Scenario 06: Cyber countermeasures against an enabling State

From International cyber law: interactive toolkit
Jump to navigation Jump to search
© NicoElNino. Licensed from Shutterstock.

A country believed to possess highly developed cyber capabilities repeatedly fails to assist other States in countering cyber attacks emanating from its territory. After yet another malicious cyber operation from the former State’s territory results in numerous casualties abroad, the said State comes under a large-scale DDoS attack. The scenario considers the international obligation of due diligence in the cyber context and the ability of States to take countermeasures in response to violations of that obligation.

1 Scenario[edit | edit source]

1.1 Keywords[edit | edit source]

Countermeasures, critical infrastructure, due diligence, international co-operation, non-State actors

1.2 Facts[edit | edit source]

[F1] State A is generally considered to possess advanced cyber capabilities, including detection and incident reaction capabilities. However, it has come under a growing criticism that it does not do enough to thwart malicious cyber activities that originate from or transit through its territory. Several States have made public statements in international fora complaining that their requests for assistance submitted to State A’s authorities had been ignored. Nonetheless, State A’s public response has been to maintain that it had attended to all reasonable requests received from other States.

[F2] Following an aircraft crash in State B involving scores of casualties, caused apparently by a fault in the air traffic control system of State B’s receiving airport, its domestic forensic investigation concludes that the fault had been caused by a premeditated cyber operation originating from the territory of State A and that it cannot be excluded that the operation will be repeated. State B’s relevant authorities submit a request for an urgent assistance to State A, asking that more information be provided on specific actors and that the infrastructure identified in the request be investigated and command and control servers shut down immediately.

[F3] Several days later, the foreign ministry of State B issues a statement claiming that no response has been received from State A to the request for assistance. The statement further says that it considers the matter to be one of utmost importance for its national security and expresses grave concerns about impact on mutual relations. On the same day, the foreign ministry of State A responds with a public statement noting that while it cannot comment on any ongoing investigation, its general policy is to attend to all reasonable requests for assistance received from other States. It adds that in any event its domestic laws do not allow it to intervene in privately owned cyber infrastructure.

[F4] One week after this exchange of statements, hundreds of State A’s government servers come under sustained large-scale distributed denial-of-service (DDoS) attacks. As a result, many public services ordinarily provided to the population of State A by online means become unavailable, including the system of e-health prescriptions, lodging of e-tax returns, and e-ticketing on public transport across the country.

1.3 Examples[edit | edit source]

2 Legal analysis[edit | edit source]

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

[L1] The legal analysis in this scenario first considers whether State A’s failure to respond to State B’s request for assistance qualifies as a violation of its due diligence obligations under customary international law or any specific treaty obligations related to interference with air traffic control. It then examines whether the measures State B took in response can be qualified as lawful countermeasures against State A.

2.1 Due diligence[edit | edit source]

Due diligence
According to the traditional formulation by the ICJ in the Corfu Channel case, every State is under an “obligation not to allow knowingly its territory to be used for acts contrary to the rights of other States”.[1] In the cyber context, the UN General Assembly urged States already in 2000 to “ensure that their laws and practice eliminate safe havens for those who criminally misuse information technologies”.[2]

It is the matter of some controversy whether the principle of due diligence reflects a binding obligation applicable to cyber operations.[3] It has also been proposed that in the cyber context, it is preferable to construe due diligence as a standard of attribution rather than as a standalone primary rule of international law.[4] Nevertheless, the present analysis proceeds on the basis that as a matter of lex lata, due diligence constitutes a general international obligation for every State not to knowingly allow its territory to be used for internationally wrongful acts using cyber means. This view has also been unanimously endorsed by the drafters of the Tallinn Manual 2.0.[5]

Due diligence does not entail a duty of prevention,[6] but rather an obligation of conduct.[7] A State breaches its due diligence obligation in the presence of the following cumulative elements:

  1. The existence of acts (by a non-State actor or a third State) contrary to the rights of a victim State,[8]
  2. which are conducted from or through the territory of the potentially responsible State (or from or through the territory or cyber infrastructure under its control),[9]
  3. which would have been unlawful if conducted by the potentially responsible State,[10]
  4. which have serious adverse consequences for the victim State,[11]
  5. with respect to which the potentially responsible State has actual or constructive knowledge,[12] and
  6. upon which the potentially responsible State can act, but fails to take all feasible measures.[13]

[L2] In the present scenario, all the elements seem to have been met. To begin with, the cyber operation against State B has resulted in the deaths of many individuals and significant destruction in that State’s territory. As such, the operation was manifestly contrary to the rights of State B (condition 1).

[L3] The indicators of compromise identified by the authorities of State B suggest that cyber infrastructure on the territory of State A had been used to mount the cyber operation in question. This information has not been opposed in any way by the territorial State. The acts in question should thus be seen as conducted from or at least through the territory of State A (condition 2).

[L4] Air traffic control systems usually form part of a State’s critical infrastructure and their proper functioning can be considered as pertaining to national security as well as part of that State’s international obligations in case of international aerial transportation. There is no doubt that a cyber operation which disables air traffic control systems, prevents the State from fulfilling its international obligations, and results in lethal effects, thus impacting on national security, constitutes a breach of international law.[14] Therefore, had the acts and omissions in question attributed to State A, they would manifestly have been internationally unlawful (condition 3).

[L5] Similarly, there is little doubt that an operation that results in scores of civilian deaths should be seen as entailing “serious adverse consequences” for the victim State. This criterion may be somewhat difficult to meet with respect to certain “below the threshold” operations. However, it is generally accepted that “operations that would comprise, were they to be directly undertaken by a State, ‘over the threshold’ of international law rules on use of force” do trigger the due diligence obligations of the potentially responsible State[15] (condition 4).

[L6] The next question is one of knowledge. It is unclear whether State A had actual knowledge of the forthcoming hostile operation from its territory in the first place. However, there are strong indications that it had at least constructive knowledge of the operation. In particular, its possession of advanced detection capabilities combined with its ignorance of repeated States’ requests for assistance suggest that State A should have known of the operation in question or that it had turned a blind eye to it. Hypothetically, if the attackers had used, at least in part, State A’s governmental cyber infrastructure, it would be easier to claim State A’s knowledge.[16] Nonetheless, by the time State B’s relevant authorities had submitted the request for assistance to State A, the knowledge element (condition 5) would likely have been met.[17]

[L7] Finally, it appears that State A has failed to take any measures to stop the hostile operation emanating from its territory. It may well be the case that State A’s domestic law proscribed it from acting against the threat, as its foreign ministry implied when it noted that State A’s authorities may not intervene in privately owned cyber infrastructure. However, this would be of no relevance as to the characterization of State A’s conduct under international law,[18] given that “domestic legal limitations do not justify a State’s failure to comply with its due diligence obligation”.[19] Similarly, State A would not be assisted by a reference to its general policy to respond to other States’ requests. In this regard, the pattern of ignoring requests apparent from prior public statements by other countries would amount to an indication that this policy is actually ineffective in practice. In sum, State A’s failure to take any specific steps designed to put an end to the hostile operations against State B strongly suggests that it had failed to act in response to the acts contrary to the rights of State B (condition 6).

[L8] In addition to these general considerations relating to the obligation of due diligence which applies as a matter of customary international law, specific States may have additional treaty obligations in similar situations of interference with air traffic control. This is particularly the case with respect to those States that are parties to the 1971 Montreal Convention.[20] The Convention prescribes that a State which has “reason to believe” that an offence of this kind will be committed has a duty to furnish any relevant information to the potential victim State.[21] Moreover, once the offence has been committed, States have an obligation to “afford one another the greatest measure of assistance” in the resulting criminal proceedings.[22] Accordingly, if States A and B both were parties to the Montreal Convention, then the failure of State A to warn and later to assist State B could additionally violate the former’s relevant treaty obligations.

2.2 Countermeasures in response to a violation of due diligence[edit | edit source]

Countermeasures
Countermeasures are “measures that would otherwise be contrary to the international obligations of an injured State vis-à-vis the responsible State, if they were not taken by the former in response to an internationally wrongful act by the latter in order to procure cessation and reparation”.[23]

An injured State may only take countermeasures against the responsible State if the following conditions are met:

  1. The existence of a prior internationally wrongful act of the responsible State against the injured State;[24]
  2. The internationally wrongful act has not ceased and “the dispute is [not] pending before a court or tribunal which has the authority to make decisions binding on the parties”, unless the “responsible State fails to implement the dispute settlement procedures in good faith”;[25]
  3. The injured State has called upon[26] the responsible State to fulfil its obligations arising from its internationally wrongful act;[27] and
  4. The injured State has notified the responsible State of its decision to take countermeasures, and offered to negotiate with that State, unless it is taking “urgent countermeasures as are necessary to preserve its rights.”[28]

Additionally, the countermeasures must fulfil the following requirements:


  1. Their aim must be to induce the responsible State to comply with the legal consequences of its internationally wrongful act;[29] in other words, the aim of countermeasures is restoration, not retribution; and the countermeasures can only target the responsible State;
  2. They “shall, as far as possible, be taken in such a way as to permit the resumption of performance of the obligations in question;”[30]
  3. They shall not affect “obligations under peremptory norms of general international law”[31], obligations under any dispute settlement procedure between the injured and responsible State, and obligations arising from the inviolability of diplomatic or consular agents, premises, archives and documents;[32] and
  4. They must be “commensurate with the injury suffered, taking into account the gravity” of the prior unlawful act and of the rights in question (i.e. the requirement of “proportionality”).[33]

Taken countermeasures must be suspended if the internationally wrongful act has ceased or if “the dispute is pending before a court or tribunal which has the authority to make decisions binding on the parties”, and they must be terminated as soon as the responsible State has complied with its (secondary) obligations.

Only the injured State may take countermeasures, unless there has been a violation of a norm of erga omnes status. Other States can only take other lawful measures.[34]

Whether a particular measure fulfils these conditions is an objective question,[35] while the burden of proof that the relevant conditions have been fulfilled falls on the injured State.[36] The exact standard of proof required is unsettled in international law and it will depend on the relevant forum. However, relevant international jurisprudence tends to rely in this regard on the standard of “clear and convincing evidence”.[37] This standard translates in practice into a duty to “convince the arbiter in question that it is substantially more likely than not that the factual claims that have been made are true.”[38] Importantly, if a State does resort to countermeasures on the basis of an unfounded assessment that a breach has occurred, it may incur responsibility for its own wrongful conduct.[39]

[L9] Assuming that due diligence is a binding international legal obligation that extends to the cyber context (a question that is analysed in the section on due diligence), an injured State may respond to the violation of this obligation by resorting to countermeasures against the responsible State.[40] Any such response must comply with the strict conditions that international law imposes on the lawful exercise of countermeasures.[41]

[L10] In the present case, whether or not the relevant criteria have been met would depend on the following considerations. At the outset, it should be noted that the analysis proceeds on the assumption that the DDoS attacks against State A’s cyber infrastructure were attributable to State B. It is also assumed that the act of launching the attacks could qualify as a breach of an international obligation of State B (for the analysis of a similar fact pattern, refer to Scenario 05). Without these preconditions, any assessment of countermeasures would be moot as there would be no relevant conduct by the injured State in question that could potentially qualify as a countermeasure.

[L11] Firstly, State A’s violation of its due diligence obligations amounts to a prior internationally wrongful act against State B as the injured State. Additionally, State A’s failure to act qualifies as a continuing internationally wrongful act for as long as it chooses not to take any specific steps designed to put an end to the hostile operations against State B.[42] This period of inaction by State A also delineates the temporary scope within which State B may, provided that the remaining conditions are met, resort to countermeasures.[43]

[L12] Secondly, the public statement by State B’s foreign ministry may be construed as a request on State A to cease its internationally wrongful conduct. It has been noted that the requirement of notification is “probably unrealistic” in the cyber context.[44] and it is the view of at least one State that prior notification need not be given if doing so would “expose highly sensitive capabilities in defending the country in the cyber arena”.[45]

[L13] Thirdly, for the DDoS attacks against State A’s infrastructure to be lawful, their object would have to have been strictly limited to inducing State A to comply with its international obligations. In particular, if State A did indeed provide the requested information to State B while the attacks were underway, State B would have to terminate the attacks given that the object of its countermeasure—the restoration of legality as between the two States—would have been achieved.[46] The aim to restore the legality on both sides implies that countermeasures should be, “as far as possible”,[47] reversible in nature;[48] this criterion is typically met by DDoS attacks such as those launched against State A in the present scenario.[49]

[L14] Fourthly, the DDoS attacks against State A would have to comply with the requirement of proportionality. To begin with, proportionality does not imply that the injured State’s response must be equivalent or reciprocal.[50] This is of crucial importance in the present case, as any active response to a failure to act will always by definition be non-reciprocal in nature.[51] In favour of proportionality, State B could argue that its action was commensurate with the injury suffered, taking into account the severe risk of further loss of life and significant damage that may result if future hostile operations emanating from State A’s territory are not thwarted. By contrast, the impact of the DDoS attacks on services wholly unrelated to State A’s unwillingness to thwart malicious cyber operations would militate against the proportionality of the supposed countermeasures. Ultimately, the assessment of proportionality is highly contextual and would depend on the specific circumstances of the case at hand.

3 Checklist[edit | edit source]

  • Due diligence:
    • Did the relevant cyber operation adversely affect the rights of a State?
    • Was the cyber operation conducted from or through the territory of another State?
    • Would the cyber operation have been unlawful if conducted by the potentially responsible State?
    • Did the incident result in serious adverse consequences for the victim State?
    • Did the potentially responsible State know of the incident in question? Alternatively, even if it only became aware of the incident through the information from the injured State, should it have known of it at the material time?
    • Did the potentially responsible State take all feasible measures to put an end to the malicious cyber activities?
  • 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?

4 Appendixes[edit | edit source]

4.1 See also[edit | edit source]

4.2 Notes and references[edit | edit source]

  1. Corfu Channel Case (UK v Albania) (Merits) [1949] ICJ Rep 4, 22.
  2. UN GA Res 55/63 (4 December 2000), Doc A/RES/55/63, para 1(a).
  3. Cf. UN GGE 2015 report, paras 13(c) and 28(e) (using non-mandatory language to express the due diligence principle in the cyber context: “States should not knowingly allow their territory to be used for internationally wrongful acts using [cyber means]” and “States ... should seek to ensure that their territory is not used by non-State actors to commit such acts”, respectively) (emphases added).
  4. See Luke Chircop, ‘A Due Diligence Standard of Attribution in Cyberspace’ (2018) 67 ICLQ 643.
  5. Tallinn Manual 2.0, commentary to rule 6, para 4.
  6. Tallinn Manual 2.0, commentary to rule 6, para 5.
  7. Cf. Case Concerning Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v Serbia and Montenegro) (Judgement) [2007] ICJ Rep 43, para 430; see further James Crawford, State Responsibility: The General Part (CUP 2013) 226–32 (on the distinction between due diligence and obligations of prevention); Rudiger Wolfrum, ‘Obligation of Result Versus Obligation of Conduct: Some Thoughts About the Implementation of International Obligations’ in Mahnoush H Arsanjani et al, Looking to the Future: Essays on International Law in Honor of Michael Reisman (Brill 2010).
  8. Corfu Channel judgment, para 22; Tallinn Manual 2.0, commentary to rule 6, para 2 and 15.
  9. Tallinn Manual 2.0, rule 6.
  10. Tallinn Manual 2.0, commentary to rule 6, para 18-24.
  11. Tallinn Manual 2.0, rule 6.
  12. Tallinn Manual 2.0, commentary to rule 6, para 37-42.
  13. Tallinn Manual 2.0, commentary to rule 6, para 43; commentary to rule 7, para 2 and 18.
  14. See, eg, UK, Statement of Attorney General Jeremy Wright QC MP, “Cyber and International Law in the 21st Century” (23 May 2018) (“If it would be a breach of international law to bomb an air traffic control tower with the effect of downing civilian aircraft, then it will be a breach of international law to use a hostile cyber operation to disable air traffic control systems which results in the same, ultimately lethal, effects.”).
  15. Dan Efrony and Yuval Shany, “A Rule Book on the Shelf? Tallinn Manual 2.0 on Cyberoperations and Subsequent State Practice” (2018) 112 American Journal of International Law 583, 55.
  16. Cf. Tallinn Manual 2.0, commentary to rule 6, para 40 (“[W]hen a State’s governmental cyber infrastructure is being exploited by another State or non-State actor for an operation, the ‘should have known’ standard is more likely to be met than in the case of the use of private infrastructure”).
  17. Cf. Tallinn Manual 2.0, commentary to rule 6, para 37 (noting that a State will be considered as having actual knowledge “if it has received credible information that a harmful cyber operation is underway from its territory”).
  18. See, ILC Articles on State Responsibility, Commentary, Part 1 Chapter 1 on Art 3, para 1 (“An act of a State must be characterized as internationally wrongful if it constitutes a breach of an international obligation, even if the act does not contravene the State’s internal law—even if, under that law, the State was actually bound to act in that way.”); see also ibid, para. 8 (“States cannot use their internal law as a means of escaping international responsibility.”)
  19. Tallinn Manual 2.0, commentary to rule 7, para 22.
  20. Convention for the Suppression of Unlawful Acts Against the Safety of Civil Aviation (signed 23 September 1971, entered into force 26 January 1973) 974 UNTS 178 (“Montreal Convention”).
  21. Montreal Convention, Art 12.
  22. Montreal Convention, Art 11(1).
  23. ILC Articles on State Responsibility, Commentary, part 3 ch 2 at para 1.
  24. ILC Articles on State Responsibility, Art 49(1); Case Concerning the Gabčíkovo-Nagymaros Project (Hungary/Slovakia) (Judgment) [1997] ICJ Rep 7, para 83.
  25. ILC Articles on State Responsibility, Art 52(3) – 52(4).
  26. ILC Articles on State Responsibility, Art 52(1)(a). According to the UK Attorney General, the UK does not feel legally obliged, when taking countermeasures in response to a covert cyber intrusion, to “give prior notification to the hostile state”. UK Attorney General, Jeremy Wright QC MP, ‘Cyber and International Law in the 21st Century’.
  27. ILC Articles on State Responsibility, Art 28-41; the list of consequences includes (i) continued duty of performance, (ii) cessation and non-repetition, (iii) reparation, and (iv) particular consequences of a serious breach of obligations under peremptory norms of general international law.
  28. ILC Articles on State Responsibility, Art 52(1)(b) – 52(2).
  29. ILC Articles on State Responsibility, Art 49(1); Case Concerning the Gabčíkovo-Nagymaros Project (Hungary/Slovakia) (Judgment) [1997] ICJ Rep 7, para 87. The list of consequences in Art 28-41 includes (i) continued duty of performance, (ii) cessation and non-repetition, (iii) reparation, and (iv) particular consequences of a serious breach of obligations under peremptory norms of general international law.
  30. ILC Articles on State Responsibility, Art 49(3).
  31. Such as the obligation to refrain from the threat or use of force as embodied in the UN Charter, obligations for the protection of fundamental human rights, and obligations of a humanitarian character prohibiting reprisals. ILC Articles on State Responsibility, Art 50(1).
  32. ILC Articles on State Responsibility, Art 50(2).
  33. Articles on State Responsibility, Art 51; Case Concerning the Gabčíkovo-Nagymaros Project (Hungary/Slovakia) (Judgment) [1997] ICJ Rep 7, para 85.
  34. ILC Articles on State Responsibility, Art 54.
  35. ILC Articles on State Responsibility, Commentary in Part 3, Chapter 2 on Art 49, para 3.
  36. ILC Articles on State Responsibility, Commentary to Part One, Chapter 5, para 8 (noting that “[i]n a bilateral dispute over State responsibility, the onus of establishing responsibility lies in principle on the claimant State”).
  37. See, eg, Trail Smelter case (United States v Canada) (Award) (1941) 3 RIAA 1905, 1965; see also 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) 624 (noting that in cases where State responsibility is involved, the required threshold tends to shift towards ‘clear and convincing’”).
  38. James Green, ‘Fluctuating Evidentiary Standards for Self-Defence in the International Court of Justice’ (2009) 58 ICLQ 163, 167 (emphasis original).
  39. ILC Articles on State Responsibility, Commentary in Part 3, Chapter 2 on Art 49, para 3.
  40. Cf. Tallinn Manual 2.0, commentary to rule 7, para 28 (“If a State is unwilling to terminate harmful cyber operations encompassed by the due diligence principle as opposed to unable to do so, the injured State may be entitled to resort to countermeasures ... based on the territorial State’s failure to comply with [its due diligence obligations].”).
  41. See ILC Articles on State Responsibility, Arts 49–53.
  42. Cf. ILC Articles on State Responsibility, Commentary in Part 2, Chapter 1 on Art 30, para 3 (noting that the notion of a continuing wrongful act extends to “situations where a State has violated an obligation on a series of occasions, implying the possibility of further repetitions”).
  43. Cf. ILC Articles on State Responsibility, Commentary in Part 3, Chapter 2 on Art 53, para 2 (“countermeasures must be terminated as soon as the conditions which justified them have ceased”).
  44. Marco Roscini, Cyber Operations and the Use of Force in International Law (OUP 2014) 106
  45. UK Attorney General, Jeremy Wright QC MP, 'Cyber and International Law in the 21st Century'.
  46. Cf. ILC Articles on State Responsibility, Commentary in Part 3, Chapter 2 on Art 49, para 7 (noting that the aim of countermeasures is “the restoration of a condition of legality as between the injured State and the responsible State”); see also 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) 638 (“the remedy of resorting to countermeasures is strictly instrumental, and not permitted as a means to retaliate. As a consequence, countermeasures are not allowed once the unlawful act has ceased”).
  47. ILC Articles on State Responsibility, commentary to Part Three, Chapter II, para 6.
  48. James Crawford, State Responsibility: The General Part (CUP 2013) 688 (“the question of reversibility is intimately linked to the instrumental function of countermeasures, to the extent that it implies a reversion to legality on both sides after the objectives of cessation and reparation have been met”).
  49. Tallinn Manual 2.0, commentary to rule 21, para 8.
  50. ILC Articles on State Responsibility, part 3, ch 2, at para 5 (‘a limitation to reciprocal countermeasures assumes that the injured State will be in a position to impose the same or related measures as the responsible State, which may not be so’).
  51. See 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) 634–35.

4.3 Bibliography and further reading[edit | edit source]

4.4 Contributions[edit | edit source]

Previous: Scenario 05: Criminal investigation Next: Scenario 07: Hacking tools