Treaties & Conventions

Eritrea - Yemen Arbitration

Documents

AWARD
Phase I: Territorial Sovereignty and Scope of Dispute

CHAPTER X – Conclusions

440. Having examined and analysed in great detail the extensive materials and evidence presented by the Parties(25), the Tribunal may now draw the appropriate conclusions.

Ancient Title

441. First there is the question of an "ancient title" to which Yemen attaches great importance; moreover the Agreement for Arbitration requires the Tribunal to decide the question of sovereignty "on the basis in particular of historic titles". Yemen contends that it enjoys an ancient title to "the islands", which title existed before the hegemony of the Ottoman Empire and indeed emanates from medieval Yemen. It contends, moreover, that this title still subsisted in international law at the time when the Turks were defeated at the end of the First World War, and that therefore, when the Ottoman Empire renounced their generally acknowledged sway over the islands by the Treaty of Lausanne in 1923, the right to enjoy that title in possession "reverted" to Yemen.

442. This is an interesting argument and one that raises a number of questions concerning the international law governing territorial sovereignty. No one doubts that during the period of the Ottoman Empire - certainly in the second Ottoman period 1872-1918 - the Ottomans enjoyed possession of, and full sovereignty over, all the islands now in dispute, and thus not only factual possession but also a sovereign title to possession. When this regime ceased in 1923, was there a "reversion" to an even older title to fill a resulting vacuum?

443. It is doubted by Eritrea whether there is such a doctrine of reversion in international law. This doubt seems justified in view of the fact that very little support for such a doctrine was cited by Yemen, nor is the Tribunal aware of any basis for maintaining that reversion is an accepted principle or rule of general international law. Moreover, even if the doctrine were valid, it could not apply in this case. That is because there is a lack of continuity. It has been argued by Yemen that in the case of historic title no continuity need be shown, but the Tribunal finds no support for this argument.

444. Yemen's argument is difficult to reconcile with centuries of Ottoman rule over the entire area, ending only with the Treaty of Lausanne (see Chapter V, above). This is the more so because, under the principle of intertemporal law, the Ottoman sovereignty was lawful and carried with it the entitlement to dispose of the territory. Accepting Yemen's argument that an ancient title could have remained in effect over an extended period of another sovereignty would be tantamount to a rejection of the legality of Ottoman title to full sovereignty.

445. The Treaty of Lausanne did not expressly provide, as the Treaty of Sèvres would have done, that Turkey renounced her territorial titles in favour of the Allied Powers; which provision would certainly have excluded any possibility of the operation of a doctrine of reversion. Yemen was not a party to the Treaty of Lausanne, which was therefore res inter alios acta. Nevertheless, none of the authorities doubts that the formerly Turkish islands were in 1923 at the disposal of the parties to the Lausanne Treaty, just as they had formerly been wholly at the disposal of the Ottoman Empire, which was indeed party to the treaty and in it renounced its sovereignty over them. Article 16 of the Treaty created for the islands an objective legal status of indeterminacy pending a further decision of the interested parties; and this legal position was generally recognised, as the considerable documentation presented by the Parties to the Tribunal amply demonstrates. So, it is difficult so see what could have been left of such a title after the interventions of the Ottoman sovereignty which was generally regarded as unqualified; and its replacement by the Article 16 regime which put the islands completely at the disposal of the "interested parties".

446. There is a further difficulty. Yemen certainly existed before the region came to be under the domination of the Ottomans. But there must be some question whether the Imam, who at that period dwelt in and governed a mountain fortress, had had sway over "the islands". Further, there is the problem of the sheer anachronism of attempting to attribute to such a tribal, mountain and Muslim medieval society the modern Western concept of a sovereignty title, particularly with respect to uninhabited and barren islands used only occasionally by local, traditional fishermen.

447. In keeping with the dictates of the Arbitration Agreement, both Parties, and Yemen especially, have placed "particular" emphasis on historic titles as a source of territorial sovereignty. They have, however, failed to persuade the Tribunal of the actual existence of such titles, particularly in regard to these islands.

448. Eritrea's claims too, insofar as they are said to be derived by succession from Italy through Ethiopia, if hardly based upon an "ancient" title, are clearly based upon the assertion of an historic title. There is no doubt, as has been shown in chapters V, VI and VII above, that Italy in the inter-war period did entertain serious territorial ambitions in respect of the Red Sea islands; and did seek to further these ambitions by actual possession of some of them at various periods. Major difficulties for the Eritrean claims through succession are, as has been shown above in some detail, first the effect of Article 16 of the Treaty of Lausanne of 1923, and later the effects of the provisions of the Italian Peace Treaty of 1947. But there is also the fact that the Italian Government, in the inter-war period, constantly and consistently gave specific assurances to the British Government that Italy fully accepted and recognized the indeterminate legal position of these islands as established by treaty in 1923. No doubt Italy was hoping that the effect of her active expansionist policies might eventually be that "the parties concerned" would be persuaded to acquiesce in a fait accompli. But that never happened.

449. So there are considerable problems for both Parties with these versions of historic title. But the Tribunal has made great efforts to investigate both claims to historic titles. The difficulties, however, arise largely from the facts revealed in that history. In the end neither Party has been able to persuade the Tribunal that the history of the matter reveals the juridical existence of an historic title, or of historic titles, of such long-established, continuous and definitive lineage to these particular islands, islets and rocks as would be a sufficient basis for the Tribunal's decision. And it must be said that, given the waterless and uninhabitable nature of these islands and islets and rocks, and the intermittent and kaleidoscopically changing political situations and interests, this conclusion is hardly surprising.

450. Both Parties, however, also rely upon what is a form of historic claim but of a rather different kind; namely, upon the demonstration of use, presence, display of governmental authority, and other ways of showing a possession which may gradually consolidate into a title; a process well illustrated in the Eastern Greenland case, the Palmas case, and very many other well-known cases. Besides historic titles strictly so-called the Tribunal is required by the Agreement for Arbitration to apply the "principles, rules and practices of international law"; which rubric clearly covers this kind of argument very familiar in territorial disputes. The Parties clearly anticipated the possible need to resort to this kind of basis of decision - though it should be said that Yemen expressly introduces this kind of claim in confirmation of its ancient title, and Eritrea introduces this kind of claim in confirmation of an existing title acquired by succession - and the great quantity of materials and evidences of use and of possession provided by both Parties have been set out and analysed in Chapter VII above, together with chapter VIII on maps and Chapter IX on the history of the petroleum agreements. It may be said at once that one result of the analysis of the constantly changing situation of all these different aspects of governmental activities is that, as indeed was so in the Minquiers and Ecrehos(26) case where there had also been much argument about claims to very ancient titles, it is the relatively recent history of use and possession that ultimately proved to be a main basis of the Tribunal decisions. And to the consideration of these materials and arguments this Award now turns.

Evidences of the Display of Functions of State and Governmental Authority

451. These materials have been put before the Tribunal by the Parties with the intention of showing the establishment of territorial sovereignty over the islands, in Judge Huber's words in the Palmas case,(27) "by the continuous and peaceful display of the functions of state within a given region." But the kind of actions that may be deployed for this purpose has inevitably expanded in the endeavour to show what Charles de Visscher named a gradual "consolidation" of title. Accordingly, the Tribunal is faced in this case with an assortment of factors and events from many different periods, intended to show not only physical activity and conduct, but also repute, and the opinions and attitudes of other governments (the different classes of materials are set out above in Chapter VII).

452. It is well known that the standard of the requirements of such activity may have to be modified when one is dealing, as in the present case, with difficult or inhospitable territory. As the Permanent Court of International Justice said in the Legal Status of Eastern Greenland case, "[I]t is impossible to read the records of the decisions in cases as to territorial sovereignty without observing that in many cases the tribunal has been satisfied with very little in the way of the actual exercise of sovereign rights, provided that the other state could not make a superior claim."(28)

453. This raises, however, a further important question of principle. The problem involved is the establishment of territorial sovereignty, and this is no light matter. One might suppose that for so important a question there must be some absolute minimum requirement for the acquisition of such a right, and that in principle it ought not normally to be merely a relative question.

454. It may be recalled that this question of principle did arise in the Palmas case, but there Huber was able to meet it by appealing to the particular terms of the compromis, which, said Huber, "presupposes for the present case that the Island of Palmas (or Miangas) can belong only to the United States or to the Netherlands and must form in its entirety part of the territory either of the one or of the other of these two Powers, parties to the dispute," and "[t]he possibility for the arbitrator to found his decision on the relative strength of the titles invoked on either side must have been envisaged by the parties to the Special Agreement."

455. The Arbitration Agreement in the present case, however, is in different and even unusual terms. The Tribunal is required only to make "an award on territorial sovereignty" and "to decide the sovereignty". The compromissory provision which led Huber to the possibility of deciding only on the basis of a marginal difference in weight of evidence cannot be said to apply in the present case.

456. There is certainly no lack of materials, evidence, or of arguments in the present case. The materials, on the contrary, are voluminous and the result of skilled research by the teams of both Parties, and of the excellent presentations by their counsel. But what these materials have in fact revealed is a chequered and frequently changing situation in which the fortunes and interests of the Parties constantly ebb and flow with the passage of the years. Moreover, it has to be remembered that neither Ethiopia nor Yemen had much opportunity of actively and openly demonstrating ambitions to sovereignty over the islands, or of displaying governmental activities upon them, until after 1967, when the British left the region. For, as shown above, the British were constantly vigilant to maintain the position effected by the Treaty of Lausanne that the legal position of "the islands" was indeterminate.

457. In these circumstances where for all the reasons just described the activities relied upon by the parties, though many, sometimes speak with an uncertain voice, it is surely right for the Tribunal to consider whether there are in the instant case other factors which might help to resolve some of these uncertainties. There is no virtue in relying upon "very little" when looking at other possible factors might strengthen the basis of decision.

458. An obvious such factor in the present case is the geographical situation that the majority of the islands and islets and rocks in issue form an archipelago extending across a relatively narrow sea between the two opposite coasts of the sea. So there is some presumption that any islands off one of the coasts may be thought to belong by appurtenance to that coast unless the state on the opposite coast has been able to demonstrate a clearly better title. This possible further factor looks even more attractive when it is realised that its influence can be seen very much at work in the legal history of these islands; beginning indeed with the days of Ottoman rule when even under the common sovereignty of the whole region it was found convenient to divide the jurisdiction between the two coastal local authorities (see paras. 132-136, above). Moreover, in the present case, the examination of the activities material itself shows very clearly that there was no common legal history for the whole of this Zuqar-Hanish archipelago; some of the evidence not surprisingly refers to particular islands or to sub-groups of islands.

459. Thus the Tribunal has found it necessary, in order to decide the question of sovereignty, to consider the several subgroups of the islands separately, if only for the reason that the different subgroups have, at least to an important extent, separate legal histories; which is only to be expected in islands that span the area between two opposite coasts. This may seem only a natural or even manifest truth, but Yemen in particular has emphasised the importance it attaches to what it calls a principle of natural unity of the islands, and some comment on this theory is therefore required.

Natural and Physical Unity

460. Yemen's pleadings insist strongly on what it calls "the principle of natural or geophysical unity" in relation to the Hanish group of islands; Yemen uses the name of the "Hanish Group" both in its texts and in its illustrative maps to encompass the entire island chain, including the Haycocks and the Mohabbakahs (the present comments do not of refer of course to the northern islands of Jabal al-Tayr and the Zubayr group, which will be considered separately later on).

461. This "principle" is described in Chapter 5 of the Yemen Memorial, where impressive authority is cited in support of it, including Fitzmaurice, Waldock and Charles de Visscher. That there is indeed some such concept cannot be doubted. But it is not an absolute principle. All these authorities speak of it in terms of raising a presumption. And Fitzmaurice is, in the passage cited, clearly dealing with the presumption that may be raised by proximity where a state is exercising or displaying sovereignty over a parcel of territory and there is some question whether this is presumed to extend also to outlying territory over which there is little or no factual impact of its authority. The Tribunal has no difficulty in accepting these statements of high authority; but what they are saying is in fact rather more than a simple principle of unity. It will be useful to cite Fitzmaurice again:

The question of 'entity' or 'natural unity'

This question can have far-reaching consequences. Not only may it powerfully affect the play of probabilities and presumptions, but also, if it can be shown that the disputed areas (whether by reason of actual contiguity or of proximity) are part of an entity or unity over which as a whole the claimant State has sovereignty, this may (under certain conditions and within certain limits) render it unnecessary - or modify the extent to which it will be necessary - to adduce specific evidence of State activity in relation to the disputed areas as such - provided that such activity, amounting to effective occupation and possession, can be shown in the principle established by the Island of Palmas case that 'sovereignty cannot be exercised in fact at every moment on every point of a territory'.(29)

462. Thus, the authorities speak of "entity" or "natural unity" in terms of a presumption or of probability and moreover couple it with proximity, contiguity, continuity, and such notions, well known in international law as not in themselves creative of title, but rather of a possibility or presumption for extending to the area in question an existing title already established in another, but proximate or contiguous, part of the same "unity".

463. These ideas, however, have a twofold possible application in the present case. They may indeed, as Yemen would have it, be applied to cause governmental display on one island of a group to extend in its juridical effect to another island or islands in the same group. But by the same rationale a complementary question also arises of how far the sway established on one of the mainland coasts should be considered to continue to some islands or islets off that coast which are naturally "proximate" to the coast or "appurtenant" to it. This idea was so well established during the last century that it was given the name of the "portico doctrine" and recognised "as a means of attributing sovereignty over off-shore features which fell within the attraction of the mainland".(30) The relevance of these notions of international law to the legal history of the present case is not far to seek.

464. Thus the principle of natural and physical unity is a two-edged sword, for if it is indeed to be applied then the question arises whether the unity is to be seen as originating from the one coast or the other. Moreover, as the cases and authorities cited by Yemen clearly show, these notions of unity and the like are never in themselves roots of title, but rather may in certain circumstances raise a presumption about the extent and scope of a title otherwise established.

465. In spite of unity theories, the fact is that both Parties have tacitly conceded that, for the purposes at any rate of the exposition of their pleadings, it may be accepted that there can be sub-groups within the main group. The nomenclature within common use indicates at least three of the sub-groups: the Mohabbakahs; the Haycocks; and what it will be convenient at least for the moment to call the Zuqar-Hanish group and its many satellite islands, islets, and rocks. These names will all be found in the British Pilot and Sailing Directions for the Southern Red Sea (Yemen has cited this publication as authority for regarding all these islands as one group, but of course if one is concerned with them as sailing hazards or landmarks when traversing the Red Sea there is really no other way to do it). There are also the two northern islands: Jabal al-Tayr, and the group of which the biggest island is Jabal Zubayr. The Tribunal will now consider its conclusions in respect of each of the three subgroups and then, finally, the northern islands.

466. Thus, in order to make decisions on territorial sovereignty, the Tribunal has hardly surprisingly found no alternative but to depart from the terms in which both Parties have pleaded their cases, namely by each of them presenting a claim to every one of the islands involved in the case. The legal history simply does not support either such claim.(31) For, as has been explained above, much of the material is found on examination to apply either to a particular island or to a sub-group of islands. The Tribunal has accordingly had to reach a conclusion which neither Party was willing to contemplate, namely that the islands might have to be divided; not indeed by the Tribunal but by the weight of the evidence and argument presented by the Parties, which does not fall evenly over the whole of the islands but leads to different results for certain sub-groups, and for certain islands.

The Mohabbakahs

467. The Mohabbakah Islands are four rocky islets which amount to little more than navigational hazards. They are Sayal Islet, which is no more than 6 nautical miles from the nearest point on the Eritrean mainland coast, Harbi Islet and Flat Islet; all three of these are within twelve nautical miles of the mainland coast. Finally, there is High Islet, which is less than one nautical mile outside the twelve-mile limit from the mainland coast, and about five nautical miles from the nearest Haycock island, namely South West Haycock.

468. Eritrea has sought to show that Italy obtained title to the Mohabbakahs along with the various local agreements Italy made with local rulers (see para. 159, above), which led to its securing title over the Danakil coast; this was not protested by Turkey and came to be recognized by Great Britain. The diplomatic history has some interest for this case, especially in highlighting the question of whether South West Haycock is a Mohabbakah island, or part of a separate group of Haycocks, or part of a larger "Zuqar-Hanish group" (see para. 215, above, for the 1930 Italian claim to sovereignty over South West Haycock).

469. Eritrea thus contends that the Mohabbakahs were comprised within what was passed to Ethiopia and so to Eritrea after the Second World War and that this is affirmed by the reference in Article 2 of the 1947 Peace Treaty to the islands "off the coast" and by the constitutional arrangements.

470. Yemen claims that the only islands Ethiopia secured jurisdiction over through local rulers were the islands in Assab Bay; and that, because formerly both coasts of the Red Sea fell under Ottoman rule; and because after the end of the First World War Yemen reverted to its "historic title"; and also because the Mohabbakahs are properly to be perceived as a unity with the Haycocks and the Zuqar-Hanish group, title to all these islands lies with Yemen. The Tribunal rejects this argument.

471. The Tribunal has already noted that there is no evidence that the Mohabbakah islands were part of an original historic title held by Yemen, even were such a title to have existed and to have reverted to Yemen after the First World War. And, even if it were the case that only the Assab Bay islands passed to Eritrea by Italy in 1947, no serious claims to the Mohabbakahs have been advanced by Yemen since that time, until the events leading up to the present arbitration.

472. The Tribunal need not, however, decide whether Italian title to the Mohabbakahs survived the Treaty of Lausanne, and passed thereafter to Ethiopia and then to Eritrea. It is sufficient for the Tribunal to note that all the Mohabbakahs, other than High Islet, lie within twelve miles of the Eritrean coast. Whatever the history, in the absence of any clear title to them being shown by Yemen, the Mohabbakahs must for that reason today be regarded as Eritrean.(32) No such convincing alternative title has been shown by Yemen. It will be remembered indeed that Article 6 of the 1923 Treaty of Lausanne already enshrined this principle of the territorial sea by providing expressly that islands within the territorial sea of a state were to belong to that state. In those days the territorial sea was generally limited by international law and custom to three nautical miles, but it has now long been twelve, and the Ethiopian territorial sea was extended to twelve miles in a 1953 decree.

473. At this point it will be convenient to look at the ingenious theory enunciated by Eritrea, based on the undoubted rule that the territorial sea extends to twelve miles not just from the coast but may also extend from a baseline drawn to include any territorial islands within a twelve-mile belt of territorial sea. Thus the baseline can lawfully be extended to include an entire chain, or group of islands, where there is no gap between the islands of more than twelve miles; the so-called leapfrogging method of determining the baseline of the territorial sea. As already mentioned, the entire chain or group of these islands consists of islands, islets, or rocks proud of the sea and therefore technically islands, with no gap between them of more than twelve miles. The only such gap is the one between the easternmost island (the Abu Ali islands) and the Yemen mainland coast.

474. The difficulty with leapfrogging in the instant case is that it begs the very question at issue before this Tribunal: to which coastal state do these islands belong? There is a strong presumption that islands within the twelve-mile coastal belt will belong to the coastal state, unless there is a fully-established case to the contrary (as, for example, in the case of the Channel Islands). But there is no like presumption outside the coastal belt, where the ownership of the islands is plainly at issue. The ownership over adjacent islands undoubtedly generates a right to a corresponding territorial sea, but merely extending the territorial sea beyond the permitted coastal belt, cannot of itself generate sovereignty over islands so encompassed. And even if there were a presumption of coastal-state sovereignty over islands falling within the twelve-mile territorial sea of a coastal-belt island, it would be no more than a presumption, capable of being rebutted by evidence of a superior title.

* - * - * - *

475. Therefore, after examination of all relevant historical, factual and legal considerations, the Tribunal unanimously finds in the present case that the islands, islet, rocks, and low-tide elevations forming the Mohabbakah islands, including but not limited to Sayal Islet, Harbi Islet, Flat Islet and High Islet are subject to the territorial sovereignty of Eritrea. It is true that High Islet is a small but prominent rocky islet barely more than twelve miles (12.72 n.m.) from the territorial sea baseline. But here the unity theory might find a modest and suitable place, for the Mohabbakahs have always been considered as one group, sharing the same legal destiny. High Islet is certainly also appurtenant to the African coast.

The Haycocks

476. The Haycocks are three small islands situated along a roughly southwest-to-northeast line. They are, from south to north, South West Haycock, Middle Haycock and Northeast Haycock. South West Haycock is some 6 nautical miles from the nearest point of Suyul Hanish, though there is the very small Three Foot Rock about midway between them.

477. As already mentioned above, the Haycocks do have a peculiar legal history and it is for this reason mainly that they need to be discussed separately here. That legal history is very much bound up with the story of the Red Sea lighthouses. But one might begin the salient points of this legal history by recalling the 1841, 1866 and 1873 firmans of the Ottoman Sultan (see para. 97, above), by which the African coast of the Red Sea and the islands off it were placed under the jurisdiction and administration of Egypt, though of course the whole of this part of the world was then under the sovereignty of the Ottoman Empire. There seems little doubt that this African-coast administration would have extended to the Mohabbakahs and the Haycocks. At this time the territorial sea was limited to three miles, and there were still grave doubts about the nature and extent of the territorial waters regime. Nevertheless there was a feeling, based upon considerations of security as well as of convenience, that islands off a particular coast would, failing a clearly established title to contrary, be under the jurisdiction of the nearest coastal authority. As mentioned above, this was sometimes called the "portico doctrine".

478. Another stage in this legal history is at the end of the nineteenth century, when the British Government was interested in the possibility of establishing an alternative western shipping channel through the Red Sea, which needed lighting if it was to be used at night. Various islands were considered as sites for a light (see paras. 203, 204, above), including South West Haycock, which is in the end proved to be the successful candidate. This involved inquiries about the "jurisdiction" under which the island would come, and the British Board of Trade satisfied itself that South West Haycock was subject to Italian jurisdiction and at any rate probably not Ottoman.

479. In 1930, when the Italians were constructing a lighthouse on South West Haycock, there was an instructive correspondence between the Italian and British Governments. An internal Foreign Office memorandum reveals the opinion that "the establishment of the Italian colony of Eritrea makes it difficult, therefore, to resist the claim that the islands off the coast of Eritrea are to be considered as an appendage of that colony"(33). This was the official reaction to a letter from the Royal Italian Government of 11 April, claiming South West Haycock, inter alia for reasons of its "immediate vicinity" to the Eritrean Red Sea coast.

480. Eritrea employs these arguments to support its claim to the Haycocks, but puts it in the form of a succession derived from the Italian colony of Eritrea, and by way of the subsequent federation of Ethiopia and Eritrea, through to Eritrean independence in 1993. There are difficult juridical problems with this theory of succession, not least the terms of the Italian armistice of 1943 and the peace treaty of 1947, whereby Italy surrendered her colonial territories for disposition by the Allies and in default of agreement amongst them, to disposition by the United Nations, which of course is what actually happened to Eritrea. However this may be, the geographical arguments of proximity to the Eritrean coast remain persuasive and accord with the general opinion that islands off a coast will belong to the coastal state, unless another, superior title can be established. Yemen has failed, in this case, to establish any such superior claim.

481. The Eritrean claim to the Haycocks also finds some support in the material provided by both Parties for the supplementary hearing on the implications of petroleum agreements. None of the Yemen agreements extends as far to the southwest as the Haycocks; the 1974 Tomen-Santa Fe agreement appears to encompass the Hanish group, but stops short of the Haycocks. On the other hand, the fully documented agreements of the Eritrean Government and Shell, Amoco and BP do cover the areas of the Haycocks, and of course the Mohabbakahs. There was no protest from Yemen, though Yemen did protest when an agreement with Shell appeared to it to trespass upon its claim to the northern islands.

482. Therefore, after examination of all relevant historical, factual and legal considerations, the Tribunal unanimously finds in the present case that the islands, islet, rocks, and low-tide elevations forming the Haycock Islands, including, but not limited to, North East Haycock, Middle Haycock, and South West Haycock, are subject to the territorial sovereignty of Eritrea. It follows that the like decision will, apart from other good reasons noted above, apply to High Islet, the one island of the Mohabbakah sub-group that is outside the Eritrean territorial sea.

483. There remains a question whether the South West Rocks should for these purposes be regarded as going along with the Haycocks. No doubt South West Rocks are so called because they lie southwest of Greater Hanish and there is no other feature between them and that island. There is some evidence that South West Rocks were, at various times, considered to form the easternmost limit of African-coast jurisdiction. While the British Foreign Office documentation relied on by both Parties reflects divergent views (referring in at least one case to Italian jurisdiction over South West Rocks as "doubtful"), the Parties agree that in the early 1890s, Italy responded to direct British inquiries concerning potential lighthouse sites with assertions of jurisdiction over all of the proposed sites, including South West Rocks. Furthermore, Italy did not object to the subsequent British suggestion that the Sublime Porte be informed of the Italian position. This thinking surfaced again in 1914, in Great Britain's initial proposal for a post-war distribution of relinquished Ottoman territory, which would have placed everything east of South West Rocks under the sovereignty of "the independent chiefs of the Arabian mainland."

484. In light of this, it seems reasonable that South West Rocks should be treated in the same manner as the other islands administered from the African coast: the Mohabbakahs and the Haycocks. South West Rocks are therefore unanimously determined by the Tribunal to be subject to the territorial sovereignty of Eritrea.

The Zuqar - Hanish Group

485. There remains to be determined the sovereignty over Zuqar and over the Hanish islands, and their respective satellite islets and rocks, including the island of Abu Ali, to the east of the northern end of Zuqar, which was for long a principal site for a lighthouse.

486. This has not been an easy group of islands to decide on, one reason for this being that, positioned as they are in the central part of the Red Sea, the appurtenance factor is bound to be relatively less helpful. A coastal median line would in fact divide the island of Greater Hanish, the slightly greater part of the island being on the Eritrean side of the line. Zuqar would be well on the Yemen side of a coastal median line.

487. The Parties have put before the Tribunal many aspects of the local legal history which are said to point the decision one way or the other. These have all been examined in detail in the chapters above. It is however already apparent from that examination that any expectation of a clear and definite answer from that earlier legal history is bound to be disappointed. The Yemeni idea of a reversionary ancient title has been discussed earlier in this chapter and found unhelpful in regard to these islands. More helpful perhaps is the material which suggests that, when the Ottomans decided in the later nineteenth century to grant to Egypt the jurisdiction over the African coast, this possibly included islands appurtenant to that coast, and according to some respectable authorities this did not include this central group of islands, both Zuqar and Hanish being regarded as still within the jurisdiction of the vilayet of Yemen. If this was so, though that position can hardly have been carried over to the present time in spite of Article 16 of the Treaty of Lausanne, it would constitute an impressive historical precedent. Hertslet's opinion about the proper distribution of jurisdiction over the islands of the Red Sea clearly impressed the British Foreign Office, but it seems to be Hertslet's view of what should be done about all the islands in the Red Sea rather than evidence of existing titles.

488. There are some echoes of the idea of Yemeni title to be found in the earlier part of the present century in for example the record of the negotiations between the Imam and a British envoy, Colonel Reilly, in which talk the Imam is said to have referred to the need to return to him certain Yemeni islands. But there is no doubt that the main grievance the Imam had in mind was the island of Kamaran and its surrounding islets, which was then occupied by the British. There was also a claim which an internal Foreign Office memorandum referred to as the Imam's claim to "unspecified islands". The British civil servants were quite prepared themselves to speculate that these islands might have included Zuqar and Hanish, which had been temporarily occupied by the British in 1915. But it is in the end difficult to attach decisive importance to a claim which could not be specified with any certainty.

489. Eritrea seeks to derive an historical title by succession, through Ethiopia, from Italy. There is no doubt that Italy had serious ambitions in respect of these central islands in the nineteen thirties and did establish a presence there. But as has been seen above that position was constantly neutralised by assurances to the British Government that Italy fully accepted that the legal status of the islands was still governed by Article 16 of the Treaty of Lausanne. And then there is also the difficulty of deriving a title from Italy in view of the provisions of the Italian Peace Treaty of 1947.

490. Then of course there are the maps. These islands are large enough to find a place quite often - though by no means always - on even relatively small-scale maps of the region. It is fair to assert that, thanks to the efforts of counsel and especially those of Yemen, the Tribunal will have seen more maps of every conceivable period and provenance than probably have ever been seen before, and certainly a very much larger collection than will have been seen at any time by any of the principal actors in the Red Sea scene. In fact, the difficulty is not so much the interpretation of a plethora of maps of every kind and provenance, as it is the absence of any kind of evidence that these actors took very much notice of, or attached very much importance to, any of them. The Tribunal is of the opinion that in quite general terms Yemen has a marginally better case in terms of favourable maps discovered, and looked at in their totality the maps do suggest a certain widespread repute that these islands appertain to Yemen.

491. As to the other aspects of the legal history of this central group, it does inevitably reflect the ebb and flow of the interest, or the neglect, as the case may be, of both sides, varying from time to time, and qualified always by the unattractive nature of these islands, relieved from time to time by occasional usefulness, as for siting navigational lights, or by their sometimes perceived or imagined strategic importance; for they have never been considered "remote" in the sense of Greenland or the Island of Palmas. Accordingly, in the Tribunal's opinion, although some of this older historical material is important and generally helpful and indeed essential to an understanding of the claims of both Parties, neither of them has been able on the basis of the historical materials alone to make out a case that actually compels a decision one way or the other. Accordingly the Tribunal has looked at events in the last decade or so before the Agreement of Arbitration for additional materials and factors which might complete the picture of both Parties' cases and enable the Tribunal to make a firm decision about these two islands and their satellite rocks and islets. The Tribunal is confirmed in this approach by the fact that both Parties have anticipated the need for such material by providing supplementary data in connection with the hearings held in July 1998. It should be added, however, that the more recent legal history of these islands shows in some respects differences between Zuqar and Hanish. Because this is so, the islands should be, and will be, considered separately. It would be wrong to assume that they must together go to one Party or the other. In this extent the Tribunal rejects the Yemen theory that all the islands in the group must in principle share a common destiny of sovereignty.

492. Of the recent events perhaps the first heading to look at is that of the Red Sea lighthouses which have featured in the arguments of both Parties. It is evident from the lighthouse history, again dealt with in detail in chapter VI above, that the undertaking by a government of the maintenance of one of these lights has generally been regarded as neutral for the purpose of the acquisition of territorial sovereignty, although it should also be remembered that, when Great Britain wished in 1892 to secure the building of a light for the proposed western shipping channel, the British Government was anxious to know which government had "jurisdiction" over the chosen site on South West Haycock, and Italy not only made a claim but had its claim to jurisdiction recognised by the British Government. Four lights have been constructed by and appear to be maintained by Yemen in the area now being dealt with (though it should be added that such lights are of course no longer manned). These are sited as follows: on the island of Abu Ali, which is some 3 nautical miles west of the northern tip of Zuqar, on the south-eastern tip of Zuqar itself; on Low Island which is off the north-eastern tip of Lesser Hanish; and on the north-eastern tip of Greater Hanish. The latter was constructed in July 1991 by Yemen and there is in evidence a picture of it with an inscription giving the name of the Republic of Yemen. It can hardly be denied that these lights, clearly intended to be permanent installations, are cogent evidence of some form of Yemen presence in all these islands.

493. Of relatively recent events, Eritrea attaches much importance to the history of Ethiopian naval patrols and the log books which evidence their occurrence, and which involved in particular the islands of Zuqar and Hanish; and this is indeed a possible factor where the islands must be taken as a group; for these were patrols in these waters generally rather than voyages to particular islands. There is no doubt that these patrols occurred on a large scale, and they are fully examined in Chapter VII and it is well known that these islands were used by the rebels, probably mainly as staging posts and relatively safe anchorages for vessels attempting to convey supplies to the rebel armies fighting on the mainland of Ethiopia, some of them possibly from Yemen, which is known to have sympathised with the rebel cause.

494. A strange aspect of these naval patrols possibly over a matter of several years - though the actual evidence Eritrea has been able to provide leaves a number of blank periods - is the lack of protest from Yemen. If Ethiopia had been patrolling the islands on the assumption that it was merely patrolling its own territory, then the lack of Yemen protest is all the more remarkable and calls for some explanation which Yemen has not altogether provided. Yemen was of course preoccupied with its own civil war between 1962 and 1970; and a good deal of this naval patrolling must have been on the high seas rather than in the territorial seas of the islands. Eritrea claims that the Ethiopian naval patrols were also enforcing fishing regulations. This seems credible for it would have provided cover for inspecting the papers of vessels even on the high seas and the rebels would hardly have confined their supply operations to ships flying the Ethiopian flag.

495. And yet these logbooks of naval patrols give relatively little evidence of activity on or even near to the islands. It is interesting to consider in this context the press statement issued by the Yemen Embassy in Mogadishu on 3 July 1973 stating "the Y.A.R. always maintains its sovereignty over its islands in the Red Sea, with the exception of the islands of Gabal Abu Ali and Gabal Attair which were given to Ethiopia by Britain when the latter left Aden and surrendered power in our Southern Yemen". This surmise was of course mistaken. But it does amount to a statement that Yemen at this time had no presence in either of these two mentioned islands and had little idea what was happening there. This, however, was the time of the Arab press rumours of Ethiopia having allowed Israel the use of certain Red Sea islands. This same press release stated that Yemen had, accompanied by journalists and press correspondents, investigated the position on "Lesser Hanish, Greater Hanish, Zuqar, Alzubair' Alswabe', and several other islands at the Yemeni coast". These were found to be "free from any foreign infiltration whatsoever". Presumably this was also the inspection by the military committee of the Arab League (see para. 321, above). This statement has the ring of truth. It most probably was the position that these islands, including Zuqar and both Hanish islands, were then normally empty of people or activity other than that of small coastal fishermen plying their traditional way of life and calling at the islands when their work took them there. But it is significant that Yemen could apparently take the above inspection party without any repercussions from Ethiopia.

496. There is much that is ambiguous and unexplained on both sides in this evidence of naval patrols. On balance the episode appears to the Tribunal to lend some weight to the Eritrean case. But again it is a matter of relative weight. There is no compelling case here for either Party. And again it is very difficult on the basis of this material to give it great weight in claims to land territory.

497. The petroleum agreements made by Yemen and by Ethiopia (and then by Eritrea) from 1972 onwards do surprisingly little to resolve the problem, for these agreements, in so far as they extended to offshore areas, were not really concerned with the islands at all, but with either the outer boundary formed by the extent of the then exploitable depths of seabed, or by the coastal median line, which was the temporary boundary actually contemplated for such agreements by the 1977 Yemeni continental shelf legislation. As was reflected by the questions put to the Parties in the closing moments of the July 1998 hearings, the agreements seemed almost to ignore the islands; not surprisingly, considering that the volcanic geological nature of the islands meant that they were totally uninteresting to the oil companies.

498. As already stated above, the Tribunal attaches little importance to the agreements by both Parties with Shell for geological investigations. The area covered by the contract activities likely traversed these islands. But the Tribunal has little doubt that Shell was operating with the permission of both Parties, and was getting information primarily for its own use, in order to decide about which areas of the continental shelf it might be worth making production agreements.

499. When it comes to actual agreements for exploitation, whether in the form of full petroleum production-sharing agreements or less than that, two of the agreements made by Yemen encompassed the Zuqar-Hanish Islands totally (one with Adair, which was very short-lived and never went into effect, and one with Tomen-Santa Fe), while the agreements made by Ethiopia (Ethiopia/Shell) avoided extending to these islands or, in the instance of the Ethiopia-IPC/Amoco agreement of 1989, cuts across Greater Hanish, the division apparently depending on precisely how one plots the coastal median line.

500. After the careful examination of the contract areas of the oil agreements of both Parties, the conclusions to be drawn from this material seem to be reasonably clear. Eritrea can and does point to the IPC/Amoco agreement with Ethiopia which cuts the Island of Hanish. There are various versions. In some versions of the attempts to draw the contract area on a map, only the tip of Hanish is within the Eritrean side of the line; in others the line appears to portray most of the island as Eritrean, leaving only a relatively small portion of it to Yemen. It is surely apparent that the contract area was defined simply in terms appropriate for the essentially maritime interests of the contracting party, and that this, in conformity with normal practice where there is no agreed and settled maritime boundary, was made the coastal median line, ignoring the possible effect of islands. It seems in effect to have been agreed and drawn on the illustrative map of the contract simply ignoring the islands. If Ethiopia had had it in mind to use the agreement for the purpose of illustrating a claim to the island of Hanish, Ethiopia would surely not have given itself only two-thirds of the island; it would have had the line make an excursion round and embrace the whole island. As it is, it seems to the Tribunal that the Ethiopian and Eritrean agreements are in effect neutral as far as the present task of the Tribunal is concerned; as indeed Eritrea argued. This does not mean that the Eritrean claim to these islands is unfounded; but it does mean that the oil agreements do little to assist that claim, except in so far as the IPC/Amoco Agreement tends to neutralize the Yemeni argument that petroleum agreements as such provide confirmation of sovereignty.

501. Yemen, besides the unconvincing suggestion that the Shell Company's seismic investigation of a large area right across the southern Red Sea somehow confirms the Yemeni claims to the Zuqar and Hanish islands, has in the Tomen-Santa Fe seismic agreement of 1974-75 referred to an agreement in which the contract does apparently embrace both Zuqar and Hanish, or most of Greater Hanish Island. This also resulted in certain activities by the company, including a collection of samples from Zuqar (see para. 409, above). This again does not establish that Yemen has validated its claim to both these islands. But as concluded above, the agreements produced by both parties fail to establish evidence of sovereignty. Perhaps it helps to see these petroleum agreements of the seventies in perspective to remember that in 1973 there was a Yemeni inspection of the islands, with journalists and representatives of the Arab League military committee, that found all these islands empty.

502. It was later that there was more activity; notably the construction in 1993 by the Total Oil Company of an air landing strip on Hanish, for the recreational visits of their employees, and as a by-product of their concession agreement with Yemen. That agreement did not encompass either Zuqar or Hanish. Nevertheless, the fact that there were regular excursion flights constitutes evidence of governmental authority and the exercise of it. Nor did it apparently attract any kind of protest from Eritrea; though of course by this time the civil war was over and Eritrea was established as an independent state.

503. As neither Party has in the opinion of the Tribunal made a convincing case to these islands on the basis of an ancient title in the case of Yemen, or, of a succession title in the case of Eritrea, the Tribunal's decision on sovereignty must be based to an important extent upon what seems to have been the position in Zuqar and Hanish and their adjoining islets and rocks in the last decade or so leading up to the present arbitration. Anything approaching what might be called a settlement, or the continuous display of governmental authority and presence, of the kind found in some of the classical cases even for inhospitable territory, is hardly to be expected. For very few people would wish to visit these waterless, volcanic islands except for a special reason and probably a temporary one. Nevertheless, it is clear from the documents mentioned earlier in this Award that both Yemen and Ethiopia had formulated claims to both islands at least by the late eighties and had indeed it would seem held secret negotiations on the claims; which negotiations, at least according to the Eritrean "Top Secret" internal report, had at first promised a compromise solution on the basis of a median line which would presumably have given Zuqar and Little Hanish to Yemen and Greater Hanish to Ethiopia. But this came to nothing. So now one must look at the effectivités for the solution.

504. Yemen has been able to present the Tribunal with a list of some forty-eight alleged Yemeni happenings or incidents in respect of "the islands", which occurred in the period between early 1989 and mid-1991. This list is not confined to the central group, for there is included for example the decisions of the 1989 London Conference on the lighthouses, and the building of a lighthouse on al Tayr in July 1989. It is evident though that Zuqar features very prominently in the list. It is also evident that Eritrea has relatively very little to show in respect of Zuqar. The Tribunal has no doubt that the island of Zuqar is under the sovereignty of Yemen.

505. In respect of Hanish the matter is not so clear cut. The Eritrean claim is well established as a claim and is clearly of great importance to that very newly- independent country. The refusal to agree to a Yemeni aerial survey of the Islands and Ethiopia's responsive claim of title to some of them is significant. So also is its arrest of Yemeni fishermen on Greater Hanish and its assertion, in response to Yemen's protest to the Security Council, that the area was within Ethiopian jurisdiction.

506. There was some emphasis by Eritrea on a scheme to put beacons on Hanish to assist Amoco's seismic testing; there is no clear evidence that they were actually installed. Any such installation of beacons covered several locations, of which Greater Hanish Island was only one, and would have been short-lived: the evidence provided by Eritrea mentions two weeks, and provides for removal of the beacons on completion of the seismic work. Moreover, the beacons were placed by the oil company, Amoco, with only a limited role for the Ethiopian government in protecting the oil company personnel and the temporary beacons from the attentions of "random individuals". Finally, there is evidence of the issuance, in 1980, of an Ethiopian radio transmitting licence to Delft Geophysical Company, which provided for a station to be located at "Greater Hanish Island, Port of Assab vicinity".

507. Yemen has more to show by way of presence and display of authority. Putting aside the lighthouse in the north of the island, there was the Ardoukoba expedition and campsite which was made under the aegis of the Yemeni Government. There is the air landing site, as well as the production of what appears to be evidence of frequent scheduled flights, no doubt mainly for the off-days of Total employees; and there is the May 1995 license to a Yemeni company (seemingly with certain German nationals associated in a joint venture scheme) to develop a tourist project (recreational diving is apparently the possible attraction to tourism) on Greater Hanish.

* - * - * - *

508. Therefore, after examination of all relevant historical, factual and legal considerations, the Tribunal finds in the present case that, on balance, and with the greatest respect for the sincerity and foundations of the claims of both Parties, the weight of the evidence supports Yemen's assertions of the exercise of the functions of state authority with respect to the Zuqar-Hanish group. The Tribunal is further fortified in finding in favour of Yemen by the evidence that these islands fell under the jurisdiction of the Arabian coast during the Ottoman Empire; and that there was later a persistent expectation reflected in the British Foreign Office papers submitted in evidence by the Parties that these islands would ultimately return to Arab rule. The Tribunal therefore unanimously finds that the islands, islet, rocks, and low-tide elevations of the Zuqar-Hanish group, including, but not limited to, Three Foot Rock, Parkin Rock, Rocky Islets, Pin Rock, Suyul Hanish, Mid Islet, Double Peak Island, Round Island, North Round Island, Quoin Island (13°43'N, 42°48'E), Chor Rock, Greater Hanish, Peaky Islet, Mushajirah, Addar Ail Islets, Haycock Island (13°47'N, 42°47'E; not to be confused with the Haycock Islands to the southwest of Greater Hanish), Low Island (13°52'N, 42°49'E) including the unnamed islets and rocks close north, east and south, Lesser Hanish including the unnamed islets and rocks close north east, Tongue Island and the unnamed islet close south, Near Island and the unnamed islet close south east, Shark Island, Jabal Zuquar Island, High Island, and the Abu Ali Islands (including Quoin Island (14°05'N, 42°49'E) and Pile Island) are subject to the territorial sovereignty of Yemen.

Jabal al-Tayr and the Zubayr Group of Islands

509. Both the lone island of Jabal al-Tayr, and the Zubayr group of islands and islets, call for separate treatment, as they are a considerable distance from the other islands as well as from each other. They are not only relatively isolated, but also are both well out to sea, and so not proximate to either coast, though they are slightly nearer to the Yemeni coastal islands than they are to the coast and coastal islands of Eritrea. Both are well eastward of a coastal median line. Here again, the Tribunal has had to weigh the relative merits of the Parties' evidence, which has been sparse on both sides, of the exercise of functions of state and governmental authority.

510. The traditional importance of both groups has been that they have been lighthouse islands (the Zubayr light was on Centre Peak, the southernmost islet of the group). It will be clear from the history of the Red Sea lighthouses (see Chapter VI, above) that, although, or perhaps even because, lighthouses were so important for nineteenth and early twentieth century navigation, a government could be asked to take responsibility or even volunteer to be responsible for them, without necessarily either seeming to claim sovereignty over the site or acquiring it. The practical question was not one of ownership, but rather of which government was willing, or might be persuaded, to take on the responsibility, and sometimes the cost, if not permanently then at least for a season.

511. It will be recollected that Centre Peak in the Zubayr group was an island in which Italy, in its 1930s period of colonial expansion, had taken a great interest; the Centre Peak light was abandoned by the British in 1932, but reactivated by Italy the following year. The British sought and obtained the usual assurances about the Treaty of Lausanne status of the island (see paras. 216-218 above). So for a time at least this group fell under the jurisdiction of the authority on the African coast.

512. Yet during the Second World War and the subsequent British occupation of Eritrea, it was decided that Great Britain was under no obligation to maintain the Centre Point light or indeed the Haycock light.

513. An important turning point in the history of the northern islands of Jabal al-Tayr and the Zubayr group was the 1989 London conference about lighthouses. This was rather different from previous conferences. This conference was to be the last of its kind, because its main purpose was to liquidate the former international arrangements for administration of the lights and the sharing of costs. The final arrangements made for the lights (which were then still of the greatest importance for navigation) were therefore intended to be permanent. No further conference was envisaged.

514. It will be remembered that Yemen was invited to the conference as an observer on the plea to the British Government that the two lighthouse islands of Abu Ali and Jabal al-Tayr, "lie within the exclusive economic zone of the Yemen Arab Republic," and that because of this Yemen was willing to take on the responsibility of managing and operating the lights. It was also the fact that Yemen had already installed new lights on both of these sites. The offer from Yemen was gratefully accepted by the conference. There had been hopes that Egypt might take on the work but Egypt was not willing to do so.

515. The matter of sovereignty was not on the agenda of the conference, nor was it discussed. Yemen's own request to be invited to the conference had wisely avoided raising the matter. Moreover, there were at the conference the usual references to the Treaty of Lausanne formula concerning indeterminate sovereignty.

516. Nevertheless, the decision of the conference to accept the Yemeni offer over the lights does reflect a confidence and expectation of the member governments of the conference of a continued Yemeni presence on these lighthouse islands for, at any rate, the foreseeable future. Repute is also an important ingredient for the consolidation of title.

517. There is also another matter where Yemen is able to show what amounts to important support for its case over these northern islands, and that is the substantially new information on petroleum agreements that was made available to the Tribunal at the supplementary hearings held for this purpose in July 1998. There are two such agreements which appear to be relevant for the islands presently under discussion.

518. First, there is the agreement made by the Yemeni Government with the Shell company on 20 November 1973. The western boundary of the contract area in this agreement is drawn so as to include within it the Zubayr group. It does not include Jabal al-Tayr, but passes at a distance which might encompass the territorial sea of that island, depending on the breadth of the territorial sea allowed to it for the purposes of a maritime delimitation.

519. The second is the Hunt Oil production sharing agreement ratified on 10 March 1985. The western contract area boundary of this agreement again includes the Zubayr group, but also appears from the illustrative map to brush the island of Jabal al-Tayr, and of course plainly includes a part of its territorial sea.

520. These agreements were not protested by Ethiopia (though it should be remembered that the Hunt agreement was made at a time when the Ethiopian civil war was still raging).

521. Neither Ethiopia nor Eritrea has made any petroleum agreements encompassing these islands. Eritrea did, however, make agreements in 1995 and 1997 with the Anadarko Oil Company, which extended in the direction of these islands and towards what appears to be an approximate median line between coasts. Yemen protested this line on 4 January 1997 as a "blatant" violation of the territorial waters of both groups and of her economic rights "in the region". This was, of course, some time after the signature of the Agreement on Principles and indeed the Arbitration Agreement initiating these proceedings.

522. The legal history of these northern and isolated islands has been mixed and varied. It has been seen that even as late as 1989 it was assumed that their sovereign status was still indeterminate in accordance with the status impressed upon them, until it should be changed in a lawful way, by the Treaty of Lausanne. Nevertheless, by 1995 it was doubtful whether any dispute over Yemen's claim to them would be agreed to be submitted to this Tribunal. Even Eritrea at one point made a proposal for an agreement in which these islands were not mentioned.

523. The Tribunal has not found this particular question an easy one. There is little evidence on either side of actual or persistent activities on and around these islands. But in view of their isolated location and inhospitable character, probably little evidence will suffice.

524. Therefore, after examination of all relevant historical, factual and legal considerations, the Tribunal unanimously finds in the present case that, on the basis of the foregoing, the weight of the evidence supports the conclusion that the island of Jabal al-Tayr, and the islands, islets, rocks and low-tide elevations forming the Zubayr group, including, but not limited to, Quoin Island (15°12'N, 42°03'E), Haycock Island (15°10'N, 42°07'E; not to be confused with the Haycock Islands to the southwest of Greater Hanish), Rugged Island, Table Peak Island, Saddle Island and the unnamed islet close north west, Low Island (15°06'N, 42°06'E) and the unnamed rock close east, Middle Reef, Saba Island, Connected Island, East Rocks, Shoe Rock, Jabal Zubayr Island, and Centre Peak Island are subject to the territorial sovereignty of Yemen.

The Traditional Fishing Regime

525. In making this award on sovereignty, the Tribunal has been aware that Western ideas of territorial sovereignty are strange to peoples brought up in the Islamic tradition and familiar with notions of territory very different from those recognized in contemporary international law. Moreover, appreciation of regional legal traditions is necessary to render an Award which, in the words of the Joint Statement signed by the Parties on 21 May 1996, will "allow the re-establishment and the development of a trustful and lasting cooperation between the two countries."

526. In finding that the Parties each have sovereignty over various of the Islands the Tribunal stresses to them that such sovereignty is not inimical to, but rather entails, the perpetuation of the traditional fishing regime in the region. This existing regime has operated, as the evidence presented to the Tribunal amply testifies, around the Hanish and Zuqar islands and the islands of Jebel al-Tayr and the Zubayr group. In the exercise of its sovereignty over these islands, Yemen shall ensure that the traditional fishing regime of free access and enjoyment for the fishermen of both Eritrea and Yemen shall be preserved for the benefit of the lives and livelihoods of this poor and industrious order of men.

ــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــــ

Go to Chapter XI

Notes - Chapter X

25.  The Tribunal wishes to note the sheer volume of written pleadings and evidence received from the Parties in this first phase of the arbitral proceedings. Each Party submitted over twenty volumes of documentary annexes, as well as extensive map atlases. In addition, the Tribunal has carefully reviewed the verbatim transcripts of the oral hearings, which together far exceed 1,000 pages. The Tribunal further notes that the majority of documents were submitted in their original language, and the Tribunal has relied on translations provided by the Parties.

26. Minquiers and Ecrehos (U.K. v. Fr.), 1953 I.C.J. 47.

27. Island of Palmas (U.S. v. Neth.), 2 R.I.A.A. 829 (1929).

28. Legal Status of Eastern Greenland (Den. v. Nor.), 1933 P.C.I.J. (Ser. A/B) No. 53.

29. 32 B.Y.I.L. (1955-56) 73-74.

30. D. O'Connell, The International Law of the Sea 185 (1982).

31. In this connection it is interesting to see the statements made in the 1977 "Top Secret" memorandum of the Ministry of Foreign Affairs of The Provisional Military Government of Socialist Ethiopia, discussed above in para. . This memorandum refers to islands in the southern part of the Red Sea that "have had no recognized owner", with respect to which Ethiopia "claims jurisdiction" and "both North and South Yemen have started to make claims."South Yemen's position is that the islands were illegally handed over to Ethiopia by the British when Britain was giving up its rights in the protectorate of Aden." It adds "the North Yemen government has now raised the question of jurisdiction over the islands. It goes on to recommend bilateral negotiations which seem in fact to have been entered into before the time of this memorandum for it goes on to say that "[b]oth states . . . have informally mentioned the possibility of dividing the islands between the two of them. The proposal is to use the median line, which divides the Red Sea equally from both countries' coastal borders, as the dividing line . . . . Ethiopia rejected this proposal as disadvantageous."

32. See D. Bowett, The Legal Regime of Islands in International Law 48 (1978), where he says of islands lying within the territorial sea of a state, "Here the presumption is that the island is under the same sovereignty as the mainland nearby"; and he also interestingly quotes Lindley, The Acquisition and Government of Backward Territory in International Law 7 (1926), writing, it may be noted, in the mid-1920s that "An uninhabited island within territorial waters is under the dominion of the Sovereign of the adjoining mainland."

33. Foreign Office Memorandum dated 10 June 1930, prepared by Mr. Orchard.