Immunities and Compromissory Clauses: Making Sense of Enrica Lexie (Part I)

Written by and

Small(ish) disputes can make for significant holdings. From Nottebohm to Lotus to AAPL v Sri Lanka, the list of relatively limited incidents prompting far-reaching judicial and arbitral pronouncements is long. We may now have to add Enrica Lexie to it. The PCA’s award of 21 May 2020, released in full earlier this month, originated in a stand-off, on 15 February 2012, between an Italian oil tanker and an Indian fishing vessel — one that had dramatic, indeed, fatal consequences for two Indian fishermen, but that, relatively speaking, was brief and discrete. And yet it resulted in proceedings spanning five years, with claims, counter-claims, and procedural objections, which have now culminated in a massive award of no less 309 pages (not even counting the individual opinions).

Introduction

Unlike Nottebohm or Lotus, Enrica Lexie cannot be reduced to that one, timeless pronouncement that will enter the legal canon. Rather, it stands out for the breadth and diversity of matters on which the Tribunal pronounced: how to find the true nature of a dispute? what does it mean for flag states to have ‘exclusive  jurisdiction’ over their ships? what is the status of marines deployed on merchant ships, and do they benefit from immunity? what is considered a maritime incident at sea? what is the content of due regard obligations, etc.? These are just some of the questions addressed by the Tribunal, and from the mere listing, it is clear that the award speaks to special aspects of the law of the sea as well as cross-cutting issues of general international law. 

Our discussion in the following focuses on one of these cross-cutting issues. We look at one aspect of the Tribunal’s handling of the immunity issue — could Indian courts exercise criminal jurisdiction over the two Italian marines that had fired the shots? More specifically, we focus on the jurisdictional side of this question: Could the Tribunal, established under UNCLOS Annex VII, entertain the question of immunity — a matter that UNCLOS does not explicitly regulate? The Tribunal felt it could; but this outcome, which was reached over the dissents of Judge Robinson and Dr Rao, was not an obvious one.

Both the outcome and the reasoning towards it are significant. They reflect the Tribunal’s understanding of its own jurisdiction under UNCLOS, and of the proper construction of jurisdictional clauses. This is a cross-cutting issue, which arises regularly when proceedings are brought on the basis of a compromissory clause. As Enzo Cannizzaro and Beatrice Bonafé noted fifteen years ago (in an EJIL article that bears (re-)reading ), ‘compromissory clauses in a treaty have a compartmentalizing effect. They tend to draw a dividing line between the category of disputes which fall within their scope … from those which fall outside their scope’. And they do so for a reason: Compromissory clauses are ‘gatekeepers’, after all (in Matina Papadaki’s helpful phrase) setting out what disputes States want to take to international courts or tribunals, and the law that the latter may use in deciding them. But ‘gatekeeping’ is easier said than done, as many disputes ‘straddle the dividing line’ (Canizzaro and Bonafé). Responsibility is an obvious example: It has long been accepted that disputes about responsibility come within a court or tribunal’s jurisdiction, whether or not the compromissory clause mentions responsibility expressly: ‘reparation [as the PCIJ noted in Chorzow Factory] is the indispensable complement of a failure to apply a convention, and there is no necessity for this to be stated in the convention itself’. In the recent ICAO Council cases, the International Court of Justice (ICJ or Court) also seemed to accept that, in assessing violations of Air Services agreements, the ICAO Council was entitled to look at other issues arising in the context of general defences (counter-measures) invoked by the parties.  

Does the same logic apply to immunities — or do States have to ‘state it in the convention itself’ that they may be judicially or arbitrally determined? This was one of the most significant questions addressed in Enrica Lexie. In this blogpost, we discuss the Tribunal’s approach, and we compare it to the approach of another court, the ICJ, which, coincidentally, has addressed very similar questions in two recent proceedings, in Certain Iranian Assets and Immunity and Criminal Proceedings. Our post will not solve the riddle of compromissory clauses and incidental questions. But by comparing these three important decisions, we hope to highlight the significance of the issue, which unlike many other questions of dispute settlement is in need of fuller analysis.

Incidental Jurisdiction over Questions of Immunity in Enrica Lexie

So how did the issue arise in the Enrica Lexie case? In the relatively standard manner that immunities are often invoked. Following the incident of 15 February 2012, when two Italian marines fired shots at the fishing vessel, St. Antony, India arrested the two marines and charged them before Indian courts. As Italy argued before the Tribunal, this not only violated the principle of exclusive flag State jurisdiction, but also the functional immunity that the marines, as state officials, enjoyed. India – again, not uncommonly, for disputes of this kind – contested the claim to immunity. But, crucially, it argued first that the questions of immunity were at the heart of the case, and that they were outside the Tribunal’s jurisdiction, which depended on the construction of Article 288 UNCLOS. According to Article 288, UNCLOS dispute settlement bodies have ‘jurisdiction over any dispute concerning the interpretation or application of this Convention’. Proceeding from Article 288, the Tribunal broached the immunity issue in two steps:

First, it had to satisfy itself that the dispute as a whole was one concerning the interpretation and application of the UNCLOS (and not one primarily about immunity, as India had suggested in its pleadings). It dealt with that first aspect relatively (and perhaps too) quickly, by identifying ‘the real issue[s] in the case and […] identify[ing] the object of the claim’ (para 231). In its view, the two States disputed which one of them could exercise jurisdiction over the Enrica Lexie/St. Anthony incident, raising ‘the alleged violations of various provisions of the Convention’ (para 235). The marines’ immunity was only ‘one out of several bases on which Italy’ alleged India’s exercise of jurisdiction to be unlawful (para 238). In the words of the Tribunal, the dispute ‘may raise, but is not limited to, the question of immunity’ (para 243). So, this was not a situation where one of the parties was masquerading a different ‘real cause’ into a dispute about UNCLOS – and, according to the Tribunal, this was sufficient for the purposes of Article 288. It is relevant in this respect, that the Tribunal (correctly; see Methymaki/Tzanakopoulos) viewed Italy’s immunity claim as a potentially relevant ‘exception’ to India’s exercise of criminal jurisdiction: as argued by Italy, it was possible that the dispute could be dealt with without any discussion of the immunity issue (paras 238-239).

But such an easy way out was not available. The Tribunal found that under the territoriality principle and as provided in Article 92 UNCLOS Italy and India had concurrent jurisdiction over the incident (para 367) — so the question of immunity did matter. But could the Tribunal pronounce upon it? This, in view of the Tribunal, required a fuller analysis than the one offered in step 1 – not one looking at the character of the dispute as a whole, but at Italy’s express claim that India had violated the marines’ immunity. Could this claim and the application of the law on immunities be said to form part of the Tribunal’s exercise of jurisdiction over ‘[a] dispute concerning the interpretation or application of this Convention’ in the sense of Article 288 UNCLOS? According to India it could not, as immunity rules were ‘in general a different body of law’ from the Convention (para 789), viz. general international law.

Italy relied on a trusted strategy to bring the immunity issue within the realm of UNCLOS (and the Tribunal’s jurisdiction): it claimed that various UNCLOS provisions, referring to the exercise of rights in accordance with ‘other rules of international law’ (or similar) in effect incorporated immunity by renvoi. This strategy failed as the Tribunal found the provisions invoked by Italy inapplicable. The Tribunal was equally unconvinced that Article 297 mattered – which in Italy’s stretched construction was said to ‘exten[d] the jurisdiction of a court or tribunal … beyond Article 288(1)’ (para 779).

Nonetheless the Tribunal found its own way towards pronouncing on the immunity question: it came up with a ‘necessity argument’, partly borrowed from Italy’s first strategy. In its view, the marines’ potential claim to immunity was impossible to ignore; quite apart from any textual references to ‘other rules of international law’, immunity ‘necessarily arises as an incidental question in the application of the Convention’ (para 809). More specifically, the Tribunal asked whether ‘the issue of the entitlement to exercise jurisdiction over the incident’, which divided Italy and India, could ‘be satisfactorily answered without addressing the question of the immunity of the Marines’ (para 805)? Thus put, the question had to be answered in the negative. As an exception to otherwise established jurisdiction, immunity (if available) would bar India’s exercise of jurisdiction. Whether the exception applied in this case, held the Tribunal, formed ‘an integral part of the Arbitral Tribunal’s task to determine which Party may exercise jurisdiction over the Marines’ (para 808). To provide ‘a complete answer’ to the question the parties brought before it then, the Tribunal had to consider incidentally whether the Marines enjoyed immunity (ibid).

Immunity, while mentioned in neither Article 288 nor in the text of UNCLOS, had crept within the scope of the Tribunal’s jurisdiction as a necessary incidental question. And so the Tribunal could proceed and hold that India had actually violated the immunity of the marines; a topic for another blogpost.

Print Friendly, PDF & Email

Categories

  • EJIL Analysis
  • Jurisdiction
  • Law of the Sea
  • Permanent Court of Arbitration

    • Tags

  • immunity
  • Jurisdiction
  • Law of the sea
  • UNCLOS

    • Leave a Comment

    Comments for this post are closed

    Comments

    Cameron Miles says

    August 27, 2020

    Without wishing to prejudice what Eleni and Christian have to say, this strikes me as a potentially dangerous development - or at least one that has the capacity to radically expand the scope of incidental jurisdiction before international courts and tribunals.

    Previously, incidental questions were confined (as the PCIJ in the Certain German Interests case seems to have understood) to matters which were logically anterior to the jurisdiction ratione materiae of the court or tribunal in question. So, in Certain German Interests, the Court held it was "hardly possible" to exercise its jurisdiction under the Geneva Convention in respect of which it was seized without first interpreting and applying Art 256 of the Treaty of Versailles.

    Put another way, the Court's jurisdiction would have been rendered a practical nullity had it not answered the incidental question, such that it was necessary to answer it if its jurisdiction was to have any meaning.

    It is difficult to see how this threshold is met in Enrica Lexie with respect to the parallel questions of immunity under UNCLOS (over which the Tribunal had jurisdiction) and immunity under general international law (over which it did not). The Tribunal could have answered the former easily without answering the latter, as the latter operates as an exception to the former: it is quite possible to determine the scope of a positive rule without also determining the scope of an exception, as they are logically separate.

    This gives rise to two further points. First, the Tribunal appears to have apprehended its role as being to resolve the entire dispute between the parties – in the sense that it felt the need to resolve every (international) legal question arising out of the factual matrix before it. This much is clear from its insistence on providing a “complete answer” (para. 808) to the immunities question; and it is what gives rise to the expansion apprehended at the start of this comment. There is a difference between answering a question logically antecedent (or preliminary) to a question clearly within the jurisdiction ratione materiae of an international court and tribunal and answering a question necessary to resolve the wider dispute between the parties. The latter is much broader than the former, and strains the consent-based system of international dispute settlement in which a court or tribunal may be given permission to resolve some legal questions, but not others.

    Of course, this gives rise to the question of how what the Tribunal did was any different to the application of a circumstance excusing wrongfulness under the law of state responsibility. The difference there is that circumstances excusing wrongfulness are general secondary rules qualifying the application of primary rules so as preclude liability. It is a different thing entirely where the ‘defence’ in question is another primary rule that may be used to establish a separate form of liability in its own right.

    Second, it is strange that the Tribunal felt the need to draw on the general proposition of incidental jurisdiction to address this point when it could have relied instead on UNCLOS Art 293(1) and its reference to “other rules of international law not incompatible with this Convention” – which has previously been used (inter alia) to allow Annex VII tribunals to resolve questions concerning the use of force (see Guyana v Suriname). Whilst I have my own reservations about the use of Art 293(1) to expand the jurisdiction ratione materiae of UNCLOS Part XV bodies, at least it doesn’t upend the way we thought compromissory clauses worked in the wider law of international dispute settlement.

    I very much look forward to reading the rest of Eleni and Christian’s posts on this matter.

    Christian J. Tams says

    August 28, 2020

    Thanks, Cameron, your comment raises crucial issues. While Eleni and I have tried to provide some context in Part II, proper responses will require more time & space — this is for articles/ monographs, not blogposts; and I doubt we‘ll solve the issues in the comments section. But let me at least flag 3 points in response:

    - The risk of overstretching compromissory clauses is real. While courts & tribunals are likely aware of this (or are reminded of it by one of the parties), they do not necessarily accord it the same weight, and seem to weigh it differently against the equally relevant consideration that a dispute must be meaningfully resolved. In reaching the preferred outcome, the framing of a dispute seems the crucial strategy, as we outline in Part II of the post.

    - As regards the relevance of Enrica Lexie,
    did ICaTs before really ‘confine‘ their exercise of jurisdiction over matter not expressly conferred to them? You mention ‘logically anterior’ questions. But the list is quite long: it also includes ‘reparation/ responsibility‘ (rarely mentioned in clauses, but routinely addressed for a near-century). ‘succession’ (Croatia-Serbia), ‘treaty exit‘ (ICAO I), ‘general defences‘ (ICAO II), modalities for bringing claims (ELSI - local remedies, not ‘tacitly dispensed with‘ in the words of the Chamber). Can all this be rationalised on the basis of a primary-secondary line? In the EJIL piece we cite in the post, Enzo Cannizzaro and Beatrice Bonafé conclude that the ICJ had “adopted quite a broad interpretation of the notion of ‘disputes on the interpretation and application of the treaty’“.

    - Art 293: I agree this is a curious omission (but have no inside knowledge as to why it did not play a more prominent role). Your reference to it points to a matter that we did not address, viz. the relationship between scope of jurisdiction (288) and law to be applied (293). The Tribunal proceeded via the jurisdictional route; but the line is fine, perhaps even finer than that between primary and secondary rules...

    Thanks again!

    Aditya Roy says

    August 30, 2020

    Dear Authors,
    Thanks for an insightful post and for bringing the contrasting approaches of ICJ AND ITLOS to our notice. However, My first argument is that some of the treaty as part of preamble have this clause that "Affirming that matters not regulated by this Convention continue to be
    governed by the rules and principles of general international law,". This clause is also part of the Preamble to UNCLOS. I was wondering why did the TRIBUNAL did not address the immunity question taking the recourse to preamble because general international law = Customary International law. Therefore, the question of immunity could have been addressed as part of Customary International Law taking recourse to preamble because preample can be taken recourse of during interpretation as per Article 31 (2) as part of context.
    My second argument is that Don't you think, the TRIBUNAL already what should be the outcome and then tried to justify it by bringing the immunity claim as part of the incidental question because they could not have been unaware about their powers deriving from the compromissory clause.
    My third argument is how do you see the relationship between the principle of Non ultra petita and jura novit curia (the court knows the applicable law).
    Thank you
    Best Wishes
    Aditya Roy