Difference between revisions of "Scenario 06: Cyber countermeasures against an enabling State"
|Line 70:||Line 70:|
by: Kubo Mačák
Reviewed by: [TBC]
Reviewed by: [TBC]
Revision as of 08:29, 15 October 2018
A developed cyber power repeatedly fails to assist other States in countering cyber attacks emanating from its territory. After yet another malicious cyber operation from that State’s territory results in a significant loss of life abroad, the 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.
Due diligence, non-State actors, international co-operation, countermeasures
State A is generally considered to possess advanced cyber capabilities. However, it has come under a growing criticism that it does not do enough to thwart malicious cyber activities that originate in 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.
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 the result of a premeditated cyber operation originating in the territory of State A and that it cannot be excluded that the operation will be repeated. State B’s authorities, namely the governmental CERT followed by law enforcement organs, submit a request for 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.
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 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.
One week after this exchange of statements, hundreds of State A’s government servers come under sustained large-scale distributed denial-of-service (DDoS) attack. 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.
Similar real-world incidents
For a general overview of the structure of analysis in this section, see Note on the structure of articles.
| 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”.
It is the matter of some controversy whether the principle of due diligence reflects a binding obligation applicable to cyber operations. 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. 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.
In the present scenario, all of 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).
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).
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 hostile 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. Therefore, had the acts in question been conducted by State A, they would manifestly have been internationally unlawful (condition 3).
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 (condition 4).
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 status as a cyber power possessing advanced capabilities suggests that it should have known of the operation in question. 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. Nonetheless, by the time State B’s governmental CERT and law enforcement organs had submitted the request for assistance to State A, the knowledge element (condition 5) would likely have been met.
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, given that “domestic legal limitations do not justify a State’s failure to comply with its due diligence obligation”. 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 would be seen as having failed to act in response to the acts contrary to the rights of State B (condition 6).
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. 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. 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. 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 would additionally violate the former’s relevant treaty obligations.
Countermeasures in response to a violation of due diligence
An injured State may only take countermeasures against the responsible State if the following conditions are met:
Additionally, the countermeasures must fulfil the following requirements:
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.
Whether a particular measure fulfils these conditions is an objective question, while the burden of proof that the relevant conditions have been fulfilled falls on the injured State. 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”. 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.” 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.
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. Any such response must comply with the strict conditions that international law imposes on the lawful exercise of countermeasures.
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 attack against State A’s cyber infrastructure was attributable to State B. Without this precondition, any assessment of countermeasures would be moot as there would be no relevant conduct by the injured State in question that could qualify as a countermeasure.
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. 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.
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.
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. The aim to restore the legality on both sides implies that countermeasures should be, “as far as possible”, reversible in nature; this criterion is typically met by DDoS attacks such as those launched against State A in the present scenario.
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. 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. 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.
Notes and references
- Corfu Channel Case (UK v Albania) (Merits)  ICJ Rep 4, 22.
- UN GA Res 55/63 (4 December 2000), Doc A/RES/55/63, para 1(a).
- 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).
- See Luke Chircop, ‘A Due Diligence Standard of Attribution in Cyberspace’ (2018) 67 ICLQ 643.
- Tallinn Manual 2.0, commentary to rule 6, para 4.
- Tallinn Manual 2.0, commentary to rule 6, para 5.
- 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)  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).
- Corfu Channel judgment, para 22; Tallinn Manual 2.0, commentary to rule 6, para 2 and 15.
- Tallinn Manual 2.0, rule 6.
- Tallinn Manual 2.0, commentary to rule 6, para 18-24.
- Tallinn Manual 2.0, rule 6.
- Tallinn Manual 2.0, commentary to rule 6, para 37-42.
- Tallinn Manual 2.0, commentary to rule 6, para 43; commentary to rule 7, para 2 and 18.
- 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.”).
- D Efrony and Y Shany, “A Rule Book on the Shelf? Tallinn Manual 2.0 on Cyber Operations and Subsequent State Practice” (3 May 2018) Hebrew University of Jerusalem Legal Research Paper No. 18-22, p. 55.
- 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”).
- 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”).
- See, eg, Articles on State Responsibility, commentary to 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.”)
- Tallinn Manual 2.0, commentary to rule 7, para. 22.
- 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”).
- Montreal Convention, Art. 12.
- Montreal Convention, Art. 11(1).
- ILC Articles on State Responsibility, Commentary, part 3 ch 2 at para 1.
- ILC Articles on State Responsibility, Art 49(1); Case Concerning the Gabčíkovo-Nagymaros Project (Hungary/Slovakia) (Judgment)  ICJ Rep 7, para 83.
- ILC Articles on State Responsibility, Art 52(3) – 52(4).
- 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’.
- 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.
- ILC Articles on State Responsibility, Art 52(1)(b) – 52(2).
- ILC Articles on State Responsibility, Art 49(1); Case Concerning the Gabčíkovo-Nagymaros Project (Hungary/Slovakia) (Judgment)  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.
- ILC Articles on State Responsibility, Art 49(3).
- 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).
- ILC Articles on State Responsibility, Art 50(2).
- Articles on State Responsibility, Art 51; Case Concerning the Gabčíkovo-Nagymaros Project (Hungary/Slovakia) (Judgment)  ICJ Rep 7, para 85.
- ILC Articles on State Responsibility, Art 54.
- ILC Articles on State Responsibility, Commentary in Part 3, Chapter 2 on Art 49, para 3.
- 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”).
- 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’”).
- James Green, ‘Fluctuating Evidentiary Standards for Self-Defence in the International Court of Justice’ (2009) 58 ICLQ 163, 167 (emphasis original).
- ILC Articles on State Responsibility, Commentary in Part 3, Chapter 2 on Art 49, para 3.
- 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].”).
- See Articles on State Responsibility, Arts 49–53.
- Cf. Articles on State Responsibility, 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”).
- M Roscini, Cyber Operations and the Use of Force in International Law (OUP 2014) 106.
- Cf. Articles on State Responsibility, 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 R Geiss and H Lahmann, ‘Freedom and Security in Cyberspace: Shifting the Focus Away from Military Responses Towards Non-Forcible Countermeasures and Collective Threat-Prevention’ in K 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”).
- Articles on State Responsibility, commentary to Part Three, Chapter II, para. 6.
- J 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”).
- Tallinn Manual 2.0, commentary to rule 21, para. 8.
- Articles on State Responsibility, part III, 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’).
- See R Geiss and H Lahmann, ‘Freedom and Security in Cyberspace: Shifting the Focus Away from Military Responses Towards Non-Forcible Countermeasures and Collective Threat-Prevention’ in K Ziolkowski (ed), Peacetime Regime for State Activities in Cyberspace (NATO CCD COE 2013) 634–35.
Bibliography and further reading
- MN Schmitt (ed), Tallinn Manual 2.0 on the International Law Applicable to Cyber Operations (CUP 2017)