[Yulia Ioffe is a DPhil (PhD) candidate and a tutor in Public International Law at University of Oxford and Olga Bozhenko is an LL.M student in International Litigation (Public International Law) at Institute of International Relations Kyiv National Taras Shevchenko University.]
In February 2018, President Trump issued his Nuclear Posture Review (NPR), which was widely anticipated by the international community. It was hoped that this document would soften the US President’s previous aggressive rhetoric (including the threat of ‘total destruction of North Korea’) and bring the US administration’s position within the orbit of international law. It is thus important to understand how the NPR captures the US current stance on the possibility of using nuclear weapons in self-defense where an armed attack has not yet occurred, particularly in light of deteriorating relations between the US and North Korea, as well as Russia’s nuclear weapons modernization efforts. We argue that the NPR is not compliant with international law standards on self-defense.
The NPR: Brief Overview
The NPR is a legislatively-mandated review, produced by the Secretary of Defense, that establishes US nuclear policy, strategy, capabilities, and force posture for the next five to ten years. Apart from being an important policy document for the US domestically, the NPR is also reflective of the most pressing issues of the universal nuclear weapons agenda. From international law perspective, it may amount to evidence of state practice as an act of executive branch (p 22) with regard to the use of force and nuclear weapons, particularly considering that some would call the US a specially affected state (p 312), as a state possessing them.
At the face value, it may seem questionable whether the NPR extends the ‘first use’ strategy right up to anticipatory self-defense. Yet, the range of scenarios potentially triggering such a first use of nuclear weapons by the US is indicative of a foreseeable possibility to use force not only in response but also in anticipation.
In fact, one can hardly imagine a more all-encompassing provision to include the option of anticipatory strike than the one envisaging the ‘employment of nuclear weapons in extreme circumstances to defend the vital interests of the United States, its allies, and partners’ (p 21), as laid down in the NPR. It is only reasonable to read this provision as covering anticipatory self-defense in view of the following clarification that the mentioned ‘extreme circumstances’ could include (i.e. are in no way limited to) significant non-nuclear strategic arms attacks.
The NPR is essentially based upon the strategy of deterrence (which Judge Oda considered ‘legitimized by international law’ in his Dissenting Opinion to the Nuclear Weapons Advisory Opinion (para 45)), presuming nuclear weapons to be used efficiently so long as they are not actually used. Still, the NPR seems to go further than that by extending the aims of retaining nuclear capability to the ‘achievement of the US objectives if deterrence fails’ (p 20) and peculiarly to ‘hedging against prospective and unanticipated risks’ (p 24). The latter includes the intention to ‘reduce potential adversaries’ confidence that they can gain advantage through expansion of nuclear capabilities’ (p 24), and to ‘respond to the possible shocks of a changing threat environment’ (p 14). This could definitely be interpreted as a provision deliberately formulated in the vaguest manner possible to create a leeway with respect to nuclear strike options, considering the context of the NPR adoption.
The current NPR, as well as Obama’s, reserves the right for the US to use nuclear weapons against non-nuclear weapon states, which are not in compliance with their non-proliferation obligations. Still, while the Obama version envisaged biological and chemical weapons development as additional scenarios to trigger nuclear response to non-nuclear threat, Trump’s NPR reserves a much broader room for action by referring to ‘the evolution and proliferation of non-nuclear strategic attack technologies’ (p 21). By including yet another option to respond by nuclear strike to conventional weapons development, the document waters down previous attempts to limit the scenarios giving rise for such an option to the potential nuclear threats only. Furthermore, such options seem to imply that there is no specific case of anticipatory self-defense against weapons of mass destruction (WMD), assimilating it within the ‘regular’ self-defense framework, where the type of weapons does not actually matter.
Assuming that we accept that international law permits anticipatory action in self-defense, it is a prevailing view (p 663) that such action is only lawful when an armed attack is imminent and the response is necessary and proportionate.
The concept of ‘extreme circumstances’ used in the NPR can be assumed to serve a bridge to the well-known Caroline formula, which is still a yardstick for measuring the legality of anticipatory self-defense. It is argued (p 11) that the standard of a threat being ‘instant, overwhelming, leaving no choice of means, and no moment for deliberation’ is too rigorous to apply for a state faced with a nuclear-related threat, particularly, the one emanating from a rogue actor. By applying such a standard, a state can end up with missing the last opportunity to avert the damage. However, when one takes into account the mode of response, which is the use of nuclear weapons, the threshold set by the Caroline formula seems justified, requiring, at the very least, to establish an undisputable need to defend state’s ‘vital interests’.
In this context, the very use of the concept of ‘vital interests’ should be viewed as problematic, having regard to the standard of ‘the very survival of a state’, which the ICJ considered as possibly justifying self-defense by means of nuclear weapons (Nuclear Weapons Advisory Opinion, para 97). It was only Judge Guillaume (para 8), who extended this standard by claiming that no legal system could deprive a state from the possibility to defend its ‘vital interests’(discussed here p 314), yet further as well referring to the ‘ultimate means’ to ‘guarantee survival’. Hence, as compared to the ICJ’s position, the NPR uses an arguably lower standard of ‘vital interests’ and, at the same time, does not limit it to the situation of an actual armed conflict.
The standard of necessity of self-defense sets a high threshold, prescribing that ‘there must be no practical alternative to the proposed use of force that is likely to be effective in ending or averting the attack’. This was confirmed, although implicitly, by the ICJ in the Oil Platforms case (para 76).
Understandably, the standard is considerably higher in respect of nuclear weapons, requiring the situation to be of the last resort kind, i.e. warranting the whole range of defensive options to be on the table. To this end, the NPR mentions that the US will attempt to stick to the strategy of the minimum destruction possible, since ‘U.S. nuclear policy for decades has consistently included this objective of limiting damage if deterrence fails’ (p 23).
Yet, even this strategy will not meet the necessity requirement, unless the US is able to establish the anticipatory nuclear strike as the only option available with no room for peaceful measures, or, at least, for conventional attack, including for the purposes of defending its vital interests or those of its allies and partners.
If it is the scale of anticipated attack and that of preventive action which must be comparable to satisfy the proportionality requirement, one can hardly imagine a conventional armed attack, however imminent, which is comparable to a nuclear strike. According to another interpretation of proportionality, supported by Judge Higgins in her Dissenting Opinion to the Nuclear Weapons Advisory Opinion (para 5), which adopts a ‘means-end proportionality’ approach with the focus on restoring state’s defensive capacity, it is possible for a nuclear strike to be lawful, even if not commensurate to anticipated attack. To this end, the NPR implicitly assumes that a limited nuclear strike could be less destructive than a large scale conventional response. The employment of tactical nuclear weapons, which allow for the precise targeting, could be justified as a measure to halt the imminent attack, which, if occurring, would require much more destructive a response.
It leaves us though with the last, often overlooked, aspect of proportionality, which relates to the extent anticipatory action remains lawful, even if it was such at the beginning (similar situation was discussed here concerning the proportionality standard for the ‘total destruction of North Korea’ as mentioned by President Trump). Perhaps it is safe to assert that any anticipatory military action, including nuclear strike, is lawful only as long as it remains defensive in nature and ceases as soon as the threat in question is eliminated. Undoubtedly, this can hardly mean that the elimination of a state is at issue. Otherwise, it is rather the case of promoting offensive strategy, but not of restoring defensive capacity.
Sources of Anticipated Attack
The NPR defines a broad range of the potential sources of the anticipated attack, with the states being the main suspects. The NPR directs special attention to North Korea and Iran calling them a ‘clear and grave threat’ (p 32) and a ‘serious concern’ (p 71) respectively, Russia which would face ‘unacceptably dire costs’ (p 30) in case of aggression, and China which challenges ‘traditional US military superiority in the Western Pacific’ (p 1). Additionally, the US continues to insist on its controversial legal position to justify self-defense, including in the context of nuclear weapons, in response to efforts by non-state actors. The NPR clearly states that a terrorist nuclear attack against the US or its allies and partners would qualify as an ‘extreme circumstance’ and any state that supports or enables terrorist efforts will experience the ‘ultimate form of retaliation’ (p 68). Thus, the NPR implicitly confirms the US’s longstanding ‘unwilling or unable’ doctrine, allowing to respond with the nuclear strike against a terrorist attack even where there is no ‘host state’ involvement. In fact, the legal evaluation of attacks by non-state actors as constituting an armed attack (let alone an anticipated attack) in the context of self-defense is still controversial (paras 14–21, 23). We are also not convinced that international law as it stands supports the ‘unwilling or unable test’ in relation to self-defense actions taken against non-state actors (discussed here and here), especially in the context of nuclear weapons.
Beyond the criticism domestically (see here and here), the NPR has attracted serious criticism from a number of states. The German government condemned ‘a spiral of a new nuclear arms race’ introduced by the NPR. The Chinese government criticized the NPR for its ‘Cold War mentality’. Russia’s Ministry of Foreign Affairs sharply criticized the NPR (see here in Russian), in particular, for lowering the threshold of the use of nuclear weapons and allowing the use of nuclear weapons in ‘extreme circumstances’, which are not limited to military scenarios and with military scenarios covering almost any use of military force. President Putin’s recent display of the new Russian nuclear weapons could also have been a reaction to the NPR. Both Russia and China emphasized that the NPR contrasts sharply with their own nuclear posture, in which they adhere to the policy of no-first-use of nuclear weapons, non-use of nuclear weapons against non-nuclear-weapon states, and where nuclear weapons are considered a strictly defensive deterrent.
The NPR, as it currently stands, is hardly in compliance with international law, since it includes a number of remote strategic inconveniences threatening the US interests (arising not only from states but also non-state actors) as triggers for the first use of nuclear weapons, effectively lowering the threshold for the use of nuclear weapons. Such use would also most likely be incompatible with the necessity and proportionality requirements.