1. Introduction
Over the last fifteen years, thousands of international lawyers have successfully crafted an elaborate and operational system of international criminal law (ICL). This project drew on precepts of criminal law as well as international human rights and humanitarian law, with the last two areas providing essential normative content as well as a familiar framework for international oversight and intervention. This article argues that in drawing on these sources, ICL also absorbed contradictory assumptions and methods of reasoning, which manifest in internal contradictions, such as inadvertent contraventions of ICL's own principles. As the urgency of articulating a recognized set of rules recedes, the time is ripe for a deeper and more systematic examination of the tapestry that we have stitched together, so that ICL may develop as a distinct and coherent discipline.
The official narrative, and widespread understanding, is that ICL adheres to fundamental principles of criminal law, and that it does so in an exemplary manner.Footnote 1 These fundamental principles distinguish a liberal system of criminal justice from an authoritarian system. An authoritarian system may deal with individuals in any manner in order to pursue its aims, whereas a liberal system embraces restraints on its pursuit of societal aims out of respect for the autonomy of the individuals who may be subject to the system. Thus while the purpose of the criminal law system as a whole may be to protect society, some further deontological, or desert-based, justification is still required for a just application of punishment to a particular individual.Footnote 2 Treating individuals as subjects rather than objects for an object lesson, or as ‘ends’ rather than solely as ‘means’, imposes principled restraints on the infliction of punishment.Footnote 3
This article will refer to three of these principles (hereafter ‘fundamental principles’), all of which are recognized by ICL. The first is the principle of personal culpability, namely that persons are held responsible only for their own conduct. ICL recognizes as ‘the foundation of criminal responsibility’ that ‘nobody may be held criminally responsible for acts or transactions in which he has not personally engaged or in some other way participated’.Footnote 4 The principle also requires sufficient knowledge and intent in relation to the conductFootnote 5 that we may find the person ‘personally reproachable’.Footnote 6 A second is the principle of legality (nullum crimen sine lege), which requires that definitions not be applied retroactively and that they be strictly construed (in dubio pro reo, rule of lenity), in order to provide fair notice to individual actors and to constrain arbitrary exercise of coercive power.Footnote 7 This principle is a ‘solid pillar’ without which ‘no criminalization process can be accomplished and recognized’.Footnote 8 A third is the principle of ‘fair labelling’, which requires that the label of the offence should fairly express and signal the wrongdoing of the accused, so that the stigma of conviction corresponds to the wrongfulness of the act.Footnote 9 This article does not argue that ICL ought to adopt certain cherished principles from any national system; instead, it examines the assumptions and reasoning techniques that lead ICL into internal contradictions with its own declared principles.
Despite ICL's claim of exemplary adherence to these fundamental principles, recent scholarship has questioned this adherence in specific areas, most notably the doctrine of ‘joint criminal enterprise’.Footnote 10 As will be seen in the examples that follow, serious issues about ICL's compliance with its fundamental principles may also be found in many other doctrines, including sweeping modes of liability, expanding definitions of crimes, and reticence towards defences. This is the puzzle that prompts the present inquiry: how is it that a liberal system of criminal justice – one that strives to serve as a model for liberal systems – has come to embrace such illiberal doctrines?
Faced with evidence of frequent departures, one might be tempted to conclude that ICL is indifferent to liberal principles and is simply more harsh and punitive than national criminal law.Footnote 11 However, such an explanation misses the curious complexity of the phenomenon. Mainstream ICL does not reject fundamental principles, but rather sees itself as fully compliant, which suggests that more subtle distortions are at work. Moreover, ICL is scrupulously generous in the guarantee of procedural rights, and succeeds in upholding them even for the most unsavoury accused.Footnote 12 Thus we see a system that strives to respect both procedural rights and principles of substantive fairness, and which succeeds in the former but falls short in the latter. A theory is needed that will account for this curious anomaly.
This article suggests that part of the problem lies in normative assumptions transplanted from human rights and humanitarian law. Insightful glimpses into some specific elements of this phenomenon have previously been offered by George Fletcher and Jens David OhlinFootnote 13 and by Allison Marston Danner and Jenny Martinez.Footnote 14 This article builds upon those insights to offer a more holistic account of observable tendencies in ICL discourse.
After decades of inactivity, ICL experienced explosive growth in the mid-1990s. This boom in institution building and norm articulation led to a sudden need for international criminal lawyers. As few such creatures existed, the vacuum was filled – at least at the outset – primarily by international lawyers with training or experience in the related fields of human rights and humanitarian law.Footnote 15 Indeed, ICL was perceived and heralded as a major advance in human rights and humanitarian law, offering a valuable remedy and means of enforcement by punishing violators. ICL professionals eagerly adopted the forms and principles of criminal law; however, these were all understood through the lens of the normative assumptions from their native domains of expertise.
These early influences have left a continuing heritage, shaping the areas of myopia in ICL. For example, this theory is compatible with the dichotomy in respect for procedural rights and substantive principles. ICL is scrupulous and generous with due process and procedural rights, since such rights are deeply familiar to and internalized by human rights lawyers. On the other hand, principles of culpability, fair warning, and fair labelling are unknown to human rights and humanitarian law. Human rights and humanitarian law focus more simply on broad and liberal construction to maximize protection for beneficiaries, and are not accustomed to the special moral restraints which arise when fixing guilt upon an individual actor.
As will be suggested below, the problem is not just unfamiliarity with these special moral restraints, but that the influence of assumptions of human rights and humanitarian law actively works at cross-purposes to fundamental criminal principles. This article will present three of the modes by which this distortion occurs. One mode is the influence of interpretive approaches from human rights and humanitarian law, such as victim-focused teleological reasoning, which not only undermines strict construction but also fosters sweeping interpretations that run afoul of culpability and fair labelling. The second mode is substantive and structural conflation – that is, the assumption that criminal norms must be coextensive with similar norms in human rights or humanitarian law, overlooking the different structure and consequences of these areas of law, and thus neglecting the special principles necessary for blame and punishment of individuals. A third mode will briefly be outlined, showing how ideological assumptions about progress and sovereignty can foster a preference for broader doctrines and a neglect of fundamental principles.
This article does not suggest that human rights and humanitarian law assumptions are the sole cause of departures from fundamental principles. Other influences may well be at play. For example, ICL deals with violations of exceptional magnitude and severity, and studies indicate that the more severe the crime, the greater the perceived pressure to convict and the greater the likelihood of perceiving an accused person as responsible for the crime.Footnote 16 Another possible influence is the incentive of judges and professionals in an emerging field to demonstrate the efficacy of their field and to increase their influence and prestige by expanding the scope and role of ICL.Footnote 17 Reputational incentives may also have a subtle impact; for example, the judge, practitioner, or scholar who espouses conviction-friendly interpretations can reliably expect to be applauded as progressive and compassionate by esteem-granting communities.Footnote 18
Moreover, even in national systems, there are occasional excesses and controversial doctrines that are arguably incompatible with fundamental principles. Current efforts to respond to terrorism and organized crime, for example, have led national systems to adopt some laws that strain liberal principles. Nonetheless, what remains striking about ICL is the prevalence and extravagance of the departures. Furthermore, and much more importantly for present purposes, the discourse is different. When national laws strain principles, there is a vibrant debate and a sense that something has been sacrificed. In ICL there is little such discourse or even awareness of incongruity – instead, participants are applying what they believe to be sound legal methods with appropriately liberal aims.
Thus, even if other factors may be in play, the impact of human rights and humanitarian assumptions remains of particular interest because it offers not only a ‘why’ but also a ‘how’. Reliance on these assumptions and methods of argumentation is not only a cause of departures from fundamental principles, but it also furnishes the analytical steps by which such departures are effected and provides the legal plausibility that allows the departures to pass unnoticed. The object of inquiry of this article is the discourse itself: the assumptions that are made and the arguments that are regarded as sound arguments with appropriate liberal aims, and how these lead to contradictions with the liberal values to which the system supposedly adheres. This article argues that our favoured reasoning methods may contain distortions and hence that we need to think about the way that we think.
The identity crisis theory helps to explain why an overwhelmingly liberal-minded profession endorses startlingly illiberal doctrines and developments. In a typical criminal law context, liberal sensitivities focus on constraining the use of the state's coercive power against individuals. In ICL, however, prosecution and conviction are often conceptualized as the fulfilment of the victims' human right to a remedy.Footnote 19 Such a conceptualization encourages reliance on human rights methodology and norms. This shift in conceptualization also shifts the preoccupation of participants in the system.Footnote 20 Many traditionally liberal actors (such as non-governmental organizations or academics), who in a national system would vigilantly protect defendants and potential defendants, are among the most strident pro-prosecution voices, arguing for broad definitions and modes of liability and for narrow defences, in order to secure convictions and thereby fulfil the victim's right to justice.Footnote 21 Whereas in a national system one may hear that it is preferable to let ten guilty persons go free rather than to convict one innocent person, the ICL literature seems to strike the balance rather differently, replete as it is with fears that defendants might ‘escape conviction’ or ‘escape accountability’ unless inculpating principles are broadened further and exculpatory principles narrowed.Footnote 22
Thus, in ICL, illiberal doctrines do not arrive in a classic authoritarian garb (e.g. that individual rights must be sacrificed to serve state or societal goals). Illiberal doctrines arrive in a liberal garb – that of human rights liberalism – and hence are readily accepted and absorbed into the system. Human rights liberalism and criminal law liberalism both arose to protect individuals from the state – in criminal law to protect the accused and potential accused from the criminal machinery, and in human rights to protect individual victims from various forms of state mistreatment. ICL practitioners apply the familiar and cherished assumptions and techniques of human rights liberalism, hence the extent to which such assumptions corrode criminal law liberalism when applied by a criminal law institution.Footnote 23 Thus, confidently maximizing the protection of victims in order to vindicate their rights may culminate in punishing individuals without fair warning, culpability, or fair labelling. In this way, human rights liberalism produces a criminal law system that is increasingly authoritarian in its disregard for constraining principles, and risks using the accused as an object in a didactic exercise rather than respecting autonomy and fairness.
The following sections will look at three modes by which assumptions of human rights liberalism distort ICL discourse: interpretive assumptions (section 2), substantive and structural conflation (section 3), and ideological assumptions (section 4). The article will use the doctrine of command responsibility as a recurring example under all three modes. As the following analysis may at times seem rather critical, some important qualifications are in order.
First, I by no means wish to suggest that ICL jurisprudence is uniformly flawed. The article provides examples to demonstrate tendencies that are common in ICL discourse. I do not suggest that the tendencies amount to an iron rule.Footnote 24 The observations are offered in the spirit of improving a relatively new discipline. Through critical awareness of these heuristics and distortions, we may consciously determine the values and principles of the system and apply them consistently.
Second, the contradictions identified are a contingent phenomenon and not an immutable fatal flaw in the ICL project. The contradiction of simultaneous allegiance to incompatible liberalisms can be addressed by choosing one or the other or developing ICL's distinct philosophical underpinnings. The doctrinal contradictions can be unearthed and resolved by reforms that bring doctrines and principle into alignment. Indeed, insofar as some distortions arise from the normative assumptions of lawyers experienced in other domains, the phenomenon may already be diminishing somewhat as the composition of the ICL profession changes. While ICL in its earliest stages drew heavily from experts in human rights and humanitarian law,Footnote 25 there is now more emphasis on criminal law expertise, which may reduce the distortions caused by exogenous heuristics. However, demographics alone are unlikely to resolve the problem completely, given the extent to which assumptions and doctrines from neighbouring fields are already internalized into ICL, and given that ICL continues to draw content from those neighbouring fields. It therefore remains important to understand and expose the two competing liberalisms of ICL and its contradictory assumptions.
Finally, where the article exposes an internal contradiction, it should not be assumed that the author believes that the doctrine is wrong and the articulation of the principle is necessarily correct. The aim of this article is to expose some of ICL's internal contradictions with its own principles, and to examine the reasoning methods and assumptions that foster such contradictions. An evaluation of how best to resolve those contradictions – which would require inter alia a philosophical analysis of the merits of particular principles and doctrines – will be part of a much larger research project and will remain for future exploration. What is significant for present purposes is that ICL falls into contradictions with the promises it makes when it articulates its principles.
While ICL draws from human rights and humanitarian law on one hand and criminal law on the other hand, it cannot faithfully comply with the strictures of criminal and non-criminal law. The choice for ICL is either to bring itself into conformity with fundamental liberal principles or to abandon the claim that it conforms to them. Is ICL simply an enforcement arm of human rights and humanitarian law, adopting incarceration of violators as a new remedy available to victims? Or is it a system of criminal justice that respects the philosophical preconditions for the punishment and stigmatization of individuals? If the latter option is chosen,Footnote 26 then supporters of ICL must critically inspect the assumptions inherited from human rights law and shed those assumptions that do not comply with liberal criminal principles. A variation on, or complement to, this exercise would be to develop specific deontological justifications for particular adaptations of principles of national criminal law.Footnote 27 Whatever route may be chosen, it is hoped that this analysis will contribute to the discipline by (i) exposing the need for such a project, and (ii) engendering critical awareness of the reasoning techniques and assumptions that distort current ICL discourse, thereby fostering more sophisticated theoretical scrutiny of doctrines in future discourse.
2. Interpretive assumptions
2.1. Victim-focused teleological reasoning
ICL jurisprudence proclaims that it follows particularly stringent standards in interpreting definitions of crimes and inculpatory rules, applying only norms that are ‘clearly’ and ‘beyond doubt’ customary law.Footnote 28 ICL emphasizes its faithful adherence to the principle of strict construction, which provides that ambiguities are to be resolved in favour of the accused.Footnote 29 As the ICTY has held,
penal statutes must be strictly construed, this being a general rule which has stood the test of time . . . A criminal statute is one in which the legislature intends to have the final result of inflicting suffering upon, or encroaching upon the liberty of, the individual . . . [T]he intention to do so shall be clearly expressed and without ambiguity. The legislature will not allow such intention to be gathered from doubtful inferences from the words used . . . [I]f the legislature has not used words sufficiently comprehensive to include within its prohibition all the cases which should naturally fall within the mischief intended to be prevented, the interpreter is not competent to extend them.Footnote 30
Similarly, Article 22(2) of the ICC Statute affirms that ‘the definition of a crime shall be strictly construed and shall not be extended by analogy. In case of ambiguity, the definition shall be interpreted in favour of the person being investigated, prosecuted or convicted.’
It would seem that, notwithstanding these proclamations of principle, ICL is heavily influenced by the distinctively ‘liberal’, ‘broad’, ‘progressive’, and ‘dynamic’ approach to interpretation that is a hallmark of human rights law.Footnote 31 While purposive interpretation may be found in any area of law, human rights law features a distinctively progressive brand, on the grounds that it aims at increasing the protection of human dignity rather than reciprocal obligations undertaken by states.Footnote 32 As ICL norms are often drawn from human rights or humanitarian law, practitioners can easily fall back on these familiar interpretive approaches,Footnote 33 and indeed ICL discourse frequently bears the fingerprints of the distinct interpretive approach of human rights.
Notwithstanding the general declarations offered by ICL, examination reveals that a technique commonly used in ICL is (i) to adopt a purposive interpretive approach; (ii) to assume that the exclusive object and purpose of an ICL enactment is to maximize victim protection; and (iii) to allow this presumed object and purpose to dominate over other considerations, including if necessary the text itself.Footnote 34 The principle of strict construction fails to constrain this technique because, as in many national systems, this principle is applied only as a final resort, after other ‘canons of construction fail to solve’ the question.Footnote 35 Applying, at a prior stage, a teleological approach that maximizes victim protection means that there is never an ambiguity left for strict construction to resolve,Footnote 36 because all ambiguities have already been resolved against the accused.
As a result, the promise of in dubio pro reo, which ICL holds out to accused, and which bolsters ICL's legitimacy, is in reality completely inverted, and the rule faced by the accused is closer to in dubio contra reo. Footnote 37
The first step in this process is the invocation of the interpretive method of the Vienna Convention on the Law of Treaties.Footnote 38 One could object that the Vienna Convention deals with interpretation of obligations undertaken between states rather than criminal prohibitions directed at individuals. Nonetheless, the application of the fairly self-evident factors listed in the Vienna Convention (textual, contextual, and purposive) is not necessarily problematic. The departure from strict construction arises when one presumed expansive purpose is allowed to prevail so decisively over text, context, and other purposes.
For example, the ICTY has often held that the purpose of the Geneva Conventions is to ‘ensure the protection of civilians to the maximum extent possible’,Footnote 39 and has used this proposition to prefer a ‘less rigorous standard’ in interpretation of its provisions.Footnote 40 However, it is doubtful that the Geneva Conventions can credibly be said to reflect such a singular purpose.Footnote 41 If the Geneva Conventions had one sole purpose of ‘maximizing’ the protection of civilians, then they would contain only a single article, forbidding any use of force or violence that could affect civilians. Instead, the Geneva Conventions contain complex provisions, indicating a much more nuanced purpose of promoting civilian protection while balancing military effectiveness and state security.Footnote 42 Thus interpretations that focus only on maximizing humanitarian protection systematically distort the balances struck in the law.
Victim-focused teleology in ICL can become so extended that it bears no relation to the actual text. For example, in Čelebići the Appeals Chamber held that
to maintain a distinction between the two legal regimes [international and internal armed conflicts] and their criminal consequences in respect of similarly egregious acts because of the difference in nature of the conflicts would ignore the very purpose of the Geneva Conventions, which is to protect the dignity of the human person.Footnote 43
Even if we welcome the regulation of internal conflicts, this particular argument warrants scepticism. The Geneva Conventions contain over 300 articles regulating international armed conflict and only one short article on internal conflicts, which was adopted only after acrimonious debate.Footnote 44 The Conventions criminalize some violations in international conflicts but, pointedly, do not do so in internal conflicts. It strains credibility to suggest that the ‘very purpose’ of the Conventions logically compels an outcome so deeply contradicted by the actual terms of the Conventions.
At times the single-issue teleological method is applied in a manner that, in addition to being objectionable for the reasons above, is also a non sequitur. For example, in interpreting the definition of crimes against humanity and war crimes, the International Criminal Tribunal for the former Yugoslavia (ICTY) has held that the aim of the ICTY Statute is to ‘make all crimes against humanity punishable’Footnote 45 or ‘not to leave unpunished any person guilty of any [war crime]’.Footnote 46 Assuming arguendo that these propositions about the Statute's purpose are correct, they shed no light whatsoever on the actual content of the definitions of crimes against humanity or war crimes. In other words, to assert that the Statute includes ‘all x’ still leaves unanswered the question of what ‘x’ entails – a question that must be answered by examining the text and the legal authorities. Nonetheless, reference to this purpose is often used in defining that content broadly.
Interestingly, in Tadić the prosecution attempted to use the ‘Statute includes all x’ technique to argue that there is no motive requirement for crimes against humanity, and the Appeals Chamber rejected the argument because it ‘begs the question . . . whether a crime committed for purely personal reasons is a crime against humanity’.Footnote 47 Yet, a few paragraphs later, the Appeals Chamber employed the very argument it had just held to be fallacious, in the course of its analysis of whether crimes against humanity require a discriminatory animus: ‘As rightly submitted by the Prosecution, the interpretation of Article 5 in the light of its object and purpose bears out the above propositions. The aim of those drafting the Statute was to make all crimes against humanity punishable.’Footnote 48
Turning to command responsibility (which will be used as a recurring example under all three modes), we find that victim-focused teleological reasoning is also prominent in the discourse on command responsibility. (Even the origins of the doctrine lie in victim-focused teleological reasoning; the doctrine was judicially created and applied in post-Second World War war crimes trials despite the absence of a provision in the Nuremberg and Tokyo Charters, after the Yamashita decision deduced the doctrine from the need to curb violations of the laws of war.Footnote 49)
To take a typical example from contemporary literature (and this example is representative rather than intended to single out any author), Greg Vetter offers a simple syllogism that is characteristic of much of ICL discourse.Footnote 50 First, to deter human rights abuses, potential perpetrators must perceive prosecution as a possible consequence of their actions.Footnote 51 Second, some features of the command responsibility doctrine in the Rome Statute are ‘less strict’ than other instruments,Footnote 52 because they set ‘an easier standard for the accused to exonerate himself or herself’.Footnote 53 Therefore, the weaker standard is plainly ‘undesirable’ because it will not deter to the same extent;Footnote 54 it reduces the efficacy of the ICC because superiors will ‘not be as fearful’ of being convicted.Footnote 55 The syllogism demonstrates that if we look exclusively at maximizing protection of victims, then analysis of doctrines becomes a simple one-dimensional task of identifying the broadest, and hence best, articulations. What is typically missing from such analyses is introspection on fundamental principles and on whether conviction is appropriate in the circumstances. For example, Vetter demonstrated the concerns with the ICC Statute provisions by showing that persons who were convicted in the Tokyo tribunals might be acquitted under the ICC standard; for example, a superior might be acquitted if crimes were committed by persons not under her authority and control,Footnote 56 or a civilian manager might be exonerated for crimes committed by civilian employees while off duty and without her knowledge.Footnote 57 However, it was not shown that persons ought to be convicted in such circumstances; in the two examples given, conviction would seem to be an affront to the principle of personal culpability.Footnote 58
Even the most sophisticated scholars frequently base their analysis exclusively on increasing victim protection, squeezing out reflection on conformity to fundamental principles. While examples abound, I will draw them from two of the most respected leaders in the field, Roger Clark and Mark Osiel. Roger Clark has argued that (i) the laws of war aim to minimize loss of life; (ii) this is best served by a broad view of command responsibility; and, therefore, (iii) a negligence standard is more appropriate than an actual-knowledge standard, to respond to the humanitarian goals of the doctrine.Footnote 59 Mark Osiel has argued of the doctrine that ‘the scope of liability is formally overinclusive, but only in order to make up for severe practical dangers of underinclusiveness’.Footnote 60 Thus ‘the hope is that the threat of serious criminal liability for “mere” negligence will lead even the most reluctant commander (in order to protect himself) to take all reasonable measures to prevent war crimes by subordinates’.Footnote 61 The arguments in question do not provide a desert-based account to justify such liability in accordance with the fundamental principles of ICL.
The problem with victim-focused teleological reasoning is that it conflates the ‘general justifying aim’ of the criminal law system as a whole – which may be a utilitarian aim of protecting society – with the question of whether it is justified to punish a particular individual for a particular crime.Footnote 62 George Fletcher has drawn an analogy to a tax regime to demonstrate the problem: while the primary function of an income tax regime is to raise revenue, it does not follow that each decision to allow or disallow a given deduction should be resolved by reference to this overriding goal.Footnote 63 In a criminal justice system adhering to liberal principles, ‘society has no warrant to treat persons unjustly in its pursuit of utilitarian gains’.Footnote 64 Thus the aim of criminal law may be to protect society from individuals, but the pursuit of that goal is qualified by principled restraints to protect the individual from society.Footnote 65
This article does not suggest that teleological reasoning is per se inappropriate. It simply argues that victim-focused teleological reasoning has engendered contradiction with the fundamental principles that the system claims to uphold. In addition to the obvious contradiction with the principle of strict construction, victim-focused teleological reasoning can foster exuberant interpretations that overlook culpability and fair labelling, as will be seen in the following illustration.
2.2. Interpretive approaches and joint criminal enterprise
The doctrine of ‘joint criminal enterprise’ (JCE) is a judicial innovation which has grown to play a major role in the charges and convictions before the tribunals.Footnote 66 As articulated in tribunal jurisprudence, JCE requires a plurality of persons acting pursuant to a common plan, design, or purpose involving the commission of a crime, and that the accused participate in the common purpose.Footnote 67 The contribution need not entail physical participation in any element of a crime.Footnote 68 The doctrine has three variations, with the ‘basic’ form being comparable with ‘co-perpetration’ or ‘common purpose’ as known in national systems,Footnote 69 while the ‘systemic’ form includes a contribution to a system of ill-treatment, and the ‘extended’ form includes crimes that were not intended but which were foreseeable.Footnote 70
The doctrine was initially welcomed as a valuable tool in securing convictions, but as the doctrine continued to develop through generous judicial interpretation, commentators have begun to question the doctrine's conformity with fundamental principles.Footnote 71 Some have argued that the expansiveness of the doctrine raises a prospect of ‘guilt by association’,Footnote 72 and in some circles JCE has been wryly referred to as ‘just convict everyone’.Footnote 73 The process by which JCE was judicially introduced is problematic in terms of the principle of legality, as will be discussed below; in addition its substance is problematic in terms of culpability and fair labelling.
First, a negligible contribution can result in massive criminal liability. The accused need not perform any part of the actus reus of any crime;Footnote 74 the accused's acts need only to contribute to a common design, and this contribution need not be ‘substantial’ nor even ‘significant’.Footnote 75 The rejection of ‘significant’ as a threshold logically entails that an insignificant contribution suffices for JCE.
Second, there are no legal limits on the scope of JCE,Footnote 76 so the resulting exposure to liability is vast. Initial justifications of JCE typically drew an analogy to contexts familiar to national law, such as a small group of criminals jointly planning a bank robbery.Footnote 77 As the doctrine has developed, it has been found to apply to atrocities across an entire region,Footnote 78 or even to a ‘nationwide government-organized system’,Footnote 79 or a ‘vast criminal regime comprising thousands of participants’,Footnote 80 in which the persons actually committing crimes are ‘structurally or geographically remote from the accused’.Footnote 81
Third, compounding these problems, the JCE doctrine deems the accused to have ‘committed’ the crimes, and indeed deems all members to be ‘equally guilty of the crime regardless of the part played by each in its commission’.Footnote 82 Thus the low-level individual making an insignificant contribution becomes liable as a perpetrator of all of the crimes in the JCE.
Fourth, the supposedly stringent requirement of ‘common intent’ which is claimed to justify such treatment and to make JCE tantamount to perpetrationFootnote 83 has also been eroded through generous judicial interpretation. Decisions have found the common intent requirement to be satisfied by dolus eventualis, which means that a person intends a consequence, even if it is not her immediate aim, if she could foresee a substantial possibility of her course of action producing that outcome.Footnote 84 This relaxed requirement undermines the supposedly crucial distinction between contribution-plus-knowledge (required for mere aiding and abetting) and contribution-plus-intent (required for JCE).
The Kvočka case held that, under dolus eventualis, awareness of contribution to the ‘everyday functioning and maintenance’ of a system is also sufficient to establish the needed mens rea.Footnote 85 Evidence that a person sought to prevent colleagues from mistreating victims, feared the consequences of quitting, or was seen as a ‘traitor’ because of sympathies for victims, would not detract from ‘knowledge’ of contribution to ‘everyday functioning and maintenance’, and hence deemed ‘intent’, and hence deemed ‘commission’ of all crimes in the JCE.Footnote 86 In this way, the supposedly stringent intent requirement which justified the harshness of JCE in fact reduces to a knowledge requirement.Footnote 87
An additional concern is that extended JCE is used to circumvent the special intent requirements that delineate particularly serious crimes. For example, as the Appeals Chamber has affirmed, ‘genocide is one of the worst crimes known to humankind, and its gravity is reflected in the stringent requirement of specific intent. Convictions for genocide can be entered only where that intent has been unequivocally established.’Footnote 88 Nonetheless, in Brධanin, the Appeals Chamber held that a person can be convicted for ‘committing’ genocide via extended JCE, which requires only that it was foreseeable that others might commit genocide as a result of the common plan.Footnote 89 The Appeals Chamber held that JCE ‘is no different from other forms of criminal liability which do not require proof of intent to commit a crime’, such as aiding and abetting.Footnote 90 This argument overlooks that aiding and abetting makes a person an accessory, whereas JCE is a form of committing the crime. As a result, a person who was involved in a JCE to carry out unlawful deportations becomes a génocidaire if other participants start to commit genocide.Footnote 91 The person still faces the stigma of a conviction for committing genocide, while having satisfied neither the actus reus nor the hitherto indispensable mens rea for genocide (in fact not even awareness of genocidal acts is required).
One might argue that prosecutorial discretion will avoid the most egregious abuses of the doctrine. However, use of over-broad doctrines with sanguine reliance on prosecutorial or judicial discretion to avoid abuses is inconsistent with the principles of legality and fair warning:Footnote 92 ‘discretion is not a reliable substitute for getting the rule right in the first place’.Footnote 93 One might also argue that the problem of assigning equal culpability to different levels of participation can be ‘adequately dealt with at the sentencing stage’.Footnote 94 However, such a solution contravenes fair labelling by lumping together radically different levels of blameworthiness under one label.Footnote 95 More importantly, it contradicts the principle of culpability to convict a person for committing thousands of murders and then take into account at the sentencing stage that he did not really do any such thing.
To summarize, JCE contradicts fundamental principles in several ways. A low-level individual making an insignificant and reluctant contribution can be deemed to be a ‘perpetrator’ of thousands of murders. It contradicts personal culpability to convict a person for ‘committing’ crimes when he or she satisfies neither the objective nor subjective elements of the offence.Footnote 96 The absence of correlation between the wrongdoing and the resulting stigma violates the principle of fair labelling.Footnote 97 Use of extended JCE to circumvent the special intent requirement of genocide is a further contravention of personal culpability and fair labelling, since the person is fixed with the enormous stigma of a conviction for committing genocide without satisfying any of the elements warranting that stigma.Footnote 98
The puzzle is, how did international tribunals composed of eminent jurists, and staffed by committed lawyers, come to embrace doctrines that are increasingly widely regarded as plainly contradictory to the fundamental principles proclaimed by the system itself? The jurisprudence does not present JCE as a necessary exception to fundamental principles, but rather as fully consistent with them. The doctrine came about through a series of analytical steps, each of which must have seemed convincing to the participants at the time.
To examine the methods of reasoning, we start with the Tadić case, which first developed the doctrine,Footnote 99 where we find that JCE was born out of victim-focused teleology, and specifically from the ‘Statute includes all x’ technique. In Tadić there was a particular charge for which the evidence did not support a conviction for commission or for aiding and abetting under Article 7(1). After acknowledging that Article 7(1) ‘sets out the parameters of personal criminal responsibility under the Statute’, and that it referred only to persons who ‘planned, instigated, ordered, committed or otherwise aided and abetted in the planning, preparation or execution of a crime’, the Appeals Chamber nonetheless held:
An interpretation of the Statute based on its object and purpose leads to the conclusion that the Statute intends to extend the jurisdiction of the International Tribunal to all those ‘responsible for serious violations of international humanitarian law’ committed in the former Yugoslavia (Article 1) . . . Thus, all those who have engaged in serious violations of international humanitarian law, whatever the manner in which they may have perpetrated, or participated in the perpetration of those violations, must be brought to justice. If this is so, it is fair to conclude that the Statute does not confine itself to providing for jurisdiction over those persons who plan, instigate, order, physically perpetrate a crime or otherwise aid and abet in its planning, preparation or execution. The Statute does not stop there.Footnote 100
Yet the text of the Statute did stop there.Footnote 101 The method of reasoning, while fully consistent with the effet utile doctrine of areas such as human rights law, is disconcerting in a criminal law context. Normally in criminal law one determines whether an accused is criminally responsible by applying the applicable statute to the acts of the accused. The Tadić approach empowers the judges to read in additional terms to fulfil the imperative of ‘ensuring that no person responsible for serious violations goes unpunished’, which in turn assumes that judges are able to determine that persons are ‘responsible’ independently of the terms of the Statute. By developing the JCE doctrine to cover the situation at hand, the Appeals Chamber was able to replace an acquittal with a conviction for ‘commission’.Footnote 102
Using this interpretive method, the judges read in a ground of liability vastly broader than any appearing in the list. The approach falls remarkably far from the declared approach of strict construction, reliance on unambiguous terms, and rejection of doubtful inferences.
The same victim-focused teleological reasoning was pursued in subsequent cases as, issue by issue, chambers opted for the more progressive option. Thus decisions held that no agreement between participants is needed,Footnote 103 that a person's contribution need not be significant,Footnote 104 that the physical perpetrators need not be part of the JCE,Footnote 105 that dolus eventualis suffices for the crucial mental element,Footnote 106 that the doctrine is not limited in scale and may include participants remote from each other,Footnote 107 and that JCE-III can be used to circumvent the special intent required for conviction for ‘committing’ genocide.Footnote 108 Through reliance on such reasoning techniques, a system that prides itself on its compliance with fundamental principles wound up amalgamating all of the most sweeping features of various national laws into a single all-encompassing doctrine divorced from culpability and fair labelling.
2.3. Victim-focused teleological reasoning aggravated by utopian aspirations
The problem of victim-focused teleological reasoning is aggravated where ICL also becomes imbued with utopian aspirations.Footnote 109 For example, whereas national criminal law seeks to manage crime – by reducing or at least visibly responding to crimes – ICL appears to aim, more ambitiously, to end the crimes. For example, the UN Security Council resolutions creating the ICTY and ICTR refer to the determination ‘to put an end to such crimes’ and express confidence that the creation of tribunals ‘would enable this aim to be achieved and would contribute to the restoration and maintenance of peace’.Footnote 110 This more urgent aspiration arguably creates greater pressure to be ruthless in the articulation of draconian norms.
The problem may be compounded further still by the grave disparity between the utopian aspirations and the dystopian realities faced by ICL. In other words, the severity and scale of the crimes and the extreme difficulty of securing arrests means that, once an accused is at trial, the desire is stronger to make a clear object lesson, and to serve the didactic function of ICL, in the desperate hope of trying to have a preventive impact in such chaotic situations. At times it seems that ICL seeks to offset its weakness on the ground through more draconian rules, or in other words, to overcompensate for material weakness through normative harshness.Footnote 111
To take an example from the literature, Professor Sherrie Russell-Brown starts from the ‘foundational precept . . . that the continuing commission of gender crimes in war must end’. Footnote 112 It is perhaps unsurprising that from a foundational precept of ending such crimes, she concludes that the already problematic doctrine of command responsibility must be rendered harsher still: ‘it is unacceptable to allow commanders to escape criminal responsibility for their subordinates' gender crimes on the basis that the commanders lacked “knowledge”’.Footnote 113 As will be discussed below,Footnote 114 the doctrine already raises principled concerns in that a person may be convicted for a serious international crime without the commission, participation, contribution, or subjective mens rea that would normally be required for such crimes. The proposal to abolish the mens rea requirement altogether would doubtless facilitate convictions, but it would deviate even further from the principle of culpability. Thus admirable but utopian objectives such as eliminating crimes, which no criminal law system can achieve, are likely to generate calls for harsher and harsher rules, and thus promote a tendency away from principled restraints.
To take an example from the case law, consider the Erdemović case.Footnote 115 In Erdemović, the majority concluded that duress may never be raised as a defence in relation to killing of civilians.Footnote 116 The decision has been criticized by some commentators for lacking sensitivity to fundamental principles,Footnote 117 departing from previous ICL pronouncements on the role of moral choice,Footnote 118 and disregarding the fact that the only way for Erdemović to be innocent was to be dead.Footnote 119 What is of interest for present purposes, however, is not whether the decision was correct or incorrect, but, rather, how the majority framed the issue and hence sidestepped the deontological discussion.
Neither the Statute nor customary law resolved the availability of a duress defence, and examination of national legal systems revealed that civil law and common law systems took different approaches. Such a clear stalemate in the sources of law presented an archetypal opportunity to apply the principle of lenity which ICL purports to follow, and to adopt the interpretation favourable to the accused.
Instead, however, the majority opted to fill the gap by reasoning teleologically from the mandate granted by the Security Council, which was to “halt and effectively redress” the widespread and flagrant violations of international humanitarian law occurring in the territory of the former Yugoslavia and to contribute thereby to the restoration and maintenance of peace'.Footnote 120 By focusing on this onerous responsibility of ‘halting’ violations, the majority could only favour sending a strong message:
We would assert an absolute moral postulate which is clear and unmistakable for the implementation of international humanitarian law . . .Footnote 121 We do so having regard to our mandated obligation under the Statute to ensure that international humanitarian law, which is concerned with the protection of humankind, is not in any way undermined.Footnote 122
The adopted aim of ‘halting’ crimes was one which only the harshest measures could hope to satisfy. Hence the chamber focused on the need to send a clear message, but overlooked whether the system's principles allowed it to use Dražen Erdemović to send that message. In other words, the majority focused not on what Erdemović did but rather on what others might do in the future.Footnote 123 It was through this process of reasoning that a system which claims that it only deals with the ‘worst of the worst’, the ‘persons most responsible’ for the most serious crimes, was transformed into a ‘criminal law that could be obeyed only by exceptional individuals’.Footnote 124
In conclusion, the cognitive dissonance arising from ICL's simultaneous allegiance to human rights liberalism and criminal law liberalism, and hence two contradictory interpretive approaches, reveals itself in curious ways. For example, at the same time that the ICTY insists that it scrupulously applies only rules that are ‘beyond any doubt customary international law’,Footnote 125 it also takes credit for having ‘expanded the boundaries of international humanitarian and international criminal law’.Footnote 126 Whichever interpretive approach one may prefer, one must recognize that this is an overt contradiction: the Tribunal cannot both apply the law strictly as it was and expand it. Such contradictions are the products of the conflicting normative assumptions simultaneously inhabiting ICL. While criminal law principles forbid the judicial expansion of norms, human rights and humanitarian law assumptions lead us to embrace and to celebrate that very expansion. Contradictorily, we both laud and deny the expansions – a product of the two incompatible liberalisms at the heart of ICL.
3. Substantive and structural conflation
3.1. Substantive and structural conflation in ICL discourse
The assumptions of human rights and humanitarian lawyers can also distort ICL reasoning through substantive and structural conflation. Many of the prohibitions of ICL are drawn from, and similar to, prohibitions in human rights and humanitarian law. Faced with familiar-looking provisions, ICL practitioners often assume that the ICL norms are coextensive with their human rights or humanitarian law counterparts, and uncritically transplant concepts and jurisprudence from other domains to flesh out their content. Such assumptions overlook the fact that these bodies of law have different purposes and consequences and thus entail different philosophical commitments.
Human rights law and humanitarian law apply to collective entities – states or parties to conflict. They focus on systems, seeking to improve the practices of states (or parties to conflict) in order to advance protection of and respect for identified beneficiaries. The remedies in each area of law are roughly comparable to civil remedies, such as a cessation of the conduct, an apology, an undertaking of non-repetition, and possibly compensation or other efforts to restore the status quo ante.Footnote 127
The primary focus of ICL, on the other hand, is on the culpability of individuals. Furthermore, while many or most of its prohibitions are rooted in human rights or humanitarian law, and are intended to reinforce those rules, the scope of ICL is much narrower: it addresses only the most serious crimes of concern to the international community as a whole.Footnote 128 Moreover, ICL is enforced through a particularly robust method – the arrest, stigmatization, punishment, and imprisonment of individual human beings found responsible for crimes. As a result, ICL features several additional restraining principles that are unknown in other legal domains, such as fair warning, culpability, and fair labelling.
Two closely related features are at work: (i) an assumption that the norms must have the same substantive content when transplanting them from other domains, and (ii) reliance on structural assumptions from human rights and humanitarian law when reasoning about the norms (for example, focusing on improving systems rather than on culpability of individuals). Substantive conflation is arguably a product of structural assumptions, since the structural assumptions blind participants to the changed context and hence the need for fresh scrutiny if rules are to be applied to individuals rather than to systems.
There are two variations of this substantive and structural conflation. The crude form is the assumption that, because a prohibition is recognized in human rights or humanitarian law, it therefore must be (or ought to be) criminalized in ICL as well.Footnote 129 Such arguments tend to overlook questions of legality and personal culpability and whether criminal law is the appropriate tool to deal with the problem.Footnote 130
The more subtle, and more interesting, variation occurs with respect to those norms that are indisputably recognized as criminalized in ICL, and which are drawn from human rights law or humanitarian law. Where an ICL prohibition is drawn from another area of law, it is understandable to assume that the norms have the same scope as they have in their original domain. Through the resulting unreflective transplantation or mimicry of human rights or humanitarian law norms, norms are absorbed into criminal law without awareness that they may be novel to criminal law and hence without scrutiny as to whether they comply with the fundamental principles peculiar to criminal law.
As an illustration, consider Common Article 3 to the Geneva Conventions, which requires that, before any sentencing of protected persons, a party must provide a trial affording ‘all indispensable judicial guarantees’.Footnote 131 A significant number of guarantees have been identified as ‘indispensable’.Footnote 132 Assume that a judge has made a ruling that an accused was too disruptive to remain in the courtroom, but in hindsight the judge is found to have applied the standard erroneously. It would follow that (i) the error breached the guarantee of the right to be present; (ii) the breach, although it may be minor and inadvertent, is nonetheless a breach; (iii) a breach of even one guarantee constitutes a failure to provide ‘all’ guarantees; and therefore (iv) the error would violate Common Article 3 and hence quite rightly require an appropriate humanitarian law remedy, which might include a reform, a new trial, an apology, or compensation.Footnote 133
Common Article 3 is recognized in ICL as a fundamental provision of IHL, such that its violation also gives rise to individual criminal responsibility for war crimes.Footnote 134 However, if an identical standard were applied in ICL, as authorities assume,Footnote 135 then by the same chain of reasoning as above it would follow that the breach of even one guarantee not only violates Common Article 3 but also constitutes a war crime. On a literal application of the provision and established ICL principles, both the actus reus and mens rea would easily be established.Footnote 136 Thus the judge would be liable to imprisonment as a war criminal, since he or she sentenced a protected person without a trial respecting all guarantees.
However, even in the most advanced legal systems of the world, trial judges are frequently found to have breached a right, and new trials are granted; we do not imprison the erring judge. Thus, although Common Article 3 has been incorporated directly into ICL, it surely requires some form of translation so that it is correctly focused on criminally blameworthy conduct rather than judicial errors.
Examples of assumptions of coextensiveness which fail to reflect on the need to translate norms to satisfy criminal law principles are readily found in ICL discourse; we shall take one example from literature and one example from jurisprudence.
For an example from literature, a thoughtful scholar in a leading commentary on the ICC Statute has voiced concern about a mens rea requirement inserted into the ICC Elements of Crimes for the crime of enforced disappearance.Footnote 137 The Elements indicate that a person denying the fact of detentions also commits enforced disappearance, but require that the person have awareness of such detentions.Footnote 138 This objection appears to be a typical illustration of a conflation with human rights. In human rights law, the knowledge of the individual denying the detention is irrelevant, because the focus is on the system as a whole and its impact on the victim. In ICL, however, the focus is on the criminal culpability of the accused individual. If, for example, a bureaucrat fails to acknowledge a detention when he or she in fact has no awareness of such detentions, then the conduct amounts to truth-telling rather than a crime. Thus assumptions that norms from other domains should be replicated exactly as found can lead to overlooking fundamental principles of criminal liability.
For an example from case law, we may turn to our recurring example of the command responsibility doctrine.
3.2. Illustration: command responsibility and conflation
Whereas the JCE doctrine has begun to attract criticism for transgressing fundamental principles, the command responsibility doctrine has received a much warmer reception. For example, Danner and Martinez, after leading the charge against JCE for its deviations from fundamental principles, concluded that ‘in contrast to JCE, command responsibility doctrine does not call for significant doctrinal reform’.Footnote 139 Similarly, Mark Osiel, while condemning JCE as too inclusive, found in contrast that it was ‘too hard to find people liable under command responsibility’.Footnote 140 However, while command responsibility may not transgress principles as flamboyantly as JCE, the doctrine does raise its own questions in terms of culpability and fair labelling.Footnote 141
The command responsibility doctrine imposes liability where (i) there is a superior–subordinate relationship; (ii) the superior knew or had reason to know that a subordinate was about to commit crimes or had done so; and (iii) the superior failed to take the necessary and reasonable measures to prevent such acts or to punish the perpetrators thereof.Footnote 142
The command responsibility doctrine allows conviction for a serious international crime committed by another person, on the basis of an omission, a ‘had reason to know’ standard, and without any causal contribution to or impact on the crime. The doctrine entails several significant shortcuts in imputing liability, which, under a liberal system of criminal justice committed to fundamental principles, would require careful attention and justification.Footnote 143 In the absence of a distinct crime of ‘breach of command responsibility’,Footnote 144 command responsibility can only be justified as a mode of liability, based on personal culpability and fair labelling, if it is to avoid accusations of ‘vicarious liability’Footnote 145 and incompatibility with the principle of culpability.Footnote 146
This article will focus on the problem of causal contribution. While some concerns have been raised about the omission standardFootnote 147 and the ‘had reason to know’ mens rea standard,Footnote 148 this article will assume that principled justifications of these features may be possible (although it remains significant that principled justifications have never been offered or considered necessary in ICL).
The problem of causal contribution arises where command responsibility is based solely on a ‘failure to punish’ after the fact, rather than any ‘failure to prevent’. Under Tribunal jurisprudence on command responsibility, a person may be convicted for an international crime based solely on a failure to punish another person for the crime, even if no fault or failing of the commander contributed to the crime, and even if the failure to punish did not contribute to any subsequent crime. It is obvious from an efficacy perspective that the threat of labelling the commander as a génocidaire may strengthen the incentive to punish diligently. But does it accord with the fundamental principles that ICL proclaims?
The principle of culpability has been declared the ‘foundation of criminal responsibility’ in Tribunal jurisprudence,Footnote 149 which found it to be ‘firmly established’ that ‘for the accused to be criminally culpable his conduct must have been proved, beyond a reasonable doubt, to have contributed to, or have had an effect on, the commission of the crime’.Footnote 150 Every other form of liability in Tribunal jurisprudence requires a ‘contribution’ to the crime for which the person is charged.Footnote 151 This comports with the ‘central role’ of causal contribution in the principle of culpability.Footnote 152 Similarly, Dressler argues that causal contribution is ‘the instrument we employ to ensure that responsibility is personal’.Footnote 153 Husak discards ‘punishment of a person for a result to which he did not contribute in any way’ as a ‘monstrosity’, raising it only as a hypothetical example of a ‘most flagrant violation of the causal principle’.Footnote 154 Most writings on the topic confirm the centrality of causation.Footnote 155
The command responsibility doctrine does not comply with this principle, since a person may be convicted for an international crime without contributing to the crime or having any effect on it. Even if we agree that failure to punish crimes is worthy of criminalization, it is simply inaccurate to label such a failure as ‘genocide’.
The main line of defence within ICL discourse is to deny that the doctrine holds the commander responsible for the underlying crimes. For example, Judge Shahabuddeen has argued that ‘reading the provision reasonably, it could not have been designed to make the superior a party’.Footnote 156 In a similar vein, the Appeals Chamber declared in Krnojelac that ‘[i]t cannot be overemphasized that, where superior responsibility is concerned, an accused is not charged with the crimes of his subordinates but with his failure to carry out his duty as a superior to exercise control.’Footnote 157 Commentators have faithfully echoed this assertion.Footnote 158 At first glance, these assertions seem reassuring: if, for example, the commander were convicted not for genocide but for ‘failure to exercise control’, then he would be convicted for his own misconduct, removing the ‘fair labelling’ or culpability problems as well as need to justify command responsibility as a mode of liability for the acts of others.
Unfortunately, however, a cursory review of the Tribunal statutes and practice shows unmistakably that the commanders are in fact charged with, and convicted for, the war crimes or crimes against humanity committed by subordinates.Footnote 159 Indeed, the tribunals have no jurisdiction over any crime of ‘failure to exercise control’. They only have jurisdiction over genocide, crimes against humanity, and war crimes, and those are the crimes of which they find people guilty and enter convictions.Footnote 160 For example, despite the assertions in Krnojelac that the accused was charged with failure to exercise control and not the underlying crimes, Krnojelac was charged with and, through command responsibility, convicted for, numerous war crimes and crimes against humanity of torture, murder, and persecution, and he was sentenced for those crimes.Footnote 161 The label attached to the charges, convictions, and sentence expressed to the world that he bore responsibility for war crimes and crimes against humanity. The tribunals are imposing such labels, for crimes bearing enormous stigma, in contravention of the principle of culpability recognized by the system, which requires a causal contribution to the crime for which one is convicted.
If the doctrine were to conform to the principle of culpability, liability for international crimes based on a failure to punish would require that the failings of the commander ‘contributed to, or . . . had an effect on’ crimes.Footnote 162 In the absence of any such contribution, to convict a person for genocide, crimes against humanity, or war crimes, and to impose the stigma that such crimes bear, contradicts the principle of culpability which ICL claims to respect.
While the difficulties with command responsibility are more subtle than those arising in JCE, it appears that the command responsibility doctrine also departs from the principles that ICL proclaims. Again, it is curious that this stark deviation has gone largely unnoticed in jurisprudence.Footnote 163
This article suggests that the inadvertent contradiction between doctrine and principle was produced in part by conflation between the humanitarian law procedural duties of commanders and the distinct question of assigning criminal liability for the acts of another.Footnote 164
Additional Protocol I to the Geneva Conventions (AP I) highlights the important distinction between the commander's procedural duty under humanitarian law and the commander's personal criminal liability.Footnote 165 Article 87 of AP I imposes, as a matter of humanitarian law, a general duty on commanders to prevent breaches by their subordinates, to report breaches to competent authorities and to initiate disciplinary or penal actions against violations.Footnote 166 This general duty – enforceable by international law remediesFootnote 167 – is supplemented by a narrower, criminal provision, in Article 86(2):
The fact that a breach of the Conventions or of this Protocol was committed by a subordinate does not absolve his superiors from penal or disciplinary responsibility, as the case may be, if they knew, or had information which should have enabled them to conclude in the circumstances at the time, that he was committing or was about to commit such a breach and if they did not take all feasible measures within their power to prevent or repress the breach.
Two features of Article 86(2) must be highlighted. First, whereas Article 87(3) includes the reporting of past crimes (‘had committed’), the penal provision of Article 86(2) refers only to ongoing or imminent crimes (‘was committing or about to commit’), thus requiring contemporaneity – and hence a possibility of influencing the subordinate's behaviour – for criminal liability. Second, the reference to ‘penal responsibility’ does not specify whether the superior must be punished for the underlying crime or for a different penal offence, such as dereliction of duty in the failure to supervise.
These differences reflect structural differences between criminal law and humanitarian law. The humanitarian law procedural duty quite reasonably does not require any causal contribution by the commander to the crimes. The purpose of the humanitarian law provision is to promote compliance of subordinates with humanitarian law by requiring a system of prevention and repression.Footnote 168 If a commander fails to punish a past crime, his government may be held liable for his breach of the rule.Footnote 169 However, this is not the same as asserting that the commander is personally guilty of the subordinate's crime; such an assertion requires consideration of fundamental principles.
Nonetheless, ICL discourse – both jurisprudence and literature – frequently assumes the coextensiveness of the humanitarian law norm and the criminal law norm, without reflecting on the different structures and consequences and hence the different principles in play.Footnote 170
For example, the drafters of the ICTY and ICTR Statutes blithely combined Article 86(2) and Article 87(3) into a single criminal provision encompassing both pending crimes and past crimes, and both failure to prevent and failure to punish.Footnote 171 In doing so, they overlooked the differences between the criminal and non-criminal provisions and insouciantly overwrote the causal contribution requirement.
Rather than detecting the problem, tribunal jurisprudence has followed this same pattern of blurring the humanitarian law procedural duty and personal criminal liability. In Blaškić, defence counsel argued that some contribution to the crimes is necessary for liability, and hence failure to punish by itself is not sufficient for liability under international law; the failure must encourage or facilitate future crimes.Footnote 172 To refute the defence argument, the Appeals Chamber relied on Article 87(3) of the Protocol, noting that it imposes a duty to punish persons responsible for past crimes.Footnote 173 The Chamber pointed out that ‘disciplinary or penal action can only be initiated after a violation is discovered’, and hence it found the defence argument ‘illogical’.Footnote 174 The Chamber's argument overlooks, however, that Article 87(3) deals with the procedural duty of humanitarian law, not the assignment of personal criminal liability. Indeed, Additional Protocol I specifically separated the broad humanitarian law duty from the narrower criminal law provision in Article 86(2). If anything, the elaborate separation of the provisions in Additional Protocol I should have given rise to an a contrario interpretation, or at least triggered reflection on whether the separation might have a principled basis.
Similarly, tribunal jurisprudence has assumed that the ‘should have known’ mental standard must have the same meaning in ICL as it does in humanitarian law, without inquiring whether the principle of personal culpability requires a different standard in criminal law.Footnote 175 This tendency to conflate may also be seen in literature, where it is frequently assumed that any differences in scope between the humanitarian procedural duty and the imposition of criminal liability are obviously ‘gaps’, ‘lacunae’, or ‘steps backward’ which will ‘allow atrocities to go unpunished’.Footnote 176
By conflating the humanitarian and the criminal norm, the tribunals appear to have unwittingly expanded the criminal prohibition. Ignoring the differences between Articles 86(2) and 87(3), the jurisprudence has effectively grafted criminal liability onto the broader humanitarian law duty, without considering whether criminal law warrants a narrower ambit. Further compounding the problem, the jurisprudence not only criminalizes the full scope of what was previously a humanitarian law duty, but it also declares the commander responsible for the underlying crimes, which carry severe stigma, without considering whether the commander's dereliction contributed in any way to the commission of such crimes. Because of conflation, judges, drafters, and commentators proceed in complacent confidence that they are simply following precedents, missing the fact that the ‘precedents’ are from different areas of law with different structures. As a result, broader norms are absorbed into ICL without triggering any alarm bells as to the need to scrutinize the proposed norm for compliance with fundamental principles.
4. Ideological assumptions (sovereignty and progress)
Finally, in addition to interpretive assumptions and conflation of norms, the ideological assumptions of human rights and humanitarian law may also influence ICL reasoning and undermine compliance with fundamental principles. Consider, for example, the narratives about progress and sovereignty in human rights law.
In a human rights instrument there is a fairly straightforward inverse relationship between the obligations undertaken and the freedom of action retained by the state. The more sovereign freedom of action retained, the narrower the human rights obligations. As a result, human rights and humanitarian law discourse routinely casts sovereignty in elemental opposition to normative progress. Sovereignty is the ‘traditional enemy’,Footnote 177 the ‘stumbling block in the advance of civil rights’,Footnote 178 the irksome vestige of ‘divine right’ ‘which human rights law is still in the process of extirpating’.Footnote 179 Sovereignty is the obstacle raised by short-sighted lawyers, diplomats, and bureaucrats,Footnote 180 and a constant threat to the project of international law.Footnote 181 This inverse relationship underlies the progress narrative of human rights, wherein a darker age of ‘sacrosanct and unassailable’ sovereignty has recently ‘suffered progressive erosion at the hands of the more liberal forces at work in the democratic societies, particularly in the field of human rights’.Footnote 182
The same assumptions about the interplay with sovereignty are replicated in ICL. Sovereignty is a ‘contradiction’ to human rights and justiceFootnote 183 and an ‘enduring obstacle’ in advancing ICL;Footnote 184 the ‘movement for global justice has been a struggle against sovereignty’,Footnote 185 such that human rights and justice must ‘trump’ state sovereignty,Footnote 186 since for sovereignty to prevail would be a ‘travesty of law and a betrayal of the human need for justice’.Footnote 187 As Robert Cryer wryly observes, ‘[w]hen sovereignty appears in [ICL] scholarship, it commonly comes clothed in hat and cape. A whiff of sulphur permeates the air.’Footnote 188
The common narrative on sovereignty in human rights and in ICL is probably in large part correct, and at least in some part overstated.Footnote 189 What is crucial for present purposes is that the importation of these perspectives routinely overlooks a significant difference between human rights or humanitarian law and ICL. In human rights or humanitarian law, where the treaties limit state behaviour, it is usually accurate to ascribe limitations in instruments to states' wishes to preserve sovereignty. Thus, in human rights discourse, the progress of ‘liberal forces’ generally means broader norms and hence more protection of individuals from the state; by contrast, a narrowing of a norm means a gain for sovereignty and hence a setback for human rights.
In ICL, however, there is an additional variable in play: the instruments circumscribe not only state freedom of action, but also individual autonomy. Thus limitations appearing in an instrument defining ICL liability might of course be attributable to preservation of sovereignty, but they also might reflect conformity to a fundamental principle. Nonetheless, in ICL it is common to overlook this important shift and to make the same assumption as in human rights discourse: to ascribe the limitation to ‘sovereignty’ or ‘compromise’, the usual business of short-sighted states failing to reflect the full potential of human rights because they cling to outdated notions.Footnote 190 Such reasoning fosters a knee-jerk rejection of provisions that may have complied with the principle of legality or culpability, thereby introducing another systemic bias against fundamental principles. Conversely, such reasoning fosters an uncritical reception of expansive interpretations as a victory of humanitarianism over sovereignty, without inquiry into fundamental principles.Footnote 191
Scholars such as Robert Cryer have convincingly demonstrated that states tend to take a wider view of definitions of crimes and principles when they are imposing them on others than when their own officials and nationals may be scrutinized.Footnote 192 While such double standards certainly warrant reproach, there still remains an unanswered (and generally unposed) question of whether the broader or narrower version of the doctrine is the more appropriate one. Confronted with such discrepancies, ICL discourse routinely adopts the same assumption as would human rights liberalism, namely that the broader norm must be the truer, better articulation, and that the narrower must be a retrenchment caused by compromise and self-interest.Footnote 193
Given the additional variable of individual autonomy, however, the reality is rather more subtle. Many conservative aspects of codification efforts may doubtless be traced to unprincipled self-interest. However, it is also possible that a deliberative codification process involving diverse participants may identify legitimate issues of principle. Self-interest may even play a productive role: participants have more incentive to engage in thoughtful examination of principles than if they were applying rules only to others. Potential exposure is likely to sharpen one's sensitivity to fairness. Conversely, judges in tribunals articulating rules to apply to ‘the other’ – that is, vanquished foes or marginalized groups – do not have the same direct incentive to scrutinize compliance with fundamental principles (and, indeed, they may have strong incentives to articulate expansive norms).Footnote 194
The ideological assumptions imported from human rights discourse obscure these dynamics, and can blind ICL participants to the origin of some doctrines in punitive victors' justice.Footnote 195 Ironically, then, ICL practitioners can embrace the most illiberal doctrines as the most ‘progressive’ and reflexively reject more modern codifications as a betrayal of the Nuremberg standard,Footnote 196 without considering compliance with fundamental principles.
The operation of this assumption may be illustrated with two examples, one from literature and one from jurisprudence.
As a first example, the Nuremberg and Tokyo Charters and the ICTY and ICTR Statutes, which were unidirectionally applied, included broad, open-ended definitions of crimes, with formulae such as ‘shall include but not be limited to’, leaving extensive room for judges to expand the definitions of crimes.Footnote 197 The creators of the Rome Statute, who could not know whether the provisions would be applied to their foes, to strangers, to friends, or to themselves, opted for a complete codification with a closed list of defined crimes. As a matter of fundamental principle, codification is generally welcomed as valuable or even essential in a modern liberal system of criminal justice in order to provide fair warning to individuals.Footnote 198 However, Alain Pellet, a highly respected scholar, and David Hunt, a highly respected judge, reacted with grave concern that states have ‘sought to codify the law’ to be applied by the judges, which in their view demonstrates a ‘deep suspicion’ and ‘unfortunate mistrust for the judges’.Footnote 199 Their analyses approach the issue entirely in terms of the benefits of expansive norms, a benefit which is perceived to be frustrated by governmental myopia and compromise. Hunt's concern is that codification will ‘preclude significantly the necessary judicial development of the law’, and he attributes such provisions to ‘fear’, ‘political compromise’, and ‘diplomatic expediency’.Footnote 200 Pellet assumes that the cause was ‘ceding to American pressure’ and ‘not trusting the judges’ and the result is that ‘the authors of the State have limited the chances of making the Court an efficient instrument in the struggle against the crimes it is supposed to repress’.Footnote 201 These illustrations show how even the leading minds in the field can assume that the broadest approach is automatically the best, and can attribute a narrower approach to myopic sovereignty-driven concerns, a heuristic which bypasses reflection on fundamental principles. Such analysis neglects the fact that open-ended criminal norms originated in victors' justice and fails to ask whether that is a pattern we wish to replicate – is an illustrative, open-ended list of crimes really the benchmark to which ICL should aspire? In this way, liberal (human-rightist, internationalist) assumptions can promote illiberal (criminal law) doctrines and condemn liberal (criminal law) doctrines.
To take an example from jurisprudence, we may return to our recurring theme of command responsibility. In the Nuremberg and Tokyo trials, where rules were being applied to vanquished foes, command responsibility was loosely defined and did not expressly require a causal contribution.Footnote 202 When a US court applied command responsibility to its own forces in Vietnam, the doctrine was tightly defined, requiring inter alia a causal contribution for personal liability.Footnote 203 In the negotiation of Additional Protocol I, states were in a position more akin to a ‘veil of ignorance’, as they did not know whether they would be the beneficiaries, relying on AP I provision for protection, or the accused; in this more neutral situation they required causal contribution.Footnote 204 The drafters of the ICTY and ICTR Statutes, applying rules unidirectionally to others, insouciantly effaced the requirement of causal contribution by blending the criminal and non-criminal provisions of AP I.Footnote 205 The drafters of the ICC Statute, once again in a position somewhat akin to the ‘veil of ignorance’, reinstated causal contribution.Footnote 206
Again the double standard is objectionable, but it leaves two possibilities: (i) causation is not required by fundamental principles, and thus the narrower provisions reflect self-interested self-protection by states, or (ii) causation is required by principles, and thus it is the broader provisions that are problematic. As was argued above, fundamental principles appear to require that a person causally contributed to crimes if he or she is to be convicted for those crimes.Footnote 207
Nonetheless, the assumptions about progress and sovereignty foster a simplistic heuristic whereby ICL participants perceive only the first possibility and reflexively condemn the narrower provision without analysis. For example, Judge Hunt, in a partially dissenting opinion, was confronted with the Rome Statute's requirement of causal contribution for command responsibility and the fact that tribunal jurisprudence accords the Rome Statute ‘significant legal value’.Footnote 208 The reinstatement of causal contribution in the Rome Statute could have offered a clue that fundamental principles are at stake, and an opportunity to discover contradictions between tribunal jurisprudence and fundamental principles. Instead, Judge Hunt simply observed that the Statute provision was the result of ‘negotiation and compromise’, as was ‘patent . . . from the vast differences between . . . those provisions and existing instruments such as the Statutes of the ad hoc Tribunals’, and hence he dismissed the Statute provision as ‘of very limited value’.Footnote 209 Among the suppressed premises in this argument are that ‘negotiation and compromise’ invalidate an outcome, and that departure from the broader (selective justice) instrument shows that the narrower (consensus) instrument is incorrect – an incorrectness attributable, of course, to the negotiation and compromise. Missing from this analysis is any reflection on fundamental principles, and whether a deliberative process as opposed to a unilateral imposition might have fostered sensitivity to principles. Similarly, several scholars have not hesitated to dismiss the causal requirement in the Rome Statute provision as a tragic concession to self-interest and, satisfied with this as a complete explanation, have failed to contemplate the possibility of a principled basis.Footnote 210
Thus assumptions about progress and sovereignty can lead ICL participants to look with suspicion upon the processes most likely to generate liberal doctrines and reflexively to favour broad but illiberal doctrines born in selective justice, without considering the fundamental principles that truly define what is progressive or regressive. Moreover, these easy conclusions lead ICL participants to overlook clues and to miss opportunities to detect contradictions between ICL jurisprudence and fundamental principles.
5. Conclusion
While ICL proclaims its exemplary adherence to fundamental principles of a liberal criminal justice system, serious departures occur in all areas, including definitions of crimes, principles of liability, and defences. This article points to an ‘identity crisis’ as part of the explanation: while ICL is a criminal law discipline, its norms and its practitioners draw a deep intellectual inheritance from human rights and humanitarian law. Interpretive, substantive, structural, and ideological assumptions of human rights and humanitarian law have been absorbed into ICL discourse, distorting methods of reasoning and undermining compliance with fundamental principles. Examples of such assumptions and modes of reasoning permeate both the jurisprudence and the literature.
First, human rights and humanitarian interpretive techniques are replicated in ICL, fostering broad, victim-focused, dynamic interpretations, a tendency which not only conflicts with the principle of legality but also encourages exuberant interpretations that contravene culpability and fair labelling.
Second, ICL judges, practitioners, and scholars also conflate substantive norms of ICL with those of human rights and humanitarian law, by importing the familiar content from those domains with the confidence that they are simply following precedent, while overlooking the fact that those ‘precedents’ come from non-criminal domains with structures that focus on improving systems rather than on individual culpability.
Third, ideological assumptions about sovereignty and progress, drawn from human rights and humanitarian law, can lead ICL practitioners to assume that the broadest norms are the most progressive, and that any narrowing must be due to self-interested sovereignty. The easy conclusions generated by assumptions about sovereignty can forestall important inquiries into compliance with fundamental principles.
The article does not suggest that the ‘identity crisis’ completely accounts for the deviations from fundamental principles.Footnote 211 Nonetheless, the influence of normative assumptions is of particular importance because it is not only a cause of departures but also the method: these assumptions furnish the analytical steps by which departures are effected and provide the legal plausibility that lets such failings go unnoticed. The focus of this article is on the discourse, the structures of argument and reasoning employed in ICL, that are assumed to be sound, legitimate, and even liberal, and yet lead to contradictions with the liberal values declared by the system.
The present inquiry has sought to expose the assumptions and reasoning techniques that lead to contradictions with the fundamental principles that the system itself proclaims. A philosophical evaluation of the merits of compliance with fundamental principles would form part of a larger project. Arguments for compliance could of course be rooted in deontological imperatives of treating persons as responsible agents as opposed to objects,Footnote 212 or in consequentialist considerations of preserving legitimacy or advancing efficacy.Footnote 213 What is important for the present inquiry is that ICL declares itself to be an exemplary system of liberal criminal justice. If ICL is to realize this declared aspiration, it will be necessary to subject its doctrines to searing scrutiny and to shed the inapposite relics, habits, and intellectual inheritance of its ‘parent’ domains of human rights, humanitarian law, and criminal law. Only in this way can ICL work out its own identity as a coherent discipline.
The choice for ICL is either to comply with the fundamental principles or to abandon the claim of compliance. More starkly, the choice is between ‘a system of justice’ and the ‘administrative elimination of wrongdoers by command of those in power’.Footnote 214 However, this seemingly stark and binary choice masks a major intellectual project for ICL. Presumably, ICL will not abandon liberal principles (which are based in respect for human dignity) in order to advance a human rights agenda, as this would create contradictions with its own values. Conversely, however, compliance with principles does not necessarily consign ICL to mimicking the principles exactly as they are known in national law.Footnote 215 This article has emphasized the need to be critical of the human rights/humanitarian law inheritance; however, there is also scope – while retaining fidelity to the fundamental liberal commitment of treating persons as ends and not solely as means – to re-examine the specific articulations of those fundamental principles as found in national law.
The project that emerges is not only to discover contradictions and the pathologies that lead to them, but to develop a more refined account of the fundamental principles appropriate for ICL. Contradictions between doctrines and principles may be resolved by correcting the doctrine, refining the principles, or both. Indeed, the special contexts of ICL may pose philosophical questions not previously considered in traditional liberal criminal theory, and hence help us to unearth new insights into the fundamental principles. Thus the project is to discover not only how liberal criminal theory may illuminate ICL, but how ICL may illuminate liberal criminal theory.
Whatever avenue is chosen in responding to the contradictions in ICL (adaptation of doctrines or principles or both), it will be an advance to have an open and principled debate rather than obscuring the contradictions, or promising x but doing y. It is hoped that the current article will engender critical awareness of (i) the contradictions in ICL, and (ii) the assumptions and distortions common in ICL reasoning, and thereby help to foster a more sophisticated and principled discourse.
