One of the main problems encountered within the Second Committee was the definition of enclosed and semi-enclosed seas. It should be recalled that some Mediterranean States, including France and Greece, at least during the early
deliberations of the Second Committee, opposed the inclusion of specific rules on enclosed or semi-enclosed seas into UNCLOS on the basis that the concept was vague, undefined and purely geographical in nature. For example, the Greek
representative stated that ‘[…] to include a vague and undefined concept in the final instrument of the Conference would lead to insuperable problems’.
It may also be noted that some States, including the USSR and Iran, tried to distinguish between enclosed and semi-enclosed seas, and even proposed different definitions for the two terms. Despite such differences there was consensus from the very early stages of UNCLOS III that the legal definition of enclosed or semi-enclosed seas should not include those ‘geographically’ enclosed and/or semi-enclosed seas bordered by only one State, neither closed seas which do not have any connection to other seas or the oceans (e.g. the Aral Sea, the Dead Sea and the Caspian Sea).
Evolution of Article 122 at UNCLOS III
There were many attempts during UNCLOS III to provide and/or clarify a legal definition of enclosed or semi-enclosed seas. The reason for this is not difficult to ascertain. If specific rules for enclosed or semi-enclosed seas were to be included in UNCLOS, it was crucial for the Convention to provide a legal definition which clearly set apart ‘enclosed or semi-enclosed seas’ from other seas and oceans.
This section deals with some of the proposals which influenced the final definition of ‘enclosed or semi-enclosed seas’ contained in Article 122 of UNCLOS and which can provide some help in the interpretation of its various ambiguous terms. It seems that the biggest influence on the legal definition of enclosed or semi-enclosed seas was exercised by the Iranian proposal which, although providing separate definitions for ‘enclosed’ and ‘semi-enclosed’ seas, envisaged a
common legal regime for both categories. While presenting the Iranian proposal at the 43rd Meeting of the Second Committee, the Iranian delegate explained his delegation’s understanding of these terms and provided examples of seas which would fall under the two separate categories. He stated that:
[…] An enclosed sea was not a fully closed sea such as the Caspian Sea or the Aral Sea, which had no outlets to the oceans. It was, instead, a small body of inland water, such as the Persian Gulf and the Baltic Sea, which had at least one outlet to the open sea.
On the other hand, he explained that the term ‘semi-enclosed sea’:
[…] could be used in a broad sense to cover larger sea basins along the margins of the main ocean basins, more or less enclosed by a land mass whether continental or insular – and with one or more narrow outlets to the oceans. […]
The Iranian delegate referred to the Mediterranean as a ‘semi-enclosed sea’. This is particularly important in the light of the divergent views existing at that time on this issue. Although Vukas argued that the Iranian proposal was not satisfactory as it contained ‘too many juridical and geographically vague notions (inland waters, open seas, narrow outlet, sea basins, main oceans basin)’, it follows that, particularly through its structure and underlying logic, it exercised a considerable influence on the final text embodied in Article 122 of UNCLOS.
Another interesting proposal which also seems to have influenced the definition of ‘enclosed or semi-enclosed seas’ is the proposal submitted at the third session of UNCLOS III by the Informal group on enclosed or semi-enclosed seas. The
Group defined a ‘semi-enclosed sea’ as:
A sea surrounded by two or more States which is of such extent and characteristics as to prevent the full and entire application of one or more of the general provisions of this Convention with regards to maritime spaces and which
is accessible from the other seas or oceans solely through straits or narrow passages traditionally used for international navigation.
It is suggested that this proposal was the result of a realization by States surrounding enclosed or semi-enclosed seas that some general provisions of UNCLOS, particularly those regulating maritime spaces (e.g. the extension of the breadth of the territorial sea up to 12 nmi, of the EEZ up to 200 nmi, etc.), were not completely appropriate and could not be fully applied in enclosed or semi-enclosed seas. The proposal therefore envisaged a new criteria for defining enclosed or semi-enclosed seas in addition to a ‘two State requirement’ and to a ‘narrow outlet’, namely ‘the extent and characteristics of such seas which prevent the full and entire application of one or more general provisions of the Convention with regard to maritime spaces’. Unlike the Iranian proposal, the latter requirement and that of the ‘narrow outlet’ were not alternative, but cumulative. Nevertheless, the Chairman of the Second Committee introduced into the ISNT a definition of an ‘enclosed or semi-enclosed sea’ modelled on the Iranian proposal, but which did not include ‘the impossibility of the entire and full application of marine spaces as provided for in UNCLOS’, as one of the criteria to determine enclosed or semi-enclosed seas.
ISNT(Informal Single Negotiating Text) and RSNT(Revised Single Negotiating Text)
Article 133 of ISNT provided that:
For the purposes of this part, the term ‘enclosed or semi-enclosed sea’ means a gulf, basin, or sea surrounded by two or more States and connected to the open seas by a narrow outlet or consisting entirely or primarily of the territorial seas and exclusive economic zones of two or more coastal States. Like the Iranian proposal, the definition contained in the ISNT used both terms ‘enclosed’ and ‘semi-enclosed seas’. However, there was a notable difference as in the ISNT the two concepts were joined in a single definition (enclosed or semie-nclosed seas). Such difference, though, is of limited practical value, since the same regime applies to both categories of seas.
It is important to emphasize that although the definition contained in Article 133 of ISNT was applicable only to the part of the Convention dealing with enclosed or semi-enclosed seas, there were many States which were not satisfied with the broad nature of this definition and which sought to amend it during the following sessions of the Conference. Notwithstanding the various proposals in this respect, the Chairman of the Second Committee did not amend the definition, which was then incorporated into the RSNT and later on, as explained below, with an important amendment which found its way into Article 122 of UNCLOS.
Following a Turkish proposal at the ninth session of the Conference, the definition was amended to read as follows:
For the purposes of this Convention, ‘enclosed or semi-enclosed sea’ means a gulf, basin or sea surrounded by two or more States and connected to another sea or the ocean by a narrow outlet or consisting entirely or primarily of the territorial seas and exclusive economic zones of two or more coastal States. This was certainly a major amendment as it extended the scope of application of what later became Article 122 to other parts of UNCLOS.
Article 122 of UNCLOS
The definition of ‘juridical’ enclosed or semi-enclosed seas found in Article 122 of UNCLOS therefore applies ‘for the purposes of UNCLOS’ and it seems possible to assert that this extremely important change was triggered inter alia by the fact that references to the concept of ‘enclosed or semi-enclosed’ are found also in some other parts of UNCLOS. As mentioned, the final text of Article 122 does not distinguish directly between ‘enclosed’ and ‘semi-enclosed’ seas, as both concepts are joined in a single definition. According to Leanza, the most notable defect of the definition contained in Article 122 of UNCLOS is that it confuses ‘geographical’ with ‘juridical’ criteria and that ultimately it is too broad.
The two State requirements
The first requirement embodied in Article 122 is that a sea, gulf or basin must be surrounded by two or more States. This is a legal and not a geographical criteria and it provides that a sea, gulf or basin which fulfils other criteria provided by Article 122 of UNCLOS, but which is not surrounded by at least two States, does not qualify as a ‘juridical’ enclosed or semi-enclosed sea. Article 122’s definition therefore excludes from its scope of application geographical enclosed or semi-enclosed seas which are not surrounded by at least two States. As this requirement is legal or political in nature it obviously means that the legal status of a particular geographical enclosed or semi-enclosed sea may change, depending on political changes on its shores (e.g. secession of States).
The fulfilment of the ‘two State requirement’ is however not enough. In addition, a certain gulf, basin or sea must fulfil at least one of the two alternative elements enumerated in the second element of Article 122: (a) it has to be connected to another sea or ocean by a narrow outlet; or (b) it has to consist entirely or primarily of the territorial seas or EEZs of two or more coastal States.
It is important to emphasize that the two requirements are not cumulative, which in turn implies a broad scope of application and consequently a considerable number of geographically enclosed or semi-enclosed seas falling under the definition contained in Article 122 of UNCLOS.
Connection to another sea or the ocean by a narrow outlet
This requirement is a geographical one although it may potentially also have a legal dimension. Independently of that, the interpretation of the first alternative requirement can only lead to the conclusion that every gulf, sea or basin bordered by two or more States connected to another sea or the oceans by a narrow outlet, qualifies as a legal enclosed or semi-enclosed sea. This holds true independently of the size of a certain sea and/or also independently of the second alternative requirement to which reference is made below.
The rationale for such conclusion was explained by Vukas who was of the opinion that ‘[…] every sea connected to another sea or the ocean by a narrow outlet, even when of considerable size, is due to its poor connection to such other sea or ocean particularly vulnerable and deserves special protection’. This is mostly a result of the slow exchange of waters through such narrow outlets which as a general rule renders such seas vulnerable to the risk of pollution from all sources. The mentioned provision is not however completely ‘problem free’ since the term ‘narrow outlet’ is not expressly defined by UNCLOS.
Another interesting question relates to cases where a certain sea, gulf or basin is connected to another sea or oceans through more than one narrow outlet and/or one or more artificial outlets. A notable example is the Mediterranean which is connected to other seas and the oceans not only through the Strait of Gibraltar, but also through the Turkish straits of Bosporus and Dardanelles (with the Sea of Marmara and the Black Sea), and furthermore through the (artificial) Canal of Suez with the Red Sea and the Indian Ocean. As UNCLOS does not provide a clear answer it is necessary to resort to its travaux préparatoires.
It should be noted that throughout the proceedings at UNCLOS III several proposals were put forward by States to change the term ‘narrow outlet’ into ‘one or more outlets’. From the Virginia Commentary it derives that the majority of the informal proposals submitted at the sixth session of the Conference suggested a change from a ‘narrow outlet’ to ‘one or more narrow straits or outlets’.
Despite substantial support, the Chairman of the Second Committee did not introduce that change into the RSNT, although it is clearly discernable from his statement at the introduction of the RSNT that this was due to the dissatisfaction of certain States with the provision requiring co-ordination of activities in enclosed or semi-enclosed seas, and not with the proposals advocating for the inclusion of the term ‘one or more narrow outlets’ in the definition of enclosed or semi-enclosed seas.
From the travaux préparatoires it may be therefore clearly discerned that the majority of States participating at UNCLOS III supported the proposals according to which also ‘gulfs, basins or seas’ which are connected to another sea or ocean through more than one narrow or even by artificial outlets, should qualify as legal enclosed or semi-enclosed seas, and it may be implied that such interpretation was accepted also by the great majority of States at the moment of the signature of UNCLOS. It is indisputable that even a sea connected to another sea or ocean through more than one narrow outlet is still facing the same problems (environmental and others) as a sea connected to the oceans only through one narrow outlet. The exchange of water through outlets is generally still extremely slow and therefore such seas are not in a much better position than gulfs, basins or seas’ which are connected to the open oceans through only one narrow outlet.
Lastly, an interesting question is posed by the meaning of the expression ‘another sea or ocean’ to which the narrow outlet is connected and particularly by the expression ‘another sea’. Does it refer to a sea which is not itself a ‘juridical’ enclosed or semi-enclosed sea? According to such interpretation, the Black Sea would not qualify as a juridical enclosed or semi-enclosed sea which would be a rather illogical conclusion. The correct interpretation seems to be that the expression ‘another sea or ocean’ does not distinguish between ‘enclosed or semi-enclosed’ and other seas and oceans and that it may refer also to another ‘juridical’ enclosed or semi-enclosed seas.
Consisting entirely or primarily of the territorial seas and EEZs of two or more coastal States
The second alternative criteria is represented by the situation where a gulf, basin or sea consists entirely or primarily of the territorial seas and EEZs of two or more coastal States. That provision is again an interesting mixture of geographical and prima facie legal requirements. It seems to aim at protecting seas of limited dimensions and, accordingly, it no longer contains the ‘narrow outlet’ requirement.
The underlying logic of the second alternative was explained again by Vukas who stated that ‘[…] small seas are susceptible of being vulnerable to pollution and over-exploitation of their natural resources […]’. Even that scholar, however, questio ned whether the drafters of UNCLOS had not gone too far by drafting a provision according to which even such large seas as the Norwegian Sea, the Greenland Sea or the Tasman Sea may be considered as ‘juridical’ enclosed or semi-enclosed seas. It seems however questionable whether, for example, the Tasman Sea may be deemed to be ‘surrounded’ by Australia and New Zealand as required by the definition of enclosed or semi-enclosed seas in Article 122 of UNCLOS, and similar considerations apply also to the Norwegian and the Greenland Sea. Furthermore, as both Articles 122 and 123 of UNCLOS and the chapeau of Part IX of UNCLOS refer to ‘enclosed or semi-enclosed seas’, it would seem that this fact should also govern the application and interpretation of the whole of Article 122 of UNCLOS, thus limiting its geographical scope of application. This seems, at least to a certain extent, to rebut the argument that ‘[…] almost any sea could be called semi-enclosed’.
Notwithstanding the above, one should be extremely cautious when using the term ‘smaller sea’ in relation to the criteria embodied in the second element of Article 122 of UNCLOS. If the maximum breadth of the EEZ (200 nmi from the baselines) is taken into account, it would follow that a sea which is entirely composed of the EEZs of its coastal States could be up to 400 nmi in width; and an obvious question is whether a sea of such dimensions is still a ‘small’ sea.
Furthermore, the definition also covers seas, bays or basins which consist primarily of the territorial seas and EEZs. The term ‘primarily’ in this context is again not precise and can lead to different interpretations. Would it be possible to say that in order for a ‘gulf, basin or sea’ to qualify as an ‘enclosed or semi-enclosed sea’ more than half or eventually two-thirds of its surface shall be covered by territorial seas or EEZs of coastal States? Neither UNCLOS nor State practice provide a straightforward answer. However, if we interpret the provision in accordance with the ordinary meaning to be given to the terms of the Treaty, it seems that the term ‘primarily’ implies a ‘gulf, basin or sea’ which is for the most part, that is mainly, composed of the territorial seas and EEZs of two or more
coastal States. While a specific percentage cannot be given, it seems possible to argue that approximately two-thirds of the area of sea would have to be composed of territorial seas and EEZs in order to fulfil the requirement of primarily within the meaning of Article 122. It is suggested therefore that seas with prevailing areas of high seas within them would not qualify as a ‘juridical’ enclosed or semi-enclosed seas, notwithstanding the fact that they are surrounded by two or more ‘States’.
Should there be a proclaimed EEZ?
An interesting question also arises as to whether the wording of Article 122 of UNCLOS refers exclusively to a proclaimed EEZ or whether it also includes ‘potential EEZs’ or sui generis zones of jurisdiction. This point again calls for an interpretation in the light of the ‘object and purpose’ of Part IX of UNCLOS in general and of Article 122 in particular.
The question at stake is therefore whether Article 122 of UNCLOS requires the formal proclamation of EEZs by two or more coastal States or whether it is sufficient that the breadth of a certain sea is such that it potentially allows for the extension of coastal State jurisdiction up to the extent referred to in the second element of Article 122. This consideration is particularly relevant in the context of the Mediterranean and Adriatic Seas where just a few States have proclaimed full EEZs; some others have proclaimed sui generis zones based on the concept of the EEZ; while many States have so far refrained altogether from extending their maritime jurisdiction. As mentioned above, the current situation in the Mediterranean may be described as one where the high seas still prevail.
From a prima facie reading of Article 122, and after taking into account the ordinary meaning of the terms used, one may be tempted to conclude that the provision in question refers exclusively to proclaimed EEZs. But is this really the case? It should be recollected that the primary reason for the inclusion of the second alternative requirement within the definition of a juridical enclosed or semienclosed sea was the need for the protection of ‘smaller seas’ which are, due to their size, particularly vulnerable to all kinds of pollution. Moreover, from the travaux préparatoires at UNCLOS III, it is possible to discern that what the majority of States had in mind were the geographical and natural characteristics of such seas and not whether or not there is one or more proclaimed EEZ. The relevant criteria, therefore, seems to be predominantly geographical in nature and not directly linked to the legal notion of the territorial sea and/or of the EEZ. In other words, it is the ecological and geographical characteristics of such seas which require the application of enhanced co-operation among bordering States and not whether a certain State has extended up to the maximum extent the breadth of its territorial sea or has proclaimed an EEZ or similar zone of jurisdiction in an enclosed or semi-enclosed sea.
It seems therefore possible to agree with Škrk that the definition found in Article 122 suggests the ‘presumed, not the established EEZ’, as otherwise one would face the absurd situation where an endangered sea of limited dimensions becomes a ‘juridical’ enclosed or semi-enclosed sea only after the proclamation of EEZs by its coastal States. Additionally, it follows that during the proceedings at UNCLOS III, not many States seriously envisaged a situation where certain States would refrain from proclaiming an EEZ and this may in turn explain why States participating at UNCLOS III did not feel the need to address the difference between the actual and potential EEZ in relation to what later became Article 122.
It is possible to conclude that reference to the territorial sea and the EEZ in Article 122 refers to the potential or presumed EEZ (and territorial sea) and not to the actual (legal) extension of jurisdiction by coastal States. Therefore, for a certain ‘gulf, basin or sea’ to qualify as a legal enclosed or semi-enclosed sea, there is no need for an actual proclamation of EEZs, nor for the actual extension of the breadth of the territorial sea up to the maximum extent permitted by international law, as an ‘inchoate’ right to do this seems to be enough.