Introductory Remarks by Pierre d’Argent

Welcome to this panel. My name is Pierre d’Argent and I am a professor of international law at Louvain University. I am also a practitioner at Foley Hoag. In a previous life, I served as First Secretary at the International Court of Justice (ICJ). I am very grateful to the organizing committee for asking me to moderate this panel, and I am joined by three friends and colleagues, who barely need any introduction. However, I am going to briefly introduce them to you, and I am going to do that in reverse order of seniority, starting with Janina Barkholdt, who is a visiting researcher at Paris I. She is also a postdoc at Freie Universität Berlin and until recently, she was an Associate Legal Officer at the International Court of Justice serving for four years as a law clerk to Judge Georg Nolte. Alison Macdonald is a barrister at Essex Court Chambers in London, and a fellow of All Souls College Oxford. Alison argues international law cases before English courts and before international courts and tribunals, and I had the privilege of teaming up with her on some of these cases before the ICJ and it has always been a real pleasure. Last, but not least, Sir Michael Wood: he had a full career in public service for more than thirty-five years in London. He was the principal Legal Adviser to the Foreign and Commonwealth Office for six years and after that he became a practitioner, a very successful and respected barrister, and also a member of the International Law Commission.

With some twenty active cases and a handful of requests for advisory opinions, it is now commonplace to note that the International Court of Justice has rarely been so busy. Furthermore, while disputes brought before the Court have until now been essentially bilateral—which is the nature of any dispute, even when obligations erga omnes or erga omnes partes are at issue—and set in a well-established procedural face-off, it is increasingly common to see cases brought by several applicants acting jointly—the case before the Court then breaks down into a cluster of bilateral disputes. Furthermore, while the Registry has, for decades, written thousands of letters under Article 43 of the Rules of Court to states parties to various multilateral conventions whose interpretation was at issue in cases brought before the Court, it is only recently that certain states, that are not parties to the dispute, have responded to this invitation and exercised their right to intervene under Article 63 of the Statute. Finally, the hearings held in advisory proceedings now bring together so many states and international organizations authorized to participate that the hotel capacity of the fine city of The Hague, transformed during these long weeks into “the place to be,” has sometimes been on the verge of saturation.

It is equally commonplace to recall that the Court has not fundamentally changed since it inherited the Statute of its predecessor and that, as an institutional body, it can therefore be considered more than a century old.

Finally, there is nothing original in asserting that institutions die if they do not reform in line with social needs.

Taken together, these various observations seem to call for an affirmative answer to the question on which the organizers of the annual conference invited Sir Michael Wood, Alison Macdonald, Janina Barkholdt, and me to reflect and give our respective opinions.

However, as a very special institution whose success—that is, its effective contribution to the truly unique mission entrusted to it by the UN Charter and the preservation of the special status that results from this in the international legal order—depends largely on the trust of those subjects over whom it has no compulsory jurisdiction, the Court is naturally resistant to the idea of reform, since trust is granted in exchange for predictability, which stems from continuity.

This is not to say, however, that the Court does not adapt. On the contrary, since the 1978 reform of the Rules, the latter have been amended many times and supplemented by Practice Directions that add an extra layer of behavioral propriety that no prudent litigant would risk ignoring. These incremental additions are sometimes welcome, sometimes somewhat regrettable—one may notably think at the rather arbitrary and obscure second paragraph of Practice Direction III—but they keep the century-old institution in a surprisingly youthful state. It would undoubtedly take much more to detect any real dynamic agility and resilience—but, once again, the preservation of a form of stability is undoubtedly written into the Court’s DNA. This reassuring stability was particularly evident when, with a striking sense of modernity, the Court leaped into the twenty-first century by organizing successful remote or hybrid hearings during the 2020–2022 pandemic, even if it meant prolonging the new habits acquired at that time for longer than strictly necessary.

Still on the subject of new institutional practices that have not been translated into texts, it is perhaps regrettable that the Court is increasingly expediting hearings on provisional measures, even going so far as to hold only one round of pleadings when all the evidence in such cases is presented orally. Admittedly, the Court is busy, and the litigants can sometimes be tedious, but depriving the parties of the right to respond to each other seems hardly compatible with the proper administration of justice. It should not be forgotten that the Court is not seized just in order to make noise or try to score points; it is seized because, in this deeply divided—not to say fragmented—world which lacks political authority and direction, recourse to the judge and the protection that it is supposed to be able to confer corresponds to a genuine social need and is often a means of last resort. Appearing not to take this need seriously by failing to allow sufficient time for a genuine adversarial debate to take place is likely to erode the trust that is so essential to the institution and which its clients, the states, are entitled to place in it.

Furthermore, if the suspicion leading to only one round of pleadings is that certain cases are brought for media purposes, refusing to allow the adversarial process paradoxically reinforces this pitfall, since the plaintiff, unaware of its opponent’s arguments, is forced to exaggerate. The same applies to the defendant if the break allocated after the claimant’s round does not allow it to adjust its pleadings. And if the time allocated between the parties’ unique rounds allows for such an adjustment, then the adversarial process becomes an unequally granted privilege, which is the very antithesis of a trial. To think that the Court’s circumstantial overload justifies moving away from what is at the heart of the trial—exchanging arguments rather than hammering them home—is undoubtedly a very short-term calculation.

In short, it is perhaps more the Court’s working methods in the interaction it organizes between the parties, in its interaction with them, and in its internal deliberative process that deserve reflection, rather than the procedural texts over which the Court has control. Also, abandoning certain new practices and returning to older, more relevant ones may sometimes be the right path of reform.

Today, our debate will focus on possible reforms concerning three incidental proceedings: provisional measures, interventions, and jurisdiction and admissibility, in other words, preliminary objections. And those three sets will be addressed respectively by Janina, by Alison and by Sir Michael. Janina, the floor is yours.

Remarks by Janina Barkholdt

Thank you very much for the kind introduction and please allow me to say at the outset that I feel very honored to be part of this panel and I would like to thank ASIL for inviting me to speak on potential reforms relating to provisional measures. Before I start and given that I left my position at the Court only recently, I would like to say that all my remarks are strictly personal and do not reflect the position of the Court or of any individual affiliated with it.

While considerable attention has recently been given to the ICJ’s substantive approach in deciding requests for provisional measures, particularly the threshold for prima facie jurisdiction and the plausibility criterion, I will focus here on a potential reform of the organization of provisional measures proceedings, an aspect that has received comparatively little attention in scholarship.

Before addressing three concrete proposals concerning the organization of the proceedings, I will briefly consider whether there is a need to change the ICJ’s current practice regarding provisional measures. In support of reforming the ICJ’s practice on provisional measures, attention is frequently drawn to three key developments since 2018.

First, the number of requests has increased significantly. Between 1946 and 2018, that is in seventy-two years, the ICJ decided on thirty-six requests for provisional measures. In the last seven years since then, parties in thirteen different cases have requested provisional measures. Notably, provisional measures have been requested in more than half of the twenty-three pending contentious cases (thirteen).Footnote ¹

Moreover, parties have increasingly submitted several “follow-up” requests, either

• for the indication of new measures under Article 73 of the Rules of Court,

• the modification of existing measures under Article 76 of the Rules of Court, or

• a request for the Court to examine in proprio motu the indication of provisional measures under Article 75 of the Rules of Court.

Thirdly, several of these requests have been brought to the Court based on the alleged violation of an erga omnes (partes) obligation.

It is possible to see these developments as a sign of trust vested by states in the Court. However, these developments have led to an increase in the workload of the ICJ with implications for the Court’s calendar and decisions of cases on the merits given that requests for provisional measures, as per Article 74, paragraph 1 of the Rules of Court, enjoy priority over all other cases before the Court.

In addition to that, some commentators have pointed to two other risks arising from these three developments. Some say that states increasingly use provisional measures for political purposes in a way that undermines the judicial function of the ICJ. Others point to a decreasing compliance by states with orders indicated by the Court.

Considering these three trends and potential risks, several options are conceivable.

I would like to focus, first, on the question whether the Court should be given the possibility to decide on requests for provisional measures without holding oral proceedings, that is by hearing the parties based only on a written procedure.

However, a limitation to a written procedure with respect to new requests for provisional measures under Article 73 is difficult to reconcile with the current text of Article 74, paragraph 3 of the Rules of Court. Article 74, paragraph 3 provides that

While it is possible to interpret “hearing” as including the possibility to “hear” the parties by way of a written procedure, the second sentence and the clear reference to “oral” suggests that the Court is required to hold oral proceedings when deciding on a request for new provisional measures under Article 73.

In contrast to that, the Rules do allow for a purely written procedure regarding requests for the modification of provisional measures (Article 76) and the indication of provisional measures in proprio motu (Article 75). Notably, the Court has opted for a purely written procedure to decide on requests for the modification of provisional measures pursuant to Article 76 of the Rules of Courts in most cases.

This means that the shift to a purely written procedure for requests under Article 73 either requires the Court to reinterpret a request for new measures under Article 73 as a modification under Article 76 or as prompting an exercise of its powers under Article 75, which does not require oral proceedings (as happened in the LaGrand proceedings).

Or, and this is in my view preferable to the reinterpretation of the request, this shift requires a change of the Rules of Court. Such a change would be, in my view, compatible with the Statute since Article 43 of the ICJ Statute does not apply to incidental proceedings.

The question is, however, if such a change would be desirable and, in particular, in which cases the Court should exercise its discretion to limit the procedure to the written form. Purely written proceedings could suffice, e.g., where the party has already obtained an order indicating provisional measures in the same case based on largely similar circumstances, or where the Court manifestly lacks jurisdiction. Other, in my view more difficult, situations, include the nature of the alleged rights (i.e., jus cogens) and the urgency of the request (as was the case in LaGrand).

Alternatively, one could consider a mix of a written and oral procedure in the course of which the parties are heard in one round of oral proceedings and respond to the other side in a second round by way of written observations. This would not require a change of the Rules of Court and also ensure that both parties are offered an equal opportunity to comment on each other’s arguments, while, at the same time, saving the days for the second round of oral proceedings.

Finally, the Court may want to consider changing its Rules of Court to concretize Articles 26 and 29 of the ICJ Statute in a way that would allow parties to have their requests heard by a special chamber on provisional measures, which may facilitate a more efficient way of engaging with requests for provisional measures in parallel to the other cases before the Court. Given that any decision on provisional measures does not prejudge the merits, the limited representativeness at that stage of the proceedings might be acceptable for parties given that the final decision will be taken by full Court. However, this mechanism would still be based on the consent of the parties. Any default mechanism that would automatically refer requests for provisional measures to a chamber would require a change of the Statute.

I would like to end by saying that I fully understand—and partly share—certain calls for change based on considerations of judicial economy—in light of the increasing workload of the ICJ—and the desire for a certain publicity of the proceedings—notably where violations of jus cogens rules are at stake.

However, in my view, these needs alone must not become the primary considerations when reflecting on the conduct of provisional measures proceedings. Rather, any changes must

• ensure the right of each party to be heard,

• respect and maintain the original purpose of this procedure that is strictly limited to the preservation of rights pendente lite and the prevention of any aggravation of the dispute between the two parties without anticipating the judgment,

• and, relatedly, preserve the judicial function of the Court, which is why such proceedings should not turn the Court from a judicial into a monitoring body.

In my view, any reform needs to bear these points in mind to maintain the trust invested by states in the Court, which is crucial to ensure their compliance with the measures indicated by the Court. Thank you so much.

Pierre d’Argent

Thank you so much Janina. And I am sure that all those proposals will be debated. Alison, you have the floor.

Remarks by Alison Macdonald

Thank you and I am delighted to be here in D.C. with so many friends and colleagues and to be participating in this really interesting discussion.

When we talk about reforming anything we have to start by asking whether there is a problem, and if so what is it? Looking at the question of interventions, I think there are some challenges, and these can be broken down into three topics which I would describe as volume, logistics, and legal basis:

Volume: We talk about increases in the work of the Court while the Court is still dealing with quite small numbers of cases in real terms. But in relative terms, it is true that there has been quite a sharp increase in states seeking to intervene—a very slow trickle over the years has become a small flood. And the turning point, I think we can agree, was the Ukraine case against Russia under the Genocide Convention, where thirty-two states intervened on the issue of jurisdiction. Then we see that a number of interventions have been filed—not nearly as many but still a slowly increasing number—in the ongoing cases brought by Gambia and South Africa, so this does seem to be a current trend.

Logistics: This is connected to the question of volume. Issues here include: how late in the proceedings can states indicate a wish to intervene? What due process issues arise if multiple states are intervening in support of one of the parties, and the other party has to deal with what they say? Do all states who are given permission to intervene have to be heard orally as well as on paper?

Legal basis: This is the question of what legal basis a state has to intervene in a particular case. And here of course as the rules stand, there are two options: Article 62: “[s]hould a state consider that it has an interest of a legal nature which may be affected by the decision in the case, it may submit a request to the Court to be permitted to intervene.” Article 63: where the construction of a treaty is in issue, any party to that treaty has a right to intervene on issues of construction.

Why do I say that there is a challenge here? Primarily because of current cases where the applicant state invokes erga omnes obligations. Does another state have an “interest of a legal nature” that may be affected by the decision, under Article 62? How should such an interest be particularized? Or should it apply under Article 63 if it is a party to the Genocide Convention, for example, which would thus confine it to making submissions on the interpretation of the provisions of the treaty, not the facts? Or should it apply under both? These are very live issues which will need to be considered by the Court in the ongoing cases under the Genocide Convention.

Looking at these challenges, is there a need for reform? And by reform of course we can include various things in the discussion, on a spectrum of formality—from amendment of the Statute itself, to amendment of the Rules, to guidance given by way of practice direction. Then of course we also have the Court’s own case law, which can and does elucidate specific issues as they arise, including for example whether a state can intervene in relation to a treaty provision in respect of which it has a reservation.

The first point to note is that the Court has already engaged in reform in this area, as part of its ongoing review of its rules and practices. In February 2024, the Court amended Articles 81, 82, and 86 of the Rules of Court.

Beginning with Article 81, the new rule clarifies the deadline for applications to intervene under Article 62, and makes it clear that these can also relate to preliminary objections, and sets a time limit in such case—something that the Court had to do ad hoc in the Ukraine case.

Article 82 deals with Article 63 interventions, and is in very similar terms. And this in fact amounts to quite a significant change in relation to Article 63, since the deadline to intervene was previously “the date fixed for the opening of the oral proceedings.” Which is impractically late, and one could imagine the chaos and unfairness caused by multiple states turning up the night before a hearing. This change is a very welcome development, and overall it is beneficial that Article 62 and 63 interventions are now procedurally aligned.

Finally, there were amendments to Article 86. Under the previous version, states intervening under Article 63 were entitled to make oral submissions at the hearing. Now the intervening state may appear during the oral phase and the Court has the power to prevent them from doing so (presumably where the Court feels it has received enough material on the papers). Again, this is a welcome introduction of procedural discretion: it remains to be seen whether the lack of a guaranteed opportunity to make oral submissions will dampen the enthusiasm of states for interventions.

The question that then arises is whether we need more changes here. I fear my answer will be quite modest. If we think about the three challenges I mentioned:

Volume: one could seek to restrict the circumstances in which states can intervene, but that would presuppose that the increased use of interventions by states is a bad thing, and to me that does not seem to be the case. It is not something one needs to discourage or restrict; it is just a question of practical management.

That takes us to my second issue, logistics. It seems to me that pragmatic procedural management is the key here. The Court has made the careful, and to my eyes very sensible, Rule amendments to adapt to particular challenges it experienced. Modest, pragmatic responses to events as they unfold seem like the way forward, within the Court’s process of review of its Rules. One question is whether states should be consulted in this process, as other courts and tribunals from time to time consult with their “users” (so for example the recent process of revision to the ICSID Rules).

Thirdly, on the issue of legal basis, it seems to me that the way forward will be the Court’s clarification of legal issues as they arise. For example the Ukraine case established that there can be interventions on the question of jurisdiction. On Article 62 and erga omnes obligations, we will see what the Court says in relation to such applications—at today’s date, from Nicaragua, Palestine, and Cuba in the South Africa case, as well as Poland in the Ukraine case.

Some commentators have made the suggestion that the amicus brief could be introduced in the ICJ, like in the European Court of Human Rights, where this has been used in interstate cases. I would be interested to hear what people think about this suggestion, but my initial reaction is that whatever merits such a process may have, it would not necessarily address any of the challenges I have identified.

To conclude with a broader point, to some extent the issue of interventions is parasitic on questions about case management more generally, including the Court’s workload more generally and the rising resort to litigation by states. My feeling is that interventions are not a free-standing target for radical reform but that they sit in the overall structure, and I hope we can discuss that more broadly.

Pierre d’Argent

Thank you also very much for this broad introduction to the topic, Alison, to which we will revert, but first we will hear the remarks from Sir Michael.

Remarks by Michael Wood

Thank you, Pierre and thank you all for coming. My answer to the question put to the panel is a firm “No!” That is assuming that by “reform” is meant some grandiose overall change; incremental changes are one matter, but “reform” tends to suggest major revision. I see no need for that. The Court has shown, particularly in recent years, that it is well able to adjust to new challenges, not least its current workload.

I could, however, point by way of example to two changes that have in my view been misguided, both in the Practice Directions. The restrictions on who can be appointed ad hoc judge are, in my view, excessive. Practice Direction VII provides that states should refrain from nominating persons who have acted as agent, counsel or advocate in another case in the three years preceding the date of the nomination. I see no justification for such a restriction, given all the other safeguards that exist. It makes it virtually impossible to find anyone with practical experience of the Court—other than former judges! Second, there is the arbitrary page-limit recently introduced into Practice Direction III. Seven hundred fifty pages may well be too few where complex facts need to be placed before the Court, especially by the applicant and especially in cases with a long historical background. Do the judges really want to be given short extracts without context? This change is hardly conducive to the proper administration of justice, and should be reversed.

I have been asked to say a word about jurisdiction and admissibility. There is much to say. For example, the practice of shoe-horning claims into a convention that happens to have an applicable compromissory clause, such as the Genocide Convention and the Racial Discrimination Convention, leads to some very artificial proceedings and should be discouraged.

I should like to make what may seem to be a radical proposal about the treatment of preliminary objections, a proposal that not all states may like. As you know, at the Court, the effect of raising preliminary objections to jurisdiction or admissibility is to stop the clock. The case is automatically diverted into a separate incidental stage, which might last a year or more. There may be good reason for this, but should it be automatic? In arbitration the tribunal is usually empowered to decide, in its discretion, whether or not the main proceedings should be halted. The Court, on the other hand, can only join the objections to the merits in exceptional circumstances. Perhaps that could be reviewed.

I shall now make two other proposals. First, I think the test for prima facie jurisdiction in provisional measures proceedings ought to be clearer, and it ought to be a higher standard than it seems to be; that would help to discourage the grandstanding that goes on in provisional measures hearings.

Second, I wonder whether provisional measure should always be binding or whether it would not have been better if the Court maintained the power to issue recommendatory measures as well as binding measures. Recommendatory provisional measures may sometimes be more effective, if only because the state to which they are directed may find it politically easier to comply. To have ruled them out, if that is what the Court did in LaGrand, would perhaps have been a mistake. In any event, as with any international instrument, just how far a measure is binding depends upon the precise wording chosen by the Court.

Pierre d’Argent

Thank you so much, Sir Michael. Janina, maybe to start the discussion, you have heard two points on provisional measures, a stricter prima facie test and the possibility for the Court to issue recommendations.

Janina Barkholdt

Thank you very much. On the prima facie jurisdiction point, I agree that the threshold needs to be a bit clearer. A different point is how exactly a higher threshold could be developed. Perhaps one could identify categories where prima facie jurisdiction is manifestly absent, for example, if there are evidently barriers or limitations to the Court’s exercise of jurisdiction. I think that the Court could develop such categories incrementally through its case law. With respect to the bindingness of provisional measures, I wonder to what extent this is possible after LaGrand given that the Court, in that case, put forward a rather long reasoning as to why provisional measures are binding under Article 41, engaging with all the rules on treaty interpretation and the travaux préparatoires, so I would be curious to see how the Court could now shift away from that reasoning.

Apart from these two specific points, I noted that there are at least three points on which there seems to be agreement among the three of us. The first one would be the concern about due process and to what the extent hearings are granted in a way that ensures equal opportunities to both parties to comment on each other. I think there should be a consistent approach by the Court in this regard, maybe even by amending the Rules of Court in a streamlined way. In the Rules of Court, the term “hearing” is often used in a way that the written procedure would suffice to grant a “hearing,” but sometimes it is understood to require an oral phase. Secondly, there seems to be a concern among the three of us about delays caused by incidental proceedings and their impact on the workload of the Court. And thirdly, that there is a need for more flexibility and discretion with respect to case management more generally. So, a thought that crossed my mind is if we might need a combination of more flexibility of the Court with respect to the organization of incidental proceedings and case management, but at the same time a stricter approach to the substantive requirements of incidental proceedings.

Pierre d’Argent

Thank you very much. Alison, you were last week in the afternoon of those hearings with Michael. Provisional measures are essentially oral proceedings, as we know, and the Court has recently been seized with modification requests, which do not call for oral proceedings, and we have seen this trend of reducing the length of the proceedings and also of limiting them to one single round. Flexibility is desirable, but is it always fair to stop after one round?

Alison Macdonald

I would say no. I think it is very difficult. I think there are fairness issues if you are going first. That is nice in a way because you get to prepare your submissions and give them, but then you have to sit through submissions to which you have no chance to respond, either on legal or factual issues, if there are things you wish to correct. If you are going second, it is really difficult, because you have got whatever was filed in writing in advance, but then you have oral submissions to which you have only a short time to respond and it is not even four hours, because you have to hand in your speech early, so it is really two hours. Of course, you can predict to a large extent what will be said, but new things can be thrown in and everything you are saying needs to be signed off. Both the morning and the afternoon party, if I can put it that way, face significant challenges in that situation and I think it is in some ways understandable given the Court’s workload, but also in some ways unsatisfactory. I think this is for one the English court lawyer in me where we have very thorough hearings, and indeed the international lawyer in me where one is used to two rounds. As a fundamental fact, it is slightly uncomfortable to have only one round, especially if it is the case that, to use a horrible phrase from arbitration, there are not post-hearing briefs. You cannot write in and even have ten pages to make some key points. If one was to keep single rounds, at least it could be mitigated by having a small written response afterwards. That is one modest idea.

Pierre d’Argent

Yes, and if there is one single round, you hope for a question by a judge and then no question comes. So, you cannot write. Michael.

Michael Wood

I think it is unfair on the applicant in a provisional measures hearing if there is only one round, because the applicant, who speaks first, usually, has no idea what will be said by the respondent. I acted for the respondent in a case like that, and I felt sorry for the applicant, because they could not respond to what we said. I think there are serious matters regarding provisional measures—they are binding and more serious than they used to be and they should be treated seriously by the Court. There should be two oral rounds; they can be brief. Alain Pellet always says he does not mind how long he has to speak, but he wants a proper gap between the sittings; to ask someone to reply in the afternoon to something that has been said in the morning is very difficult. Why are the judges doing this? I suppose it is partly because they are busy, but they are not really saving a lot of time by rushing the case in this way. The judges may not realize what the pressures are. I think they should think about that more carefully.

I had one other point. You did not ask me about this at all, but advisory opinions are a bit of a mess. They are very important, but I think the General Assembly puts the Court in a very difficult position when it asks what, while technically legal, are in fact highly political questions. President Salam said as much when he spoke last year before the ILC. He suggested that a number of cases submitted to the Court showed the dysfunction of the political organs of the UN; that states are referring matters the Court that should be dealt with by the political organs of the UN, and this posed a risk of politicizing the Court. A second point is this: the wording of the questions is sometimes quite obscure, which may reflect a wish to maximize the votes but may simply be bad drafting. The General Assembly’s rules of procedure provide that the draft question should be put to the Sixth Committee. I am not sure that would help, but anyway they should think about the questions carefully and whether they really need to ask them. Third, on advisory opinions, I think the procedure needs real improvement. The French alphabetical order is particularly bad if you are Germany, for example, because you speak first and then everyone criticizes you. It is usually better to come toward the end, which provides at least some opportunity to respond to points made by others. Of course, coming first one can set the scene, which may be important if you are a central player, as Palestine has recently been. Perhaps the Court should use the English alphabetical order sometimes. Or perhaps do what the General Assembly does, which is take the first name out of the hat. Then at least you are not always the first or the last to speak. I think there are other things that could be thought about carefully with advisory opinion. Having a single round is pretty hopeless. Having two rounds may be impractical if ninety countries are going to take part. But something could be done to ensure that there is a bit of a dialogue between the various sides, and the trouble is there are numerous sides.

Pierre d’Argent

Thank you for bringing the issue of advisory opinions to the debate. I am looking forward to the discussion.