Corps de l’article

The Crimes Against Humanity and War Crimes Act[1] was adopted to fulfil two main objectives : 1) to implement Canada’s obligations under the Rome Statute of the International Criminal Court[2] in order to ensure its ability to cooperate fully with investigations and prosecutions by the International Criminal Court (ICC) ; and 2) to retain and enhance Canada’s capacity to prosecute and punish persons accused of the “core” international crimes, namely genocide, crimes against humanity and war crimes[3]. At the heart of the system put in place by the ICC to ensure accountability for these crimes lies the principle of complementarity. States bear the primary responsibility to prosecute those responsible for international crimes. The ICC will exercise its jurisdiction only if the competent State is “unable” or “unwilling” to do so[4].

The criminalisation of core international crimes in domestic criminal legal systems is therefore essential for the success of the ICC and, more broadly, of the international criminal justice enterprise. The Rome Statute has spurred an effort by State parties to enact implementing legislation for such purpose, as well as to facilitate cooperation with the international body. Canada was one of the first States to adopt such legislation[5]. The War Crimes Act’s provisions with respect to jurisdiction, which provide for universal jurisdiction, as well as those regarding the substantive definitions of crimes, general principles of liability and available defences, are central for the fulfilment of the second objective cited above, namely the strengthening of Canada’s ability to prosecute the core international crimes. With respect to these various aspects, the Act presents an interesting mosaic of applicable law. Its definitions regarding offences refer essentially to international law, whereas the available defences, justifications and excuses are those of both Canadian law and international law, and the modes of participation in offences are exclusively those of Canadian law.

Without a doubt, principles of liability are fundamental in determining the criminal responsibility of a person. In this regard, contrary to the legislation of other states[6], the Canadian War Crimes Act does not rely on international law and does not, at least explicitly, allow judges to base themselves upon international law principles in the event that domestic ones are inadequate or insufficient to capture the essence of international crimes. The present study aims at offering a preliminary assessment — though limited in scope — as to whether this legislative choice is likely to create difficulties in prosecutions under the Act or unjustifiable incoherence with international law.

The issue of individual criminal responsibility is quite controversial in both the practice and theory of international criminal law. Stefano Manacorda notes the following :

Traditionally, the difficulty in attributing criminal behaviour (as defined in international criminal law) to individuals derives from the fact that classical criminal law categories are built on a “mononuclear” paradigm (one author, one fact, one victim). Such categories therefore seem largely inapplicable to macrocriminality. Oversimplifying, one could say that “individual” criminal responsibility, as a principle, is inadequate for explaining “collective” criminality[7].

The International Criminal Tribunal for the former Yugoslavia (ICTY) recognised in its early jurisprudence that international crimes “constitute manifestations of collective criminality [and that] the participation and contribution of the other members of the group is often vital in facilitating the commission of the offence in question[8]”. Individual culpability does not diminish with distance from the actual acts committed against specific victims. The reverse is often true. Adolf Hitler, Adolf Eichmann and the likes never laid a hand on their victims, but sent millions to their deaths[9]. Blameworthiness often increases “in tandem” with the distance separating the mastermind from the direct perpetrator[10].

Article 25 of the Rome Statute now contains detailed rules of individual criminal responsibility, a positive development from earlier codifications whose rules governing participation in offences tended to be sketchy and archaic. At Nuremberg, the description of modes of liability was quite basic[11]. The ad hoc tribunals could rely on somewhat more detailed statutory provisions[12], and they progressively adopted rules aimed at capturing the collective nature of international crimes, notably through the (controversial) doctrine of joint criminal enterprise.

A comparative assessment of Canadian law and international criminal law, as codified in the Rome Statute and as interpreted by the international tribunals, is highly relevant. Differences between the two sets of principles, or gaps in either, could potentially lead to admissibility issues before the ICC or at least create inconsistency in the scope of responsibility for the same crimes, depending on whether they are prosecuted in Canadian courts, or before the ICC or other international jurisdictions. For instance, if Canadian law was found not to contemplate a principle of liability encompassed by the Rome Statute, or to provide for one more limited in scope, Canada could find itself unable to prosecute in its domestic courts a person who would be punishable by the ICC for the same acts, thereby opening the door to a finding of admissibility of the case before the international body[13]. Moreover, if Canadian law was found to be broader than the Rome Statute, by allowing the prosecution of a participant in an offence that could not be prosecuted pursuant to the Rome Statute or under customary international law, this could potentially violate the principle of legality in cases of retroactive prosecutions or the Canadian courts be said to lack a basis in international law for the exercise of universal jurisdiction[14]. Finally, principles that were developed in international criminal law regarding liability have taken into account the particular nature of international crimes. Some forms of liability indeed allow persons who would be treated as accomplices under domestic principles to be treated as principals under international law. Other principles allow catching into the net of criminal culpability individuals who might escape liability under traditional domestic law principles[15]. Canadian modes of participation in offences may be found to be maladapted or in need of a different interpretation where applied to international crimes. The analysis of international jurisprudence in the interpretation of domestic principles of liability may therefore be of crucial importance, as was aptly confirmed by the Supreme Court of Canada in Mugesera[16]. The application of international criminal law in Canadian administrative proceedings emanating from the immigration and refugee fields should also not be underestimated.

Considering Canada’s position on international justice issues generally, and its role in the development of international criminal law in particular, the analysis of Canada’s legislative choices provides a good case study of the rapidly evolving trend of prosecuting international crimes at the national level. This may shed light on the fundamental issues that will be determining factors of its success, or failure. This paper forms part of a broader research project on various aspects of the War Crimes Act that looks, for instance, into jurisdictional issues, substantive definitions of the core crimes and of the separate offence of superior and command liability, as well as available defences, justifications and excuses. The project also explores in a comparative light other principles of liability and inchoate offences not addressed herein, such as counselling, complicity after the fact, attempt and conspiracy. Needless to say, only the overall picture can represent comprehensively the promises and limits of Canadian law as regards participation in international crimes. This paper thus offers a limited review of a few modes of participation in offences applicable to prosecutions under the War Crimes Act. It first examines the rules regarding principal liability (1), before turning to the principles substantiating accomplice liability, namely aiding, abetting and extended liability arising out of a common purpose (2). It further aims at offering a preliminary assessment of whether such domestic legal principles are well-suited to the prosecution of international crimes.

1 The “Commission” of Genocide, Crimes Against Humanity and War Crimes : Principal Liability

Article 25 (3) of the Rome Statute draws upon the experience of the ICTY and ICTR regarding principles of liability. The detailed provision on modes of participation in offences is accompanied by an obligation for the Prosecutor to include, in the document containing the charges against an accused, the legal characterisation of the facts as they accord not only with the substantive crimes defined at articles 6-8 of the Statute, but also with the precise forms of participation under articles 25 and 28[17]. Furthermore, once the Pre-Trial Chamber decides, in the confirmation of charges hearing[18], that there is sufficient evidence to establish substantial grounds to believe that an accused has committed a crime as a principal, it will not look into the accused’s liability according to the modes of accessory liability provided for at paragraphs (b) to (d) of article 25 or according to superior responsibility under article 28[19]. This approach is theoretically different from that of Canadian law, which imposes no obligation on the Prosecutor to specify whether a person is accused as a principal or as an accomplice[20].

Article 25 (3) (a) of the Rome Statute provides the most comprehensive provision thus far in an international statute regarding principal liability. It states that

a person shall be criminally responsible and liable for punishment for a crime within the jurisdiction of the Court if that person :

(a) Commits such a crime, whether as an individual, jointly with another or through another person, regardless of whether that other person is criminally responsible.

The Rome Statute thus criminalises individual commission but also, explicitly, joint commission and commission through another person, regardless of whether that other person is criminally responsible. The first part of this paper compares principal liability for genocide, crimes against humanity and war crimes pursuant to the War Crimes Act with equivalent principles under the Rome Statute or as they have been applied by the ad hoc tribunals.

1.1 Individual Commission

As regards individual perpetration, the applicable principles are relatively straightforward in both legal systems and should not raise any difficulties as far as their respective scopes of application are concerned. In the decision on the confirmation of charges of Germain Katanga and Mathieu Ngudjolo Chui, the Pre-Trial Chamber described as a principal for “commission of the crime as an individual” he who “physically carries out all elements of the offence”[21]. Individual perpetration is defined similarly in Canadian law[22]. Since the Rome Statute has provided principal liability with a broader scope by including explicitely co-perpetration and perpetration through another person, the ICC will less likely follow the tendency of the jurisprudence of the ad hoc tribunals to expand the notion of individual perpetration.

The ad hoc tribunals’ statutes do not provide guidance as to what is meant by “commission”, leaving it to the Chambers to articulate its proper meaning. In Tadić, the Appeals Chamber defined it as “the physical perpetration of a crime by the offender himself, or the culpable omission of an act that was mandated by a rule of criminal law[23]”. However, the notion of perpetration was extended by the ad hoc tribunals, in an effort to reflect properly the collective nature of international crimes. The Appeals Chamber in the same judgment concluded that commission was not limited to direct and physical perpetration, but also contemplated participation in a joint criminal enterprise, a doctrine which will be discussed in more detail below[24]. In the same vein, the ad hoc tribunals have also either read into the notion of commission the explicitly absent concept of co-perpetration[25] or given an expansive interpretation of perpetration so as to include those whose actions were “as much an integral part of the genocide as were the killings which [they] enabled[26]”.

Such liberal interpretations of the tribunals’ statutes illustrate the judges’ uneasiness with a narrow interpretation of principal liability for international crimes and the need to reflect adequately upon various degrees of direct, even if remote, participation in a crime. The ad hoc tribunals’ insistence that those bearing serious responsibility for international crimes be recognised as principals rather than as accomplices, may be explained by a general tendency to impose harsher sentences on principals[27]. As we shall see, the differentiation between principals and accomplices does not carry similar consequences as regards sentencing in Canadian law, at least in principle[28]. It may be that Canadian judges will feel more comfortable than their international counterparts in applying principles of accomplice rather than principal liability for masterminds and other similar “non-physical” participants to international crimes. Complicity will be discussed further below. It is fitting at this juncture to examine the notions of co-perpetration and perpetration through another person as they exist under the Rome Statute and in Canadian law.

1.2 Co-perpetration

Apart from individual commission, the Rome Statute also explicitly criminalises co-perpetration. A Pre-Trial Chamber of the ICC, in its decision on the confirmation of charges of Thomas Lubanga, developed rather complex principles in this regard[29]. First, the Pre-Trial Chamber elaborated on the conceptual distinction between principals and accomplices. The Pre-Trial Chamber rejected what it termed objective and subjective approaches : “The objective approach […] focuses on the realisation of one or more of the objective elements of the crime. From this perspective, only those who physically carry out one or more of the objective elements of the offence can be considered principals to the crime[30].” This is the approach of Canadian law. As for the subjective approach, it focuses on the state of mind of the accused : “only those who make their contribution with the shared intent to commit the offence can be considered principals to the crime, regardless of the level of their contribution to its commission[31]”. As the Chamber noted, this is the approach adopted by the ad hoc tribunals through aspects of the doctrine of joint criminal enterprise.

The Chamber elaborated on a third approach, which is not restricted to the physical perpetrators but also includes those who control the commission of the crime in one way or another. This approach identifies principals as those who control the commission of the offence and who are aware that they have such control because they either (1) physically perpetrated the objective elements of the offence (commission) ; (2) controlled the will of those who physically perpetrated the objective elements of the offence (commission through another person) ; or (3) held, with others, the control of the offence by reason of the essential tasks that were assigned to them (co-perpetration)[32].

With respect to co-perpetration, the Chamber established an intricate test aimed at determining whether the alleged co-perpetrator had sufficient control over the offence. Control was said to exist where objective elements were established comprising (1) the existence of a plan or agreement between two or more persons and (2) an essential contribution by each co-perpetrator to the realisation of the offence. Subjective elements also needed to be established requiring (1) that the suspect have the necessary mens rea regarding the crime at issue ; (2) that the suspect and other co-perpetrators share in the knowledge that the realisation of the objective elements of the crime might result from the implementation of their plan ; and (3) that the suspect know of the factual circumstances allowing him to exercise joint control over the commission of the crime[33].

It remains to be seen whether this approach will become solid ICC jurisprudence. The current principles reveal a limited similarity with Canadian law on the issue. Section 21 (1) (a) of the Criminal Code, despite its use of the singular, certainly does not exclude a plurality of perpetrators[34]. The distinction between a co-perpetrator and an accomplice turns on the issue whether the parties acted “in concert, pursuant to a common motive[35]”. There must be “[a]n intention to act in concert with another [which] is […] an integral part of the doctrine that, where one person commits one element of the offence and another person commits another element of the offence, both are co-perpetrators of the offence[36]”. The analogy with the Rome Statute ends here. In Canadian law, it is also essential that the accused must have actually committed one of the essential elements of the actus reus : “il ne suffit pas de contribuer à la perpétration de l’infraction mais bien d’en réaliser au moins un élément essentiel (sans quoi plus rien ne départage les alinéas a) et b) de l’article 21 (1)[37]”. It is not the importance of the contribution of a person in the commission of an offence that will determine whether he or she is a co-perpetrator or an accomplice, but whether he or she has accomplished a part of the actus reus[38]. There lies the most fundamental difference between the current interpretation of the Rome Statute and Canadian law. In Lubanga, the Pre-Trial Chamber clearly stated that the contribution needed not be linked to the actus reus at the execution stage and thus rejected the view that only those who physically commit an aspect of the actus reus can be considered principals to the crime[39].

Co-perpetratorship in Canadian law is thus much more limited in scope than under the Rome Statute, which was strongly influenced by continental European legal models. It excludes individuals who have not physically committed an element of the actus reus of an offence that was materially perpetrated by others, but to which they have nonetheless contributed in an essential manner and according to a common concerted plan.

1.3 Perpetration Through Another Person

Article 25 (3) (a) of the Rome Statute also criminalises commission “through another person[40]”. The concept of innocent agency is known to Canadian law. In such instances, a person who has not personally committed any material element of the offence may nevertheless be considered a principal to it[41]. The innocent agent is used by the perpetrator to commit an unlawful act. He or she is either unaware that the act is unlawful, or cannot be held responsible because of mental illness or minority, for instance, or because he or she was acting under duress.

However, article 25 (3) (a) of the Rome Statute does not limit itself to perpetration through an innocent agent. It specifies that commission through another person occurs “regardless of whether that other person is criminally responsible”. It thus recognises the possibility of perpetration through a guilty agent, a concept which it distinguishes from co-perpetration. This concept is foreign to Canadian law and may originate from similar doctrines found in States within the continental European legal tradition[42]. In Katanga and Chui, the ICC Pre-Trial Chamber I elaborated on this doctrine, which it also identified by the terms : “perpetrator behind the perpetrator (Täter hinter dem Täter)[43]”. It explained :

The underlying rationale of this model of criminal responsibility is that the perpetrator behind the perpetrator is responsible because he controls the will of the direct perpetrator. As such, in some scenarios it is possible for both perpetrators to be criminally liable as principals : the direct perpetrator for his fulfilment of the subjective and objective elements of the crime, and the perpetrator behind the perpetrator for his control over the crime via his control over the will of the direct perpetrator[44].

As to the applicability of the doctrine to cases most relevant to international criminal law, the Chamber noted that they “are those in which the perpetrator behind the perpetrator commits the crime through another by means of ‘control over an organisation’ (Organisationsherrschaft)[45]”. It stated :

The leader must use his control over the apparatus to execute crimes, which means that the leader, as the perpetrator behind the perpetrator, mobilises his authority and power within the organisation to secure compliance with his orders. Compliance must include the commission of any of the crimes under the jurisdiction of this Court[46].

The somewhat controversial[47] notion of perpetration through a guilty agent seems destined to a great future. On 4 March 2009, an ICC Pre-Trial Chamber issued an arrest warrant against the President of Sudan, Omar Al Bashir, for crimes committed in Darfur, based on this concept[48]. The majority of the Chamber found that there were reasonable grounds to believe that Omar Al Bashir and other high-ranking Sudanese political and military leaders directed the branches of the State apparatus and led them, in a coordinated manner, to implement jointly a common plan consisting of a counter-insurgency campaign against rebel groups, as well as an unlawful attack on a part of the civilian population of Darfur perceived as being close to the rebels. The targeted tribes were to be subjected to forcible transfers and acts of murder, extermination, rape, torture, and pillage[49]. The Chamber also found that “there are reasonable grounds to believe that Omar Al Bashir, as de jure and de facto President of the State of Sudan and Commander-in-Chief of the Sudanese Armed Forces at all times relevant to the Prosecution Application, played an essential role in coordinating the design and implementation of the common plan[50]”. He stands accused of numerous charges of crimes against humanity and war crimes.

Perpetration through a guilty agent and co-perpetration will likely play an essential role at the ICC in securing accountability of masterminds and the likes who control the commission of core crimes without themselves executing an aspect of the actus reus of such crimes. Considering the absence, or limited scope, of these modes of liability under Canadian law, principal liability will be inapplicable for a wide range of persons responsible for the “most serious crimes of concern to the international community as a whole[51]”. It may, or it may not, be a consequential limitation of Canadian law. It will in fact be problematic only if participants to international offences escape justice because their contribution cannot be encompassed adequately by other applicable principles.

This leads us to a comparative analysis of accomplice liability and the manner in which it may apply to international crimes in Canada. This enquiry may shed light on the intricate question of whether the War Crimes Act, by referring exclusively to principles of liability available in Canadian law, has equipped Canadian courts well enough to embark, as they must, on the difficult journey of war crimes prosecution.

2 Complicity : Aiding, Abetting and Common Purpose

Accomplice liability is called to play an essential role in the endeavour of imputing liability to the numerous individuals who may have contributed to the commission of international crimes. This is particularly so in situations where a restrictive view of principal liability is adopted, as we have seen is the case in Canadian law. The aim of this second Part is to assess, in light of the international principles, whether and how Canadian principles of accomplice liability can reflect the reality of international crimes, in full respect of the principle of complementarity, while staying true to the canons of domestic criminal law.

2.1 Modes of Participation Included in the Crimes against Humanity and War Crimes Act

Sections 4 (1) and 6 (1) of the War Crimes Act state that “[e]very person is guilty of an indictable offence who commits” [emphasis added] one of the core crimes. It specifically adds, at paragraph (1.1), that “[e]very person who conspires or attempts to commit, is an accessory after the fact in relation to, or counsels in relation to, an offence referred to in subsection (1) is guilty of an indictable offence.”

The Act does not explicitly include important forms of participation in a crime provided for at section 21 of the Criminal Code, namely aiding, abetting and extended liability arising out of a common purpose[52]. This has led to debates and arguments as to whether complicity (before the fact) can form a basis for prosecution of a core international crime in Canada[53]. Fortunately, the negative view does not withstand scrutiny. Indeed, there is no doubt that section 34 (2) of the Interpretation Act[54] allows the provisions of the Criminal Code on modes of participation to apply to offences prosecuted under the War Crimes Act[55].

Any other interpretation would be both unjustifiable and unduly restrictive and would run contrary to the principle of complementarity, which requires that national prosecutions be able to be conducted for the whole range of crimes and perpetrators within the jurisdiction of the ICC. The inclusion of paragraph (1.1) was probably deemed necessary because the offences designated therein are autonomous offences in the Criminal Code and not modes of participation in crimes. Interestingly, as we shall see, some of these offences are not criminalised as such in international criminal law.

2.2 Aiding and Abetting

In Canadian law as well as in international criminal law, aiding and abetting are separate forms of secondary liability, but they are typically used and referred to together[56]. They will be treated in a similar fashion here except where distinctions are warranted.

Paragraphs 21 (1) (b) and (c) of the Criminal Code provide :

(1) Every one is a party to an offence who […]

(b) does or omits to do anything for the purpose of aiding any person to commit it ; or

(c) abets any person in committing it.

Article 25 (3) (c) of the Rome Statute provides that

a person shall be criminally responsible and liable for punishment for a crime within the jurisdiction of the Court if that person : […]

(c) For the purpose of facilitating the commission of such a crime, aids, abets or otherwise assists in its commission or its attempted commission, including providing the means for its commission.

Aiding and abetting is likely to play a less important role before the ICC than before Canadian courts, because of the wide scope given to principal liability in the Rome Statute. Similarly, aiding and abetting has had a limited significance before the ad hoc tribunals, which relied extensively on joint criminal enterprise as a mode of principal liability[57]. For this reason, Canadian judges, when applying the notion of aiding and abetting to international crimes, should not only be guided by the principles developed by the international tribunals as regards aiding and abetting. Canadian jurisprudence should also consider as informative the ICC jurisprudence on principal liability, notably co-perpetration and commission through another person, as well as the ad hoc tribunals’ jurisprudence on the doctrine of joint criminal enterprise. This will be particularly important where the accused is a leader, a mastermind of crimes or a mid-level participant whose contribution to the actual crime(s), and links to the physical perpetrators, are more remote than is often the case with the accomplice under traditional common law.

2.2.1 Comparative Analysis

Before turning to differences, it may be useful to note at the outset that Canadian law and international criminal law (as interpreted by the ad hoc tribunals) coincide in various respects regarding aiding and abetting. For instance, both legal systems are in broad agreement concerning the consequences of mere presence and omissions. Though there are nuances, it will not be necessary to delve into this issue here[58]. Furthermore, in both systems, the guilt of the aider or abettor does not depend upon the principal being prosecuted or even identified[59]. Moreover, it is entirely possible that the principal may not even be aware of the accused’s contribution[60]. Both legal systems also criminalise complicity of an attempted offence, though the consequences of such conduct may end up being quite different. Indeed, article 25 (3) (f) of the Rome Statute criminalises attempt, but the language used seems to indicate that liability will be incurred for the main crime that was attempted rather than for a self-standing inchoate crime of attempt[61]. This distinction has an impact, among other things, on aiders and abettors of an uncompleted offence. Indeed, under the Rome Statute, an accomplice to an offence that was only attempted will be prosecuted for that offence, namely genocide, crimes against humanity and war crimes[62]. In Canada, the same person would be prosecuted for the self-standing offence of attempt[63]. Normally, under Canadian law the sentence is much more lenient for attempt than for the actual crime[64]. However, exceptionally, the War Crimes Act applies the same sentence for the committed crime as for the inchoate offences contemplated therein, including attempt, thereby ensuring greater consistency with the Rome Statute[65]. There have been no convictions for attempt under international criminal law, so it is difficult to anticipate how its gravity will be assessed for the purposes of sentencing. The Court would no doubt refer to general principles, in accordance with article 21 (2) of the Statute, and probably find that virtually all legal systems punish attempts less severely than crimes that actually take place.

A first major difference between international criminal law and Canadian law has already been identified but should be highlighted in more detail here. There has been a tendency in international criminal law to consider aiders and abettors as having a lower degree of criminal responsibility than principals. Consequently, it was held that “aiding and abetting is a form of responsibility which generally warrants a lower sentence than […] responsibility as a co-perpetrator”[66]. In Canada, as in English common law, principal and secondary parties are charged with the same offence and are subject in principle to the same sentence[67]. The trial judge, however, retains full discretion to impose a sentence that is in accordance with the degree of culpability of each participant[68]. The practical impact of the principal/secondary dichotomy is therefore much less significant in Canada. However, the distinction between principals and accomplices remains relevant because, among other things, the requirements are different regarding the material and mental elements, causality and defences. Considering the narrow scope of principal liability, the limits of the law on secondary participation will in many cases represent the limits of Canadian law regarding liability for international crimes.

Regarding the actus reus of aiding and abetting, it may be noted that international criminal law requires that the assistance have an effect, usually described as substantial, upon the perpetration of the crime. The classic formulation of this principle emanates from the Tadić appeal : “The aider and abettor carries out acts specifically directed to assist, encourage or lend moral support to the perpetration of a certain specific crime (murder, extermination, rape, torture, wanton destruction of civilian property, etc.), and this support has a substantial effect upon the perpetration of the crime[69].” Though the act or omission must have a substantial effect, there is no requirement of a causal connection[70].

In Canadian law, the effect of the assistance or encouragement on the commission of the crime is probably irrelevant, and is, in any case, certainly insufficient. As Côté-Harper, Rainville and Turgeon have explained, “[l]a culpabilité du complice n’est […] en rien tributaire de l’efficacité de son aide […] Le droit permet en définitive la condamnation du complice en l’absence de preuve de l’utilité de son rôle dans la perpétration du crime[71]”. This is mainly because aiding (and abetting) requires further proof of the specific intent to assist in the commission of the crime[72]. Therefore, one can safely advance that the ad hoc tribunals’ requirement that the aiding and abetting have a substantial effect upon the committed or attempted crime is a notable difference with Canadian law (where, however, the conduct of an “accomplice” that is in and of itself a “significant contributing cause” of a prohibited result would qualify for principal liability)[73]. The effect of the aiding or abetting on the perpetration of the crime is thus not a decisive issue in Canadian law, if at all it is an issue. Notably, the Rome Statute does not — at least expressly — require that the aiding or abetting make a substantial contribution to the crime[74].

As regards the mens rea of aiding and abetting, there is a noteworthy difference between Canadian law and international criminal law : Canadian law requires a specificintention to assist, whereas the ad hoc tribunals’ jurisprudence requires the accused to have knowledge that his or her conduct will assist a specific crime. In Tadić, the Appeals Chamber makes this clear : “In the case of aiding and abetting, the requisite mental element is knowledge that the acts performed by the aider and abettor assist the commission of a specific crime by the principal[75].” Whether and to what extent there exists a significant difference between intent, on the one hand, and knowledge, on the other hand, is unclear. In Canada, the intention to assist has sometimes been described in terms of knowledge. In Hibbert, in an attempt to define the meaning of “purpose” in the context of section 21 (1) (b) of the Criminal Code, the Supreme Court of Canada agreed with commentators “that a person who consciously performs an act knowing the consequences that will (with some degree of certainty) flow from it ‘intends’ these consequences or causes them ‘on purpose’, regardless of whether he or she desired them[76]”. Don Stuart adds : “Subjective mens rea is a concept of cognitive awareness and no more[77]”. Noting that purpose had been equated with intention, Eric Colvin and Sanjeev Anand write that “[i]t is therefore possible for someone to aid an offence in the sense that it is known that aid will result[78]”. Kent Roach, on the other hand, declares that “[k]nowledge is a slightly lower form of subjective mens rea than intent or purpose[79]”. It may be that the only true distinction would be if intent was interpreted as importing “the idea of ‘desire’ into the definition”[80]. This approach was rejected by the Supreme Court in Hibbert[81].

The distinction may also be dependant upon the degree of likelihood of the consequence. In Chartrand, the expression “with intent to” at section 281 of the Criminal Code was interpreted by the Supreme Court of Canada as requiring proof that the accused either had a conscious purpose to cause the event or that the consequence was “foreseen by the accused as a certain or substantially certain result[82]”. According to this view, which also finds support in English common law[83], the degree of likelihood of the consequence serves to distinguish between intent and recklessness. With respect to the former, the result must be certain or substantially certain whereas the latter will be satisfied with the accused’s subjective awareness of the possibility or probability of the result occurring. Therefore, both intention and recklessness can be expressed in terms of subjective “knowledge”. The distinction will depend on the knowledge of the degree of likelihood of the occurrence of the prohibited result[84]. The Rome Statute provides further support for this interpretation. Article 30 (2) states :

2. For the purposes of this article, a person has intent where :

(a) In relation to conduct, that person means to engage in the conduct ;

(b) 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 [emphasis added].

The latter provision thus allows intent to be established either in the presence of a conscious purpose (the accused means to cause the consequence) or if there exists subjective foresight that the result is certain or substantially certain to occur (willoccur in the ordinary course of events). It may be added that paragraph 3 under article 30 defines “knowledge” exactly like “intent” as regards consequences : “‘knowledge’ means awareness that […] a consequence will occur in the ordinary course of events”. Accordingly, article 30 likely excludes recklessness as a basis for liability. The ICC Pre-Trial Chamber in Lubanga confirmed that recklessness is not included under article 30, but then adopted a very expansive interpretation of intention and knowledge, equating them with what Canadian law would consider to be recklessness. Indeed, the Pre-Trial Chamber accepted that the volitional element included not only conscious purpose and foresight of the necessary (certain) outcome of the conduct, but also awareness of a risk, that could besubstantial or even low[85]. However, another Pre-Trial Chamber, in a posterior and well-reasoned decision, was unequivocal that article 30 is limited to “actual intent or will to bring about the material elements of the crime” and awareness “that those elements will be the almost inevitable outcome of his acts or omissions”. It clearly rejected any lower standard[86].

The ad hoc tribunals have not made clear whether the accused must know that his assistance “will” or “may” or “will probably” assist in the commission of the crime. Some commentators have rejected that “knowledge” in this case includes recklessness[87]. This view is supported by the first element (strangely) included in the actus reus of aiding and abetting by the ad hoc tribunals, namely that “it must be proven that the alleged aider and abettor committed acts specifically aimed at assisting, encouraging, or lending moral support for the perpetration of a specific crime[88]”.

Assuming intent/purpose is a stricter requirement than knowledge, it is important to note that the Rome Statute seems in any case to impose a higher degree of mens rea for aiding and abetting than the ad hoc tribunals’ jurisprudence. Article 25 (3) (c) reads : “For the purpose of facilitating the commission of such a crime, aids, abets or otherwise assists in its commission or its attempted commission[89]”. This wording is reminiscent of section 21 (1) (b) of the Criminal Code. It will be recalled that the Supreme Court of Canada interpreted “purpose” in that provision as “being essentially synonymous with ‘intention’”[90]. It remains to be seen how the ICC will interpret the purpose requirement, and speculations in this regard have been many[91]. However, at least until such time as the ICC decides conclusively on the matter, the similarity in language as between the Criminal Code and the Rome Statute favours an interpretation of Canadian law regarding mens rea in harmony with the Rome Statute, if not with international customary law as it has been interpreted by the ad hoc tribunals. Complicity by recklessness is therefore excluded both in Canadian law and under the Rome Statute, and specific intent to assist is required[92].

As a further mens rea requirement of aiding and abetting, international criminal law and Canadian law require that the accused be aware, at a minimum, of the essential elements of the substantive crime for which he or she is charged with responsibility as an aider and abettor[93]. The ad hoc tribunals have concluded that this includes awareness of the mental state of the physical perpetrator[94]. This requirement of knowledge of the physical perpetrator’s intent applies equally to specific-intent crimes or underlying offences such as genocide or persecution as a crime against humanity[95]. However, the accomplice need not share the genocidal intent of the principal[96].

The knowledge-based mens rea in international law for aiding and abetting genocide has been criticised by some commentators for failing to respect the fundamental characteristic at the heart of the crime of genocide, namely its specific intent[97]. Others have opined that there is no practical difference between wilful assistance with subjective knowledge of the principal’s intent to destroy a protected group and intention to destroy such group[98]. In light of the above discussion, provided that it is subjective knowledge at the level of reasonable certainty that is required of the accomplice to genocide with respect to the specific intent of the principal, there is indeed no effective difference with requiring that he or she share that intent.

Despite the standard for murder adopted in Kirkness[99], the international wording should be adopted in Canada for genocide, crimes against humanity and war crimes. Framing the requisite intent in terms of subjective knowledge of the principal’s specific intent (at the level of reasonable certainty) keeps close track with the reality of assistance to the worst crimes, while yielding effectively the same result as requiring that the accomplice share that intent. Furthermore, any debate as to whether the accomplice is held to a lower form of mens rea than the principal (knowledge vs. specific intent) becomes less relevant in the Canadian context, where the specific intention to assist or to encourage is always a requirement for aiding or abetting. This specific intent, coupled with subjective knowledge of the principal’s specific intent at the level of reasonable certainty, creates no asymmetry in the requisite mens rea of the accomplice and that of the principal.

There is therefore relative consistency of the applicable principles in both legal systems regarding accomplice liability. However, Canadian courts, like their international counterparts, will be faced with the lingering difficulty of establishing individual criminal responsibility in the context of collective criminality. For instance, is the criminal liability of a high-ranking official adequately captured by the current interpretation of aiding and abetting ? Can or should the liability of a president who masterminds the commission of massive and systematic crimes be determined according to whether he or she “intended to assist” in the commission of those crimes ? Clearly, his or her intent and actual role in the execution of the crimes goes beyond what is traditionally understood as accomplice liability. Furthermore, aiding and abetting is built around the idea that the material element of complicity is intrinsically linked to the material element of the impugned offence[100]. Though the material element requirement for aiding and abetting is relatively easy to establish — a few words of encouragement will suffice for abetting, for instance —, it remains closely related to the actual crime that was aided or abetted[101]. However, an accused’s contribution to international crimes may be quite remote from the principal’s (material perpetrator’s) actions. Aiding and abetting under section 21 (1) (b) and (c) must be interpreted so as to cover adequately the various types of participation in large collective enterprises leading to massive crimes.

The ad hoc tribunals have gone to great lengths in attempting to establish individual responsibility in that context. They have developed the doctrine of joint criminal enterprise in response to the observation that the commission of genocide, crimes against humanity and war crimes frequently occur through the implication of a plurality of persons, some of whom remain at a distance from the actual perpetrators[102]. The doctrine has indeed served as a basis for a large proportion of the convictions entered by the ad hoc tribunals[103]. The Rome Statute has not espoused the doctrine of joint criminal enterprise to its full extent. However, to respond to the same concerns and with comparable results, it adopts, as we have seen, a broad definition of principal liability through an expansive interpretation of co-perpetration and an innovative concept of perpetration-by-means. With these recent developments in mind, the following analysis of the joint criminal enterprise doctrine will allow us to assess whether and how aiding and abetting in Canadian criminal law can respond specifically to the reality of collective mass criminality.

2.2.2 Joint Criminal Enterprise

The recognition as part of international customary law[104] of the joint criminal enterprise doctrine within principal liability[105] has been the object of substantial academic and jurisprudential debate. The leading judgment is once again the Tadić Appeal. In that case, the ICTY Appeals Chamber identified three classes of cases where liability arising from a common purpose might arise. These categories have come to be labelled basic joint criminal enterprise (JCE 1), systemic joint criminal enterprise (JCE 2 ; the concentration camp cases) and extended joint criminal enterprise (JCE 3)[106]. The actus reus was said to be common to all three types of joint criminal enterprise :

i. A plurality of persons. They need not be organised in a military, political or administrative structure […]

ii. The existence of a common plan, design or purpose which amounts to or involves the commission of a crime provided for in the Statute. There is no necessity for this plan, design or purpose to have been previously arranged or formulated. The common plan or purpose may materialise extemporaneously and be inferred from the fact that a plurality of persons acts in unison to put into effect a joint criminal enterprise.

iii. Participation of the accused in the common design involving the perpetration of one of the crimes provided for in the Statute. This participation need not involve commission of a specific crime under one of those provisions (for example, murder, extermination, torture, rape, etc.), but may take the form of assistance in, or contribution to, the execution of the common plan or purpose[107].

Later cases contributed in clarifying the nature of the actus reus. For our purposes, suffice it to note the following :

  1. Membership in the group having a common criminal purpose is not sufficient per se[108]. There is no guilt by “mere association”[109].

  2. The doctrine of joint criminal enterprise is not restricted to small-scale cases and can apply to large criminal enterprises[110]. Furthermore, “[w]hile a Trial Chamber must identify the plurality of persons belonging to the JCE, it is not necessary to identify by name each of the persons involved. Depending on the circumstances of the case, it can be sufficient to refer to categories or groups of persons[111].”

  3. The accused need not have performed any part of the actus reus of the perpetrated crime[112]. The contribution of the accused need not be necessary or substantial, but some chambers have affirmed that it should at least be a significant contribution to the common criminal purpose[113]. Several factors may serve to evaluate the level of participation, such as the accused’s position of authority, efforts made to impede the effective functioning of the system, etc.[114] Causation is irrelevant[115].

  4. If the plan fundamentally changes, the accused is only liable for the crimes which related to the original plan to which he had subscribed[116]. However, a joint criminal enterprise can come to embrace expanded criminal means (additional crimes not originally encompassed in the joint criminal enterprise), as long as the evidence shows that the members agreed (explicitly or not) on this expansion of means to achieve the common objective[117].

  5. The actual perpetrator need not be a party to the common purpose. Members of a joint criminal enterprise can incur liability for crimes committed by principal perpetrators who were non-members of the enterprise, provided that it has been established that the crimes can be imputed to at least one member of the enterprise and that this member — when using the principal perpetrators — acted in accordance with the common objective[118].

Following Tadić, the level of mens rea changes according to the category of joint criminal enterprise which is concerned :

With regard to the first category, what is required is the intent to perpetrate a certain crime (this being the shared intent on the part of all co-perpetrators). With regard to the second category (which, as noted above, is really a variant of the first), personal knowledge of the system of ill-treatment is required (whether proved by express testimony or a matter of reasonable inference from the accused’s position of authority), as well as the intent to further this common concerted system of ill-treatment. With regard to the third category, what is required is the intention to participate in and further the criminal activity or the criminal purpose of a group and to contribute to the joint criminal enterprise or in any event to the commission of a crime by the group. In addition, responsibility for a crime other than the one agreed upon in the common plan arises only if, under the circumstances of the case, (i) it was foreseeable that such a crime might be perpetrated by one or other members of the group and (ii) the accused willingly took that risk[119].

Despite the use of the term “foreseeable”, rather than “foreseen”, which hints at a requirement of objective liability, the Chamber made clear that in the third category of joint criminal enterprise, “more than negligence is required. What is required is a state of mind in which a person, although he did not intend to bring about a certain result, was aware that the actions of the group were most likely to lead to that result but nevertheless willingly took that risk. In other words, the so-called dolus eventualis is required (also called ‘advertent recklessness’ in some national legal systems)[120].” This passage nevertheless opens the door to criminal liability for genocide on the basis of recklessness, contrary to the specific intent normally required for this crime[121], an issue to which we will return below.

As mentioned, the Rome Statute does not explicitly provide for joint criminal enterprise. Article 25 (3) (d) contemplates a “residual” form of liability for crimes committed by a group of persons acting with a common purpose[122]. It envisages a form of accomplice liability, rather than principal liability as the doctrine of joint criminal enterprise was construed by the tribunals[123]. It reads :

[A] person shall be criminally responsible and liable for punishment for a crime within the jurisdiction of the Court if that person […]

(d) In any other way contributes to the commission or attempted commission of such a crime by a group of persons acting with a common purpose. Such contribution shall be intentional and shall either :

(i) Be made with the aim of furthering the criminal activity or criminal purpose of the group, where such activity or purpose involves the commission of a crime within the jurisdiction of the Court ; or

(ii) Be made in the knowledge of the intention of the group to commit the crime [emphasis added].

The first category of joint criminal enterprise is clearly encompassed by the interpretation given to co-perpetration in Lubanga[124]. The second type could possibly be encompassed by article 25 (3) (d). However, the Statute probably does not cover the extended liability for crimes not initially contemplated in the common criminal design. The Lubanga confirmation of charges decision[125] and the general provision on mens rea found at article 30 of the Rome Statute, which excludes recklessness, prevent an interpretation of any paragraph under article 25 (3) in a manner akin to the third category of joint criminal enterprise liability as construed by the ad hoc tribunals[126]. Interestingly, though, article 25 (3) (d) (i) refers to intent to contribute to the common purpose of a group which involves commission of a crime within the jurisdiction of the Court[127]. Arguably, and paraphrasing article 30 (2) of the Rome Statute, should it be established that the accused was subjectively aware that the additionaloffences would (certainly or almost certainly) occur in the ordinary course of events, the intentional requirement would be satisfied. The Rome Statute thus probably excludes only the “outer limits of type three joint criminal enterprise”[128].

In light of the above discussion, let us look at a particular provision of the Criminal Code which extends the liability of an accomplice to crimes that were committed as part of a common criminal purpose. Section 21 (2) reads :

(2) Where two or more persons form an intention in common to carry out an unlawful purpose and to assist each other therein and any one of them, in carrying out the common purpose, commits an offence, each of them who knew or ought to have known that the commission of the offence would be a probable consequence of carrying out the common purpose is a party to that offence.

A cursory glance at this provision suffices to notice a certain resemblance with the third category of joint criminal enterprise developed by the ad hoc tribunals, which also provides for extended liability for additional offences arising out of a common purpose. Unsurprisingly, both doctrines have been criticized for imposing a sort of vicarious liability, and for violating the presumption of innocence by diminishing the minimum level of mens rea for crimes which require specific intent for principals. Still, there exist noteworthy differences between the two doctrines.

For our purposes here, three points will be stressed regarding section 21 (2) and the manner in which it could operate in the context of prosecutions for genocide, crimes against humanity and war crimes in Canada. First, section 21 (2) only applies to additional offences which were not agreed to between the participants in the joint criminal enterprise[129]. Thus section 21 (2) cannot serve to capture joint criminal enterprises identified by the ad hoc tribunals of the first and second categories.

Second, it seems that the requirement of a common illegal design is more stringent in Canadian law than it is under the international joint criminal enterprise doctrine. Indeed, as noted above, in international law there is no need for the plan, design or purpose to have been previously arranged or formulated. It can materialise extemporaneously[130]. In Canada, the temporal element of the common purpose is not fully settled, but arguably, the wording of the provision, which, in French, imposes the existence of a “projet”, as well as some judicial interpretations, point towards the need for the criminal purpose to have been formed prior to the commission of the crime[131]. It would thus exclude spontaneous group violence[132].

A related point should be noted concerning the requirement of a common purpose : section 21 (2) requires that two or more persons “form an intention in common to carry out an unlawful purpose” (“forment ensemble”). This requirement, added to the necessity that there be intention mutually to assist each other in the commission of the offence, implies that the participants to the criminal enterprise must be identifiable, and must include the actual perpetrator(s). This requirement effectively shields leaders of criminal organisations, who are usually more remote from the principal[133]. It also prevents application of the doctrine to large criminal enterprises where full membership is difficult to establish. The difficulty in applying section 21 (2) in such cases is not linked to the leader’s intention to carry out an unlawful purpose or even to assist therein[134], but to the requirement that such intention was actually formed in common with the main perpetrator. The jurisprudence of the ad hoc tribunals is much less stringent in this regard. First, the doctrine is now probably applicable to large joint criminal enterprises[135] and it is not necessary to establish precisely the membership of the joint criminal enterprise[136]. Secondly, recent developments impose liability on members even if the main perpetrator is not part of the joint criminal enterprise. As already noted, there is no need for an understanding or agreement to commit a particular crime between the accused and the principal perpetrator[137]. It would be difficult, as it is currently interpreted, to allow such an extension of the common purpose at section 21 (2)[138]. This is probably a good thing. The broad reach of the third category of joint criminal enterprise as interpreted by the ad hoc tribunals is one of the most criticised aspects of international criminal law, as is evidenced by the explicit exclusion of its outer limits in the Rome Statute.

Our third remark is concerned with the level of mens rea required of the accomplice for the extended liability to arise under section 21 (2) of the Criminal Code for international offences proscribed by the War Crimes Act. Liability for offences other than those agreed to in the common plan will be restricted in Canada to offences actually foreseen by the accused. Subjective foresight of the additional offences will be required. Indeed, the Supreme Court of Canada has decided that a few offences require a special degree of mens rea by reason mainly of the social stigma that attaches to conviction for those offences. These offences include murder[139], attempted murder[140] and crimes against humanity and war crimes[141]. For these offences, subjectivemens rea will be required as a principle of fundamental justice. Importantly, “that minimum degree of mens rea is constitutionally required to convict a party to that offence as well”[142]. The words “ought to have known” are thus to be deleted where section 21 (2) is applied to the international crimes contemplated by the Act.

Surprisingly very little has been said about what is meant by “subjective” foresight as the constitutionally-required mens rea for certain offences. Generally, “subjective” mens rea comprises both intention and recklessness[143]. With respect to murder, the constitutional minimum was described as intent to cause death or intent to cause bodily harm that the accused knows will likely cause death[144]. As mentioned above, while there is an element of “likelihood” which tracks closely the usual description of recklessness, it was held that the intent to cause bodily harm that is known will likely cause death was only a “slight relaxation” and quasi-identical to the intent to kill[145]. A particularly high level of recklessness is thus required, so that awareness of a risk or of a possibility of death is excluded in the case of murder[146]. Attempted murder requires intent to kill and therefore satisfies the constitutional requirement of subjective foresight[147]. As concerns crimes against humanity, the subjective criterion adopted in Finta was rearticulated in Mugesera in accordance with customary international law (though the constitutional basis was not explicitly reaffirmed). Most interestingly, it explicitly includes recklessness :

[I]n addition to the mens rea for the underlying act, the accused must have knowledge of the attack and must know that his or her acts comprise part of it or take the risk that his or her acts will comprise part of it […] The person need not intend that the act be directed against the targeted population, and motive is irrelevant once knowledge of the attack has been established together with knowledge that the act forms a part of the attack or with recklessness in this regard […] In Finta, the majority of this Court found that subjective knowledge on the part of the accused of the circumstances rendering his or her actions a crime against humanity was required. This remains true in the sense that the accused must have knowledge of the attack and must know that his or her acts are part of the attack, or at least take the risk that they are part of the attack[148].

Therefore, it seems that the minimal constitutional requirement of subjective foresight may include recklessness, with different levels of awareness of the risk depending on the offence at play. The constitutionally required subjective mental element of crimes against humanity includes recklessness, according to the Supreme Court of Canada[149]. The rationale would likely extend to war crimes, which also require that the perpetrator “was aware of factual circumstances that established the existence of an armed conflict[150]”. However, the crime of genocide, which is characterised by the specific intent “to destroy, in whole or in part, a national, ethnical, racial or religious group, as such[151]”, would likely preclude recklessness, though the question as to the constitutional minimum remains open. Obviously, murder can constitute an underlying offence of all core international crimes. In such cases, a “lower form” of recklessness (awareness of a risk) would be excluded for the underlying offence but permitted for the required knowledge of the contextual elements (attack or armed conflict) that qualify the crime as a crime against humanity or a war crime.

What is the situation of an accomplice under section 21 (2) with respect to the different levels of subjective mens rea required of the principal as a constitutional minimum ? In Logan, Lamer C.J. decided that “[w]hen the principles of fundamental justice require subjective foresight in order to convict a principal of attempted murder, that same minimum degree of mens rea is constitutionally required to convict a party to the offence of attempted murder[152].” Consequently, the Supreme Court of Canada declared unconstitutional, in such cases, the objective component of section 21 (2)[153]. However, it did not reject any other wording of that provision and maintained the subjective foreseeability to the level of probability[154]. Therefore, recklessness — with a high degree of consciousness — would appear as sufficient mens rea for liability to arise under section 21 (2), including for specific-intent crimes such as attempted murder[155] and, arguably, genocide. This conclusion converges with certain decisions of the ICTY which have held that the additional offences must have been foreseen as “likely” or “most likely”, though most ICTY cases state a lower standard of awareness by referring to a “possibility” of occurrence[156].

As noted above, extended liability of the participant in a joint criminal enterprise on the basis of recklessness for genocide and other specific-intent crimes has been criticised. In Canada, the softening of the mens rea requirement for the participant in a joint criminal enterprise may be justified insofar as the law requires that he or she be subjectively aware of the specific intention of the principal[157]. The degree of likelihood of the additional offence, explicitly set at section 21 (2) to a probability and not merely a possibility, also imposes a high threshold for the reckless conduct to attract criminal sanction. Furthermore, it reflects the particular danger of criminal groups for the public and the additional moral guilt attached to engaging in a joint criminal enterprise that the accused knows will probably escalate[158]. Finally, under section 21 (2), the accomplices will always be required to have a specific intent to “carry out an unlawful purpose and to assist each other therein”. This specific intent, coupled with the subjective foresight of the probability that additional offences be committed, is in the end a moderately softened mental element.

However, under that interpretation, the accomplice is still held to a lower standard than the principal for the additional specific-intent offences. The Supreme Court of Canada has not accepted that “the principles of fundamental justice prohibit the conviction of a party to an offence on the basis of a lesser degree of mens rea than that required to convict the principal”. However, it left open the question whether this proposition could be supported “in a situation where the sentence for a particular offence is fixed[159]”. The War Crimes Actmandates a sentence of life imprisonment where intentional killing is the basis of the international offence[160]. This could serve as a basis for an argument that an accomplice, who faces the same penalty as the principal, should be required to have the same level of (subjective) mens rea, which, at least for genocide and other specific intent crimes, excludes any form of recklessness[161]. Admitedly, the Supreme Court of Canada pronounced itself in the context of a discussion on the objective/subjective dichotomy. Still, its reasoning could be extended to apply to an asymmetry in the different degrees of subjective mens rea applicable to the accomplice and the principal[162]. This interpretation is supported by the Rome Statute — an interpretative guide for Canadian judges for prosecutions under the Act — which excludes liability on the basis of recklessness as a general rule[163].

In closing this discussion, it should be noted that nearly all cases in Canada concerning the application of international criminal law have emanated from the administrative level, that is from various quasi-judicial bodies or from courts (including the Federal Court of Appeal and the Supreme Court on one occasion) sitting in judicial review of such proceedings. The importance of these decisions can be considerable, as exemplified by the Mugesera decision which has influenced profoundly the domestic approach to international crimes. “Complicity” in international crimes has been the object of an impressive jurisprudence emanating from these fields. It is outside the scope of this article to offer a complete review of this rich and vast jurisprudence[164]. However, two points will be stressed.

First, an analysis of this case law shows that it bears many similarities with the international joint criminal enterprise doctrine, though the two “have developed virtually in isolation of each other[165]”. Both apply to large-scale criminal enterprises ; both acknowledge that membership in a criminal organisation is not enough (though Canadian bodies have developed an exception for those organisations limited to a brutal purpose) ; and both recognise that the significance of the accused’s contribution need not be linked to the actual perpetration of the material elements of an offence and can be determined among other things considering “[a]n accused’s leadership status and approving silence […] the size of the enterprise, the functions performed by the accused and his efficiency in performing them, and any efforts made by the accused to impede the efficient functioning of the joint criminal enterprise[166]”. Second, and consequently, it is apposite that an analysis be undertaken as to whether and how Canadian criminal law may be inspired by this case law. With respect to complicity, the applicable principles have developed taking into account the collective nature of international crimes and were influenced by the “international” character of the crimes[167].


In summary, the state of Canadian law on principal liability and complicity (excluding counselling), when analysed in conjunction with international criminal law, is such that :

  1. Commission and co-perpetration (principal liability) will not apply to those who participate in the commission of international crimes without however executing one of the essential elements of the actus reus, thereby excluding an important proportion of participants therein and most leaders or masterminds.

  2. Aiding and abetting is therefore called to play a crucial role in the endeavour of imputing liability at the domestic level for the core international crimes.

  3. Aiding and abetting is in broad terms similar in Canadian and international criminal law, but there is no requirement in Canadian law that the assistance have a substantial effect on the commission of the crime.

  4. Aiding and abetting has a stricter mens rea requirement in Canadian law than the knowledge-based mens rea required by the ad hoc tribunals (though the Rome Statute may be interpreted as also requiring specific intention).

  5. Despite a slightly different standard for murder, Canadian law should continue to uphold the general requirement that the aider and abettor need not share the intent of the principal. His or her knowledge of the intention of the principal, including the latter’s specific intent, will suffice. This is in line with international customary law.

  6. Aiding and abetting, however, does not easily capture the collective nature of international crimes. It does not cover joint criminal enterprises of the first and second categories, which are themselves probably encompassed by the detailed modes of perpetration provided for under article 25 (3) (a) or under article 25 (3) (d) of the Rome Statute. This could exclude those participants whose role may be essential for the realisation of the common criminal design, but who may be more difficult to link to a particular crime or to a particular physical perpetrator.

  7. The extended liability of section 21 (2) — with the applicable Charter guarantees — does reflect the collective nature of international crimes. It may serve to cover situations akin to those classified under the third category of joint criminal enterprises by the ad hoc tribunals, though its scope as to the nature of the joint criminal enterprise is — perhaps thankfully — narrower. However, the recognition of the collective nature of the core crimes applies only to additional offences that may have been committed as a probable consequence of the criminal design.


Momčilo Krajišnik was a prominent member of the Serbian Democratic Party (SDS), and President of the Bosnian-Serb Assembly. He was found by the ICTY to have been at the center of a joint criminal enterprise, the common objective of which was “to ethnically recompose the territories under [Bosnian-Serb leadership] control by expelling and thereby drastically reducing the proportion of Bosnian Muslims and Bosnian Croats living there”[168]. Mr Krajišnik’s material contribution to the joint criminal enterprise was to “help establish and perpetuate the SDS party and state structures that were instrumental to the commission of the crimes. He also deployed his political skills both locally and internationally to facilitate the implementation of the [joint criminal enterprise’s] common objective through the crimes envisaged by that objective[169]”. For instance, he was found to have “[participated] in the establishment, support or maintenance of SDS and Bosnian-Serb government bodies at the Republic, regional, municipal, and local levels […] through which [he] could implement the objective of the joint criminal enterprise[170]”. He was also found to have supported, encouraged, facilitated or participated in the dissemination of information to Bosnian Serbs “that they were in jeopardy of oppression at the hands of Bosnian Muslims and Bosnian Croats, that territories on which Bosnian Muslims and Bosnian Croats resided were Bosnian-Serb land, or that was otherwise intended to engender in Bosnian Serbs’ fear and hatred of Bosnian Muslims and Bosnian Croats or to otherwise win support for and participation in achieving the objective of the joint criminal enterprise[171]”. With the requisite intent, these contributions and comparable others were sufficient for liability to arise for the crimes against humanity of deportation and forced transfer, as well as persecution by way of deportation and forced transfer, which were the original crimes forming the common objective. Central to this determination was the fact that “the participation of an accused person in a JCE need not involve the commission of a crime, but that it may take the form of assistance in, or contribution to, the execution of the common objective or purpose[172]”.

Could Momčilo Krajišnik have been successfully prosecuted in Canada ? What about Omar El-Bashir (regardless of applicable immunities) ? Could someone who is accused of having directed the branches of the state apparatus and led them, in a coordinated manner, to implement a joint common plan consisting of ethnic cleansing or unlawful attacks against civilians be successfully prosecuted before Canadian courts ?

In the ad hoc tribunals’ jurisprudence and arguably in the detailed scheme of article 25 (3) of the Rome Statute, the distinction between principal and secondary liability mainly serves to reflect degrees of criminal liability[173]. In Canada, modes of principal and secondary liability mostly serve classificatory purposes. Degrees of criminal responsibility are reflected at the sentencing level. Hence, a mastermind of massive crimes who can only be convicted using secondary liability principles can theoretically receive a sentence that adequately reflects the extent of his or her guilt[174].

However, a gap in the net cast by the Canadian legal regime may be anticipated in cases where the contribution of the accused — mostly leaders and masterminds, but also mid-range perpetrators who will often find themselves in a grey zone — will be more difficult to link to the material elements of a particular crime. Their role may be essential for the realisation of the common criminal design, but may, for instance, be harder to link to a particular crime of murder or rape. Indeed, this is the fundamental distinction between joint criminal enterprise and aiding and abetting at international law[175].

Reliance on the law governing aiding and abetting in Canada to ensure liability for participation in collective crimes yields another unsatisfactory consequence. Apart from the extended liability arising from section 21 (2) of the Criminal Code, the collective nature of the crime, the existence of a plan or the involvement of the accused in an illicit association may not be relevant at all when establishing guilt[176]. Though it may at times facilitate the prosecutor’s task (low-level aider or small-scale crimes with few participants), it may at other times prevent effective prosecution (mid-level participants, masterminds or large-scale criminal enterprises). In some cases, it may lead to inadequate results. Indeed, criminal law is usually more severe regarding those who associate themselves with illicit organisations or who take part in group crimes. This is sometimes reflected in the gravity of the sentence imposed[177], in the very extended liability of section 21 (2) for additional offences arising out of a joint criminal enterprise, and in specific offences which criminalise certain forms of association with criminal organisations regardless of the commission of a specific offence[178]. The limited recognition of the reality of group criminality in the law applicable to international offences is unfortunate, both because it effectively prevents fair labelling of the impugned offences and participation therein, and because it can impact on the perceived gravity of certain conducts and on the attached sentences. This is in stark contrast with the principles developed by a Pre-Trial Chamber of the ICC in the first decision regarding co-perpetration, where the existence of a plan or agreement is essential, as well as with the concept of perpetration through a guilty agent, which will often be “organisational” in nature. It is also at odds with the doctrine of joint criminal enterprise developed by the ad hoc tribunals. In fact, it is the requirement of a plurality of persons sharing a common criminal purpose which partly justifies the co-existence of the doctrine of joint criminal enterprise with other modes of criminal liability[179].

Furthermore, while in most cases the use of secondary liability principles will not prevent the imposition of individual criminal liability, in certain cases it just might do so. Aiders and abetters under Canadian law can invoke defences that are not available, or available to a lesser extent, to the principal. For instance, an accomplice could theoretically invoke a defence of voluntary intoxication not similarly available, if at all, to the principal[180]. Therefore, the “indirect perpetrator” accused of international crimes in Canada will at times be subject to a less stringent legal regime than if he were considered a principal, or if his or her participation in a criminal enterprise were recognised. It should be noted that while this may create a more favourable regime for an accused under Canadian law, it is another issue whether this leads to “incapacity” in terms of the ICC complementarity scheme. Indeed, the same defence may also be available before the ICC, regardless of whether the accused is considered a principal or an accomplice[181].

Within the aiding and abetting regime, apart from extended liability for additional offences arising out of a common purpose, liability as an abettor for being a party to a conspiracy is an interesting route to explore. It could indeed be argued that participation in a conspiracy constitutes encouragement or counselling to commit the crime. In such a scenario, a conspiracy would have to be proved, in a similar fashion to common criminal purpose under section 21 (2), and liability could be incurred for both conspiracy and the substantive crime eventually committed (or only for the latter if the merger is seen as unjustified)[182]. However, for liability to attach to a participant in the main crime of genocide, crimes against humanity or war crimes, it will always be necessary to meet the conditions of abetting that particular crime (in the scenario under discussion, this would involve among other things that the physical perpetrator be a party to the conspiracy). In cases such as those of Bashir or Krajišnik and a majority of leaders or masterminds party to vast criminal enterprises, such a link between the high-level perpetrator, physical perpetrators and identifiable crimes on the ground may prove difficult to establish.

In addition to principal liability, aiding, abetting and common purpose, which were the central targets of this study, other modes of liability as well as inchoate offences will likely be of use in the endeavour of imputing liability at the national level for international crimes. Among those, liability for conspiracy as a separate inchoate offence offers an alternative for prosecutions aiming at reflecting the collective nature of international offences. It is an alternative with limited potential, however. First, it is unsatisfactory, as liability will then not be incurred for a core crime (genocide, crimes against humanity, war crimes). Furthermore, difficulties may arise in prosecutions for conspiracy under the War Crimes Act. Indeed, the Act expressly allows prosecution of conspiracy for all core crimes, whereas conspiracy is only criminalised for genocide in international law, and then again, it is not criminalised as such under the Rome Statute[183]. The prevailing view currently is that “conspiracy does not exist as a form of liability for war crimes or crimes against humanity” in international customary law[184]. This discrepancy between Canadian law and international law needs to be assessed in light of the principle of legality, enshrined in section 11 (g) of the Charter[185]. It may also serve as a basis for an argument alleging the (inexistent) basis at international law for the exercise of universal jurisdiction over these offences. A constitutional argument is indeed conceivable for prosecutions for conspiracy to commit crimes against humanity or war crimes, at least those committed prior to the entry into force of the War Crimes Act.

Moreover, counselling, both as a form of participation in offences and as an inchoate offence, will certainly prove crucial in domestic prosecutions for international crimes. There, too, a comparative analysis shows that Canadian law presents characteristics that may prove difficult to reconcile with international criminal law. Nevertheless, section 22 (2) of the Criminal Code, which allows for the extended liability of the counsellor for “every offence that the other commits in consequence of the counselling” will be of particular interest, despite the difficulties inherent in proving a causation link and in linking, for instance, a “high-level” counsellor with a “low-level” perpetrator.

Courts called upon to apply the Canadian principles of liability in prosecutions under the War Crimes Act should seek guidance in the international jurisprudence and, where relevant, the jurisprudence of the Federal Court in matters of immigration and refugee law. In these decisions, the collective nature of international crimes has influenced the law in a manner which may find an echo in the Canadian criminal law context. For instance, there is no principled reason why the law on aiding and abetting should not be developed so as to encompass contributions to — even large — joint criminal enterprises directed at the commission of mass atrocities. The accused’s contribution to the core crimes and the requisite intent should not be exclusively dependent upon the material elements of a particular offence, but should be assessed also considering an accused’s leadership status and role within an organisation or illicit association as well as his or her efforts in achieving, or impeding, the objective of the criminal purpose. Such an interpretation may be essential if Canadian law is to succeed in capturing not only low-level perpetrators — which it does in many instances — but also mid-range participants, leaders and masterminds of international crimes, particularly in the context of large criminal enterprises with a plurality of participants at different levels. Arguably, since the material element of aiding and abetting in Canadian law does not require that the act or omission of the accused have any effect whatsoever on the commission of the crime, and because it is not required that the principal even be known, it could be interpreted so as to allow more remote contributions to be considered sufficient, provided the requisite intent is present. The War Crimes Act forces Canadian criminal law to evolve and to move away from the “mononuclear” paradigm of one author, one fact, one victim.

The inexistence of other modes of participation particularly relevant to international crimes, such as planning and ordering, further militates for such an interpretation of the law. Liability as a superior for failing to prevent or repress the commission of crimes committed by subordinates complements the regime on secondary participation in crimes, but will apply to different factual scenarios. Furthermore, superior liability is a distinct offence under the Act[186] and, in cases of participation in joint criminal enterprises, will often not adequately reflect individual criminal liability for the core crimes.

The present limited analysis shows that Canadian law can and must adapt to the reality of international crimes. The purpose of the system put in place by ICC — at the heart of which is the principle of complementarity — is to close the “impunity gap” for genocide, crimes against humanity and war crimes. In order for this system to function effectively, States must be able to exercise their jurisdiction — including universal jurisdiction — against the full range of possible perpetrators, including those at the top of military nd political structures (rules on immunity permitting). They must also be able to exercise their jurisdiction whatever the complexity of the structure of the collective criminal venture giving rise to the worst international crimes. Whether this is done through principal or secondary liability principles should remain the object of an interesting theoretical contest between different legal systems. However, what this legislative preference cannot do is create reason for “incapacity”, or a justification for “unwillingness”.