Paper presented at the Seminar on current maritime issues and the work of the International Maritime Organization. Twenty-Third Annual Seminar of the Center for Ocean Law and Policy, University of Virginia School of Law, IMO, January 6-9, 2000.

Introduction

It should not be forgotten that an intense treaty making activity was in progress at IMO well before the Third United Nations Conference on the Law of the Sea (UNCLOS III) started its deliberations in 1973; by the end of these deliberations in 1982 most of the main IMO treaties had been adopted and some of them were uncontestedly considered as "generally accepted". This paper highlights some aspects of the evolution of the correlation IMO-Law of the Sea; it also includes some views on how the relationship between IMO conventions and the 1982 United Nations Convention on the Law of the Sea (hereinafter, "UNCLOS" or "the Convention") could be analyzed in terms of treaty law.

The subject of IMO's interface with the 1982 United Nations Convention on the Law of the Sea (hereinafter "UNCLOS" or "the Convention" ) has been comprehensively addressed in two documents produced by the IMO Secretariat after the adoption and entry into force of this Convention. The second document, "Implications of the entry into force of the United Nations Convention on the Law of the Sea for the International Maritime Organization (LEG/MISC/2, dated 6 October 1997, hereinafter, "the IMO Study") should be considered as a necessary reference for the better understanding of this paper.

I Historical overview

Four main periods define the evolution of the interrelation between the elaboration of IMO safety and antipollution regulations and developments in the field of the law of the sea. From 1959 to 1973, conventions and protocols were adopted at IMO without the existence of a comprehensive law of the sea treaty framework. Between 1973 and 1982 this framework was elaborated by UNCLOS III, in parallel to the adoption of the most important IMO treaties. From its adoption until its entry into force (1982 to 1994) UNCLOS served as an importance reference to the always prolific on-going regulatory work undertaken by IMO. The last period, running from 1994 to present, features the dynamic interaction established between UNCLOS as a treaty in force on the one hand and IMO treaties on the other.

1959-1973

The IMO Study highlights the importance UNCLOS as a comprehensive jurisdictional framework defining the features and extent of State jurisdiction for the implementation of IMO regulations. It is accordingly appropriate to start our research by recalling what happened when this framework did not exist, namely at the time when IMO (then "IMCO" - Intergovernmental Maritime Consultative Organization-) initiated its activities in 1959. The adoption of safety regulations primarily enforceable by the flag State was not be particularly affected by the failure of UNCLOS I to provide a single comprehensive jurisdictional treaty to guide the work of organizations such as IMCO. Accordingly, the adoption of the International Convention on Load Lines, 1966 (LL) did not present major jurisdictional problems as the first expression of an IMO treaty setting global safety rules and standards to be applied on board ships.

Legal uncertainty stroke upon IMO when in the aftermath of the Torrey Canyon incident the Organization faced the challenge of implementing its antipollution mandate by addressing questions involving coastal State jurisdiction. The 1958 Convention on the Territorial Sea and Contiguous Zone regulated the status of both sea zones but not their extension and a second conference convened in 1960 equally failed in this regard. Customary law did not provide a solution because traditional rules which had been widely accepted until the end of the second world war were being continuously contested by increasing unilateral State action leading to de facto extension coastal state jurisdiction over the sea and sea bed.

This situation had important repercussions at the 1969 Diplomatic Conference which adopted the International Convention relating to the Intervention on the High Seas in Cases of Oil Pollution Casualties, 1969 (the "Intervention Convention"). In the view of some countries the convention should apply to all seas, without making distinctions between the territorial and the high seas since these areas "had different meanings in different parts of the world...In the view of many countries...the territorial sea only extended for 3 miles; but in the view of other countries it extended to 12,100 or even 200 miles from the shore. If, , therefore, the Convention mentioned only the high seas, there would be a doubt as to whether large areas, extending up to 200 miles from the coasts, were covered..."1 Other countries felt that to apply the convention to territorial sea " would mean limiting the sovereign rights of coastal States. For example, such States would have to enter into consultation with other States and actions would have to be decided not in local courts, but through consultation or arbitration. That would be a dangerous precedent..."2 The alternative of an IMO diplomatic conference regulating the rights and extension of the powers of the coastal State would be unthinkable today due to the existence of UNCLOS as a the sole treaty which defines jurisdictional rights and obligations of the coastal State. In fact, the Intervention Convention has in the opinion of many been made redundant by the provisions of UNCLOS regulating the extension and regime of the different sea zones and the right of intervention to avoid pollution in article 221.

While the role of regulating coastal State jurisdiction at a global level (as it is the case of the Intervention Convention) is no more part of IMO's mandate, the so called approval role, namely, the adoption of routeing provisions at the request of coastal States along particular sea zones adjacent to their coast has been continuously expanded after receiving its initial legitimation in the Convention on the International Regulations for Preventing Collisions at Sea, 1972 (COL REG) in connection with the adoption of traffic separation schemes (TSS).

A reference to the 1959-73 period cannot exclude mention of the International Convention on Civil Liability for Oil Pollution, 1969 (CLC) and the International Convention on the Establishment of an International Fund for Compensation for Oil Pollution Damage, 1971 (The FUND Convention). For the purposes of including such mention within a law of the sea context suffice to point out at the importance of extending the mandate of a UN agency to regulate the liability of private persons in view of the diversity of interests affected by oil pollution damage: the emergence of the concept of damage to the marine environment as a distinctive subject matter regulated by a liability convention and the establishment of compulsory insurance arrangements to cover for such damage is a milestone in the history of the evolution of private maritime law towards a regulatory context where private and public law loose their traditional distinctive features This evolution is reflected now in article 235 of UNCLOS. Damage was limited to the territorial sea in a moment where its extension had not been established in international law, thus leaving states parties with a great discretionary power to determine the territorial scope of application. This uncertainty has now been solved by the delimitation of the territorial sea, the establishment and delimitation of the EEZ and the explicit extension of the concept of damage to the EEZ in the 1992 Protocols to the CLC and the FUND Convention.

1973-1982

The period 1973 -1982 during which the Third UN Conference on the Law of the Sea was deliberating is the most prolific in the history of IMO. The most important IMO treaties (SOLAS, MARPOL, STCW) were adopted while, at the same time, the Third Conference was finally developing a comprehensive convention which would provide an appropriate jurisdictional framework for the enforcement of IMO treaties.

The interaction between treaty making activities at the Third Conference and those at IMO is reflected not only in continuous consultations to avoid overlapping but also in temporary clauses included in IMO treaties indicating that these treaties should not be interpreted as prejudicing the codification and development of the law of the sea being undertaken by UNCLOS III. Another proviso establishes the basis for the avoidance of jurisdictional conflicts between the law of the sea and IMO rules and standards. It stipulates that nothing in these treaties shall prejudice present or future claims and legal views of any State concerning the law of the sea and the nature and extent of coastal and flag state jurisdiction. No reference is made to port state jurisdiction, strictly speaking not a law of the sea concept in itself but a corrective device which was being continuously developed by IMO to remedy non-compliance with obligations initially imposed upon the flag State. Nevertheless, the notion of port State jurisdiction was introduced in that veritable environmental mini-convention contained in Part XII of UNCLOS.

1982-1994

After its adoption UNCLOS became a reference for IMO's work in the sense that the elaboration of IMO regulations were bound to bear in mind the need to conform with the "Constitution of the oceans". But it was a Constitution not in force and it was not clear whether it would ever be. Accordingly, attitudes towards the new treaty at IMO fora were at the beginning rather sceptic.

This situation changed not only as a result of the progressive recognition of UNCLOS but also through debates within IMO related to the ever problematic issue of the approval role of the Organization, not only in connection with TSS but also with routeing measures in general. This was a subject which throughout the history of the Organization had continuously been the source of discussions focusing on the extent and features of IMO's role in the adoption of routeing measures depending on the sea zone where these measures should apply. On top of this, the adoption of routeing measures became related not only related to safety of navigation but also to environmental considerations and this feature made necessary a closer analysis of IMO's environmental mandate in the light of Part XII of UNCLOS. It is important to highlight some of the steps which signalled the progressive recognition of the role of UNCLOS at IMO:

Resolution A.572(14) on General Provisions on Ships Routeing adopted by the IMO Assembly on 20 November 1985 3 states that these provisions shall not prejudice the provisions of UNCLOS (3.16). The resolution includes novel regulations which establish distinctions between TSS and other routeing systems lying within or beyond the territorial sea 4.

During 1992 and 1993 the Legal Committee and an ad-hoc informal working group considered legal issues regarding the adoption of mandatory ship reporting to Vessel Traffic Services (VTS) bearing in mind the framework established by UNCLOS 5.

Resolution A.720 on Guidelines for the designation of special areas and the identification of particular sensitive sea areas adopted by the IMO Assembly on 6 November 1991 6 contains an explicit recognition of UNCLOS (with the exclusion of part XI) as "widely accepted as customary international law (1.3.7)".

1994-1997

The last important element of the story was added when UNCLOS moved from its customary law status to that of a treaty in force. Entry into force of UNCLOS meant that IMO instruments, rather than simply taking into account UNCLOS, had to conform with its regulations. Hence the detail in which the question of compatibility between UNCLOS and IMO conventions is addressed in the 1997 IMO Study. A rather striking example of how things had changed is shown by the introduction of explicit references to UNCLOS as source of obligations for States Party to SOLAS in connection with the adoption of ships' routeing and ship reporting systems. As a consequence of amendments to Chapter V of this convention, regulations V (8)(j) and (8-1)(i) indicate respectively that all adopted ships' routeing systems and ship reporting systems "shall be consistent with international law, including the relevant provisions of the 1982 United Nations Convention on the Law of the Sea". These provisions have no parallel in any other IMO treaty instrument.

II UNCLOS jurisdictional framework for the development of IMO treaties

The preceding historical overview explains the dynamic interrelation developed throughout forty years between the international law of the sea and the international law of safety of navigation and prevention of marine pollution formulated by IMO. Main features governing this interrelation are the inclusion of clear references to IMO's safety and antipollution standards and the strengthening of several important features related to questions of jurisdiction and enforcement. For instances:

The global mandate of IMO is implicitly but firmly acknowledged in cases where the expression "competent international organization" is used in singular, in connection with the adoption of international shipping rules and standards in matters concerning safety, efficiency of navigation and the prevention and control of marine pollution from vessels and by dumping.

Paramount for the implementation of IMO regulations is the requirement contained in article 94 of UNCLOS that every State "shall effectively exercise its jurisdiction and control in administrative, technical and social matters over ships flying its flag" and the comprehensive set of references included in the same article to the duty of the flag State to implement regulations which as it is explained in the IMO Study are recognized without dissent as being IMO shipping rules and standards.

UNCLOS has solved legal uncertainty in setting the jurisdictional features of IMO's approval role related to the exercise of coastal State jurisdiction in the case of traffic separation schemes, designation of sea lanes ship reporting and vessel traffic services and, in general, adoption of routeing measures. Of particular importance in this regard is the different language used by UNCLOS to define the obligations of coastal States depending on the sea area where the se measures apply. For instances, in the case of sea lanes and traffic separation schemes applicable to the territorial sea, states must "take into account" IMO recommendations. Instead, in the case of straits used for international navigation, the language becomes mandatory: the designation of sea lanes and traffic separation schemes "shall conform to generally accepted international regulations" (article 41 (3)).

The notion of port State jurisdiction was initially developed by IMO as a corrective remedy, namely, as a limited set of procedures which port States can implement in order to correct deficiencies in the exercise of flag State jurisdiction resulting in non-compliance with safety and antipollution regulations by foreign ships voluntarily in port. Being primarily focused on the need to achieve a balance between flag and coastal State jurisdiction, UNCLOS does not include general regulations on port State jurisdiction but it does so in Part XII. Due to this peculiarity, the relationship between UNCLOS and IMO provisions changes in connection with the subject of the protection of the marine environment.

The degree to which UNCLOS obliges parties to it to implement IMO regulations is expressed in a language which varies depending on subject matter: states parties must "take account of", "conform to", "give effect to" or "implement" "generally accepted international regulations" "applicable international instruments", "generally accepted international regulations, procedures and practices", etc. Some of these terms invite a wide interpretation of how the "umbrella" character of UNCLOS should operate as a tool to effectively bind Parties to apply IMO regulations. Discussions on this subject frequently focus on two main issues. The first one relates to the degree to which states parties to UNCLOS are bound to implement generally accepted IMO regulations even if they are not parties to the IMO treaties in which these regulations are contained. Closely related to this question are the different views on how the criteria of general acceptance (as applicable to IMO regulations) should be construed.

III Binding degree of UNCLOS references to IMO regulations

The IMO Study provides some useful tools to analyze this question under the tittle "Legal status of IMO treaties in accordance with international law and the law of the sea". In general, the Study suggests a case by case analysis bearing in mind the language used in UNCLOS and the particular IMO rules and standards under consideration. As we have seen in the case of routeing measures, it is not the same to say that Parties to UNCLOS should "take into account" and to prescribe the obligation to "conform to" or "implement", " generally accepted" rules and standards.

In principle, it seems beyond discussion that in many cases UNCLOS contain general obligations to apply rules and standards contained in IMO treaties. But after asserting this principle, distinctions must be made: UNCLOS ' language is general and as such of a restricted operative character. In this regard there seem to be consensus in support of an assertion of paramount importance contained in the IMO Study: "UNCLOS is acknowledged to be an "umbrella convention" because most of its provisions, being of a general kind, can be implemented only through specific operative regulations in other international treaties" . This assertion implies that IMO rules and rules and standards are very precise technical provisions which cannot be considered as binding among States unless they are parties to the treaties where they are contained. UNCLOS provisions certainly aim at the effective implementation of substantive safety and antipollution rules, but in the end they remain basically jurisdictional provisions, namely, provisions which regulate the features and extent of state jurisdiction but not the enforcement of measures regulated in other treaties. Compliance with IMO rules and standards cannot be dissociated from the treaty framework in which these rules and standards are contained. Thus UNCLOS obligations to enforce IMO rules and standards should be understood as operative on condition that parties to UNCLOS also become parties to the IMO conventions where these rules and standards are contained. SOLAS, MARPOL or STCW rules and standards can only be properly implemented if flag and port States multilaterally bind each other in accordance with the terms of these treaties.

It follows that UNCLOS obligations to apply IMO rules and standards should be interpreted in accordance with the carefully drafted UNCLOS provisions on compatibility with other treaties in general (article 311) and in connection with environmental treaties (article 237). It is through the operation of these rules on compatibility that UNCLOS recognizes the importance of other treaties as a source of international obligations.

The interpretation contained in the preceding paragraphs qualifies the view according to which parties to UNCLOS are obliged to implement generally accepted IMO rules and standards irrespective of whether they are or not party to the treaty where these rules and standards are contained. Such interpretation could result in discouraging states from becoming Parties to the main IMO Conventions and decide instead to enact IMO provisions as national legislation in order to enforce them without complying with the corresponding obligations: it could be argued that its is irrelevant for States parties to UNCLOS to become parties to basic IMO treaties since, in any case, UNCLOS would be considered the basic legal source of obligations to comply with all generally accepted IMO rules and standards. If this interpretation were to be implemented many states would seek enforce IMO rules and standards in respect of foreign vessels as national legislation without complying with their obligations as port States: there would be simply no treaty law relationship obliging them to do so. In the environmental field this risk is noticeable in the cases of states which enact MARPOL provisions in order to enforce them in respect of foreign vessels, without providing the corresponding reception facilities prescribed by MARPOL.

The unqualified view that parties to UNCLOS should implement IMO regulations even if they are not parties to the IMO treaties containing these rules could also result in the introduction of confusion in connection with the meaning of "general acceptance". For instances:

in the case of amendments frequently introduced to SOLAS and MARPOL in accordance with the system of tacit acceptance , parties these treaties have the right not to accept the amendments and preserve this right even if the new provisions come to be "generally accepted". On the contrary, parties to UNCLOS which are not parties to SOLAS or MARPOL would be obliged to implement these new provisions. The assessment of their obligations in accordance with UNCLOS would depend on rather uncertain views on general acceptance. On the contrary, parties to SOLAS and MARPOL would rely on a solid treaty law situation authorizing them not to accept amendments.

The preceding example might be largely theoretical at present in the light of the fact that so far, amendments to IMO conventions hardly ever meet formal declarations of non acceptance. But what happens with MARPOL ? This Convention entitles parties to opt out from its annexes III, IV and V and many States have made use of the opt-out clause. It follows that even if provisions in any of these annexes became "generally accepted" parties to MARPOL which made used of the opting out clause would still be entitled not to apply them. On the contrary, parties to UNCLOS which are not parties to MARPOL would not be entitled to opt out and would accordingly be obliged to apply rules contained in those annexes.

In the case of the two preceding examples, it can be argued that once regulations become generally accepted even parties to the corresponding IMO convention which have preserved the right not to apply these regulations would, in accordance with UNCLOS, have the duty to finally accept them. An assessment of such situation would imply considerations no how to balance on the one hand the character of UNCLOS as a convention imposing upon parties the need to implement measures which have become generally accepted as a result of an increasing political will at an international level, and on the other the preservation of basic treaty law principles of compatibility contained in articles 311 and 237.

If all these comments are taken into account, what is in the end the real extent of UNCLOS obligations to apply IMO rules and standards? By way of conclusion, this question can be answered by recognizing that it is true that being these provisions fundamentally shipping provisions, they prescribe a commitment for parties to UNCLOS to comply with those standards as flag States. However, basic principles of treaty law indicate that once these shipping provisions become part of a treaty they cannot be considered in a unilateral way. They must be correlated to the obligations imposed upon port and coastal States. Otherwise the treaty law structure of UNCLOS and the main IMO conventions would break and legal uncertainty would follow.

Finally, and once more, any interpretation should primarily bear in mind the different nature of the general "umbrella" provisions of UNCLOS and that of the extremely precise IMO regulations: it is true that UNCLOS involves clear obligations for states not to allow the navigation of substandard ships, but it would be simply too daring to assert, for instances, that a State party to UNCLOS and not party to SOLAS is obliged to ensure that all ships flying its flag locate a collision bulkhead at a distance from the forward perpendicular of not less than 5% of the length of the ship. And it cannot be established beyond doubt that UNCLOS is being violated by a State Party to it if ships flying its flag do not situate a stem gland in a watertight shaft tunnel or other watertight space separate from the stem tube compartment.

Further distinctions should be made in connection with the implementation of Part XII. For instances, it can be asserted that, even in the high seas, a ship Party to UNCLOS violates the Convention if it does not comply with discharge requirements under MARPOL, because discharges not allowed by MARPOL conform the definition of pollution contained in UNCLOS. But it cannot be asserted straightforward that lack of MARPOL equipment on board a ship which is not polluting, means per se a violation to UNCLOS.

IV What is "general acceptance"?

The need to consider IMO rules and standards as intrinsically associated to the treaty in which they are contained is also relevant to provide a consistent legal interpretation to the requirement of their "general acceptance". It is precisely in connection with this requirement that the question of the ambiguity of UNCLOS expressions has been frequently raised. This question will be analyzed here by way of the example provided by the discussions on the meaning of general acceptance of IMO rules and standards held at the working session of the Committee on coastal State jurisdiction relating to marine pollution at the sixty-seventh Conference of the International Law Association (1996) 7.

Chapter 3 of the report presented for the consideration of the Committee contains an analysis of the meaning of expressions such as generally accepted international rules and standards. The report starts with a warning: "Indeed, at the outset of UNCLOS III a gentleman's agreement was adopted imposing consensus the general rule to adopt provisions, aimed at arriving at a package deal acceptable to all. This resulted first of all in the vague and ambiguous terminology sometimes encountered in this document, which has most certainly not to be attributed to poor draftmanship. Secondly, as a result of this, some of the provisions just referred to have to be understood as an agreement between participants to further disagree, a sort of lowest common denominator acceptable to all the parties involved" 8

It is precisely bearing in mind these views that some assertions in the same report should be carefully qualified. This is the case of conclusions expressed in the same document that a central element to determine the generally accepted character of a specific rule and standard appears to be the practice of states, followed by the suggestion that the legal instrument containing the respective rule and standard is only of secondary importance 9. This last suggestion seems to be wrong if we follow the line of reasoning of the warning quoted above: it is precisely on account of the vague and ambiguous terminology and the political background of the preparatory works at UNCLOS III that the legal features of instruments where IMO rules and standards are contained becomes a matter of paramount importance. UNCLOS provides the basic legal framework for general obligations to comply with IMO rules and standards. However, as an umbrella convention reflecting minimum consensus on a "lowest common denominator", UNCLOS needs to be complemented with further treaty law structure provided by the IMO conventions and protocols in order to ensure that obligations to implement the extremely precise technical rules and standards contained in such conventions and protocols become binding in accordance with international law.

Once the importance of the treaty law features of IMO rules and standards has been properly acknowledged, it becomes necessary to add further precision to the concept of state practice as a criteria to establish when these rules and standards can be considered as generally accepted. References to a merely de facto notion of state practice, while valuable from a statistical, pragmatic point of view is of limited relevance for the purpose of a legal analysis of treaty law obligations such as those established in IMO instruments. The acknowledgment of the explicit consensual element which is at the core of any treaty law obligation leads to the necessity of analysing not only what states do but what they ought to do in accordance with written law. When merely of a general, customary, non written character, international law can rely merely on general practice as the sole source of obligations. Once international law is written into a detailed text full of technical specifications, the concept of general acceptance must be necessarily be construed bearing in mind the formal expression of consent given by states to be bound by the treaties in which those technical specifications are contained. This is particularly important in the case of IMO rules and standards which, by virtue of their own nature can only be expressed in terms of written law. Hence the criteria sustained in the IMO Study according to which the most important single element to consider in connection with the concept of general acceptance is the degree of acceptance of a treaty expressed in the number of formal ratifications or accessions.

In the case of the main IMO shipping conventions the requirement of general acceptance is ensured by combining two requirements, namely the number of states parties to them and the condition that contracting states represent at least 50% of the world tonnage. The most important IMO treaties have not only been ratified by a high number of countries but also reflect the formal consent given by these states to ensure that between 90 and 99 per cent of the world merchant fleet complies with IMO rules and standards.

V The environmental UNCLOS and IMO rules and standards

It has been suggested in section II that due to the peculiar features of UNCLOS Part XII the relationship between UNCLOS and IMO provisions changes in connection with the subject of the protection of the marine environment. This paper should not come to an end without referring to this important question.

The UNCLOS feature as "umbrella convention" is greatly altered in Part XII, which includes provisions which are per se of an operative kind, which can be directly implemented and as such should be read together with other operative provisions contained in IMO treaties, notably MARPOL. See for instances the provisions on investigations of foreign vessels contained in article 226 and compare them with those of MARPOL, article 5. Both articles indicate how certificates should be inspected, measures to be taken when things are not in order, etc.

This apparent overlapping of provisions can be solved with an interpretation of both UNCLOS Part XII and MARPOL which helps to avoid inconsistencies by pointing out at the similarities and differences of both bearing in mind their ultimate purpose. Both treaties aim at the protection of the marine environment by means of ensuring that antipollution preventative measures are properly implemented. However, UNCLOS focuses more on measures to be taken to prevent and penalize discharges in ocean spaces while in the case of MARPOL violations are not only related to illegal discharges but also to the non-compliance of preventative measures to be applied on board irrespective of whether discharges take or not place. The distinction has important consequences in connection with the application of penalties. It should not be forgotten that in accordance with UNCLOS (article 230) penalties other than monetary ones can be imposed only in case of a wilful and serious act of pollution in the territorial sea. It is obvious that the expression "act of pollution" should be interpreted bearing in mind the concept of "pollution of the marine environment" established in article 1(4) of UNCLOS. In other words, there must be an act of wilful misconduct in the territorial sea, resulting in the introduction into the marine environment of a polluting substances to authorize the imposition of a prison sentence. Violations to MARPOL rules resulting in substandard navigation without both wilful misconduct and polluting discharges can only be sanctioned with monetary penalties.