Guest Post: The ICTY Loses its Way on Complicity – Part 1
[James G. Stewart is an Assistant Professor at the University of British Columbia. He is also presently a Global Hauser Fellow at New York University School of Law.]
The International Criminal Tribunal for the former Yugoslavia (ICTY) is undoubtedly one of the most important institutions in the history of international law, not only for its catalytic effect in generating trials for international crimes before both international and domestic courts but also for breathing new life into both international humanitarian and criminal law. Yet, the ICTY Appeals Chamber recently rendered a judgment on the law of complicity in Prosecutor v Momčilo Perišić (see here), that could undo much of its legacy. In this first of two posts, I will set out the background to this case and consider the problem of “specific direction” as an element of the actus reus, which the Appeals Chamber has newly adopted. In a second post, I will focus on the mental element of complicity, showing how a more traditional approach to mens rea can address the underlying concerns without so seriously disrupting the law of complicity.
Two weeks ago, I attended a roundtable dedicated to the law of complicity at the University of San Diego. Over the course of two days, a dozen of the best criminal theorists in the English-speaking world came together to debate four competing accounts of complicity. On the flight home, however, I was more than slightly surprised to learn that the ICTY had just announced a new understanding of the doctrine that is without equivalent in any national law, very different from the Tribunal’s earlier jurisprudence and at odds with the views of all experts congregated at the roundtable I had just attended. Indeed, the new understanding of complicity that the ICTY adopts in Perišić appears inconsistent with foundational principles of criminal law in ways that seriously compromise the doctrine. Below, I explain why this new position is so troublesome, before I go on to suggest a safer path the Appeals Chamber could have followed.
Momčilo Perišić was the Chief of the General Staff of the Yugoslav Army (VJ), making him the highest ranking officer in that army. Between August 1993 and November 1995, he provided extensive military and logistical aid to the Army of Republika Srpska (VRS), lead by the infamous Radovan Karadžić and Ratko Mladić. At trial, Perišić was convicted of aiding and abetting international crimes perpetrated by the VRS, most notably for crimes associated with the sniping campaign used to terrorize civilians within Sarajevo and for the terrible bloodletting at Srebrenica. Perišić unquestionably provided the VRS with large quantities of weapons, seconded officers involved in these crimes to the VRS (Mladić included), and supported the VRS in a host of other ways. Was all this support innocuous assistance of a general type or criminal complicity in the international crimes undertaken by the VRS?
The Appeals Chamber favored the former option, overturning the Trial Chamber’s conviction for aiding and abetting various international crimes, including persecution. In order to exclude general assistance from the scope of complicity, the Appeals Chamber announced that “specific direction” is now an element of the actus reus of aiding and abetting, implying that Perišić’s assistance for the VRS’s war effort must not just cause international crimes, it must be “specifically directed” towards this end. So even though the Trial Chamber had found that “the VRS [waged] a war that encompassed systematic criminal actions against Bosnian Muslim civilians as a military strategy and objective,” (see here, para. 1621) the Appeals Chamber found that Perišić’s considerable assistance was not “specifically directed” towards the concrete crimes with which he was charged. Whatever one might think about this verdict, the law makes no sense.
To begin, “specific direction” has no real grounding in customary international law. In identifying the “specific direction” standard, the Chamber placed great weight on casual language in the Tadić Appeals Judgment, rendered many years ago. In that judgment, a differently constituted Appeals Chamber had opined that (Tadić Appeal Judgment, para. 229, my emphasis):
“[t]he 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.”
While a number of international decisions have rote cited this passage since, none had ever given meaning to “specifically directed” in a methodologically serious manner. The Appeals Chamber’s recent judgment gives the impression that the wording enjoyed a long pedigree in practice, but none of this history stems from sources prior to Tadić and few of the international tribunals that make passing mention of the concept ever treat the phrase as an element to be proved.
This shaky foundation is not altogether surprising—no previous understanding of complicity ever included a requirement of this type. In my earlier research on complicity (see here), I showed how international courts and tribunals frequently borrow their understandings from domestic criminal systems. Complicity is no exception to this rule; the international understanding of complicity was heavily influenced by German, English and French equivalents. However, in each of these contexts, national legal systems do not use anything remotely similar to “specifically directed” to limit the actus reus of complicity. Instead, causation and an accompanying auxiliary concept (called legal causation, remoteness or objektive Zurechnung in different jurisdictions), usually performs this work (see here, p. 19). Thus, the ICTY’s new requirement for complicity stands alone in a sea of inconsistent state practice as one of a kind. It cannot, therefore, reflect customary international law.
How is this standard problematic? First, the “specifically directed” test draws a series of untenable distinctions at the level of actus reus. For instance, the Appeals Chamber declares, “the provision of general assistance which could be used for both lawful and unlawful activities will not be sufficient alone, to prove that this aid was specifically directed to crimes of principal perpetrators.” (para. 44). But in words of Cambridge Professor Glanville Williams, accomplices are almost always “otherwise innocent,” and acts that help crime are frequently commonplace. There is, for example, nothing inherently illegal in driving a car away from a bank, but this conduct becomes a paradigmatic example of complicity when it assists a bank robbery. Similarly, driving a car is commonplace, which in no way deprives the action of an ability to constitute complicity when it makes a causal contribution to the realization of a pre-established crime. In either case, causation is the relevant inquiry (see here), not “specific direction.”
“Specific direction” is all the more anomalous when one of the leading criterion for limiting the over-inclusive consequences of causation—the substantial contribution standard—is already applied in international doctrine. The standard refrain before the ICTY is that the aider and abettor must have a “substantial effect on the commission of crimes.” Admittedly, not everyone would use this same language. Scholars debate a range of doctrines for excluding causal contributions that are too remote, strange or unfair. These include the frequently misunderstood terms “remoteness,” “harm-within-the-risk” and, most significantly, “substantial contribution” (see here, p. 19). In practice, the very purpose of doctrine like objektive Zurechnung (normative attribution) in Germany and proximate cause in the United States is largely captured by this reference to substantial contribution, so it is odd that the Appeals Chamber created a “specific direction” standard instead of dedicating more energy to assessing Perišić’s alleged complicity within established categories.
The standard the Appeals Chamber adopts is also exceptionally hard to define. For instance, in elaborating a definition of “specifically directed”, the Appeals Chamber intimates that two inquiries are relevant. First, Perišić’s copious assistance could be “specifically directed” if “the VRS was an organisation whose sole and exclusive purpose was the commission of crimes” (Appeal Judgment, para. 52). I doubt, however, that it is possible for any organization to only act criminally: if members of the Gestapo or the Interahamwe drove on the correct side of the road, purchased food from local merchants, or fell in love and got married over the course of their respective genocides, I presume that one is not complicit in their many international crimes by helping them carry out these peripheral activities. Once again, causation is the concept that usually distinguishes these innocuous cases from those where an accomplice’s actions actually assist atrocities. By implication, this first prong of the “specifically directed” test runs counter to the preordained role of causation in the criminal law.
The second test is equally problematic. In addition to the criminal organizations limb, the Appeals Chamber declared that Perišić’s actions might be “specifically directed” at international crimes if he “endorsed a policy of assisting VRS crimes.” (Appeal Judgment, para. 52, my emphasis). However, this reasoning seems to inappropriately conflate the physical and psychological elements of accomplice liability. This tendency is exacerbated by findings that: “specific direction may involve considerations that are closely related to questions of mens rea. Indeed, as discussed below, evidence regarding an individual’s state of mind may serve as circumstantial evidence that assistance he or she facilitated was specifically directed towards charged crimes.” (Appeal Judgment, para. 48). Unfortunately, merging psychological dispositions like endorsement with the conceptually distinct question of whether an accused’s conduct adequately assisted the crime in question is incompatible with basic ideas about criminal law, which strive to keep the two apart for sound reasons.
The Appeals Chamber’s position on remoteness is no more convincing. At several points in its analysis, the Chamber appears to opine that the remoteness of Perišić’s contribution necessitates the “specifically directed” test, such that this new standard would not be necessary if the accomplice was present when the crime was committed (see Appeal Judgment, paras. 39, 70). Thus, “specific direction” is not an invariable element of aiding and abetting; it only arises where the accomplice is not at the scene of the atrocity. Yet, this reasoning seems draconian, when presence at the scene of the crime was stripped of any normative significance in national understandings of complicity many decades ago. In the common law, for instance, a principal in the second degree was someone who assisted a crime while present at the scene, whereas accessories were absent. Canada abandoned the distinction in 1897, and all Anglo-American jurisdictions have followed suit since, precisely because the distinction is procedurally complicated and morally irrelevant. In this light, this aspect of the position in Perišić is highly regressive.
For similar reasons, one must be skeptical about the Appeals Chamber’s view that “geographic distance” is an appropriate measure of remoteness (see Appeal Judgment, paras. 38-40). In a day and age where technology enables individuals to cause untold harm across the entire planet, an attempt to limit complicity by geography seems to underappreciate contemporary capabilities for occasioning harm. If, for example, an independent medical company willfully sells vast quantities of a biological contagion to a terrorist group in the Netherlands so that they can use it as a weapon, then this terrorist group launches this weapon at New Zealand on the diametrically opposite side of the globe, surely geographic distance plays no part in calculating the medical company’s responsibility for the international crime that transpires when the weapon arrives at its destination. Likewise, “temporal distance” (Appeal Judgment, para. 40) need not be relevant in assessing remoteness either—what difference does it make if the missile takes 15 minutes or 15 years to reach its target? I think none.
The final difficulty is that, in large part, this approach was unnecessary. In Part 2 of this criticism, I will show how the misgivings in Perišić are really issues for the mental element of complicity, which can be resolved there without unduly disrupting the integrity of the doctrine. For now, it suffices to observe how that Appeal Chamber’s new understanding fails to square with defensible principles, previous jurisprudence or national equivalents. If left unaddressed, this reasoning will leave a black mark on that institution’s important contribution to international criminal law, and erect an unjustifiable impediment to the accountability of those who assist atrocities. In a world where complicity’s importance transcends international criminal justice itself (extending to the UN Global Compact, the Ruggie Process, the Alien Tort Statute and beyond), these shortcomings raise real concerns.