Might contain traces of Lotus: The limits of exclusive flag state jurisdiction in the Norstar and the Enrica Lexie cases

Abstract

The article scrutinizes some of the surprising commonalities in the reasonings of two recent decisions by two separate judicial forums: the ITLOS’s judgment in the M/V Norstar case and the award of an ad hoc arbitral tribunal in the Enrica Lexie case. One key connection between the two decisions is their heavy reliance on the Lotus judgment of the PCIJ. Another similarity between the two disputes is that both of them revolve around the concept of exclusive flag state jurisdiction under UNCLOS Article 92(1) and adjacent questions of jurisdiction on the high seas. The article is going to subject both decisions to criticism and argue that some of the more problematic positions adopted by the tribunals in both cases amount to no more than obiter dicta – thus establishing an additional parallel with Lotus, which also received heavy criticism for its controversial obiter dictum. The two tribunals’ new-found interest in Lotus also provides an opportunity to discuss the utility and legal weight of Lotus as a precedent in the face of a century of developments in treaty law and judicial practice. In this sense, this article builds on and attempts to continue the recent trend in scholarship advocating for a renewed appreciation of the Lotus case against the backdrop of decades of criticism against it. Accordingly, the article aims to facilitate a better understanding of all three disputes, the principles they applied, and the dynamics of international adjudication and international law in general.

1. Introduction

Following a century of relative neglect from the International Court of Justice (ICJ), the S.S. Lotus (France v. Turkey) ¹ case was endorsed by two prominent international tribunals in two high-stakes disputes within a year’s time: in the International Tribunal for the Law of the Sea’s (ITLOS) 2019 judgment in the M/V Norstar (Panama v. Italy) case ² and in an ad hoc arbitral tribunal’s award in the Enrica Lexie Incident (Italy v. India) case. ³ While these two decisions serve as the immediate backdrop of the present article, this article aims to be more than a combined case note of some sort: instead, my goal is to contextualize the legal issues raised in these cases and juxtapose them with the looming spectre of the Lotus judgment.

While both Norstar and Enrica Lexie are law of the sea disputes, they have wildly different factual backgrounds. Norstar concerned ‘bunkering’ activities committed on the high seas and Enrica Lexie is about a shooting incident. Hence, pairing the two cases might seem arbitrary at first. But, on a closer look, there are many similarities between them: the applicants prevailed in both cases; Italy participated in both cases (winning one and losing the other), and two of the five arbitrators in Enrica Lexie had – at different times – also been judges at the ITLOS, so there is a partial overlap in the identity of the adjudicators as well. Furthermore, the Enrica Lexie tribunal cited and endorsed Norstar’s holding on the scope of exclusive flag state jurisdiction. Finally, and most importantly, both decisions heavily build on the century-old Lotus judgment of the Permanent Court of International Justice (PCIJ).

Contrasting Norstar and Enrica Lexie with Lotus, one notices further commonalities among them: all three cases revolved around a ship (which ended up becoming the cases’ namesake – admittedly, not a rare occurrence in law of the sea disputes); in all three, a narrow majority rendered the decisions, sparking vociferous dissents; and all three disputes involved a developed nation and a developing nation as the two parties.

There is further correspondence concerning the subject matter of the three disputes. One recurring theme across all three cases is the three distinct types of jurisdiction (enforcement, adjudicative, and perspective jurisdiction). Lotus can be seen as the ‘cradle’ of this trichotomy of jurisdictions in international law and the two new cases take these themes in interesting new – and sometimes controversial – directions. Commentators have already praised and criticized both Norstar and Enrica Lexie for some of their holdings. ⁴ Thus, the primary aim of the present article is not to rehash these commentaries, but rather to deploy a different vantage point and analyse the tribunals’ reasoning and adjudicatory techniques from a more pragmatic perspective. I will show that some of the more controversial findings of the two decisions were not integral to the tribunals’ final decisions; as such, they are obiter dicta. This provides another strange parallel to Lotus, which also received more criticism for its famous obiter than for its ratio decidendi.

Recent years have seen a shift in scholars’ general attitude towards Lotus. After having been shunned for decades, the judgment now seems to be going through somewhat of a renaissance: one author, in particular, called for ‘Letting Lotus Bloom’, prompting readers to reconsider some of their old reflexes about the case. ⁵

In light of this, the aim of the present article is twofold: first, to look at the Lotus case with fresh eyes, through the lens of the legal issues presented in the two more recent disputes, and test Lotus’s strength against a century of legal development. Second, the article scrutinizes Norstar and Enrica Lexie through Lotus’s lens, and weighs Lotus’s utility in tackling contentious issues in a present-day legal landscape and the track record of present-day judicial bodies in avoiding some of the pitfalls often associated with the PCIJ’s Lotus judgment. All of this should help in gaining an enhanced understanding of the three disputes, the legal principles they engaged, and the dynamics of international adjudication and international law in general.

2. Lotus

Few international judgments have attracted as much attention – and criticism – as Lotus has. ⁶

Despite the over-abundance of academic commentary on the judgment – or possibly because of it – nothing is straightforward about this case. First of all, there are at least two distinct dimensions to the judgment: on the one hand, there is the PCIJ’s pronouncement (in an obiter dictum) of the ever-infamous Lotus principle, which allegedly gives states a carte blanche to do as they please unless a certain conduct is explicitly prohibited by international law. On the other hand, there is Lotus’s more narrow holding on the exercise of concurrent adjudicative jurisdiction on the high seas, which, to this day, is the cornerstone of any discussion on the ‘international law of jurisdiction’. ⁷ Enrica Lexie and Norstar engage – mostly – with this second dimension, thus, it makes sense for the article to do the same. However, delimiting, so to say, the two dimensions is not without its challenges.

For this reason, the article will follow the unorthodox method of first grappling with the many layers of reflection the judgment has gathered over almost a century – i.e., Lotus’s ‘legacy’ – and only then will it turn to the analysis of the judgment itself. This might seem counterintuitive at first, but it should assist in having a more focused and streamlined understanding of the judgment and its key holdings.

2.1 Lotus’ legacy

The Lotus case is unique in international jurisprudence for at least four reasons.

First, Lotus is often cited as the only major – but certainly the most well-known – instance when the PCIJ’s (or ICJ’s) core holding in a dispute was subsequently overturned by treaty law. ⁸

As the story is reproduced in legal textbooks ⁹ and both commentaries to the United Nations Convention on the Law of the Sea (UNCLOS), ¹⁰ Lotus’s holding on the legality of the exercise of concurrent jurisdiction by France and Turkey ‘produced alarm among seafarers’, and a long campaign ensued against the decision to avoid multiple prosecutions for the same offence. ¹¹

This led to the codification of rules providing for the exclusive jurisdiction of the flag state over criminal matters in cases of ‘collision or any other incident of navigation’. The rules were codified first in the 1952 Brussels Convention, then later in the 1958 High Seas Convention, and took their final shape as codified by UNCLOS Article 97(1), providing for the exclusive jurisdiction of the flag state for penal or disciplinary proceedings ‘[i]n the event of a collision or any other incident of navigation concerning a ship on the high seas’.

While this provision (and its precursors) are sometimes described as the ‘overruling’ ¹² of a key holding, this article proposes that Article 97(1) is better understood as a carve-out rule. Accordingly, in incidents involving vessels or high seas that are not collisions or incidents of navigation, Lotus is still good law. ¹³ What is more – as confirmed by the Enrica Lexie award – Lotus is the applicable law for these issues. ¹⁴

Second, the Lotus case, despite its fame, is barely ever referenced in the ICJ’s decisions. Handeyside, in a recent article, identified only three occasions when the ICJ referenced the Lotus case in a merits decision, and even in these decisions, the Court refused to meaningfully engage with the Lotus ruling. ¹⁵ Commentators identified further cases, where the spectre of Lotus seemed to be lingering in the background – the Corfu Channel case, ¹⁶ the Kosovo opinion, ¹⁷ and the Arrest Warrant ¹⁸ case are such examples – but no explicit references were made to the judgment itself. It is almost as if Lotus was considered as taboo or, at the very least, an overly sensitive topic by the Court. (Against this background, the comparative relevance of the Norstar and Enrica Lexie tribunals’ engagement with Lotus is even more significant.)

Such aversion on the ICJ’s part is not so surprising, considering that some judges of the Court heavily criticized Lotus in their separate and dissenting opinions: the Separate Opinion of Judges Higgins, Kooijmans, and Buergenthal to the Arrest Warrant case characterized Lotus’s position on the law of jurisdiction as ‘the high water mark of laissez-faire in international relations, and an era that has been significantly overtaken by other tendencies’. ¹⁹ Higgins criticized Lotus elsewhere too, noting that it should be looked at ‘with great caution’. ²⁰

The third reason why Lotus is special is that discourse on the judgment has been largely diverted from what the case was actually about. Granted, the PCIJ was, at least partially, responsible for this diversion through the formulation of an obiter dictum, whose controversy grew so large that it overshadows the narrower legal point of the case in its ratio decidendi. ²¹ According to the infamous adage – the so-called Lotus principle – ‘[r]estrictions upon the independence of States cannot therefore be presumed’. ²² It is unnecessary to engage here either with the holding itself or the criticisms levelled against it. I will limit myself to making a small observation on how the PCIJ’s holding grew to be – at least partially – misconstrued.

The root of the problem was probably Judge Loder’s dissenting opinion to the Lotus majority judgment, in which he wrote: ‘under international law everything which is not prohibited is permitted … in other words, on the contention that, under international law, every door is open unless it is closed by treaty or by established custom’. ²³ This recharacterization has become all but synonymous with the original judgment’s holding. ²⁴ There are two problems with this. First, Judge Loder’s rephrasing was not identical in meaning to the original. ²⁵ Second, Loder used this quote to summarize Turkey’s position in the proceedings – although, admittedly, according to Loder, the majority adopted essentially the same position. ²⁶

Thus, it would seem that an obiter dictum, ²⁷ taken out of context and mischaracterized by a dissenting judge, was largely influential in casting a negative light on the judgment in toto. This is all the more astonishing, considering the conflict between the obiter and the decision’s narrower ratio decidendi. Guilfoyle notes that ‘if the Lotus principle means what it literally says, then the PCIJ in Lotus failed to apply it’. ²⁸ This is correct. What is more, the Lotus principle was not only irrelevant to the Court’s final holding, it was at odds with it, as will be shown below.

Finally, no case of the PCIJ or the ICJ – save for the highly problematic South West Africa judgment ²⁹ – has been subject to the amount of scholarly criticism Lotus has. Already in the wake of the judgment, many commentators voiced their concerns about it. ³⁰ Criticism remained the norm for the decades to follow: a representative example is Mann’s 1964 Hague Academy Lecture, where he made the point that ‘nothing was decided in the Lotus case’. ³¹

Thus, it is all the more surprising that recent years have seen a slight shift in scholars’ attitudes towards Lotus. By offering a critique of the criticism, Hertogen defended Lotus from a policy perspective, arguing that what we understand under the ‘Lotus principle’ is based on a mistaken reading of the majority judgment and that it is the prevailing reading of Lotus that should be ‘discarded as inaccurate’ and not the judgment itself. ³²

Others have also followed course, calling for a more nuanced reading of the judgment, considering the Lotus principle ‘in its immediate textual context or framed against the wider circumstances of the case’. ³³ According to another recent article, it is only the ‘conceptual origins’ of Lotus that are problematic and not the ‘core content’ of the case. ³⁴ The same article talks about a ‘scholarly ambiguity about the current standing of the Lotus rule in international law’. ³⁵ Considering the original approach of outright hostility towards the case, ‘ambiguity’ is already a great step forward for Lotus and its legacy.

2.2 Jurisdiction in Lotus

An undeniable privilege one enjoys when contributing to a century of commentary on perhaps the most famous inter-state dispute is that one can remain cursory in lining out the facts of the case. Lotus, in the PCIJ’s very own words, was about:

a collision on the high seas between two vessels flying different [French and Turkish] flags, on one of which [the French vessel, the ‘Lotus’] was one of the persons alleged to be guilty of the offence, whilst the victims were on board the other [the Turkish vessel, the ‘Boz-Kourt’]. ³⁶

Following the incident, Turkish authorities arrested the Lotus’s first officer, Lieutenant Demons, and a Turkish court sentenced him to 80 days of imprisonment (and a fine of £20). ³⁷

The question before the Court was whether Turkey had breached international law by instituting criminal proceedings against Lieutenant Demons. The Court found in favour of Turkey. The Court’s decision was supported by the narrowest possible margin of majority: six judges for and six judges against, with President Huber’s casting vote. ³⁸

A brief summary of the PCIJ’s reasoning is reproduced below. I will only quote segments of the award that will be directly relevant in the context of the article’s discussion on Norstar and Enrica Lexie as well.

The Court first spelled out the general rule of prohibition on the extra-territorial exercise of jurisdiction. Notably, this paragraph followed directly after the Court’s pronouncement of the Lotus principle and is in stark contrast to it (or, alternatively, it is a carve-out of massive proportions):

Now the first and foremost restriction imposed by international law upon a State is that—failing the existence of a permissive rule to the contrary—it may not exercise its power in any form in the territory of another State. In this sense jurisdiction is certainly territorial; it cannot be exercised by a State outside its territory except by virtue of a permissive rule derived from international custom or from a convention. ³⁹

The PCIJ then introduced another carve-out to the above rule (i.e., an exception to the exception):

It does not, however, follow that international law prohibits a State from exercising jurisdiction in its own territory, in respect of any case which relates to acts which have taken place abroad, and in which it cannot rely on some permissive rule of international law … Far from laying down a general prohibition to the effect that States may not extend the application of their laws and the jurisdiction of their courts to persons, property and acts outside their territory, it leaves them in this respect a wide measure of discretion which is only limited in certain cases by prohibitive rules; as regards other cases, every State remains free to adopt the principles which it regards as best and most suitable. ⁴⁰

The distinction between these two paragraphs is among Lotus’s most valuable contributions to international law. What the Court did here is it introduced a distinction between extraterritorial enforcement jurisdiction on the one hand (the exercise of which, as a general rule, is prohibited), and extraterritorial prescriptive (legislative) and adjudicative jurisdiction on the other (the exercise of which, as a general rule, is permitted). ⁴¹ In case of the latter categories (prescriptive and adjudicative jurisdiction), the judgment makes this pronouncement explicit, while in case of the first category (enforcement jurisdiction), the reference is implicit (and also follows from the process of elimination, i.e., if there is a special rule governing prescriptive and adjudicative jurisdiction, the only type of jurisdiction ‘remaining’ is enforcement). This trichotomy still provides, to this day, the basic framework of how we think of jurisdiction in international law. ⁴²

As the facts of the Lotus case concerned the exercise of Turkey’s adjudicative jurisdiction, the case could have been decided (with the same result) based on the above holding alone. Granted, the rest of the reasoning still added some welcome nuance. A couple of pages later, the Court applied the above general considerations to the specific – law of the sea – context of the case with the expected results. It is useful to quote the relevant section in full, as it played an important role in Norstar and Enrica Lexie as well:

It is certainly true that—apart from certain special cases which are defined by international law—vessels on the high seas are subject to no authority except that of the State whose flag they fly. In virtue of the principle of the freedom of the seas, that is to say, the absence of any territorial sovereignty upon the high seas, no State may exercise any kind of jurisdiction over foreign vessels upon them. Thus, if a war vessel, happening to be at the spot where a collision occurs between a vessel flying its flag and a foreign vessel, were to send on board the latter an officer to make investigations or to take evidence, such an act would undoubtedly be contrary to international law.

But it by no means follows that a State can never in its own territory exercise jurisdiction over acts which have occurred on board a foreign ship on the high seas …

This conclusion could only be overcome if it were shown that there was a rule of customary international law which, going further than the principle stated above, established the exclusive jurisdiction of the State whose flag was flown …

In the Court’s opinion, the existence of such a rule has not been conclusively proved. ⁴³

Thus, the Court applied the same dichotomy between enforcement and prescriptive–adjudicative type of jurisdiction in the context of the law of the sea and found no lex specialis rule to necessitate a diversion from its original approach.

Notably, both the Norstar and the Enrica Lexie tribunals relied on the italicized first sentence of the first paragraph in the quote. ⁴⁴ However, they did so without also quoting the second italicized segment of the text, thus breaking the internal logic of the reasoning and distorting its meaning. ⁴⁵ This will be discussed in more detail below.

The final Lotus holding that is relevant for our purposes concerned the exercise of concurrent jurisdiction (in case of incidents on high seas):

The offence for which Lieutenant Demons appears to have been prosecuted was an act of negligence or imprudence having its origin on board the Lotus, whilst its effects made themselves felt on board the Boz-Kourt. These two elements are, legally, entirely inseparable, so much so that their separation renders the offence non-existent. Neither the exclusive jurisdiction of either State, nor the limitations of the jurisdiction of each to the occurrences which took place on the respective ships would appear calculated to satisfy the requirements of justice and effectively to protect the interests of the two States. It is only natural that each should be able to exercise jurisdiction and to do so in respect of the incident as a whole. It is therefore a case of concurrent jurisdiction. ⁴⁶

The finding on concurrent jurisdiction is not particularly controversial in itself. However, the Court, in reaching this conclusion, also held that vessels should be regarded as the ‘floating territory’ of the state. ⁴⁷ This legal fiction has attracted much criticism from the dissenting judges, and has since been deemed to have ‘fallen into disrepute’. ⁴⁸ The Danish Company Tax Liability case offers a helpful explanation as to why this fiction is ‘misleading’. Just one reason is that vessels would have their own territorial sea around them if state territory was interpreted verbatim. ⁴⁹ This is understandably problematic.

This, however, does not mean that Lotus’s holding, in this context, is erroneous. Rather, it seems to be the case that it is permissible to treat vessels as the floating territory of the state for the (narrow) purposes of exercising adjudicative jurisdiction as established in Lotus, ⁵⁰ but not otherwise.

2.3 Jurisdiction after Lotus

The Lotus case concerned the narrow issue of states’ right to exercise their adjudicative (criminal) jurisdiction in a law of the sea context. In spite of this, the judgment has contributed greatly to the discipline of jurisdiction in international law in general. Lotus is the locus classicus for rules on concurrent jurisdiction, the objective territoriality principle, the ‘effects’ doctrine, and collision on the high seas (in this last regard, with the caveat of UNCLOS’s carve-out on collisions and incidents of navigation). ⁵¹

The trichotomy that was never spelled out, only implied in Lotus, has come to govern our understanding of international law rules on exercising jurisdiction. ⁵² The Third Restatement provides a concise taxonomy of the three types of jurisdiction:

Prescriptive jurisdiction: “to prescribe, i.e., to make its law applicable to the activities, relations, or status of persons, or the interests of persons in things, whether by legislation, by executive act, or order, by administrative rule or recognition, or by determination by a court”.

Adjudicative jurisdiction: “to adjudicate, i.e., to subject persons or things to the process of its courts or administrative tribunals, whether in civil or in criminal proceedings, whether or not the state is a party to the proceedings”.

Enforcement jurisdiction: “to enforce or compel compliance or to punish noncompliance with its laws or regulations, whether through the courts or by use of executive, administrative, police, or other nonjudicial action”. ⁵³

Most authorities follow this trichotomy. ⁵⁴ There are good reasons for this, as ‘[e]ach form of state jurisdiction is distinct. They are subject to different tests and differ in scope …’. ⁵⁵ Some scholars, however, only recognize the two major categories of ‘legislative/prescriptive’ and ‘enforcement’ jurisdiction, making only passing mention of adjudicative jurisdiction. ⁵⁶

This article proposes that it is advisable to follow the three-pronged approach of the Restatement, as adjudicative jurisdiction possesses sufficiently distinct characteristics. Thus, treating it as analogous to either enforcement or prescriptive jurisdiction could prove to be problematic. The Norstar dispute is a prime example why.

3. Norstar

3.1 Summary

The M/V ‘Norstar’ was an oil tanker owned by a Norwegian company, chartered to a Maltese company, flying the flag of Panama. From this flurry of jurisdictions, it is the Panamanian one that is going to be relevant for our purposes, as it was Panama, the ‘Norstar’s’ flag state, that initiated proceedings against Italy before the ITLOS. The dispute was about a Decree of Seizure issued by the Italian authorities that eventually led to the arrest and seizure of the ship by Spanish authorities in Spanish territorial waters. ⁵⁷

From 1994 to 1998, the ‘Norstar’ engaged in supplying gas oil to mega yachts in international waters ‘off the coasts of France, Italy and Spain’. ⁵⁸ The sale of gas oil to vessels at sea is known as ‘bunkering’ or ‘offshore bunkering’. ⁵⁹ In 1997, the Italian fiscal police initiated an investigation against the ‘Norstar’ and its affiliated broker company registered in Italy. The investigation revealed that the ‘Norstar’s’ activities were in breach of the applicable tax laws:

the M/V Norstar was involved in the business of selling the fuel purchased in Italy in exemption of tax duties to a clientele of Italian and other EU leisure boats in the international waters off the coasts of the Italian city of Sanremo. ⁶⁰

Italian authorities initiated criminal proceedings against a number of individuals involved, including the ‘Norstar’s’ captain. ⁶¹

On 11 August 1998, the Public Prosecutor at the Court of Savona, Italy, issued a Decree of Seizure against the ‘Norstar’, proclaiming that the vessel ‘as well as the oil product transported therein must be acquired as corpus delicti’. ⁶² The same day, the Italian prosecutor’s office also sent a request for judicial assistance to Spanish authorities to enforce the Decree of Seizure. The request was based on the European Convention on Mutual Assistance in Criminal Matters (European Convention). On 25 September 1998, Spanish authorities in Palma de Mallorca seized the M/V ‘Norstar’. ⁶³

In 2003, the competent Italian court acquitted the persons that had been charged and also revoked the ‘Norstar’s’ seizure, ordering that the ship be released to its owner. ⁶⁴ According to Panama, the owner was never made aware of this. ⁶⁵ Finally, the ‘Norstar’ reached a sad fate when a waste management company bought it at a public auction in 2015 to convert it into scrap. ⁶⁶

The case marked the first preliminary objections judgment in the ITLOS’ history. ⁶⁷ The tribunal delivered the judgment on 4 November 2016, ruling that it had jurisdiction to adjudicate the dispute and Panama’s claims were admissible. Although the dispositif did not make this clear, based on the judgment’s reasoning, the tribunal in effect only upheld its jurisdiction to Panama’s claims under UNCLOS Articles 87 and 300, and found that the rest of the provisions that Panama had relied on ‘[could not] be invoked in the present case’. ⁶⁸ The tribunal’s decision on upholding its jurisdiction enjoyed a 21–1 majority – with Judge Treves, the ad hoc judge appointed by Italy, dissenting – while the decision on admissibility was adopted with a 20–2 ratio. ⁶⁹

One controversial aspect of the preliminary objections judgment was the tribunal’s decision that Italy (and not Spain) was the proper respondent. Italy argued that the tribunal had no jurisdiction ratione personae as Spain was the proper respondent. Italy’s defence rested on the question: how exactly could the arrest and seizure of the ‘Norstar’ – both carried out by Spanish authorities – be attributed to Italy? The tribunal was unable to offer a satisfying answer to this conundrum, ⁷⁰ and held that ‘[as the Tribunal] found that Italy is the proper respondent, the question whether the conduct of Spain is attributable to Italy is irrelevant for the purpose of determining the proper respondent’. ⁷¹ Other than being slightly tautological, this holding is probably the most baffling of all in both Norstar judgments, and one that had conceptual and legal repercussions for the merits judgment as well.

Admittedly, the tribunal’s analysis is on point about the integral part that Italy played in the proceedings. Nevertheless, this only proves that there would have been no case against Spain as a sole respondent. It does not prove that there was a case against Italy as a sole respondent.

The solution would have been for Panama to sue both Italy and Spain together. There seems to be no indication on the fact record as to why Panama did not involve Spain in the proceedings or whether it had attempted to do so. According to Panama, ‘Spain has not been mentioned, summoned, cited, or even referred to in this case either as defendant or as a third party, nor has it shown any interest in participating through any of the possible methods accepted by the Convention’. ⁷² One can only wonder why Spain had not lined up to be on the receiving end of a claim that was in the hundreds of EUR millions.

The final leg of the tribunal’s analysis was the rejection of Italy’s admissibility objection on the basis that Spain was an indispensable third party in the proceedings. ⁷³ This outcome followed logically from the tribunal’s holding that Italy was the proper respondent. On the whole, the holding is aptly described as ‘a decision that while technically consistent with prior authority, was markedly inconsistent with reality’. ⁷⁴

On the merits, the tribunal also found in Panama’s favour, the victory was, however, a pyrrhic one. In terms of the compensation to be paid, the tribunal found that the Italian court’s 2003 judgment ordering the ‘Norstar’s’ unconditional release interrupted ‘the causal link between the wrongful act of Italy and damage suffered by Panama’. Therefore, Panama was not allowed to claim any damages for the loss of the ‘Norstar’ sustained after this date. In addition to this, the tribunal rejected the rest of Panama’s monetary claims (for the shipowner’s loss of profits, the payment of wages, payment of fees and taxes, damage to the charterer, material and non-material damage for natural persons). Thus, the final amount awarded was US$ 285,000 plus pre-award interest compounded since 1998, amounting altogether to approximately US$ 500,000. This was less than one percent of Panama’s original claim. ⁷⁵

3.2 Freedom of navigation and exclusive flag state jurisdiction in the judgment

The tribunal’s (re-)framing of Panama’s claims in its judgment on jurisdiction created problems for the merits analysis as well. Italy viewed the tribunal’s judgment as ‘curtailing’ the dispute and limiting it to the lawfulness of Italy’s Decree of Seizure. ⁷⁶ According to the tribunal, however, the dispute involved ‘not only the Decree of Seizure and the Request for its execution’ – undertaken by Italy, the respondent – but ‘also the arrest and detention of the Norstar’ – undertaken by Spain. ⁷⁷

Next, the tribunal turned to the facts of the case, and found that ‘the Decree of Seizure and its execution concern both alleged crimes committed in the territory of Italy and bunkering activities conducted by the “Norstar” on the high seas’. ⁷⁸ Naturally, for the purposes of the dispute, only the activities conducted on the high seas were relevant. ⁷⁹

The tribunal then undertook a detailed analysis under UNCLOS Article 87(1), which is, without doubt, the most exciting and controversial part of the merits judgment. Article 87(1) lays down the fundamental rule on states’ right to freedom of navigation. In its reasoning under Article 87(1), the tribunal extensively relied on Article 92(1) as interpretative guidance. Article 92(1) provides for the general rule of exclusive flag state jurisdiction on the high seas:

[s]hips shall sail under the flag of one State only and, save in exceptional cases expressly provided for in international treaties or in this Convention, shall be subject to its exclusive jurisdiction on the high seas …

The most recent – 2017 – commentary to the UNCLOS limits the scope of Article 92(1) to enforcement jurisdiction. According to the commentary, the provision:

creates only a prohibition on exercising enforcement jurisdiction over foreign vessels on the high seas; multiple States may still attach legal consequences to acts committed on a vessel on the high seas as a matter of prescriptive jurisdiction. ⁸⁰

In addition to being ‘canonical’ by virtue of its inclusion in the commentary, this is also the mainstream viewpoint shared by the majority of scholars. ⁸¹ While the UNCLOS commentary helps limit the scope of Article 92(1), it, unfortunately, does not adhere to the three-pronged approach in qualifying the different types of jurisdiction. Accordingly, it does not make an explicit reference to adjudicative jurisdiction. Nonetheless, for the purposes of Article 92(1), adjudicative jurisdiction should logically be grouped with prescriptive jurisdiction, and thus remains unaffected by the provision.

Turning back to the analysis in Norstar, the tribunal’s linking of Article 87(1) to Article 92(1) was not entirely without textual support: Article 87(1) itself provides that ‘[f]reedom of the high seas is exercised under the conditions laid down by this Convention and by other rules of international law’. The tribunal made the point that ‘[f]reedom of navigation would be illusory if a ship – a principal means for the exercise of the freedom of navigation – could be subject to the jurisdiction of other States on the high seas’. ⁸² In support of this statement, the tribunal relied on the holding of the Lotus judgment already quoted above:

It is certainly true that – apart from certain special cases which are defined by international law – vessels on the high seas are subject to no authority except that of the State whose flag they fly. In virtue of the principle of the freedom of the sea, that is to say, the absence of any territorial sovereignty upon the high seas, no State may exercise any kind of jurisdiction over foreign vessels upon them. ⁸³

This is the majority judgment’s only reference to Lotus, however, it is strategically placed. The tribunal followed up the reference with a confirmation that Article 92(1) UNCLOS was the reflection of the above Lotus quote. Hence, Lotus helped provide a crucial link between Articles 87(1) and 92(1). It is somewhat unfortunate that the PCIJ’s own caveat ⁸⁴ to the quote did not make it into the Norstar judgment, as it would have offered valuable guidance for the analysis that followed.

Next, the tribunal confirmed – relying on consistent ITLOS case law – that bunkering activities fell within the scope of freedom of navigation. ⁸⁵

The tribunal then continued its legal analysis of Article 87(1), holding that, in addition to physical or material interference, ‘even acts which do not involve physical interference or enforcement on the high seas may constitute a breach of the freedom of navigation’. ⁸⁶ The tribunal further noted that ‘Italy recognizes the possibility that acts falling short of enforcement action on the high seas could be relevant in terms of a breach of article 87 of the Convention, if such acts produce some “chilling effect”’. ⁸⁷ Thus, it seems that it was common ground between the parties – and the tribunal – that acts that did not amount to physical interference could breach Article 87(1).

In the next paragraph, however, the tribunal discarded Italy’s qualification requiring a ‘chilling effect’ and stated instead that:

Regardless of such effect, any act which subjects activities of a foreign ship on the high seas to the jurisdiction of States other than the flag State constitutes a breach of the freedom of navigation, save in exceptional cases expressly provided for in the Convention or in other international treaties. Thus Italy’s application of its criminal and customs laws to bunkering activities of the M/V ‘Norstar’ on the high seas could in itself, regardless of any chilling effect, constitute a breach of the freedom of navigation under article 87 of the Convention. ⁸⁸

For the analysis of the actual breach, the tribunal relied on Article 92(1) (or, in the tribunal’s formulation, ‘the principle of exclusive flag state jurisdiction’) to hold that the provision:

prohibits not only the exercise of enforcement jurisdiction on the high seas by States other than the flag State but also the extension of their prescriptive jurisdiction to lawful activities conducted by foreign ships on the high seas. ⁸⁹

Quite bafflingly, the tribunal offered no authority or further justification on this point, which the whole case turned upon. Actually, in the 121-page long award, this was the first mention of the term ‘prescriptive jurisdiction’ other than in the summaries of the parties’ positions. The tribunal concluded by stating that:

if a State applies its criminal and customs laws to the high seas and criminalizes activities carried out by foreign ships thereon, it would constitute a breach of article 87 of the Convention, unless justified by the Convention or other international treaties. This would be so, even if the State refrained from enforcing those laws on the high seas. ⁹⁰

One can only wonder why the ITLOS decided to adopt such a controversial position and why it did so in such a cursory manner. Leaving the legal correctness of the decision aside for the moment, the tribunal could certainly have made more of an effort to come up with justifications to support a position that ‘run[s] against the grain of prevailing scholarly opinion’. ⁹¹

Other commentators have made compelling points on why such an expansive reading of the rule on exclusive flag state jurisdiction is problematic from a policy perspective. ⁹² I do not wish to repeat these arguments here. I would instead like to make the point that other than erroneous; the tribunal’s key holding was superfluous.

This is because (i) the issuance of the Decree of Seizure did not constitute an exercise of prescriptive jurisdiction, rather it qualified as an exercise of adjudicative jurisdiction; (ii) accordingly, on the facts of the case, it was unnecessary for the tribunal to attribute such an extensive scope to Article 92(1); and therefore, (iii) the tribunal’s holdings are essentially obiter dicta, at least to the extent that they concern prescriptive jurisdiction.

According to the ITLOS, Italy’s breach of Article 87(1) entailed the following acts: (i) the adoption of the Decree of Seizure; (ii) the request for judicial assistance addressed to Spain; and (iii) the arrest and detention of the ship. ⁹³ The question is: which of these actions qualified as an exercise of Italy’s prescriptive jurisdiction? The arrest and detention of the ship most certainly did not: they are clear examples of the exercise of enforcement jurisdiction (leaving aside, for the moment, the fact that the arrest was carried out by Spain and not Italy). That leaves the issuance of the Decree and the request for judicial assistance.

The Decree was adopted by the Italian Prosecutor’s office in the course of proceedings before a criminal court. ⁹⁴ Italy’s aim with the adoption of the Decree was to ‘subject persons or things to the process of its courts’. ⁹⁵ Thus, the adoption of the Decree was an exercise of Italy’s adjudicative jurisdiction. Italy’s prescriptive jurisdiction was exercised through the adoption of its laws that served as the basis of the adoption of the Decree of Seizure. However, the Norstar judgment did not discuss this regulatory framework as part of the factual background and provisions of Italian law were only referenced in passing. Indeed, the tribunal limited its focus and analysis to the Decree, which, however, is not a legislative act. ⁹⁶ And this is unsurprising in light of the mainstream view, according to which ‘… the mere exercise of prescriptive jurisdiction, without any attempt at enforcement, will not normally have to pass the test of international law’. ⁹⁷

Italy’s request for execution under the European Convention is harder to classify. As the request was issued by an Italian court in the course of the criminal proceedings, it should also belong under the category of adjudicative jurisdiction. However, it could be argued that – as the request entailed the arrest and detention of the ship – it is better characterized as an exercise of enforcement jurisdiction. Either way, the request was not an exercise of prescriptive jurisdiction.

In light of this, there seems to be no good reason why the tribunal chose to attribute an unnecessarily broad reading to a legal standard in a case, where the factual background did not even warrant the application of that legal standard. Even the narrower grounds suggested by this article would still leave room for criticism: it is not self-evident that the issuance of the Decree of Seizure as an exercise of adjudicative jurisdiction impaired the ‘Norstar’s’ freedom of navigation. But at least the judgment would be conceptually and legally more consistent, and its holdings would not have the far-reaching policy implications that are referenced by its critics.

3.3 Freedom of navigation and exclusive flag state jurisdiction in the joint dissenting opinion

Seven judges of the ITLOS penned a joined dissenting opinion, heavily criticizing the award. ⁹⁸ The dissent’s core point was that Article 87(1) did not apply to the case and, even if it had been, Italy had not violated it. Notably – and somewhat surprisingly – the dissenting judges accepted and adopted the majority’s classification of Italy’s acts as an exercise of its prescriptive jurisdiction, and formulated their arguments accordingly.

Where the dissenters did disagree with the majority judgment was the scope of Article 87(1). The dissenting judges attributed a narrower reading to Article 87(1), holding that it protected the free movement of vessels ‘primarily from the exercise of enforcement jurisdiction by non-flag States on the high seas’. ⁹⁹ Notably, just a couple of paragraphs later, the opinion seemed to concede that Article 87(1) ‘may also protect vessels on the high seas from the prescriptive jurisdiction of non-flag States’ under certain conditions (not met in Norstar). ¹⁰⁰

In another two paragraphs, the dissenters returned to their main argument against the majority’s broad interpretation of exclusive flag state jurisdiction. ¹⁰¹ Thus, while the dissenting judges were willing to concede that Articles 87(1) and 92(1) might, in certain cases, be breached through the exercise of prescriptive jurisdiction, they disagreed with the ‘wholesale’ prohibition approach of the majority. To support this point, they relied on a string of citations, referencing Guilfoyle’s contribution to the 2017 UNCLOS commentary, who, in turn, had referenced a French treatise from the 1930s and – as it should be no surprise to the reader by this point – the Lotus case. As stated above, Guilfoyle’s views on prescriptive jurisdiction represent the majority view on this issue. ¹⁰² Also, the position seems to enjoy some post hoc support as well by subsequent commentary and criticism of the majority decision. ¹⁰³

The dissent also picked up on the majority judgment’s perplexing phrasing of its main holding: ‘[exclusive flag state jurisdiction] prohibits … the extension of their prescriptive jurisdiction to lawful activities’. ¹⁰⁴ According to the dissenters, ‘[t]his would seem to suggest that a non-flag state is not excluded from extending, in conformity with international law, its prescriptive jurisdiction to the unlawful activities of foreign vessels or of persons on the high seas’. ¹⁰⁵ Indeed, the majority’s qualification here is unhelpful. ‘Lawful’ – under which set of laws? International law or domestic law? In vacuo, every activity is ‘lawful’ until it becomes subject to regulation. There is also a temporal side to the question: as it were the case in Norstar, the Italian courts ended up finding that the ‘Norstar’s’ activities had been lawful all along. This, however, could not have been foreseen at the time of the issuance of the Decree of Seizure (or at the time of adopting the relevant Italian legislation, for that matter).

The opinion closed with the argument that ‘a State may exercise its prescriptive criminal jurisdiction with respect to conduct on the high seas where such conduct is integral to an alleged crime committed in the State’s territory, not when it is justified or allowed by international law to do so, but when it is not prohibited by international law to do so’. ¹⁰⁶ The statement is supported by a reference to the Lotus case, elegantly tying together Lotus’ narrower point on the limits of exclusive flag state jurisdiction, with its meta-points on the structure of international law. In legal terms, the argument is somewhat superfluous in an otherwise mostly convincing dissenting opinion, also for the reason that it engages with the more controversial aspects of Lotus’ legacy.

Nevertheless, it shows the considerable might of the Lotus decision that it guided the majority as well as the minority judges in their considerations. At the same time, it should caution us that it was also the Lotus case – together, of course, with the applicable treaty provisions – that allowed the two groups of judges to arrive at polar opposite conclusions.

In the following section, the article will discuss the Enrica Lexie award. As will be shown, the Enrica Lexie tribunal was more successful in utilizing the Lotus judgment than the ITLOS judges, particularly in its interpretation of concurrent jurisdiction over the incident. Alas, the arbitrators ran into some of the same pitfalls when discussing the scope of UNCLOS Article 92(1).

4. Enrica Lexie

4.1 Summary

The facts of the Enrica Lexie case revolved around the near-collision between the ‘Enrica Lexie’, an Italian-flagged vessel, and the ‘St. Antony’, an Indian vessel, ¹⁰⁷ and a shooting incident costing the lives of two Indian fishermen. The ‘Enrica Lexie’s’ crew included a six-member vessel protection detachment of Italian marines. ¹⁰⁸ The marines’ task was to protect the ship from threats of piracy.

On 15 February 2012, as the ‘Enrica Lexie’ was passing through India’s contiguous zone, the ship’s crew noted that the ‘Enrica Lexie’ was on a collision course with the ‘St. Antony’. Given that the attempts of the ‘Enrica Lexie’s’ crew to contact the ‘St. Antony’ through radio and visual signals were unsuccessful, the crew had been under the impression – mistakenly, as it later turned out – that the ‘St. Antony’ was a pirate ship targeting the ‘Enrica Lexie’. In reaction to this – and also in an attempt to avoid a collision – the marines aboard the ‘Enrica Lexie’ fired three consecutive rounds of warning shots. After the third round of shots, when the two vessels were 30 metres apart, the ‘St. Antony’ changed course, avoiding the two vessels’ collision. ¹⁰⁹

The ‘St. Antony’ was an Indian fishing boat, and the shots fired by the marines killed two Indian fishermen. The reason why the ‘St. Antony’ did not react to the ‘Enrica Lexie’s’ attempts to communicate was that the boat’s captain and most of the crew were asleep after a long night of fishing. ¹¹⁰

Following the incident and upon the Indian coast guard’s request, the ‘Enrica Lexie’ changed its course towards India’s territorial sea. Once the ship arrived at the Kochi anchorage, the Indian authorities commenced an investigation and later arrested the two Italian marines. ¹¹¹

While the crux of the case was whether India’s exercise of jurisdiction (prescriptive, adjudicative, and enforcement) was lawful and whether the marines enjoyed immunity from jurisdiction, Italy brought claims under several other UNCLOS provisions as well. India also brought counterclaims against Italy, arguing, inter alia, that the marines’ actions violated India’s right to freedom of navigation.

Already on these facts, the similarities between Enrica Lexie and the Lotus case are apparent. The factual background of Lotus (‘a collision on the high seas between two vessels flying different flags, on one of which was one of the persons alleged to be guilty of the offense, whilst the victims were on board the other’ ¹¹² ) is almost identical to facts in Enrica Lexie. One important difference is that there was no collision in Enrica Lexie, only a near-collision. The fact that the events of the Enrica Lexie case took place in India’s contiguous zone is, however, not a real distinguishing feature: by virtue of UNCLOS Article 58(2), the Convention’s provisions on the high seas are also applicable to the contiguous zone, and the case was decided under these provisions.

Enrica Lexie provides an excellent opportunity to pick up the threads of Norstar on the importance of distinguishing between the distinct types of extraterritorial jurisdiction on the high seas: Italy alleged that India had breached international law through (i) the exercise of its prescriptive jurisdiction (by adopting its Maritime Zones Act); (ii) the exercise of its adjudicative jurisdiction (by bringing criminal proceedings against the marines for acts committed on the high seas aboard an Italian vessel); and (iii) the exercise of its enforcement jurisdiction (through ‘interdicting’ and ‘escorting’ the ‘Enrica Lexie’ to an Indian port.)

Similarly to the Norstar case, Enrica Lexie was an UNCLOS-based dispute, however, it was adjudicated by an ad hoc Annex VII arbitral tribunal, and administered by the Permanent Court of Arbitration. Prior to the rendering of the award in 2020, two provisional measures orders were delivered: in the first order, the ITLOS, acting under Article 292 of the UNCLOS, ruled on the suspension of all criminal proceedings in India and Italy in 2015; ¹¹³ and in the second order, the Annex VII arbitral tribunal ruled in 2016 that India release one of the two marines held in its custody, due to his health conditions. ¹¹⁴

In the merits award, delivered in May 2020, the tribunal rejected most of Italy’s claims on India’s violations of various UNCLOS provisions but upheld the most foundational one: India’s exercise of jurisdiction was found to have violated the marines’ immunity, and the tribunal ordered the marines to be released to Italy to be tried. The tribunal also upheld one of India’s counterclaims, finding that the shooting incident amounted to a violation of India’s freedom of navigation.

Most of the award’s key issues were decided based on a narrow majority of three arbitrators against two. In an unusual matrix of dissenting opinions, the two dissenting arbitrators penned individual dissents and a joint dissent as well. ¹¹⁵ While these dissenting opinions offer valuable criticism and insights on the points they raised, this article will not subject them to closer scrutiny, as the legal issues highlighted here all happened to enjoy unanimous support among the five arbitrators.

The Enrica Lexie award was better received than Norstar, and most commentators limited their observations to smaller criticisms on more nuanced points. These criticisms were focused mostly on the tribunal’s ruling on the marines’ functional immunity, and whether the tribunal had jurisdiction to adjudicate upon this incidental issue. ¹¹⁶

While these are certainly the ‘central’ issues of the case and they break new legal ground, the singular attention devoted to them is somewhat regrettable. This is the case all the more so as the award is exceptionally rich in legal detail, addressing questions which have rarely been in the focus of international adjudication: leaving aside the issues of jurisdiction and applicable law, the case engaged – at least – 14 different provisions of the UNCLOS. ¹¹⁷ This article aims to remedy this, at least to some extent, by focusing on two less-explored issues in Enrica Lexie, also guided by this article’s underlying Lotus theme.

4.2 Concurrent adjudicative jurisdiction in the trichotomy of jurisdictions

From the many issues discussed in the award, there is one that directly engages with the Lotus case: the question of whether Italy had exclusive jurisdiction to initiate criminal proceedings against the marines, or, alternatively, Italy and India both had concurrent – adjudicative – jurisdiction over the incident (the same way as France and Turkey in Lotus).

The issue bore great importance as had the tribunal sided with Italy, the dispute would have been over, and there would have been no need for the tribunal to discuss the thorny issue of the marines’ immunity (and the equally complex question of its own jurisdiction over the question of the marines’ immunity).

The tribunal’s analysis on this point is murky because it is split into two (or, potentially, three) ¹¹⁸ different parts throughout the award. The first part of the analysis is located under Italy’s first claim on how the adoption of certain maritime acts – i.e., the exercise of India’s prescriptive (legislative) jurisdiction – allegedly breached numerous UNCLOS provisions. It is in this context that the arbitral tribunal first analysed whether India had a potentially valid basis under international law for exercising its jurisdiction. ¹¹⁹

However, even before commencing the analysis, the arbitral tribunal already discarded Italy’s claim, seeing ‘no need to address that issue in the context of the present dispute’. ¹²⁰ Thus, the Enrica Lexie award said nothing – neither in the reasoning ¹²¹ nor in the ratio decidendi ¹²² – on the lawfulness of India’s exercise of its prescriptive jurisdiction.

In the very next paragraph, the tribunal jumped to the question of the legal basis of India’s exercise of its jurisdiction ‘over the “Enrica Lexie” incident and the Marines’. ¹²³ This was a stark shift to the questions of adjudicatory and enforcement jurisdiction, and one that was not made explicit by the tribunal. ¹²⁴ The award found that the application of the objective territoriality principle to ships and aircraft was well-founded in international law, and was not the same as treating the ship as the floating territory of the state. The key reference here was, once more, the Lotus case. ¹²⁵

Having found that the objective territoriality principle was a valid basis for India’s exercise of jurisdiction, the tribunal had to test this basis against (i) the rule of exclusive flag state jurisdiction under Article 92(1); and (ii) Article 97(1)’s carve-out rule on collisions and incidents of navigation of the UNCLOS. ¹²⁶

Regarding UNCLOS Article 92(1), the tribunal held that India’s exercise of jurisdiction was:

not only compatible with the Convention, but justified by Article 92, paragraph 1, of the Convention, which provides for the principle of exclusive flag State jurisdiction. Pursuant to this principle, India, as the flag State, has exclusive jurisdiction over the ‘St. Antony’ and may assert its jurisdiction in respect of the offence that was allegedly completed on board its vessel in the exclusive economic zone, in the same way as Italy, as the flag State, has exclusive jurisdiction over the ‘Enrica Lexie’ and may assert its jurisdiction in respect of the offence that was allegedly commenced on board its vessel. ¹²⁷

While it is not spelled out explicitly, it is relatively clear that the tribunal was referring to the exercise of India’s adjudicative jurisdiction. Thus, according to the tribunal, at least in the context of adjudicative jurisdiction, there was no conflict between Article 92(1)’s requirement of exclusive flag state jurisdiction and the Lotus-based exercise of concurrent jurisdiction based on the simultaneous application of the objective and subjective territoriality principles. The two could be reconciled.

Whether the same considerations applied to enforcement and prescriptive jurisdiction, the award did not say. The tribunal later returned to the interpretation of Article 92(1) in the context of India’s exercise of its enforcement jurisdiction, which is something I will address in more detail in Section 4.3.

The tribunal’s analysis of the exercise of India’s adjudicative jurisdiction was, however, incomplete at this point. The award picked up the thread approximately 80 pages later. ¹²⁸ A final hurdle was the question of whether Article 97(1)’s carve-out rule on collisions and incidents of navigation applied to the case.

While the Enrica Lexie incident was decidedly not a collision, it was possible to raise the argument (and Italy did so) that it constituted an incident of navigation, and, under Article 97(1), only Italy – the flag state – had jurisdiction to initiate criminal proceedings against the ‘Enrica Lexie’s’ crew members.

In this segment, the tribunal also elaborated upon the facts and the afterlife of the Lotus judgment, then continued with an in-depth analysis of the interpretation of the term ‘incident of navigation’. ¹²⁹ Finally, based on a fact-driven analysis, the tribunal found that the incident did not qualify as an ‘incident of navigation’, so there was no exception to the applicable rule of concurrent jurisdiction. ¹³⁰ Accordingly, India would have won the case, had it not been for the question of the marines’ immunity, which finally decided the case in Italy’s favour.

Figure 1 depicts the tribunal’s thought process in the framework of the UNCLOS’s provisions.

Figure 1. Flowchart modelling the reasoning of the arbiral tribunal in the Enrica Lexie case

In light of the above analysis, some of the criticism levelled against the tribunal’s treatment of Articles 92 and 97 can be, at least partially, rebutted. Devaney and Tams, for example, argue the following in connection with the two articles:

… the expansive reading of Article 92 creates tensions within the system of UNCLOS Part VII. The Tribunal struggles to explain the existence of UNCLOS Article 97 as a special provision regulating the exercise of penal or disciplinary jurisdiction in relation to collisions or shipping incidents. Put simply, if UNCLOS Article 92 already precludes the exercise of prescriptive jurisdiction by non-flag states, would a further provision be required to exclude such states from applying the law against a ship’s crew? And given that it envisages such proceedings in the flag state or in the state of nationality, would UNCLOS Article 97—widely read as a restrictive rule limiting adjudicative jurisdiction—not rather be an exception to exclusive flag state jurisdiction? ¹³¹

Contrary to the above observation, there is not necessarily a ‘tension’ between the tribunal’s reading of Articles 92 and 97, or rather: any tension is reconciled by the tribunal’s application of the Lotus rule. The tribunal held that (i) the Lotus rule on concurrent jurisdiction and Article 92(1) can be reconciled (at least in the case of adjudicative jurisdiction); and that (ii) Article 97 creates an exception to the Lotus rule (and not (directly) to Article 92(1)). So, for the purposes of adjudicative jurisdiction, the analysis is coherent. That being said, Devaney and Tams were otherwise correct to criticize the tribunal’s expansive reading of Article 92(1), which issue is further explored in the next section.

4.3 Enrica Lexie’s endorsement of Norstar

The tribunal analysed Article 92(1) in more detail under a separate claim: Italy’s claim for a breach of the provision through India’s exercise of its enforcement jurisdiction by directing, interdicting, and escorting the ‘Enrica Lexie’ to an Indian port. Notably, Italy brought no claims under Article 92(1) concerning India’s exercise of its prescriptive or adjudicative jurisdiction. But this did not stop the arbitral tribunal from attributing the same expansive interpretation to Article 92(1) that the Norstar tribunal did, finding that the provision prohibited the exercise of all kinds of jurisdiction, including both prescriptive and adjudicative jurisdiction. ¹³²

The tribunal supported its conclusion by recalling that the concept of jurisdiction under international law is generally understood to entail these three sub-categories, therefore, Article 92(1)’s prohibition must encompass all three of these domains. ¹³³ While the explanation is far from satisfactory, it is already more detailed than what the ITLOS provided in Norstar.

As it was stated in the Norstar Dissenting Opinion, the 2017 UNCLOS commentary adheres to a different interpretation, limiting Article 92(1)’s scope to a prohibition on non-flag-state enforcement. ¹³⁴ Unfortunately, the Enrica Lexie tribunal did not use the opportunity to reflect upon this, all the more so, as the tribunal otherwise heavily relied on the 2017 UNCLOS commentary, citing it on 17 different counts.

Another issue is the apparent contradiction between this holding and the tribunal’s holding on how Article 92(1) can be reconciled with the exercise of concurrent jurisdiction. How can Article 92(1) be at the same time an all-encompassing prohibition and the flexible rule that allows for Lotus-based carve-outs for the exercise of concurrent adjudicative jurisdiction? The tribunal offered no answer.

Next, the tribunal turned to a fact-driven analysis of India’s conduct and found that India’s actions – in interdicting and escorting the ‘Enrica Lexie’ – did not amount to an exercise of enforcement jurisdiction, so there was no breach of Article 92(1). While, on the facts of the case, this holding is not entirely unproblematic either, in a strictly legal sense, the analysis is sound.

The tribunal’s endorsement of Norstar’s expansive interpretation of Article 92(1) can be criticized for the same policy reasons as the Norstar decision, and so it has been. ¹³⁵ Indeed, as Honniball notes, ‘[f]or readers of the M/V Norstar Judgment, this oversight is a moment of déjà vu’. ¹³⁶ Less evident, however, is that Enrica Lexie’s holding on Article 92(1) – similarly to Norstar’s – is also an obiter dictum, although in a way that is somewhat different from Norstar.

First, as argued above, the tribunal made no pronouncements on India’s exercise of prescriptive jurisdiction. According to the award’s dispositif, there was ‘no need to address the question of the compatibility with UNCLOS of India’s 1976 Maritime Zone Act and its 1981 Notification’. ¹³⁷ The tribunal made this pronouncement unanimously. None of Italy’s additional claims engaged with India’s exercise of prescriptive jurisdiction. Thus, the award’s findings on the application of Article 92(1)’s prohibition of prescriptive jurisdiction were not integral to any of the findings in the dispositif; as such, they are obiter dicta.

Second, the article already reflected upon the tribunal’s analysis of India’s adjudicative jurisdiction. In essence, the tribunal held that Article 92(1) can be reconciled with the Lotus-type exercise of concurrent adjudicative jurisdiction. Thus, the tribunal either (i) held (or rather implied) that Article 92(1) did not encompass the exercise of adjudicative jurisdiction and later contradicted itself when it endorsed Norstar’s broad scope or, more probably, (ii) treated the Lotus rule as a carve-out to Article 92(1)’s prohibition. Either way, the tribunal’s holding on the broad prohibition under Article 92(1) was, once more, not integral to the decision on the exercise of adjudicative jurisdiction.

The most recent edition of Akehurst’s Modern Introduction to International Law makes the same observation: ‘[i]n Enrica Lexie … [t]he Tribunal seemed to be speaking generally about jurisdiction, including prescriptive jurisdiction, but India’s activities involved in that case were essentially about enforcement jurisdiction’. ¹³⁸

While it is challenging to piece together the Enrica Lexie tribunal’s reasoning, most of the award is legally sound and offers valuable guidance to less-discussed issues of the law of the sea. The Enrica Lexie tribunal’s utilization of Lotus is also commendable as it confirms the important role Lotus continues to play in the law of concurrent jurisdiction. In addition to Lotus, the arbitral tribunal also made good use of other timeworn cases, such as the Muscat Dhows case ¹³⁹ and the Owners of the Jessie case. ¹⁴⁰ Finally, the tribunal relied heavily on the available commentaries to the UNCLOS and other scholarly publications. This is a strong (and welcome) departure from what the international community is used to in terms of the ICJ’s – or the ITLOS’s – practice of referencing secondary sources outside of their own output. ¹⁴¹

It is regrettable and, indeed, peculiar that, following the ITLOS’s footsteps, the Enrica Lexie tribunal also chose to adopt a controversial position on UNCLOS Article 92(1), especially as the position was not in any way needed for the tribunal to arrive at its final ruling.

5. Jurisdiction after Norstar and Enrica Lexie

Having taken a closer look at both decisions, it seems that, in spite of Enrica Lexie’s apparent endorsement of Norstar, the two cases contradict each other in a key aspect: in Norstar, Italy’s exercise of adjudicative jurisdiction over acts committed on the high seas was found to have violated UNCLOS Article 92(1); in Enrica Lexie, India’s exercise of its adjudicative jurisdiction over acts committed on the high seas was found to conform with UNCLOS Article 92(1) (and only violate customary international law on immunity).

Moreover, the two cases opened the door to diverging interpretations of UNCLOS Article 92(1). A 2022 edition of a law of the sea treatise already makes mention of this, noting the ‘two strands of opinion’. ¹⁴² In reality, however, there are three diverging interpretations: (i) the original mainstream reading of the provision under the 2017 UNCLOS commentary, limiting the scope of Article 92(1) to enforcement jurisdiction; (ii) the view proposed by Italy in the Norstar case, according to which ‘in certain exceptional circumstances an act that falls short of enforcement action may still become relevant from the perspective of article 87, for instance when it produces some “chilling effect”’; ¹⁴³ and (iii) the position adopted by the Norstar majority and endorsed by the Enrica Lexie tribunal, holding that any exercise of extraterritorial jurisdiction (including prescriptive jurisdiction) is considered a breach of UNCLOS Article 92(1) (and, by extension, UNCLOS Article 87).

Finally, the present article has made the argument that there is a ‘mismatch’ between the tribunals’ holdings and the underlying facts of the two cases; as a logical result of this mismatch, neither of the two decisions provides an analysis or even a description of state legislation that is contrary to UNCLOS Article 92(1): in both cases, the underlying state legislation was only of secondary importance to the tribunals.

Regardless, based on the abstract reading of the obiter dicta in the two decisions, a door is now open for litigants to use the UNCLOS’s compulsory dispute settlement mechanism to bring claims against states based on the adoption ¹⁴⁴ of extraterritorial legislation, ‘even if the State refrained from enforcing those laws on the high seas’. ¹⁴⁵

Only when (and if) such claims are brought will it be possible to see the real effect and reach of the tribunals’ broad reading of UNCLOS Article 92(1).

6. Conclusion

An important takeaway, and one that remains true in light of Norstar and Enrica Lexie, is that it is for good reason that the Lotus decision is integral to our discussions on the international law of jurisdiction. Much of the judgment’s holdings on fundamental distinctions between different types of jurisdiction remain valid today.

This article made the argument that legal doctrine should follow the three-pronged (legislative, adjudicative, and enforcement) categorization of the types of jurisdiction, as each type possesses sufficiently distinct characteristics to warrant the application of different legal tests. Some of the errors in the Norstar judgment also stem from the fact that the majority did not make adequate distinctions between the respective types of jurisdiction.

While the Enrica Lexie tribunal fared better in this respect, the arbitrators were not consistent in their approach, which made their reasoning contradictory at some points. The Enrica Lexie tribunal’s analysis on concurrent jurisdiction and the relationship between UNCLOS Articles 92 and 97 was more nuanced, in no small part thanks to the wisdom received from Lotus. Finally, and regrettably, both the Enrica Lexie and Norstar tribunals exercised strategic selectivity when relying on authorities in general, and Lotus, in particular.

The three cases also highlight the strange fate of the rule on exclusive flag state jurisdiction enshrined under UNCLOS Article 92(1). At the time of Lotus, the provision had not existed, and yet, the ‘spirit’ of the provision is very much present in the tribunal’s reasoning. In the decades to follow, Lotus continued to provide the framework for discussion on the content and scope of the rule.

In Norstar, Article 92(1) took centre stage – and provoked the ire of commentators – even though Panama did not rely on the provision in its prayer for relief. Conversely, in Enrica Lexie, Italy did rely on Article 92(1) in its request, if only in the context of India’s enforcement jurisdiction. In both of these cases, the tribunals attributed a broad scope to Article 92(1) in holdings that did not engage with the facts of the respective disputes. Accordingly, as this article argued, these holdings are no more than obiter dicta. Ironically, Lotus’s most controversial holding was also an obiter dictum.

This brings us (back) to Lotus, a decision, which at certain times seemed to be in peril of being consumed by its own shadow. For a large part of the twentieth century, it seemed that the commentary, the legacy, the meta-discussion: the ‘baggage’ has almost become more important than the judgment itself. For these reasons, it is vindicating to see Lotus back in its natural, law of the sea habitat, to be utilized as a guiding authority on the legal point it originally decided. Norstar and Enrica Lexie re-positioned Lotus into its traditional setting. As it turns out, the judgment sits quite comfortably and continues to serve its purpose as a judicial authority on a narrow point of law. It remained largely unfazed by a century of legal developments and will continue to be the applicable law for issues concerning concurrent jurisdiction on the high seas that do not qualify as collisions or incidents of navigation.

It is also for the same reason, however, that the two tribunals’ endorsement of Lotus serves as a cautionary tale: when relying on previous decisions, international courts and tribunals should exercise vigilance not to distort the normative content of these holdings by quoting them out of context or in an incomplete manner. This is particularly true for ‘old’ cases, such as Lotus. This does not mean that Lotus or other early precedents should be discarded; rather, just like ancient artifacts, they should be handled with extra care.