“License to Kill” in Salisbury: State-sponsored assassinations and the jus ad bellum

Above: U.K. Ambassador to the U.N. Jonathan Allen speaks at an urgent meeting of the Security Council on the recent nerve agent attack in Salisbury, U.K. on March 14, 2018. (Spencer Platt/Getty Images)

Ever since Sergei Skripal, a former Russian spy turned defector, and his daughter Yulia were found unconscious on a bench in Salisbury on March 4, the world’s attention has turned to this little-known city in Wiltshire, England. Investigations by British officials suggest that Skripal and his daughter – both of whom remain in critical condition – were deliberately poisoned with “a military-grade nerve agent of a type developed by Russia,” known as “Novichok.” Several individuals who may have been in contact with the substance were briefly hospitalized. One police officer remains seriously ill.

The Salisbury attack is believed to be the latest in a string of assassination attempts allegedly orchestrated by Russia in recent years. The most famous precedent concerns the murder of that other FSB defector Aleksandr Litvinenko, who died in London in 2006 after being poisoned with radioactive polonium. As in 2006, Moscow has denied any involvement.

The Salisbury incident has put further strain on the relations between London and Moscow, which were already seen as being at an “all-time low.” In an address to the House of Commons on March 13, Prime Minister Theresa May set an ultimatum, calling upon Moscow to account for how the Russian-produced nerve agent could have been deployed in Salisbury and to provide full disclosure of the “Novichok” program to the relevant international body, the Organization for the Prohibition of Chemical Weapons. The ultimatum expired without a positive response from Russia.

Interestingly, May has asserted that, absent a satisfactory explanation from Russia, the action is regarded as an “unlawful use of force by the Russian State against the United Kingdom” (see here and here, and see also the statement by the British UN ambassador during the emergency meeting here). The UK has also referred to an “attack” on British soil – without, however, speaking of an “armed attack” (see here). The same is true for the statement issued by the North Atlantic Council (see here).

The different statements have spurred a flurry of reactions on social media, with several scholars joining the fray, and essentially discussing three main issues:

(1) whether the Salisbury attack might qualify as a “use of force” in the sense of Article 2(4) UN Charter (as May’s statement would suggest);

(2) whether it would simultaneously qualify as an “armed attack” under Art. 51 UN Charter; and

(3) if an armed attack, whether the necessity and proportionality criteria leave room for a forcible response in self-defense.

Nothing suggests that the UK is considering any forcible measures against Russia in response to the attack. Instead, London has announced various non-forcible measures, including the expulsion of 23 Russian diplomats, which hitherto seem to fit the label of mere “unfriendly” retorsions (note: the permissibility of UK countermeasures has been discussed by Ashley Deeks over at Lawfare). Given the improbability of a forcible British reaction, the questions identified above are mostly of a theoretical interest – notwithstanding the added stigma that admittedly comes with a State being held responsible for an unlawful recourse to force as opposed to, for instance, a breach of another State’s territorial integrity or the violation of an individual’s right to life. Still, May’s explicit reference to a “use of force” against the UK makes for an interesting precedent – one that invites us to test the incident against the jus ad bellum framework.

Assassination (attempts) on foreign soil as a “use of force”?

Can targeted killings of a single individual on foreign soil qualify as a “use of force” in the sense of Article 2(4) UN Charter? If one were to follow the interpretation of the Independent International Fact-Finding Mission on the Conflict in Georgia (IIFFMCG), established by the Council of the European Union, the answer must surely be negative. In its report of 2009, the IIFFMCG effectively takes the view that “very small incidents” do not possess the required gravity to qualify as uses of force (at 242). Examples of such incidents specifically include “the targeted killing of single individuals” (note 49). The position of the IIFMCG has been most forcefully defended by Olivier Corten, who finds that targeted operations remain beyond the scope of the prohibition on the use of force if they (1) remain small-scale; (2) do not result in direct confrontations with troops of the territorial state, and; (3) are not directed against the territorial state (in the sense that there is no intention to use force against the territorial state itself) (at 55, 85). According to Corten, this position “undoubtedly reflects customary international law” (ibid., at 856). O’Connell has supported a similar approach (here, at 102). In response to the speech by Theresa May, O’Connell cautioned on Twitter that “calling a crime… a use of force” constitutes a “dangerous departure from the law.”

But is it really? Authors arguing that Art. 2(4) UN Charter is subject to a de minimis threshold have emphasized that States do not generally employ jus ad bellum terminology when condemning (small-scale) targeted killings and the like. That is of course correct. Recent illustrations such as the aforementioned murder of Litvinenko, or the assassination with a VX nerve agent of Kim Jong-Il’s half-brother Kim Jong-nam in a Malaysian airport early 2017, confirm this general picture. Still, as I have explained elsewhere, there are several arguments that dispel the idea of a genuine gravity threshold for “uses of force.”

First, at a conceptual level, the deliberate projection of lethal force into the territory of another (non-consenting) State is perfectly capable of qualifying as a use of force “in the international relations” between States. This is so, even if the targeted persons are private individuals, rather than state organs, and even if no actual damage is done to state infrastructure. Second, States that refrain from publicly condemning targeted killings on their soil as a “use of force” may have various reasons for doing so, including a desire not avoid diplomatic escalation, or to deny the presence of known terrorists on their soil. The implication is that it is difficult to deduce from these instances a genuine opinio juris in support of a de minimis threshold. Third, States have oftentimes treated similar incidents in jus ad bellum terms. By way of illustration, the assassination of PLO member Khalil al-Wazir by Israeli commandos in Tunis in 1988 was clearly regarded as a breach of Article 2(4) UN Charter (see SC Res. 611(1988)), as was the (foiled) plan to assassinate then-former President George H.W. Bush in Kuwaiti territory in 1993 (see the UNSC debate here)). It is equally telling that  the United States frequently invokes the right of self-defense to justify targeted killings outside “areas of active hostilities.” The idea of a gravity threshold for “uses of force” moreover sits uneasily with the position of the International Committee of the Red Cross that there is “no requirement of a specific level of intensity of violence to trigger an international armed conflict” (see the 2017 Commentary to Common Article 2 of the Geneva Conventions, at § 265). Finally, from a policy-oriented perspective, the paradoxical effect of introducing a gravity threshold for Art. 2(4) is that it would be easier to justify small-scale forcible acts, for instance, as “countermeasures.” Is this truly where we want to go?

In the end, the better view is that any deliberate projection of lethal force onto the territory of another state can trigger Article 2(4) of the UN Charter. When tested against this mold, the Salisbury incident (if imputable to Russia) does amount to a “use of force.” Either way, even those favoring the IIFMCG position must acknowledge that the speech by Theresa May constitutes an important fragment of State practice contradicting the existence of any gravity threshold.

The Salisbury incident as an “armed attack”?

The more difficult question is whether the Salisbury incident can simultaneously be regarded as an “armed attack” under Article 51 UN Charter – language that PM May has refrained from using. I must confess being somewhat surprised to see several Twitter messages by expert colleagues affirming the existence of an “armed attack” without much hesitation. Inasmuch as one holds the view that the “killing of a single individual” is insufficiently grave to amount to a “use of force” (see supra), it follows a fortiori that such an act cannot amount to an “armed attack” triggering the right of self-defense. Yet even if the Salisbury attack is regarded – correctly, in the present author’s view – as a “use of force,” the traditional interpretation of Article 51 of the UN Charter nonetheless assumes that a “gap” exists between the “use of force” and an “armed attack,” and that only the “most grave” uses of force trigger Article 51. This is the position that was famously set forth by the International Court of Justice (ICJ) in the Nicaragua case (at § 191) and reaffirmed by the Court subsequently. While not uncontested, it probably remains the majority position in legal doctrine today. And while the ICJ has acknowledged that the gravity threshold should not be set too high, the poisoning of a private individual on foreign soil is of course far removed from “the mining of a single military vessel” (i.e., the example alluded to in the Oil Platforms judgment (at § 72)).

Against this background, what options are left to construe the Salisbury incident as an “armed attack”? As suggested above, some simply reject the gap between “use of force” and “armed attack” – a position also embraced in the 2016 Department of Defense Law of War Manual (at para. and the State Department’s Legal Adviser, on behalf of the U.S. government, in 2012. As I have argued elsewhere, there is indeed considerable support for the view that isolated hostile encounters between military units or small-scale territorial incursions by foreign troops can qualify as “an armed attack” (in spite of the ICJ’s categorical –and much-criticized- exclusion of “mere frontier incidents”), allowing for limited on-the-spot reaction in self-defense (see Armed Attack, at section 3.3).

Whether that means the ICJ’s gravity requirement can also be set aside in other scenarios such as the present one is, however, far less obvious. The present author is not aware of any precedents where the targeted killing of a single individual gave rise to a claim of self-defense on the part of the territorial State. This is so in spite of the fact that assassination attempts in foreign territory are hardly a recent invention of the Kremlin. Examples of similar operations go back much further in time. Apart from the KGB/FSB, other institutions such as the CIA or Mossad have their own established track record in the domain (remember those Iranian nuclear scientists killings a few years ago?). To my knowledge, such operations have never given rise to an appeal to, much less actual recourse to, self-defense (although the underlying reasons may of course be manifold). What is more, the same appears to be true for attacks against political leaders or diplomatic envoys abroad (see Armed Attack, at section 3.4.2). Attacks against individual diplomats, for instance, have taken place on numerous occasions, but have generally been framed as “crimes” punishable before a court of law, rather than as “armed attacks,” warranting the exercise of self-defense (note, however, that these attacks have mostly been carried out by non-State armed groups). And while Israel did draw attention to a number of attacks against its diplomats to justify its military intervention in Lebanon in 1982, these attacks formed part of a larger list of terrorist attacks which Israel relied upon. Several States, including the UK, moreover dismissed the idea that the assassination of an ambassador could in any way justify attacks against Lebanese towns and villages (see here, at 3).

A rare example of an assassination plot giving rise to action justified on the basis of the right of self-defense pertains to the US military strikes against the Iraqi intelligence headquarters in Baghdad in 1993 after the discovery of the  above mentioned plans to murder then-former President George H.W. Bush. Interestingly, in the subsequent Security Council debate, many States expressed understanding for the strikes, with several – including nota bene Russia – confirming that the US acted in accordance with Article 51 UN Charter. At the same time, one should not lose sight of the unique features of the case. In the words of the representative of New Zealand, “[a]ny nation that seeks to assassinate the Head of State or a member of the senior political leadership of another State commits and act of aggression. Such actions are at the most serious end of the scale because Heads of State symbolize the sovereignty and integrity of their country” (see here, at 23).

The contrast with the Salisbury attack is obvious. Even if he was reportedly a British citizen at the time of the attack, Sergei Skripal was not an acting British official, let-alone a high-level official “symbolizing the sovereignty and integrity” of the United Kingdom. Skripal was presumably targeted because he was an FSB defector, not because he was somehow regarded as representing the UK. Accordingly, the Salisbury attack can hardly be construed as an attack against a manifestation of the British State, or an attack aimed at damaging British infrastructure. Or, to paraphrase the wording of the ICJ in the Oil Platforms judgment, there is no evidence that the attack “was aimed specifically at” the UK or “with the intention of harming” the UK (§ 64).  All things considered, even accepting that (large-scale) attacks against private individuals are not a priori excluded from the scope of Article 51 UN Charter, the Salisbury incident seems insufficiently grave of its own to be regarded as an “armed attack” in the sense of Article 51. The fact that, apart from Sergei Skripal and his daughter, a number of other persons were temporarily hospitalized seems insufficient to alter this conclusion.

One might object that the Salisbury attack was not an isolated incident, but rather part of a pattern of State-sponsored assassinations on British soil. Accordingly, it could be argued that, even if each single incident is insufficiently grave to be regarded as an armed attack, when taken cumulatively, the attacks do come within the scope of Article 51 of the UN Charter. This reasoning mirrors the so-called “accumulation of events” (or so-called Nadelstichtaktik theory), which holds that several minor attacks may be “accumulated” for purposes of assessing self-defense claims. The impact of the doctrine is twofold. On the one hand, it suggests that “less grave uses of force” can, when forming part of a chain of events, add up to an actual “armed attack.” On the other hand, the doctrine also suggests that the cumulative nature of the attacks must be taken into account to determine what reaction in self-defense would be “proportionate.”

The “accumulation of events” theory, while not uncontested, enjoys considerable support in legal doctrine, especially in relation to attacks by irregulars and armed bands. It has also on numerous occasions been invoked by States (see Armed Attack, at section 3.2.2.b). In the context of the 1964 Gulf of Tonkin incident, for instance, the UK representative in the Security Council stressed how there “[had] not been merely one isolated attack on United States warships in international waters,” but rather “repeated attacks, the nature of which is such as to indicate that they were deliberately mounted” (S/PV.1140, § 90). Furthermore, the ICJ seems to have implicitly endorsed the doctrine in a number of judgments, including the Nicaragua case (at § 231), the Oil Platforms case (§ 64) and the Armed Activities case (§ 147). At the same time, the doctrine is not without problems. It has frequently been invoked in the past in attempts to justify interventions that were manifestly disproportionate and in relation to cross-border actions by oppressive regimes against national liberation movements. In all, the “accumulation of events” theory should arguably be construed narrowly, as applying only to situations where consecutive attacks are linked in time, cause and source.

Whether in the present case the alleged pattern of State-sponsored assassinations is sufficiently regular and interlinked is open to debate. The Russian government is known to have passed legislation in 2006 giving its agents a license to engage in extra-judicial killing of terrorists and dissidents outside Russia (see e.g. here). Apart from the murder of Aleksandr Litvinenko later that same year (which the UK has blamed on Russia), however, it is unclear how many “State-sponsored” assassinations may have occurred on British soil. A 2017 BuzzFeed News investigation made reference to 14 suspicious deaths in UK territory which US intelligence services have linked to Russia. The same report nonetheless observes that British police “have ruled out foul play in every last case.” Some persons were deemed to have died of suicide, others of natural causes, etc. Even if British authorities may be re-opening some investigations with renewed scrutiny, it remains very difficult to determine that specific individuals were murdered, and that these murders can be imputed to the Russian State. And even assuming that some deaths can be linked to the Russian security services, it remains questionable whether a limited number of State-sponsored assassinations spread over a period of twelve years can qualify as a protracted campaign of interlinked attacks, so as to cumulatively give rise to an “armed attack” in the sense of Article 51 UN Charter.


A response with fire and fury?

Even assuming – for the sake of argument – that several assassinations can be imputed to the Russian Federation and that these can cumulatively be regarded as an “armed attack,” one final question remains: what forcible action could the UK possibly undertake in accordance with the UN Charter? Pursuant to the necessity and proportionality criteria, self-defense should in principle be limited to the halting and repelling of an armed attack. Yet, in the present case, even if Sergei Skripal and his daughter are still fighting for their lives, the “attack” is factually over. Does that mean the case for self-defense is now closed?

The answer to this question arguably hinges on whether the UK has compelling evidence that further attacks on British soil are imminent. If such evidence were to exist, this would arguably satisfy the necessity limb of self-defense. By contrast, absent such evidence, it would be hard to regard a forcible response as anything other than a (punitive) reprisal, which is prohibited under international law. That being said, the border between (lawful) self-defense and (unlawful) reprisals is not always easy to draw, and some operations more akin to a reprisal (such as the above mentioned 1993 US strikes against Baghdad) have at times been met with understanding, or even support.

In so far as the military option is not on the table, it is of course mere speculation to consider what action might plausibly constitute a “proportionate” response to a (string of) targeted assassination(s). By the UK’s own standard (see here, at 3), large-scale military strikes would unsurprisingly be excluded. By contrast, targeted forcible action against foreign agents involved in State-sponsored assassinations might pose fewer problems at least from a proportionality perspective.

In the end, notwithstanding the despicable nature of an assassination attempt on foreign soil, especially one using a chemical weapon posing a serious health risk to innocent bystanders, it remains very difficult to see how the Salisbury attack could qualify as an “armed attack” under the UN Charter. If Moscow is indeed behind the attack, this certainly merits a strong rebuke by the British authorities, but also by the broader international community. A further escalation of the row between Russia and the UK by a British invocation of Article 51, however, seems both politically unwise, and difficult to reconcile with the existing international legal framework. 

About the Author(s)

Tom Ruys

Professor of International Law at Ghent University, Founder of the Ghent Rolin-Jaequemyns International Law Institute (GRILI)