Scenario 16: Cyber attacks against ships on the high seas

This scenario considers a series of cyber operations against merchant vessels and warships on the high seas from the perspective of public international law. It analyses in particular issues related to jurisdiction and freedom of navigation on the high seas, as well as whether the cyber operations amounted to a prohibited use of force.

Keywords
high seas, jurisdiction, freedom of navigation, sovereign immunity, countermeasures, use of force, intervention

Facts
[F1] State A and several other States have all agreed to pass domestic legislation that prevents companies incorporated in their territory from selling certain prohibited goods to State C. State A argues that this is justified by security concerns about how these goods would be used. The legislation grants State A the power to request the cargo manifests of ships bound for State C which depart from any of State A’s ports as well as the power to search any of State A’s flagged ships for the prohibited goods. State B has refused to follow State A’s actions and publicly stated that it does not believe the goods should be prohibited, and would continue to permit the sale of these goods to State C. The sale of these goods has not been prohibited under international law.

[F2] State A has since made numerous public statements criticising State B for not passing similar legislation. In these statements, State A has suggested it might stop and investigate ships that depart from State B and are under State B’s jurisdiction that are bound for State C if there are reasonable grounds to believe that there are prohibited goods on-board. In response, State B has sent two warships to the high seas that are adjacent to the exclusive economic zone of State A, to prevent any searches.

[F3] A month after State A enacted the legislation, State B discovers that the cyber infrastructure belonging to government custom officials in State B’s main port has been breached. State B determines that the breach originated from a targeted spear-phishing campaign against the computer systems in the customs agency headquarters. Malware installed as a result of the spear-phishing campaign has enabled the perpetrator to gain access to the shipping schedules and cargo manifests belonging to many of the ships that have departed from State B, destined for State C. This information has been classed as sensitive by State B in the aftermath of the legislation passed by State A and others. In the following days, State B traces the cyber operation to the Cyber Branch of State A’s military (incident 1).

[F4] Two weeks after the discovery of the unauthorised access into the computer systems of the customs agency, State B’s coastguard (a branch of State B’s military force) receives a distress signal from a merchant vessel registered to State B. The captain of the vessel states that their on-board navigational systems are suffering from significant interference as they sail through the high seas adjacent to State A’s exclusive economic zone, making navigation dangerous and increasing the risk to the crew and the vessel. An investigation traces the interference to an Advanced Persistent Threat (APT) group that has achieved a foothold in the ship’s network through a spear-phishing campaign targeting crew members. The network was not segmented, thus giving the APT group real-time access to multiple systems across the ship, including the navigation systems. The sophistication of the APT operation, and the similarity with the above described spear-phishing campaign against State B’s government customs officials, leads State B to attribute the operation to the Cyber Branch in State A (incident 2).

[F5] To protect the merchant vessel from any possible boarding threat, State B’s coastguard commands one of the nearby warships dispatched to the high seas to assist the merchant vessel. The warship reports that there has been disruption to its navigational radar capabilities and that it is unable to locate the merchant vessel. State B traces the source of the interference to the same APT group responsible for the interference with the merchant vessel’s on-board navigational systems. State B therefore attributes the interference to the Cyber Branch in State A (incident 3).

[F6] To end both operations, State B infects a computer in the Foreign Office of State A with a virus designed to render the machine permanently unusable, and that spreads across the network to other computers. In the aftermath of incident 1 and whilst the cyber operation was continuing against the customs officials, this weakness was identified by State B as being exploitable if needed. State A’s Foreign Office shares a building with the Cyber Unit, and it is believed by State B that they may share the same network. The purpose is to stop the cyber operations of State A against State B, conducted by State A’s Cyber Branch (incident 4).

[F7] All concerned States are parties to the United Nations Convention on the Law of the Sea (UNCLOS).

Examples

 * Cyber interference against vessels in the Persian Gulf and Gulf of Oman (2019)

Legal analysis
For a general overview of the structure of analysis in this section, see Note on the structure of articles.

[L1] The analysis below will consider whether any of incidents 1–3 would be classed as a violation of international law, with a particular focus on the cyber operations against the two vessels. The analysis will then consider whether the response in incident 4 would be lawful under international law.

State organs and exercise of governmental authority
[L2] The Cyber Branch of State A’s military, which is responsible for the cyber operations in incidents 1, 2 and 3, is an organ of State A; its conduct is therefore attributable to State A.

Prohibition of intervention
[L3] With respect to the first prong of the test, the targets of the cyber operations in incidents 1 and 3 were predominantly governmental and run by State B, i.e., its customs agency and warships assisting the merchant vessels. This means that State B normally had a significant amount of freedom in deciding how they are used, and thus the operations may well have had a bearing on its domaine réservé.

[L4] Incident 2 could also fall within the domaine réservé of State B''. Although shipping is subject to some international regulation (notably the regime of the UN Convention on the Law of the Sea or UNCLOS), States retain considerable freedom to pursue their own policies concerned with shipping, and thus as long as “these did not otherwise violate international law”, shipping would likely fall within the domaine réservé of the relevant States.  In fact, targeting shipping has been recognised as a form of economic coercion (a form of intervention), which further suggests that it falls within the domaine réservé of State A.

[L5] It is less clear whether the second prong of the test, the criterion of coercion, was met. The cyber operations in incidents 1-3 must be coercive in nature, meaning that the operations had the potential for compelling State B to act (or refrain from acting) in a way that it would not otherwise have done (under the first approach to coercion described above), or to deprive State B of its ability to control its internal or external affairs (under the second approach).

[L6] For this reason, the cyber operation in incident 1 is unlikely to be coercive. The operation resulted in sensitive data being accessed by State A. However, this alone is unlikely to compel State B to change the way it acts, or to deprive it of its control over the sale of certain goods to State C, thus lacking the coercive element needed to satisfy the conditions above.

[L7] In contrast, incidents 2 and 3 are more likely to possess the necessary coercive element. Under the first approach to coercion, which requires that the act should compel “the target State to engage in an action that it would otherwise not take”, it is likely that the cyber operation in incident 2 would achieve this. If State B feared that cyber operations against ships bound for State C might pose serious risk to the safety of its crew and the vessels themselves, then this apprehension may indeed bring about a change in State B’s actions. This would also be true under the broader second approach to coercion, since State B would be deprived of its ability to control or govern matters within its domaine réservé, such as its shipping policies.

[L8] The cyber operation in incident 3 would also likely possess the required coercive element by potentially compelling State B to refrain from assisting its merchant vessels, or from dispatching warships to that area of the high seas. [L9] Incidents 2 and 3 are therefore likely to qualify as a prohibited intervention under international law. However, this is subject to an important caveat; if these incidents were considered as amounting to a legitimate maritime law enforcement operation (see paras L21–L25 below), they would not constitute an intervention prohibited by international law.

Flag State jurisdiction
[L10] State B is the flag State of the merchant vessel in incident 2, thus it possesses flag State jurisdiction over the vessel. In accordance with the principle of the exclusive jurisdiction of the flag State, no State other than State B is permitted to interfere with the vessel unless one of the exceptions outlined above may be identified.

[L11] The domestic legislation passed in State A provided that State with the authority to request and collect the cargo manifests of ships flying its flag and travelling to State C, and also to search any of such ships. However, this does not amount to an authority to search or interfere with ships that do not fly the flag of State A.

[L12] State A is not the flag State of the vessel in question. State A therefore cannot exercise its enforcement jurisdiction against the vessel. As such, the cyber operations against the vessel, which resulted in the interference with the vessel’s navigation systems would amount to a breach of State A’s international obligations, unless it could be shown that an exception existed which justified the interference (discussed below).

Freedom of navigation
[L13] State A’s cyber operation against State B’s merchant vessel (incident 2) could also amount to a breach of the freedom of navigation. The cyber operation against the vessel’s navigation systems has impeded that ship’s navigation, resulting in the captain’s decision to transmit a distress signal. That the interference was conducted through cyber means is immaterial as it has been held that an interference does not need to be physical in nature in order to qualify as a breach of the freedom of navigation.

[L14] There are two possible grounds that State A could advance as justifications for its interference with the merchant vessel. It would be entitled to exercise enforcement jurisdiction in relation to State B’s vessel if it had either a “specific allocation of authority under international law” (i.e., the exceptions provided for in UNCLOS and other regimes), or the consent of State B to act.

[L15] Under UNCLOS, State A would have to have had “reasonable grounds to suspect” that one of the relevant exceptions had existed on board the merchant vessel. However, State A did not make any claim that any of these exceptions applied in relation to State B’s merchant vessel. On the contrary, State A’s public statements that it would stop and search ships that may have prohibited goods on board suggest that its motivation for interference with the merchant vessel was extraneous to the grounds permitted by UNCLOS. In addition, the scenario does not mention any other applicable treaty that would permit such an interference, nor any relevant resolution of the UN Security Council. As such, it would be difficult to argue that there existed a specific allocation of authority under international law for State A to justifiably exercise its enforcement jurisdiction.

[L16] Similarly, State B has not given its consent for State A to exercise enforcement jurisdiction over the merchant vessel.

[L17] The cyber operation against the merchant vessel by State A would therefore amount to a violation of State B’s freedom of navigation on the high seas.

Sovereign immunity
[L18] There was no international armed conflict between States A and B. Accordingly, the immunity of State B’s warship targeted by the cyber operation in incident 3 did not cease to apply as between those States.

[L19] As such, the warship had complete immunity and could not be interfered with in any way. The radar disruption suffered by the warship amounted to an impermissible interference, as the cyber operation significantly impaired the radar infrastructure by undermining its ability to locate the merchant vessel.

[L20] Therefore, the cyber operation against State B’s warship qualified as a breach of the sovereign immunity afforded to the warship under the UNCLOS regime.

Maritime law enforcement operations
[L21] The cyber operation in incident 2 may be considered as a law enforcement operation. Circumstances that would support such a conclusion include that the cyber operation was not intended to defend State A against a threat to its sovereignty, but rather to prevent the passage of a vessel. If the cyber operation was preceded by the attempted arrest and charging of those on the merchant vessel, this would also support qualifying the operation as law enforcement. However, the isolated nature of the cyber operation against the merchant vessel, and the lack of prior circumstances suggesting a law enforcement action may undermine the operation’s classification as maritime law enforcement.

[L22] If the cyber operation in incident 2 was characterised as a law enforcement operation, it would be likely to be classed as unlawful. The lack of a valid jurisdictional basis would support a finding of unlawfulness. State A had no jurisdictional basis to conduct a cyber operation against the merchant vessel, as it was not the flag State of the vessel, nor did it have the flag State’s consent. The exceptions that can grant jurisdiction over foreign flagged vessels also do not apply in this scenario (see paras L14-L16 above).

[L23] In addition, the actions taken would likely have also breached the standards imposed on enforcement operations. The cyber operation was the first measure used, thus not meeting the requirement that auditory or visual signals should be used first. Moreover, the interference with the navigation systems caused by the cyber operation had the potential to result in the death or injury to the crew and severe damage to the ship. State A would thus likely have failed to make all efforts to ensure life would not be endangered.

[L24] In contrast, incident 3 is more likely to concern a use of force under Article 2(4) of the UN Charter. This is the general presumption when vessels possessing sovereign immunity are subject to a foreign state’s enforcement jurisdiction.

[L25] In addition, the cyber operation targeting the warship interfered with its ability to assist and to protect the merchant vessel. There were also no pre-existing circumstances that could be considered as being suggestive of law enforcement, given that the cyber operation was the first occurrence of any action against the warship in incident 3. The military nature of the Cyber Branch in State A responsible for the cyber operation may support the finding of a military situation with two military forces in opposition to each other.

Use of force
[L26] Incident 1 would be unlikely to be considered as a use of force. The severity of the operation was minimal; the data which was accessed was not corrupted or deleted, and the infrastructure, in which it was stored remained functional after the incident. Though the access was invasive, the lack of significant consequences militates against considering the operation as a use of force.

[L27] Incidents 2 and 3 would have to be distinguished as potential uses of force under Article 2(4) of the UN Charter rather than maritime law enforcement operations (see section 2.2.5 above) in order for the following analysis to apply.

[L28] With respect to incident 2, the cyber operation would likely lack the necessary severity due to the absence of physical harm to the individuals on board of, and damange to the vessel and because the navigation’s loss of functionality was only temporary in nature. As such, it is unlikely that the cyber operation would reach the threshold of a prohibited use of force.

[L29] In contrast, the cyber operation against the warship (incident 3) impacted the functionality of the radar to the point that the merchant vessel could no longer be located. This interference with the warship’s systems could be considered as severe. Though the operation did not result in direct physical harm, it arguably interfered with State B’s critical national interest by impacting the functionality of one of its warships and by undermining its ability to assist and protect its merchant vessels.

[L30] Moreover, the military nature of the vessel means the targeted systems were of a greater concern to State B. The cyber operation would also be considered military in character as it can be attributed to the military Cyber Branch in State A. Both of these considerations would further support classifying the cyber operation as a use of force.

[L31] An isolated attack against a warship can be sufficient to engage the right to self defence provided it reaches the threshold of an armed attack. However, it is unlikely that the cyber operation in incident 3 would be considered as a grave use of force sufficient to amount to an armed attack due to the absence of death or injury to persons, or the destruction of property.

Permissible responses by State B
[L32] Once it has been established that State A’s conduct amounted to an internationally wrongful act, State B may have the right to resort to countermeasures.

Countermeasures
[L33] State B launched a cyber operation against State A’s Foreign Office with the aim of ending cyber operations against it (incident 4). International law permits State B to take countermeasures in order to bring to an end those actions by State A that qualify as internationally wrongful acts. Incidents 2 and 3 arguably amounted to internationally wrongful acts by State A (see paras L12, L17, and L20 above).

[L34] Countermeasures cannot be anticipatory; they must only be taken in response to an internationally wrongful act. The vulnerability in the Foreign Office was identified prior to the cyber operations against the vessels; however, the operation against its computer system was not launched until after the cyber operations in incidents 2 and 3. Therefore, the countermeasure would not qualify as anticipatory.

[L35] State B, before taking countermeasures, should normally call upon State A to fulfil its international obligations (i.e., to cease the operations against State B) and notify State A of its intention to take countermeasures. However, this has been recognised as incompatible with the purpose of many cyber countermeasures (as it may enable the opposing state to mitigate or avoid the effects of the countermeasure). This position has also been recognised by several States. The duty to notify State A may then be precluded due to the urgency of the countermeasures that are needed to preserve the rights of State B. Given the immediacy of the internationally wrongful act that was being committed against the two vessels, it is unlikely the countermeasures would require notification, as to notify State A of it would likely undermine its purpose and provide State A with an opportunity to mitigate the consequences of the operation.

[L36] By contrast, the countermeasure used may fail the test of proportionality. This requires that “the effects of a countermeasure must be commensurate with the injury suffered” considering the “gravity of the internationally wrongful act and the rights in question”. This determines the type and intensity of the countermeasure available to State B.

[L37] In incident 2, the cyber operation had the potential to result in death of or injury to the crew, or result in substantial damage to the vessel itself. Similar consequences were possible in incident 3. However, the actual consequence was a loss of functionality, albeit for critical systems essential when a vessel is under way, which created the distinct possibility of physical harm to persons or physical damage to the vessel.

[L38] In contrast, incident 4 concerned the spreading of a virus through a computer network which aimed to irreparably damage the compromised infrastructure. State B was targeting the Foreign Office on the basis that they may share the same network as the Cyber Branch resulting in widespread effects across State A departments.

[L39] Judging whether a supposed countermeasure is proportionate is necessarily an ‘approximation’. Ultimately, this can make reaching a conclusive judgment on proportionality difficult, as in this case. It could be argued that the possible consequences of the operations in incidents 2 and 3 exceeded the consequences in incident 4, thus making the countermeasure commensurate. If the consequences in incidents 2 and 3 escalated to death of or injury to the crew, or significant damage to the ship as a whole, the countermeasures would likely be proportionate. However, as these possible consequences did not materialize, it may be that the widespread permanent damage to the cyber infrastructure in State A’s Foreign Office would not be commensurate with the injury suffered by State B.

[L40] Finally, the purpose of the countermeasure should be to induce State A to comply with the legal consequences of the internationally wrongful act – the purpose cannot be retribution. Therefore, countermeasures are to be used in such a way as to ‘permit the resumption of performance of the obligations in question’. This has been considered to amount to an obligation to “limit conduct to means that are reversible as far as possible”, though this principle is not absolute.

[L41] The virus that State B has used against the computer in State A’s Foreign Office offered little chance of any reversibility, as its purpose was to make the computers unusable and spread quickly for maximum impact. The irreversibility suggests that the purpose was not to compel State A to comply with the legal consequences of the cyber operation, but rather to retaliate in response.

[L42] Overall, the response by State B would likely be considered as exceeding the bounds of permissible countermeasures under international law.

Checklist

 * Flag State jurisdiction
 * Has there been an unlawful interference by a foreign State with a vessel under the jurisdiction of another?
 * Did any of the exceptions permitting the foreign State to exercise its jurisdiction exist at the time?
 * Freedom of navigation
 * Has the cyber operation impacted the navigation of a foreign vessel?
 * Sovereign immunity
 * Has there been an interference with a foreign vessel that enjoys sovereign immunity?
 * Is there an international armed conflict in existence that would remove the sovereign immunity of the vessel?
 * Maritime law enforcement or use of force
 * Does an “objective evaluation” of the circumstances suggest the operation qualified as law enforcement, or as a use of force?
 * If law enforcement, have the standards for using forcible means in a law enforcement operation been adhered to?
 * If a potential use of force, has the cyber operation reached the threshold necessary for classification as a prohibited use of force under Article 2(4) of the UN Charter?
 * Countermeasures
 * Is the injured State responding to a prior internationally wrongful act of the responsible State?
 * Would the injured State’s conduct taken in response amount to an internationally wrongful act if not justified as a countermeasure?
 * Do the measures taken in response by the injured State meet the conditions prescribed for the lawful resort to countermeasures under international law?

Bibliography and further reading

 * Russell Buchan ‘Cyber Attacks: Unlawful Uses of Force or Prohibited Interventions?’ (2012) 17 Journal of Conflict and Security Law 212.


 * Gary P. Corn and Eric T. Jensen, ‘The Use of Force and Cyber Countermeasures’ (2018) 32 Temple International and Comparative Law Journal 127.


 * Cordula Droege, ‘Get Off My Cloud: Cyber Warfare, International Humanitarian Law, and the Protection of Civilians’ (2012) 94 IRRC 533.


 * Robin Geiss and Henning Lahmann, ‘Freedom and Security in Cyberspace: Shifting the Focus Away from Military Responses Towards Non-Forcible Countermeasures and Collective Threat-Prevention’ in Katharina Ziolkowski (ed), Peacetime Regime for State Activities in Cyberspace (NATO CCD COE 2013).


 * Terry D. Gill, ‘Non-intervention in the Cyber Context’ in Katharina Ziolkowski (ed), Peacetime Regime for State Activities in Cyberspace (NATO CCD COE 2013).


 * Douglas Guilfoyle, ‘The High Seas’ (2015) in Donald Rothwell, Alex Oude Elfernik, Karen Scott and Tim Stephens (eds), The Oxford Handbook of the Law of the Sea (OUP 2015).


 * Wolff Heintschel von Heinegg, ‘Warships’ in Rüdiger Wolfrum (ed), Max Planck Encyclopedia of Public International Law (OUP 2008, updated October 2015).


 * Albert J. Hoffmann, ‘Freedom of Navigation’ in Rüdiger Wolfrum (ed), Max Planck Encyclopedia of Public International Law (OUP 2008, updated April 2011).


 * Maziar Jamnejad and Michael Wood, ‘Prohibition of Intervention’ (2009) 22 Leiden Journal of International Law 345.


 * Natalie Klein, ‘Maritime Security’ in Donald Rothwell, Alex Oude Elfernik, Karen Scott and Tim Stephens, The Oxford Handbook of the Law of the Sea (OUP 2015).


 * Natalie Klein, Maritime Security and the Law of the Sea (OUP 2012).


 * Doris König, ‘Flag of Ships’ in Rüdiger Wolfrum (ed), Max Planck Encyclopedia of Public International Law (OUP 2008, updated April 2009).


 * Philip Kunig, ‘Prohibition of Intervention’, in Rüdiger Wolfrum (ed), Max Planck Encyclopedia of Public International Law (OUP 2008, updated April 2008).


 * Patricia Kwast, ‘Maritime Law Enforcement and the Use of Force: Reflections on the Categorisation of Forcible Action at Sea in the Light of the Guyana/Suriname Award’ (2008) 13 Journal of Conflict and Security Law 1.


 * Michael N Schmitt, ‘Below the Threshold Cyber Operations: The Countermeasures Response Option and International Law’ (2014) 54 Virginia Journal of International Law 697.


 * Michael N Schmitt, ‘Computer Network Attack and Use of Force in International Law: Thoughts on a Normative Framework’ (1999) 37 Columbia Journal of Transnational Law 885.


 * Yoshifumi Tanaka, ‘Navigational Rights and Freedoms’ in Donald Rothwell, Alex Oude Elfernik, Karen Scott and Tim Stephens (eds), The Oxford Handbook of the Law of the Sea (OUP 2015).


 * Katja Ziegler, ‘Domaine Réservé’, in Rüdiger Wolfrum (ed), Max Planck Encyclopedia of Public International Law (OUP 2008, updated April 2013).

Contributions

 * Scenario by: Matt Kuningas
 * Analysis by: Matt Kuningas
 * Reviewed by: Jeffrey Biller, Robert McLaughlin