Archive of posts for category
Featured Posts

Comments on Atrocity Speech Law by Gregory Gordon

by David Simon

[David J. Simon is the Director of the Genocide Studies Program at Yale University.]

It is something of a cliché to call a newly published book an “achievement.”  I can think of no better word, however, to describe Gregory Gordon’s Atrocity Speech Law.  This is the rare book on legal doctrine that is engaging and digestible to lawyers, legal academics, and non-lawyers alike.  Professor Gordon describes horrifying episodes of atrocity speech in recent history, meticulously develops and dissects the ensuing case law that unfolded, and cogently argues for a more comprehensive vision of future standards by which atrocity speech should be proscribed, prosecuted, and adjudicated upon.  These quite different component parts are so well executed that it is almost easy to forget by the magisterial end that the whole project begins with the coining of the term “Atrocity Speech Law.”  Gordon has set an extremely high standard for future books on the subject.

Gordon begins the task of elaborating what this field encompasses by surveying a disparate/scattered set of jurisprudential concerns surrounding speech related to the commission of atrocity crimes.  Finding coherence lacking, he proceeds to articulate both a jurisprudential fix and a set of real life motivations for doing so.  Not being a legal academic myself, I will mostly refrain from commenting upon the former.  Suffice it to say that Gordon’s reverse engineering of atrocity speech case law exposes and illustrates an unholy trinity of haste, political expediency, and under-competent judging that all too frequently characterizes the formation (and execution) of international humanitarian and criminal law.

The fix he proposes to the doctrinal mess he describes is not merely a matter of bringing order to a chaotic corner of the legal world.  For Gordon,

…folding all speech-related delicts into a single provision provides for a better organized exposition of the law and calls greater attention to speech as a potent factor in provoking and fueling atrocity.  As a practical matter, rather than having a fragmented collection of rules scattered in different parts of a statute, the proposed consolidated provision puts the proper emphasis on speech crimes and facilitates simultaneous consideration of the offenses so that prosecutors can make more efficient and effective charging decisions (385).

However, as my own interests and expertise, such as it is, lie in the field of atrocity prevention, I focus on the implications of Gordon’s work there.  In a narrow reading, there is benefit – indirect but important – to doctrinal coherence:  too much of international prevention efforts are ad hoc, uncoordinated, and therefore come off as collectively amateurish.  Like its international post-conflict justice component, the prevention regime itself, therefore, seems easily circumventable, characterizable as the project of one-world dreamers with little bite except in occasional instances of post hoc convenience.  One might reasonably expect a legal code to be the exception, but as Gordon shows in his survey of the field’s fragmentation, this is far from the case.  The call of coherence is also a call for the prevention project to be taken seriously, and for it to have the tools necessary for that to be so.

Displaying the righteous zeal of a former prosecutor (and I mean that as an unmitigated compliment), Gordon offers a more direct payoff projected from this project.  He notes that the operationalization of his doctrinal fixes

would promote greater consistency in prosecuting atrocity speech in both municipal and transnational fora around the world.  And it would more effectively put would-be verbal hatemongers on notice that speech in atrocity contexts will not be the collateral postscript to filling mass graves.  Rather, it will serve as humanity’s alarm bell, auguring and triggering punishment before it is too late (395).

The logic underpinning this bold claim is straightforward:  several efforts to establish constructs by which to understand genocide have, as Gordon cites, emphasized the role that speech plays in the process of effecting genocide and other atrocities. For example, Gregory Stanton’s “Ten Stages” theory gives significant weight to the role of speech acts in fomenting atrocities.  Nearly half of Stanton’s stages may substantially involve speech: Classification (Step 1), Symbolization (2), Dehumanization (4), Polarization (6), and – controversially, as Gordon acknowledges, Denial (10) (see here).

If conceptualizations like these are accurate, to prevent genocide requires disrupting the willful acts core to each step of the process.  And when those acts include speech acts, a framework for prevention necessarily must include a strategy for identifying speech that has the potential to cause atrocity – and doing something about it. A comprehensive and consistent framework for defining atrocity speech, as Gordon provides, enables the first task.

The second, however, remains a challenge, particularly as one moves away from relying on the knowledge of the possibility of ex post adjudication and punishment having a deterrent effect, and towards ex ante identification and intercession.

At the basis of the difficulty of ex ante intercession is Gordon’s completely accurate conclusion that the harm of atrocity speech is not necessarily a product of the injury (or violence) it causes, but in its very potential to cause violence.  Thus, Gordon rightly argues that to respond appropriately (i.e., preventively) to incitement, the doctrine must recognize its inchoate nature. To invoke a phrase from another field, the anti-atrocity speech regime must feature (proactive) police patrols as well as (reactive) fire alarms.  However, convincing courts to play their proactive role is a tough sell on the criminal side.  International courts can barely maintain their capacity to render ex post judgments, much less take on ex ante cases.  Both international and national courts have been loath to separate guilt from the causation of harm, even when statutes dictate otherwise.  Furthermore, it is hard to imagine atrocity speech, which for better or for worse always has a tentacle in the realm of politics, being given a hearing which itself would not be accused of being political.  In most countries, a state, through its municipal courts, would be unlikely to look into the alleged speech transgressions of (fellow) state actors out of concern for its own legitimacy.  Prosecutions of non-state actors would be more likely to occur, but inevitably smack of overreach, an attempt of to use the power of the state to quash dissent.

The guidance Gordon gives, while delineating with remarkable clarity the nature of speech that should elicit a response anterior to atrocity actions, leaves unanswered questions  as to who ought to identify the transgressions and who ought to initiate the prosecutions.  As with much of the genocide prevention effort, political expedience concerns will likely override implementation even where municipal jurisdictions have adopted it.  Skepticism of internationalism combined with realist conceptions of foreign policy will mute its application at the international level. The outsize role the United States plays in establishing global norms is especially challenging for the implementation of a prevention agenda around atrocity speech:  not only is skepticism of global governance high, but there is extreme reluctance on the part of the United States to countenance ex ante speech limitations of any sort, a handicap to the adoption of anti-atrocity speech norms.

I suspect – and by that I mean ‘fear’ – that the arc of the Responsibility to Protect (R2P) norm may represent the most likely trajectory.  R2P emerged from an idealistic set of principles, gained life as a body of UN-elaborated language, and withered to something that is cited in international policy when not significantly inconvenient to major powers.  The unwillingness to place principle above politics (and, barring that, fiscal constraints) relegates R2P to the status of secondary concern.

And yet the story of R2P offers a glimmer of hope, as well.  The advent of R2P has introduced language of responsibility into the proceedings of various junctures of global governance.  When global actors use or endorse the language disingenuously, that disingenuousness comes with a price in terms of reputation and credibility (albeit one that countries, over and over again, have been willing to pay.)  Where R2P tends to succeed is below the Security Council level, in the rhetoric of middle powers, in the actions of young regional organizations like the African Union’s Peace and Security Council, or the Office of Security and Co-operation in Europe.  Politics and other concerns can still supersede the sway of the norm at this level, but with fewer veto players holding trump cards at higher levels, the doctrine stands a greater chance of gaining a foothold.

I would argue that the task for the atrocity speech crimes project is to build an implementation infrastructure that does not rely solely on judges and criminal complaints.  The architecture of atrocity speech policy must acknowledge legal realism, which recognizes the enduring political barriers to implementation of a regime based solely on ideals. In doing so, the goal would not be to pare back, or make less inclusive, the anti-atrocity speech framework.  Rather, the task is to populate the regime with institutions other than politically disinterested municipal courts or under-empowered international ones.

The task is to develop an administrative architecture that could preside over an essentially civil law application of Gordon’s framework.  The architecture might involve a network of non-governmental organizations paired with independent international bodies, perhaps set up at regional levels.  A core mandate of these bodies would be to hear such cases of alleged atrocity speech, and deliver injunctions accordingly.  While the bodies would possess no policy power, the failure of the enjoined to adhere to the injunctions then would become legal matter in its own right, triggering referral to either a higher juridical body (like the ICC) or a political one (like the Security Council, or its equivalents on regional government organizations).

There exist blueprints in the world of global governance for such a mid-level institution. The European Court of Human Rights has injunctive powers (albeit among others). Alternatively, treaty-derived institutions like the International Criminal Court or the International Atomic Energy Agency interface with the United Nations but were separately established, and thereby exercise a measure of independence.  Thus Article 6 of Gordon’s proposed treaty could be expanded to specify a mid-level international organization that could be charged with identifying or responding to atrocity speech complaints.  Certainly, this would not be without controversy, and the institution(s) would be vulnerable to capture.  Yet compromisable new institutions remain preferable to vesting all anti-atrocity speech prerogatives with national and international organs that have, to this point, created a muddled body of law, and whose politicization is already manifest.

Technology, Speech, Hate, Virtuality … and the Path of the Law

by Mark Drumbl

[Mark A. Drumbl is the Class of 1975 Alumni Professor of Law and Director, Transnational Law Institute at Washington and Lee University.]

Atrocity Speech Law is a hefty book. It is, as Professor Gordon himself describes it, a ‘tome’. Atrocity Speech Law is rigorous and ambitious: packed with information, breathtakingly detailed, brimming with integrity, and vivified by important themes of law reform. In contrast to the absurd invective it seeks to deter, Gregory’s arguments are measured and modulated, poised and principled.

Although Gregory has invested so much in this book, and the book contains so much, he also is wisely modest that much more remains to be said. So I want to say something about one thing Gregory has identified as something that remains to be said, and then I want to say something about something else that he hasn’t identified as such.

But, first, to the core argument: Gregory posits that atrocity speech law exists but is hampered by the fact it is fragmented. The fragmentation, I would hasten to add, is not deliberate or intentional. Like much of law, the law regulating atrocity speech emerged by virtue of bricolage. That said, and however charmingly organic, bricolage quickly brushes up against important limits: concerns over coherence, predictability, and consistency. In fact, simply by labeling the issue at hand as ‘atrocity speech’, Gregory already contributes structure and parsimony, in that the regulation of speech that encourages violence currently remains cobbled together from various specific crimes such as incitement to genocide, hate speech, ordering, instigation, and persecution as a crime against humanity. By advancing arguments of unification and codification (he proposes a draft convention on atrocity speech), Gregory situates himself within a venerable tradition of international law-making. Although I often find myself attracted to (a bit of) messiness, and believe the value of Cartesian organization in the world of law may be overrated and too hungrily stated, Gregory makes a very compelling case for his vision.

Gregory gestures ahead near the end of the book. He points to historical research, denial, sentencing, and empirical inquiry on the actual effects of hate speech as areas for further research. Among these, I think that sentencing is of particular salience. Gregory finds unexplainable variation among sentences issued for hate speech crimes at the international level.  I am not surprised. I have long questioned the rationalities of sentencing at the international criminal courts and tribunals and the operational coherence of the sentences that actually are issued. Although greater predictability has arisen over time, which is good to see, pockets of concern persist. Assembling together the various crimes that involve speech, which Gregory has done, and providing a heuristic of sentences for these crimes, which Gregory also has done, exposes these underdeveloped aspects of sentencing. Although retribution and deterrence are taken as the two major goal of international sentencing, as re-announced recently by the ICC in Bemba and Al Mahdi, how would these goals apply in the case of atrocity speech? Can persons convicted for speech crimes be rehabilitated? Are they in a position, unlike other criminals, to undo what they had done, to disclaim what they had claimed, to correct the record, to retract, to unwind? If so, how could these remedies form part of the punitive schematic?

Turning now to the something not spoken of: Atrocity Speech Law largely absents conversations about technology. This surprises me. The book has a bit of a last century vibe to it. Sure, some societies are more technologically embedded and uploaded than others, but we are all well beyond newspapers and radio broadcasts. Our world is less one of RTML and Kangura and megaphones and Nuremberg rallies than it is one of social media, anonymous (and instantaneous) information, and virality; of YouTube and the internet; of ‘fake news’ and doctored events; of Instagram and Facebook and  Twitter and Whatsapp and Iphones. Widespread and systematic access to the Web, so to speak, permits everyone to become a speaker (anonymous or otherwise) and to be heard. So hate speech metastasizes much faster than ever before, it can stain so many listeners so quickly, and can become pandemic. Anyone with an internet connection can start it.  State-run television, cable given over to the interests of an ideology, and the press each certainly and assuredly is important. But I would wish to hear from Gregory as to how his (re)construction of atrocity speech law would map onto these new virtual media which definitively change how and through whom and from where ‘information’ is obtained.  How does technology challenge (or not) the many legal elements of the crime?

Gregory closes his book with an appeal for both application of penal law and preventative measures. So, then, preventatively, how to ventilate the spaces of the Web in which ‘speakers’ vent their furor? Is social host liability, liability for providing space, an answer? If so, what further cascades might such liability present for freedom of expression? Any vexation? What opportunities for reparations? What can we learn from internet regulation domestically, for example, when virtuality is used to cyberbully or humiliate individuals?

Gregory’s book is a tour-de-force. We are all the better for it. We are lucky to have him, and his work, to guide us through these rapidly growing thickets.

Comment on Atrocity Speech Law by Gregory Gordon

by Roger S. Clark

[Roger S. Clark is the Board of Governors Professor of Law at Rutgers Law School.]

Several important themes are developed alongside one another in Gregory Gordon’s remarkable book on the activity for which he coins the term “atrocity speech law.”  They are captured largely in his sub-title “Foundation, Fragmentation and Fruition” and in his summary of the “fruition” points at pp. 16-24 of the work.  I cannot do justice to all his exhaustive scholarship and his insights in a short blog post,  so I shall comment on the general picture and then outline, idiosyncratically no doubt, a couple of the points that particularly struck me.

Gordon assays a concrete legal definition of his subject area in the text of a creative proposed “Convention on the Classification and Criminalization of Atrocity Speech Offenses” (pp. 378-81).  Article 1 of the treaty has the Parties confirming that “hate speech, when uttered for purposes of triggering or supporting atrocity crimes, should be accurately categorized in its different penal forms and criminalized and punished according to the actual form it takes.”  “Such speech,” it adds, shall be considered to give rise to what is known as ‘atrocity speech liability’”.  Article 2 states that “[f]or purposes of this Convention, ‘atrocity crimes’ means genocide, crimes against humanity, and war crimes, as defined respectively in Articles 6, 7, and 8 of the Rome Statute of the International Criminal Court.”  Obviously, I admire his boldness in trying his hand at treaty drafting, which is why I mention at the outset his global solution to the problems elucidated in the previous three hundred-odd pages.  I’ll return to the Convention later.

The “foundation” part of the argument expands on his understanding of the “historical record,” namely that “instances of mass atrocity have always been accompanied by communications campaigns designed to demonize the intended victims and inflame the passions of would-be perpetrators.”  (P. 5, and see his excellent typography of the various forms of demonization at pp. 284-91, where he relates them especially to the “direct” requirement in respect of public incitement to genocide.)

The first attempt to deal with such activities through the criminal law was at Nuremberg where Julius Streicher was charged, convicted and executed on the basis of the persecution leg of crimes against humanity.  The confusion about whether an inchoate incitement theory, which Gordon espouses, would be appropriate in such cases begins with Streicher, although the issue was not pressing there, as the evil deeds had occurred at the time of trial and could be attributed to him and persecution filled the bill.

It should be noted, though, that the Nuremberg prosecution also sought to use an inchoate conspiracy theory.  Had it been developed further, that might have been a viable way to deal with situations which are of concern to Gordon, where there is no necessary causal link between what the accused did and the atrocities actually committed.  But the Tribunal held that the inchoate conspiracy provision in the Charter applied only to crimes against peace (aggression).

The London drafters of Nuremberg had grappled in a desultory fashion with conspiracy, an important feature of the enterprise for Justice Jackson that was not well understood by the others drafters, even the British.  Sometimes, as I read his transcript of the proceedings, I wonder if Jackson himself really understood what he was doing.  In particular, he and the British were not always clear whether they were talking about the inchoate conspiracy doctrine that the British and American legal systems share, or the peculiarly American doctrine that a conspirator may become liable for the crimes committed by other conspirators in somewhat peripheral situations where they would not be held liable by virtue of basic complicity principles (the “Pinkerton” doctrine which remains a feature of federal law).  More on this later too.

Streicher’s persecution leg of crimes against humanity, which might not necessarily include violence, was essentially spun off to become the core of genocide, as defined in the Genocide Convention.  Genocide appears almost identically in the later Statutes of the ad hoc Tribunals for Former Yugoslavia and Rwanda, and of the ICC, although it still finds an overlapping place in definitions of crimes against humanity.

The Nuremberg Charter contains no suggestion that there should be responsibility for attempted crimes under international law; the Tribunal itself was not enthusiastic about conspiracies to commit war crimes and crimes against humanity. The Genocide Convention, however, picks up both attempt and conspiracy responsibility, along with direct and public incitement — three inchoate offenses. It included in its definition not only genocide (as a principal) and complicity in genocide, but also that trio of preparatory offenses. These inchoate offenses travelled, uniquely with reference to genocide, into the Statutes of the ad hoc Tribunals.

If this can be regarded simply as an example of the vagaries of the drafting process, the drafting of the Rome Statute is even more curious.  In Article 25 of the Statute, dealing with “individual criminal responsibility,” attempts (defined essentially in the Model Penal Code language of “a substantial step”) can be prosecuted for all Statute offenses (Article 25 (3) (f)).  This, as I see it, includes not only genocide, crimes against humanity and war crimes, but also the crime of aggression, whose inclusion within the effective jurisdiction of the Court is on the horizon.  Direct and public incitement of others to commit genocide is also included (Article 25 (3) (e)).

This type of criminality is, however, not extended to the other crimes within the Court’s jurisdiction.  I was involved as a representative of Samoa in the negotiations on Article 25.  As best I understood the argument, in a process that proceeded on the basis of consensus, it was that genocide was unique and the type of incitement involved did not need to be extended to the other offenses.

Gordon disagrees with this result and I am inclined to think that he has a point. Conspiracy was even more puzzling for the negotiators.  This time, no reference to inchoate conspiracy for genocide based on the Genocide Convention was carried forward into the Rome Statute.  My impression here was that many players, like their predecessors in London in 1945, had no clue about the difference between inchoate conspiracy and conspiracy as a form of complicity.  They thought they had solved the problem with Article 25 (3) (d) of the Statute, a kind of residual complicity provision which renders criminally responsible a person who:

“In any other way contributes to the commission or attempted commission of such a crime by a group or persons acting with a common purpose.  Such contribution shall be intentional and shall either: (i) Be made with the aim of furthering the criminal activity or criminal purpose of the group … or (ii) Be made in the knowledge of the intention of the group to commit the crime.”

This is a mode of participation in a crime (or an attempt) when that crime or attempted crime actually occurs, not an inchoate offense for which there is responsibility regardless of what happens afterwards.  If it is a “conspiracy” theory, it is one closer to the Pinkerton doctrine (and its international counterpart Joint Criminal Enterprise) but one which, unlike some versions of those, requires a minimum mens rea of knowledge on the part of the “co-conspirator.”  (Inchoate) conspiracy to commit genocide, from the Genocide Convention, got lost in the shuffle.  Those who worked on defining the crime of aggression, in fulfillment of the expectations of Rome that it would be finalized later, saw no reason to carry forward Nuremberg’s conspiracy to commit aggression, given that it had disappeared in Rome in respect of genocide.  (They did think that Article 25 (3) (d) might apply to aggressors in some cases, as long as they were leaders; see Article 25 (3 bis).)

The “fragmentation” part of the book analyzes, in particular, the relative incoherence of the jurisprudence, especially that of the Rwanda Tribunal.  There are difficulties especially with the ways in which the Tribunals have dealt with the “public” and “direct” aspects of incitement to genocide, and with the definitions of persecution and of the complicity words “ordering” and “instigation.”  There is, moreover, the absence of the crime of incitement to commit war crimes.  Gordon argues for improved definitions and for creating the crime of incitement to commit war crimes.  I think he should probably argue for the sake of consistency for responsibility for inciting crimes against humanity also, but I could not find him doing so explicitly.  He has argued in a separate book chapter for a kind of incitement to commit aggression.  He does not appear to be a fan of conspiracy as an inchoate offense, in spite of its tenuous history in relation to aggression at Nuremberg or in the Genocide Convention.

Incidentally, Professor Gordon notes (p. 357) “Conspiracy, direct incitement, and attempts to commit, as well as complicity in the commission of, crimes against peace, crimes against humanity and war crimes” are defined as criminal in the United States Army Field Manual 27-10.  He also notes the complete absence of application or of expert commentary on this provision.  Note also that the Military Commissions Act of 2009 (echoing the earlier 2006 version) claims jurisdiction in Commissions for attempts, conspiracy and (inchoate) “solicitation” to commit the motley bunch of substantive offenses (not including aggression) outlined in the statute.  I assume that ‘solicitation” here is the same as “direct incitement” in the Field Manual.  The requirement of directness perhaps protects some speech, a legitimate concern for Gordon.  The conspiracy part of the MCA has been the subject of extensive litigation, not least in the Hamdan saga.  The legislative assertion that “[t]he provisions of this subchapter codify offenses that have traditionally been triable by military commission” has to be taken with a large dose of salt.

So, to the “fruition” part of the argument.  I think that his modest proposal is to fix Article 25 (3) of the Rome Statute, at least extending incitement (without the public requirement which the case-law suggests is unnecessary) to cover war crimes (p. 351) and presumably crimes against humanity.  The more radical proposal is the proposed treaty on Atrocity Speech Offenses, noted above, which would replace (or run parallel?) to the present less than coherent assemblage of offenses.  The basic idea is apparent from the definition that I quoted earlier.  Article 5 of the draft includes the gamut of different kinds of conduct involving hate offenses which the parties are required to criminalize: “Incitement,” “speech abetting” (encouragement during ongoing atrocities without proof of causation), “instigation” and “ordering” with freshened definitions in each case.

There is plenty to mull over here and examining the Convention in detail would take up a book of blogs, not a mere 1800 words.

A Set of International Crimes without Coherence or a Proper Name: The Origins of “Atrocity Speech Law”

by Gregory Gordon

[Gregory Gordon is Associate Professor of Law, Associate Dean for Development and External Affairs and Director of the Research Postgraduates Programme at The Chinese University of Hong Kong Faculty of Law.  He was formerly a prosecutor with the International Criminal Tribunal for Rwanda and the U.S. Department of Justice, Office of Special Investigations.]

I have always felt that great scholarship is born of great frustration. And that’s what inspired me to write Atrocity Speech Law: Foundation, Fragmentation, Fruition (Oxford University Press 2017). Why was I so frustrated? The answer goes back to my salad days as a lawyer with the Office of the Prosecutor at the International Criminal Tribunal for Rwanda, when I was assigned to the “Media” team. We investigated, and eventually indicted, certain newspaper and radio executives/employees responsible for inflammatory rhetoric disseminated in the lead up to and execution of the Rwandan Genocide. But there were few legal precepts, and even less jurisprudence, available to guide us. What little there was emanated from Nuremberg, where rabid Jew-hating journalist Julius Streicher, Nazi Radio Division head Hans Fritzsche and Reich Press Chief Otto Dietrich had been prosecuted. So, from a legal perspective, we had to be resourceful as we constructed our ICTR media cases centering on charges of direct and public incitement to commit genocide and hate speech as the crime against humanity of persecution – only the latter having been charged against the just-mentioned Nazi propagandists (other possible speech-related charges available to us were instigation and ordering). But given that we were venturing onto what was largely legal terra nullius (especially with respect to incitement), we often had to grope in the doctrinal dark. So that was the first stage of frustration.

Still, from a law development perspective, I remained sanguine. The key “Media” defendants were ultimately convicted – Ferdinand Nahimana and Jean Bosco Barayagwiza (founders of the infamous Radio Télévision Libre des Milles Collines or RTLM, otherwise known as “Radio Machete”), RTLM announcer Georges Ruggiu, and extremist Hutu newspaper editor Hassan Ngeze. The judgments against them, along with that of Mayor Jean-Paul Akayesu for incitement to genocide (the first in history) – offered hope that a decent foundation had been laid for a law that could effectively govern the relationship between speech and atrocity. But that hope turned out to be misplaced.

Over time, for example, it became clear that there were problems with the formulation and application of the incitement crime, comprising the elements of “direct,” public,” “mens rea,” “incitement” and, possibly causation. Issues arose with respect to each of these. I will not deal with each of them here but will provide some brief examples for illustrative purposes. For instance, thanks in large part to the Akayesu judgment’s paying wishy-washy obeisance to both French- and English-language sources, treatment of the “direct” element was schizophrenically situated somewhere between Common Law and Civil Law conceptions. Unfortunately, the French word for “incitement” – inciter – was also the French word for “instigation” – one of whose elements is resultant violence. So that seemed to engender confusion with respect to causation. Incitement, as an inchoate crime, should not require causation. But Akayesu and its progeny were examining causation in the factual portions all the same and the Akayesu judges even went so far as to assert the need to prove “a possible causal link” between the relevant speech and subsequent violence in that case.

As for the “public” element, its inadequacies were exposed in the so-called “roadblock cases” at the ICTR. Even though inflammatory speech uttered at roadblocks was in a “public” place — because on public roads accessible to all citizens — the speech was held not to be “incitement” because “members of the public” were not present. But if enough persons were present at the roadblock, then “members of the public” could be considered in attendance and the speech could be considered “incitement.” This was a distinction without principle and did not seem to be justified from a policy perspective. Moreover, an advocate’s voice in closer proximity to a listener is arguably more compelling than one from a distance. In other words, private incitement can be just as lethal, if not more, than public.

There were problems with crimes against humanity(CAH)-persecution too. That crime consists of (1) knowingly uttering speech as part of a widespread or systematic attack against a civilian population; (2) resulting in a severe fundamental group-rights deprivation (motivated by group-status); and (3) considered as being of the same gravity level as the other CAH acts. Unfortunately, the ICTR and ICTY have adopted polar opposite positions regarding the issue of whether hate speech not calling for violence can serve as the actus reus for persecution as a crime against humanity. The ICTR concluded that such rhetoric blatantly deprives the target ethnic group of fundamental rights and thus could be the basis for charging persecution. But the ICTY, in the Kordić judgment, found that hate speech not directly calling for violence did not constitute persecution because it failed to rise to the same level of gravity as the other enumerated crimes against humanity acts (such as imprisonment or deportation, for example). And so doctrinal gridlock ensued.

Instigation, the prompting of another to commit an offense (resulting in commission of the offence) – with a connection between the prompting and the crime (i.e. a “contribution”) has also been plagued with problems. As we have seen, it has been consistently confused with incitement and this has exacerbated the muddled jurisprudence regarding causation in both bodies of law. Moreover, there is no consistent approach to the crime’s “contribution” requirement, leading to a series of disjointed pronouncements regarding the degree of contribution and, reading last year’s horrid Šešelj judgment, arguably importing a “but for” causation requirement into the jurisprudence. Ordering, essentially instigation in the context of a superior-subordinate relationship, has been deficient as it permits the superior to escape liability if the command is not carried out (clearly problematic when juxtaposed with incitement to genocide, where there is no superior-subordinate relationship between speaker and listener but the speech utterance itself – regardless of resultant violence — carries liability).

In addition to such individual offense problems, I was also beginning to realize these modalities did not function well together as an ensemble, thus creating significant liability loopholes. For example, in the law’s current state, liability for “incitement” — an inchoate crime — is limited to genocide. Crimes against humanity and war crimes are also horrific atrocity offenses. Why was there no incitement liability connected to those crimes? Similarly, speech uttered in support of contemporaneous mass violence — and with knowledge that the violence is occurring — is limited to the offense of persecution as a crime against humanity. Why was there no speech-specific liability for rhetoric uttered knowingly in support of ongoing acts of genocide or war crimes? The answer could not lie in generic accomplice liability, as it does not recognize the unique power of speech to provoke mass atrocity in the first place.

Upon deeper reflection, it occurred to me that this problem owed to the piecemeal development of the entire body of the law from its inception. On an ad hoc basis, according to individual exigencies at different times, this doctrinal assemblage had been cobbled together by taking a hodgepodge of legal concepts, such as inchoate or accomplice liability, and willy-nilly fastening them to different speech activities. Thus, it is only by historical chance that incitement, a form of inchoate liability, only applies to genocide and not to crimes against humanity or war crimes. The resulting gaps frustrate prevention efforts and help encourage repressive regimes to take advantage of the ambiguity and suppress legitimate speech.

So, in the full measure of time, the growing body of jurisprudence was clearly not allaying my initial sense of frustration. Perhaps, I began thinking, fellow academics and other experts were also noticing these problems and offering solutions. But a review of the literature also left me disappointed – it was as fragmented as the doctrine. Despite excellent individual pieces by talented scholars such as Susan Benesch, Carol Pauli, Wibke Timmermann, Diane Orentlicher and Richard Wilson, among others (myself included – guilty as charged!), there had been no comprehensive study of this body of law. Some articles and books had dealt with parts of it but no single volume had furnished a comprehensive analysis of the entire jurisprudential output and the relation of each of its parts to one another and to the whole (although Wibke’s monograph Incitement in International Law, published after I started writing my book, came closest). No one had yet bothered to step back, systematically consider what has been produced, and provide holistic, constructive analysis and suggestions for change.

And thus was the Atrocity Speech Law project born. After completing my research, I found the book logically divided into the three components of its subtitle. Part 1, “Foundation,” begins with a brief history of atrocity speech, focusing on the Armenian genocide, the Holocaust, and mass killing in the former Yugoslavia and Rwanda in the 1990s. It then looks at the modern treatment of hate speech in international human rights treaties (such as the International Covenant on Civil and Political Rights) and in domestic jurisdictions. This serves as a bridge to a history of atrocity speech law focusing on its origins at the Nuremberg trials. Flowing from this, the book examines the development of speech crimes as formulated in the Genocide Convention and the statutes of the ICTR, ICTY and ICC. It then analyzes the relevant decisions issued by these courts, including the seminal ICTR Akayesu, Ruggiu, and Media Case judgments as well as the ICTY’s Kordić decision. It concludes by considering the general framework and the elements of the crimes established by these decisions.

Part 2, “Fragmentation,” goes on to identify the discrepancies within that framework, its inconsistent applications and other problems the framework engenders, as discussed above. Finally, Part 3, “Fruition,” recommends how the law should be developed going forward to deal with these issues. It begins by proposing how to fix the various problems within each individual speech offense. Then it suggests a more comprehensive and elegant solution: a “Unified Liability Theory” that would replace the current patchwork of speech offenses surrounding atrocity — e.g., incitement only applying to genocide, speech-specific inchoate liability not applying to crimes against humanity or war crimes, and speech offenses related to war crimes being limited to instigation/ordering — and create four general categories of speech offenses: (1) “incitement” (an inchoate mode of liability applying to all three core crimes — genocide, crimes against humanity and war crimes — but eliminating the “public” element from the liability portion of the crime and attaching it to sentencing considerations — while removing “direct” from the title only, not the prima facie elements to help protect free speech); (2) “speech-abetting” (a type of accomplice liability for speech knowingly delivered simultaneously with commission of atrocities, and also applying to all three core crimes); (3) “instigation” (a form of commission liability applying to all speech causally related to subsequent atrocity and thus also linked to the three core crimes); and (4) ordering (criminalizing commands to commit atrocity within a superior-subordinate relationship and incorporating inchoate liability).

And all of these reforms can be operationalized through promulgation of a new treaty, “The Convention on the Classification and Criminalization of Atrocity Speech Offenses,” and/or through amendment of the Rome Statute to include Article 25bis — “Liability Related to Speech.” That new provision, whose equivalent could also be placed in domestic statutes as well as the constituent instruments of existing and/or new ad hoc international tribunals, would contain all four types of speech liability — incitement, speech abetting, instigation and ordering — connected to all three core international offenses — genocide, crimes against humanity and war crimes. The point is that speech’s unique and potent contribution to atrocity should be recognized and criminalized in its own right. It should not become lost in a set of scattered provisions, relegated as a functionally invisible adjunct to other criminal law concepts in the general “modes of responsibility” sections of statutes, charters and codes. And the set of principles it gives rise to should have a name commensurate with its elevated status. That name should capture the entire range of the doctrine and its intimate relationship with mass violence. And this book coins that name: “atrocity speech law.”

Symposium: Gregory Gordon’s “Atrocity Speech Law”

by Chris Borgen

Over the next three days we will have an online discussion concerning Gregory Gordon’s new book Atrocity Speech Law: Foundation, Fragmentation, Fruition (Oxford 2017).

We welcome Professor Gordon (The Chinese University of Hong Kong Faculty of Law), as well as Roger Clark (Rutgers Law), Mark Drumbl (Washington and Lee School of Law), and David Simon (Yale Dept. of Political Science), who will comment on themes from the book.

We look forward to the conversation!

The Conundrum of International Organisations Immunities: Jam et al v International Finance Corporation  (Part 2 of 2)

by Rishi Gulati

[Rishi Gulati is a Barrister at the Victorian Bar, Australia; Dickson Poon Scholar of Law at King’s College London; and Academic Expert, Bretton Woods Law, London. This is the second of a two part post concerning recent litigation against the International Finance Corporation (‘IFC’) in US courts. Part I is available here.]

The recent case of Jam et al v International Finance Corporation (.pdf) is significant jurisprudence testing the boundaries of the express waiver of immunities granted to multilateral financial institutions. The 24 March 2016 decision of the US District Court for the District of Columbia can be found here (.pdf) (‘First Instance decision’); and the decision of the US Court of Appeals for the District of Columbia Circuit delivered recently on 23 June 2017 is here. Both those decisions are discussed below.

The First Instance decision– propagating the ‘1945’ approach to IO immunities

As discussed in my previous post, after being denied justice within the IFC’s institutional structures, the plaintiffs approached the national courts of the US to seek a remedy against the IFC for the injury caused due to the development and construction of the Tata Mundra Plant (the ‘Plant’).

Unsurprisingly, the IFC asserted its immunities under the US International Organisations Immunities Act of 1945 (‘IOIA’). The District Court upheld its immunity. Two matters are crucial for understanding the decision.

The first issue concerned the interpretation of the IOIA in terms of the standard of immunities that applies to international organisations such as the IFC. Specifically, does the IFC enjoy ‘absolute’ or ‘restricted’ immunity? Notably, the IOIA grants IOs immunities similar to those enjoyed by states. In 1945, states enjoyed virtually absolute immunity, and thus IOs enjoyed similar immunities. However, the evolution of the restrictive doctrine of State immunity culminating in the introduction of the Foreign Sovereign Immunities Act in 1976 (‘FSIA’) provided that states were only entitled to immunities in respect of their sovereign or governmental functions, not their commercial acts. The question then became whether this restricted view of state immunity was transposable to the immunities of IOs subject to the IOIA regime? Relying on precedent (Atkinson v Inter-American Development Bank, 156 F.3d 1335 (DC Cir 1998), the decision at First Instance (.pdf) concluded that under the IOIA, IOs continued to enjoy immunities  as envisaged in 1945. Thus, the restrictive doctrine did not apply  to IOs such as the IFC.

Second, there was the question of how the IFC’s broadly phrased express immunity waiver was to be interpreted. The court used the language of a ‘corresponding benefit’ to read the IFC’s express waiver provision narrowly, ‘with careful attention to “the interrelationship between the functions of the [IFC] set forth in the Articles of Agreement and the underlying purposes of international immunities.”’ (First Instance decision, p. 8). The language of a ‘corresponding benefit’ has been presumably developed by the courts to address the problematic drafting of the express immunity waivers included in the constituent instruments of several international financial institutions. In Mendaro, it was said that ‘most organizations would be unwilling to relinquish their immunity without receiving a corresponding benefit which would further the organization’s goals’. Immunity in that case was only restricted for ‘actions arising out of [an organisation’s] external commercial contracts and activities’. The rationale is that if external parties do not have an option to seek legal redress against an organisation concerning such actions, they might not be minded to do business with it. Hence, the corresponding benefit to the IO would demand a waiver in such circumstances.

In Jam v IFC, the plaintiffs argued that as the claims arose out of the IFC’s external activity, the express waiver of immunities was triggered. The Court rejected this argument, distinguishing the current case from other examples, and read the express waiver very narrowly as applying only when the claimant is an external party in a direct contractual or business-like relationship with the organisation (First Instance decision, p. 9). The Court went on to weigh the costs and benefits of any waiver. It is apparent that the more the Court is required to scrutinise the internal functioning of an organisation, the less likely a waiver would apply, with the court concluding  that ‘suits like the plaintiffs’ are likely to impose considerable costs upon IFC without providing commensurate benefits. Hence, IFC has not waived its immunity to this suit’ (First Instance decision, pp 11-12).

The Appeal decision: The disjunction of State and IO immunities

On 23 June 2017, in Jam et al v International Finance Corporation, the US Court of Appeals for the District of Columbia Circuit dismissed the plaintiffs’ appeal (‘Appeal Decision’). Despite noting that the IFC did not deny the harm caused to the plaintiffs, two out of the three judges endorsed the reasoning of the District Court (Appeal decision, p.1). Although concurring with the result, Judge Pillard’s reasons deserve a special focus for they demonstrate the need for a fundamental shift in approach. Judge Pillard observed: ‘I agree that Atkinson and Mendaro, which remain binding law in this circuit, control this case. I write separately to note that those decisions have left the law of international organizations’ immunity in a perplexing state. I believe both cases were wrongly decided, and our circuit may wish to revisit them.’ (Appeal decision, p. 11)

First, on the question whether the FSIA’s restrictive doctrine of State immunity should also apply to IOs, preferring the approach taken by the Third Circuit, in the Appeal Decision, Judge Pillard said:

We took a wrong turn in Atkinson when we read the IOIA to grant international organizations a static, absolute immunity that is, by now, not at all the same “as is enjoyed by foreign governments,” but substantially broader. Reading the IOIA to dynamically link organizations’ immunity to that of their member states makes sense. The contrary view we adopted in Atkinson appears to allow states, subject to suit under the commercial activity exception of the FSIA, to carry on commercial activities with immunity through international organizations…Were I not bound by Atkinson, I would hold that international organizations’ immunity under the IOIA is the same as the immunity enjoyed by foreign states. Accord OSS Nokalva, Inc. v. European Space Agency, 617 F.3d 756, 762¬64 (3d Cir. 2010) (declining to follow Atkinson and holding that restricted immunity as codified in the FSIA, including its commercial activity exception, applies to international organizations under the IOIA).

Judge Pillard’s preference for restricting IO immunity may yield some benefits for plaintiffs seeking justice against international organisations in the US. However, if eventually adopted, it should be limited to the specific context of the IOIA. Brief remarks on the issue are warranted here.

It is unfortunate that the IOIA links the immunities of IOs to that of states. The basis of state and IO immunity is different. It is well known that state immunity is a basic tenet of international law justified by sovereign equality, whereas IO immunities are treaty-based and functional. Equating state immunity with IO immunity confuses the regime. The immunity of an IO is to be determined with reference to its functions, which may be commercial in nature, as is the case with the multilateral development banks. That is perhaps the reason why the express waiver is written into the IFC’s Articles of Agreement. Thus, the scope of IO immunities ought to be determined by reference to their underlying rationale – functional necessity. Indeed, the problematic nature of using the commercial v non-commercial exception to assess the IFC’s immunity was pointed out by Judge Silberman (with Judge Edwards agreeing):

Ironically, the line of cases applying Mendaro ended up tying waiver to commercial transactions, so there is a superficial similarity to the commercial activities test that appellants would urge us to accept. But whatever the scope of the commercial activities exception to sovereign immunity, that standard is necessarily broader than the Mendaro test; if that exception applied to the IFC, the organization would never retain immunity since its operations are solely “commercial,” i.e., the IFC does not undertake any “sovereign” activities (Appeal Decision, p. 7).

Clearly, the linking of state immunity with IO immunity in the IOIA is a structural problem that remains. Be that as it may, the judiciary must work within the statutory framework. Judge Pillard’s views highlight the bizarre state of affairs where states, the creators of IOs, possess less immunities than IOs. When seen in that light, and taking into account the unsatisfactory structural framework of the IOIA, one can have much sympathy for Judge Pillard’s call for a new approach.

Importantly though, in the case at hand, it is the operation of the IFC’s express waiver provision that ought to be closely scrutinised. According to the criterion set out by the corresponding benefit test, the fundamental question is whether the IFC would receive a sufficient enough benefit should its immunity be restricted, and the plaintiffs’ claims be allowed to proceed to a hearing of the merits? Interestingly, while Judge Silberman (with Judge Edwards agreeing) acknowledged that ‘the term “benefit” is something of a misnomer’, it was concluded that the plaintiffs could not satisfy the substance of the corresponding benefit test set out in Mendaro (Appeal Decision, p. 9). However, Judge Pillard sought to demonstrate the arbitrariness of the corresponding benefit test, stating: ‘It is not entirely clear why we have drawn the particular line we have pursuant to Mendaro. Why are suits by a consultant, a potential investor, and a corporate borrower in an international organization’s interest, but suits by employees and their dependents not?’ (Decision of Judge Pillard in Appeal decision, p. 7)

Some brief conclusions

Ultimately, the concept of a ‘corresponding benefit’ is an attempt by a national judiciary to undo the consequences of problematic treaty language. The Court speculates whether or not a particular waiver will promote or restrict the goals or functions of an IO. It weighs the costs and benefits of any waiver of immunity to the IO. Overall, the more the courts are required to pass judgment on the internal operations and policies of an IO, the less likely it is that a waiver would apply. Critically though, the key problem with the corresponding benefit test is that costs and benefits can be perceived in different ways depending on the preferences of individual judges. Further, any suit against an IO will impact on the internal decision making of an IO in some way. The question will always be of degrees. So far, the courts seem to have addressed the risk of arbitrariness by reading the exception as narrowly as possible, with claims raising issues that would cause any intrusion into an IO’s policy sphere jurisdictionally barred. This narrow approach results in a denial of justice to already vulnerable plaintiffs.

As is demonstrated   by the case at hand, the IFC does not deny the harm caused to the plaintiffs by breach of the loan agreement. Further, it is apparent that the IFC disregarded the recommendations of the CAO, its very own accountability mechanism. In such circumstances, the effect of the decision of the national courts is that the plaintiffs, despite bringing a claim in the home forum of the IFC, have been denied justice and a fair hearing of their claims, which on all accounts appear to have significant merit. The plaintiffs have suffered grave injury, and their injury has now been compounded by denial of access to a remedy. The present case is another example of IO immunities prevailing over access to justice. Significantly, access to justice concerns have clearly seemed too impact Judge Pillard’s views:

Our cases seem to construe charter-document immunity waivers to allow suits only by commercial parties likely to be repeat players, or by parties with substantial bargaining power. But the opposite would make more sense: Entities doing regular business with international organizations can write waivers of immunity into their contracts with the organizations…Sophisticated commercial actors that fail to bargain for such terms are surely less entitled to benefit from broad immunity waivers than victims of torts or takings who lacked any bargaining opportunity, or unsophisticated parties unlikely to anticipate and bargain around an immunity bar (Decision of Judge Pillard in Appeal decision, p. 7).

Jam et al v International Finance Corporation joins a long line of cases where individuals suffering grave harm have been denied justice both at the institutional and the national level. Had the IFC positively responded to the CAO’s recommendations, national proceedings would not have been needed. But where justice is denied within the institutional legal order, considerations around the perceived independence of an organisation continue to trump the individual right to access justice generally, and a fair trial specifically. It is most unfortunate that international organisations such as the IFC who purport to deliver ‘lasting solutions for development’ choose to engage in conduct that leads to grave denials of justice to already vulnerable people who ought to be the very beneficiaries of their development activity.

The Conundrum of International Organisations Immunities: Jam et al v International Finance Corporation  (Part 1 of 2)

by Rishi Gulati

[Rishi Gulati is a Barrister at the Victorian Bar, Australia; Dickson Poon Scholar of Law at King’s College London; and Academic Expert, Bretton Woods Law, London]

Following the conclusion of the much discussed Haiti Cholera Class Action in US courts, the immunities of international organisations (IOs) have again been tested in the courts of that country in claims filed against the International Finance Corporation (‘IFC’), a financial institution within the World Bank Group. This is the first of a two part post. In Part I, I canvass some general matters around the IFC’s immunities, and provide the context to the litigation. In Part II of this post, I will discuss the jurisprudence arising out of this litigation so far.

The IFC’s immunities under the IOIA

The provisions of the US International Organisations Immunities Act 1945 (‘IOIA’) provide the applicable immunities regime for the IFC. Pursuant to 22 USC § 288a(b) of the IOIA, ‘[i]nternational organizations, their property, and their assets, wherever located and by whomsoever held, shall enjoy the same immunity from suit and every form of judicial process as enjoyed by foreign governments, except to the extent that such organizations may expressly waive their immunity for the purpose of any proceeding or by the terms of any contract’.

With international financial institutions such as the IFC essentially operating in the commercial marketplace, their founding instruments expressly allow for waivers for immunities from suits brought by private litigants in certain types of claims. In the case of the IFC, Article VI, Section 1 of the IFC’s Articles of Agreement grants immunities in classical functional terms. However, Article VI, Section 3 contains an express waiver, stating:

Actions may be brought against the Corporation only in a court of competent jurisdiction in the territories of a member in which the Corporation has an office, has appointed an agent for the purpose of accepting service of process, or has issued or guaranteed securities. No actions shall, however, be brought by members or persons acting for or deriving claims from members…

On its face, the express waiver to the IFC’s immunity is broad. According to the plain meaning of the words, the functional immunities of the IFC or similarly situated organisations will have little work to do. Indeed, courts have referred to such treaty language as ‘somewhat clumsy and inartfully drafted’ (Mendaro v World Bank, 717 F2d 610 (DC Cir 1983) (‘Mendaro’). Such criticism is well justified. This is because the language used in the express waiver contained in Article VI, Section 3 seemingly provides for a blanket waiver of immunity in respect of all suits save for the ones expressly prohibited. Should the words used be given effect to in accordance with their plain meaning, the IFC, and similarly situated organisations, would lose  their immunities even in respect of claims arising out of their functional activity. Thus defeating the very grant of functional immunities in the first place. To deal with this anomaly, courts have employed creative but unsatisfactory ways of addressing the challenge presented. The recent case of Jam et al v International Finance Corporation (.pdf), considered below, is just one example where the limits of the express waiver of immunities has been tested. The 24 March 2016 decision of the US District Court for the District of Columbia is here (.pdf) (‘First Instance decision’); and the decision of the US Court of Appeals for the District of Columbia Circuit delivered recently on 23 June 2017 is here. Note again, I discuss both these decisions in Part II of this post.

Jam et al v International Finance Corporation – a background

The IFC loaned USD 450 million to Coastal Gujarat Power Limited (‘CGPL’), a subsidiary of Tata Power. The loan was towards the construction of the coal-fired Tata Mundra Power Plant in India (the ‘Plant’). Critically, though, the IFC remains responsible for the monitoring and supervision of CGPL. This is to ensure CGPL’s compliance with internal IFC policies, including those set out in its ‘Performance Standards on Environmental and Social Sustainability’ (‘IFC Standards’). Those Standards established a mechanism for ‘the assessment, avoidance, minimization, and mitigation of environmental and social risks’ (First Instance decision, p. 3). The detailed facts are set out in the First Instance decision here (.pdf).

In summary, the complainants, a group of Indian nationals, who live, fish and farm within proximity of the Plant allege that the main legacy of the Plant has been ’environmental and social harm—to the marine ecosystem, to the quality of the air, to plaintiffs’ health, and to their way of life’ (First Instance decision, p. 1). Seeking justice within the IFC’s institutional structures, a group of Indian persons filed a complaint with the IFC’s Compliance Advisor Ombudsman (‘CAO’), the IFC’s ‘independent recourse and accountability mechanism’. Tellingly, with respect to the complaint of the affected individuals in the Plant, as was pointed out at First Instance, the CAO concluded that: ‘IFC had failed adequately to consider the environmental and social risks to which plaintiffs would be exposed as a result of the Plant’s development’; and that in ‘the CAO’s estimation, IFC then compounded that error by failing to perform an environmental and social impact assessment “commensurate with project risk,” and by failing to “address [subsequent] compliance issues during [project] supervision.”’

Ultimately, it is apparent that the CAO was not satisfied with the steps taken by the IFC to address the plaintiffs’ grievances. Nevertheless, as the CAO does not have enforcement powers, its views were not binding on the IFC, and the grievance remained unaddressed (First Instance decision, p. 4).

The plaintiffs finally resorted to the national courts, contending that the IFC was liable for the injuries caused by failing to comply with its own policies and standards, and failing to enforce the terms of its loan agreement with the CGPL. The plaintiffs sought equitable relief or, alternatively, compensatory and punitive damages before the US District Court for the District of Columbia (First Instance decision, p. 1). More specifically, the plaintiffs contended that ‘the irresponsible and negligent conduct of the International Finance Corporation in appraising, financing, advising, supervising and monitoring its significant loan to enable the development of the Tata Mundra Project in Gujarat, India’ caused them injury. The plaintiffs pursued claims in ‘negligence, negligent supervision, public nuisance, private nuisance, trespass, and breach of contract.’ (First Instance decision, p. 5).

More to follow in Part II of this entry.

The ICC Appeals Chamber Was Not Wrong (But Could Have Been More Right) in Ntaganda

by Luigi Prosperi

[Luigi Prosperi received his PhD in “International legal order and human rights” from Sapienza University of Rome, honorary fellow in International Law at Sapienza University, formerly Associate Legal Officer at ICTY.]

Last week, the Appeals Chamber (AC) of the ICC unanimously rejected Bosco Ntaganda’s Defence appeal against the 4 January decision of the Trial Chamber VI (TC), which had found that the Court has jurisdiction over the war crimes of rape and sexual slavery (provided for in Article 8(2)(b)(xxii) and 8(2)(e)(vi) of the Statute) when the perpetrators and the victims are members of the same armed force or group.

The decision caused a stir among several prominent ICL experts, and was harshly criticized in this post by Professor Heller. The aim of this post is not so much to respond to their criticisms, but to shed some light on a fundamental step in the AC’s reasoning, which I will refer to as the “systematic argument”. Since the AC’s systematic approach reverberates throughout the decision, in this blog post I will present the arguments in the same order as the judges.

First of all, the AC identifies the scope of the appeal:

  1. The principal issue arising in this appeal is whether the Trial Chamber erred in law when it held that victims of the war crimes of rape and sexual slavery listed in article 8(2)(b) and (e) do not have to be “protected persons” in the sense of the Geneva Conventions of 1949 (“Geneva Conventions”) or “[p]ersons taking no active part in the hostilities” in the sense of Common Article 3 to the 1949 Geneva Conventions (“Common Article 3”) (…).

Interestingly, the first argument wielded by the TC against this conclusion had been a systematic one. According to the judges, provided that “the Statute is first and foremost a multilateral treaty which acts as an international criminal code for the parties to it”, in order to determine if the crimes of rape and sexual slavery provided for in Article 8(2)(b)(xxii) and 8(2)(e)(vi) are limited to acts constituting grave breaches of the Geneva Conventions or serious violations of Common Article 3, they have first to consider the “statutory framework”, and then eventually turn to the laws and customs of international and non-international armed conflicts (TC decision, paras. 35-39).

Even though it upholds this systematic argument, the AC does not refer to the TC’s interpretation of the law in this regard, but proposes a different reading. On the one hand, the AC subscribes to the TC’s findings that neither the chapeaux of Article 8(2)(b) or (e), nor the two provisions criminalizing rape and sexual slavery expressly circumscribe the group of potential victims (para. 46), and that only in relation to “other form[s] of sexual violence” the Statute makes an express reference to grave breaches of Geneva Conventions or serious violations of Common Article 3, with the aim of requiring that the unenumerated conducts meet a “gravity threshold” (para. 49). On the other hand, the AC rejects the theory that including the “Status Requirements” would run contrary to the structure of Article 8, and render the world “other” in the chapeaux of Article 8(2)(b) and (e) meaningless (para. 47). It also finds no indications that the drafters intended to avoid any potential overlaps between the different categories of war crimes listed in the sub-paragraphs of Article 8(2) (para. 48).

However, at para. 48 the AC supports one of the TC’s main findings in relation to the structure of Article 8, and holds that:

while the drafting history is silent as to whether the drafters intended the war crimes of rape and sexual slavery under article 8(2)(b)(xxii) and (e)(vi) to be subject to the Status Requirements, it is clear that the drafters intended these crimes to be “distinct war crimes”, as opposed to merely illustrations of grave breaches of the Geneva Conventions or violations of Common Article 3.

Accordingly, the AC concludes that it is not possible to infer that the drafters intended to circumscribe the protection afforded under Article 8(2)(b)(xxii) and (e)(vi) to a particular group of potential victims. Therefore, the TC did not err in law in finding that, based the ordinary meaning, context and drafting history of the above-mentioned provisions, the statutory framework does not require the victims of the crimes of rape and sexual slavery to be “protected persons” in terms of the Geneva Conventions or “persons taking no active part in the hostilities” in terms of Common Article 3 (paras. 50-51).

In other words: given that the Statute is a multilateral treaty which acts as an international criminal code for its parties, the judges are bound to apply its provisions in light of Article 21 and of the rules of treaty interpretation before turning to secondary sources. It is in fact the ordinary meaning to be given to the terms of Article 8(2)(b) and (e) in their context, together with the drafting history of these provisions, that militate in favour of the view that under the Rome Statute, rape and sexual slavery may amount to war crimes when committed against members of the same armed group as their perpetrator.

According to this very principle, the AC has to consider whether any limitations in relation to the Status Requirements may arise from the broader international legal framework.

Pursuant to Article 21 of the Statute, in the absence of any indication to the contrary, the AC considers that the reference to “the established framework of international law” permits recourse to customary and conventional international law “regardless of whether any lacuna exists” in that provisions (para. 53). It is because of this reference (and not in light of a duty to apply Article 8 consistently with pre-existing norms of customary or conventional international humanitarian law) that the Court may introduce additional elements in respect of a given war crime listed in its sub-paragraphs, if required under customary or conventional international law (para. 54).

In relation to the Status Requirements, it should therefore be demonstrated that as a general rule, international humanitarian law circumscribes the group of potential victims of war crimes to persons protected under the Geneva Conventions or persons not taking active part in hostilities under Common Article 3, or that a specific exclusionary rule exists in relation to the war crimes of rape and sexual slavery. In other words, the AC finds that in order to determine that the Court has no jurisdiction over such crimes, it has to be demonstrated that Geneva Conventions and Common Article 3 categorically exclude the members of the same armed group from protection, either in general or in relation to the crimes of rape and sexual slavery.

In this regard, the AC argues that on the one hand, Geneva Conventions I and II respectively afford protection to the wounded and sick on land, and the wounded, sick and shipwrecked at sea, irrespective of their belonging to enemy armed forces or to a party’s own armed forces (para. 59); and on the other, Common Article 3 “provides for unqualified protection against inhumane treatment irrespective of a person’s affiliation” – unless (s)he is taking active part in the hostilities (para. 60).

Moreover, the AC relies upon the updated commentary of the ICRC, which observes that Common Article 3 “protects members of armed forces against violations committed by the armed force to which they belong”, referring as supporting practice to paras. 78-80 of the PTC II decision on the confirmation of charges issued in this very case (para. 61).

Accordingly, the AC finds that “international humanitarian law does not contain a general rule that categorically excludes members of an armed group from protection against crimes committed by members of the same armed group” (para. 63).

As to the argument that an exclusionary rule would be specifically applicable to the war crimes of rape and sexual slavery, the AC upholds the TC’s finding that despite most of the express prohibitions appear in contexts protecting civilians and hors de combat in the power of a party to the conflict, the prohibitions of rape and sexual slavery in armed conflict are well established under conventional (see Article 75 of AP I and Article 4(2)(f) of AP II) and customary international humanitarian law (see Rules 93 and 94 of the ICRC Study on Customary IHL). The AC also agrees that “there is never a justification to engage in sexual violence against any person” in the context of armed conflicts (para. 65). Therefore, the AC concludes that there is “no reason to introduce Status Requirements” to Article 8(2)(b)(xxii) and 8(2)(e)(vi) of the Statute “on the basis of the ‘established framework of international law’ ” (para. 66).

This is indeed the only part of the decision which I find slightly disappointing, since it could have been based on a more solid reasoning.

In particular, having taken a strong stance on the Rome Statute’s nature, I would expect the AC to thoroughly apply the rules on interpretation, as provided for in Article 21. Pursuant to Article 21(3) of the Rome Statute and Article 31(1) of the Vienna Convention on the Law of Treaties, Article 8 shall in fact be interpreted also in light of “internationally recognized human rights” and of the object and purpose of the Rome Statute – or, as the Prosecution had put it before the PTC, pursuant to “a purposive or teleological interpretation of article 8(2)(e)(vi), consistent with the protective rules applicable to children during warfare” (Prosecution submissions, para. 183).

Indeed, I find it somehow disappointing that the AC did not refer to the Statute’s object and scope (to fight impunity “for the sake of present and future generations”), nor to the “internationally recognized human rights” – including everybody’s right not to be victim of rape and sexual slavery. In this spirit, it is of the utmost relevance the fact that under international human rights law, there is never a justification to engage in sexual violence against any person, irrespective of this person’s status.

Moreover, the AC may have held that the Court should especially exercise its jurisdiction over rape and sexual slavery committed against members of the same armed group, when such status had been determined by the perpetrator’s unlawful conduct. In this regard, the AC could have enhanced the TC’s references to the duty not to recognise situations created by certain serious breaches of international law, and to the principle that one cannot benefit from one own’s unlawful conduct (TC decision, para. 53). These considerations may in fact contribute to defining, in the specific circumstances of the case, the “purpose” of fighting against impunity (of individuals that had enlisted the same underage kids they would abuse of).

In other words: provided that international humanitarian law does not generally and categorically prohibit the prosecution of those conducts, it is the evolutionary nature of the Rome Statute, which should at any time be interpreted in light of international human rights law, to justify the adoption of the broadest possible interpretation of Article 8(2)(b)(xxii) and 8(2)(e)(vi).

In order to prevent a likely objection in relation to the (unwanted) effects of this decision, in the final paragraphs the AC underlines that it does not blur in any way the distinction between war crimes and ordinary crimes:

  1. The Appeals Chamber emphasises in this context that the Elements of Crime for each war crime contain an express nexus requirement which must be established in each particular instance. Thus, it must be established that the conduct in question “took place in the context of and was associated with an armed conflict” of either international or non-international character. In the view of the Appeals Chamber, it is this nexus requirement, and not the purported Status Requirement, that sufficiently and appropriately delineates war crimes from ordinary crimes. To that end, as rightly observed in the Impugned Decision with reference to the judgment of the ICTY Appeals Chamber in Kunarac, the Trial Chamber may have regard, inter alia, to “the fact that the perpetrator is a combatant; the fact that the victim is a non-combatant; the fact that the victim is a member of the opposing party; the fact that the act may be said to serve the ultimate goal of a military campaign; and the fact that the crime is committed as part of or in the context of the perpetrator’s official duties.” The Appeals Chamber considers that any undue expansion of the reach of the law of war crimes can be effectively prevented by a rigorous application of the nexus requirement.

According to the AC, any undue expansion of the reach of the law of war crimes may in fact be prevented by a rigorous application of the principles concerning the nexus requirement. It is in fact the absence of a link to the armed conflict situation (which according to the same Kuranac AC consists in the fact that the crime “is shaped by or dependent upon” the armed conflict) that would put hundreds of potential war crimes (of rape and sexual slavery) back into Pandora’s box.

Finally, a further consideration has to be made in relation to the impact of this decision.

Considering that it is jurisdictional in nature, and that the ICC judges are not bound by a principle of precedent, it is foreseeable that the saga is far from over. In the meantime, it is noteworthy that Prosecution witnesses have testified about the child soldiers’ status and their participation to the UPC military operations.

No Justice for Sri Lankan Male Survivors

by Charu Hogg

[Charu Lata Hogg is the Director of the All Survivors Project at the UCLA School of Law and an Associate Fellow, Asia Program, Chatham House. This post is based on primary research conducted by Mirak Raheem, Deanne Uyangoda and Marisa De’Silva in Sri Lanka.]

Eight years since Sri Lanka’s nearly three-decade conflict came to a brutal end following the defeat of the Liberation Tigers of Tamil Eelam, the country is still coming to grips with the legacy of massive human rights violations committed by all sides. Egregious abuses by state security forces and armed groups also took place during other insurgencies in Sri Lanka, specifically during the armed insurrections in the south by the Marxist Janatha Vimukthi Peramuna (JVP) first in 1971 and then 1987-1989. However, sexual violence against men and boys, particularly in the context of detention under a previous infamous security legislation, has only recently begun to be recognised as among the numerous abuses that have taken place within the context of the armed conflict. A proposed Counter Terrorism Act (CTA) which is awaiting Parliamentary approval threatens to continue prolonged detention without charge. If approved this counter-terrorism bill could continue to facilitate human rights abuses in detention.

A 2015 report resulting from UN investigations concluded that men were as likely to be victims of sexual violence as women within the context of detention (The investigations were mandated by the Human Rights Council (Resolution 25/1) to investigate allegations of violations of international law following the breakdown of the 2002 Ceasefire Agreement to the end of the conflict. See, Report of the Office of the High Commissioner for Human Rights Investigation on Sri Lanka (OISL)), but the issue remains largely hidden and stigmatized. At the core of this silence is the lack of legal protection for men and boys against sexual violence which informs social attitudes and contributes to an environment in which violations can take place with virtual impunity. In addition to law reform, a fundamental gap in knowledge and expertise on preventing, investigating and responding to sexual violence constitutes a significant barrier. Training of all relevant government, official and professional stakeholders including members of law enforcement and security forces, the judiciary, national human rights institutions, medical and mental health professionals and humanitarian workers could provide a crucial first step.

In numerous cases of sexual violence against men and boys documented by the UN and other international organisations, victims were detained under the 1979 Prevention of Terrorism Act (PTA) which has been widely used to detain LTTE suspects and under which many fundamental rights and guarantees to protect detainees are suspended. There have been repeated calls for its repeal and the new government which came into power in 2015 has promised to do so. Yet, the PTA remains on the statute books and continues to be applied, permitting the administrative detention for up to 18 months with only limited judicial supervision, irregular access to legal representation, and without the possibility of challenging the legality of detention. The PTA also creates an environment in which forced confessions are effectively encouraged by permitting statements made to police officers at any time while in custody to be admitted in court as evidence, with the burden placed on the accused to prove that the confession was extracted under duress.

A complex set of factors prohibit male victims of sexual violence in Sri Lanka from coming forward to report this violation. But even if they did, Sri Lanka does not recognise the possibility of male rape under the law. Rather men are defined only as perpetrators of rape. Article 363 of the Penal Code states “a man is said to commit rape who has sexual intercourse with a woman under any of the following descriptions…” (See Section 363 of the Penal Code (Ordinance No. 2 of 1883 as amended). Similarly, the prohibition of statutory rape applies only to girls under the age of 16 years and not to boys (Section 363(e) of the Penal Code in describing conditions for statutory rape refers to “…with or without her consent when she is under sixteen years of age).) Similarly, the prohibition of statutory rape applies only to girls (under the age of 16 years) and not to boys (Section 363(e) in describing conditions for statutory rape refers to “…with or without her consent when she is under sixteen years of age” [emphasis added]). Other provisions under the Penal Code mischaracterize or define sexual violence in a way that they do not reflect the lived experience of survivors, are inconsistent with the more inclusive, gender-neutral definitions under international law, or are otherwise inadequate for prosecuting sexual violence against males.

This legal discrimination against men and boys which is so acutely reflected in Sri Lankan law also permeates other institutions and sectors and contributes to the pervasive lack of medical, counselling and other support services available to male survivors. While services for women are equally inadequate, increased attention on this issue during the post-war phase has triggered a recognition of this issue which remains entirely absent in the case of male survivors. Some medical teaching institutions continue to put forward the understanding that only women can be raped and do not recognise the rape of men. Other representatives of the humanitarian community have admitted to All Survivors Project that sexual violence against men is neither monitored nor responded to during the conflict or outside of it. Added to this is the widespread discrimination against homosexuals which is prescribed in law and the criminalization of consensual same sex acts which may also discourage male survivors from reporting or accessing services for fear that they may be accused of homosexual activity. Specifically, Sections 365 and 365A criminalize certain homosexual acts categorizing them as ‘unnatural offences’. These provisions have been used to persecute members of the LGBTIQ community and serve to reinforce discriminatory gender stereotypes.

During our research, (more…)

William Bradford Fails Upward — and Is Still Lying About His Credentials

by Kevin Jon Heller

When last we met William Bradford, he had just published an article in the National Security Law Journal (NSLJ) accusing centrist national-security-law professors of treason and advocating prosecuting them for providing material support to terrorists. After many scholars, including me, pointed out that the article was both absurd and deeply offensive, the NSLJ repudiated the article. (Alas, the journal has since scrubbed the repudiation from its website.)

Bradford’s article was not his first brush with controversy He was forced to resign from Indiana University at Indianapolis after Inside Higher Education revealed that he had lied about his military service, falsely claiming, inter alia, that he had fought in Desert Storm and Bosnia and had won a Silver Star. Bradford then later resigned from West Point — whose decision to hire him still boggles the mind — after it came to light that he had falsely claimed that he had been an assistant professor at the National Defense University (NDU), run by the Department of Defense. According to the NDU, to quote the Guardian, “he was not a professor there, nor even a staff employee…. He is said to have worked for a Waynesboro, Virginia-based translations and business consultant, Translang, which had a contract with the university.”

You would be forgiven for thinking that someone who has accused respected law professors of committing treason and who was forced to resign from two academic institutions for lying about his credentials might have a difficult time finding a new — and more important — position. But if you do think that, you have never met Donald J. Trump, for whom no one is too dishonest or too incompetent to hire. Because Trump has recently appointed Bradford to the be the Director of the Office of Indian Energy at the Department of Energy (DoE).

That’s appalling in and of itself. But the awfulness doesn’t end there, because Bradford is still lying about his credentials. Here is a screenshot of Bradford’s bio on the DoE website (in case the DoE reads this and decides to scrub it):

Notice the text inside the red rectangle: Bradford is still claiming to have been a faculty member at the NDU — the same claim that led to his resignation from West Point.

In any sane administration, Bradford would be fired in the next 48 hours. But this is the Trump administration, so I’m not holding my breath.

The Downing of the Syrian Fighter Jet and Collective Self-Defence

by Kinga Tibori-Szabó

[Kinga Tibori-Szabó is the author of the monograph Anticipatory Action in Self-Defence, which won the 2012 Francis Lieber Prize. She currently works at the Kosovo Specialist Chambers. The views expressed herein are those of the author and do not necessarily reflect the views of the employing organization.]

On 18 June 2017, a US Navy Super Hornet shot down an SU-22 fighter jet of the Syrian government south of Tabqah, in Syria. The US-led anti-ISIS Coalition’s statement indicated that the Syrian fighter jet was shot down after it dropped bombs near Syrian Democratic Forces (SDF) positions in the area. The Coalition statement averred that the action was taken

“in accordance with rules of engagement and in collective self-defense of Coalition partnered forces”.

Let’s analyse that statement. The action seems to have been a standard exercise of unit self-defence, a tactical level application of national or collective self-defence. It allows commanders to defend their unit, other units of their armed forces as well as other specified units (friendly forces) against hostile acts or hostile intent (see annex D of the San Remo Rules of Engagement (ROE) Handbook). But as aptly discussed in a post on Opinio Juris, unit self-defence should be viewed as a ROE concept that has underlying legal consequences but is not a legal term in and of itself. So the fact that the downing of the Syrian aircraft was in accordance with ROE does not answer the question of its legal basis in jus ad bellum. That brings us to the second part of the statement, the reference to collective self-defence.

According to the statement, the Coalition acted in collective self-defence to protect the SDF against an attack by the Syrian regime without itself seeking to fight that regime.

First, a few points need to be made regarding the reference to collective self-defence to underline that the Coalition does not seek to fight the Syrian regime.

  • The jus ad bellum basis of Coalition actions on Syrian territory has attracted criticism because of reliance on a controversial interpretation of collective self-defence. In a nutshell, the request of Iraq for the assistance of the international community to fight ISIS was interpreted as a legal basis for the exercise of collective self-defence not only on the territory of Iraq, but also on the territory of Syria (see opinions on this here, here and here)
  • Be that as it may, the Coalition has repeatedly emphasized and the quoted statement expressly states that “[t]he Coalition does not seek to fight [the] Syrian regime”. This cautious approach was somewhat overshadowed by the evident US support for ‘vetted Syrian opposition forces’, but until recently, the US and its Coalition partners had avoided direct confrontations with the Syrian regime. However, the US strikes against the Syrian airbase in early April amounted to use of force against the Syrian regime and have initiated an international armed conflict between the two states in the sense of Common Article 2. The downing of an Iranian UAV in early June and of a second one two days after the shooting of the Syrian aircraft made it clear that these confrontations are no longer isolated incidents, but clashes in the context of an international armed conflict between Syria and the leader of the Coalition (US).
  • Nonetheless, the existence of an armed conflict between the US and Syria is a separate issue from the question whether (collective) self-defence can be invoked by the Coalition as a justification for some of the clashes within the context of that armed conflict. Simply put, the invocation of collective self-defence in the Coalition statement does not protect the US from the obvious: that it is now in armed conflict with the Syrian regime and potentially with its allies.

Second, the exercise of collective self-defence requires, in addition to the core conditions of self-defence, at least a request for assistance from a state. This is unsurprising, as the right of self-defence enshrined in Article 51 of the UN Charter pertains to states rather than armed groups. While it is widely accepted that an armed attack may be carried out by an armed group and self-defence may be exercised against such a group, the possibility of armed groups having the right to invoke self-defence is not contemplated. Nonetheless, it seems that the Coalition invoked collective self-defence to protect a ‘friendly’ armed group notwithstanding the fact that such a group does not have the right to invoke self-defence in the first place and request assistance on that basis. The question thus arises whether there is room for an expansive view on the right of self-defence permitting a state to invoke ‘collective’ self-defence on behalf of an armed group and assist the group on that basis. The wording of Article 51 of the UN Charter leads to a prima facie negative answer. Nonetheless, the proliferation of conflicts involving armed groups and the current reality of states defending such groups without themselves being the target of the same attack may bring this question to the fore and further complicate the interpretation and application of self-defence in relation to non-state actors.

ICC Appeals Chamber Says A War Crime Does Not Have to Violate IHL

by Kevin Jon Heller

One of the most basic assumption of ICL is that an act cannot be a war crime unless it violates a rule of international humanitarian law (IHL). Article 6(b) of the London Charter criminalised “War Crimes: namely, violations of the laws or customs of war.” Article 3 of the ICTY Statute provides that “[t]he International Tribunal shall have the power to prosecute persons violating the laws or customs of war,” while Article 4 of the ICTR Statute provides that “[t]he International Tribunal for Rwanda shall have the power to prosecute persons committing or ordering to be committed serious violations of Article 3 common to the Geneva Conventions of 12 August 1949 for the Protection of War Victims, and of Additional Protocol II thereto of 8 June 1977.” And Article 8 of the Rome Statute criminalises “[g]rave breaches of the Geneva Conventions of 12 August 1949”; “[o]ther serious violations of the laws and customs applicable in international armed conflict”; [i]n the case of an armed conflict not of an international character, serious violations of article 3 common to the four Geneva Conventions of 12 August 1949″; and “[o]ther serious violations of the laws and customs applicable in armed conflicts not of an international character.” In each and every case, war crimes are limited to violations of IHL.

No more. The Appeals Chamber (AC) at the ICC has just unanimously held in Ntaganda that a perpetrator can be convicted of a war crime even if his act does not violate IHL. That decision is not simply “unprecedented,” as the AC openly acknowledges. It is simply incorrect — as this post will demonstrate.

The judgement itself addresses allegations that Ntaganda is criminally responsible for two war crimes — rape and sexual slavery — involving children forcibly recruited into his organised armed group, the UPC/FPLC. Ntaganda challenged that allegation, arguing that “crimes committed by members of armed forces on members of the same armed force do not come within the jurisdiction of international humanitarian law nor within international criminal law.” The Trial Chamber (TC) disagreed, in a judgment ably discussed and critiqued by Yvonne McDermott. Ntaganda appealed, giving rise to this judgment. Here is the AC’s “key finding”:

2. Having regard to the established framework of international law, members of an armed force or group are not categorically excluded from protection against the war crimes of rape and sexual slavery under article 8 (2) (b) (xxii) and (2) (e) (vi) of the Statute when committed by members of the same armed force or group.

Before turning to the logic of the judgment, it is important to be very precise about the terms of my quarrel with the AC. I completely agree with the AC that there are situations in which a member of an armed force can, in fact, commit the war crime of rape or the war crime of sexual slavery against a member of the same armed force. As the AC rightly notes, although the Third and Fourth Geneva Conventions do not apply to acts committed by a combatant against someone from the same side of the conflict — whether by virtue of membership in that same armed force (GC III) or by nationality (GC IV) — the First and Second Geneva Conventions contain no such limitation:

59. In contrast, Geneva Conventions I and II, which protect the wounded and sick on land and the wounded, sick and shipwrecked at sea respectively, provide protection “in all circumstances […] without any adverse distinction founded on sex, race, nationality” and prohibit violence against them. Importantly, such protected status is not limited to persons belonging to enemy armed forces, but includes wounded, sick or shipwrecked members of a party’s own armed forces, a rule that corresponds to the understanding of the scope of protection since the first Geneva Convention was adopted in 1864. It follows from the above that the notion of grave breaches under Geneva Conventions I and II includes violations committed against the wounded, sick or shipwrecked committed by members of their own armed force.

Nothing in GC I or GC II suggests, however, that IHL protects all members of the armed forces against member-on-member violence. On the contrary, let’s take a look at the AC’s statement again, with the critical language in bold:

59. In contrast, Geneva Conventions I and II, which protect the wounded and sick on land and the wounded, sick and shipwrecked at sea respectively, provide protection “in all circumstances […] without any adverse distinction founded on sex, race, nationality” and prohibit violence against them. Importantly, such protected status is not limited to persons belonging to enemy armed forces, but includes wounded, sick or shipwrecked members of a party’s own armed forces, a rule that corresponds to the understanding of the scope of protection since the first Geneva Convention was adopted in 1864. It follows from the above that the notion of grave breaches under Geneva Conventions I and II includes violations committed against the wounded, sick or shipwrecked committed by members of their own armed force.

Under GC I and GC II, in other words, member-against-member violence violates IHL only if the victim is wounded, sick, or shipwrecked. If the victim is none of those things — if he or she is not hors de combat — that violence may well violate a state’s domestic criminal law, but it does not violate IHL.

If the AC had limited the scope of its judgment to rape and sexual slavery committed against child soldiers who were hors de combatdefined by the ICRC, in relevant part, as “anyone who is defenceless because of unconsciousness, shipwreck, wounds or sickness” — it would have been on firm ground. But that is not what it has done. On the contrary, the AC goes to great lengths to make clear that member-against-member rape and sexual slavery are war crimes even if the victim is an active combatant –– ie, one who is not hors de combat. Here is the relevant paragraph (emphasis mine):

64. With regard to the second issue – namely whether Status Requirements exist in international humanitarian law specifically for the war crimes of rape and sexual slavery – the Appeals Chamber observes that the prohibitions of rape and sexual slavery in armed conflict are without a doubt well established under international humanitarian law. As noted by the Trial Chamber, protection under international humanitarian law against such conduct generally “appear[s] in contexts protecting civilians and persons hors de combat in the power of a party to the conflict”. In this regard, the question arising before the Appeals Chamber is whether such explicit protection under international humanitarian law suggests any limits on who may be victims of such conduct. In the view of the Appeals Chamber, there is no conceivable reason for reaching such a conclusion.

Notice the bold language, because it’s critical — and wrong. IHL protection does not “generally” apply only to civilians and combatants hors de combat. On the contrary, each and every IHL convention applies only to those two categories of individuals. As we have seen, the AC itself acknowledges that limitation with regard to all four of the Geneva Conventions. It cites no other source of IHL, instead simply noting that the ICRC states in its new commentary to GC I “that Common Article 3 protects members of armed forces against violations committed by the armed force to which they belong.” But that statement is incomplete and misleading, because the ICRC makes unequivocally clear that CA3’s prohibitions apply only to individuals who are hors de combat:

518  Subparagraph (1) covers all ‘[p]ersons taking no active part in the hostilities, including members of armed forces who have laid down their arms and those placed hors de combat by sickness, wounds, detention, or any other cause’. The article does not expand on these notions and this part of the article did not give rise to much discussion at the 1949 Diplomatic Conference. The protection afforded under this subparagraph requires that the person be in the power of a Party to the conflict (see section E.4).
519  The protection of persons not or no longer participating in hostilities is at the heart of humanitarian law. The persons protected by common Article 3 are accordingly described by way of explicit delimitations: ‘persons taking no active part in the hostilities, including members of armed forces who have laid down their arms and those placed hors de combat by sickness, wounds, detention, or any other cause’ (emphasis added). Parties to a non-international armed conflict are under the categorical obligation to treat these persons humanely, in all circumstances and without any adverse distinction.

The Trial Chamber’s judgment is no better. The TC rests its conclusion that member-against-member rape is a war crime even when the victim is an active combatant solely on two things: the Martens Clause and Art. 75 of the First Additional Protocol (AP I). Here is paragraph 47:

While most of the express prohibitions of rape and sexual slavery under international humanitarian law appear in contexts protecting civilians and persons hors de combat in the power of a party to the conflict, the Chamber does not consider those explicit protections to exhaustively define, or indeed limit, the scope of the protection against such conduct. In this regard, the Chamber recalls the Martens clause, which mandates that in situations not covered by specific agreements, ‘civilians and combatants remain under the protection and authority of the principles of international law derived from established custom, from the principles of humanity and from the dictates of public conscience’. The Chamber additionally notes that the fundamental guarantees provisions [in Art. 75] refer to acts that ‘are and shall remain prohibited at any time and in any place whatsoever’ and as such apply to, and protect, all persons in the power of a Party to the conflict.

I don’t have time to get into a detailed discussion of the Martens Clause. Suffice it say here that it is very unlikely that the Clause can ever be relied upon to expand IHL not only beyond conventional law, but even beyond customary IHL — and as the AC itself acknowledges (para. 60), there is literally zero state practice indicating that member-against-member mistreatment is a war crime even when the victim is an active combatant. Even Antonio Cassese, no stranger to judicial activism, dismisses this “norm-creating” reading of the Martens Clause as “radical.” As he says, “[s]urely the Clause does not envisage — nor has it brought about the birth of — two autonomous sources of international law, distinct from the customary process.”

As for Art. 75 of AP I, the Protocol’s “fundamental guarantees” provision, the TC’s position is deeply problematic. Here is n. 111:

Article 75 of Additional Protocol I refers to ‘a Party to the conflict’ (emphasis added) and therefore does not limit the fundamental guarantees to persons in the power of the opposing party.

The TC conveniently fails to note that Art. 75 applies only to international armed conflict — and that Art. 4 of AP II, the “fundamental guarantees” provision in the NIAC Protocol, is specifically limited to “persons who do not take a direct part or who have ceased to take part in hostilities” (ie, civilians and combatants hors de combat).

Given that conventional IHL uniformly requires the victim of member-against-member mistreatment to be hors de combat, on what basis does the AC hold that the status of the victim is irrelevant? The answer comes from this paragraph (emphasis mine):

65. The Appeals Chamber agrees with the Trial Chamber’s finding that “there is never a justification to engage in sexual violence against any person; irrespective of whether or not this person may be liable to be targeted and killed under international humanitarian law”. Accordingly, in the absence of any general rule excluding members of armed forces from protection against violations by members of the same armed force, there is no ground for assuming the existence of such a rule specifically for the crimes of rape or sexual slavery.

This is simply incorrect. To begin with, there is a specific rule excluding active combatants from the war crimes of rape and sexual slavery in member-against-member situations: namely, the rule that says violence in member-against-member situations violates IHL only when the victim is hors de combat. The AC’s judgment suggests that states not only had to specify that rule in the various IHL conventions, they also had to add: “oh, and by the way, this limit means that mistreating active combatants doesn’t violate IHL.” But that’s silly: the former implies the latter. After all, expressio unius est exclusio alterius is a basic rule of treaty interpretation.

But even if that was not the case, there would still be a general rule excluding active combatants from the war crimes of rape and sexual slavery in member-against-member situations: the rule that says a war crime must involve a violation of IHL. As noted at the beginning of this post, that is one of the most basic assumptions of IHL. Not all violations of IHL are war crimes, but all war crimes are violations of IHL. So the burden of proof was not on Ntaganda to show that rape and sexual slavery cannot be war crimes in member-against-member situations if the victim is an active combatant. The burden was on the prosecution to prove that such acts actually violate IHL. Because if they don’t — and they don’t, as we have seen — the Court has no jurisdiction whatsoever over Ntaganda’s acts, at least insofar as they are legally characterised as war crimes.

In the end, the AC’s decision in Ntaganda is little more than the latest iteration of the Court’s willingness to rely on teleological reasoning when the Rome Statute does not protect victims as much as the judges think it should. No one is in favour of raping and sexually enslaving child soldiers. But the solution isn’t to detach the law of war crimes from its moorings in IHL by holding — if only implicitly — that an act can be a war crime even if it does not violate IHL. To do so is not only legally indefensible, it risks delegitimising both the Court and the law of war crimes itself.