The United States and Allies Sharing Intelligence with Ukraine

A remarkable feature of the ongoing Russo-Ukrainian conflict is the strategic and tactical role played by the sharing of intelligence between the US (and other Western allies) and Ukraine. That intelligence sharing started even before the war, in support of claims made publicly and privately about an imminent Russian attack against Ukraine (on which see more here). But it has continued since, including at an operational or tactical level, and has only gained in pace, as shown by recent reporting by NBC News and the New York Times.

In the initial stages of the conflict, the sharing of intelligence with Ukraine appeared to be constrained by one legal concern – the United States did not want to become a party to the international armed conflict between Russia and Ukraine by providing direct targeting support. An NBC report dated 4 March reported a senior US congressman as stating that:

We want to support the Ukrainians in every way we possibly can, without going to war with Russia … When it comes to intel-sharing and targeting, that’s a fine line. …We’re not doing [intelligence for real-time targeting], because that steps over the line to making us participating in the war. So the Pentagon is really struggling and walking that very fine line.

Other lawmakers made the same point, by way of criticism or not, but the US National Security Adviser and the Pentagon spokesperson denied that intelligence sharing was held up by legal concerns. Nonetheless, on 13 April NBC reported that then, after more than a month later:

The U.S. is also sharing more and better intelligence with the Ukrainians, the senior defense official and an administration official said. Last week, Defense Secretary Lloyd Austin issued new guidance to the intelligence community about what it can provide with respect to the Donbas. The protocols for providing intelligence needed to be updated for the re-prioritized fight in the Donbas, the senior defense official said, explaining that “the guidance was loosened up a bit.”

On 16 April, a new NBC report reiterated the narrative that there was a loosening of legal and policy constraints and a consequent ramping up of intelligence sharing:

Once the invasion got underway, lawyers in the U.S. defense and intelligence bureaucracy imposed guidance that in some cases limited the sharing of targeting information that could enable lethal Ukrainian strikes against Russians. But as Russia’s aggression has deepened, and under pressure from Congress, all of those impediments have been removed, officials say.

Earlier this month, for example, the director of National Intelligence withdrew and replaced a memo that prohibited intelligence sharing for the purposes of regaining captured territory or aiding Ukrainian strikes in Crimea or the Donbas, officials said.

The two reports reference two new memoranda – one by the Defence Secretary and the other by the DNI – that resulted in some shift of policy, but obviously these are not public. Yet the same report noted that even at the early stages of the invasion the US was providing Ukraine with actionable intelligence for both defensive and offensive purposes. For example

the U.S. gave Ukrainian forces detailed intelligence about exactly when and where Russian missiles and bombs were intended to strike, prompting Ukraine to move air defenses and aircraft out of harm’s way, current and former U.S. officials told NBC News.

That near real-time intelligence-sharing also paved the way for Ukraine to shoot down a Russian transport plane carrying hundreds of troops in the early days of the war, the officials say, helping repel a Russian assault on a key airport near Kyiv.

The intelligence shared also included specific information about threats to the life of President Zelensky.

And on 4 May the New York Times published a piece reporting that ‘U.S. intelligence support to the Ukrainians has had a decisive effect on the battlefield, confirming targets identified by the Ukrainian military and pointing it to new targets.’ In particular,

The United States has provided intelligence about Russian units that has allowed Ukrainians to target and kill many of the Russian generals who have died in action in the Ukraine war, according to senior American officials. … Ukrainian officials said they have killed approximately 12 generals on the front lines, a number that has astonished military analysts. The targeting help is part of a classified effort by the Biden administration to provide real-time battlefield intelligence to Ukraine. That intelligence also includes anticipated Russian troop movements gleaned from recent American assessments of Moscow’s secret battle plan for the fighting in the Donbas region of eastern Ukraine, the officials said. Officials declined to specify how many generals had been killed as a result of U.S. assistance.

The United States has focused on providing the location and other details about the Russian military’s mobile headquarters, which relocate frequently. Ukrainian officials have combined that geographic information with their own intelligence — including intercepted communications that alert the Ukrainian military to the presence of senior Russian officers — to conduct artillery strikes and other attacks that have killed Russian officers.

According to the New York Times, the US is not the only state sharing actionable intelligence with Ukraine, although it is probably the largest contributor – ‘Other NATO allies also give real-time intelligence to the Ukrainian military.’ The Times piece ends with a quote from a statement of the Chairman of the Joint Chief of Staff at a Senate hearing, where he said that ‘There’s a significant amount of intelligence flowing to Ukraine from the United States … We have opened up the pipes.’

The reporting by the Times was not exactly welcomed by the administration. After the piece was published, ‘a National Security Council spokeswoman, said in a statement that the battlefield intelligence was not provided to the Ukrainians “with the intent to kill Russian generals.”’ A Pentagon spokesperson similarly pushed back against the Times reporting by stating that

We do not provide intelligence on the location of senior military leaders on the battlefield or participate in the targeting decisions of the Ukrainian military … [The Ukrainians] make their own decisions, and they take their own actions.

Finally, on 5 May both NBC News and the New York Times reported that the US shared intelligence with Ukraine that helped it sink the Russian flagship Moskva in the Black Sea. Accounts differ as to whether the US intelligence only confirmed the Moskva’s location and other information Ukraine already had, or was of greater import and impact. Still, according to NBC, the ‘U.S. did not know in advance that Ukraine was going to target the Moskva, officials said, and was not involved in the decision to strike.’

So, what are we to make of these intelligence sharing arrangements from the standpoint of international law? Readers will recall that I had previously dealt with some of the relevant issues in my article on complicity and intelligence sharing in International Law Studies, and in a series of posts on EJIL: Talk! (here, here, here and here).

The most prominent question here – that so obviously informs the US intelligence-sharing practices – is whether the sharing of actionable/battlefield intelligence can somehow make the US a participant in the Russia-Ukraine IAC. This is a pure issue of IHL, and a complex one. We could see from the overall narrative arc above that intelligence sharing gradually expanded, and it seems that there was some shift in the internal US evaluation of the law here, but likely more one of degree than one of kind. Yet the concern to avoid becoming a belligerent in the conflict remains a pressing one – note in particular how with respect to the most recent reported incidents (the targeting of Russian generals and the sinking of the Moskva) US officials insist that they do not participate in the targeting decisions of the Ukrainian forces and that the Ukrainians are combining the shared information with their own intelligence to inform their decisions.

Some of these disclaimers appear forced. It may well be true that the US was not providing ‘intelligence on the location of senior military leaders on the battlefield,’ i.e. telling Ukraine that Russian general X or Y was in such and such location, but that the US was nonetheless informing Ukraine about the location of Russian mobile headquarters in the field, in which the generals just happened to be located. It may well be true that the US did not intend the Russian generals to be killed, but it was certainly aware of a real possibility – even a very high level of probability – that this will happen. It may well be true that the US did not know with certainty that Ukraine would launch missiles at the Moskva, i.e. no Ukrainian officer told their US liaisons that they would do so, but it does not require a leap of the imagination to conclude that confirming to the Ukrainian forces the location of a key enemy vessel would very possibly lead to an imminent strike against it. And so on – the leaks to the press from some US officials are being mitigated by others, publicly maintaining a degree of plausible deniability, as it were. (For a more detailed discussion of state knowledge and intention evoked by some of the terminology used in reported statements, see here).

The line that the US is drawing here – supplying intelligence v. being involved in targeting decisions – is a sensible one from the standpoint of assessing whether the US is participating in the armed conflict. If supplying a drone or a howitzer to Ukraine does not make the assisting state a participant in the conflict, then surely supplying information can’t cross that threshold either. But actually making decisions or specific suggestions on what targets should be struck, when and how, is a material step further. So while the line the US is attempting to navigate is a fine one, from what we know now it doesn’t seem to have been crossed.

On the general issue of participation in the conflict by states assisting Ukraine, I would direct readers to this excellent post by Alexander Wentker. There clearly are legal consequences attached to establishing participation. But I would suggest that these consequences are, practically speaking, of much less relevance here than the optics and considerations of policy. It would simply be a substantial escalatory step for the US, or one of its allies, to legally be seen as a participant in the Russia-Ukraine IAC. This is what the US administration seeks to avoid. And it’s no wonder therefore that intelligence sharing will, in practical terms, be constrained by this concern.

The second question is whether the sharing of intelligence by the US and other states with Ukraine could somehow be legally wrongful, i.e. violate some obligation under international law that would prohibit such sharing. It was precisely scenarios such as these that were the focus of my ILS piece and the blog series.

As I explained then, there are two basic ways in which intelligence sharing (or the receipt of shared intelligence) can be contrary to international law. First, it can be unlawful as such, in the sense that a rule of international law may specifically prohibit the sharing of intelligence with a partner, regardless of how the partner intends to use that intelligence. Second, the sharing or receiving of intelligence may be unlawful not because they are prohibited as such, but because they become prohibited due to their connection with an unlawful act of a partner. In such cases intelligence sharing is a form of complicity in the partner’s wrongful act, which the shared intelligence facilitates.

Bearing in mind the nature of the information being shared, such as the location of military assets (the collection of which would not, in principle, implicate the human right to privacy), the only rule that could conceivably prohibit the sharing of intelligence in this context would be the duty of neutrality. Thus, sharing actionable/battlefield intelligence could be inconsistent with a third state’s obligation to remain neutral between the belligerents of the Russia/Ukraine IAC. But for reasons extensively examined in various essays on the Ukraine crisis, neutrality today cannot reasonably be interpreted as categorically prohibiting this form of assistance, just like it could not be interpreted as requiring third states to refrain from supplying Ukraine with weapons. This is either because, in the post-UN Charter world, neutrality remains only in an attenuated form, where third states are in fact allowed to assist a victim of aggression (which is what Ukraine is), or because collective self-defence would operate as a circumstance precluding the wrongfulness of any breach of neutrality.

As for issues complicity, as things stand the sharing of intelligence is not wrongful simply because it is not assisting any wrongful act by Ukraine. Ukraine is committing no wrong by killing Russian combatants, including generals, or destroying enemy vessels. This would be so even if (contrary to disclaimers by US officials) the information shared pertained to the location of specific individuals, such as high-ranking military commanders. Targeting these individuals with lethal force is at least in principle lawful under applicable international humanitarian law and human rights law. Facilitating such actions – which the sharing of intelligence undoubtedly does – is therefore not wrongful either. It could become wrongful only if Ukraine started attacking civilians or committing other violations of international law facilitated by shared intelligence, and there is no evidence of such wrongs being committed. That said, the duty to respect and ensure respect for IHL (and, in my view, similar duties under human rights law), do require the US to assess the risk of facilitating potential wrongful acts by Ukraine before intelligence is shared.