After Intervention: The Persistent Absent Third Party Issue in the Sapodilla Cayes Case

On 19 March, the ICJ delivered a judgment allowing Guatemala to intervene as a non-party in Sovereignty over the Sapodilla Cayes/Cayos Zapotillos (Belize v. Honduras). In this case, Belize is seeking a declaration that, as opposed to Honduras, it “is sovereign over the Sapodilla Cayes”, a number of islands off the shores of Belize, Guatemala, and Honduras (Application Instituting Proceedings, para. 22). The Sapodilla Cayes are also part of the subject-matter of a dispute pending at the Court between Guatemala and Belize, which concerns all of Guatemala’s territorial claims against Belize (Guatemala’s Territorial, Insular and Maritime Claim (Guatemala/Belize) (on the background of the two cases, Guatemala’s intervention and the Court’s decision, see here and here).

While the intervention issue is now resolved, it remains unclear whether the Court can decide the Sapodilla Cayes case on the merits. As argued here, Guatemala could constitute an indispensable third party and thus bar the Court from exercising its jurisdiction under the Monetary Gold principle. This post will first examine how Monetary Gold could apply to a territorial dispute like the Sapodilla Cayes cases. It will then show how neither Guatemala’s consent to adjudicate the issue in its dispute with Belize nor its intervention in Sapodilla Cayes would resolve the issue. However, a potential solution could rely on the already established “givens” doctrine, i.e., relying on legal determinations made in other cases. While that doctrine has traditionally relied on past decisions, this case would require a slight procedural innovation enabling the Court to rely on simultaneous determinations.

Monetary Gold and territorial claims

The Monetary Gold principle goes back to the eponymous 1954 case of Monetary Gold Removed from Rome in 1943 (Italy v. France, UK and US), in which the Court held that “[w]here […] the vital issue to be settled concerns the international responsibility of a third state, the Court cannot, without the consent of that third state, give a decision on that issue […]” (Monetary Gold, p. 33). This enshrines a higher standard than that required for intervention, namely that the third state’s legal interest must not merely be affected by the decision, but form its “very subject-matter” (Monetary Gold, p. 32). The Court has since elaborated and developed this framework as it relates to the responsibility of absent third parties (see, e.g., Certain Phosphate Lands in Nauru (Nauru v. Australia) Preliminary Objections (1992)).

However, the exact contours of how Monetary Gold applies to territorial disputes remain unclear. While the Court has always stressed the importance of safeguarding absent states’ rights in delimitation cases (Land and Maritime Boundary between Cameroon and Nigeria, Preliminary Objections (1998) para. 116), it has never declared a case inadmissible because of it. Instead, it has usually protected absent third states by indicating the hypothetical vector of boundaries between the parties without fixing a precise end point, sometimes explicitly relying on Monetary Gold to do so (see, e.g., Territorial and Maritime Dispute between Nicaragua and Honduras in the Caribbean Sea (2007) para. 312; see Thienel at p. 350).

Moreover, the Court held that certain types of territorial disputes, such as the location of a tripoint along an established boundary with a third state (Land and Maritime Boundary between Cameroon and Nigeria, Preliminary Objections (1998) para. 79), or the existence of a condominium between the parties potentially shared by a third state (Land, Island, and Maritime Frontier Dispute (El Salvador / Honduras), Application by Nicaragua for Permission to Intervene (1990) para. 73), did not meet the threshold of the Monetary Gold principle. This has prompted Kolb to state that “[i]t is only in highly exceptional circumstances that one can imagine the Monetary Gold rule being applied to the delimitation process.” (Kolb at p. 578).

However, the Sapodilla Cayes case is different in that it does not concern the delimitation of a boundary, but sovereignty over distinct islands. The Court highlighted this difference in the Sapodilla Cayes Intervention Judgment:

In [Territorial and Maritime Dispute (Nicaragua v. Colombia)], the Court recognized that, although Costa Rica possessed a legal interest in the maritime area under consideration, its interest would not be “affected” by the Court’s decision since it remained possible for the Court, when delimiting the parties’ maritime boundary, to end the line “before it reaches an area in which the interests of a legal nature of third states may be involved” […]. Such an approach is not open to the Court in the present case, where the question does not concern the drawing of a maritime boundary, which is susceptible to being indicated in a particular direction “until it reaches the area where the rights of third states may be affected”, thereby avoiding any potential intrusion upon the rights (or interests) of a third state […]. Rather, Belize and Honduras have asked the Court to determine the holder of sovereignty over specific maritime features in respect of which Belize and Honduras, as well as Guatemala, have advanced competing claims. For the Court to answer the question before it in the present case, it must necessarily decide whether Belize or Honduras (or neither) has sovereignty over the cays. Consequently, any judgment rendered on the merits may have a direct bearing upon Guatemala’s interest, since it is one of the claimants to sovereignty over the same maritime features in a separate case before the Court.

(Sapodilla Cayes Intervention Judgment, para. 40 (references omitted))

Of course, this statement deals with the lower standard of Art. 62 of the Statute and not Monetary Gold. However, what the Court identified as the “exclusive” nature of sovereignty determination (Sapodilla Cayes Intervention Judgment at para. 45) might prove to be a problem under Monetary Gold as well. As shown above, in the past, the Court has used the line-indication solution which it rejects here not only to deny the existence of a legal interest, but, as in Territorial and Maritime Dispute between Nicaragua and Honduras, has decided infra petita on the merits explicitly to protect the rights of third states. Where this is not possible, Monetary Gold might come into play, leading the Court to refrain from exercising its jurisdiction in the Sapodilla Cayes case. Likewise, it might decide infra petita in Guatemala/Belize and refuse to decide the claims concerning sovereignty over the Sapodilla Cayes.

Potential Solutions? Intervention, consent, and “givens”

First, under the Court’s jurisprudence, Guatemala’s non-party intervention is most likely not enough to resolve the Monetary Gold issue in the present case. The Monetary Gold doctrine requires the third state’s consent to the Court’s jurisdiction. Non-party intervention does not require a jurisdictional link between the parties and the intervenor. Thus, it does not provide the Court with jurisdiction over the intervenor the same way it would have over a party, and the intervenor is not bound by the judgment like a party would be (Sapodilla Cayes Intervention Judgment at para. 61). Consistent with this, where states have intervened as non-parties, the Court still analyzes whether their interests are implicated in an impermissible way, including under the Monetary Gold principle (See, e.g., Land and Maritime Boundary between Cameroon and Nigeria, Merits (2002) para. 238; Jurisdictional Immunities of the state (Germany v. Italy: Greece intervening) (2012) para. 127).

Second, the fact that all three states have submitted this issue to the Court’s jurisdiction in separate proceedings is unlikely to be enough to pre-empt the Monetary Gold problem. The principle does not simply require that the Court has jurisdiction over the absent third party in the subject matter. Rather, the absent third party must either be a party to the specific proceedings or consent to the Court settling the trilateral dispute without its involvement (on the latter possibility see Arbitral Award of 3 October 1899 (Guyana v. Venezuela) Preliminary Objection (2023) paras. 96-107).

However, the Court might use these existing titles of jurisdiction to resolve the issue under the so-called “givens” doctrine. This exception to the Monetary Gold principle states that the Court may rely on prior authoritative legal findings concerning the legal interest of absent third parties. This exception was already implicit in the Court’s reasoning in Monetary Gold itself, where it did not take issue with Albania’s responsibility vis-à-vis the United Kingdom, which it had already determined in Corfu Channel. The Court later more explicitly entertained the possibility in East Timor that legal determinations in UN General Assembly and Security Council resolutions could serve as givens (East Timor (Portugal v. Australia) (1995) para. 32). Most recently, ITLOS relied on the ICJ’s Chagos Advisory Opinion as a given in Delimitation of the Maritime Boundary between Mauritius and Maldives in the Indian Ocean (Mauritius/Maldives) Preliminary Objections (2021) (see generally Stendel & Wentker at pp. 589-92).

Concurrent givens?

Thus, using the givens doctrine, the ICJ could take advantage of the dual pending cases before it and the fact that all parties to the sovereignty dispute did consent to its jurisdiction over this dispute. However, since both cases have indispensable third parties concerning the same issue, the Court cannot resolve one before the other. Therefore, this case would require two simultaneous decisions which mutually provide each other the given legal determination on Guatemala’s and Honduras’s respective claims to sovereignty over the Sapodilla Cayes. Besides the broader question of whether the Court would go for such an Escherian solution, this raises the technical issue whether it even has the ability to render two judgments at the exact same time.

Joint judgment or common action?

In consolidated or joined cases, the Court has previously issued joint judgments (see South West Africa (Ethiopia v. South Africa; Liberia v. South Africa) Second Phase (1966); North Sea Continental Shelf Cases (Federal Republic of Germany / Denmark; Federal Republic of Germany / Netherlands) (1969)). As these judgments form integrated decisions announced by the Court at the same time (see, e.g., the Court’s decision announcement in South West Africa), there can be little doubt that all parts of these judgments became binding on the parties at the same time (see Rosenne at p. 110, speaking of “formal joinder of proceedings leading to a single judgment (of which the operative clauses may be different)”; see also North Sea Continental Shelf at para. 11). However, while in the two cases at hand, a formal joinder is not impossible, it is unlikely because Guatemala/Belize has a significantly broader scope and both parties in the Sapodilla Cayes have opposed joinder (see, e.g., the statement of Honduras’s Co-Agent at para. 19, which also referenced Belize’s unpublished memorial to that effect; material overlap and party consent as factors may not be strictly necessary, but highly relevant, see, e.g., Kolb at p. 1000; cf. Toh at pp. 18-19).

However, under Article 47 of the Rules, the Court may also direct less formal “common action” short of a joinder to improve coordination. This is also most likely the legal basis for what the Court referred to in the Sapodilla Cayes Intervention Judgment as “certain case management measures [which] might improve co-ordination between the two sets of proceedings” (at para. 42). While, the Court also referred to certain limits of these measures, it has on several occasions delivered judgments in parallel cases on the same day and in the same sitting (see, e.g., the decision announcements in Nuclear Tests, Aerial Incident at Lockerbie, and Legality of Use of Force; but see Fisheries Jurisdiction, where the Court delivered the judgment in the case brought by the UK in the morning and in the case brought by Germany in the evening session of the same day). However, this raises the question of whether same-day judgments would be enough to qualify as concurrent determinations.

When exactly does the ICJ render a judgment?

Art. 94(2) of the Rules of Court states that “[t]he judgment shall become binding on the parties on the day of the reading.” A plain reading of this provision would suggest that all judgments rendered on the same day therefore become binding on the parties at the same time. However, some scholars have raised objections to this interpretation since that would give the obligations retroactive effect. Thus, Rosenne suggested in his Commentary on the 1978 Rules that this provision should be interpreted to mean “the time at which the judgment is read in open court (when copies are also handed to the parties)” (Rosenne at p. 194). In a similar vein, Khan suggests in the Statute Commentary that the judgment should become binding at “the hour and minute of the closure of the sitting at which the judgment has been read” to increase publicity (Khan at para. 35).

Thus, the solution depends on the interpretation of Art. 94(2). However, it seems that regardless of which point in time one adopts as decisive, it should be possible for the Court to ensure that the judgments do become binding at the same time. If it delivers the judgments at the same sitting as in Nuclear Tests, Aerial Incident at Lockerbie, and Legality of Use of Force, the plain reading as well as Khan’s interpretation would be satisfied. Rosenne’s interpretation seems more difficult to satisfy, but at least handing the copies out simultaneously seems reasonable. Alternatively, and to remove uncertainty, the Court could declare in the judgments the exact point at which it considers that they become binding on the parties, either as an act of interpretation of Art. 94(2) of the Rules or as an adoption of an (informal) procedural rule.

Outlook: A legitimate solution?

Of course, one might question whether this is more than a procedural parlor trick to introduce trilateral dispute resolution through the backdoor. However, in any run of the mill Monetary Gold type case, the interest in protecting absent third states is already in inherent tension with the court’s objective of peacefully settling disputes. Where the absent state has consented to the Court’s jurisdiction over the same issue in concreto, but just not with regard to all parties involved, it arguably requires less protection, which could tip the scale in favor of settling the dispute. Concurrent givens would provide the Court with an avenue of effecting this balancing act without straying from its established jurisprudence.

Granted, this solution seems to be very similar to the second one presented and rejected above, i.e. merely relying on Guatemala’s consent in the Guatemala/Belize case. However, the givens doctrine stands on more solid doctrinal ground. The only step required is extending it to simultaneous determinations, which seems less dramatic than ITLOS’s shift to include Advisory Opinions as givens in Mauritius/Maldives. In effect, this would allow the Court to continue its trend of carving out exceptions to the Monetary Gold principle, as it has done since coining its modern formulation in Nauru and East Timor.