Separate opinion of Judge Xue

SEPARATE OPINION OF JUDGE XUE
1. Notwithstanding my vote on subparagraph (4) of the operative part of the Judgment, I
wish to place on record my disagreement with the reasoning in relation to the location of the
starting-point of the land boundary between the Parties and the way in which this issue is treated in
the Maritime Delimitation case.
2. First of all, I am of the view that, under the 1858 Treaty of Limits, the Cleveland Award
and the Alexander Awards, the starting-point of the land boundary should be located on the
north-eastern end of the Harbor Head Lagoon rather than at the end of the sandspit of Isla Portillos
at the mouth of the San Juan River (right bank).
3. In this joint case, identification of the starting-point of the land boundary is an essential
issue, both for the determination of the territorial sovereignty of the coast in dispute and for the
maritime delimitation between the Parties in the Caribbean Sea. The Parties do not disagree that the
1858 Treaty of Limits, the Cleveland Award and the Alexander Awards constitute the legal basis
for the determination of the land boundary between the two countries. Notwithstanding the
continuous geographical changes in the Isla Portillos in the course of the last one and half century,
the provisions of the 1858 Treaty and the terms of the arbitral awards remain applicable. In other
words, the starting-point of the land boundary has to be determined on the basis of these legal
documents.
4. By its Order of 31 May 2016, the Court decided to appoint two experts to conduct site
visits to the coast of the northern part of Isla Portillos and to advise the Court regarding the state of
the coast between the point suggested by Costa Rica and the point suggested by Nicaragua in their
pleadings as the starting-point of the maritime boundary in the Caribbean Sea. The Court put the
following questions for the experts to answer:
“(a) What are the geographical co-ordinates of the point at which the right bank of the
San Juan River meets the sea at the low-water line?
(b) What are the geographical co-ordinates of the land point which most closely
approximates to that identified by the first Alexander Award as the starting-point
of the land boundary?
(c) Is there a bank of sand or any maritime feature between the points referred to in
subparagraphs (a) and (b) above? If so, what are their physical characteristics? In
particular, are these features, or some of them, permanently above water, even at
high tide? Is Los Portillos/Harbor Head Lagoon separated from the sea?
(d) To what extent is it possible, or probable, that the area concerned will undergo
major physical changes in the short and long term?”
5. In their report, the Court-appointed experts located the original starting-point of the land
boundary, now submerged under the sea, in accordance with the terms of the 1858 Treaty and the
arbitral awards, and marked the geographical co-ordinates of the land point which most closely
approximates to that identified by the first Alexander Award as the starting-point of the land
boundary.
- 2 -
6. The experts’ report demonstrates that the initial segment of the land boundary, including
its starting-point, remains identifiable and actually identified. What is left of Harbor Head Lagoon
and the accreted sandbar separating the lagoon and the sea is a broken part of the land boundary,
now enclaved within Costa Rica’s territory. The experts’ answer to the first question in fact
identified the current location of the point at which the San Juan River reaches the sea, in other
words, the place where the original land boundary breaks.
7. In the present Judgment, the Court considers that it has determined the starting-point of
the land boundary in its 2015 Judgment in the case concerning Certain Activities Carried Out by
Nicaragua in the Border Area, where it interpreted the 1858 Treaty as providing that “the territory
under Costa Rica’s sovereignty extends to the right bank of the Lower San Juan River as far as its
mouth in the Caribbean Sea” (Certain Activities Carried Out by Nicaragua in the Border Area,
(Costa Rica v. Nicaragua) and Construction of a Road in Costa Rica along the San Juan River
(Nicaragua v. Costa Rica), Judgment, I.C.J. Reports 2015 (II), p. 703, para. 92). A closer perusal
of the relevant paragraphs of the 2015 Judgment shows that this interpretation of the said Judgment
is questionable.
8. In the first place, the location of the starting-point of the land boundary did not fall within
the scope of the jurisdiction of the Court in that case. That issue depends on the determination of
the sovereignty over the coast of the northern part of Isla Portillos. If the coast belonged to
Nicaragua, the land boundary should extend eastward to Harbor Head Lagoon; otherwise, the
boundary would start at the mouth of the river on the western side of Isla Portillos. In the present
case, the question whether the watercourse that channelled the San Juan River and Harbor Head
Lagoon still bears on who has the sovereignty over the coast of the northern part of Isla Portillos.
By interpreting paragraphs 69 and 70 of the 2015 Judgment, the Court states that, “no decision was
taken by the Court in its 2015 Judgment on the question of sovereignty concerning the coast of the
northern part of Isla Portillos, since this question had been expressly excluded”. This conclusion
means that the status of the last segment of the land boundary including its starting-point, as
determined by General Alexander, was yet to be determined.
9. Moreover, should the Court have determined the starting-point of the land boundary as at
the mouth of the San Juan River in the 2015 Judgment, it would not have been reasonable for the
Court to instruct the experts to relocate the original starting-point of the land boundary and to find
the geographic co-ordinates of the land point that most closely approximates to the original
starting-point identified by General Alexander, because the boundary would follow the natural
course of the San Juan River to the sea and it would be pointless to identify these points.
10. Although the drafters of the 1858 Treaty and the arbitral awards well anticipated that the
land boundary would necessarily be affected by gradual or sudden coastal changes in the future,
they did not specifically spell out what principles of international law would apply in the event of
such changes. Although it was mentioned that “[t]he ownership of any accretion to said Punta de
Castilla is to be governed by the laws applicable to that subject” (see 1888 Cleveland Award,
United Nations, Reports of International Arbitral Awards (RIAA), Vol. XXVIII, p. 209), the
situation of what now stands as partial disappearance of the watercourse was not envisaged.
11. It is true that General Alexander made it clear that in the practical interpretation of the
1858 Treaty, the San Juan River must be considered a navigable river. However, if the startingpoint
of the boundary is to be automatically determined by the river’s outlet to the sea, it would be
difficult to explain why both Parties agree that Harbor Head Lagoon belongs to Nicaragua rather
than Costa Rica; since the watercourse has now reached the Caribbean Sea at the mouth of the
- 3 -
San Juan River, what is on the right bank of the River, including Harbor Head Lagoon, should
automatically be merged with Costa Rica’s territory.
12. When the Court, on the basis of the experts’ report, determines that there is no longer any
water channel connecting the San Juan River with Harbor Head Lagoon and therefore the coast of
the northern part of Isla Portillos belongs to Costa Rica, it virtually states that the land boundary is
disrupted at the mouth of the San Juan River by the natural change of the coast (see sketch-map
No. 2 of the Judgment, reproduced below).
13. The Court’s decision that Harbor Head Lagoon and the sandbar separating it from the
Caribbean Sea are under Nicaragua’s sovereignty (Judgment, para. 73) cannot simply be attributed
to the agreement of the Parties; the underlying reason is Costa Rica’s recognition that the line
around Harbor Head Lagoon still constitutes part of the land boundary, albeit disconnected with the
rest of the land boundary.
14. Situations with water boundaries vary from case to case. There is no established rule of
customary international law governing the legal impact of watercourse change on boundaries
(e.g. Anzilotti, Bardonnet, Bouchez, Caflisch)1. In the present case, so far as the land boundary is
1 D. Bardonnet, “Frontières terrestres et frontières maritimes”, Annuaire français de droit international, Vol. 35,
1989, pp. 10-11, citing D. Anzilotti’s 1914 study of State practice published in the Rivista italiana di diritto
internazionale; L.J. Bouchez, “The Fixing of Boundaries in International Boundary Rivers”, International and
Comparative Law Quarterly, Vol. 12, 1963, p. 807; L. Caflisch, Règles générales du droit des cours d’eau
internationaux, Recueil des cours, Vol. 219, 1989, p. 82.
- 4 -
concerned, there are two relevant factors that should be taken into account. First, the starting-point
of the land boundary, even after being relocated, remains in an unstable situation. As the experts
pointed out in their report,
“the position of the mouth of the San Juan River experiences continuous variations,
mainly related to changes in the spit of Isla Portillos, i.e., westward growth by
accumulation of sand and destruction by erosion. The growth of the spit by sediment
accretion is a progressive process, whereas its destruction, including the opening of
channels, may occur rapidly by strong waves (e.g., hurricanes) and floods of the
San Juan River. Consequently, the mouth of the San Juan River and its right bank are
highly mobile.” (Expert Opinion, 30 April 2017, p. 42, para. 117; emphasis added.)
To maintain stability and certainty of the boundary, more weight should be given to its legal title
than to the factual change on the ground. Secondly, the enclave resulting from the break-up of the
land boundary is not a self-standing geographical feature as such; until the Court’s present decision
on the sovereignty of the coast of the northern part of Isla Portillos, it formally constituted part of
the land boundary. Whether Harbor Head Lagoon and its frontal sandbar would eventually
disappear as a result of coastal recession, as claimed by Costa Rica, the enclave, as it currently
stands, should form part of the geomorphological circumstances of the coast for the maritime
delimitation, a point I now turn to.
15. The location of the starting-point of the land boundary is not a hard issue. So far as the
land boundary is concerned, it does not matter where to locate its starting-point; whether it is
identified at the eastern headland of Harbor Head Lagoon, or at the mouth of the San Juan River as
a result of the disappearance of the water channel, the boundary continues to serve its purposes
well despite the break-up of the initial segment of the boundary. What matters in this case is the
impact of the coastal change on the maritime delimitation. Although the Court takes cognition of
the great instability of the coastline in the area of the mouth of the San Juan River, it does not give
sufficient consideration to the coastal relationship between the Parties. With Costa Rica’s coast
now situated between Nicaragua’s territories, Harbor Head Lagoon on the eastern side and the river
mouth on the western side, it would be difficult, if not impossible, to choose a starting-point on
land that would genuinely reflect a median point. Either way, there would be some cut-off effect to
the detriment of one Party. That is to say, to use the point suggested by Nicaragua as the startingpoint
of the maritime boundary, Costa Rica’s coast would be cut off from the sea. To use the point
suggested by Costa Rica, on the other hand, there would also be some cut-off effect on Nicaragua’s
enclave.
16. The Court stated in the Nicaragua v. Honduras case that, “[n]othing in the wording of
Article 15 suggests that geomorphological problems are per se precluded from being ‘special
circumstances’ within the meaning of the exception, nor that such ‘special circumstances’ may only
be used as a corrective element to a line already drawn” (Territorial and Maritime Dispute between
Nicaragua and Honduras in the Caribbean Sea (Nicaragua v. Honduras), Judgment,
I.C.J. Reports 2007 (II), p. 744, para. 280). In the present case, the geomorphological conditions of
the coast of the northern part of Isla Portillos and the break-up of the land boundary constitute such
special circumstances.
17. I agree with the majority that given the prevailing circumstances of the coast and the
current location of the mouth of the San Juan River, it is reasonable and equitable to draw the
provisional median line from the coast on the western side of Isla Portillos near the mouth of the
San Juan River. I doubt, however, the wisdom to select as the starting-point of the maritime
boundary a point on the solid land closest to the mouth of the river, currently identified as point Pv.
- 5 -
This is because, first, that point is equally unstable and secondly, little consideration is given to
Nicaragua’s access to Harbor Head Lagoon.
18. In paragraph 105 of the Judgment, the Court recognizes that the situation of the enclave
is a special circumstance and calls for “a special solution”. It nevertheless considers that “[s]hould
territorial waters be attributed to the enclave, they would be of little use to Nicaragua, while
breaking the continuity of Costa Rica’s territorial sea”. Therefore, the delimitation in the territorial
sea between the Parties will not take into account any entitlement which might result from the
enclave. In my opinion, this is not a convincing reasoning to ignore Nicaragua’s entitlement from
the enclave, no matter how small it is.
19. The prevailing geographical phenomenon of the coast is instability. In considering the
special circumstances in the delimitation of the territorial sea, the Court does not attach much
importance to the experts’ advice that the overall coast will undergo continuous changes due to
coastal erosion and whether the mouth of the San Juan River would move further westwards or
eastwards is unpredictable. Indeed, the Court cannot base its decision on the prediction of future
changes, but on the factual situation of today. To treat the enclave as negligible, however, in my
view, cannot be regarded as “a special solution”.
20. In order to overcome the difficulty arising from the repositioning of the starting-point of
the land boundary at the mouth of the San Juan River as a result of the disappearance of the
watercourse along the coast, the starting-point of the maritime delimitation, in my opinion, can be
detached from the starting-point of the land boundary. To provide access to Harbor Head Lagoon
for Nicaragua, in light of the geographical situation of the coast, the maritime boundary may start
from a fixed point (the same as the hinge point) on the median line at a distance of 2 nautical miles
from the coast without being connected with a mobile line to a point on land. Although with
2 nautical miles’ territorial sea undelimited, this approach would place the Parties in a better
position to manage their coastal relations, particularly in respect of navigation. It would not be the
first time that a delimitation begins at some distance out to the sea; the judicial and arbitral
practices support such a resolution where there is an uncertain land boundary terminus (see, for
example, Territorial and Maritime Dispute between Nicaragua and Honduras in the Caribbean
Sea (Nicaragua v. Honduras), Judgment, I.C.J. Reports 2007 (II), p. 756, para. 311; Case
concerning the delimitation of the maritime boundary between Guinea and Guinea-Bissau, Award
of 14 February 1985, United Nations, RIAA, Vol. XIX, pp. 149–196).
(Signed) XUE Hanqin.
___________