The Baltic Sea Cable-Cuts and Ship Interdiction: The C-Lion1 Incident
According to widely held Russian perceptions, Russia is at war with NATO. In September 2024, Vladimir Putin stated that lifting restrictions on Ukraine’s use of long-range NATO missiles to attack Russia “will mean that NATO countries—the United States and European countries—are at war with Russia. And if this is the case, then, bearing in mind the change in the essence of the conflict, we will make appropriate decisions in response to the threats that will be posed to us.” On 17 November 2024, the United States lifted the restrictions.
The C-Lion1 Incident
That same day, a submarine telecommunications cable connecting Sweden and Lithuania was cut. The next day, the C-Lion1 submarine telecommunications cable linking Finland and Germany was cut. Both cable-cuts (I refer to both as the C-Lion1 incident) occurred in Sweden’s exclusive economic zone (EEZ). German Defence Minister Boris Pistorius stated that “[n]o one believes that these cables were cut accidentally.” In his view, the cable-cuts were “probably sabotage.”
Swedish and Finnish authorities launched a criminal investigation into the C-Lion1 incident focusing on the Chinese bulk carrier Yi Peng 3 that has a Russian captain. On 19 November, as the ship was exiting the Baltic Sea via the Danish Great Belt, it was accompanied by Danish Navy warships. The Yi Peng 3 stopped in the Kattegat, where Denmark and Sweden have established a long and narrow EEZ corridor. As of 25 November, the Yi Peng 3 remains anchored in the Danish part of the Kattegat close to the outer limit of the Danish territorial sea. In addition to the Royal Danish Navy presence, Russian and German warships and Swedish coast guard ships have been sighted in the area. According to media reports, the second anchor of the Yi Peng 3 appears damaged.
It is unclear whether the coastal authorities have boarded the ship in the Kattegat. The coastal States are seeking the flag State’s (China) consent to inspect the ship. It is unclear whether China will grant the consent. Irrespective of this, this post demonstrates that the coastal authorities may rely on additional legal bases for interdicting the Yi Peng 3.
Map of the location of the Yi Peng 3 in the Kattegat where the ship remained anchored for days (as of 19 November 2024).
Source: Marine Traffic
The Wider Context
Submarine cables are difficult to protect against intentional cutting. For example, the submarine cables of Norway, the United Kingdom, France, and Taiwan have repeatedly been cut in recent years, likely by foreign ships in their maritime area. In some cases, such cuts lead to disruptions to the internet and banking services in the affected coastal areas.
Like many other coastal States, the Baltic States and Finland are highly dependent on submarine cables for their internet connections. The Nord Stream explosions of September 2022, the Balticconnector incident of October 2023, and the recent C-Lion1 incident of November 2024 underscore the vulnerability of Western States to intentional damage caused to their critical offshore infrastructure. These maritime nations are much more dependent on maritime connections than their land-based rivals China, Russia, and Iran.
In 2023, a Chinese bulker carrier dragged its anchor along the seafloor of the Gulf of Finland for close to 200 km, damaging a Baltic Sea gas pipeline. The suspiciously analogous act by another Chinese ship in the C-Lion1 incident demonstrates how foreign adversaries may collect valuable intelligence on the resilience of Baltic offshore infrastructure. The dragging of the anchor proves that some objects can be effectively damaged by more primitive means, while other incidents demonstrate that sometimes more specialised equipment is required (e.g., the Russian special force on underwater warfare). Such information can then be used to cut off the energy and telecommunications connections of a particular region (e.g., the partly enclaved Baltics or Taiwan) either in hybrid warfare or for enforcing a blockade.
In the above-described geopolitical context, sabotage cannot be excluded. But this post is limited to examining potential legal bases for interdicting the Yi Peng 3 based on the peacetime legal framework of maritime law enforcement. On the meaning of sabotage, see Tramazzo. On the application of sabotage in respect of the Nord Stream explosions, see Lott.
The Legal Loopholes
The United Nations Convention on the Law of the Sea (UNCLOS), widely considered as the Constitution of the Oceans, left certain areas of the law of the sea loosely regulated. One such area is the protection of critical offshore infrastructure outside the limits of the territorial sea. According to the dominant view, Article 113 of UNCLOS, which regulates the breaking or injury of a submarine cable or pipeline, does not provide for universal jurisdiction. Criminal jurisdiction can be implemented by the relevant coastal State, and only if the cable is damaged by a national of the coastal State or a ship flying its flag (International Law Association (ILA), 2024 Report, p. 16).
At the time of the drafting and adoption of the UNCLOS (1973-1982), use of the internet was mostly a privilege of a few government and university employees. By the time the UNCLOS entered into force (1994), the internet was gradually becoming the cornerstone of everyday life in Western societies, supported by the massive laying of submarine telecommunications cables on the seafloor. Apparently, the drafters of UNCLOS did not anticipate such a landslide societal change. They granted hardly any protection to the submarine telecommunications cables that carry 95 to 99 percent of intercontinental data flows. By now, these cables serve as the lifelines of our contemporary internet-based world economy.
The current protection regime of inter-continental submarine cables rests mostly on Article 10 of the 1884 Paris Convention for the protection of cables. Denmark and Sweden are parties to the 1884 Paris Convention, while China, the flag State of the Yi Peng 3, is not. The United States considers Article 10 to reflect customary international law (ILA, 2024 Report, p. 16).
Article 10 of the 1884 Paris Convention might enable Sweden and Denmark to board the Yi Peng 3 if they consider this provision reflects customary international law. The customary law-status would ultimately be for a court or tribunal to decide. Alternatively, based on a literal interpretation of Article 10, an argument might be made that boarding and questioning the crew can be enforced by a coastal State party to the treaty against a flag State that is not party to the treaty, such as China in respect of the Yi Peng 3 (see Papastavridis, p. 34). But given the unclear legal status of this provision and very limited State practice in its implementation, it is unlikely that either Denmark or Sweden is interested in testing the limits of Article 10 of the 1884 Paris Convention against the Yi Peng 3, especially given that boarding authorities would be only entitled to check the ship’s documents and question the crew.
Robert Beckman has concluded that “[i]t is evident … that there are serious security gaps in the current legal regime and that neither the 1884 Cable Convention nor UNCLOS adequately address the issue of intentional damage caused to submarine cables” (p. 289; see also Davenport, p. 6). Beckman has stressed the need for a new global treaty on the protection of submarine cables “wherever the act took place, whatever the nationality of the perpetrator, and regardless of their motive or purpose” to close the current gaps in the legal framework (p. 288).
What additional legal basis could permit Danish or Swedish enforcement measures against a ship suspected of damaging submarine cables in the Baltic Sea? There may be lessons to learn from the Balticconnector incident.
Lessons from the Balticconnector Incident
In the Balticconnector incident of October 2023, two out of eight telecommunications submarine cables serving as the backbone of Estonia’s digital society were damaged in the Estonian EEZ. A submarine gas pipeline was also damaged in the Finnish EEZ. China admitted that their NewNew Polar Bear caused the damage by its anchor, but has not cooperated with the investigators.
Finland was ready to intervene if the NewNew Polar Bear entered the Finnish territorial sea but considered that it has no right to take any law enforcement measures against the ship such as boarding, inspection, and arrest, while the ship remained in Finland’s EEZ. However, I have argued elsewhere that Finland might have had sufficient legal basis under UNCLOS and general international law to board, inspect, and potentially arrest the NewNew Polar Bear (e.g., under arts. 79(2) and 221 UNCLOS).
The Balticconnector incident illustrates that while the coastal State may use her right of hot pursuit against a ship that has caused an oil spill or gas leakage with her anchor (arts. 79(2) and 221 in combination with UNCLOS art. 111(2)) on the grounds of protecting and preserving of the marine environment (UNCLOS, art. 56(1)(b)(iii)), a similar right does not exist in relation to submarine cables that a ship damages. This is because the breaking or damaging of a cable does not cause marine environmental pollution.
In the immediate aftermath of the recent C-Lion1 incident, the Finnish Defence Minister signalled a significant change of State practice in the Baltic Sea regarding ships suspected of damaging critical offshore infrastructure, based on lessons learned from the Balticconnector incident. He notes that the previous interpretation of the law of the sea was too conservative to the extent that it did not support the interdiction of a foreign ship outside the limits of the territorial sea, based on such a suspicion. Minister Häkkänen stated that it is quite clear that if critical infrastructure has been destroyed or seriously damaged, and it is necessary to find out who did it, then it cannot be a situation where the victim State does not intervene.
By contrast, the Estonian Minister of Defence is still of a different opinion. A year after the Balticconnector incident, Minister Hanno Pevkur commented on the incident and claimed that “checking a ship that starts its journey from a Chinese port somewhere and ends in St. Petersburg is essentially excluded if it remains solely in international waters.” This echoes a previous statement of the representative of the Estonian Ministry of Defense in November 2023 that direct coercion aboard a foreign ship is impossible if the ship is in so-called international waters. This position is not correct, and it is rather short-sighted in the context of presumed intentional damaging of the coastal State’s critical offshore infrastructure.
If an act against critical offshore infrastructure constitutes a hybrid attack, then the entire operation targets the so-called legal loopholes identified above. To create deterrence and legal resilience against such operations, the coastal State should be ready to implement enforcement measures against the suspected ship by creating a precedent that would allow a court or tribunal to assess the legality of its measures (see also Ringbom).
Complementary Legal Bases
In the Arctic Sunrise case, the Annex VII Arbitral Tribunal demonstrated its willingness to consider potential legal bases for Russia’s law enforcement measures beyond the scope of the UNCLOS and based on general international law (para. 235). In this context, coastal States may consider taking enforcement measures against suspected foreign ships in response to incidents like the C-Lion1 based on the customary law of necessity (art. 25, Draft Articles on State Responsibility with commentaries; in the context of the Balticconnector incident, see the discussion in Ringbom and Lott, p. 186-187).
Complementary legal bases for coastal States to act in response to damage to critical offshore infrastructure outside the territorial sea stem from the concepts of piracy (art. 101, UNCLOS) and terrorism. However, when applied in practice, they appear to have even more limitations than the previously discussed legal bases under UNCLOS and the 1884 Paris Convention.
The Annex VII Arbitral Tribunal found in the Arctic Sunrise case that “[a]n essential requirement of Article 101 is that the act of piracy be directed ‘against another ship’” (para. 238). Nonetheless, Article 101(a)(ii) of UNCLOS still clearly stipulates that piracy consists of acts against property in a place outside the jurisdiction of any State. Submarine cables meet the definition of property, and they are mostly located outside the jurisdiction of any State. Thus, the concept of piracy could, in principle, be interpreted dynamically.
However, Article 101 of UNCLOS also stipulates that piracy must include the following elements: “any illegal acts of violence or detention, or any act of depredation, committed for private ends.” It might be more difficult to meet these criteria, because the damaging of submarine cables and pipelines is mostly not committed for private ends, even if it might be considered an act of violence. In the context of hybrid warfare at sea, States, rather than private actors, benefit from the breaking or damaging of critical offshore infrastructure of their perceived adversaries. Furthermore, when conducting hybrid operations, States tend to seek to maintain plausible deniability for their involvement in the operations to avoid their responsibility for the internationally wrongful acts.
Under the Convention for the Suppression of Unlawful Acts Against the Safety of Maritime Navigation (SUA) and its 2005 Protocol, an offence includes the discharge of explosives from a ship or the use of a ship in a manner that causes death or serious injury or damage (art. 3bis(1)(a)(i)(iii)). However, the coastal State cannot board the suspected ship without the express authorization of the flag State (SUA Protocol, art. 8bis(5)(c)). Similar limitations apply under the Convention for the Suppression of Terrorist Bombings (see also the ILA, 2024 Report, p. 52).
Yet in the Arctic Sunrise case, the Annex VII Arbitral Tribunal indicated that it might be lawful for the coastal State to take additional preventive enforcement measures in the EEZ against a suspected terrorist attack:
One of the rights of a coastal State in its EEZ that may justify some form of preventive action against a vessel would derive from circumstances that give rise to a reasonable belief that the vessel may be involved in a terrorist attack on an installation or structure of the coastal State. Such an attack, if allowed to occur, would involve a direct interference with the exercise by the coastal State of its sovereign rights to exploit the non-living resources of its seabed. It is not, however, necessary for this Tribunal to determine the extent of any power to take such preventive action. This is because on the facts here there was no reasonable basis for Russia to suspect that the Arctic Sunrise was engaged in or likely to engage in terrorist acts (para. 314).
In the EEZ, coastal States can also make a more proactive and coordinated use of the possibility to board ships suspected of intentionally damaging submarine cables based on the captain’s consent for the verification of the ship’s documents and cargo. The U.S. Commander’s Handbook on the Law of Naval Operations makes it clear that this practice is well-rooted in international law (p. 4-7; see also Fink, p. 34-35). But this would have a rather limited effect, because this right does not permit the coastal State to inspect the vessel nor adopt additional enforcement measures. Furthermore, presumably only a handful of suspected ship captains would be willing to provide such consent.
The Special Circumstances of the Danish Straits
In addition to the previously discussed claim of necessity, the potentially most effective legal basis for adopting enforcement measures against ships suspected of damaging submarine cables in the Baltic Sea might stem from the legal regime applicable to the Danish Straits. Is Denmark entitled to enforce its domestic laws in the Great Belt or Øresund against a ship suspected of having damaged critical offshore infrastructure in the Baltic Sea?
The key question here is how to interpret the 1857 Copenhagen Convention. Article I(1) stipulates that: “No ship for the future shall under any pretext whatsoever be seized or subjected to any stoppage on its way through the Sound and the Belts.” Erik Brüel concluded in his seminal monograph on international straits that “it seems illogical to say that it prevents Denmark carrying out civil arrests” (p. 44).
Thus, it can be argued that Denmark has a limited control over commercial ships transiting through the Danish Straits under the 1857 Copenhagen Convention and Article 35(c) of UNCLOS. In the Case Concerning Passage Through the Great Belt before the International Court of Justice, Denmark maintained (and Finland appeared to concur) that, in general, the right of (strait-specific non-suspendable) innocent passage applies to the Great Belt ( p. 3ff).
The question thus arises whether the Copenhagen Convention excludes Danish enforcement measures against non-innocent passage in the Belts or Øresund?
Brüel writes that:
What the treaty prevents are measures which force all vessels to stop, especially those which have no connection whatever with the land but are merely passing by. The right to make arrests, even apart from the fact that the question will hardly arise in one out of 10,000 ships, affords no opportunity for evading the provisions of the treaty: It therefore seems compatible with the principle of the freedom of passage which is laid down in the treaty (p. 44-45).
Brüel concludes that “Denmark should be recognised as possessing a right to make civil arrests on board vessels which are passing through” (p. 45). If a civil arrest that depends on civil law claims over a particular ship is possible in the Danish Straits, then Denmark and Sweden might also be entitled to enforce their laws against a ship that has violated the rules of innocent passage in the Great Belt or Øresund.
Conclusion
It is difficult for me to believe that Denmark would not take any measures against a foreign ship if its passage were prejudicial to the peace, good order, or security of Denmark within the meaning of Article 19(1) of UNCLOS. A ship suspected of conducting a hybrid attack against a Danish NATO ally and that transits through the Danish territorial sea in the Great Belt or Øresund, thereby posing a threat to the numerous objects of critical offshore infrastructure located therein, also poses a security threat to Denmark. Ultimately, though, any enforcement measures pursuant to this line of argumentation would depend on the interpretation of the 1857 Copenhagen Convention in the context of the progressive development of the international law of the sea, including the regime of straits in the 20th century.
***
Dr Alexander Lott works as a researcher at the Norwegian Centre for the Law of the Sea at the UiT – The Arctic University of Norway.
Photo credit: Rajavartiolaitos (Finnish Border Guard)