Tag Archives: icc

ICC PTC authorises investigation in Bangladesh/Myanmar: some thoughts

Big news, came out of the ICC yesterday, a mere few days after the proceedings initiated by Gambia at the ICJ against Myanmar: Pre-Trial Chamber III has authorised the opening of an investigation in the situation of Bangladesh/Myanmar. This decision follows a decision in September 2018 where PTC I found that the Court could theoretically exercise jurisdiction in such a situation, despite Myanmar not being a State Party (I commented on that decision here) and the Prosecutor’s request last summer (which I reflected upon here).

There are a number of interesting takeaways from the decision which I want to briefly address here.

1) Assessment of evidence

Back in July, I raised the point that the OTP request did not contain an explanation of how it assessed available information. I found that problematic, especially given the challenges that exist when using open source material, such as NGO and UN reports.

Apparently, this is not a problem for the PTC, which provides no indication of its understanding of the standard of proof during a preliminary examination, nor does it provide any methodological explanation of how it independently assessed available information. Essentially, the PTC seems to have taken any “finding” in a human rights report at face-value, which, even at such an early stage of the proceedings, cannot be an adequate approach for a criminal court. 

2) The territorial exercise of jurisdiction question

In order to deal with the central issue here (the fact that the alleged incidents all took place on the territory of a non-State party, with the involvement of nationals of a non-State party), the PTC had to discuss several matters. I found this part of the decision actually quite pleasant and intellectually stimulating to read, but I must confess, I did not always get how its different parts fit one with another. It is as if several really interesting preparatory memos were copied and pasted together in the final decision without the final polishing work on making sure that it all fit in a coherent argumentation.

For example, the PTC had to decide what “conduct” in Article 12(2)(a) meant, which was important in order for the ICC to be able to satisfy the territorial precondition to the exercise of jurisdiction. Article 12(2)(a) of the Rome Statute provides that:

the Court may exercise its jurisdiction […] if The State on the territory of which the conduct in question occurred or, if the crime was committed on board a vessel or aircraft, the State of registration of that vessel or aircraft;

The PTC felt the need to discuss how the words “conduct” and the words “crime” in Article 12(2)(a), with the conclusion that “the notions of ‘conduct’ and ‘crime’ in article 12(2)(a) of the Statute have the same functional meaning” (par. 48).

This might be true, but it is not clear how this in fact helps us to understand what “conduct” actually means. More specifically, it does not really explain how the PTC immediately jumps to the conclusion in the next paragraph: ” For these reasons, the word is used in a factual sense, capturing the actus reus element underlying a crime subject to the jurisdiction ratione materiae of the Court”. Why should “conduct” and “actus reus” be equated in this way? This is not explained clearly.

Neither is there any explanation for the extraordinary logical leap from conduct, to actus reus, to consequences in par. 50-51:

50. Further, depending on the nature of the crime alleged, the actus reus element of conduct may encompass within its scope, the consequences of such conduct. For instance, the consequence of an act of killing is that the victim dies. Both facts concerning the act and the consequence (i.e. the killing and the death) are required to be established.

51. In respect of certain crimes within the Statute, the particular consequence may be that the victim behaves, or is caused to behave, in a certain way as a result of conduct attributable to the alleged perpetrator. The negative corollary is that, should those consequences not follow from the conduct of the perpetrator, the crime cannot be said to have occurred (although the suspect’s conduct may constitute attempt).

There is for me a counter-intuitive finding here, which flows from the automatic conflation between “conduct” and “actus reus”: that the consequences of a conduct should also be considered as conduct, through the magical mediation of “actus reus”. But this is far from obvious. Conduct, in any traditional meaning of the term, is what you actually do (or don’t do in the case of omissions), not its consequences. The fact that a certain consequence might be part of the actus reus elements of a crime does not automatically mean that it is technically “conduct” for the purposes of Article 12(2)(a).

The next step in the reasoning is equally puzzling in its construction. The PTC have to discuss how much of a “conduct” must take place on the territory of a State party to trigger Article 12(2)(a). Here, the Judges go into an interesting discussion on possible ways to interpret domestic practice on the exercise of criminal jurisdiction, with the ultimate conclusions that (par. 58):

first, under customary international law, States are free to assert territorial criminal jurisdiction, even if part of the criminal conduct takes place outside its territory, as long as there is a link with their territory. Second, States have a relatively wide margin of discretion to define the nature of this link.

In other words, this provides little help to actually interpret Article 12(2)(a), neither does it support the conclusions in the following paragraphs (par. 60-61):

since the States Parties did not explicitly restrict their delegation of the territoriality principle, they must be presumed to have transferred to the Court the same territorial jurisdiction as they have under international law.

The only clear limitation that follows from the wording of article 12(2)(a) of the Statute is that at least part of the conduct (i.e. the actus reus of the crime) must take place in the territory of a State Party. Accordingly, provided that part of the actus reus takes place within the territory of a State Party, the Court may thus exercise territorial jurisdiction within the limits prescribed by customary international law.

But how is there anything “clear” about the “wording of article 12(2)(a)” in that respect? Article 12(2)(a) makes no mention of “part of the conduct” or even “actus reus”! That is the whole point of having the discussion in the first place. If everything was that “clear”, there was no need to spend 8 pages on the matter in the decision…

Ultimately therefore, it is striking to note that there is in fact little support in the decision for the proposition that 1) conduct means actus reus of the crime, which also includes the consequences of conduct and 2) that it is sufficient for part of the actus reus of the crime to be committed on the territory of a State party for Article 12(2)(a) to be triggered.

In this sense, I wonder if there are not in fact strong reasons to have a restrictive understanding of the ICC’s exercise of jurisdiction, whether from the angle of the respect for sovereignty, or, more simply perhaps, from the angle of the application of the principle of strict interpretation of criminal law statutes (Article 22 of the RS). But this is too much to deal with here.

3) The material jurisdiction question

I don’t have much to say on the identified alleged crimes themselves (deportation and persecution).

In relation to deportation, I do note that the PTC chooses to not reopen the debate on whether it is a discrete crime under the Rome Statute (which has as a consequence that crossing a border becomes an element of the crime, which in turn allows for the exercise of territorial jurisdiction). I’m sure this matter will arise again in subsequent proceedings, as it should.

In relation to alleged persecution, the PTC’s way of linking it to Bangladesh (in order to trigger the territorial precondition to the exercise of jurisdiction) is not very developed. It simply states that: “The Chamber is further satisfied that the Prosecutor could reasonably believe that the alleged coercive conduct leading to the Rohingya’s deportation to Bangladesh was directed against an identifiable group or collectivity” (par. 109). This is a very minimalist explanation.

I note more particularly that the PTC stays away from discussing the “right of return” that had bee put forward in the initial request by the OTP. It would be premature at this stage to deduce from that a skepticism on the part of the judges for that argument, but it is interesting nonetheless that they chose not to address it.

Finally, on a broader level, if we accept for the purposes of the conversation that the PTC is right, this approach by the PTC, which particularly equates “consequences” to “conduct” for the purposes of the Article 12(2)(a), actually has an potential impact on the scope of the exercise of jurisdiction by the ICC which is not really explored in the decision. The PTC summarizes its position in the following (par. 124):

The Chamber recalls its determination regarding jurisdiction ratione loci where it found that the Court can exercise jurisdiction where a part of the actus reus of a crime within the jurisdiction of the Court is committed on the territory of a State Party. Consequently, the Chamber authorises the commencement of the investigation for crimes committed at least in part on the territory of Bangladesh.

With the following footnote 254: “For example if a person is shot on the territory of Myanmar and dies, as a result, on the territory of Bangladesh”.

But this logic, 1) bringing in consequences in the framework and 2) accepting that only part of the actus reus must be on the territory of a State Party, actually means that theoretically far more crimes could be brought into the discussion than the ones currently discussed in the decision (deportation and persecution). Indeed, if consequences are taken into account, by definition, the lasting consequences of all crimes will “follow” the victims into Bangladesh.

Moreover, pushing the logic even further, genocide (especially the “Deliberately inflicting on the group conditions of life calculated to bring about its physical destruction in whole or in part” aspect of it) could be arguably brought into the fray. In that respect, I had already noted the absence of such discussion in the initial OTP request. We will have to see how things progress during the actual investigation, especially given the broader context of the proceedings initiated by Gambia at the ICJ.

4) interests of justice

Obviously, as the first Article 15 decision to follow the Afghanistan decision, it was going to be interesting to see how it dealt with the question of “interests of justice”.

I noted, in relation to the OTP request, that the choice had apparently been made to simply ignore the Afghanistan decision and remarked that:

More discussion would appear all the more justified that this situation seems to fall squarely in the Afghanistan logic, given that it involves alleged conduct that took place entirely (putting the legal nicety of the « crossing of the border » criteria) on the territory of a non-State party by non-State party nationals, with virtually zero chance of cooperation. If there was ever a time to explicitly challenge the Afghanistan understanding of the « interests of justice » in a request, it was now.

Well it seems that the OTP adopted the “right” strategy, because the PTC’s handling of “interests of justice” is minimalist at best. The Judges devote juste 7 lines to it in the decision:

As regards the interests of justice, the Prosecutor has stated that she has identified no substantial reasons to believe that an investigation into the situation would not be in the interests of justice and the Chamber has no reason to disagree with this assessment. This view is reinforced by the fact that, according to the Registry’s Final Consolidated Report, ‘all victims representations state that the victims represented therein want the Prosecutor to start an investigation in the Situation.

The PTC has therefore also chosen to completely ignore the Afghanistan decision, makes no effort to define the “interests of justice”, the applicable legal and procedural framework, or how it applies to the current situation.

Given the importance of the upcoming appeal, this is quite surprising. In fact, one might wonder whether there should not have been some judicial restraint here and whether it should not have been preferable to not issue the decision until the matter is settled on appeal.

5) the scope of the authorised investigation

On this point there is also a serious departure from the Afghanistan precedent. The PTC makes it very clear that (par. 126):

the Chamber wishes to emphasise that the Prosecutor is not restricted to the incidents identified in the Request and the crimes set out in the present decision but may, on the basis of the evidence gathered during her investigation, extend her investigation to other crimes against humanity or other article 5 crimes, as long as they remain within the parameters of the authorised investigation. Similarly, the Prosecutor is also not restricted to the persons or groups of persons identified in the Request.

I don’t necessarily disagree with that (as I noted in my critique of the Afghanistan decision at the time), but what is striking here is that there is zero reference to the Afghanistan decision. It is simply ignored as if it doesn’t exist. which means that currently, on this point, we have conflicting case law from ICC Pre-Trial Chambers, which is obviously not very helpful for the future practice of the Court.

I also note that from a temporal perspective, the PTC says that “crimes of a continuous nature” can be investigated “even when such crimes commenced before 1 June 2010 (or the date of entry into force of the Statute for any other relevant State Party) in so far as the crimes continued after this date” (par. 132). Given the long history behind the current situation, this temporal expansion of the jurisdiction of the Court could have a serious impact on the scope of the investigation.

6) conclusion

Legally, this decision is not really satisfactory. Its argumentation on a number of key issues is somewhat unconvincing and would have deserved more discussion, especially in such a sensitive situation bringing in a non-State Party.

Whether the PTC is technically right or not, I must admit, from the beginning of this process, some uneasiness at how the Rome Statute framework is here being interpreted in a way that allows, based on an arguably tenuous link to Bangladesh, the Court to investigate events which should normally be beyond the natural jurisdiction of the Court: on the territory of a non-State Party, with the involvement of non-State party nationals.

Beyond the legal issues raised by the decision, one can legitimately raise questions about the opportunity for the OTP (and the Court) to get bogged down in what is likely not to be a very easy situation to navigate, at a time when the Court is regularly criticised for not delivering on its promise for justice (a completely unreasonable promise to be making in the first place, but that is a different matter). Yet, seeing the initial reaction by some NGOs, I realise that the market for hope is still going strong…

On the other hand, this decision does contribute to the “new” narrative that the ICC is moving away from Africa. FIDH for example insists here on the fact that it is the first official investigation in Asia (So, Georgia, while being categorized as an “eastern european state” in UN/ICC regional groupings, is technically in Asia, but we get the point). How successful this move away from Africa will be is another question…

ICC OTP requests authorisation to open an investigation into the Myanmar situation: three comments

Last Friday, 5 July 2019, the Prosecutor if the International Criminal Court filed a request under article 15 to be able to open an investigation into the situation of Myanmar. This is no surprise, as she had notified her intention to do so on the 12 June 2019, which led to the constitution of a Pre-Trial Chamber to that effect.

I don’t necessarily have too many comprehensive comments on the substance of the request at this points, but having read through the request, I wanted to put forward three initial thoughts I had, which mostly relate to choices in terms of drafting.

  • The jurisdictional issue

As those following these proceedings will know, the current request was preceded by a decision last September, following a request from the Prosecutor under Article 19(3) of the Rome Statute, whereby a majority of PTC I considered, in a nutshell, that even though Myanmar was not a State party, the ICC could exercise jurisdiction in relation to the crime of deportation to the extent that one element of the crime (the crossing of an international border) did take place on the territory of a State party, here Bangladesh. The PTC also found that the ICC could possibly exercise jurisdiction over other Rome Statute crimes if it could also be established that one element of the crime took place on the territory of a State party.

Putting aside my doubts about whether it was appropriate to use article 19(3) like this in the first place (see here), I note that in the current request, the Prosecutor essentially takes the September 2018 jurisdictional decision as a given and does not explain this newly composed Pre-Trial Chamber what exactly what the reasoning of the other PTC was in relation to 1) considering deportation as a discrete crime under the Rome Statute and 2) its understanding of « conduct » in article 12(2)(b).

Given the discussions surrounding the use of 19(3) and the substance of the September 2018 decision, I would have expected the Prosecutor to do a little more explaining (convincing) in this respect, because it cannot be assumed that the new Pre-Trial Chamber will simply follow the prior decision in this respect.

Three additional small points in respect to jurisdiction:

– It is interesting to note how the OTP brings in other « crimes » (such as killings and rape) as evidence of the coercive nature of the displacement. This is quite smart I would say, but it does questions. Indeed, these « crimes » were entirely committed on the territory of a non-State Party and the Court cannot directly exercise jurisdiction over them, which means that any potential Accused cannot be charged for them. However, concretely they would be part and parcel of the discussion of deportation and any pronouncement of the Court on such questions would necessarily involve discussion of individual responsibility. In other words, any potential Accused would in effect be charged and convicted for these acts under the guise of deportation, which could reasonably be construed as an impermissible circumvention of the jurisdictional limitations of Article 12(2).

– At this point in time, the OTP reasoning on the link between these coercive acts and the deportation is quite embryonic. Indeed, logically, you would expect the Prosecutor to demonstrate a link between specific groups of refugees in Bangladesh and the alleged crimes constituting the coercive acts. However, the Prosecutor simply states that: « The coincidence in time between the peak of the violence and large numbers of Rohingya crossing into Bangladesh may, of itself, show a causal link between the coercive acts and the victims’ displacement to Bangladesh » (par. 113). While this might be sufficient at this stage of the proceedings, this should not be sufficient moving forward if an actual investigation takes place because conflating correlation and causation is not an acceptable investigatory methodology.

– Given the Prosecutor’s flexible inclusion of other « crimes » as underlying acts of coercion, I’m genuinely curious why genocide was not discussed. Now, I’m not familiar enough with the current debate going on about this, but there are a lot of claims of genocide being made. It would therefore interesting to know whether the non-inclusion of genocide is due to 1) lack of jurisdictional territorial link given the particular circumstances of this situation, 2) lack of actual evidence of genocide in the evidence provided to the OTP (as opposed to ethnic cleansing for example) and/or 3) a strategic choice to stay away from the issue.

  • The evaluation of the « interests of justice »

The request includes the usual (based on recent practice of the OTP) couple of pages indicating that “The Prosecution has identified no substantial reasons to believe that an investigation into the situation would not be in the interests of justice” (par. 290).

I have no strong views on this either way, but I do wonder whether this question might not have deserved a change of approach, given that this request in the first one to be filed since the Afghanistan decision. This latter decision and the fact that the OTP appealed is only mentioned in passing in a footnote (fn 774).

However, one could have imagined that the Prosecutor would present in more detail than it does here at least its understanding of the legal framework applicable to « interests of justice » determination, especially when it comes to the margin of discretion (if any) afforded to a PTC in reviewing an OTP request, rather than just referring the Judges to its appeals brief.

More discussion would appear all the more justified that this situation seems to fall squarely in the Afghanistan logic, given that it involves alleged conduct that took place entirely (putting the legal nicety of the « crossing of the border » criteria) on the territory of a non-State party by non-State party nationals, with virtually zero chance of cooperation. If there was ever a time to explicitly challenge the Afghanistan understanding of the « interests of justice » in a request, it was now.

Maybe there is a strategic choice here to not insist too much on this, on the double assumption that 1) the current PTC will know what the debate is about and 2) possibly that the Afghanistan decision is and will remain an isolated decision. Time will tell whether it was a good choice.

Of course, logically if I were the current PTC, I would not issue any decision until the Afghanistan appeal has been resolved one way or another.

  • The OTP’s approach to the assessment of available information.

As is well known, during the preliminary examination, the Prosecutor does not have autonomous investigation powers and must rely on publicly available information or information voluntarily shared by various sources (States, NGOs, etc.). This investigatory dependence on outside sources raises the question of what methodology needs to be followed by the OTP in assessing the seriousness of such information.

I recently submitted a communication to the OTP on this issue, specifically addressing methodological difficulties arising from relying extensively on human rights reports during a Preliminary Examination, such as the over-use of anonymous hearsay, the unverifiability of sources, the impossibility of assessing the credibility of alleged witnesses or the sometimes less ideal legal analysis put forward.

What is particularly relevant for the current discussion is that in that report I support my claim that the OTP should rigorously assess third-party material partly on the fact that the OTP itself has in the past, both in policy documents and in requests, clearly highlighted its autonomous obligation to assess the credibility and reliability of third-party reports.

For example, in the Afghanistan request, the Prosecutor indicated (par. 29):

The Prosecution has evaluated sources and their information following a consistent methodology based on criteria such as relevance (usefulness of the information to determine the commission of crimes within the jurisdiction of the Court), reliability (trustworthiness of the provider of the information as such), credibility (quality of the information in itself, to be evaluated by criteria of immediacy, internal consistency and external verification), and completeness (the extent of the source’s knowledge or coverage vis-à-vis the whole scope of relevant facts). It has endeavoured to corroborate the information provided with information available from reliable open and other sources.

This paragraph is directly copied from the Burundi request (par. 26) a few months prior.

In a similar fashion, in the Georgia request, the Prosecutor indicated par. 48):

Notwithstanding the low threshold that is applicable at this stage, neither theProsecution nor the Chamber should rely on information that is not credible or reliable. This is clear from the statutory requirement of determining whether the information available establishes a reasonable basis to believe that one or more crimes within the jurisdiction of the Court have been committed. Similarly, the Prosecutor, and the Chamber, must analyse and evaluate the seriousness of the information and the reliability of the source. To hold otherwise would require the Court to take any allegation made by any source at face value.

Given these systematic methodological clarifications in the most recent OTP requests, I was surprised that they have disappeared altogether from the Myanmar request, which simply states that: « The sources relied upon in this Request are amongst those considered by the Prosecution to be sufficiently reliable and credible for the proposition for which they are relied on » (par. 29).

Of course, you might consider that this is just a cosmetic difference and that this does not mean that the OTP did not concretely seriously assess available information. Maybe (although I note that the terms reliable, reliability, reliable, credible, credibility, authenticity, corroborate, corroboration and corroborated appears, taken all together, 13 times in a 146-page request). But I am still young and naive and believe that there is a reason why a party makes certain claims or not. As a result, I wonder why the OTP felt it necessary to explicitly lay down some methodological points in all recent requests, but not in the Myanmar request.

As an aside, I don’t understand why the annex listing information used is not publicly available (at least in redacted form), given that arguably a large number of sources will be public sources.

In conclusion, and generally, this request is a perfect case study in the ongoing debate about the balance that needs to be struck between the aspirational vocation of the ICC and its need to be realistic about what it can achieve (see recently Mark Kersten on this), and the related discussion about managing expectations created by ICC activities among affected communities.

I would think those who consider that the ICC cannot give up on being aspirational will likely think that the balance has been struck adequately here, within the jurisdictional limitations of the Court (although the genocide issue will obviously be a sticking point, given the weight of this issue in public discourse on the situation of the Rohingya).

I’m personally not too sure. The fact remains that both legally (the jurisdictional argument is not as solid as it should be) and practically (irrespective of the legal argument on deportation, this is essentially a situation relating to a non-State party), the ICC could be seen to be straying here on the very periphery, at best, of what it can deal with and the chances of anything concrete coming out of this are very slim.

This does not necessarily mean that the OTP should not have moved forward, but, at the very least, those reporting on this (NGOs and journalists alike) should be careful on how they present this move and its possible outcome, so as not to create unrealistic expectations that,  if they should  not be met, will contribute to plague the Court’s legacy in years to come.

Short post on ICC Presidency underwhelming decision in relation to the Judge Ozaki imbroglio

Yesterday, the Presidency of the ICC issued a decision relating to various requests filing by the Ntaganda Defence following the decision of the plenary of Judges to allow Judge Ozaki to continue to sit on Trial Chamber VI during the deliberations of the case, while taking up a position as Japanese Ambassador to Estonia.

I just wanted to share a few quick thoughts:

1/While I understand it procedurally, I find the distinction between fitness to be a Judge generally and fitness to be a Judge on a particular trial somewhat artificial. The Presidency suggests in its decision that only the latter can be raised by the Defence in a particular case, because general unfitness does not necessarily entail bias in particular proceedings. But if a Judge is unfit to be a Judge at all, how is that not something that can be raised by a Party to a specific case? It does not make sense to me practically.

I also note that the decision considers that determining whether a Judge is fit to be a Judge (independent, impartial, etc) is a purely “administrative” function. This bothers me, because such a determination is at the heart of the judicial process and someone should be accountable for it. Another victory for managerialism?

2/ I find the harsh criticism of the Defence filing strategy, which is accused of obfuscating things and creating unnecessary delays wholly inappropriate. It is not unreasonable for the Defence, faced with an unprecedented scenario, to take all necessary measures to make sure that no procedural mistake is made. But it seems it is always a lose-lose situation for the Defence. If they had gone straight to TC VI with a disqualification request, they might have been told to address the Presidency. Or they might have been told that they lack a factual foundation for the request. If they go to the Presidency with a request for disclosure in order to support their subsequent request, they are told that they are wasting time. The bad faith here is breathtaking.

All the more breathtaking,  given the Judges’ collective failure to avoid this mess in the first place, therefore putting the Defense in this difficult situation.

In that respect, the setting of a strict deadline for the Defense to file a disqualification motion before TCVI (par. 24) is frankly incomprehensible. There is no legal basis given for that and there is no reason the Presidency (an administrative body, by its own admission) should be setting deadlines for Parties in a particular case instead of the Trial Chamber to which the case has been assigned. This is wholly inappropriate. The disposition of the decision even goes further as to also set a deadline for any response by the Parties as well as a deadline for Judge Ozaki to respond.

A footnote (fn 30) tells us that “One Judge was of the view that the Presidency could not impose a deadline for the filing of a request for disqualification”. Thank you for letting us know, but it would have been nice to have some insight in what actually justifies this legally.

3/ On the disclosure of information, the Presidency continues to refuse to disclose any information relating to the process that led to the Plenary decision in the first place. I do not understand the lack of transparency here, especially given the storm the decision created. The Presidency cannot hide behind the fact that this is an administrative decision because, as noted above, such a decision goes at the heart of the judicial function of the Court and should be subject to public scrutiny, if only by the Parties to the proceedings.

Apparently, “One judge expressed a contrary view and, rather, would have favoured partial disclosure of information (other than the records of deliberations amongst the judges) as necessary to safeguard the human right to defence of Mr Ntaganda including the right to ask and receive information required to properly exercise his right to a defence” (par. 29). I agree with this anonymous Judge that the Defence should be given an opportunity to access relevant information that might be relevant for any disqualification motion.

My guess is that ultimately, this whole process will go nowhere and will not end with the disqualification of Judge Ozaki. But in the meantime, the whole procedure continues to give us invaluable insight in how the current Judges at the ICC think, and think of themselves, and this is fascinating.

Guest post: The Appeals Chamber’s Chastisement of PTC II for its Article 87(7) Referral Gameplaying

[Hemi Mistry is an Assistant Professor in Law at the University of Nottingham, where she researches and teaches in the area of international criminal justice. Her current research focuses upon how judicial procedure before international courts and tribunals affects how those institutions pursue their mandates. She wishes to thank Dov Jacobs for hosting this guest post.]

It’s a been a fiery time in the world of ICL: first, the decision by Pre Trial Chamber (PTC) II not to authorise an investigation into the situation in Afghanistan situation and second, this week’s bumper decision by the Appeals Chamber (AC) in Al Bashir, concerning Jordan’s non-cooperation in the execution of the arrest warrant for Omar Al Bashir and – more widely – the matter of immunities under customary international law. With the judicial division of the ICC seemingly intent upon institutional (self-)immolation, the international legal Blogosphere and Twittersphere have watched on (and fanned the flames) in horror. Amidst the flames, I wish to highlight one important aspect of Monday’s AC decision which represents a positive development in the Court’s caselaw. Specifically, this post concerns the Appeal Chamber’s reversal, by majority, of PTC II’s decision under Article 87(7) to refer Jordan’s non-cooperation in the arrest and surrender Omar Al Bashir to the UN Security Council and ICC Assembly of States Parties. Against the backdrop of the bigger ‘immunities’ question (for first of what will likely be many takes, see Jacobs and Akande) and the post-Afghanistan decision ‘crisis’, the corrective issued by the AC may at first glance appear to be a case of fiddling while Rome burns. However, as I will argue, this aspect of the AC decision represents an important attempt by the Chamber to reign in, or at least draw a line under, one of the most problematic and (excuse the pun) unruly traits in recent PTC decision-making. But first, a lot has happened, so to understand the significance of this week’s decision it is necessary to go back to basics…settle in for a story…

Article 87(7) represents one of the limited number of tools available to the ICC to respond to failures by its States Parties to comply with their obligations under the Rome Statute and to provide cooperation to the Court in the forms requested by its organs. It reads:

Where a State Party fails to comply with a request to cooperate by the Court contrary to the provision of this Statute, thereby preventing the Court from exercising its functions and powers under this Statute, the Court may make a finding to that effect and refer the matter to the Assembly of States Parties or, where the Security Council referred the matter to the Court, to the Security Council.

Article 87(7) determinations of non-cooperation and referrals to the Security Council have become a standard response to the failure by States Parties to arrest and surrender Omar Al Bashir upon his presence within their jurisdictions. The first three ‘referrals’ of non-cooperation by PTC I of Sudan in Harun and Ali Kushayb and of Kenya and Chad in Al Bashir, all issued in 2010, were not technically referrals under Article 87(7), but the exercise rather a summarily-claimed inherent power to ‘inform’ the Security Council of the accused’s presence upon the territory of a State Party, or the situation-State. However since then, PTC II, initially configured as PTC I, made three Article 87(7) referrals of Sudanese non-cooperation in violation of its obligation to cooperate with the Court established under Security Council Resolution 1593: two in Al Bashir (here, and here) and one in Hussein (here). Further to this, Chambers have issued non-cooperation decisions against eight States Parties for their failures to arrest and surrender Al Bashir upon his presence within their respective jurisdiction: Chad (2011, and 2013),  Malawi (2011), Nigeria (2013), the DRC (2014), South Africa (2015), Uganda (2016), Djibouti (2016), and Jordan (2017). In all but two instances (Nigeria (2013) and South Africa (2015)) of these instances of non-cooperation the Chamber went on to refer that non-cooperation to the Security Council and the Assembly of States Parties. Monday’s AC decision, however, reverses the PTC’s decision to refer Jordan’s non-cooperation to the Security Council.

PTC II’s caselaw on Article 87(7) is characterised by two themes: 1) Its growing frustration with the failure of states to engage with – even acknowledge – their obligations to cooperate under the Rome Statue, and 2) the PTC’s growing frustration with the Security Council and its failure to do anything to follow up on the Chamber’s referrals once they had been made. These frustrations manifested themselves in a number of ways. One way has been the gradual escalation in the language and detail with which the Chamber exhorts the responsibility of both States Parties and the Security Council to cooperate with and support the Court. Another way this frustration has manifested is the Chamber’s attempt to leverage whatever coercive tools it has within the Rome Statute toolkit to nudge States Parties towards increasing constructive engagement with the Court around the matter of cooperation, with the ultimate goal being fulfilment of cooperation requests. One such tool is the discretionary component of Article 87(7).

The early jurisprudence in Al Bashir adopted a straightforward interpretation of Article 87(7). The permissive, or discretionary, nature of the power to determine non-cooperation and refer was not acknowledged and, instead, once the Chamber had established the State Party was under an obligation to cooperate and two had failed in fact to provide that cooperation, the referral of that finding of non-cooperation was automatic (see, for example, Chad (2011, and 2013),  Malawi (2011)). In these early decisions the Chamber was presented with quite straightforward non-cooperation: the states in question simply did not engage with or acknowledge their obligations under the Statute, asserting instead the immunity of then-President Al Bashir. With non-cooperation findings and referrals stacking up and gathering dust untouched by the Security Council, in 2013 the Chamber was presented with a new situation when, for the first time, it was presented with non-cooperation by a State Party – Nigeria – which did, at least, acknowledge and accept its obligations under the Statute, which professed its commitment to fulfilling those obligations and which provided explanations for its failure on that occasion meet the Court’s cooperation request. In response, for the first time the PTC acknowledged the permissive ‘may’ in Article 87(7) and in doing so emphasised the discretionary nature of its powers under Article 87(7) (para.10). Although it appeared to implicitly acknowledge that Nigeria’s failure to arrest and surrender had been a breach of its obligations to cooperate, the Chamber did not make a formal finding to that effect. Instead, after ‘taking note’ of the explanation offered by Nigeria for its inaction, it concluded simply that ‘it is not warranted in the present circumstances to refer the matter’ (para.13).

Yet, the following year, when the DRC argued that the Chamber should not refer its non-cooperation in materially similar circumstances to Nigeria and having made similar representations as to its good faith commitment to its obligations under the Rome Statute, the Chamber refused to follow its Nigeria decision. Rather, it did two things. First, it appeared to distinguish DRC’s non-cooperation on the facts: referring in its dispositive paragraphs to the ‘deliberate’ refusal of the DRC to arrest and surrender Al Bashir, and elsewhere in the decision it described the DRC’s explanation for its non-cooperation as ‘unconvincing’ (para.13). With no objective difference between the motives of, and explanations provided by, Nigeria and the DRC respectively, it would seem that this distinction on the facts was made on the basis of a subjective assessment of the good faith credentials of the relevant parties. The exercise of discretion on this basis is not itself problematic – indeed the value of judicial discretion is in its ability to allow judges to be responsive to the circumstances of the decision at hand. However, the Chamber’s refusal to be convinced by the DRC’s explanation for its non-cooperation is itself difficult to accept given that the previous year it was ready to accept on face value that same explanation when offered by the Nigerians. That aside, the most problematic aspect of the Court’s decision in the DRC case is the second thing it did: it introduced a new and additional obligation, the obligation to consult with the Chamber in accordance with Article 97 of the Statute. On the basis of two findings – 1) the ‘deliberate’ failure to arrest Al Bashir (which was no more deliberate than Nigeria’s failure) and 2) the DRC’s failure to engage in Article 97 consultations, the Chamber referred the DRC’s non-cooperation to the Security Council, even though no mention of Article 97 was ever made in the previous Nigeria decision. It is this manoeuvre by the Court – the moving of the goalposts for the exercise of its judicial discretion – that is difficult to accept as legitimate.

Up until now, the structure of the Chamber’s analysis under Article 87(7) had been broadly the same. Notwithstanding the implicit acknowledgement of Nigeria’s factual non-cooperation, in all cases, the Chamber swept together the non-cooperation finding and the referral of that finding. However, in 2015, PTC II once again departed from its caselaw, confronted now by a state – South Africa – whose own Supreme Court had determined that it had failed to provide cooperation to the Court, contrary to its obligations under the Statute. Yet, not only did South Africa itself advance this Supreme Court ruling as evidence of its commitment to its obligations under the Statute and – more broadly to the rule of law – it had gone further, and had engaged the process of consultations under Article 97 which the Chamber had, in its DRC decision, identified as a decisive factor in the exercise of judicial discretion. Thus, the Chamber was forced to uncouple the non-cooperation finding and the referral: it made the finding of fact that South Africa had failed to cooperate in accordance with its obligations under the Statute, but nevertheless it decided not to refer that finding to the Security Council.

Having moved the goalposts once in the DRC decision, and having exhausted the (very limited) panoply of obligations under the Rome Statute that it could add to the justifications for the exercise of its discretion in South Africa and in doing so move the goalposts again, the PTC II had – it appeared – exhausted the leveraging potential of its discretion under Article 87(7). Yet, when it came to address Jordan’s failure to arrest and surrender Al Bashir in 2017 although Jordan had objectively overcome all the hurdles set out by the PTC II in Nigeria, then DRC and South Africa, the PTC refused to follow its decision in South Africa. Instead, it determined that what Jordan claimed was engagement with the consultation process in Article 97 was not, in fact, engagement with the consultation process (para.46-49). It further argued that Jordan could not have engaged in consultations for the purpose of Article 97 because the obstacle impeding cooperation that it advanced as the subject of those consultations – i.e. the question of immunities – had been addressed, resolved and removed in the course of the consultations on the same matter undertaken by South Africa (para.54-55). Now, therefore, it would seem that if Jordan wished to invoke the consultation procedure to qualify for the Chamber’s consideration of not to refer its non-cooperation, it needed to have come up with a novel obstacle to cooperation to consult over. Yet again, the goalposts were moved. This manoeuvre was one step too far and, on Monday, the AC reversed the PTC’s decision to refer Jordan’s non-cooperation. Not only did the AC conclude that the PTC made an ‘error’ in mischaracterising Jordan’s attempts to engage the Article 97 consultation process, but that the differential treatment of Jordan and South Africa constituted ‘an abuse of judicial discretion’ (para 210-211).

This aspect of the AC decision, and especially the strongly worded rebuke of the PTC’s conduct, should be welcomed. On one hand, the Chamber’s experiment with leveraging its discretion to refer non-cooperation to the Council to influence the behaviour – even if it is just the argumentative behaviour – of States Parties was partly successful. The DRC – rather than not engaging with the cooperation obligation – acknowledged, like Nigeria did, its obligation under the Rome Statute, stated its commitment to its Rome Statute obligations, but explained the practical impediments to its ability to discharge those obligations at the case at hand. Similarly, Jordan sought to engage the same consultation process under Article 97. Even though, as the Chamber itself pointedly observed, the Council has consistently failed to act upon its referrals – such that there is little ‘hard’ consequence from such a referral – States Parties, first the DRC and more recently Jordan, have taken steps to align their non-cooperation with that of Nigeria and South Africa to avoid a referral to the Security Council. Whereas PTC I has argued that the referral under Article 87(7) are meant to be ‘value neutral’ and ‘not designed to sanction or criticise the requested State’ (para.33) (in the same breath, however, it appeared to concur with PTC II in characterising the referral power as a tools ‘“to use at a certain point in time as a last resort measure or as part of a comprehensive strategy to promote cooperation (para.24)), it would seem that PTC II has had some success in persuading States Parties that a non-referral is a ‘carrot’ worth striving (or at least arguing) for.

On the other hand, having achieved that, the Chamber’s refusal to treat like cases alike and its frustration of the legitimate expectation of states, simply, reeks of bad faith. One of the fundamental tenants of the rule of law is the expectation that, under the rule of law, like cases will be treated alike. It is one if not the factor that distinguishes the politics of legality from those other – less desirable – forms of politics that legality exists in opposition to (Shklar 1964). It is through the giving of reasons that legitimate expectations are created and stability and predictability in the exercise of judicial discretion are guaranteed. In the face of the considerable political, diplomatic and – indeed – legal controversies arising out of the Al Bashir arrest warrant, it was reasonable for states to look to the PTC’s caselaw for guidance and to rely upon that caselaw. The development of – and adherence to – stable and predictable principles might do little to incentivise the kinds of changes in state behaviour (i.e. the movement towards cooperation) that the strategy of leveraging its discretion has been aimed towards. Arguably, for this reason, the experiment with decoupling of the non-cooperation determination and the referral to leverage the referral discretion was always doomed to folly. However, PTC I has also accepted the decoupling of the non-cooperation determination and the Security Council referral, having characterised non-cooperation findings as value neutral objective assessments of fact (Al Islam 2014 at para.23) and referrals as ‘one of the tools available to the Court “to use at a certain point in time as a last resort measure or as part of a comprehensive strategy to promote cooperation”’ (Al Islam 2014 at para.23). As such, the discretionary nature of the referral determination is likely here to stay and, going forward, Chambers need to develop transparent principles to guide that discretion if they are to rebuild some of the trust of States Parties. In the face of the failure of the Security Council to take political action in response to PTC, the manoeuvring of the goalposts for the Article 87(7) non-referral was clearly a proactive attempt to enhance the effectiveness of the Court. However, judicial chambers are not the Security Council or the Assembly of States Parties; they are judicial bodies, not political bodies. With the composition of PTC II now changed, this week’s decision by the AC reminds the new crop of judges that – going forward – if they too wish to engage in politics, they must play the game of legality and abide by the basic rules of that game.

You have just entered Narnia: ICC Appeals Chamber adopts the worst possible solution on immunities in the Bashir case

This morning, 6 May 2019, the Appeals Chamber issued its Judgment on the Appeal filed by Jordan against an 11 december 2017 decision by Pre-Trial Chamber II whereby it was found that Jordan failed to comply with an order to arrest and transfer Bashir to the ICC because Bashir did not benefit from Head of State immunity in the context of a UNSC Referral. As a consequence, Jordan’s non-cooperation was referred to the UNSC and the ASP.

It was a long-awaited decision, touching upon fascinating issues of public international law, treaty interpretation, customary international law, effects of UNSC resolutions, etc. The process that led to the Judgment was itself fairly novel, the Appeals Chamber having invited and received amicus briefs from the AU, the Arab League and a dozen law professors, who, in addition to their written briefs, were heard and questioned by the Chamber at length over several days in September 2018.

This is a really difficult blog post to write, because the Judgment comes at the conclusion of 8 years of debates (since the 2011 Malawi decision), with many sub-plots and twists. Explaining comprehensively why this is a terrible decision therefore would ideally require some prior knowledge of what the stakes are and would require me to fill way too many pages for a blog.

Therefore, for a comprehensive overview of the issues and explanations of why the reasoning of the Appeals Chamber is not convincing, I simply refer you to the innumerable blog posts I’ve written on the topic (particularly my initial reaction to the Malawi decision back in 2011, which is relevant here given the fact that the Appeals Chamber dug it up from nowhere) and my comprehensive chapter on the issue (an earlier draft of which you can find here).

What I will do here is simply pick and choose some particular problematic or noteworthy aspects of the Chamber’s reasoning.

First of all, at the heart of the Appeals Chamber’s reasoning is the conceptual idea that there exists a concept of “international tribunal” that is not simply the “pooling” of the exercise of jurisdiction by States (see par. 115 of the Judgment). These “international tribunals” would have a different nature and would therefore not be subject to the same rules of international law than States are, which allows the Appeals Chamber to say that while Head of State immunity continues to apply between States, it does not apply before international tribunals. However, this reasoning is very problematic. Indeed, it relies on a moral, rather than legal basis, as seen in the claim that “international tribunals” act in the name of the “international community as whole” rather than individual States. But what is the “international community” as a whole from a legal perspective? It is of course never defined in the Judgment.

Moreover, where does the “pooling” of jurisdiction end and the “international tribunal” start? For example, if France and Belgium, who separately would not be able to arrest and prosecute a foreign head of State, create an “international tribunal” through a bilateral agreement, would that new institution be able to prosecute that same Head of State? I doubt it, because States cannot grant to an IO a power they do not possess. This is basic common sense. You can’t just avail yourself of this simple rule by essentially claiming the moral high ground. Also, what arrogance to claim that you are acting in the name of the “international community as a whole”, especially when you see in practice the opposition that exists to the ICC (whether it is justified or not is another issue).

Given the reasoning adopted by the Appeals Chamber on this point, I find it somewhat ironic that the Judges go on the explain that “The law does not readily condone to be done through the back door something it forbids to be done through the front door” (par. 127). Not only is there a slight Judge Dredd feeling about this sentence (who is the “law” exactly? “I am the law!”), but it is exactly what the Judges did to remove immunity in the first place: treaty rules might be a problem? let’s use the back door of the “international tribunal” and magically go through the cupboard into the magical Narnia world when standards rules of international law disappear under the benevolent gaze of the “international community as a whole”…

The key issue should not so much be what an abstract category of “international tribunal” would look like, but rather, in each specific circumstances how a tribunal was created and what effects it might have on third States. This requires no creativity whatsoever, but simply a basic application of public international law rules. In the case of the ICC, as a treaty body, there is simply no reason why it should bind third States and that should be the end of the discussion (the PTC in the Rohinga decision attempted to justify that the ICC could have an effect on third-States as an IO with objective legal personality, but their reasoning was not very convincing either). 

Second of all, as a consequence of the Judgment, Article 98(1) is rendered mostly meaningless. Of course, Article 98(1) does not actually list what immunities need to be respected by cooperating States, but it is somewhat difficult to imagine that if the drafters of the Rome Statute really thought that Article 27 removed all immunities, even in the horizontal relationship between States, as an established rule of customary international law, it would have bothered to introduce Article 98(1) in the first place.

Third of all, the consequences for third States are quite big.

1) this Judgment means that immunities cannot be claimed by nationals of non-State parties, even when the situation is not referred to the Court by the UNSC. In other words, even if a situation is opened through a State referral or a proprio motu decision by the Prosecutor, State Parties would have to arrest and surrender nationals of non-State parties who would otherwise benefit from immunity.

2) This is probably one of the weirdest consequences of the Judgment: because the Appeals Chamber claims the existence of a rule not just in the Rome Statute, but in customary international law, that there are no immunities before “international tribunals”, one could arguably claim that the obligation to arrest and surrender a person would rest not only on State parties but also on non-State parties, because customary law is binding on all States… this is of course a ridiculous proposition, but it shows the absurdity of the Judgment.

Fourth, I note that the AC makes no mention of the Malabo Protocol, which explicitly provides for Head of State immunity. Presumably, by the AC’s standards, the tribunal constituted by the Malabo Protocol would be an “international tribunal”. Therefore, it would be acting in the name of the “international community as a whole”, in claiming that immunities exist before “international tribunals”. Why would this not equally be evidence of a contrary customary international law?

Fifth, I note that the Appeals Chamber mentions in one paragraph that Jordan’s obligation to cooperate would also stem from the Convention against Genocide. This was the position taken in a separate opinion to the South Africa decision by Judge Brichambaut (see my commentary here). I was not entirely convinced by the argument at the time, but at least the Judge made an effort to analyse the Genocide Convention specifically and explain how it would be related to the Rome Statute. Here, there is no such effort and it is impossible to understand what the relevance of the Genocide convention is in the current proceedings. Maybe it was included as a possibility in a earlier draft of the Judgment and someone forgot to remove it, because as it stands, these few lines, which seem to be added as an afterthought, with no explanations, 1 footnote and no references, are completely useless.

Sixth, I don’t have much to add that I haven’t said in the past to the analysis provided by the Appeals Chamber of the “UNSC route”. Just a few quick thoughts:

1) It’s not entirely clear why the AC bothers with this section of the Judgment at all, given that the questions raised become essentially moot given the customary law avenue taken by the Chamber.

2) I do note that the reasoning provided by the AC is interesting when it comes to determine under what sections of the Rome Statute Soudan would be obliged to cooperate with the Court. I must say that on first reading, I am convinced with the argument according to which, given the language of the UNSC Resolution, Soudan would have to respect the cooperation provisions relating to State parties rather than non-State parties. However, this does not automatically mean that Soudan would be bound by Article 27, and on this the Judgment is less convincing.

3) I also note that the AC did not follow my friendly advice: it did not actually asked those States on the UNSC who wrote the referral if in fact they did intend to remove immunities. This would have been all the more interesting as some of these States have made public Statements to the contrary (see here).

Seventh, in relation to the referral of Jordan to the ASP and the UNSC, I remain a bit lost at what the legal framework is. In such discretionary matters, it seems more of a divination exercise than one of legal reasoning, so I don’t have much to say about that. I just wonder, maybe naively, when a referral will ever be justified. Indeed, when the Jordan cooperation issue comes up, there is 7 years of constant (if not consistent) case-law asking State parties to arrest and surrender Bashir. I have all the sympathy in the world for Jordan’s legal position, which I mostly share, but practically, in those circumstances, it is difficult to see Jordan’s non-cooperation as anything else but a clear refusal to comply with a clear order from the Court. How is that not sufficient for a referral? I’m wondering, but this is just me thinking out loud, whether the real issue is to avoid that this discussion be forced on the ASP/UNSC, bodies which might make statements that the Judges would not want to hear. Indeed, it would not look good if the UNSC or the ASP (under the pressure of the AU States) actually came out in defense of Jordan…

Eighth, while I have yet to read the separate opinion (I might blog on it later), I am not sure what to make of its existence. Indeed what are we taking about?  it is a separate concurring opinion which focuses on written by 4 of the 5 judges in relation to the 2  grounds of appeal for which the decision was unanimous. So basically, did the 5th Judge not agree with the content of the separate opinion? because if she had, it would be signed by all 5 judges, concurring with themselves and then this would just be another part of the Judgment… Also, as also noted by Hemi Mistry, there are regular cross-references between the Judgment and the concurring opinion, which raises the question of the exact status of the document. I hope to have more clarity when I actually read it.

In conclusion (for now), I honestly believed that the AC would play it safe and stir away for the Customary Law route. I was wrong. As noted by Dapo Akande, this radical view justifies even more that the AU move for an ICJ advisory opinion on the matter. They have nothing to lose now.

From the perspective of the ICC, the Appeals Chamber has sadly confirmed I was right when I chose “the frog that wanted to be an ox” title, for my book chapter on immunities. This is again a case of the ICC Judges trying to be more (and to make the Court be more) than it actually is. In the fable, the frog actually exploded at the end… it’s of course just a metaphor, but given recent developments at the Court, it’s increasingly becoming a concrete risk for the institution.