crimes – contrary to many “ordinary” offenses – typically  requires the cooperation of many persons, who may all have different motives and goals. If ICL wishes to prevent such crimes it should not limit criminal liability to those who pursue a limited “purpose” and thus refrain from punishing all persons who consciously join their efforts to commit the offense.

(2) Does the wording of Art. 25 (3) (c) of the ICC Statute compel a different result? I don’t think so.

(a) Let us start with semantics. The “purpose” a person pursues describes his objective but not his motive. Therefore, an assistant under Art. 25 (3) (c) does not have to lend help because he wishes to bring about the offense. Nor does the commission of the offense have to be his sole purpose: Even if his main goal is to make money by selling arms, he may well act with the (secondary) purpose of facilitating the crime committed with those very weapons.

I also doubt that “purpose” necessarily coincides with a desire or with positive feelings about the (known) objective of one’s acts. For example, if a robber threatens to kill me unless I give him my wallet, and I comply with his demand, one can certainly say that I act with the purpose of satisfying his demand (so that he leaves me alone). But I certainly do not desire or even approve of his making off with my wallet. In sum, “purpose” describes one of the objectives of one’s act but does not say much about one’s attitude (of approval or disapproval) about that objective.

(b) This consideration leads to the general question about the role of “volition” in intent (or purpose). Some of the contributors to this blog seem to assume that – as James Stewart has put it – “the reference to purpose requires a volitional commitment to the consummated offense”. But what is the exact meaning of “volitional”, and why should it matter? If D shoots at V from close range, killing him, can D deny that he “willed” to kill V? Even if intention contains a “volitional” element, the actor’s volition will necessarily follow from his knowledge of the result that he is going to bring about. D may claim that he really did not like killing V (as in the robbery example above), and that he is sorry that he did – but still he “willed” V’s death, otherwise he would not have shot at him from close range. Contrary to  Flavio Noto, I would not say that it is a “fiction” that “certain knowledge about an undesired but anticipated side-effect is tantamount to a volitional commitment to that side-effect”. If the assailant in Flavio Noto’s airplane example knows that his rocket will kill all passengers, then it is his will to kill them all – although his motive for firing the rocket may have been his hate of one particular passenger. As in most other cases, the actor’s motive is irrelevant for his liability. His will flows from his act – if he had not willed the foreseen result he would not have acted.

This, by the way, is the hidden wisdom in the convoluted definition of intent and knowledge in Art. 30 of the ICC Statute: Normally, mens rea requires no more than a person’s awareness that a result will occur in the ordinary course of events. A further “volitional” element is necessary only where an offense definition requires that the perpetrator “intend” results beyond those brought about by the actus reus. If larceny is defined as taking someone else’s chattel with the intent of possessing it (as is the definition in § 242 German Penal Code), then it must be proved that the defendant “wished” to keep the chattel for himself – because that future development is not part of the actus reus of larceny. But assisting as defined in Art. 25 (3) (c) does not have this structure. The purpose of the assistant relates exactly to what she does: facilitate someone else’s (criminal) conduct. Therefore, the assistant’s volition is a necessary and undeniable consequence of her cognition.

(c) After having clarified what “purpose” may mean, we can turn to the question to what the assistant’s “purpose” must relate under Art. 25 (3) (c). 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. For example, 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).

(d) In what I said so far, I assumed as true the widely shared assumption that the words “for the purpose” describe a special mental element of assisting under Art. 25 (3) (c). But there is a plausible alternative reading of these words, which has been spelled out by Antje Heyer in her excellent and extensive analysis of liability for aiding and abetting in ICL (published in 2013 in German under the title Grund und Grenze der Beihilfestrafbarkeit im Völkerstrafrecht, pp. 500-501; for a similar interpretation, see Katherine Gallagher, ‘Civil Litigation and Transnational Business’, 8 JICJ 745 at 765 (2008)). “For the purpose of facilitating the commission” can also be interpreted as an element of the actus reus of assisting: the assistant’s conduct must be specifically shaped in a way as to be of use to the perpetrator. Under this interpretation, conduct that is part of a person’s normal business would not qualify as assistance, because that conduct would not have the objective purpose of facilitating someone’s crime. If, for example, an arms trader sells weapons to a dictator at their regular price and under regular conditions, he would not be an assistant to crimes against humanity even if he is aware that such crimes will be committed using these weapons. But if the trader sells the weapons at a higher price because of an existing embargo, or if he sells weapons that have been specifically designed for killing civilians, he would be liable because this particular deal has been accommodated to serve the specific “purpose” of committing the crime. Under that interpretation, the regular mens rea requirements (as described in Art. 30) would apply – the arms dealer would only have to be aware of the specific elements that give the arms deal its “purpose”.



The U.S. Model Penal Code’s Significance for Complicity in the ICC Statute: An American View

Adil Ahmad Haque, Professor of Law and Judge Jon O. Newman Scholar, Rutgers School of Law, Newark

I’ve been enjoying the discussion of complicity on this blog, but as a scholar of both American and International Criminal Law, I confess that I’m slightly confused by the discussion of the U.S. Model Penal Code (MPC)’s significance for aiding and abetting in the ICC Statute. I’m grateful to James Stewart for offering me the chance to comment on these three authors’ arguments in this respect.

I see two plausible approaches. The first is that the drafters of the Rome Statute considered the MPC, both 2.06(3) and 2.06(4), and deliberately decided to depart from it and require purpose with respect to all components of “a crime” (conduct, result, circumstance). If the drafters wanted to incorporate the 2.06(3)-2.06(4) framework then they would have done so. Since they did not, we should presume that they had their reasons and intended something different.

The second is that the drafters intended to track the MPC. According to 2.06(3), an accomplice to a conduct crime aids with the purpose of facilitating the perpetrator’s conduct. According to 2.06(4), an accomplice to a result crime is, first, “an accomplice in the conduct causing [the prohibited] result” who, second, “acts with the kind of culpability, if any, with respect to that result that is sufficient for the commission of the offense.” At the first step, we apply 2.06(3) to determine whether the defendant is an accomplice to the perpetrator’s conduct, ie, if the defendant aided the perpetrator with the purpose of facilitating the perpetrator’s conduct. Only at the second step do we ask whether, in addition, the defendant had whatever mental state with respect to the results of that conduct is required for commission of the crime. So 2.06(4) adds to, and does not subtract from, the purpose requirement of 2.06(3).

This is pretty clear from the MPC Commentary: “[2.06(4)]deals with a special case that arises when an actor is an accomplice in conduct within the meaning of [§ 2.06(3)], and when a criminal result—anticipated or unanticipated—flows from that conduct.” “The most common situation in which Subsection (4) will become relevant is where unanticipated results occur from conduct for which the actor is responsible under Subsection (3). His liability for unanticipated occurrences rests upon two factors: his complicity in the conduct that caused the result, and his culpability towards the result to the degree required by the law, that makes the result criminal.”

A good illustration is Riley v. State, Court of Appeals of Alaska, 2002. 60 P.3d 204: P fires into a crowd, recklessly injuring two people. A assisted P with the purpose that P engage in certain conduct (firing into the crowd), and was reckless with respect to whether P’s conduct would cause injury. P and A are both guilty of first-degree assault (recklessly causing serious physical injury by means of a dangerous instrument). Interestingly, this case reads 2.06(4) into a statute that, like the Rome Statute, on its face tracks only 2.06(3).

On either view, an accomplice must act with the purpose of facilitating the perpetrator’s conduct. Neither view supports complicity on the basis of knowingly or recklessly facilitating the perpetrator’s conduct. Whether or not the Rome Statute departs from the MPC with respect to results, it certainly follows the MPC with respect to conduct.

Why require that an accomplice aid with the purpose of facilitating the perpetrator’s conduct? I think about it in this way: to perpetrate a crime, one must not only posses the mental states explicitly required by the offense definition but also perform a voluntary act, which in turn requires an intent to perform the bodily movements that constitute the prohibited conduct or cause the prohibited results. It follows that, to be complicit in a crime, one must not only posses the mental states required by the offense definition but also posses the intention that the perpetrator perform the relevant voluntary act. Any lesser standard would create a gap between the perpetrator’s responsibility for his/her own conduct and the accomplice’s responsibility for the perpetrator’s conduct.

A Comparative Perspective on the “Purpose” Requirement

Dr Cassandra Steer – Cassandra is the author of a forthcoming book Translating Guilt: Identifying Leadership Liability for Mass Atrocity (T.M.C Asser Press, 2015). She is a Lecturer and Researcher at the Amsterdam Centre for International Law, but about to join McGill Law School.

As with so much terminology in the Rome Statute, the debates on the content and meaning of Article 25 are fueled by problems of translating domestic criminal law notions to the international context. It is tempting to copy-paste the MPC approach because the wording looks similar, however the Rome Statute is a patchwork of legal traditions, and no single domestic interpretation will provide the clarity that is sought. Concepts and words in one system do not necessarily have the same connotations as they do in others. The concern that the Rome Statute departs from customary law is in my view not the central issue, since one could argue that the case law of the ad hoc tribunals does not itself amount to custom; all that has emerged is differing trends in different international and hybrid tribunals, each of which draw on various domestic criminal law models. Instead I argue a broader theory of liability should be applied to solve questions of interpretation; namely a theory that differentiates between principal and assistant liability.

Aiding and abetting comes from old English common law terminology, when there used to be a legal distinction between principal in the first degree (the physical perpetrator), principal in the second degree (anyone who was physically present and aiding the crime), and an accessory who was, according to the Blackstone commentaries; “not the chief actor, nor present at its performance, but is some way concerned therein, either before or after the fact”. The distinction was made because the death penalty applied to every felony for principals, but not for accessories. However over time this legal distinction was eradicated in the common law tradition. Although the terms “principal” and “accessory” remain, every actor is considered equally liable for the full commission of the crime, regardless of his or her actual contribution. This amounts to a functionally unitary system.

In the US, the MPC attempted to make a clear distinction between the fault elements ranging from intent, purpose, knowledge to recklessness, as applicable to different crimes, however it is unclear whether there should be differing requirements for different participants in a crime, given that there is no legal distinction between them. James Stewart’s argument for a dynamic system of volitional requirements would make sense in this context; whatever is required for the crime should apply to all participants.

The different interpretations of “purpose” that appear in US case law with respect to aiders and abettor agree that in any case it should not be seen as equivalent to motive. As one judge put it in a 1940 case, even the person who sells a gun to another, knowing it will be used for murder, cannot escape liability by saying the gun was sold merely for profit and not for the purpose of the crime:

“Guilt as an accessory depends, not on ‘having a stake’ in the outcome of the crime [. . . ] but on aiding and assisting the perpetrators; and those who make a profit by furnishing to criminals, whether by sale or otherwise, the means to carry on their nefarious undertakings, aid them just as truly as if they were actual partners with them” (Backun v United States (1940) 112 F2d 635 (Fourth Circuit Appeals Court) p 637.)

This somewhat loose standard was refined by Judge Learned Hand in Peoni: the accused must “in some way associate himself with the venture, that he participates in it as something he wishes to bring about, that he seeks by his action to make it a success. All the words used [. . . ] carry an implication of purposive attitude towards it.” (United States v Peoni (1938) 100 F2d 401 (Second Circuit Court of Appeals) p 402)

The original MPC draft had included “knowingly” as a loose standard for accomplice liability, however the final draft follows Judge Learned Hand’s formulation, and requires
“purpose to facilitate the crime”. Most federal and state courts follow this standard, however as the other two excellent blogs in this discussion point out, interpretation of the standard can sometimes differ.

In Canada there is a definitional difference between an aider and an abettor; s. 21 of the Criminal Code requires for the aider that she “does or omits to do anything for the purpose of aiding any person to commit it” (emphasis added), whereas no such requirement exist for the abettor. Despite this, the same functionally unitary system of liability applies where there is no legal distinction between a principal and an assistant. As was stated in the leading Thatcher case: “[this provision] is designed to make the difference between aiding and abetting and personally committing an offence legally irrelevant. It provides that either mode of committing an offence is equally culpable, and, indeed, that whether a person personally commits, or only aids or abets, he is guilty of that offence.” (R v Thatcher S.C.J. No. 22, 1 SCR 652, p 690.)

Generally in Canadian case law there are high mens rea levels required for parties to a crime who do not actually commit the crime, such as knowledge or purpose, regardless of the mens rea requirement for the crime committed, as a way of placing some limits on the deliberate policy attempt to broaden the net of liability. The Supreme Court has stated that “the more peripheral the accused’s involvement to the completed crime, the more sense it makes to require a higher form of subjective mens rea.” (R v Roach (2010) 2 SCR 98 (Supreme Court of Canada) para 36.)

There must be a double intent; an intent to assist the physical perpetrator, as well as knowledge of at least the type of crime that is to be committed, though not necessarily the exact crime. The purpose requirement for the aider is interpreted with flexibility in the case law, since it “would be fulfilled if he had either intent or knowledge of both the crime, and of the intent of the perpetrators to commit the crime.” (R v Briscoe (2010) 1 SCR 411 (Supreme Court of Canada) para 16.) At the same time, willful blindness would also suffice in the place of actual knowledge.

Since abetting is generally understood to entail encouraging, instigating or promoting, it implies some intention to see the crime committed, and is interpreted more strictly.

It is difficult to compare the civil law tradition to these definitions, since the English language terms “aiding and abetting” can only be translated by approximation, and there are many different forms of assistant participation. While many jurisdictions that follow the civil law tradition have modes of liability that amount to instigation, soliciting or assisting, the definitions and mental fault elements differ. Suffice to say there is a double intent requirement in Argentina, Belgium, Germany, and the Netherlands, such that an actor must both intend her own participation and also intend the commission of the predicate crime (or a lesser crime. She is only liable for a further crime if it was foreseeable in the course of the crime she intended to participate in.) Whether this intent is the same as or stricter than “purpose” is uncertain for reasons of translation, but it would appear in case law trends that something similar to “knowledge” would usually suffice.

James Stewart has pointed out that in many of these jurisdictions dolus eventualis suffices whenever intent is required. Many common law lawyers are uncomfortable with this notion, but the best way to understand this is a lower limit of intent, the same as willful blindness. There must be evidence of the actor’s acceptance of the risk of a criminal outcome; this is what separates it from recklessness. Where the MPC has attempted to make clear distinctions between intent, purpose and knowledge, in most civil law jurisdictions there is only a clear distinction between intent and recklessness; the difference is that under intent there are many judicial interpretations which offer a sliding scale, the lower limit of which is dolus eventualis. Thus a term such as “purpose” may fit into this sliding scale without having a specific statutory definition. It would be something more than dolus eventualis and something less than pure intent, but as long as there is something willful (and therefore blameworthy) about the assistant’s actions, it would fit the generic test.

The point of the double intent requirement in these jurisdictions is the same as the reasoning that emerges in Canadian jurisprudence, namely that when extending liability to those who have not physically committed the crime, it is necessary to compensate the lesser physical contribution with a greater requirement of mental fault. However this has an even more important role in these systems where a normative legal differentiation exists between principals and assistants. A principal is considered to be more morally blameworthy, and therefore receives a greater punishment and the full weight of the conviction for having committed the crime. An assistant (including but not limited to aiders and abettors) has contributed less and is therefore less morally blameworthy, and receives a lesser legal qualification and usually a lesser punishment.

This distinction is embedded in the notion of moral agency; there is a difference between what I should do ‘simpliciter’, or the morality of principalship, and what I should do by way of contribution to what you do, or the morality of complicity. The latter is still wrong if you commit a crime, but it is a secondary wrong. For this reason a principal is convicted for the commission of the crime proper, whereas a secondary participant is convicted for her role in assisting the principal, but legally speaking she is not convicted for having committed the crime herself, since she did not fulfil the elements of the crime. She is not a genocidaire, but an assistant to the genocidaire, and is legally qualified as such.

In interpreting the wording of Article 25 of the Rome Statute, the terminology in paragraph (3)(b) seem to be drawn from civil law models (and then translated into English terminology, already risking translation problems!) and those in paragraph (c) are familiar to the common law model. Schabas suggests that since in practice the two paragraphs overlap very considerably, they should not be viewed as two different or distinct bases of liability, but rather as an effort to codify exhaustively.

This is why I believe the inclusion of “purpose” for the aider and abettor is justifiable; because he has contributed less, there must be some compensation for this in the form of requiring purposeful facilitation, in order to protect against guilt by association. Similarly, paragraph (3)(d) requires intentional contribution for an assistant to a group with a common purpose.

This aligns with what James Stewart states, that the “purpose” requirement goes to contribution, but the mens rea fault element to be determined still depends on the predicate crime. However I disagree with James that this necessarily amounts to a unitary theory of liability, for the reasons argued here. In fact it matches a differentiated theory particularly well. If the ICC continues its interpretation of Article 25 as differentiated, therefore requiring less physical contribution for the morally less blameworthy modes of liability, then it needs compensate this with high standards of volition with respect to the participation. This still leaves room for a dynamic theory for the mens rea with respect to the predicate crime. There must be safeguards against guilt by association for the assistant modes of liability, and the “purpose” requirement under (3)(c) as well as the “intentional” requirement under (3)(d) fulfill this need.

A comparative perspective on James’ final question “what does purpose mean” may still mean that there is some flexibility in interpreting the precise meaning of these safeguards. As long as guilt by association or recklessness are excluded, it may be possible to include knowledge, willful blindness or dolus evenutalis, especially since in civil law jurisdictions these all amount to gradations of intent. So as long as there is evidence of wilful participation, regardless of the motive, it would be possible to include business men and women who act for profit, but in full knowledge of the crimes they are facilitating.

Putting “Purpose” in Context

Dr Sarah Finnin is the author of the book Elements of Accessorial Modes of Liability: Article 25(3)(b) and (e) of the Rome Statute of the International Criminal Court (Martinus Nijhoff, 2012). She is presently an Associate Legal Officer at the ICTY

Nema Milaninia is a Legal Officer, Office of the Prosecutor, International Criminal Tribunal for the former-Yugoslavia.

 The views expressed herein are those of the authors alone and do not necessarily reflect the views of the International Criminal Tribunal for the former Yugoslavia or the United Nations in general.

The terms of Article 25(3)(c) of the Rome Statute diverge from the standard definition of the mental element required for aiding and abetting under customary international law. Article 25(3)(c) requires that an accused act “[f]or the purpose of facilitating the commission” of a crime. In doing so, it provides for a mental element different from, and in addition to, the “knowledge” or “intent” requirements as defined in Article 30 of the Statute. Article 25(3)(c) echoes the approach originally developed by the American Law Institute in its Model Penal Code (“MPC”), which requires that an accomplice act “with the purpose of promoting or facilitating the commission of the offense” [Section 2.06(3)(a)(ii)]. By contrast, customary international law, as reflected in the jurisprudence of the ICTY [Šainović et al AJ, para. 1649], ICTR [Kalimanzira AJ, para. 86], SCSL [Taylor AJ, para. 436] and ECCC [Chea and Samphan TJ, para. 704], requires only that the accused know that his or her acts “assist in the commission of the offense”.

Given that the term “purpose” is not defined in the Rome Statute, the International Criminal Court (“ICC” or “Court”), will have to interpret the term in accordance with Article 21 on applicable law, and with general principles of treaty interpretation as set out in the Vienna Convention on the Law of Treaties (“VCLT”). Under Article 31 of the VCLT, the terms of Article 25(3)(c) must be interpreted “in good faith in accordance with [their] ordinary meaning”, “in their context and in the light of [the Rome Statute’s] object and purpose”. That “object and purpose” will, in our view, include the limitations on the Court’s jurisdiction stemming from the principles of gravity and complementarity [Rome Statute, Articles 1, 17(1)(d), 53(2)(c)].

Article 25(3)(c) represents a political compromise resulting from disagreement amongst common law and civil law representatives at the diplomatic conference for the establishment of the ICC, who had difficulty agreeing on the most appropriate means of limiting the scope of application of Article 25(3)(c) in the specific context of the ICC [Taylor AJ, para. 435; Scheffer Amicus Curiae Brief in John Doe v. Nestle, S.A., No. 10-56739, pp. 11-13]. The ICC was never intended to prosecute the full range of individuals who make a culpable contribution to the commission or attempted commission of a crime within the jurisdiction of the Court [Rome Statute, Article 17(1)(d)]. The Court was designed to function in a manner that is complementary to national criminal jurisdictions, which themselves maintain the primary responsibility for trying the vast majority of perpetrators.

This is not to say that the approach adopted in Article 25(3)(c) was an appropriate means for limiting the scope of cases which come before the ICC. In fact, an additional “purpose” requirement is problematic for a number of reasons. For example, it might have the effect of protecting individuals from liability where they take advantage of situations of armed conflict for financial gain, knowing that their conduct makes a substantial contribution to the commission of international crimes. Though one may be tempted to look to the MPC in an effort to resolve some of these problems, the Court may only have reference to the MPC itself, its commentaries, or the interpretation of MPC under domestic law, in one of two ways: (i) as a supplementary means of interpretation in accordance with Article 32 of the VCLT, to the extent that they informed the “preparatory work of the [Rome Statute] and the circumstances of its conclusion”; or (ii) as just one source, amongst others, from which the Court might derive a general principle of law under Article 21(1)(c) of the Statute.

It may not be necessary, however, to have recourse to the MPC to develop a workable interpretation of “purpose”. There is scope for the Court to interpret the “purpose” requirement broadly, and in a manner that minimises the divergence from customary international law. First, as a matter of evidentiary proof, where there is evidence that an accused had knowledge that his or her conduct would facilitate the commission of a crime, and nevertheless engaged in that conduct, the Court could infer that the accused acted for the purpose of facilitating the commission of that crime. Second, the Court could interpret the terms of Article 25(3)(c) as requiring only that facilitating the commission of a crime be a purpose of the accused’s conduct, but not the sole purpose. For example, evidence of a financial motive would not itself exclude a finding that the accused also acted for the purpose of facilitating the commission of a crime.

Regardless of the interpretation ultimately adopted by the Court, Article 25(3)(c) was not intended to reflect State practice and opinio juris and thus codify customary international law. Article 10 of the Rome Statute itself provides that its provisions should not “be interpreted as limiting or prejudicing in any way existing or developing rules of international law” for purposes other than cases directly before the Court. As David Scheffer, head of the U.S. delegation to the Rome Conference, has recently argued before the U.S. Ninth Circuit Court of Appeals, “[w]hile some other articles of the Rome Statute ended up reflecting customary international law, Article 25(3)(c) is not one of them.” [Scheffer Amicus Curiae Brief in John Doe v. Nestle, S.A., No. 10-56739, p. 9].

Yet ironically, in some instances, Article 25(3)(c) has been treated at the domestic level as if it were a reflection of customary international law. This has occurred in two ways: (i) through the direct incorporation of the Rome Statute (including Article 25) into domestic legislation by States implementing their obligations under the Statute; and (ii) where domestic courts improperly rely on Article 25 as a source of customary international law regarding individual criminal responsibility in their own jurisprudence (for example, in U.S. Alien Tort Statute litigation where reliance on Article 25(3)(c) is having a real and immediate impact on the scope of corporate liability for aiding and abetting international crimes) [Aziz v. Alcolac, Inc., 658 F.3d 388, 399–400 (4th Cir. 2011); Presbyterian Church of Sudan v. Talisman Energy, Inc., 582 F.3d 244, 259 (2d Cir. 2009)].

It would be wrong for States to limit their jurisdiction over international crimes in this way. Eschewing the recognised standard under customary international law in favour of the political compromise contained in Article 25(3)(c) would create an “impunity gap” which the Rome Statute’s principle of complementarity was designed to avoid. The very purpose of complementarity is that States maintain primary responsibility for prosecuting international crimes. It does not mean that the jurisdiction of States needs to mirror that of the ICC. The ICC was envisioned as an additional tool in the fight against impunity that would exercise jurisdiction over only those cases of most serious concern to the international community as a whole. States, however, can and should maintain the customary international law standard for aiding and abetting, so as to ensure that all parties who contribute to the commission of international crimes are held accountable.