The accompanying article by Matt Halling, about which I have been asked to comment, argues for an amendment to Article 7(2) of the Rome Statute. Article 7(2) imposes a limitation on the scope of crimes against humanity by requiring that they be perpetrated ‘pursuant to or in furtherance of a State or organizational policy’. The justification for such a change to the Statute is the concern that ‘deserving crimes admissible under the complementarity and gravity requirements cannot be prosecuted’ as a result of ‘compromised language’ that creates ‘a . . . loophole, admittedly small but very real, where crimes well within the purpose of the ICC are non-justiciable because of the policy requirement’.Footnote ¹

To demonstrate that the ‘loophole’ is ‘very real’, Halling presents three scenarios. The first is a ‘rogue general’ with a genocidal bent and access to nuclear weapons. In most countries, such a lunatic would get early retirement, but the scenario is completed by the failure of the country to prosecute because the racist general is so popular and action against him might spark a civil war. Why an amendment to the Rome Statute would assist is not clear. Wouldn't transfer to The Hague of this much-admired war hero also provoke the dreaded civil war?

The second is the ‘terrorist blogger’. Why is a blogger who points to the shortcomings in a state's security apparatus deemed a terrorist? Perhaps as a serial blogger myself, I'm overly sensitive to such attacks on my freedom of expression.Footnote ² The fact that some genuine terrorists read the blog and manage to exploit the information it provides (the state's own incompetent security people don't seem to read blogs; otherwise, they'd have fixed the problem before the terrorist attack) hardly makes the blogger an accomplice. It isn't clear why the blogger in scenario 2 is a criminal at all, let alone a criminal against humanity, with or without the policy element.

The third is the ‘corporate hacker’ who obtains confidential information about members of a political party and then disseminates this on the Internet. The result is ‘bankrupting and otherwise disrupting the lives of hundreds of thousands of party B members’. Am I missing something here? How could the dissemination of such information be a crime against humanity, even without the policy element? Halling blithely describes it as ‘persecution’, but it looks to me like little more than a breach of the right to privacy, albeit on a large scale. Bankrupting and disrupting lives sounds like business as usual at Goldman Sachs and Lehman Brothers. As in scenario 1, government officials are being blackmailed, so they will not prosecute. It is not apparent why these terrified souls would overcome their fears and co-operate with the International Criminal Court.

Let me try and help out here. How about the scenario where the world is attacked by a powerful extraterrestrial, who lands on an ice floe outside any state's territorial waters and then launches microwave attacks on ethnic minorities. We would need an amendment not only to Article 7(2) but presumably, also, to Article 12, because neither extraterrestrials nor ice floes fall within the scope of the Court's personal and territorial jurisdiction. Here's a loophole worth worrying about. A Security Council resolution could cure the defect, because under Chapter VII the Council has jurisdiction over extraterrestrials. But even if the Security Council were to act, what if we were to discover that the extraterrestrial was only sixteen years of age? This too could lead to impunity. We'd need to amend Article 26 as well!

Or suppose an evil mastermind manages to shoot down a US aircraft carrying nuclear warheads, and then hides them in an underwater cave. He threatens to attack major urban centres in Europe and Asia unless apartheid is reinstated in South Africa. He is captured by a glamorous British agent, but judges in the United Kingdom deny extradition out of concerns that a trial would be unfair. Writers like Ian Fleming were dreaming up such scenarios long before the Preparatory Committee of the International Criminal Court began tinkering with the definition of crimes against humanity adopted at Nuremberg.

Of course, the examples are all preposterous. Can't Matt Halling provide a real example, instead of a fantasy drawn from science fiction, Hollywood, and comic books? He should show us how Pol Pot or Eichmann or an Idi Amin would escape justice because of shortcomings in the definition of crimes against humanity, given the lack of a policy element. The attentive reader will note that all of these real examples involve agents of states acting pursuant to policy. The lone maniac in the Halling scenarios is a figment of literature, Hollywood, and video games.

A second argument invoked in the article is the desire for coherence in the Statute. Halling explains that there is no policy requirement for genocide, so, he asks, why should there be one for crimes against humanity? He too quickly dismisses the good arguments for an implicit requirement of a policy for genocide, not to mention some language in the Elements of Crimes that suggests something akin to policy. But whatever the interpretation, does not the consistency argument work equally well in the other direction? Why not amend the definition of genocide in order to require explicitly that it be perpetrated as part of a state or organizational policy? That way, we would ensure consistency with crimes against humanity. There are arguments on both sides of this divide, and both sides cite the need for consistency.

The article makes the questionable assertion that states are required to employ the definition of crimes against humanity set out in the Rome Statute, failing which they will be deemed unwilling or unable, thereby opening the door to prosecution by the International Criminal Court. We do not yet have any authority for this, because to date all rulings on admissibility have confined themselves to cases where the state in question is ‘inactive’. It seems excessive to expect the Court to intervene where a state has conducted a serious prosecution for crimes against the person yet failed to label the act genocide, crimes against humanity, or war crimes in accordance with the Statute. Would not the Court have better things to do with its limited resources?

There is a strong presumption in the Rome Statute that national courts are the natural forum for the prosecution of ‘the most serious crimes of concern to the international community as a whole’. Institutionally, however, there seems to be a powerful pull on the Prosecutor and on judges to snatch cases away from states, perhaps to provide a relatively inactive Court with work. Hopefully, interpretations and practices will develop by which we return to a system that is not too strict or demanding on national systems, and that seeks to leave them with the cases wherever possible. Pedantic, technical arguments in these early days by which the Court trumps national prosecutions cannot work in the long run. Moreover, they run contrary to the underlying principles of the Statute.

Halling says that ‘several countries . . . have abolished the policy requirement’, adding that they are listed in the famous footnote in the Appeals Chamber decision in Kunarac, but I am afraid that it is not clear what he is talking about. Perhaps we could be given some examples of states that purport to exercise universal jurisdiction over crimes against humanity allegedly perpetrated by individuals such as the rogue general, the hacker, and the online terrorist (I will concede that there have been no universal jurisdiction cases targeting extraterrestrials, even juvenile ones, or the mastermind at the head of SPECTRE). But it would certainly make sense for states to abolish the policy requirement to the extent that they are prosecuting crimes against humanity committed on their own territory. For that matter, it would make sense for them to forgo prosecution of crimes against humanity altogether. States possess jurisdiction for ‘ordinary’ crimes of murder and so on committed on their territory. They need neither the policy element nor any other ingredient from the concept of crimes against humanity because they are not engaged in international prosecution.

If we pursue the logic of Halling's argument, why not eliminate the concept of crimes against humanity altogether, to the extent that it is unnecessary for national courts exercising traditional modes of jurisdiction. He no doubt will reply that this might reduce the stigma of prosecution for a serious international crime. But that, it seems to me, is an argument for narrowing the definition of crimes against humanity rather than broadening it. The narrower the definition, the greater the stigma.

The history of criminal law, and not just international criminal law, is one of plugging loopholes as they arise. But it is consistently a response to real crimes, rather than contrived hypotheses. There remain many ‘loopholes’ in the Rome Statute. One of them that, to be fair, is signalled in the Halling article, is terrorism. Indeed, this very much gets to the heart of the real debate about the scope of crimes against humanity, as it is unlikely that the Court will concern itself with serial killers or motorcycle gangs, let alone the Joker or an extraterrestrial. Many have argued that the requirement that the policy be attributable to an ‘organization’ as well as a state was inserted to ensure that terrorism could be addressed. It does not seem inadvertent that two of the three hypothetical scenarios proposed by Halling expressly contemplate ‘terrorists’. But even the first scenario, of Dr Strangelove, is in reality one of terrorism. Probably there is a reluctance to consider the threat to use nuclear weapons as an act of terrorism because this might make the permanent members of the Security Council, not to mention India, Pakistan, and Israel, a bit uneasy. It is actually a very good example of a weapon whose effectiveness as a threat rests almost entirely on its ability to instil terror.

The argument that ‘crimes against humanity’ do, or should, deal with terrorist crimes meets with serious obstacles. When the United Nations was considering how to address the 2005 assassination of the Lebanese politician Rafiq Hariri, there was an attempt to characterize it as a crime against humanity. In the Secretary-General's report on the draft statute of the Special Tribunal for Lebanon, a fairly detailed reference to the discussion of this issue in the Security Council appears, along with the conclusion that ‘considering the views expressed by interested members of the Security Council, there was insufficient support for the inclusion of crimes against humanity within the subject matter jurisdiction of the tribunal’.Footnote ³ This controversial remark prompted an unprecedented declaration by the Legal Counsel to the United Nations: ‘The text of the statute, the language of the report, the preparatory work and the background of the negotiations clearly demonstrate that the tribunal will not be competent to qualify the attacks as crimes against humanity.’Footnote ⁴ It was a case of damage control. The Secretary-General's attempt to push the boundaries of crimes against humanity had provoked a fierce reaction from ‘interested members of the Security Council’. Halling seems puzzled by the exclusion of crimes against humanity from the Statute of the Special Tribunal for Lebanon, and treats the fact as being entirely neutral. Actually, the paper trail indicates strenuous opposition to the concept within the Security Council. An additional element to this debate is the failure to incorporate terrorist crimes into the Rome Statute. The prospect had been entertained in the Final Act of the Rome Conference, but the attempt by the Netherlands to put the matter on the agenda of the Review Conference met with failure.Footnote ⁵

This obsession with loopholes in the subject-matter jurisdiction of the Rome Statute is difficult to understand. Isn't the real issue the fact that the International Criminal Court can prosecute only a handful of cases at any one time? There seems to be plenty in Articles 6, 7, and 8 to keep it busy for many years to come, without the need for amendments and additions. The Prosecutor of the International Criminal Court recently set out a three-year strategic plan according to which he expected to finish the three trials now under way or about to begin, and to start ‘at least one new trial’ before he completes his mandate in 2012. In addition, he said that he intended to continue ongoing investigations in seven cases, and conduct ‘up to four new investigations of cases’.Footnote ⁶ Assuming that he fulfils these goals, that will mean that in its first nine years of operation the International Criminal Court will have completed the trials of four people, started the trial of another individual, and investigated cases involving another 11 individuals, for a total of 16. By comparison, the International Criminal Tribunal for the former Yugoslavia (ICTY) had completed 23 cases after nine years of operation (coincidentally, the Nuremberg tribunal also judged a similar number, finishing its trial less than fourteen months after its work began).Footnote ⁷

But if the past is any guide, it would seem unlikely that the Prosecutor will meet even these exceedingly modest forecasts. Six years ago he proposed a budget that was based on the proposition that ‘[i]n 2005, the Office plans to conduct one full trial, begin a second and carry out two new investigations’.Footnote ⁸ A flow chart derived from the Prosecutor's forecasts indicated that the first trial before the Court would be completed by August 2005.Footnote ⁹ He became somewhat less ambitious in 2006, when a three-year strategic plan proclaimed the expectation that the Court would complete two ‘expeditious trials by 2009, and . . . conduct four to six new investigations’.Footnote ¹⁰ In fact, by 2009 not one trial was even close to completion. Thus, if experience is any guide, the exceedingly modest targets set by the Prosecutor in the recent strategic plan are unlikely to be attained.

There remains much to be clarified about the interpretation of Article 7 of the Rome Statute. That, of course, is an issue quite separate from the proposal that Article 7(2)(a) be amended. As the Statute now stands, the policy element cannot be ignored. Pre-Trial Chamber I, in the decision authorizing the investigation by the Prosecutor in Kenya, was divided on the scope to be given to the term ‘organizational’. It should be borne in mind that this was an ex parte ruling, and that the judges did not have the benefit of a vigorous challenge to the Prosecutor's submissions. In assessing the organizational issue, the two-judge majority held that ‘a number of the attacks were planned, directed, or organized by various groups including local leaders, businessmen, and politicians associated with the two leading political parties, as well as by members of the police force’.Footnote ¹¹ Furthermore, ‘the supporting material contains references pointing to the involvement of Kalenjin leaders, businessmen and ODM politicians, including cabinet ministers’.Footnote ¹² As for ‘the attacks emanating from the police, it is reported that the killings of the suspected Mungiki members occurred pursuant to a government campaign aimed at the suppression of this gang while the killings of suspected SLDF members and Mt Elgon residents reportedly occurred in the context of a government joint military–police operation’.Footnote ¹³ This Kristallnacht-like situation certainly hints at involvement of the state, or of some elements within it. The majority decision is hardly good authority for the large and liberal approach to the word ‘organizational’ that some have proposed.

Judge Kaul's compelling dissenting opinion was even stricter. He held that the organizations contemplated by Article 7(2)(a) ‘should partake of some characteristics of a state. Those characteristics eventually turn the private “organization” into an entity which may act like a state or has quasi-State abilities’.Footnote ¹⁴ Judge Kaul's discussion of the historic context of the adoption of crimes against humanity and the function this concept has played in international law is both rare and refreshing. It also provides a very eloquent rejoinder to the arguments advanced by Halling in his article.

There is no doubt that the policy element debate, whether applied to crimes against humanity or genocide, represents a significant difference between the law applicable at the International Criminal Court and the ICTY. The judges at the International Criminal Court must contend with explicit texts, imposed by states through the Statute itself and the Elements of Crimes. In the case of crimes against humanity, a ‘State or organizational policy’ is required, whereas for genocide, this must be ‘a manifest pattern of similar conduct directed against that group or . . . conduct that could itself effect such destruction’.Footnote ¹⁵ The more laconic provisions of the Statute of the ICTY opened the door for judges to declare that no such policy requirements existed under customary international law. In Kunarac, as Halling recounts, they tried to prove this using a long list of authorities. On closer examination, however, the references did not really establish the point. The argument that there is no policy element in genocide rests on a literal construction of the definition, and divorces the concept from its context. But the orientation of the judges at the ICTY on this point is probably best explained by the low-grade caseload they were given in the institution's early years. A policy element was an inconvenient obstacle when the roster of accused consisted of thugs and hooligans such as Tadić, Kunarac, and Jelisić. These were not the masterminds of the atrocities. Later, when the Tribunal began dealing with the higher ranks, it discovered the value of a policy, especially in the application of its joint criminal enterprise theory.

Halling argues for an amendment to the Rome Statute, so his case has to rest on policy considerations rather than law. His contention is not that the Statute should be interpreted in a certain way, nor does he rely on trying to reconcile the applicable law of the Court with what some may claim to be customary international law. He says that the definition may enable certain villains to escape the Court's jurisdiction, and that the loophole needs to be closed. In my view – and this, too, is a consideration of policy – the Court needs more loopholes, not fewer. Let us use the metaphor of a fishing net. Fishermen use nets of various sizes depending on the fish that they wish to catch. If you want small fish, then you have a fine net. You will catch some big fish, too, because they will not be able to swim through the small holes. But if you want to confine your energies to big fish, you need larger holes. Otherwise, you'll spend all your time picking out the sardines, minnows, and plastic bottles. One of the great challenges of international criminal justice is the selection of cases. Resources must be focused using the concept of ‘those who bear the greatest responsibility’ or something similar. To do so, we need a large net with big holes, not a fine one. The policy element for crimes against humanity (and genocide) is one way to achieve this.