The Definition of the ‘Undefined’ Straits Used for International Navigation

Abstract

It is true that the ocean is subdivided into marginal basins and seas, all of which are interconnected by straits. From the purely geographical point of view, straits are considered to be “a narrow passage between two large bodies of water which have different density stratification and sea level elevations due to the independent, natural processes in each basin”. In this regard, the International Hydrographical Organisation defined geographical straits as “a narrow passage between landmasses or islands or groups of islands connecting two larger sea areas”. Also from a geographical viewpoint, although more concise, is the concept provided by G. Gidel, who stated that straits are “un passage maritime resserré entre deux terres”.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

It is true that the ocean is subdivided into marginal basins and seas, all of which are interconnected by straits.¹ From the purely geographical point of view, straits are considered to be “a narrow passage between two large bodies of water which have different density stratification and sea level elevations due to the independent, natural processes in each basin”.² In this regard, the International Hydrographical Organisation defined geographical straits as “a narrow passage between landmasses or islands or groups of islands connecting two larger sea areas”.³ Also from a geographical viewpoint, although more concise, is the concept provided by G. Gidel, who stated that straits are “un passage maritime resserré entre deux terres”.⁴

However, these purely geographical definitions, although they serve as a reference, they are not valid as regards providing a legal concept for this maritime area from the point of view of Public International Law. Taking into account that not all the geographical straits come within the area of application of the specific regime stipulated in Part III of the 1982 Convention, but only those straits used for international navigation, as stated in this third Part

As pointed out by Hyde⁵ it is evident that ‘there are straits and straits’. And only the straits “used for international navigation” are classified as ‘international straits’, and only these straits fall within the specific regime stipulated in Part III, sections 2 and 3, of the 1982 Convention.⁶

In fact, Part III of the UNCLOS is entitled: “Straits used for International Navigation”. The use of these terms shows that the 1982 Convention clearly follows in the wake of the 1958 Geneva Convention which conventionally confirmed this expression following the dictates of the ICJ in the Corfu Channel Case in 1949; although the UNCLOS equates this with the changes carried out in the Law of the Sea, such as those concerning the breadth of territorial waters or the appearance of new maritime areas such as the exclusive economic zone or the waters of archipelago.

However, the UNCLOS did not just retake the expression “straits used for international navigation”, but it is also presented as the heir to the ‘definition’, or better the ‘indefinition’ of the expression throughout its article. In the words of J.N. Moore, “UNCLOS has not altered or clarified the existing uncertainty in customary international law over the definition of ‘straits used for international navigation’”.⁷

In fact, the Convention does not define what straits used for international navigation are and, therefore, does not specify which straits come within the scope of Part III; likewise it fails to establish the differentiation between straits which are used for international navigation and those which are not.

There is no doubt that this accumulation of circumstances means that it is necessary to carry out the previous, complex task of properly conceptualising this legal shape which is the subject of our study, insofar as this is possible. Although we must point out that the name, the terminology, is not the decisive factor in the definition of straits, as shown by the fact that areas which clearly come within the category of international straits are termed “channels” (St. George’s Channel, North Channel, Santa Lucia Channel,…), or “passages” (Martinique Passage, Aruba-Paraguana Passage, Jacques Cartier Passage,…). Besides the particular denomination which they might receive in a determined language, as occurs with “belt” in Danish (Litlle Belt, Great Belt), “proliv” in Russian (Proliv Karskiye, Proliv Friza,…), “kaikyo” and “suido” in Japanese (Tokara Kaikyo, Tsugaru Kaikyo, Shikotan Suido, Taraku Suido,…) and “selat” in Malayan–Polynesian (Selat Sunda, Selat Lombok, Selat Alas,…). Thus, this aspect has no judicial importance.

The same occurs with the maps as regards whether a determined strait is identified or not. These do not have probative value in this respect; and constitute mere tools of an auxiliary nature.

3.1 Some Proposals of Definitions Put Forward Within the Framework of the III Conference

The representative of the German Government said in his reply to the request for information from the Preparatory Codification Committee of The Hague of 1930 that “the notion of straits has not yet been fixed in Public International Law”.⁸ This statement continues to be in force after the adoption of the Geneva Convention of 1958 which, although, as we have seen, it establishes a specific regime for the straits used for international navigation in its article 16.4, it does not contain definition of these.⁹ And, although it may seem strange (especially if we take into account the repeated doctrinal vindication concerning the need to fill this gap), it still remains unresolved by the UNCLOS.

It is true that, throughout the III Conference of the United Nations on the Law of the Sea, the advisability of including a definition in this regard was discussed, although the difficulty involved in achieving a concept that would be generally accepted was also pointed out. In this regard, several explicit and implicit proposals of a definition of the “straits used for international navigation” were presented.

One proposal, which was co-sponsored by ten Arab States, referred to the straits which communicate two parts of the high seas and are habitually used for international navigation.¹⁰

The proposal submitted by Canada was more precise, as Canada was especially interested in expressly including the reference to habitual use for international navigation; its intention was to exclude the Northwest Passage from this classification, a group of straits only used sporadically and seasonally. In this respect, Canada defined an international strait in the following terms:

An international strait is a natural passage between land formations which:

• (a)

  • (i) Lies within the territorial sea of one or more States at any point in its length and

  • (ii) Joins two parts of the high seas…

• (b) Has traditionally been used for international navigation.¹¹

The British proposal stated that the articles on the passage of straits used for international navigation should be applied to any straits or other water course in the geographical sense, which are used for international navigation and connect two parts of the high seas; the proposal of the “Private Group of States” contained an almost identical description.¹²

However, neither the articles project co-sponsored by Malaysia, Morocco, Oman and Yemen, nor the text submitted by the “Group of coastal States of Straits”, contained any reference to the geographical factor, and stressed the fact that the straits affected were the ones included in the territorial sea of one or several States, that is to say, the legal factor.¹³

Finally, the Conference decided not to include a definition in the text of the Convention, and, as we stated above, this endorses its Part III to the “Straits used for international navigation”, but it did not specify what these terms meant at any point.

In this regard, L.M. Alexander stated that the expression “is so inexact as to render the phrase itself almost meaningless”;¹⁴ a circumstance which constitutes a clear controversial factor as there may be a number of interpretations and a lack of essential precision.¹⁵

In such a situation, we are clearly addressing a problem of interpretation – one of the many which the UNCLOS gives rise to, as we will see – which must be resolved through recourse to the interpretational criteria included in article 33 of the 1969 Vienna Convention on the Law of Treaties, as this is a treaty authenticated in several languages; its bona fide nature, the text, the context and the current meanings of the terms are the first instruments of interpretation, not forgetting the complementary recourse to the preparatory work and the circumstances in which it was drafted. It is a question of adopting the common sense interpretation, as “a strict (or ‘mechanical’) interpretation could well be found to be inappropriate in certain settings”.¹⁶

In this respect, we understand that the content of the proposals submitted by the States throughout the III Conference, referred to above, constitute a fundamental aspect of this clarifying work. Bearing these proposals in mind, it is possible to state that, to a greater or lesser extent, these are all tributaries of the three factors which has traditionally constituted the doctrinal definition doctrinal of ‘international straits’ or ‘straits used for international navigation’: geographic, legal and functional.

As stated by Hart, “a better approach seems to be that of seeking a ‘concept’ which consists of criteria, rather than a ‘definition’”.¹⁷

This leads us to conclude that the precise conceptualisation of the “straits used for international navigation” must include the determination of the geographic, legal and functional factors so that, in order to be able to identify an international strait,¹⁸ these three components must be employed as a type of Holy Trinity.

3.2 The Geographical Factor

According to E. Brüel, a strait in the geographical sense is understood to be “a contraction of the sea between two territories, being of a certain limited width and connecting two seas otherwise separated at least in that particular place by the territories in question”.¹⁹ Based on this concept and following this author, we can state that there are four features or defining geographical characteristics of an international strait: a natural sea passage, a contraction of the sea, separating two land areas, and uniting two areas of the sea.

(a) In the first place, it must be a part of the sea not created artificially. Thus, the adjective ‘maritime’ distinguishes the straits from other water channels which do not form part of the sea, as is the case of rivers which constitute fluvial areas. While the adjective ‘natural’ differentiates these from other maritime water courses artificially created by man, such as canals, for example the Kiel, Suez or Panama Canals, which are subject to the absolute sovereignty of the States they cross, unless there is an agreement otherwise. Evidently, rivers and artificial canals do not fall within the category of straits.²⁰

As regards this final aspect, R. Lapidoth²¹ puts forward an interesting question on what the author classifies ‘semi-natural passages’, that is to say, the case of strait which becomes navigable as a result of excavation work carried out by a State. Is this subject to the legal regime of straits? The ICJ made a pronouncement on this particular point with regard to the Fisheries Case (United Kingdom vs. Norway), in relation to the legal classification of l’Indreleia – a typical geographical formation of the ‘skjærgaard’, literally a coast full of farallones and rock walls –, a navigation route in the interior of the Norwegian straight, base lines, stating that:

The Court is bound to observe that the Indreleia is not a strait at all, but rather a navigational route prepared as such by means of artificial aids to navigation provided by Norway. In these circumstances the Court is unable to accept the view that the Indreleia, for the purposes of the present case, has a status different from that the other waters included in the ‘skjægaard’.²²

R. Lapidoth²³ holds that, unlike the majority opinion which understands that the Court has given a negative response to the inclusion of these semi-natural passages within the regime of straits, this conclusion cannot be reached with this extract.

We should remember that the case in question involved passages located in interior waters of Norway. The Court was not called on to pronounce on the application of the regime of straits to these passages, but on the thesis of the United Kingdom that these passages are located in territorial waters and not in interior waters; on rejecting this thesis, the Court specified that it was going to analyse the question “for the purposes of the present case”,²⁴ that is to say, from the categorisation of the waters as interior or not. This specification and the particular circumstances of the case enable us to conclude with R. Lapidoth that the decision of December 18, 1951 in the Fisheries Case has not laid down a general rule in relation to semi-natural passages.

From our point of view, we consider that the response to this interesting question is found in the practice of the States. In this regard, we observe that there are many important straits recognised as such – for example, the Turkish Straits, the Straits of Malacca and the Straits of Singapore –, were the subject of artificial, technical work in order to make them more navigable and it has never been suggested that they have ceased to be straits and become artificial canals. In addition, if each navigation route enabled due to the execution of coastal, technical work is withdrawn from the regime on straits, this will enable the ‘Straits’ States to exclude a strait from this regime as they wish.²⁵

There is also the case of the appearance of a new strait, that is to say, a passage which arises or becomes navigable as a result of an earthquake. We consider that, in this case, the natural character of the strait as an essential requirement of the geographical component is fully complied with, which would entail its inclusion in this category on condition that the other components required are present.

In any case, we understand that the navigable or non-navigable character of a strait and, consequently, the execution of artificial work on the coast in order to make a natural passage navigable, or to make it more safely navigable, do not influence the geographical component which will always exist when there is a natural passage of maritime waters; this particular aspect will only occur in relation to the functional component which we will refer to below.

(b) In the second place, there must be a contracting of the sea in such a way that the water in the straits, compared with the adjacent waters, is more reduced in extension than these. This aspect is very imprecise as, from a geographical point of view, there is no specification of the width the contraction must have. The determination of the maximum width is important for the purposes of the inclusion of the strait as an international strait although this question is part of the legal component.

In addition, the geographical point of view is only limited to the superficial aspect of the strait and does not take other factors into account such as the depth.

(c) Furthermore, it must separate two land areas, and it is indifferent whether these are two continental territories (the Strait of Gibraltar, Bab El Mandeb, the Strait of Hormuz,…), two islands (the Menorca Channel, the Strait of Bonifacio, the Cook Strait,…), or a continental territory and an island (the Strait of Messina, the Strait of Corfu, the Channel of Zanzibar,…).

However, we believe that it is important to point out that, although it is irrelevant as regards the definition of straits, the type of territories separated do have legal consequences as concerns the navigation regime applicable to the straits, for example, in the case that mainland and island territory of the same State are separated with more than one option of passage, the navigation regime is different as the right of passage in transit will not be in force, but the right of innocent passage which cannot be suspended, as we will see below.

(d) Finally, the straits must unite two sea areas, two maritime areas where the straits are the natural connection between both. The size of the waters connected is not a determining factor either. These may be two oceans – inter ocean straits – (the Magellan Strait, Le Maire Strait, the Torres Strait,…), two parts of the same ocean – intra-ocean straits – (the Bass Strait, the Robenson Channel, the Strait of Juan de Fuca,…), two seas – inter-sea straits – (the Sound, the Western Chosen Channel, the Straits of Makassar,…), or two parts of the same sea – intra-sea straits – (the Strait of Messina, the Kalmar Sound, the Strait of Otranto,…). A fifth category of Straits are those which join an ocean with a sea – ocean-sea straits – (the Strait of Bering, Bab el Mandeb, the Strait of Gibraltar,…).

As regards this last note and as in the previous case, an important legal question arises concerning the union between the geographical and legal components and the proof of the interdependence between the defining components of an international strait. This involves the nature of the maritime areas connected by the waters of the straits. In this regard, we find that most of the doctrine analyses this matter exclusively from the geographical point of view. However, we do not share this point of view as we understand that the classification of the marine areas connected is a question, which, together with its geographic aspect, falls fully within the legal area as it is International Law and not geography which defines and delimits which spaces may be considered to be the high seas, exclusive economic zones, territorial seas or interior waters of the States.

As regards this matter, the ICJ in the Corfu Channel Case set out the decisive criteria for the definition of an international strait as its situation as a connection component between two parts of the high seas.²⁶ However, the 1958 Geneva Convention extended the areas connected referring to its article 16.4, not only to those which join a part of the high seas, but also those where the straits connect the high seas with the territorial seas of a foreign State; which gives rise to two passages. There are no doubts as regards the customary nature of the first type of passage, which connects two areas of the high seas, but it is not very clear whether this nature can be duly applied to the second type of passage. In fact, this second case does not appear in the preliminary work of the ILC and was incorporated at the end in order to promote the Israeli claims as concerns a navigation in the Gulf of Aqaba and through the Strait of Tiran.²⁷

In the opinion of R. Lapidoth,²⁸ there is no doubt that this involves a customary norm and its affirmation is founded on several declarations made by maritime powers. However, this does not take into account the many contrary declarations made by other States – especially the Arab States²⁹ –, stating the exclusively conventional nature of this rule; nor the fact that neither doctrine nor case law had referred to these types of passages before 1958.³⁰ Therefore, it is inadvisable to classify the norm included in the 1958 as customary as regards this second type of passage.

Evidently, when the III Conference and the adoption of the UNCLOS took place, this polemic was overcome. In this regard, the range of waters connected by international straits is again extended in the 1982 Convention, and logically includes the new maritime areas which arise in the III Conference of the United Nations on the Law of the Sea, such as the exclusive economic zone (EEZ) and archipelagic waters. Consequently, several types of passages can be distinguished. One is the classical strait which connects two areas of the high seas which is included in the news EEZ area and is a connection between the high seas and the EEZ and the high seas or the EEZ (article 37). The polemic kind concerning the connection between the high seas or the EEZ and the territorial waters of another State was also codified (article 45).

However, two more cases were added: the straits which separate an island from a mainland State, when the other side of the island has a similar high sea or EEZ corridor (article 38); and the exceptional case of the connection of the interior waters of a State, in the hypothesis that the tracing of the straight base line has resulted in enclosing waters which were not considered as such as interior waters (article 35(a)).

Although it is not expressly include in Part III, it may be inferred that the stipulations in Part IV regarding the “Archipelagic States”, which in the case of international straits between the islands of an archipelagic State, that is to say, in the case of “Archipelagic Straits” the waters of these straits, as well as the maritime areas connected by these are basically archipelagic waters of the coastal State of the Straits on condition that this State is an archipelagic State and has delimited its maritime areas with archipelagic baselines (article 47). This occurs especially with Indonesia (Bangka Passage, Strait of Lombok, Selat Alas, Strait of Sunda,…) and the Philippines (Balintang Channel, San Juanico Strait, Strait of Maqueda, Strait of Polillo,…); although we also find straits of this type in the Bahamas (Crooked Passage, Northeast Providence Channel, Northwest Providence Channel), Trinidad and Tobago (Galleons Passage), Papua New Guinea (Vitiaz Strait, Dampier Strait,…), Salomon Islands (Indispensable Strait, Manning Strait,…), and Fiji (Vatu-i-ra Channel). Although, in this last case of ‘archipelagic straits’ the legal regime applicable to these is not that of Part III, but that of Part IV.

In line with this last observation and in relation to the distinct nature of the maritime waters connected by the ‘international straits’, we must finally note the existence of a gap in the framework of Part III in particular, and in the UNCLOS itself in general. This lack refers to the possibility that a strait connects an area of the high seas or EEZ with the archipelagic waters of a State. This is not a hypothesis, it is a reality. This situation arises in the Wetar Strait, between Timor Leste and Indonesia, which connects the EEZ or high seas of the Sea of Arafura with the Indonesian archipelago waters.

3.3 The Legal Component

It is usually established that there are several hundred straits on the planet in the geographical sense of the term explained above, but only certain number of these are included in this category of the Law of the Sea. In order to be included, certain legal aspects which we have mentioned above and which are balanced between the geographical and the legal must be complied with, such as those related to the land surfaces separated, and the maritime spaces connected, and a specific legal component, namely, that the waters of the strait form part of the territorial sea of one or more States. It is not necessary that the totality of the waters of the straits are included in the territorial sea, it is sufficient that overlapping occur in some parts of the strait.³¹

This component is implicit in the Geneva Convention of 1958 as it is included in the part dedicated to the territorial sea, and more specifically in section III dedicated to the right of innocent passage of foreign vessels through the territorial sea. Years before, several authors, such as J. Westlake³² had pointed this out at the beginning of the twentieth century, when he stated that the waters of the strait form part of the territorial sea of on or more States as the Straits considered in International Law were those whose width is less than double the extension of the territorial sea of the coastal States of the strait in question. A thesis which was included in the proposals of the Preparatory Committee of the Hague Conference of 1930, and included in the Conference itself when the Principal Committee pointed out that the waters of the straits which do not form part of the high seas are territorial seas.³³

For legal purposes, therefore, an international strait is a geographic strait in which the territorial sea of the coastal State or States does not permit the existence of a freely navigable through this seaway.³⁴ If this is not so, in the event that double the width of the territorial sea is exceeded, that is to say, that the width of the strait exceeds 24 nautical miles in all its parts (on condition that the width of the respective territorial sea is 12 miles, as, if it is less, the maximum limit referred to would be proportionally reduced), this would be geographic strait but not an international strait as part of its waters would not be submitted to the sovereignty of any coastal State and there would be a freely navigable route and freedom to fly over the strait as it would be high seas or an EEZ.

It is important to stress, therefore, that there is a direct relationship between the extension given to the territorial sea and the number of straits considered as such for legal purposes; only when the width of the strait takes in the sums of the territorial seas of the coastal States (regardless of the extension which have these have decided to attribute to themselves), the strait will be legal. If there is a band of high seas or an EEZ between these, it will not be legal.

In this regard, it will be necessary to take into account if there are islands in the strait belonging to one or more coastal States, which must inevitably be taken into consideration for the purposes of establishing the width of the straits and, therefore, its legal nature. The existence of islands in a strait may convert a purely geographic strait into a legal one. Evidently, these islands will have their own territorial seas attributed and these may overlap with the territorial sea of the mainland coast, which will mean that the waters of the strait are territorial sea and, therefore, the legal component which would not be complied with otherwise is verified.

This circumstance will obviously depend on the breadth of territorial sea which the coastal State decides to attribute to its mainland and to its islands located in the strait. If it chooses 12 miles, it will probably mean that the strait becomes a ‘strait used for international navigation’ in the sense of Part III. This occurred with the Bass Strait, between the island of Tasmania and the Australian mainland where there are about 50 islands; Australia’s attribution of a territorial sea of 12 miles to all of these meant the transformation of this strait, traditionally ‘free’, into an ‘international strait’. A circumstance which was repeated in the Torres Strait, between Australia an Papua New Guinea; the existence of islands belonging to both States, with their own territorial seas, reduced the minimum distance of the navigable channel in Australian territorial waters to 20 miles.

However, it may occur that the coastal State or States decide to attribute an area of territorial sea of less than 12 miles, which restricts this metamorphosis. This occurs, for example, with the Kattegat, strait of the North Sea which is over 24 miles wide and takes as reference points the respective mainland coasts of Denmark and Sweden; therefore, a priori, it does not comply with the legal component. However, there are two islands in this strait (the Danish islands of Laesø and Anholt) and if these are taken into consideration, they divide the strait into two channels; neither of the channels exceeds 24 miles in width. In fact, the width is reduced to 12 miles in the narrowest part between the Danish peninsula of Jutland and the island of Laesø, and is 22 miles wide in the east channel, between this island and Sweden. This circumstance would make the Kattegat comply with the legal component and, consequently, it would enter within the framework of application of Part III of the UNCLOS. However, faced with the common extension of the territorial seas to 12 miles, at the end of 1979, Denmark and Sweden agreed to attribute a territorial sea of 3 miles to each other, leaving a strip for free navigation in both channels of the strait amounting to 6 and 16 miles, respectively; which means that the provisions of Part III are not applicable to the Kattegat.

This also occurs in the Strait of Kadet, between Denmark and Germany, and in several Japanese straits.

In this regard, it should be pointed out that, in fact, the effect of islands on a strait is that it divides the strait into one or more channels which require the application of the legal component test in order to classify each of these as ‘international straits’. Channels which rarely appear individualised as straits in practice. This individualisation, however, occurs in a certain number of cases. Such is the case of the Strait of Korea – between Korea and Japan –, where the Japanese island of Tsushima divides this into the Western Chosen or West Korea Strait (23 miles), and the Tsushima or East Korea Strait (25 miles); however, neither of these enter into the scope of application of Part III because there is a high sea route in these, as consequence of Korea’s and Japan’s decision to attribute themselves a territorial sea of 3 miles in this area.

Mention should also be made of the Strait of Luzon – between Taiwan and the Philippines –, divided into three channels as a result of the multiplicity of islands: Bashi Channel (53 miles), Balintang Channel (24 miles) and Babuyan Channel (14 miles); as a result of this individualisation, the last two mentioned are clearly ‘international straits’.

This legal factor, which we refer to, is expressly mentioned in article 36 of the UNCLOS which stipulates the following:

This Part does not apply to a strait used for international navigation if there exists through the strait a route through the high seas or through an exclusive economic zone of similar convenience with respect to navigational and hydrographical characteristics; in such routes, the other relevant Parts of this Convention, including the provisions regarding the freedoms of navigation and overflight, apply.

As examples of geographical and functional straits which clearly do not enter within the scope of application of Part III of the UNCLOS, as stated in article 36, because their widths exceed 24 nautical miles, we can cite: the Strait of Otranto, Malta Channel, Skagerrak, Chetvertyy Kurilskiy Proliv, Strait of Florida, Yucatan Channel, Caicos Passage, Anegada Passage, Strait of Taiwan, Balabac Passage, and the Mozambique Channel, among many others.

Besides the traditional case we have referred to, which stipulates that the waters which form part of the straits are territorial seas, we must take into account another two special situations of international straits whose waters are not territorial seas, and which are also referred to in the UNCLOS.

Firstly, the particular case included in article 35(a) of the Convention concerning straits whose waters have become interior waters as a result of straight base lines whereas previous to these lines being drawn, they were not interior waters, that is to say, they were territorial waters or high seas; following the provisions in article 5.2 of the Geneva Convention of 1958. It is evident that, in this hypothesis, despite the applicability of Part III, the waters which form part of the straits continue to be interior waters and not territorial seas.

Secondly, there is the case referred to above of the ‘archipelagic straits’ included in Part IV of the UNCLOS. As we have seen, in this case the waters which form part of this special type of international straits are not territorial seas either, but archipelagic waters.

3.4 The Functional Component

Together with the geographical and legal components, the straits must comply with a final criterion of functional nature: the use of the straits for international navigation. A condition which, in turn, presupposes the intrinsic requisite that this seaway must be navigable, if it is not, evidently it would not enter the legal category being analysed at this point.

We must state that the navigation requirement refers exclusively to maritime navigation and not air navigation, which clearly does not have the navigability problems arising on the sea. In fact, in the light of an analysis of the doctrine on the subject, as well as the opinions of the States in this regard, except for the United States, for obvious reasons, the conclusion can clearly be drawn that, when reference is made to the use of a strait for international navigation or not, this is done thinking solely of sea navigation, and, surface navigation, not underwater. This is the most logical position if it is taken into account that we refer to a maritime area whose waters are territorial sea, where innocent passage referring exclusively to maritime surface navigation has traditionally been in force.

Until the arrival of the UNCLOS, the overflight of aircraft and underwater navigation were not allowed in territorial seas including straits. Thus the reference to the straits used for international navigation, codified in the conventional regime of 1958 and reproduced in the regime of 1982, must be understood with regard only to maritime surface navigation. In fact, the straits which are not used for maritime navigation or are not navigable do not give rise to overflight problems or, even that underwater navigation may be feasible. However, it is understood that they do not comply with the functional requirement, remain outside Part III, and are regulated by the general regime of Part II concerning territorial seas.

This is the case of the Dampier Strait in Papua New Guinea, located between the Island of Umboi and New Britain, which joins the Sea of Bismarck to the north with the Salomon Sea to the south, but is not navigable.

Three of the four straits which separate the Andaman Islands, a group of five islands which form part of the Indian territory of the Union of the Andaman and Nicobar Islands, located in the Bay of Bengal; specifically, the Strait of Austin between North and Middle Andaman, the Strait of Homfray between Middle Andaman, Baratang and the northern point of South Andaman, and the Strait of Andaman between Baratang and South Andaman are not navigable. Only the Strait of Macpherson between South Andaman and Rutland Island is used for international navigation.

The Russian straits located in the Arctic, Petukhouskii Shar, Sharapov Shar, Iugorshii Shar and Red Army are not suitable for surface sea navigation, due basically to the intense and continual fog.

Furthermore, McClure Strait in the Canadian Arctic archipelago is not navigable either due to the fact that it is blocked by ice throughout the year, although some submarines can cross it submerged.

On this point, as we have stated, all these straits which form part of the territorial seas of one or more States and which, despite their geographical situations, are not navigable or are not used for international navigation, remain outside Part III and they are, therefore, subject to the provisions of the Part II of the 1982 Convention on the right of innocent passage which may be suspended in.

Undoubtedly, when the UNCLOS refers to “straits used for international navigation” it is expressly including the functional component which characterises an international strait, and copies the same formula used by the 1958 Geneva Convention, which is initially found in the Corfu Channel Case. In this case, the ICJ stressed the second determining criteria, although not the most decisive one, for the definition of an international strait (the fundamental criteria was the geographical one): the fact that the strait is used for the purposes of international navigation,³⁵ – maritime surface navigation.

But what does ‘used for international navigation’ mean exactly?

Undoubtedly, the question of the ‘use’ of the strait does not have mere theoretical relevance but is crucial and very difficult to interpret. While, the geographical and legal components are easy to determine, it is extremely complex to specify the functional usefulness as this is, in the opinion of S.C. Truver,³⁶ a very vague criteria. In fact, this is the component which generates the greatest discrepancy between States as those who use the straits always hold that this seaway is used for international navigation, while the coastal States challenge its international nature and reject such use, It occurred in the 1970s with the Strait of Malacca, which Indonesia and Malaysia refused to classify as international strait, and with Canada in relation to the Northwest Passage, which Canada holds is part of Canadian interior waters and is not open to international navigation.

We can also state that one of the main reasons which put the successful termination of the III Conference at risk was the fact that the majority of the ‘straits’ States were blinded by the desire to exclude their own straits from such use, and, therefore, from this definition.³⁷

As pointed out by R.R. Baxter, “it is impossible to answer in the abstract how many straits meet this requirement of being ‘useful’ for the international navigation”;³⁸ and the opinions of the doctrine in this regard are varied and even disparate. Therefore, the proposals such as those of C. Jiménez Piernas are not off the mark. In his opinion, this minimum criterion of usefulness must be corrected through the introduction of a number of objective factors, such as the number of vessels and flags, the tonnage and the value of the cargoes,³⁹ together with the use of such straits by warships.

In this same regard, E. Brüel points out that, in order to determine the importance of a strait, subjective criteria are not sufficient, but impartial considerations are necessary such as, “the number of ships passing through the strait, their total tonnage, the aggregate value of their cargoes, the average size of the ship, and especially, whether they are distributed among a greater or smaller number of nations”.⁴⁰ All these factors are guidelines and none of them can be considered to be decisive. The fundamental point for this author is that the interest in the use of the strait has a world nature.

Similarly, Ch. de Visscher pointed out that the data used by the Court in the Corfu Channel Case, are not conclusive. In his opinion, the number of vessels, the total tonnage, the value of the cargo and the number and variety of flags represented are the four useful factors for determining the importance of a strait.⁴¹

J.D. González Campos has a similar opinion and states that the idea of usefulness expressed by the ICJ is satisfied with a minimum of special usefulness of the sea passage, which not only constitutes an essential or necessary route for maritime navigation, but is possibly a useful or convenient route.⁴²

However, O’Connell⁴³ holds that mere potential usefulness is insufficient. An opinion shared by T. Treves.⁴⁴ Contrary to this, R. Lapidoth understands that a weak flow of traffic may be sufficient on condition that the strait can serve for navigation.⁴⁵

Two very clear conclusions can be obtained from the opinions given here. The first is that there is no agreement in the doctrine regarding what must be understood by the expression “used for international navigation”. The second is that there is no unanimity as regards the degree of use of a strait required for it to be considered as international. This use has never been “quantified”.

In our opinion, we understand that ‘use’ cannot be confused with ‘importance’. Moreover, criteria such as those explained above: tonnage of vessels, value of the cargo, types of vessels, variety of flags, etc. do not refer to establishing the usefulness of a strait for navigation, but its importance its strategic, commercial, defensive or communications importance has no relevance whatsoever as regards the applicable legal regime. In addition, these criteria vary with time; therefore, they do not contribute legal security in this respect. The functional component does not distinguish between straits; it only requires that the strait be used for navigation, regardless of the type of use. The fact that the States prefer to use alternative communication routes does not subtract from the internationality of a strait which is navigated at some time.

It is evident that we are faced with a problem of interpretation. Thus, in order to reach a conclusion it is essential to bear in mind the rules of article 33 of the Vienna Convention on Law of Treaties of 1969 – which, in the end, refers to those set out in articles 31 and 32 –, and in particular the authentic texts of the UNCLOS because all of them attest.⁴⁶ Thus the French text refers to “Détroits servant à la navigation internationale”, while the Spanish is similar to the English and uses the expression “Estrechos utilizados para la navegación internacional”. Taking the current meaning of the terms “used”, “servant” and “utilizados” as the first criteria of interpretation in accordance with the stipulations in article 31.1 of the Convention of 1969, in our opinion, it is clear that the verbal tense which the legislator wished to use is the ‘present continuous’, and in consequence that the use is not potential but real.

This conclusion is strengthened by recourse to another objective component of interpretation which is the very text of the treaty as component of the context (article 31.2 Convention of 1969). In this regard, within the Section dedicated to passage in transit, article 37 of the English text of the UNCLOS stipulates that, “this section applies to straits which are used for international navigation”.⁴⁷ The use of the present tense is more than evident and, consequently, it is also evident that there is necessary that the use of the strait would be current, which corroborates our previous thesis.

As regards the terms “which are” that, as we have seen, only appear in the English text, S.N. Nandan and Sh. Rosenne⁴⁸ hold that these introduce a temporary aspect to the functional criteria, in the sense that only those straits which are being used for international navigation at the time the Convention comes into force will be governed by passage in transit. We do not share this conclusion. On the contrary, we understand that the meaning and effect which must be granted to these terms is purely descriptive and ‘atemporal’; which is also confirmed by the versions in other authentic languages.⁴⁹ The legal regime of Part III will be applied to the international straits which, at any time, comply with the three components we have stated above. The disappearance of any of these at any time would exclude a strait from the category analysed here, in the same way that if a purely geographical strait in the future complies with the legal and functional components, it would fall within the category of international strait and, therefore, Part III of the UNCLOS would apply to it.

In addition, we consider that in this clarifying work must also retain the preparatory work of the Convention, as a complementary interpretation resource which confirms our previous conclusion (article 32 of the 1969 Convention). In this regard, several proposals were presented at the III Conference in order to include the term “customarily used”.⁵⁰ As we saw above, Canada proposed a definition of a strait which included that it “has traditionally been used”.⁵¹ Chile was also in favour of the ‘traditional use’.⁵² However, none of these proposals was finally accepted. In contrast, when referring to the right of passage through archipelagic waterways, the article 53.4 of the UNCLOS uses expressions such as “normal passage routes”, or “normal navigational channels”. The inclusion of these references in Part IV and the omission of ‘normally’ in Part III⁵³ means that it is not necessary that this be a normal channel, but that any type of navigation use can be relevant in order to classify a strait as international.

In fact, no evidence of use may be excluded a priori as the exceptional or non-traditional use. Both new and remote use can be taken into consideration. The use may be civil or military, or both. The use does not have to be regular nor reach a predetermined level,⁵⁴ in such a way that, subsidiary straits which are not essential for international navigation may be included in Part III. It is true that, when the authors refer to ‘international navigation’, they are normally referring merchant navigation, but we must also consider military navigation. In fact, the important changes introduced by the UNCLOS into the regime of international straits regarding the transit of the passage mainly benefit surface warships, military aircraft and submarines.⁵⁵

Thus, we agree more with the opinion of H. Caminos, who states that “as long as some degree of use actually exists, any strait connecting two parts of the high seas or an exclusive economic zone with another, or the high sea or exclusive economic zone with the territorial sea of a foreign State, can be classified as ‘a strait used for international navigation’”.⁵⁶

Evidently, as we pointed out, at any time it may occur that a strait included in Part III ceases to be used, or cannot be used (because it becomes non-navigable) for international navigation. In such a case, this maritime area will not comply with the functional component, thus it would cease to belong to the category of ‘straits used for international navigation’, and, consequently, the provisions of Part III would not apply to it, however those of Part II regarding the territorial sea and the right of innocent passage with the possibility of suspension would.

Moreover, the supposition may be put forward inversely. It is possible that a strait is not used for international navigation or is not navigable, but it does have the geographical and legal components, and becomes navigable and starts being used for international navigation. On this point, one inevitably must think of the technology applied to navigation. The technological progress can lead to better, more powerful ships with more sophisticated technology which enables them to navigate through straits where navigation was impossible before, and the functional requirement is not complied with. Logically, if this occurs, this strait would become an ‘international strait’ as from this point and would fall within the scope of application of Part III of the UNCLOS (a position which, as we stated above, is counter to what is maintained by H.B. Robertson, as we have seen above).

This hypothesis has arisen recently as regards the aforementioned Northwest Passage,⁵⁷ a group of straits which constitute the east-west route between the Atlantic Ocean and the Pacific Ocean through the Canadian Arctic Archipelago, and especially since 2007, because that year the ice around the North Pole has reduced to the minimum and a maritime route lying the Pacific and the Atlantic Ocean has been open. Since its waters are frozen over throughout almost the whole year, this passage is not usually used for international merchant navy navigation, except for a short period of time (2 or 3 months a year in summer) when it is navigable although normally with the aid of icebreakers. In this regard, as pointed out by D. Pharand, there are five routes in the Northwest Passage.⁵⁸ Only two of these are really feasible for the navigation of deep draught vessels in arctic class 7 or more, which require a minimum of 20 m depth; and what we will call Route 1 is better than Route 2 because the climatic conditions are less severe.

Specifically, Route 1 runs along the following seaways: Lancaster Sound (between Bylot and Devon Islands, with a width of 51 miles), Barrow Strait (between Prince Leopold and Devon Islands, with a width of 15.5 miles), Viscount Melville Sound, (between Victoria and Melville Islands, with a width of 66 miles), Prince of Wales Strait (between Banks and Victoria Islands, with a width of 6.5 miles), and the Gulf of Amundsen (separates the mainland from Banks and Victoria Islands, with a minimum width of 56 miles). As we have pointed out, this is the most feasible route for the icebreakers and was chosen by the Dome Petroleum Company as the main route. Route 2 is an alternative to Route 1, it follows the same direction as far as Viscount Melville Sound, but then continues through the McClure Strait, which separates Banks Island from Prince Patrick and Melville Islands, and has a minimum width of 43 miles; this Strait is wider that the Prince of Wales Strait in Route 1, but, as we pointed out, it is blocked by ice throughout the year, which makes it impossible to navigate.⁵⁹

The proof of the operability of Route 1 as regards the technical plan was provided in 1969, when the American oil tanker ‘Manhattan’, weighing 151,000 tons and horse power of 43,000, crossed the Northwest Passage for the first time, accompanied by an ice breaker.⁶⁰ More recently, the ‘MSFram’ started the September 24, 2009 a cruise from Oslo to New York, crossing Greenland and the fiords of Canada.

This has led to serious discrepancies as regards the classification of the Northwest Passage in general, and the series of straits which make it up in particular, as ‘international strait’.⁶¹ Today the majority doctrine excludes the whole of the Northwest Passage from the classification as an international strait because it cannot be used for international navigation. A doctrine which is preached and fostered by the coastal State of these straits, Canada, due to evident strategic and defensive interests, which are vital interests for the States which are omnipresent in the classification of international straits.

However, this situation could change in the not too distant future, and make this debate a sterile one if the current global heating of the planet and the resulting melting of the arctic zones, currently the layer of ice is 35% thinner than three decades ago. In this sense, in summer of 2007 the Northwest was fully open to navigation and the ice layer has reduced to 4.24 millions of square kilometres, The Intergovernmental Panel of Climate Change of United Nations upholds that in the 2040s the Arctic would be without ice in the summers.⁶² If this happens, the Northwest Passage would become a navigable maritime channel, and would undoubtedly be much used for international navigation taking into account the fact that it considerably reduces the distance and, therefore, the time required to reach the Atlantic from the Pacific: A voyage from Tokyo to London involves 15,000 miles by Cape Horn, 12,600 miles by the Panama Canal, 8,600 miles by the Suez Canal (which only admits small ships), and 7,900 miles by the Northwest Passage.⁶³

There would be no doubts about compliance with the functional component which is the subject of discussion today. The response to the question put forward by D. Pharand, in the sense that, if international navigation takes place “would this result in the internationalisation of the passage?” and, more specifically, “will the number of foreign flags and foreign transits be sufficient to put the Northwest Passage in the category of straits being used for international navigation?”⁶⁴ Our response is categorical: Yes, and we believe that time will prove us right.

As concerns this last case that a non-navigable strait may become navigable and be used for international navigation, a final consideration may be put forward as regards the fact that this circumstance occurs thanks to the navigation aids or the technical work carried out by the coastal State of the strait. This question would require differentiating the type of aid. Thus, these aids which do not have to be facilitated by the coastal State, but may be provided by the State which uses the strait while it navigates it; this is the case of the icebreakers which cross the arctic straits. Evidently, in this case, the fact that the aid given in order to navigate the strait does not have to be provided by the coastal State means that the functional component is not affected, nor is the international nature of the strait called into question.

The situation arises when it is a question of technical aid or work which can only be carried out by the coastal State. As occurs with the two lighthouses placed by Australia in the eastern area of the Bass Strait (on Deal Island and on King Island) in order to make navigation feasible on this route, which is quite complicated and difficult. The coastal State does not have any obligation to carry out this aid, if it does not do so the strait will not be navigable, it will not comply with its functional component and, consequently, it will not fall within the category of ‘international strait’ nor will the provisions of Part III be applicable to it. However, if the coastal State facilitates this aid or work so that the strait is navigable and it begins to be used for international navigation, from our point of view, from this time the coastal State of the strait is obliged to maintain this aid or work due to the application of the legal institution of ‘estoppel’; that is to say, it has been ‘prevented’, it is bound by its own acts, although the coastal State refuses to see it as such. We can say that the functional component is complied with and that the strait in question passes to the category of ‘strait used for international navigation’.

3.5 Finalising a Concept of ‘International Strait’

From the above, it is possible to conclude a final concept of what must be understood by ‘strait used for international navigation’, which is to say, an ‘international strait’. This would be, “a natural maritime passage which entails a contraction of the waters no greater than double the width of the territorial sea of the respective coastal States, which separates two land masses, and communicates a high seas or EEZ area with another high seas or EEZ area, or a high seas or EEZ area with the territorial sea of another State or, possibly, with its interior waters or its archipelago waters, and is used for international navigation”.

The subsequent question is inevitable. Considering the hundreds of straits in the geographic sense which exist on the planet and which in fact appears in geophysical studies,⁶⁵ which of them are subject to the rules of Part III of the UNCLOS? Or, in other words, which are the international straits in the world?

The response is far from easy and is extremely complex, especially for a jurist as other factors of an exclusively geographic and geo-morphological nature are added to the purely legal factors, such as the corresponding widths, the pertinent coordinates, or the navigation channels, which are beyond the mind of the lay person as regards this matter. Besides these points there are differences and contradictions between papers on the crucial question of the minimum width in each of the straits. For example, this occurs with the Guadeloupe Passage, which, according to the Kennedy Report,⁶⁶ is 20 miles wide, while according to Lay, Churchill and Nordquist,⁶⁷ the width is 28 miles. The same happens with the Strait of Kara, to which Lay, Churchill and Nordquist⁶⁸ attribute a width of 19 miles, while it appears in The Times Atlas ⁶⁹ with a width of 29 miles, and W.E. Butler⁷⁰ reduces this to 13.5 miles at its narrowest channel. Such a variation may make a determined strait oscillate between the balance of the legal and the merely geographical depending on the source of reference.

Another equally conflictive aspect refers to the effect the existence of islands may have on the establishment of the width of a strait, a point we have already referred to. When determining the minimum width of a specific strait, at times neither geographers nor jurists take into account the existence of islands in the strait belonging to one or both coastal States, and only take the respective mainland coasts of the States of the strait as points of reference. However, as we pointed out, on occasions there are islands throughout the State, islands which have their own territorial sea, therefore they must be taken into consideration and, consequently, this would considerably reduce the minimum distance between both sides of the strait, especially if the sovereign coastal State decides to attribute 12 mile territorial waters to these.

Among other examples, this situation arises in the Strait of Florida, between the United States and Cuba in the Caribbean Sea. However, in this case, although the width of 82 miles commonly facilitated⁷¹ does not take the aforementioned islands into consideration, if these were taken into account, this would not entail a change as regards the legal regime, bearing in mind that that the location of these continues to give a width greater than 24 miles, that is to say, a strait not included in Part III.

However, in other cases, the existence of islands in the strait with its own territorial sea reduces the width of the strait to such an extent that straits initially not included in Part III due to exceeding the width of 24 miles would become international straits.

Among others, this is the case of the Barrow Strait in the Canadian artic archipelago, which has a width of 27 miles according to studies made.⁷² However, along the length of the strait there are several islands (Griffith, Garrett, Lowther, Young, Hamilton), which divide the strait into several channels, none of which exceed 16 miles in width – in fact, the minimum width is 8 miles and the maximum 15.5 miles. Thus, if we take these islands into consideration and grant them a territorial sea of 12 miles, the legal effect is that all the waters of the Strait of Barrow would become territorial seas of Canada and, consequently, the strait would be included in Part III of the UNCLOS.⁷³

The same occurs with the aforementioned Bass Strait, in Australia. The width usually attributed to this strait is approximately 100 miles,⁷⁴ in consequence it would be out from Part III. However, this figure obviously does not take into account the chain of islands which are abundant in the eastern area of the strait, from Tasmania to Victoria, which has its own territorial sea, and converts the waters of the eastern entry of the strait into a territorial sea and, therefore, into an ‘international’ strait. It should also be mentioned that this is the case of the Torres Strait, which we have also referred to. Or the Bering Strait, between Russia and the United States, which is usually excluded from the list of international straits as it is considered to exceed 24 miles in width, without taking into account the two islands in the centre of the channel, Big Diomede which belongs to Russia, and Little Diomede which is under U.S. sovereignty, which are separated by scarcely 2 miles of sea, and these divide the straits into two channels, whose widths are no greater than 19 miles; thus, the eastern channel is a territorial sea of the U.S.A., while the western channel is a territorial sea of Russia.

All these obstacles make it difficult to achieve an exhaustive catalogue of international straits, bearing in mind the non-existence of an ‘official’ inventory of such straits.

It is true that a large number of legal works on the straits consulted refer to a map drawn by the Office of the Geographer of the Department of State of the United States of America,⁷⁵ which gives a figure of 116 straits which would be affected by the extension of territorial waters to 12 miles. However, while we do not wish to question the precision and importance of this work, we must bear in mind the words of the ICJ in relation to the value of cartography “as evidence of an auxiliary or confirmatory kind”, and “they cannot be given the character of a rebuttable or ‘iuris tantum’ presumption”.⁷⁶ Following the pronouncement made by the Court, only those maps incorporated into treaties, which are the expression of the will of the States will have legal force. Moreover, the map of the United States Department of State does not take into account other data which as we have seen can exclude the inclusion of a strait in Part III, such as the category of waters which connect or the functional requirement; or the possible existence of an EEZ or high seas corridor, when the coastal States do not extend their territorial waters to 12 miles. Therefore, the value of this map is merely referential.

Furthermore, the number varies considerably in other works consulted. Thus, L.M. Alexander⁷⁷ gives a figure of 265 straits used internationally for navigation. Smith⁷⁸ considers 220 straits to be international. While the list provided by Lay, Churchill and Nordquist⁷⁹ places the figure at 136. Larson⁸⁰ considers that there are 134 international straits. The French National Commission on Toponimy cites less than a hundred in its “Nomenclature of Maritime Areas”.⁸¹ However, Brüel⁸² mentions 35 important straits. While, the report prepared for the First Conference on the Law of the Sea by the frigate captain R.H. Kennedy⁸³ describes 33 straits used as navigation routes and whose width does not exceed 26 nautical miles; a number which is increased to 34 by R. Lapidoth.⁸⁴

In the light of this oscillation of figures, we fully share the opinion of O’Connell in the sense that:

The statistics are a little misleading, because many of these straits are dubiously ‘international’ in the sense of being habitually used by shipping as required by the principles laid down by the International Court of Justice in the Corfu Channel case, and indeed find themselves in the list only because of assiduous cartographical scrutiny on the part of those who seek to produce arresting figures.⁸⁵

Therefore, it is not surprising, as Beckett points out, “that the works of jurists do not contain comprehensive lists of international straits, since they are concerned to expound principles and not to provide catalogues”.⁸⁶

This does not mean that we don’t wish to accept the challenge involved in identifying the ‘international straits’, but simply point out the limits which this task entails and, therefore, the fact that it is impossible to have a complete identification. Thus, the inclusion of the ‘straits used for international navigation’ in each of the categories included in Part III of the Convention,⁸⁷ which we will deal with in the following chapter, is not intended to be an exhaustive list.