The International Law of Intelligence Sharing in Multinational Military Operations: Concluding Thoughts

Having canvassed the various conceptual questions of state complicity in the prior posts in the series, we can now return to the two basic intelligence sharing scenarios I outlined in my first post – the sharing of intelligence facilitating a wrongful act, and the receipt of intelligence that was unlawfully collected and/or shared. In both scenarios the causal nexus element can generally be met, but their resolution will depend on an assessment of the assisting state’s fault. Thus, under Article 16 ASR, an assisting state will be responsible for sharing intelligence with another state that facilitated the wrongful act of the latter state if the assisting state (i.e. its organs or agents) did so:

1. with the purpose/direct intent of facilitating the wrongful act, that is in a situation in which the facilitation of the wrongful act was the conscious object of the assisting state’s conduct, regardless of any deficits in the assisting state’s knowledge;
2. with the oblique/indirect intent of facilitating the wrongful act, that is while being practically certain that the assisted state will commit the wrongful act and practically certain that it will use the intelligence provided to do so, and consciously choosing to provide the intelligence despite this practical certainty;
3. with willful blindness, that is with the purpose of avoiding acquiring information but for which the assisting state would have acted with indirect intent.

The culpability element of Article 16 would apply to intelligence sharing that contributes to any violation of international law, regardless of the subject-matter, so long as the two states are bound by the same rule. This culpability element also forms the minimum baseline for regime-specific complicity rules, the fault requirements of which can only be looser, more expansive than those of Article 16. Therefore, the sharing of intelligence that assisted a partner’s violation of (say) a third state’s sovereignty or the principle of non-intervention, e.g. through a cyber operation interfering with elections, would trigger the assisting state’s responsibility under Article 16 if the assisting state acted with one of the modes of fault above.

As explained, the IHL and IHRL-specific complicity rules have the added benefit of applying to state assistance, including the sharing of intelligence, to non-state actors. For serious violations of IHL and IHRL they arguably have the further added benefit of employing risk-based forms of fault. That is, the state sharing intelligence would be liable under these complicity rules if it was aware of a degree of risk that its assistance would be used to facilitate a serious violation, such as torture, but consciously disregarded that risk and proceeded to provide the assistance nonetheless. The risk threshold would certainly be satisfied if it was more likely than not that the assisted partner would commit the wrong and would use the aid provided to do so. It would probably also be satisfied if the risk was below likelihood but was nonetheless serious, substantial, real, or clear.

The emergence of a clear pattern of ongoing violations by the assisted partner will both elevate the assisting state’s appreciation of the risk of future violations and be evidence that any mitigation measures are proving to be ineffective. Thus, for example, the Intelligence and Security Committee of the British Parliament found that between 2001 and 2010 there were ‘232 cases recorded where it appears that UK personnel continued to supply questions or intelligence to liaison services after they knew or suspected (or, in [in the Committee’s] view, should have suspected) that a detainee had been or was being mistreated.’ The Committee also found that UK intelligence agencies ‘also suggested, planned or agreed to rendition operations proposed by others in 28  cases [and there were] a  further  22  cases  where  SIS  or  MI5  provided  intelligence  to  enable  a  rendition  operation  to  take  place.’ In each of these cases specific intelligence facilitated a specific wrongful act. While it is possible that when individual decisions were being made, especially in the early stages of the ‘war on terror,’ British officials did not know to practical certainty or even appreciated a degree of risk that the intelligence would facilitate a wrongful act by third parties, such as the United States, or hoped that mitigation measures would reduce that risk, that notion becomes impossible to sustain after a pattern of behavior emerges over time.

As for the second scenario, involving the receipt of unlawfully obtained and shared intelligence, it is a question of fact whether the receiving state is aware of some degree of risk that the information was shared or obtained unlawfully – in real life the users of, for example, a fused intelligence product may not actually have any idea about the sourcing of a given item of information. That said, even if the requisite degree of knowledge is there, this scenario does not in principle engage the general complicity rule in Article 16 ASR, which only applies to prospective assistance to an ongoing or future wrongful act. The provision of assistance after the wrongful act was expressly excluded by the ILC from the scope of Article 16; a fortiori, the mere passive receipt of information unlawfully obtained cannot violate that rule either because it does not actually facilitate the commission of the wrong, which has already been completed.

Article 16 could be engaged, however, if the factual pattern was such that B was not a mere passive recipient of unlawfully obtained information, but actually provided assistance to A’s wrongful act of unlawfully collecting and/or sharing intelligence or actively solicited that act. This could for instance be the case in situations of continuing cooperative relationships in which each party provides the other with some quid pro quo – B could be providing money, its own intelligence or other assistance to A, while A would continue sharing its intelligence in a revolving cycle of interactions. If the underlying violation was a serious breach of jus cogens, this could also potentially engage the rule in Article 41(2) ASR, which prohibits states from assisting the maintenance of a situation created by such a breach.

Similarly, a pattern of receiving information from a state that acquired it in violation of international law can evolve into an indirect form of prohibited assistance to that state. As noted by the UK Parliament’s Joint Committee on Human Rights in the context of information procured through torture:

[I]f the Government engaged in an arrangement with a country that was known to torture in a widespread way and turned a blind eye to what was going on, systematically receiving and/or relying on the information but not physically participating in the torture, that might well cross the line into complicity. Systematic, regular receipt of information obtained under torture is in our view capable of amounting to “aid or assistance” in maintaining the situation created by other States’ serious breaches of the peremptory norm prohibiting torture. As a number of witnesses to our inquiry put it, the practice creates a market for the information produced by torture. As such, it encourages States which systematically torture to continue to do so.

Thus, especially in situations of longer-term partnerships and those in which a pattern can be discerned in the behavior of the state or non-state actor sharing the intelligence, e.g. of use of torture, the receiving state may no longer be regarded as merely a passive recipient of tainted information. Rather, it may be seen as encouraging or actively assisting the production and sharing of such information, including by creating a market for such violations. Consider, for example, the UK’s receipt of intelligence from third states procured by the mistreatment of detainees in the ‘war on terror;’ Parliament’s Intelligence and Security Committee found ‘198 cases recorded where UK personnel received intelligence from liaison services obtained from detainees whom they knew had been mistreated, or with no indication as to how the detainee had been treated but where, in our view, they should have suspected mistreatment.’ Such situations can, depending on the facts, fall within the scope of Articles 16 and 41 ASR and of the regime-specific complicity rules.

Finally, the actual use of the unlawfully obtained and shared information would generally not taint some operation of the recipient state that would otherwise be lawful. For instance, if B used information obtained through the use of torture by A to target an individual in a drone strike, and in fact that individual turned to be a combatant and the strike was only directed against a military target, there would be no violation of IHL as a consequence of the strike regardless of the fact that B relied on information obtained through torture. Similarly, on an IHRL analysis, if a state used lethal force against an individual who in fact posed an imminent threat to others and there was no other way to neutralize the threat, so that the use of force would be compatible with the right to life, the lawfulness of the use of force would not depend on the fact that some of the relevant information pointing to the threat was obtained unlawfully and shared with the recipient state – if, again, the information was objectively accurate. On the contrary, the rules of both IHL and IHRL may positively require resort to such intelligence, regardless of the fact that it was collected and shared unlawfully. If, say, the receiving state was put on notice through the shared tainted information, e.g. data obtained in violation of the right to privacy, that there were civilians in an area in which it intended to pursue a strike, ignoring that information would put the receiving state in violation of its duty to take all feasible precautions in attack and verify that all targets are military ones.

In sum, as we have seen, the sharing and receiving of intelligence may either be contrary as such to some rule of international law (e.g. the human right to privacy, or a bespoke treaty regime), or may trigger state responsibility on the basis of complicity in a partner’s violation of international law. The complicity scenarios pose the greatest set of difficulties.

The complicity rules of international law are underdetermined. But as I have tried to show we nonetheless possess an ample toolbox with which to address complicity scenarios, including those involving the sharing of intelligence in the course of multi-partner military operations. Article 16 ASR is the most powerful tool in that toolbox because of its general applicability and its usefulness as a normative baseline for other, regime-specific complicity rules. The variegated fault element I argue for is normatively justifiable and practicable, allows us to make sense of the internally conflicted ILC work product, and enables us to move beyond the false dilemma between intent and knowledge that has burdened much of the literature so far.

Regime-specific complicity rules coexist with the generally applicable ones and can be of great practical import. Both IHL and IHRL contain such rules, which can apply to State assistance in multinational military operations. The existence of the IHL rule is not in doubt. Again, we are speaking here of the negative duty of restraint – not to encourage or facilitate violations of IHL – a duty with which any State can and should comply, and not about the much broader positive obligation to exert influence on third parties to put an end to their violations, which remains controversial. Even if some of its elements remain indeterminate, the IHL-specific complicity rule applies in all circumstances, regardless of whether the assisting State is involved in the armed conflict in which violations of IHL might occur, and regardless of the legal classification of that conflict. As for the IHRL rule, it is in the process of consolidation, but its potential impact should not be underestimated.

Therefore, while the web of legal rules that apply in complicity scenarios, such as those of intelligence sharing in military operations, is a complex one, the conceptual and practical difficulties in interpreting, applying, and developing these rules are not insurmountable. As with all legal rules, their practical implementation will depend primarily on a gradual process of norm internalization and the building up of a culture of compliance for which capacity-building efforts within governments are indispensable, as are external and internal pressure, advocacy, and litigation. That said, the reality of coalition operations is that the relevant partners may be subjected to different treaties and mechanisms of outside scrutiny. But it is inevitable that the greater constraints to which some of the states involved may be subject will have spillover effects on their partners, who will in practice have to take them into account. As far as legal risks for partners are concerned, domestic and international litigation under IHRL will produce the most acute such risks, a tendency that will only grow in the future.

This post is part of a series; the other essays in the series are available here: Part I, Part II, Part III.