Editor’s Note: This is a featured post. Newer posts, including those in our online symposium on The Constitutionalization of International Law, appear below
The ICJ has now officially announced that it will deliver its advisory opinion in the Kosovo case on 22 July. This essay/post is intended to serve as a preview of the many issues raised in the case, of the main lines of argument by states before the Court, and of the several possible avenues that the Court might take in deciding the case. It is meant to be structured in such a way to help the reader to better understand where the Court is coming from once it hands down its much anticipated opinion, and I hope that it will prove helpful. But, before that:
A massive disclaimer: I have had the immense honour and privilege to work as an advisor to the Serbian legal team. I am therefore at least doubly biased, since I am both a Serb and was paid to be one, and this post might reflect that bias despite my best efforts. That notwithstanding, this post is based on my opinions and my opinions alone, and does not in any way, shape or form reflect the position of the Serbian legal team. And when I say ‘opinions,’ I will do my best merely to explain the various lines of argument and the options before the Court as objectively as possible, rather than advocate for some of them. I will likewise deliberately say nothing on the possible practical consequences of the Court’s ruling, nor comment in any way on the broader political context. Finally, I would truly very much appreciate it if anyone citing this post in whatever fashion did NOT refer in doing so to my role within the Serbian legal team, as again this post is written solely in my personal capacity.
Further technical note: the written pleadings in the case are available here, and the oral pleadings here. In describing the various arguments, I will generally refer to (some of the) states making them, but I will not be giving precise citations. Likewise, readers are certainly aware that the Court has not only invited states to make their arguments on Kosovo’s independence, but has also asked the authors of the unilateral declaration of independence (UDI) of Kosovo to provide their own contributions, without prejudice to Kosovo’s status. As the readers will shortly see, the identity of the authors is actually a live issue in the case. For the sake of simplicity, however, I will refer to ‘Kosovo’ both to designate the geographical area, and the (‘so-called,’ ‘self-proclaimed,’ etc.) entity of Kosovo, whose UDI is the object of the case, and whose representatives spoke before the Court. Finally, the length of the disclaimers by itself indicates that this will be a relatively longish post, so sorry about that.
Now, with that out of the way – in the course of giving its advisory opinion, the Court will have to address in one way or another four broad groups of issues:
(1) Various preliminary objections (put forward e.g. by France and Albania) mainly as to the propriety of the exercise of its advisory jurisdiction, that seek to persuade the Court to decline to render an opinion. These objections are neither likely to succeed nor are themselves particularly interesting, and I will therefore not be discussing them in this post;
(2) The meaning of the question that the UN General Assembly has posed to the Court. As we will see, the scope of the question is an absolutely crucial issue, that will have ripple effects on all other issues in the case;
(3) The lawfulness, vel non, of the UDI under general international law;
(4) The lawfulness, vel non, of the UDI under UN Security Council Resolution 1244 (1999).
With the exception of (1), I will now address each of these broad areas in turn.
The Question Question
UNGA resolution A/RES/63/3 asked the following question of the Court: “Is the unilateral declaration of independence by the Provisional Institutions of Self-Government of Kosovo in accordance with international law?” The Court obviously has to interpret this question before attempting to answer it, and much will depend on how it proceeds to do so.
Both Kosovo and its supporters (e.g. the US and UK) have criticized this question as being absurd or nonsensical, or at the very least so narrow that it renders the Court’s advisory function almost totally meaningless. In their view, the GA has asked the Court (in a formulation proposed by Serbia, which Serbia itself referred to as narrow) to proclaim on the legality of a verbal act adopted on a given day by (what in Serbia’s argument was, and remains) a non-state actor – the UDI of 17 February 2008. However, except in the most limited of circumstances (e.g. incitement to genocide), international law has absolutely nothing to say on the verbal acts of non-state actors, as it does not regulate them with regard to their content. Therefore, the question whether the UDI was ‘in accordance’ with international law either cannot be answered, or that answer is very simple indeed – there was no norm of international law prohibiting it, and hence it is in full accordance with international law.
For their part, Serbia and its supporters (e.g. Spain, Cyprus and Argentina) argue that the question is neither all that narrow nor absurd. Not only do some of the objections above presuppose the content of the norms in dispute, but the General Assembly cannot be thought to have asked the Court a question without any meaning or purpose. Rather, in Serbia’s view, the question is not confined to the legality of a purely verbal act by a non-state actor, but pertains to the lawfulness of a purported act of state creation, i.e. the secession of a part of Serbia’s sovereign territory. The question can be said to be narrow only to the extent that it excludes from its ambit the lawfulness, vel non, of the acts of third states, namely the recognition by some states of the alleged new state of Kosovo.
This account is of course a bit simplistic, but I hope that it adequately captures the arguments of the opposing sides. Bearing them in mind, the Court has three basic options as to how it will interpret the question:
(1) A narrow reading will mean that the Court will confine itself to examining the legality of a purely verbal act adopted by a non-state actor on a given day;
(2) A moderate reading will require the Court to examine the lawfulness of the secession as such;
(3) An expansive reading will have the Court venture not only into examining the lawfulness of the secession, but also into the further issue of what are the consequences of its finding of (il)legality. Most importantly, after examining the lawfulness of the secession, the Court could establish whether the entity of Kosovo fulfils the legal criteria for statehood, i.e. possesses a sufficient factual predicate for it in the absence of any legal impediments.
Subtle variations in these three readings are of course possible, as each is underpinned by its own policy considerations and by how the Court perceives its adjudicative role in the international community. One thing to watch for is whether the Court will resort to explicitly reformulating the question posed by the General Assembly, as it has done in several cases in the past. But again, I cannot emphasize this strongly enough, a great deal depends on which option the Court chooses. In the following analysis, I will proceed from the assumption that the Court will choose either a moderate or an expansive reading of the question.
The UDI and General International Law
If the UDI is seen as a purported act of secession and state creation, what does general international law have to say about it? In the submission of Kosovo and its supporters – nothing much. International law is essentially neutral towards secession, treating it as a fact that is neither legal nor illegal. In some cases, international law may explicitly prohibit secession, for example when it is a direct consequence of the use of inter-state force, as was the case with Northern Cyprus. In others, it may explicitly create a right of secession, as with external self-determination. In between these two extremes, however, international law remains in a posture of neutrality.
This in Serbia’s view incorrect. Far from being neutral towards secession, modern international law posits territorial integrity as one of its foundational principles. Unless it creates an explicit and exceptional right to secession, as with external self-determination, international law generally prohibits it in favour of the mother state’s integrity.
To this argument Kosovo responds by saying that Serbia and its supporters mischaracterize the legal nature of the principle of territorial integrity. It is indeed a foundational rule of modern international law, but it is one which applies only between states. In other words, it is other states which (up to a point) had a duty to respect Serbia’s territorial integrity, but that duty did not extend to individuals and non-state actors within Serbia itself. Such a duty may have been imposed, say, by Serbian domestic law, but individuals and non-state actors most certainly do not have it under general international law.
The written and oral arguments made by dozens of states before the Court are of course a bit more complex. They also involve much speaking past each other, a large degree of confusion, as well as conceptual and terminological inconsistencies. But ultimately, in my view, the general international law issue boils down to a choice between two competing models of secession: a two-part model advocated by Serbia and its supporters, where secession is prohibited unless expressly permitted; and a three-part model advocated by Kosovo and its supporters, in which there are narrow zones of both prohibition and explicit permission, and a wider zone of neutrality, in which secession is neither prohibited nor expressly permitted.
Now, crucially, this is again where the ‘question question’ comes into play. Kosovo and its supporters thus argue that it is only if Kosovo’s secession comes into in the ‘prohibited’ zone of either model that it would be necessary for the Court to venture into the issue of self-determination. This is so simply because the GA’s question does not ask the Court to rule whether Kosovo had a right to declare independence, but whether its UDI was in accordance with international law – a requirement that a mere absence of a prohibition would presumably satisfy.
It is because of this procedural posture and the formulation of the question that the issue of self-determination has played a distinctly subsidiary role in the proceedings (to the initial surprise of many, myself included). Kosovo did its best to persuade the Court that it simply doesn’t need to reach it – indeed, if one takes a look at Kosovo’s first written contribution, self-determination takes up only two (2!) out of hundreds of pages. On the other hand, many states have put a greater emphasis on the self-determination issue, and the Court may well have something to say about it. As the readers are aware, that issue raises several distinct sub-issues: are the Kosovo Albanians a ‘people’; does a non-colonial people gain the remedial right to external self-determination on account of state oppression and denial of human rights and/or internal self-determination; if so, has that right lapsed because of passage of time and Serbia’s offer of full internal autonomy.
The UDI and Resolution 1244
One strand of the case is thus the conformity of the UDI with general international law. The other is its validity under UNSC Resolution 1244 (1999).
The main issues basically revolve around the proper interpretation of Resolution 1244. The arguments are very sophisticated, and I am incapable of summarizing them in a truly coherent way (I would in that regard refer the readers to the oral arguments of Sean Murphy on behalf of Kosovo and Andreas Zimmermann on behalf of Serbia). Suffice it to say that the main thrust of Kosovo’s argument is again the absence of any prohibition. In their view, there is nothing in the resolution which prevented Kosovo from adopting a declaration of independence. In particular, the preambular reference to the FRY’s territorial integrity was nothing more than a statement on the Council’s part of the legal situation as it was when the resolution was passed – that Kosovo was then, in 1999, a part of the FRY. It was not, however, a guarantee that Kosovo will always remain a part of Serbia.
Serbia, on the other hand, argues that the resolution asked the parties to reach a settlement on Kosovo’s final status, and that the term ‘settlement’ by definition excludes solutions imposed unilaterally, such as the UDI. In fact, both the Security Council and the Special Representative of the Secretary-General (SRSG) have previously declared as invalid and contrary to Resolution 1244 acts adopted by the Kosovo Assembly which in one way or another tried to assert a claim of independence.
Kosovo responds by saying that today, unlike in previous cases, there was indeed a ‘settlement,’ which was not unilateral, as it followed a political process of negotiation, the appointment of Martti Ahtisaari as the UNSG’s special envoy, and his conclusion that the final status should be independence. Serbia rejoins, however, by saying that the Security Council never endorsed Ahtisaari’s plan and Kosovo’s independence.
These are the main arguments in a nutshell, but they are interwoven in a very complex web with several other more-or-less ancillary arguments, some of which I think are absolutely fascinating. They are moreover easy to miss if the case is looked at only casually. They include:
(a) The identity of the authors of the UDI: One issue, that at first glance might seem to be quite marginal or even peculiar, is how to characterize the authors of the UDI. In Serbia’s view, the authors of the UDI were undoubtedly the Provisional Institutions of Self-Government (PISG), namely the Kosovo Assembly, President and Government, that were set up by UNMIK under the auspices of Resolution 1244. Not only is this a fact that the Court can adduce from the available evidence, but in its question to the Court the General Assembly has actually made this factual determination, when it asked to Court to rule on whether ‘the unilateral declaration of independence by the Provisional Institutions of Self-Government of Kosovo [was] in accordance with international law.’
Kosovo responds by saying that the authors were not in fact the PISG, i.e. were not acting in that capacity, but were acting organically as the ‘democratically elected representatives of the people of Kosovo.’ The UDI was not adopted as an act of the Kosovo Assembly as part of the PISG, but was signed by the assembled dignitaries more or less in their personal capacity, as is shown, inter alia, by the original vellum manuscript of the UDI itself.
Thus, the parties in the case actually dispute the very identity of the authors of the UDI. At the oral hearings (e.g. at 28, 29, 30, etc.), in order to avoid any appearance of prejudice, President Owada thus repeatedly referred to the authors as the ‘authors of the unilateral declaration of independence by the Provisional Institutions of Self-Government of Kosovo.’ This semantically simply magnificent formulation drew a smile from at least one or two other members of the bench – but we’ll see in a minute why any of this actually matters.
(b) Legal nature of UNMIK regulations: One prong of Serbia’s argument is that the UDI not only violates Resolution 1244 as such, but also the whole legal regime established by it through regulations passed by the SRSG/UNMIK as the international civilian administrator. Among these regulations the most important is the Constitutional Framework for Kosovo, which sets out the competences of the PISG. In Serbia’s view, the UDI and subsequent developments were manifestly ultra vires acts under the Constitutional Framework.
Kosovo and its supporters respond by saying that even if the UDI was a violation of the Constitutional Framework, the GA’s question only asks the Court to assess the conformity of the UDI with international law, which the regulations are not, as they are or were the local, municipal law of Kosovo.
Not so, says Serbia. Not only were the regulations passed by UNMIK, a subsidiary organ of the UN established by a Security Council resolution, and were hence international by definition, but Kosovo’s argument presents a false choice to the Court. There is no reason why the Constitutional Framework and other regulations couldn’t at the same time be regarded as being both international and domestic in nature.
And this is where the identity of the authors of the UDI starts to come in – if they were not the PISG, and were acting in some sort of organic capacity as the democratic representatives of their people, a pouvoir constituant, then they could not have acted ultra vires the Constitutional Framework because it was not the Framework which gave them their vires to act. Serbia then says, oh come on, this is just a self-serving construction designed to avoid the binding effect of the Constitutional Framework. These were the same people who were elected as the PISG and they didn’t just go poof! one minute and transform from the PISG into some ill-defined group of individual representatives.
(c) Binding effect of UNSC resolutions on non-state actors: The case also raises a similar, but more fundamental issue – could the UDI, as an act of a non-state actor, even violate a UNSC resolution? Under Art. 25 of the Charter, it is UN member states who have a duty to comply with UNSC resolutions, but there is nothing in the Charter that says that individuals or non-state actors generally have such a duty. If, in other words, Resolution 1244 was not even binding on the authors, how could the UDI have violated it?
Now, this is an argument that was noticed by a very small number of states – and it is an issue very, very rarely addressed in the literature. Even those that have noticed the argument chose not to employ it in such a categorical way. Some have merely expressed doubts as to whether the UNSC can bind non-state actors; others have opted for an interpretative presumption, saying that even if the UNSC can bind non-state actors, it needs to do so explicitly, and there is no such language in the resolution.
Serbia has several responses to this line of argument. First, because the authors were the PISG (here again the identity issue comes into play), and the PISG were created by UNMIK, which was itself created by Resolution 1244, the PISG must themselves be bound by 1244, and indeed this is what the Constitutional Framework explicitly says. Likewise, the Council has in its presidential statements also explicitly referred to the need for the PISG and all other actors to comply with 1244, and has accordingly regarded the resolution as being binding on non-state actors. Finally, even if the authors were not the PISG, all actors in Kosovo were subject to 1244, as this is simply a consequence of the nature of international territorial administration.
(d) Acquiescence by the UNSC and the UNSG/SRSG: Acquiescence is a major argument deployed by Kosovo and its supporters in favour of their preferred interpretation of Resolution 1244 as requiring neither acceptance by Serbia nor explicit endorsement by the Security Council for a ‘settlement’ of Kosovo’s final status to be reached. This purported acquiescence takes two forms. First, the UNSG and the SRSG have acquiesced in the UDI by not annulling it, as they had the authority to do, an authority that the SRSG exercised on several previous occasions. Second, the UNSC has also acquiesced, as it has adopted no decision condemning the UDI, nor has it instructed the UNSG/SRSG to disregarded it. All this indicates that the primary actors tasked with interpreting and applying Resolution 1244, to whose judgment the Court should defer, did not consider it to preclude the UDI.
Serbia responds as follows: first, the UNSG and the SRSG have quite explicitly taken a position of ‘strict neutrality’ towards the UDI. A position of neutrality by definition cannot be taken as a tacit endorsement of the UDI’s legality. As for the UNSC, we know exactly why it has been unable to adopt any decision regarding the UDI – because a lack of political agreement among its permanent members. This is neither the first nor the last such time. Considering not only the text of the Charter which sets out the positive requirements for a Council decision, but also the context and the way in which the Council operates, it would be unacceptable as a matter of both law and policy to say that a mere failure to condemn equals acquiescence. Otherwise, the Council could, for example, be said to have acquiesced in any unauthorized use of force that it fails to condemn because of a potential P-5 veto (see, e.g., Iraq).
This, I think, covers most of the issues with regard to the UDI’s conformity with Resolution 1244. The ‘big’ question, if I may be excused for paraphrasing George W. Bush, is who is ultimately the decider. Does a ‘settlement’ have to be negotiated, i.e. can Serbia indefinitely prevent Kosovo’s independence simply by withholding its agreement? Or can Kosovo impose a solution on Serbia, so long as it has the support of a large section of the ‘international community’? Or is it rather the Security Council who has to conclusively say that a settlement has been reached, and what that solution has to be? The big question aside, we mustn’t underestimate the importance of the ancillary issues I have outlined above, and their interplay with the interpretation of Resolution 1244.
Conclusion: The Court’s Options
Having thus looked at the various issues and arguments, what are the options before the Court? I can see six, again bearing in mind that the Court can interpret the GA’s question narrowly, moderately, or expansively:
(i) Dismissal: The Court can refuse to answer the question and dismiss the case on various grounds of judicial propriety. This is unlikely, though not impossible.
(ii) Narrow reading: The Court can read the question very narrowly, in which case its answer will be very narrow as well, confined to the formal examination of the UDI as a purely verbal act adopted by a non-state actor on a given day. Whatever the Court’s decision, it will legally be neither here nor there with respect to the broader issues at play.
(iii) Moderate reading – Serbia wins: If the Court adopts a moderate reading of the question, and looks at the legality of the secession as such, it can then conclude that the secession was unlawful as a violation of general international law, Resolution 1244, or both. However, it might then rule that it is not for it to say what the consequences of such a finding of a violation would be. Specifically, it could leave open the question of Kosovo’s statehood for resolution (or not) within the UN political bodies or within the political process generally. This would allow Kosovo to argue that though its secession may have been unlawful (in the view of the Court), its statehood is a question of political fact and is untouched by the Court’s decision.
(iv) Moderate reading – Kosovo wins: Similarly, the Court might adopt a moderate reading of the question, find that there was no norm of international law prohibiting Kosovo’s secession and that the UDI was therefore ‘in accordance’ with international law, only to then abstain from ruling on Kosovo’s statehood. This would allow Serbia to argue that though Kosovo’s secession may not have been unlawful (in the view of the Court), Kosovo still does not fulfil the criteria for statehood, as e.g. its independence from Serbia is entirely dependent on the (status-neutral) international presence in Kosovo which still operates under the auspices of Resolution 1244.
(v) Expansive reading – Serbia wins: If the Court interprets the question expansively, and finds in Serbia’s favour, it could then proceed to examine the issue of Kosovo’s statehood and rule that the entity of Kosovo is not a state under international law.
(vi) Expansive reading – Kosovo wins: Alternatively, the Court could read the question expansively and rule in Kosovo’s favour, finding also that it does indeed fulfil the legal criteria for statehood. (It is also theoretically possible, but quite unlikely, that the Court will find that secession was unlawful, but that Kosovo is a state, or vice versa).
These are the results that I think are legally and logically open to the Court (of course, the Court may well disagree, and come up with something completely different). I would, however, dispute the conventional wisdom that the Court will render a Solomonic opinion that will be equally (un)satisfactory for both sides. The formulation of the question and the consequent array of arguments appear to exclude such an option. Except in scenarios (i) and perhaps (ii), there will be a clear winner and a loser, though in scenarios (iii) and (iv) the extent of that win or loss could be somewhat relativized (and these scenarios are for that reason perhaps more likely). Finally, whatever the majority decides, it is certainly possible for there to be a serious split within the Court, which would probably be bad news for everyone, above all for the Court itself. That said, again, the inevitable political spin on the Court’s decision and its broader ramifications are not something that I wish to comment on.