MJIL Symposium: A Response to Darryl Robinson by James Stewart

[James Stewart is an Assistant Professor at the University of British Columbia, Faculty of Law. He is currently undertaking a Global Hauser Fellowship at New York University School of Law.]

This post is part of the MJIL 13(1) symposium. Other posts in this series can be found in the related posts below.

It is a pleasure to be invited to comment on Professor Darryl Robinson’s excellent new article How Command Responsibility Got So Complicated. His meticulous research has, once again, advanced our understanding considerably. Indeed, this particular article is but the most recent manifestation of Professor Robinson’s groundbreaking commitment to marrying criminal theory and international criminal doctrine in ways that shed new light on dilemmas that have plagued scholars and practitioners for too long. In this piece, he focuses on the much-disputed physical contribution of the failure to punish limb of superior responsibility. Some say that a superior can be convicted of genocide, for example, for failing to punish acts of her subordinates who perpetrated the crime, but Professor Robinson joins others who protest that this violates the principle of culpability. How can you be held responsible for a crime to which you did not contribute? Conversely, those who argue that failures to punish can be re-imagined as a separate conduct-type crime stripped of consequences to overcome the participation problem ignore that international law does not support that reading. Instead, Professor Robinson concludes that subsuming superior responsibility within everyday notions of accessorial liability offers a more elegant solution.

I feel compelled to start my review of the piece with a confession of sorts. In my former incarnation as an Appeals Counsel at the International Criminal Tribunal for the former Yugoslavia (‘ICTY’), I had a hand in formulating the Prosecution’s position on superior responsibility in a range of the cases that are key to Professor Robinson’s argument (Hadžihasanović, Orić, and Halilović). In part, I admit this in order to disclose a potential impurity in my views on the topic (although, for balance, my own views were different from the position ultimately adopted by the Prosecution in these cases, contrary to the conclusion ultimately reached by the Appeals Chamber that ruled on them, and I may have changed them again since reading Professor Robinson’s provocative article). For present purposes, though, this experience is also germane since it leads me to think that Professor Robinson might be too quick in arguing that the ICTY has not wrestled with these issues; to the contrary, all sides were engaged in a frenzied review of much of the literature Professor Robinson cites in an attempt to deal with precisely these problems, although no one came close to addressing the topic with anything approaching the sophistication Professor Robinson now offers. Sometimes an absence of judicial reasoning just conceals issues too complex to articulate.

Turning to the substance of his argument, I also have to admit that Professor Robinson’s excellent work may have prompted a change of heart in me, but one that runs counter to his reasoning. In my earlier work, I too have argued that failures to punish cannot be squared with causation in the original offense, for reasons very close to those Professor Robinson eloquently sets out in this article (see pp 19–29). I now doubt us both. In particular, I now wonder whether failures to punish cannot be assimilated to ex post facto aiding and abetting, in ways that solve these causal problems. To illustrate, consider a case called Blagojević, where a key issue was whether the accused was responsible for aiding and abetting genocide by reburying the bodies of exterminated Bosnian Muslims, some time after the massacre at Srebrenica. In dealing with the issue, the Tribunal rightly reasoned that ex post facto aiding and abetting is possible if there is some pre-established agreement prior to the massacre, such that an accused assisted the realisation of the criminal plan at the time it materialised. So for Blagojević, he could be guilty of genocide if and only if his agreement to dispose of the bodies made some causal difference to the decision of the perpetrators to carry out the executions. Although preliminary, I wonder whether a variant of this reasoning may also justify treating failures to punish the original crime in causal terms, contrary to the portion Professor Robinson, myself, and most everyone else have argued for.

How would this work? Well, a superior who fails to punish might make a causal contribution to an international crime that has already materialised, if her willingness to turn a blind eye to her subordinates’ crimes is made known prior to the massacres and this makes some difference to the subordinates’ decision to carry out the bloodletting. In the comparable field of corporate criminal liability, scholars often posit that corporations aid and abet the crimes perpetrated by their employees by creating a culture that enables the crime (see here, p 38). By analogy, it seems at least plausible that the superior who makes it known that she will not hold underlings to account for international crimes, creating or at least increasing a sense of complacency about the absence of formal repercussions for this type of violence, herself makes a difference to the crimes occurring. To modify Blagojević, the commander creates a culture of impunity characterised by a tacit agreement that she will not hold subordinates responsible for atrocities, and this influences the decision to carry out the international crimes in question. Of course, this demotes the failure to punish to a mere point of evidence establishing a manifest disinterest in punishing crimes prior to their commission, but this is a minor normative shift relative to the radical surgery proposed by all other options.

All the same, on the assumption that failures to punish cannot be reconciled with causation, there are still one or two less significant areas where Professor Robinson and I might part ways. First, if failures to punish do not have causal power for the original offence, I agree entirely that fashioning failures to punish in acausal terms violates the principle of culpability, but I doubt whether this is an immutable feature of blame attribution in all its possible manifestations. My own view is that a contribution is made necessary by the way international crimes are defined. There are, for instance, coherent accounts of criminal justice that dispense with causation, participation, or a contribution altogether in favor of rules that make attempt the paradigm for all criminal responsibility (see here, pp 10–11). On these accounts, murder is established by pulling a trigger. But international criminal law is not one of these systems — causing particular consequences is frequently made essential by the identity of international crimes themselves. In other words, to the extent that failures to punish operate as an acausal ‘mode of liability’, it misrepresents responsibility in international criminal law (‘ICL’) because it holds a superior out as having satisfied elements of a crime she has not. The key point is that a contribution is necessitated by the nature of international crimes, not some feature of superior responsibility alone.

This insight has great repercussions for ratification theories of responsibility Darryl countenances, which apologise for treating failures to punish as modes of liability even without causation. According to these theories, ‘the commander, by failing to prosecute the crime, “ratifies”, endorses or acquiesces in the crime, and thus voluntarily absorbs liability for it ex post facto’ (p. 46). Having reviewed arguments by Christopher Kutz and Amy Sepinwall, undoubtedly two exceptional scholars, Professor Robinson concludes that ‘[f]or the time being it seems that there are still difficulties to be overcome before ratification theory can offer an explanation and justification for command responsibility as a mode of liability without causal contribution’ (p. 51). But this equivocation appears inconsistent with the remainder of his compelling argument, especially when he moves immediately back into a tone of disdain at the ‘extraordinary position’ that presently characterises acausal accounts of failure to punish (p. 51). Simply stated, I believe Professor Robinson cedes too much ground to excellent scholars who argue for ratification theories, with whom he cannot coexist so peaceably.

Try as I might, I cannot follow Professor Robinson down this path. To be sure, there are few scholars whose work I revere more than Kutz and Sepinwall, both of whom bring exceptional degrees of sophistication to a whole raft of issues of great salience to modern international criminal justice. And yet, to my mind, both offer models of individual responsibility that are not available in international criminal justice as presently constituted, precisely because they are not minded to tailor their theories of responsibility to the specific identity of international crimes as they currently exist. As I mention earlier, international crimes are constructed in ways that make causal inquiries necessary — a causal contribution is manifestly indispensible for murder as a crime against humanity. In other words, causation is an ingredient of the disapprobation we seek to communicate in convicting an individual of many international crimes, so describing someone as a murderer when they have not satisfied this element miscommunicates what it means to be responsible for this crime. As a consequence, ratification theories violate the principle of culpability Professor Robinson rightly champions, and for these reasons, I suspect that he had to criticise these explanations more forcefully to maintain analytical consistency.

This brings us to the idea of a separate crime of dereliction of duty, where I have a related quibble. While I accept that almost everyone considers a separate conduct-type crime of dereliction of duty that disregards consequences of the superior’s failure as a defensible alternative to failures to punish as a ‘mode of liability’, I fear that this reframing seriously understates a superior’s responsibility in the vast majority of conceivable cases. If failures to punish promote future crimes, as is generally the case in campaigns of violence, the conduct-type crime of dereliction of duty is manifestly inadequate as an expression of a superior’s true guilt. In these situations, failures to punish unequivocally cause subsequent atrocities. The same is true if I am correct that failures to punish can also cause the original murder by constituting ex post facto aiding and abetting. So Professor Robinson’s claim that ‘the “separate offence” approach would indeed address the culpability problem’ (p. 32) is only true in the negative sense that the separate crime would not constitute an illiberal excess; true enough, but it would also grossly understate the superior’s responsibility by effectively excising the ensuing massacres the superior contributed to from the moral assessment of her guilt. If causation carries moral significance, as international criminal justice formally assumes, this conceptual myopia is an aberration. So I doubt whether reimagining failures to punish as a separate conduct-type offense is much of a solution.

Leaving all this to one side, I sense that the quintessential issue with superior responsibility is causation through omissions. While the article draws on an extensive literature treating this topic, I am not entirely convinced that scholarly positions on the ability of omissions to generate causal power are as uniform as Professor Robinson intimates; it strikes me that causation through omission is one of the perennial points of disagreement that constitutes the field of criminal theory. But regardless of the relative popularity of any particular account, I believe that these theories need to be aired, discussed and criticised more openly in the context of international crimes. My own modest struggles with causation in this field did not deal with omissions, but they did reveal major conceptual problems with positions espoused in leading domestic criminal systems. Despite this, international courts called to articulate an account of causation by omission have merely cited scholars who appeal to other scholarly authorities, thereby inviting us to take the position on trust. Based on previous experience, I am convinced this methodology is unsafe.

In short, I hope we can convince Professor Robinson to turn his considerable talents to the theory of causation by omission soon. To gently provoke this, let me offer some superficial misgivings thus far. To some extent, Professor Robinson reconciles himself with the International Criminal Court’s (‘ICC’) argument that an omission need merely ‘increase the risk’ of a resulting crime. While I appreciate the rationale for this approach, the downsides also seem marked — does this embrace of probabilistic causation not undermine the entire basis for having causation in the first place? My driving a car negligently is certainly wrong because it creates important risks, but my driving a car negligently in ways that cause death increases my culpability by an order of magnitude. Indeed, this is the prime reason we have causation in the criminal law at all. Why should we struggle with the enormous complexities it engenders if risking was sufficient all along? The overarching moral concern is that treating ‘risk’ as sufficient for causation robs international criminal justice of the true moral significance that arises when the risk of atrocity actually materialises. So if the ICC’s position is defensible, I hope we will soon hold it up to the light for scrutiny.

Let me end, then, with a final and perhaps predictable set of thoughts. Professor Robinson describes his attempt to sublimate superior responsibility within complicity as ‘an elegant solution,’ and I agree wholeheartedly. But why stop there when an even more elegant alternative awaits by pursuing this analytical trajectory to its logical end? The truth is, international criminal justice is replete with jurisprudence indicating that superior responsibility is more culpable than simple aiding and abetting or perpetration, and a host of cases treat superior responsibility as akin to perpetration, not complicity. To my mind, this is further confirmation that the distinction between committing and participating, to which advocates of the differentiated model of blame attribution assign so much normative weight (without ever defending), is really quite vapid. If this is true, should we not take Professor Robinson’s game of Russian dolls one step further, and encase complicity within perpetration too? Doing so leads us to the unitary theory of perpetration I have advocated for here, and to the abolition of superior responsibility outright. In the end, I view this as a preferable normative account of international blame attribution, including for crimes perpetrated by superiors who fail to discharge legal duties.

Nonetheless, I congratulate Professor Robinson on an extremely thorough, original, provocative, and important piece of research. I look forward to further exchanges over the years to come.