Hostname: page-component-745bb68f8f-b6zl4 Total loading time: 0 Render date: 2025-02-11T09:49:26.624Z Has data issue: false hasContentIssue false
  • English
  • Français

Push the Envelope – Watch It Bend: Removing the Policy Requirement and Extending Crimes against Humanity

Published online by Cambridge University Press:  22 November 2010

MATT HALLING
Rights & Permissions [Opens in a new window]

Abstract

This article argues for amending the Rome Statute to remove the state or organizational policy requirement from the classification of crimes against humanity. After a brief look at the requirement itself, the article presents arguments to show how the policy loophole creates an accountability loophole in international criminal law, and how removing it both resolves inconsistencies in the Rome statute and facilitates prosecutions for international crimes. The article's final section examines and responds to leading arguments for keeping the policy requirement. The article is intended to show how the policy requirement limits international criminal law's scope in unwelcome ways and to challenge the use of state or organizational policy as a dividing line between domestic and international crimes.

Keywords

crimes against humanityInternational Criminal Courtinternational criminal lawpolicy requirementRome Statute
Type
HAGUE INTERNATIONAL TRIBUNALS: International Criminal Court and Tribunals The Policy Element of Crimes against Humanity
Information
Leiden Journal of International Law , Volume 23 , Issue 4 , December 2010 , pp. 827 - 845
Copyright
Copyright © Foundation of the Leiden Journal of International Law 2010

Justice for crimes against humanity must have no limitations.

Simon Wiesenthal Footnote 1

This article will discuss the wisdom in amending the Rome Statute's definition of crimes against humanity to remove the state or organizational policy requirement. Section 1 will explain the state or organizational policy requirement (S/O policy) and will explore the regrettably reasoned case opinions regarding its abolishment, section 2 will examine the original intentions behind the requirement, section 3 will assert several arguments as to why the Rome Statute should be amended to remove S/O policy as a necessary element of crimes against humanity, and section 4 will respond to a series of common arguments made for keeping policy in the definition. The paper concludes that amending the Rome Statute to exclude the S/O policy requirement is a worthwhile and appropriate step towards preventing impunity for perpetrators of crimes against humanity.

In addition to advocating a specific change to the Rome Statute, this article also offers theoretical implications for international criminal law. One way to draw the line between national and international crimes is to argue that the latter emanate from some broader policy that necessitates international intervention. If, as argued throughout, this premise is false and some non-policy crimes deserve international criminal law classification, then what is and is not an international crime begins to unravel. Divining some more appropriate way to conceptualize an ‘international crime’ is beyond the scope of this article; unravelling will have to do for now.

1. Defining the policy requirement of crimes against humanity and why the international criminal tribunals no longer require it

Under Article 7(1) of the Rome Statute, someone commits a crime against humanity when certain actions are committed as part of a widespread or systematic attack directed against any civilian population.Footnote 2 Article 7(1) goes on to list the actions that could give rise to a crime against humanity, including murder, extermination, forcible transfer of population, and persecution.Footnote 3 The widespread or systematic context serves to distinguish inhumane crimes handled exclusively by national jurisdictions from crimes against humanity, which is a category of international crimes.Footnote 4 To commit a crime within the scope of Article 7(1) the perpetrator generally needs to have had the intent and to have knowledge of the wider context.Footnote 5

The S/O policy requirement is an additional contextual requirement added in Article 7(2)(a). The requirement reads: ‘“Attack directed against any civilian population” means a course of conduct involving the multiple commission of acts referred to in paragraph 1 against any civilian population, pursuant to or in furtherance of a State or organizational policy to commit such attack.’Footnote 6 The S/O policy can be inferred and need not be expressly stated under the jurisprudence of the ICC Pre-Trial Chamber (PTC) and the International Criminal Tribunal for the former Yugoslavia (ICTY).Footnote 7 The ICC's official Elements of Crimes reveals that the state or organization has to ‘actively promote or encourage such an attack;’Footnote 8 tolerance or acquiescence of the state is not enough.Footnote 9 The only time when the policy requirement can be met by a failure to take action is when there is a ‘deliberate failure to take action, which is consciously aimed at encouraging such attack’.Footnote 10 The Elements of Crimes makes it clear that the policy requirement is not met if a state is unwilling to prosecute an offender for reasons other than encouraging the attack.

The policy of a ‘state or organization’ has been interpreted by the ICC PTC as being a policy ‘made by groups of persons who govern a specific territory or by any organization with the capability to commit a widespread or systematic attack against any civilian population’.Footnote 11 When Pre-Trial Chamber II applied this test, an S/O policy was deemed to have been met by finding that attacks were ‘following the same pattern’,Footnote 12 and that civilians were ‘regularly threatened’.Footnote 13 The recent landmark decision authorizing an investigation into the situation in Kenya discusses how the key is whether the group ‘has the capability to perform acts which infringe on basic human values’.Footnote 14

These interpretations all imply a fairly intricate level of control by and cohesion within the ‘state or organization’. The Kenya Decision gives a list of non-exhaustive factors as to what an organization looks like,Footnote 15 but it does not discuss the outer limits of what an organization could be. Looking at UN Convention language, the broadest definition of ‘organization’ that the ICC could look to is Article 2 of the UN Convention against Transnational Organized Crime.Footnote 16 An organization there is ‘any structured group of three or more persons, existing for a period of time and acting in concert with the aim of committing one or more serious crimes’, with a ‘structured group’ defined as being not ‘randomly formed . . . and that does not need to have formally defined roles for its members, continuity of its membership or a developed structure’.Footnote 17 It is clear that under both the ICC ‘groups of people’ definition and the Transnational Organized Crime Convention there is no way in which a single person could qualify as an organization.

Prior to the ICC PTC decisions, the S/O policy requirement had been ignored in the recent ad hoc tribunal jurisprudence because, in what has quickly become one of the most famous footnotes in the history of international criminal law, the ICTY Appeals Chamber abolished the requirement in Prosecutor v. Kunarac.Footnote 18 As the ICTY statute did not expressly require S/O policy, the court had the authority to decide that crimes against humanity did not have a policy element on the ground that state practice ‘overwhelmingly supports the contention that no such requirement exists under customary international law [CIL]’.Footnote 19 Despite the shocking brevity of the footnote's analysis, Kunarac has been cited favourably by the International Criminal Tribunal for Rwanda (ICTR)Footnote 20 and in the report of the UN Commission of Inquiry into Darfur.Footnote 21

Ironically, this well-cited analysis from a prestigious court is perhaps the weakest argument for the S/O policy requirement to be amended out of the Rome Statute. Although Guenael Mettraux notably has carried out a more exacting analysis of the authorities than the Kunarac Appeals Chamber and reached the same conclusion as that court,Footnote 22 William Schabas analyses many of the same authorities and concludes that the Kunarac footnote reflects a results-oriented political decision rather than a profound analysis.Footnote 23 Schabas carried out a close reading of the Kunarac footnote and found that the Appeals Chamber had cited Nuremberg precedent out of context, relied on lower court authority when national Supreme Court cases had reached opposing results, and never even mentioned the Rome Statute, a major contrary authority which had been completed four years before the Appeals Chamber's analysis.Footnote 24 Robert Cryer et al.'s textbook also reviewed the relevant precedents and the authors similarly concluded that, despite the Appeals Chamber's words to the contrary, Kunarac does not make clear the approach that S/O policy states should follow in order to comport with CIL.Footnote 25

The end result of surveying the law surrounding the policy requirement is a bizarre debate where the ad hoc tribunals are in broad agreement, on the basis of largely discredited reasoning, on the omission of policy from the CIL definition of crimes against humanity. Rather than delving beyond the scope of this article to make yet another attempt at resolving this issue, it is sufficient and reasonable to say that the conflicting authorities are revealing that there just may not be any customary international law on this point.

The S/O policy requirement's uncertain status in CIL begs the question whether an S/O policy requirement should be part of crimes against humanity. Before examining those arguments, it is worth looking at the history of the ICC's policy requirement to see why non-S/O crimes are almost wholly excluded from the Rome Statute.

2. The history behind the S/O policy requirement and the concerns of jurisdiction over ‘unorganized’ crimes

The origins of the policy requirement stem primarily from concerns over the scope of crimes against humanity. At the Rome Conference, the debate over the scope of the contextual requirement of crimes against humanity split the conference into two groups: those who wanted jurisdiction over either widespread or systematic crimes versus those who wanted the requirement to be for widespread and systematic crimes.Footnote 26 The result is a compromise in which, in exchange for the disjunctive ‘widespread or systematic’, the policy requirement was added to placate this second group.Footnote 27 Although not explicitly described as ‘policy’, the International Law Commission's Draft Code of Crimes similarly demanded that a crime against humanity must be ‘instigated or directed by a Government or by any organization or group’.Footnote 28

The policy requirement was chosen from among a series of contextual elements that have been added to crimes against humanity since its beginnings in order to internationalize sufficiently this category of crimes. ‘Crimes against humanity’ was originally applied to fill a legal gap at the Nuremberg trials; the laws of war crimes or wars of aggression (then called ‘crimes against peace’) did not adequately cover the Holocaust, where Germans were perpetrating atrocities against their own citizens.Footnote 29 This large intrusion into state sovereignty has always led to cautious crimes against humanity definitions that made sure that the crimes at issue were of a type warranting international intervention. Older definitions of crimes against humanity have required that the crimes be committed in the context of armed conflict (called the ‘war nexus’ requirement)Footnote 30 or be committed with discriminatory animus.Footnote 31 The policy requirement was first codified in the Rome Statute, but Schabas has argued that policy had been implicitly used as an element of crimes against humanity as early as Nuremberg.Footnote 32

Why was the policy requirement also not tossed into the dustbin with the war nexus and discriminatory animus requirements? As Schabas puts it, leaving ‘widespread or systematic’ as the only contextual factor for a crime against humanity ‘has the potential to make crimes against humanity applicable to serial killers, mafias, motorcycle gangs, and small terrorist bands’.Footnote 33 The concern is that giving random or unrelated crimes the status of crimes against humanity adds to the mix crimes that should belong solely in national jurisdictions, which goes against the object and purpose of internationalizing the crime.Footnote 34

This policy requirement debate is indicative of a fundamental tension in the ICC itself, where the court is designed to have jurisdiction over the worst international crimesFootnote 35 but is also complementary to national justice systems. The ICC is designed to admit only those cases where the crimes are of sufficient gravityFootnote 36 and, under the complementarity principle, when states have either not prosecuted an offender or have been demonstrably unwilling or unable to prosecute.Footnote 37 Crimes such as murder and rape are serious in any context, but internationalizing them is unnecessary, because national justice systems are expected to take on cases where non-state actors commit crimes in opposition to sovereign values.Footnote 38 The general logic here is that the ICC's intrusion into state sovereignty is only warranted when the state or a large, powerful organization within the state perpetrates or encourages the attacks.

The logic makes sense, but the whole system breaks down if deserving crimes admissible under the complementarity and gravity requirements cannot be prosecuted. The compromised language in Article 7 creates a crimes against humanity loophole, admittedly small but very real, where crimes well within the purpose of the ICC are non-justiciable because of the policy requirement.

3. Why the S/O policy requirement should be removed

3.1. A policy requirement can lead to impunity

The following three hypotheticals are all crimes against humanity under ad hoc tribunal jurisprudence, but, because of the S/O policy requirement outlined in Rome Statute Article 7(2)(a) and the Elements of Crimes, are all non-justiciable in the ICC.

The rogue general. A rogue general in country A gains control over a nuclear weapon using his insider access to military launch codes. He threatens to launch the weapon against his own country unless all Jews leave the state. As intended, tens of thousands of Jews are forcibly displaced from country A and the weapon is never fired. Country A is afraid to prosecute the rogue general because his action was so popular that a prosecution could start a civil war.

The online terrorist. A terrorist mastermind from country A starts a blog on ways to exploit holes in country B's airport security. His motives are that a realized threat of another ‘11 September’-style bombing would lead to country B pulling its troops out of the Middle East. He does not know and does not directly communicate with anyone who reads his blog. On the basis of his advice, hijackers are able to capture several aircraft and destroy some large office buildings in country B. Tens of thousands of lives are lost on the basis of the mastermind's blog advice. Following the attacks the mastermind is located in country A. Wary of their capacity to handle such a complicated and expensive prosecution and fearful that their national would be tortured if extradited to country B, country A refers their situation to the ICC.

The corporate hacker. A computer hacker in a large corporation in country A believes that his country can never truly prosper unless the country's political party B is destroyed. By secretly using corporate infrastructure he is able to hack into party B's government computer networks and capture important personal records of party B members. He then disseminates this stolen information widely on the Internet. As intended, massive identity theft occurs, bankrupting and otherwise disrupting the lives of hundreds of thousands of party B members. Country A knows where the hacker is in its territory and is aware of his crime, but the hacker has also tapped into country A's government computer systems. The hacker is blackmailing government officials with highly sensitive information; country A is unwilling to prosecute even though it strongly disagrees with the hacker.

3.1.1. Why the hypotheticals are not crimes against humanity

The rogue general, online terrorist, and computer hacker hypotheticals are all admissible crimes against humanity in every respect but for the S/O policy requirement. All are widespread or systematic attacks against civilian populations. All are attacks under Article 7(1): the rogue general is a situation of forced transfer of population,Footnote 39 the online terrorist is a case of murder,Footnote 40 and the computer hacker relates to persecution.Footnote 41 None of these crimes is committed pursuant to a state policy, as the rogue general's crime is not being promoted or encouraged by the state. All three crimes are committed by individuals who do not qualify as organizations under the ICC PTC test or the Transnational Organized Crime Convention. Gravity and complementarity are both met, but the Prosecutor cannot bring a charge of a crime against humanity because of the policy requirement.

Although no court has interpreted S/O policy in this way, it is possible that the relevant policy could come into existence as the attack unfolds. Such could be the case in the online terrorist hypothetical, where the plane hijackers are forming organizations with the requisite S/O policies even though the mastermind had no such policy in place when he instigated the crime. However, the Kayeshima trial chamber discusses how systematic attacks must be carried out pursuant to a ‘preconceived policy or plan’.Footnote 42 Further, even if Kayeshima's language was distinguished or interpreted favourably, the other two hypotheticals would remain unresolved. The state enables the rogue general's ethnic cleansing under duress, but is not ‘actively promoting or encouraging’ the attack as required by the Elements of Crimes. The computer hacker's crime could unfold without any organization at all; like poisoning a water supply,Footnote 43 he merely sets up a chain of events and forces beyond his control do the rest.

3.1.2. Why the hypotheticals are not war crimes or genocide

War crimes cannot be committed unless there exists an armed conflict of either an international or non-international character.Footnote 44 None of the hypotheticals are occurring in the context of armed conflict.

Genocide is more relevant to the hypotheticals, as it is possible for a single individual to commit genocide under the ICC's Elements of Crimes. Every act of genocide under Article 6 of the Rome Statute either had to take place in the context of ‘a manifest pattern of similar conduct or was conduct that could itself effect such destruction’.Footnote 45 The phrase ‘manifest pattern of conduct’ is basically inserting a policy requirement.Footnote 46 However, the phrase ‘or conduct that could itself effect such destruction’ is not qualified by any pattern or policy and leaves room for prosecutions against non-organized genocidal criminals. Unlike in Kunarac, the drafters of the Elements of Crimes seem to have made room for the ICTY case rejecting a policy requirement for genocide, Prosecutor v. Jelisić.Footnote 47 The Jelisić trial chamber discussed the theoretical possibility of a lone individual committing genocide,Footnote 48 and the ICC allows for that possibility.

However, genocide is highly difficult to prove and functions as a kind of ‘aggravated crime against humanity’;Footnote 49 other elements of the crime doom all the above hypotheticals. First, genocide has a specific-intent requirement where the criminal must intend to destroy the targeted group ‘in whole or in part’.Footnote 50 The rogue general lacks specific intent because he intends to ethnically cleanse his country rather than destroy the group; the notion that ethnic cleansing is not genocide has been accepted by both the ICCFootnote 51 and the International Court of Justice.Footnote 52 The online terrorist's wish to stop military action in the Middle East also fails on specific intent. Second, genocide only exists when the specific intent is to destroy four specific groups: ethnic, national, racial, or religious.Footnote 53 An intention to destroy a political or social group, such as a political party in the computer hacker hypothetical, is insufficient for genocide.

3.1.3. So what?

The hypotheticals are constructed to show how a very serious crime can evade ICC jurisdiction solely on grounds of the policy requirement of a crime against humanity. What is needed to expose the S/O policy loophole is (i) a crime against humanity of sufficient gravity in every respect except S/O policy; (ii) a state that is unwilling or unable to prosecute; and (iii) a crime that lacks the demanding intent requirements needed to meet the threshold for genocide. Every crime meeting these three criteria can be committed at will and the ICC is powerless to stop them.

But how often will this happen? Schabas described the notion of a lone génocidaire as an hypothèse d'école Footnote 54 – a kind of scholarly constructed crime that is so remotely possible in real life that it can safely be ignored. One could argue that the above hypotheticals are comparably remote, but there are several problems with casually dismissing them. First, it would be one of the darkest days in the history of international criminal justice for even a single crime of such magnitude to be committed with impunity. Second, modern trends are moving in directions that make a non-organized crime against humanity more and more possible: weapons get smaller and more powerful, organizations become less hierarchical and more nebulous. Third, the longer the loophole exists the more opportunity there is for clever criminals and their lawyers to disguise intentionsFootnote 55 and craft crimes exploiting the loophole. Fourth, it is completely unacceptable for there to be any loopholes when it comes to crimes so massive that they warrant ICC intervention; jurisdiction for these kinds of crime should aim to be ‘watertight and inescapable’.Footnote 56 Fifth, the International Criminal Court exists precisely to prosecute unfathomable crimes. Nobody could ever imagine anything like the Holocaust before it happened, but it still happened. Waiting until such a crime is committed before amending Article 7 would be to wait too long.

3.2. Removing the policy requirement resolves inconsistencies in the Rome Statute

The policy requirement makes Article 7 inconsistent with itself and creates disparate treatment of policy across the range of crimes in the Rome Statute.

First, S/O policy as qualified by the Elements of Crimes has nullified the disjunctive compromise reached in Rome, because widespread attacks also have to be sufficiently systematic in order to satisfy the policy requirement. Influential commentators such as Kevin Jon Heller have argued that the disjunction has little meaning in practice because proving policy means having to show systematic aspects to widespread attacks.Footnote 57 Kai Ambos and Steffen Wirth have also argued that Article 7, when combined with the language in the Elements of Crimes, amounts to ‘deleting the “widespread” alternative from the Statute’.Footnote 58 As a counterpoint, Darryl Robinson emphasizes that the original idea was for S/O policy to require a lower evidentiary threshold and not just be a synonym for ‘systematic’.Footnote 59 Robinson's explanation has the advantage of giving separate meanings to ‘systematic’ and ‘policy’, but distinguishing the terms only according to the extent of the evidence needing to be shown begs the question how the ICC could actually make such a distinction on a given set of facts.

The ICC PTC jurisprudence on this point has so far rejected Robinson's view entirely, and has confirmed the other scholars’ comments that the policy requirement functionally removes the ‘widespread’ alternative from the crimes against humanity definition. The Katanga Confirmation Decision has interpreted policy as being a high threshold: the policy behind a widespread attack must nevertheless be ‘thoroughly organised and follow a regular pattern’.Footnote 60 The attacks must also follow a ‘common policy involving public or private resources’.Footnote 61 The standards the court is describing for ‘policy’ may look familiar; it is how the ICTR in Akayesu defined a ‘systematic attack’.Footnote 62 The Bemba Confirmation Decision also looked to a ‘system of attack aimed at creating a climate of fear’ when ruling on the S/O policy requirement.Footnote 63 The PTC so far has indeed interpreted ‘policy’ and ‘systematic’ synonymously; the disjunctive language fought for at Rome has lost all meaning in the court's first cases.

Second, it is a nonsensical glitch in the court's jurisdiction that a non-organization can commit genocide, but not a crime against humanity, in the ICC schema. The language used in the aforementioned ‘conduct that could itself effect such destruction’ in the Elements of Crimes Footnote 64 made room for a lone génocidaire, even if the possibility of such a crime is remote. Schabas has stated that Jelisić had just as little authority for its lone génocidaire theory as did Kunarac in the crimes against humanity context;Footnote 65 why did the Rome Statute's drafters accept one analysis and not the other?

The most likely explanation for this inconsistency is that the drafters in Rome had no desire to disrupt well-accepted, previously negotiated language. When drafting the 1948 Genocide Convention, a state policy requirement was considered,Footnote 66 but ultimately did not make the final definition. Among the arguments made six decades ago for excluding such a requirement were that policy ‘unduly restricted the concept of genocide’Footnote 67 and, although government complicity was an ‘element’ of genocide, it need ‘not necessarily be a condition of its existence sine qua non’.Footnote 68 The 1996 ILC draft seems to have concurred with these arguments, or at least saw no need to renegotiate the 1948 definition, when they commented how their definition of ‘the crime of genocide would be equally applicable to any individual who committed one of the prohibited acts with the necessary intent’.Footnote 69

Even though exactly the same arguments made in 1948 and 1996 could apply just as easily to crimes against humanity as to genocide, the logic was not extended and no distinguishing explanation was presented. The ILC explains that isolated criminal conduct would not constitute a crime against humanity because ‘it would be extremely difficult for a single individual acting alone to commit [a crime against humanity]’.Footnote 70 Is it not just as difficult for an individual to commit genocide as to commit crimes against humanity? Why does the ILC not take the same position on policy that it took six pages previously when discussing genocide? Regardless of what one thinks about S/O policy for genocide and crimes against humanity, the position on policy should be the same with respect to both crimes.

The genocide–crimes against humanity policy inconsistency could be resolved in two ways: either the crimes against humanity policy requirement can be removed from Article 7 or S/O policy can be made mandatory for both crimes against humanity and genocide. The former approach has the added advantage of distinguishing widespread attacks from systematic ones, and is therefore the preferred alternative.

3.3. Removing the policy requirement better allows the ICC to prosecute all the worst terrorism crimes

The online terrorist and computer hacker hypotheticals show that many cases that fall into the policy requirement loophole are going to be terrorism crimes. Terrorism was one of the crimes most difficult to omit when the Rome Statute was being drafted; it was not removed from the ILC draft statute until 1995Footnote 71 and, along with drug trafficking, had the most support for inclusion in Rome among all the ‘missing crimes’.Footnote 72 Given 11 September and other serious terrorism crimes committed since the 1998 conference, it is possible that terrorism will eventually be added into the statute as a distinct crime.

However, another way of incorporating terrorism into the Rome Statute is to give a ‘liberal interpretation of crime against humanity’;Footnote 73 abolishing the S/O policy requirement is clearly a liberalizing push in terms of the crime's definition.

Several reasons are advanced why terrorism was not included in the Rome Statute. Patrick Robinson summarizes this debate by concluding that terrorism was felled by (i) definitional concerns; (ii) time constraints on negotiations; (iii) the need to complete the treaty in Rome; (iv) misgivings over classifying terrorism as an ‘international crime’ rather than a ‘treaty crime’; and (v) the fact that states have incentives to prosecute terrorists and prosecutions can happen more efficiently at the national level.Footnote 74 Despite the existence of several available terrorism definitions,Footnote 75 controversy was created over finding a definition that criminalized terrorism while respecting the fundamental right to seek self-determination.Footnote 76

The hypotheticals show that there is a limit to how far states may be willing to go to prosecute serious terrorism crimes, and removing S/O policy as a requirement for crimes against humanity creates an appealing balance between prosecuting terrorism and respecting self-determination. Fighting for self-determination demands protection under international law, but it makes sense to classify actions as crimes against humanity if the group's methods reach such a destructive level that the gravity and widespread or systematic thresholds are reached under the Rome Statute. The policy requirement would have to be removed to ensure that these threshold-crossing crimes are prosecuted, as highly decentralized networks such as that of the online terrorist hypothetical run a significant risk of failing the ICC PTC's ‘groups of persons’ definition of an organization.

3.4. Removing the policy requirement can facilitate complementarity

It is easier for states to prosecute crimes against humanity effectively if they do not have the added burden of proving S/O policy, thus reducing the need for ICC intervention. Complementarity requires that states prosecute crimes as they are spelled out in the Rome Statute; the prosecutions have to be for ‘crimes against humanity’, not the murders, rapes, and so on that underlie the charge of crimes against humanity.Footnote 77 It is overwhelmingly likely that a national prosecution for crimes against humanity with an appropriate sentence will satisfy the ICC Office of the Prosecutor and the Pre-Trial Chamber for complementarity purposes, even if S/O policy is not in the national definition.Footnote 78 Needless to say, states could change their statutes to remove policy as a requirement without having to wait for an ICC amendment; many states have already done so.Footnote 79

4. Counterarguments and responses

4.1. Domestic legal systems are capable of addressing serious non-policy crimes

Schabas insists that ‘concerns that requiring a State policy will leave a so-called impunity gap are misplaced. Most so-called non-State actors find themselves more than adequately challenged by various national justice systems’.Footnote 80 This statement is generally true, but the hypotheticals show how states may not be capable of addressing these crimes. The rogue general's state is afraid of the violence to which a prosecution would lead, while the computer hacker's government is paralysed by a potential scandal. As regards the online terrorist case, the ICC Appeals Chamber has held that a self-referral satisfies complementarity in situations where no criminal investigations were ongoing.Footnote 81 The fact that national justice systems can generally handle these cases ignores the exceptional cases where they cannot, and the ICC is designed to handle cases of an exceptional character.

4.2. Removing the policy requirement infringes state sovereignty

Looking at the hypotheticals one can see how the kinds of situation where policy holds up prosecution are those where ICC members would actually be amenable to giving up some sovereignty. The ICC has always been a lightning rod for criticism, especially in the United States, because it both threatens state sovereignty and is an ‘unprecedented attempt to check the power of the UN Security Council’.Footnote 82 Some countries are not and may never be ICC members, regardless of whether S/O policy is contained in the definition of crimes against humanity; the better question is whether current parties to the statute who already accept Article 7 would object to this extra intrusion into their sovereignty.

If a state is willing to cede some sovereign power to the ICC, removing the S/O policy requirement should not change their position. The only case where an admissible crime against humanity fails only because of the policy requirement is when states are not encouraging the crime against humanity but are otherwise unwilling or unable to prosecute. If they were helping the perpetrator or deliberately not prosecuting in order to encourage the attack somehow, then there would be a state policy! Given complementarity, gravity, and the widespread or systematic requirements, the only cases of crimes against humanity that currently turn on S/O policy are when states do not undertake prosecution against their will; it is unlikely there would be much resistance to giving the ICC authority over these cases.

4.3. Removing the policy requirement gives the ICC authority to prosecute decidedly non-international crimes

This is the ‘serial killer argument’ made by Schabas and given above.Footnote 83 The fear is that ‘widespread’ or ‘systematic’ is insufficient as a contextual element by itself because groups such as serial killers or events that are widespread, but random, such as crime waves, could trigger ICC jurisdiction if the S/O policy requirement were removed.

This argument ignores the fact that these kinds of crime would still be inadmissible for other reasons. It is overwhelmingly unlikely that a serial killer can ever last long enough to commit a ‘widespread’ attack on a civilian population, given that courts have interpreted ‘widespread’ to mean ‘massive, frequent, large-scale action’Footnote 84 or, if a single act, one of ‘extraordinary magnitude’.Footnote 85 Unlike in the case of the hypotheticals, individual states have plenty of incentives to prosecute serial killers. Even if a widespread attack admissible under complementarity could be committed, it is difficult to conceive how a serial killer meets the gravity threshold intended by the Rome Statute. The PTC has recently endorsed a factor test for gravity where ‘the issues of the nature, manner and impact of the [alleged] attack are critical’ and a gravity analysis ‘should not be assessed only [from] a quantitative perspective, i.e. by considering the number of victims; rather, the qualitative dimension of the crime should also be taken into consideration’.Footnote 86 Although this test may be subject to a future appeal, it is the only one to date that has not been overruled by the Appeals Chamber.Footnote 87 It is very hard to imagine that a lone serial killer would be able to satisfy the PTC's factor test when his acts are considered alongside the impact, scale, and ‘qualitative dimension’ of other crimes worthy of prosecution at the ICC.

Policy is a redundant check for keeping these uncontroversially non-international crimes out of the ICC.Footnote 88 Simon Chesterman describes how the policy requirement essentially reiterates a point on which everyone can hopefully agree: that isolated and random acts cannot amount to crimes against humanity.Footnote 89 Admissibility checks and prosecutorial discretion preserve this intention even without an S/O policy requirement. These crimes are still excluded, and serial killers are not going to The Hague in chains anytime soon.

4.4. Crimes falling into the policy loophole can be handled by a transnational criminal law approach

Another point worth mentioning is that the S/O policy loophole could potentially be closed by a transnational criminal law approach – that is, impunity could be resolved without ICC involvement by extraditing the perpetrator to a country that would be willing to prosecute. A transnational approach could thus resolve the impunity gap created by the loophole while maintaining the current definition of crimes against humanity.

However, a transnational approach applied to the above hypotheticals at best reaches the same result as removing the S/O policy requirement. First, all three governments in the hypotheticals know where the perpetrators are within their countries and are unwilling to prosecute them. These crimes are different from crimes such as drug-trafficking that lend themselves to co-ordinated transnational solutions. Second, a transnational approach assumes that another state is willing and able to get involved in prosecuting these cases. This may not reflect reality; it is a lot to ask of a state with no vested interest in a case to that it take it on despite the complexity, expense, and attendant security risks of such cases. In contrast, the ICC's sole function is to seek and prosecute persons ‘for the most serious crimes of international concern’.Footnote 90 Third, arguing for a transnational approach in these matters is consistent with amending the Rome Statute. A transnational approach could indeed address these loophole crimes, leaving the ICC with an inadmissible case because of complementarity. There is nothing wrong with this result, but if this complete transnational harmony existed in real life then there would be no need to create the ICC, because it would never have an admissible case. The problem with the loophole crimes comes when no one else wants to prosecute them, and international criminal law should be designed to confront such a possibility.

4.5. Crimes falling into the policy loophole can be handled by creating new ad hoc tribunals

Another possible solution to the loophole is for the UN Security Council to create an ad hoc tribunal whenever this problem arises. The UN has done this several times in recent memory, but all the tribunals were necessary because the ICC lacked jurisdiction over the relevant crimes. The ICTY and ICTR both pre-date the ICC's creation, and the Special Court for Sierra Leone,Footnote 91 although created after the Rome Statute, was necessary because the relevant crimes were committed before the ICC's entry into force in 2002.Footnote 92

Assuming that the ICC continues to exist and functions well, creating new international ad hoc tribunals with jurisdiction over ICC crimes can too easily be harm disguised as good. One major reason for having the ICC is that it serves as a streamlining force for international criminal law; it is a place, as Chesterman once wrote, to resolve issues across courts by ‘providing a firmer basis for a coherent jurisprudence’.Footnote 93 Making extra ad hoc tribunals just to cover S/O policy loopholes in crimes against humanity diminishes this streamlining effect. Further, creating ad hoc tribunals to close loopholes in the Rome Statute is inefficient; for every new tribunal new venues have to be found, new statutes negotiated, new codes of procedure drafted, and so on.

The most important reason for wanting ICC jurisdiction over every admissible international crime is the problem of what I call tribunal shielding. If the court's jurisdiction is properly triggered over a particular territory,Footnote 94 the ICC has authority to prosecute anyone from the relevant starting point going forward, regardless of the nationality of the perpetrator.Footnote 95 All sides are fair game for prosecution under the Rome Statute, and referrals cannot slice out jurisdiction over particular parties, despite the best efforts of the Security Council to do so with the 2005 Darfur referral.Footnote 96 In contrast, when the Security Council creates an ad hoc tribunal it can carve up jurisdiction in ways that, depending on the wishes of the council and of its five permanent members in particular, can create selective justice. Tribunal shielding is when the UN creates an ad hoc tribunal for mixed motives, arguing legitimately that no proper domestic venue exists to hear the cases at issue while disguising a second intention: to craft justice in such a way that Security Council members and their allies are kept out of the courtroom.

The Special Tribunal for Lebanon (STL) may well be the first example of tribunal shielding. It was created in 2007 for the prosecution of those who in 2005 killed the Lebanese prime minister, Rafiq Hariri, and 21 others. The STL has jurisdiction only over the perpetrators of the bombing which killed the prime minister and over other connected attacksFootnote 97 that occurred between 1 October 2004 and 12 December 2005.Footnote 98 Notably, the period covered by Security Council Resolution 1757 excludes the entire period in 2006 when Israel waged war in Lebanon.

It is worth considering why an ICC referral does not seem to have been considered, despite the fact that the Secretary-General's report on establishing the STL considered and briefly presented a prima facie case that the bombing and connected attacks rose to the level of a crime against humanity.Footnote 99 A debate in the Security Council, referenced although unpublished, resulted in the exclusion of crimes against humanity from STL jurisdiction.Footnote 100 If crimes against humanity are present on these facts, then the absence of an ICC referral or at least an international tribunal with broader jurisdictional grant is suspicious. Needless to say, the United States would almost certainly use its Security Council veto over any tribunal that could lead to Israeli leaders facing criminal prosecutions for war crimes or crimes against humanity.

It is not yet known if the category of crimes against humanity was not added to the STL statute because the Security Council found crimes against humanity to be irrelevant on these facts, because the ICC's widespread or systematic element was not met, because the policy requirement was not met, or simply conveniently to avoid the question of why the jurisdictional grant was sliced so thinly. Perhaps time will tell just what happened at the STL negotiations, but the relevant point here is that a narrow crimes against humanity definition makes it easier for the Security Council to dismiss the ICC option and to engage in tribunal shielding.

The hypotheticals are crimes of enormous magnitude and the ICC was painstakingly negotiated to handle just these kinds of case. Every exception made of this nature, which means that the court cannot access it, diminishes the institution, creates opportunities for selective justice, and opens up a Pandora's box that comes with new ad hoc tribunals having overlapping jurisdiction with the ICC.

4.6. Removing the policy requirement cheapens crimes against humanity

Ronald Slye and Beth Van Schaack call this a ‘semantic inflation’ argument, where ordinary crimes are given the label and corresponding stigma of international crimes, blurring the distinction to the point where the term ‘international crime’ loses its content.Footnote 101 Even if they are never going to be prosecuted at the ICC, the concern is that if groups like Hell's Angels can commit crimes that satisfy the elements of crimes against humanity, then that cheapens a category of crimes originally designed to prosecute Nazis at Nuremberg for the Holocaust. Putting aside the fact that the S/O policy requirement may not actually succeed in excluding motorcycle gang crimes from crimes against humanity (such crimes can be interpreted as being in furtherance of a gang's organizational policy), this is perhaps the most worthy argument for keeping a policy requirement.

Two points need to be made. First, the category of crimes against humanity has become progressively broader over the years and, despite its Holocaust beginnings, customary international law and Article 7 now give this label to a much larger variety of behaviour. As time passes ‘crimes against humanity’ have had to broaden in scope to encompass new behaviour, and requirements once deeply embedded, like the war nexus, have been abandoned.Footnote 102 The original necessity for applying ‘crimes against humanity’ after the Second World War was that it would serve as a kind of ‘gap filler’ for international criminal law.Footnote 103 In fact, there was so little historical precedent at Nuremberg for charging crimes against humanity that the war nexus was created to ‘strengthen its legality by connecting crimes against humanity to the more established notion of war crimes’.Footnote 104 When the category of crimes against humanity is seen in this historical context, as a backstop to prevent impunity, then removing the S/O policy requirement seems less like cheapening these crimes and more like the next evolutionary, gap-filling step taken by ‘crimes against humanity’.

Second, it is unfair to dwell on the expressive disadvantages of removing the policy requirement while not recognizing that expressive advantages exist as well. The hypotheticals are all well-deserving of the status and stigma associated with crimes against humanity; all three examples involve tens of thousands of lives being ruined or destroyed. Focus in particular on the online terrorist hypothetical, which is essentially a copycat of the 9/11 terrorist attacks on the United States. Is it semantically satisfying to say that a repeat performance of a decade-shattering terrorist attack is not a crime against humanity? Is that what we want to say?

Fears over semantic inflation are certainly defensible, and no one can deny that ‘crimes against humanity’ has been hyperbolically attached to a lot of conduct that clearly does not rise to the level deserving the label ‘international crime’. However, these fears should not prevent conduct deserving of the label from being designated as such. More importantly, ‘crimes against humanity’ should not be defined so narrowly that even one crime that looks like the hypotheticals can be committed with impunity.

5. Conclusion

Despite all the criticism levelled at the assessment in Kunarac of the policy requirement, the court has it right for the wrong reasons. The unlikelihood that a crime falls into the policy loophole is more than offset by the backlash and legal chaos that is risked if even a single crime were to find it. Removing the policy requirement from crimes against humanity allows for every individual perpetrator of admissible crimes to be tried at the ICC, be it for crimes against humanity, genocide, and/or war crimes. There is a satisfying symmetry in that result.

Perhaps the likeliest result of this debate is that, rather than overriding the strong academic forces that wish to retain S/O policy in Article 7, the ICC judges will stretch the definition of ‘organization’ into oblivion should a loophole crime ever come up. This result solves the impunity issue, but the result is similar to changing Article 7: the policy requirement becomes abrogated by interpretation rather than by amendment. The difference between either form of abrogation is small, but so long as the policy requirement exists there is always the risk that the judges will choose to give it legal content. Abrogation by interpretation is also problematic because Article 121 of the Rome Statute envisions amendments being exclusively within the purview of the Assembly of States Parties.Footnote 105

The international community should not have to desire that a rule be ignored so that justice is done. Article 7 should be amended consistent with ICC procedure under Article 121Footnote 106 to remove the policy requirement. The remaining definition would be that a crime against humanity exists when anyone commits the statutorily specified acts as part of a ‘widespread or systematic attack directed against any civilian population, with knowledge of the attack’.

References

1 Simon Wiesenthal Center, ‘Quotes’, available at www.wiesenthal.com/site/pp.asp?c=lsKWLbPJLnF&b=4441355.

2 1998 Rome Statute of the International Criminal Court, 37 ILM 999 (hereinafter Rome Statute), Art. 7(1).

3 Ibid.

4 R. Cryer et al., An Introduction to International Criminal Law and Procedure (2007), 187.

5 Rome Statute, supra note 2, at Arts. 7(1), 30. When the enumerated acts in Art. 7 indicate no mens rea, then, per Art. 30, these crimes have the default mens rea of needing to be committed with intent and knowledge.

6 Ibid., Art. 7(2)(a) (emphasis added).

7 Prosecutor v. Katanga and Chui, Decision on the Confirmation of Charges, Case No. ICC-01/04–01/07–717, PTC I, 30 September 2008 (hereinafter Katanga Confirmation Decision), para. 396; see also Prosecutor v. Blaškić, Judgement, Case No. IT-95–14-T, T.Ch. 3 March 2000, at para. 204 (ICTY authority on this point).

8 Finalized Draft Text of the Elements of Crimes, UN Doc. PCNICC/2000/1/Add.2 (2000) (hereinafter Elements of Crimes), available at www.icccpi.int/library/about/officialjournal/Element_of_Crimes_English.pdf, Art. 7.

9 Ibid., Art. 7, n. 6. See also Ambos, K. and Wirth, S., ‘The Current Law of Crimes against Humanity – An Analysis of UNATAET Regulation 15/2000’, (2002) 13 Criminal Law Forum 1CrossRefGoogle Scholar, at 31–3 (the Elements of Crimes language precludes toleration from fulfilling policy).

10 Elements of Crimes, supra note 8, at Art. 7.

11 Prosecutor v. Bemba, Decision Pursuant to Article 61(7)(a) and (b) of the Rome Statute on the Charges of the Prosecutor against Jean-Pierre Bemba Gombo, Case No. ICC-01/05–01/08–424, PTC II, 15 June 2009 (hereinafter Bemba Confirmation Decision), at para. 88; Katanga Confirmation Decision, supra note 7, at para. 396.

12 Bemba Confirmation Decision, supra note 11, at para. 115.

13 Ibid.

14 Situation in the Republic of Kenya, Decision Pursuant to Article 15 of the Rome Statute on the Authorization of an Investigation into the Situation in the Republic of Kenya, Case No. ICC-01/09–19-Corr, PTC II, 31 March 2010 (hereinafter Kenya Decision).

15 The factors are (i) whether the group is under a responsible command, or has an established hierarchy; (ii) whether the group possesses, in fact, the means to carry out a widespread or systematic attack against a civilian population; (iii) whether the group exercises control over part of the territory of a state; (iv) whether the group has criminal activities against the civilian population as a primary purpose; (v) whether the group articulates, explicitly or implicitly, an intention to attack a civilian population; and (vi) whether the group is part of a larger group, which fulfils some or all of the abovementioned criteria. Ibid., para. 93. But see ibid., para. 51 (Judge Kaul, dissenting) (arguing more narrowly that an organization should partake of some of the characteristics of a state).

16 2000 United Nations Convention against Transnational Organized Crime, UN Doc. A/Res/55/25/Annex (hereinafter Transnational Organized Crime Convention’), available at www.uncjin.org/Documents/Conventions/dcatoc/final_documents_2/convention_eng.pdf.

17 Ibid.

18 Prosecutor v. Kunarac, Judgement, Case No. IT-96–23, A.Ch., 12 June 2002, para. 98, n.114. The footnote reads,

There has been some debate in the jurisprudence of this Tribunal as to whether a policy or plan constitutes an element of the definition of crimes against humanity. The practice reviewed by the Appeals Chamber overwhelmingly supports the contention that no such requirement exists under customary international law. See, for instance, Article 6(c) of the Nuremberg Charter; Nuremberg Judgement, Trial of the Major War Criminals before the International Military Tribunal, Nüremberg, 14 November 1945 – 1 October 1945, in particular, pp 84, 254, 304 (Streicher) and 318–319 (von Schirach); Article II(1)(c) of Control Council Law No 10; In re Ahlbrecht, ILR 16/1949, 396; Ivan Timofeyevich Polyukhovich v The Commonwealth of Australia and Anor, (1991) 172 CLR 501; Case FC 91/026; Attorney-General v Adolph Eichmann, District Court of Jerusalem, Criminal Case No. 40/61; Mugesera et al. v Minister of Citizenship and Immigration, IMM-5946–98, 10 May 2001, Federal Court of Canada, Trial Division; In re Trajkovic, District Court of Gjilan (Kosovo, Federal Republic of Yugoslavia), P Nr 68/2000, 6 March 2001; Moreno v Canada (Minister of Employment and Immigration), Federal Court of Canada, Court of Appeal, ?1994g 1 F.C. 298, 14 September 1993; Sivakumar v Canada (Minister of Employment and Immigration), Federal Court of Canada, Court of Appeal, ?1994g 1 F.C. 433, 4 November 1993. See also Report of the Secretary-General Pursuant to Paragraph 2 of Security Council Resolution 808 (1993), S/25704, 3 May 1993, paras 47–48; Yearbook of the International Law Commission (ILC), 1954, vol. II, 150; Report of the ILC on the work of its 43r session, 29 April – 19 July 1991, Supplement No 10 (UN Doc No A/46/10), 265–266; its 46th session, 2 May – 22 July 1994, Supplement No 10 (UN Doc No A/49/10), 75–76; its 47th session, 2 May – 21 July 1995, 47, 49 and 50; its 48th session, 6 May – 26 July 1996, Supplement No 10 (UN Doc No A/51/10), 93 and 95–96. The Appeals Chamber reached the same conclusion in relation to the crime of genocide (Jelisić Appeal Judgement, para 48). Some of the decisions which suggest that a plan or policy is required in law went, in that respect, clearly beyond the text of the statute to be applied (see e.g., Public Prosecutor v Menten, Supreme Court of the Netherlands, 13 January 1981, reprinted in 75 ILR 331, 362–363). Other references to a plan or policy which have sometimes been used to support this additional requirement in fact merely highlight the factual circumstances of the case at hand, rather than impose an independent constitutive element (see, e.g., Supreme Court of the British Zone, OGH br. Z., vol. I, 19). Finally, another decision, which has often been quoted in support of the plan or policy requirement, has been shown not to constitute an authoritative statement of customary international law (see In re Altstötter, ILR 14/1947, 278 and 284 and comment thereupon in Ivan Timofeyevich Polyukhovich v The Commonwealth of Australia and Anor, (1991) 172 CLR 501, pp 586–587).

19 Prosecutor v. Kunarac, supra note 18.

20 Prosecutor v. Semanza, Judgement, Case No. ICTR-97–20-A, A.Ch., 20 May 2005, at para. 269.

21 Report of the International Commission of Inquiry on Violations of International Humanitarian Law and Human Rights Law in Darfur, UN Doc. S/2005/60 (2005), para. 179.

22 G. Mettraux, ‘Crimes against Humanity in the Jurisprudence of the International Criminal Tribunals for the Former Yugoslavia and for Rwanda’, (2002) 43 Harvard International Law Journal 237, at 271–83.

23 W. Schabas, ‘State Policy as an Element of International Crimes’, (2008) 98 Journal of Criminal Law & Criminology 953.

24 Ibid. at 953, 960–3.

25 Cryer et al., supra note 4, at 197.

26 D. Robinson, ‘Defining “Crimes against Humanity” at the Rome Conference’, (1999) 93 AJIL 43, at 47.

27 Ibid.

28 Report of the International Law Commission on the Work of Its Forty-Eighth Session, UN Doc. A/51/10 (1996) (hereinafter ILC 1996 Report), at 93.

29 Van Schaack, B., ‘The Definition of Crimes against Humanity: Resolving the Incoherence’, (1999) 37 Columbia Journal of Transnational Law 787, at 791Google Scholar. Genocide was not conceived of as an international crime until the 1948 Genocide Convention, which was two years after the Nuremberg International Military Tribunal.

30 Cryer et al., supra note 4, at 191–2. For more on the end of the war nexus requirement, see generally Van Schaack, supra note 29.

31 Cryer et al., supra note 4, at 192.

32 Schabas, supra note 23, at 960–2 (looking at Nuremberg precedents when analysing Kunarac's customary international law argument).

33 Ibid., at 960.

34 Ibid., at 982 (‘Indeed, the existence of a State policy may be the best criterion in distinguishing between individual crimes that belong to national justice systems, and international crimes’).

35 Rome Statute, supra note 2, preamble.

36 Ibid., Art. 17(1)(d).

37 Ibid., Art. 17. Art. 17 allows for the court to assess the gravity and complementarity grounds on their own initiative, but they may not rule over a third admissibility check constraining the Office of the Prosecutor: the Prosecutor additionally may decline jurisdiction over a case when doing so ‘is in the best interests of justice.’ Ibid., Art. 53.

38 R. Slye and B. Van Schaack, ‘The Internationalization of Crimes’, in Slye and Van Schaack (eds.), International Criminal Law: The Essentials (2008), 111.

39 Rome Statute, supra note 2, Art. 7(1)(d). This conduct may also fall within the persecution provision.

40 Ibid., Art. 7(1)(a).

41 Ibid., Art. 7(1)(b). This conduct may also fall within the extermination or other inhumane acts provisions.

42 Prosecutor v. Kayeshima, Judgement, Case No. ICTR-96–3-T, T.Ch., 12 May 1999, para. 123 (emphasis added).

43 Shortly after the Second World War, Jewish poet Abba Kovner led a plot with the aim of poisoning various urban water supplies in an effort to kill six million Germans, which would equal the number of Jews killed in the Holocaust. The plan was found out before most of the poison had been released, but had it succeeded it would have been both genocide and a crime against humanity. See generally T. Segev, The Seventh Million: The Israelis and the Holocaust (2000), 140–53.

44 Rome Statute, supra note 2, at Art. 8.

45 Elements of Crimes, supra note 8, at Art. 6.

46 Schabas, supra note 23, at 967 (the difference between manifest pattern and policy ‘would appear to be largely semantic’).

47 Prosecutor v. Jelisić, Judgement, Case No. IT-95–10-A, A.Ch., 5 July 2001, para. 48.

48 Prosecutor v. Jelisić, Judgement, Case No. IT-95–10-T, T. Ch., 14 December 1999, para. 100. Although they do not use the words ‘lone individual committing genocide’, the fact that the Jelisić Appeals Chamber clearly rejects a policy requirement for genocide is confirmation that they also agree that lone génocidaires are possible.

49 W. Schabas, ‘The Genocide Convention at 60’, Crimes of War Project, 6 February 2009, available at www.crimesofwar.org/onnews/news-schabas.html (‘Subject to some insignificant technical quibbles, the crime of genocide constitutes the most aggravated or extreme form of crimes against humanity’).

50 Rome Statute, supra note 2, at Art. 6.

51 Prosecutor v. Al Bashir, Decision on the Prosecution's Application for a Warrant of Arrest against Omar Hassan Ahmad Al Bashir, Case No. ICC-02/05–01/09, PTC I, 4 March 2009 (hereinafter Bashir Arrest Warrant Decision), para. 144.

52 Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia & Herzegovina v. Serbia & Montenegro), Judgement of 26 February 2007, [2007] ICJ Rep. 70, para. 190.

53 Rome Statute, supra note 2, at Art. 6.

54 W. Schabas, ‘Darfur and the “Odious Scourge”: The Commission of Inquiry's Findings on Genocide’, (2005) 18 LJIL 871, at 877.

55 The Al Bashir PTC recognizes that a well-disguised genocidal intent should not be a barrier to prosecution, but it also acknowledges that the rarity of finding direct evidence of intent ‘renders the establishment of the dolus specialis of genocide particularly difficult’. Bashir Arrest Warrant Decision, supra note 51, at para. 27. The PTC dismissed genocide charges against Al Bashir on grounds of specific intent; the decision was reversed on appeal for applying an erroneous legal standard and was remanded for a new determination. See Prosecutor v. Al Bashir, Judgment on the appeal of the Prosecutor against the ‘Decision on the Prosecution's Application for a Warrant of Arrest against Omar Hassan Ahmad Al Bashir’, Case No. ICC-02/05–01/09–73, A.Ch., 3 February 2010.

56 Prosecutor v. Tadić, Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, Case No. IT-94–1-AR72, A.Ch., 2 October 1995, at para. 91 (discussing the scope of Common Art. 3 in relation to ICTY jurisdiction).

57 K. J. Heller, ‘The Rome Statute in Comparative Perspective’, in K. J. Heller and M. D. Dubber (eds.), The Stanford Handbook of Comparative Criminal Law (2009), 45.

58 Ambos and Wirth, supra note 9, at 33.

59 Robinson, supra note 26, at 51.

60 Katanga Confirmation Decision, supra note 7, at para. 396.

61 Ibid.

62 Cryer et al., supra note 4, at 195, n. 41.

63 Bemba Confirmation Decision, supra note 11, at para. 115.

64 Elements of Crimes, supra note 8, at Art. 6.

65 Schabas, supra note 23, at 966–7 (presenting arguments as to why the conclusions in Jelisić on non-policy genocide do not accurately reflect opinio juris).

66 Report of the Ad Hoc Committee on Genocide, UN Doc. E/AC.25/SR.4 (1948), 3–6 (fourth meeting).

67 Ibid., at 6.

68 Ibid., at 4.

69 ILC 1996 Report, supra note 28, at 89.

70 Ibid. at 95.

71 Report of the International Law Commission on the Work of Its Forty-Seventh Session, UN Doc. A/50/10 (1995) (hereinafter ILC 1995 Report), 66 (making the decision to exclude terrorism and five other crimes).

72 P. Robinson, ‘The Missing Crimes’, in A. Cassese et al. (eds.), The Rome Statute of the International Criminal Court: A Commentary (2002), 497–525 (explaining the history behind crimes that were considered, but not included, in the Rome Statute).

73 M. Di Filippo, ‘Terrorist Crimes and International Co-operation: Critical Remarks on the Definition and Inclusion of Terrorism in the Category of International Crimes’, (2008) 19 EJIL 533, at 566.

74 Robinson, supra note 72, at 517.

75 A list of nine terrorism conventions in existence before the Rome Conference is included in the annex of the 2000 International Convention for the Suppression of the Financing of Terrorism, UN Doc. A/54/109.

76 Ibid.

77 W. Schabas, Introduction to the International Criminal Court (2007), 181–2 (inferring this requirement from ICC PTC jurisprudence); see also Prosecutor v. Lubanga, Decision on the Prosecutor's Application for a Warrant of Arrest, Case No. ICC-01/04–01/06–8-Corr 17–03-2006, PTC I, 10 February 2006 (hereinafter Lubanga Arrest Warrant Decision), paras. 31–40 (national proceedings must encompass both the person and the conduct which is the subject of the case before the Court to be inadmissible). The Court has not made it clear if a legitimate terrorism prosecution is prosecuting the same conduct as crimes against humanity for complementarity purposes.

78 See K. J. Heller, ‘The Shadow Side of Complementarity: The Effect of Article 17 of the Rome Statute on National Due Process’, (2006) 3 Criminal Law Forum 255 (arguing that complementarity makes it broadly permissible and easier for states to convict criminals, even to the point of eroding due process guarantees).

79 Several countries that have abolished the policy requirement are listed in the Kunarac footnote. See text accompanying note 18supra. But see Schabas, supra note 23, at 963 (explaining that, because they had been overruled on this point by the country's Supreme Court, the Canadian precedents were erroneously included). The United States does not have a criminal crimes against humanity statute on point, and relevant civil cases under the Alien Tort Statute (ATS) give little guidance as to whether policy must be demonstrated when alleging crimes against humanity in the ATS context. See Cabello v. Larios, 402 F.3d 1148, 1161 (11th Cir., 2005) (policy not addressed when declaring how crimes against humanity involve proving a widespread or systematic attack directed against any civilian population.); Wiwa v. Royal Dutch Petroleum Co., 2002 WL 319887, 9–10 (SDNY, 2002) (purporting to follow Rome Statute Art. 7, but ignores the policy requirement and defines crimes against humanity only as widespread or systematic attacks against a civilian population).

80 Schabas, supra note 23, at 982.

81 Prosecutor v. Katanga and Chui, Judgement on the Appeal of Mr Germain Katanga against the Oral Decision of Trial Chamber II of 12 June 2009 on the Admissibility of the Case, Case No. ICC-01/04–01/07–1497, A.Ch., 25 September 2009, paras. 80–83.

82 Goldsmith, J., ‘The Self-Defeating International Criminal Court’, (2003) 70 University of Chicago Law Review 89, at 101CrossRefGoogle Scholar.

83 Schabas, supra note 23, at 960.

84 Prosecutor v. Akayesu, Judgement, Case No. ICTR-96–4-T, T.Ch., 2 September 1998, at para. 580.

85 Prosecutor v. Blaškić, supra note 7, at para. 206.

86 Prosecutor v. Abu Garda, Decision on the Confirmation of Charges, ICC-02/05–02/09–243-Red, PTC I, 8 February 2010 (hereinafter Abu Garda Confirmation Decision), at para. 31.

87 Situation in the Democratic Republic of the Congo, Judgment on the Prosecutor's appeal against the decision of Pre-Trial Chamber I entitled ‘Decision on the Prosecutor's Application for Warrants of Arrest, Article 58’, Case No. ICC-01/04–169, A.Ch., 3 July 2006, at paras. 68–82 (reversing a gravity test advanced by the PTC in the Lubanga Arrest Warrant Decision). The author takes no position as to whether the new gravity test from the Abu Garda Confirmation Decision would satisfy the Appeals Chamber; it is only considered in this article because it is the only gravity test that has, at present, not been overruled by the jurisprudence.

88 Katanga Confirmation Decision, supra note 7, at para. 396.

89 Chesterman, S., ‘An Altogether Different Order: Defining the Elements of Crimes against Humanity’, (2000) 10 Duke Journal of Comparative & International Law 307, at 316Google Scholar.

90 Rome Statute, supra note 2, Art. 1.

91 2002 Statute of the Special Court for Sierra Leone, 2178 UNTS 138.

92 Rome Statute, supra note 2, at Art. 11(1).

93 Chesterman, supra note 89, at 342.

94 Rome Statute, supra note 2. The court's jurisdiction can be triggered by referral from a state party (Art. 14), referral from the Security Council (Art. 13(b)), or on the initiative of the Office of the Prosecutor (Art. 15). Normally the court only has territorial jurisdiction over states parties, but territorial jurisdiction may also be conferred by Security Council referrals or by non-parties on an ad hoc basis (Art. 12(3)).

95 Ibid., Art. 12(2)(a).

96 Para. 6 of UN Security Council Resolution 1593 states that referrals exclude members of non-State parties for ‘alleged acts or omissions arising out of or relating to operations in Sudan’. UN SC Res. 1593, UN Doc. S/Res/1593 (2005), at para. 6. Regardless of whether para. 6 is legal under the UN Charter or a jus cogens legal obligation, the Prosecutor has made it clear in other contexts that exceptions to personal jurisdiction will not be effective. Situation in Uganda, Decision to Convene a Status Conference on the Investigation in the Situation in Uganda in Relation to the Application of Article 53, Case No. ICC-02/04–01/05–68, PTC II, 2 December 2005, paras. 3–4. See also Schabas, supra note 77, at 157 (Whatever the legality of para. 6 of Resolution 1593, it is ‘most certainly incompatible with the Rome Statute’.)

97 Report of the Secretary-General on the Establishment of a Special Tribunal for Lebanon – Attachment – Statute for the Special Tribunal for Lebanon, UN Doc. S/2006/893 (2006), Art. 1.

98 The parties to the STL may change the end date with Security Council approval. Ibid.

99 Report of the Secretary-General on the Establishment of a Special Tribunal for Lebanon, UN Doc. S/2006/893 (2006), para. 24. The relevant paragraph reads as follows: 24. Mindful of the differences in scope and number of victims between the series of terrorist attacks committed in Lebanon and the killings and executions perpetrated on a large and massive scale in other parts of the world subject to the jurisdiction of any of the existing international criminal jurisdictions, it was nevertheless considered that the 14 attacks committed in Lebanon could meet the prima facie definition of the crime, as developed in the jurisprudence of international criminal tribunals. The attacks that occurred in Lebanon since 1 October 2004 could reveal a ‘pattern’ or ‘methodical plan’ of attacks against a civilian population, albeit not in its entirety. They could be ‘collective’ in nature, or ‘a multiple commission of acts’ and, as such, exclude a single, isolated or random conduct of an individual acting alone. For the crime of murder, as part of a systematic attack against a civilian population, to qualify as a ‘crime against humanity’, its massive scale is not an indispensable element.

100 Ibid., para. 25; see also Report of the Secretary-General on the Establishment of a Special Tribunal for Lebanon: Addendum, UN Doc. S/2006/893/Add.1 (2006) (statement by Nicolas Michel confirming the exclusion of crimes against humanity from the STL).

101 Slye and Van Schaack, supra note 38, at 108.

102 See generally Van Schaack, supra note 29. See also Kenya Decision, supra note 14, paras. 8–9 (Judge Kaul, dissenting) (emphasizing the importance of the distinction between international and national crimes).

103 Van Schaak, supra note 29, at 819 (certain inhumane acts that defined the Nazi era could never be ‘shoehorned’ into crimes against the peace or traditional notions of war crimes).

104 M. Cherif Bassiouni, Crimes against Humanity in International Criminal Law (1992), 186.

105 Rome Statute, supra note 2, at Art. 121.

106 Ibid.