Initial Thoughts on the UK Attorney General’s Self-Defence Speech

This is part of a series of posts discussing the UK Attorney General’s speech on the Modern Law of Self-Defence. See also the other posts in the series by Monica Hakimi and Marko Milanovic.

It’s a pleasure to be able to contribute to this EJIL:Talk! discussion of the speech this week by the UK Attorney General, Jeremy Wright QC MP, on “The Modern Law of Self-Defence”. There are two elements of the speech that strike me as especially notable, and on which I’d like to give my initial thoughts here: the invocation of the so-called ‘unwilling or unable’ test and, particularly, the meaning of ‘imminence’ in relation to anticipatory self-defence.

Unwilling or Unable

The explicit acceptance by the UK of the ‘unwilling or unable’ concept, while brief, is a conspicuous feature of the Attorney General’s speech. The speech roots itself in tradition, with nods to the power and weight of history (stretching right back to the 1795 Jay Treaty, as well as, of course, including the obligatory self-back-patting over Britain’s role in the end of the international slave trade). However, there’s no hiding the novelty of the UK’s acceptance of the hugely controversial notion of responding to armed attacks (actual or imminent) even in cases where there is no ‘host state’ involvement whatsoever, simply on the basis of the unwillingness or inability of the state to prevent a non-state actor attack. The US has espoused the ‘unwilling or unable’ doctrine for years, of course, but the UK has not, at least not explicitly.

Admittedly, the Attorney General’s speech is not the first British invocation of unwilling or unable. In November 2015, David Cameron, then Prime Minister, argued before Parliament that the UK’s action in Syria was justified because “the Assad regime is unwilling and/or unable to take action necessary to prevent ISIL’s continuing attack on Iraq” (as well as making the same assertion, the same week, in a memorandum to the Foreign Affairs Select Committee). However, these statements by the Prime Minister were the first clear articulations of the British acceptance of an unwilling or unable test, and were expressed very specifically in relation to action taken against ISIS in Syria. To my knowledge, the Attorney General’s speech acts as the first unequivocal confirmation that the UK has adopted unwilling or unable in genere. This is not a surprising fact, of course, but – to my mind – it is not a positive one either.

Put simply, and leaving aside policy, like Kevin Jon Heller (and many others) I remain unconvinced that state practice supports an unwilling or unable test in relation to self-defence actions taken against non-state actors (and, by unavoidable extension, the state(s) on/from which they are operating). The question of whether the law should allow for military action in such circumstances is a different matter: one that I will unapologetically sidestep. As the law stands, though, for my money, it does not.

The Meaning of Imminence

Although the Attorney General stated that the focus of his speech was “the law of self-defence, in particular against non-state actors”, and while the brief references to unwilling and unable are striking, the main substantive contribution of the speech, in my view, is its attempt to engage with the concept of imminence in relation to anticipatory action: an issue that, legally, is actually just as relevant to anticipatory action against states sensu stricto as it is to action against non-state actors.

While there are those who maintain that any form of anticipatory action is unlawful unless an armed attack has occurred [see, e.g., Olivier Corten, The Law Against War: The Prohibition on the Use of Force in Contemporary International Law (Hart, 2010), 407-411], it is fair to say that the majority of states and commentators now accept that anticipatory action may, exceptionally, be lawful if an armed attack is imminent. It is well-known that the UK takes this position: as the Attorney General states, “the long-standing UK view is that Article 51 of the UN Charter does not require a state passively to await an attack, but includes the ‘inherent right’…to use force in self-defence against an ‘imminent’ armed attack…”.

The problem with the conclusion – assuming that one accepts it – that international law allows for anticipatory action in self-defence in relation only to imminent attacks is that it tells us very little unless it is clear what is meant by imminence. This problem isn’t new, but it remains crucial: if imminence is the standard, then we need to know what it requires as a matter of law. As I have noted elsewhere [(2015) 2 Journal on the Use of Force and International Law 97, 104-105], there have been relatively few serious attempts to engage with not merely the need for imminence but also the meaning of imminence in relation to anticipatory self-defence. Having said that, one excellent recent consideration of this question appeared in Noam Lubell’s chapter in Marc Weller’s impressive Handbook [‘The Problem of Imminence in an Uncertain World’ in Marc Weller (ed), The Oxford Handbook of the Use of Force in International Law (OUP, 2015) 695]; another perspective on the issue can be found in Sir Daniel Bethlehem’s piece in the American Journal of International Law [(2012) 106 AJIL 769] (upon which the Attorney General relies heavily). Nonetheless, there remain comparatively few such attempts to put meat on imminence’s bones.

It is, therefore, commendable (and pleasing) that the Attorney General – on behalf of the UK Government more generally – has attempted to tackle this question explicitly, and to set out a position on it at a level of depth not previously done by the UK (or most other states). However, as Monica Hakimi has already noted in her contribution to this discussion of the Attorney General’s speech, “the devil is very much in the details here.” In essence, in approaching the question of what ‘imminence’ entails, the Attorney General repeats and endorses Sir Daniel’s 5 criteria for the (context-specific) evaluation of whether an attack is imminent. These criteria have been set out and discussed in both Monica’s post and that of Marko Milanovic, so I won’t repeat them again here.

Nonetheless, I will say that I agree with Marko that the restatement of these criteria does not meaningfully break new ground (at least in terms of the argument made, irrespective of whether one accepts it), and that I concur with both Monica and Marko that the criteria are unclear and open to interpretation (when has it ever been otherwise when it comes to interpretations of ad bellum standards?). In particular, the Bethlehem criteria potentially allow for the use of force against threats that are temporally remote, especially if the effects of the threat materialising would be significant (again, as discussed by Monica Hakimi). That seems rather too open-ended to me. Moreover, I find the Attorney General’s restatement of another claim made by Sir Daniel in relation to imminence in his AJIL piece especially troubling:

“[t]he absence of specific evidence of where an attack will take place or of the precise nature of an attack does not preclude a conclusion that an armed attack is imminent for purposes of the exercise of a right of self-defense, provided that there is a reasonable and objective basis for concluding that an armed attack is imminent.”

The Attorney General doesn’t merely repeat this statement, he is explicit that it both “reflects and draws upon what has been a settled position of successive British Governments” and “must be right”.

Personally, I think the Wright/Bethlehem position here goes further than other recent notable formulations of imminence (such as Lubell’s notion that the attack needs to be “specific and identifiable” [supra, 702], or the statement of the Georgia Commission that there must be “an objectively verifiable, concretely imminent attack” [Report of the Independent International Fact-Finding Mission on the Conflict in Georgia, vol I (2008) 254]). I find it difficult to reconcile the need for a specific, identifiable and concrete imminent attack with the argument that all that is required is a general finding that there will be an imminent attack of some kind, somewhere. As a policy matter, for me, the UK’s understanding of imminence as presented by the Attorney General allows for too much eye-of-the-beholder discretion and, thus, is open to abuse. And, as a legal matter, I’m not at all sure such a reading of imminence is supported by state practice (other than by the usual suspects that the Attorney General identifies – the US, Canada, Australia and New Zealand – and a handful of others). This, thus, feels to me more a statement of what the UK would like imminence to mean, not what it actually means.

Those who take a traditional position on the law governing self-defence (sometimes, somewhat reductively, called ‘restrictionists’) may well applaud the reaffirmation by the Attorney General of the UK’s long-maintained view that ‘non-imminent’/pre-emptive self-defence is unlawful. I certainly do. But, in practice, the implications of this position are contingent on the adopted understanding of imminence. The Attorney General is very careful to state that the UK’s “approach does not…in any way dispense with the concept of imminence.” However, it is worth recalling that the infamous US National Security Strategy of 2002 didn’t either, at least terminologically. While imminence was to all intents and purposes abandoned by the US’s ‘Bush Doctrine’, the 2002 National Security Strategy actually argued that the US “must adapt the concept of imminent threat”: the US didn’t claim that imminence wasn’t the standard, it just interpreted it in a way that went beyond all recognition.

While the interpretation of imminence presented by the Attorney General – based on Sir Daniel’s formulation – obviously doesn’t come close to the unrestricted credibility-bypass of the Bush Doctrine, and while I’m pleased that a genuine effort has been made by the UK Government to spell out what it sees imminence as entailing, I nonetheless have concerns about the wide scope of interpretation that it allows, especially in the context of ‘imminent’ attacks that cannot be pinpointed, even loosely, to a specific time or target.