No First Use of Nuclear weapons, abbreviated to NOFUN or NFU (the latter is adopted in this paper), is on the face of it a simple idea, whose adoption is long overdue as a confidence and security building measure. When it comes to working out its implementation in such areas as force structures, and the readiness at which nuclear forces are held, a number of difficulties emerge. For one thing, as far as is known, no nation ever has configured its forces because of having adopted a policy of NFU. Rather, the experience has been that nations have declared a policy of NFU when they find themselves in a position of having no need to use nuclear weapons first in any likely circumstance. Thus nations have found themselves in a position, vis-à-vis any obvious enemy, where their forces were sufficient to absorb and defeat any plausible form of attack. A declaration of NFU might then confer some presentational advantage, perhaps in terms of gaining moral high ground.

Examples of NFU declaration

A number of examples come to mind. The Soviet Union, at the height of the Cold War, claimed adherence to a policy of NFU at a time when its preponderance of conventional forces on the ground made it plausible to believe that any attack launched by the Warsaw Pact against the West could succeed by the use of conventional forces only. The Soviet Union’s very substantial nuclear armoury was, on this supposition, being held back for use in retaliation only, if the West should initiate a nuclear exchange, as was indeed NATO’s policy under the doctrine of ‘flexible response’. (This is not, however, to beg two obvious questions. First: did the Soviet Union in fact ever countenance a policy of attacking the West in this way? It is no longer in doubt that military plans of this kind existed, but ample reason to suppose that NATO’s forces on the ground, backed by the equally vast armoury of US tactical and strategic weapons, must have sufficed to deter such an adventure in any remotely plausible circumstances. Secondly: if the Soviet Union nevertheless had felt compelled to attack, is there any good reason to suppose that it would have adhered to its own declared policy of NFU? The answer is plainly no. Indeed the military plans just referred to implied an all-out assault accompanied by nuclear tactical bombardment from the outset). The point is that the massive overhang of conventional force made a declaratory policy of NFU plausible, rather than that the forces were configured on the basis of an a priori decision by the Politburo to adhere to such a doctrine. Whether this declaration scored any points for the Soviet Union in the ideological struggle is very doubtful. When that struggle had plainly been lost, as a result of the collapse of the Warsaw Pact and of the Soviet Union itself, and the ratio of forces between Russia and the West in Europe had thus been reversed by force majeure, Russia duly abandoned its declared NFU policy.

A second example is India. Because her conventional forces outnumber those of Pakistan by a factor of some three to one (very similar, as it happens to the original Warsaw pact preponderance against NATO) she has seen fit to announce – since becoming an overt possessor of nuclear weapons – that she would never be the first to use them. In her battle for the favours of the west generally and the US in particular, this is no doubt seen as politically useful. The point is, once again, that the preponderance of conventional force, which has been in place for as long as the two countries have existed, is what makes NFU plausible as declaratory posture. It is not the case that Indian force planners have been obliged to go for such large forces in response to a political decision to adopt NFU as a practical war plan. Nor, once again, is there any good reason to suppose that a policy of NFU would be adhered to if – against all the odds – Pakistan was seen to be prevailing dangerously in an all-conventional conflict.

China is a third example. Against her traditional adversary India she possess more than sufficient conventional force to prevail, if she were determined to, in any border conflict and has demonstrated as much. (Against Taiwan she cannot at present field enough conventional forces, - air power, assault shipping - to mount a successful invasion. But any suggestion of using or threatening to use nuclear weapons to coerce Taiwan, even if that country were to go for a unilateral declaration of independence, would be rendered quite unreal by the near-certainty of a massive American response. American strategic nuclear forces outnumber China’s by at least a hundred to one). Here again it is the correlation of forces that enables a policy of NFU to be declared rather than Chinese purity of intention governing the way the Chinese armed forces are built up.

Similar considerations apply to the so-called Negative Security Assurances (NSAs) given by the United States, Russia, France and the UK in connection with the Nuclear Non-Proliferation Treaty (NPT). These amount to an undertaking not to use nuclear weapons against non-nuclear weapons states parties to the Treaty. Similar undertakings have been given to states parties to treaties establishing various Nuclear Weapon Free Zones (NWFZs). These assurances are purely declaratory and are hedged about in various ways. But the essential point is that they were given to states most unlikely to intend an attack or to find the means to carry one out successfully by conventional means against the states giving the Assurances. In no case can it be shown that the giving states have in any sense reconfigured their conventional or nuclear forces to match their NSA undertakings. The reverse is again the case. Because they have in the past had more than sufficient conventional forces to overwhelm any potential opponent among those to whom they were giving these guarantees they were able to give them essentially cost-free.

NFU in various shapes and sizes

Since real examples are lacking it is worth exploring some theoretical alternatives. At one extreme one might postulate a condition of existential or virtual NFU. In such a case a state would be self-deterred from using nuclear weapons, by simple fear of the consequences, whether or not it had made any such declaration, adopted such a policy, or even recognised doctrinally that this was in fact the case. The late Lord Carver, one-time British Chief of the Defence Staff and for long a proponent of a NFU posture for NATO, believed that whatever the policy of ‘flexible response’ might prescribe, in the event Britain, France and the US would never have been the first to use nuclear weapons against any invasion by the Warsaw Pact. The policy in fact required NATO forces to meet a conventional attack with conventional defence until such time as it was in imminent danger of being overwhelmed. At this stage it was envisaged that NATO would attempt to retrieve the military situation on the ground by using nuclear weapons in a measured and tactical manner in the first place, moving up to an all-out assault on key aspects of Soviet state power only when all else had failed. Substantial tactical or ‘theatre’ nuclear forces were provided for this purpose. Lord Carver believed that any such use on NATO’s part would in reality be self-deterred by fear of escalation. Confronted by the prospect of massive devastation of their countries, European leaders would seek an accommodation with the invader, preferring in the last resort to be ‘red’ rather than ‘dead’. Of course NATO could never admit this or its whole deterrent posture would have been undermined. But this, Carver believed, was the underlying truth. His argument has great force and he may well have been right. The same may, of course, have been true of the Russian leaders, in which case a genuine state of mutual existential NFU existed in those years. The point can never be proved one way or the other. It is clear, however, that such a condition of NFU could never be ‘implemented’ in the sense of force levels and readiness being adjusted in the light of it, since it could never be admitted to exist, let alone declared.

At the other extreme one can imagine a situation in which two adversaries, both in possession of nuclear weapons, were both prepared to adopt a policy of NFU, seriously intending to implement it, and to conclude a formal legally binding treaty to that effect. If for the moment we disregard the position of any third parties, then so long as the two parties stand by their word, no nuclear war could ever take place between them. This would indeed have consequences for force structures and planning. One of these would be that, since their nuclear weapons could never be used, it would be logical to dispose of them. In this case what we are really talking about is an agreement to do away with nuclear weapons altogether. This might well be desirable, but it would be better to address the goal directly rather than by stealth, using an NFU declaration as the Trojan Horse.

As a middle case one could imagine a nuclear weapon state, confronting a power of roughly equal size and also in possession of nuclear weapons, deciding as a deliberate act of policy to adopt a policy of NFU. It might hope that its adversary might reciprocate, in which case a formal treaty could be concluded with results along the lines just discussed. But it might also decide to go ahead on its own if necessary, perhaps for ethical reasons, or as a confidence building measure to foster a better atmosphere (a ‘sunshine’ policy), or on a simple calculation that it would make the world a safer place. Such a country would then indeed have to set about implementing the measures it judged necessary to make up for the loss of a nuclear first-use option. There are two obvious moves it would have to consider. One would be to increase the size of its conventional forces to the point where its potential enemy could no longer overwhelm it by conventional means. It is impossible to give any general indication of the size of increase that would be needed or its political implications, including the extra financial costs, and whether of not such a change would be generally beneficial. But an interesting instance of this process in reverse was provided by the build-up of NATO in the early years of the Cold War. In 1952 the North Atlantic Council at its meeting in Lisbon authorised a force goal of ninety six divisions. Of these some thirty five to forty were to be in the line and ready at all times, the rest were to be reservists. But the countries concerned could not afford to provide forces on this scale. From December 1954, accordingly, the North Atlantic Council authorised the Supreme Allied Commander Europe to base his plans on the use of atomic weapons from the outset. This was followed by a revision of the Lisbon goal to a figure of thirty divisions for the Central Front. There was no longer any idea that there could be a non-nuclear war in Europe. As Fred Mulley said: (The Politics of Western Defence, Thames and Hudson, London 1962, page 123): ‘A characteristic feature of the development of nuclear weapons, no doubt influenced to a considerable extent by the mounting cost of defence budgets … was the reduction in the size of standing armies’. In logic, therefore, one consequence of foregoing a nuclear first-use option would be to increase the size and/or technical sophistication of conventional forces, biting on the bullet of possibly much higher costs, with consequent penalties to spending on social goals and the greater militarisation of society (for example conscription). This might be a very bad bargain.

The other obvious way of setting oneself up for a NFU policy would be to adjust the status of one’s nuclear forces into a survivable second-strike mode. This might well involve increasing their numbers. It is no surprise that the nuclear weapons states at present professing a NFU policy are those that are actively increasing the size of their nuclear forces. American intelligence believes that the Chinese force of some 20 intercontinental missiles is to be tripled over the next decade or so. China will no doubt claim that this is being done in response to the American development of anti-ballistic missile defences. More probably China, despite the priority given to economic development, would have gone ahead with this expansion anyway. India also has plans to increase her nuclear missile force, perhaps putting them in submarines, a wholly logical implementation of a second-strike posture. Other possible responses are to disperse the missile sites, to go for mobile missiles or provide anti-ballistic missiles to defend them. (The ABM Treaty – now defunct – allowed for this. The US briefly possessed such a system and Russia still does). Another likely corollary of adopting a survivable second-strike posture would be to raise the alert status of nuclear forces. If the assumption is that these forces will be used at the time and place of a country’s own choosing then in normal circumstances a relatively relaxed posture can be adopted and the notice to fire reduced in a measured manner, when needed. If, on the other hand, nuclear weapons can be used only in retaliation for a nuclear strike on oneself then this is to give control of the timetable to the enemy. In logic a much shorter period of notice would have to be routinely observed and possibly authority to fire in the last resort permanently decentralised to individual missile force commanders. Worse still, such a country might feel compelled to adopt a policy of ‘launch on warning’ - the most unstable posture of all.

NFU of Weapons of Mass Destruction (WMD)

A variant of great topical relevance is NFU related to weapons of mass destruction (WMD). The idea is that a country might give an undertaking never to use any weapon of mass destruction save in response to such a use against itself. In practice, since the nuclear weapons states have already forsworn any use of biological or chemical weapons, this would amount to a threat to use their own nuclear weapons in response not only to a nuclear, but also to a chemical or biological attack. There are a number of difficulties over this suggestion. For example, it would at first sight conflict directly with the Negative Security Assurances by the USA, Russia, France and Britain (and the blanket NFU policy of China) given to all non-nuclear weapons parties to the NPT. This is not an open and shut issue however. The ‘Axis-of-Evil’ states (North Korea, Iran and Iraq), while all being parties to the NPT, could be regarded as not in good standing on the grounds of having attempted to circumvent the Treaty. It is certain that North Korea and Iraq have been in violation of the treaty as a result of concealing their illicit production of fissile material. The US intelligence services insist that they have good evidence of a secret nuclear weapons programme in Iran. But these states remain parties to the Treaty. Since they have not been shown to be in ‘material breach’ of the treaty (i.e. by the actual production of nuclear weapons) it cannot be legally claimed that they have forfeited their immune status under the NSA guarantees.

A related argument derives from the doctrine of belligerent reprisal. Traditionally an important mechanism that can be brought to bear is an act of retaliation, otherwise in itself illegal, carried out by one party to a conflict in response to an illegal attack upon itself and intended to cause the attacker to comply with the law. It is easy to see how the United States, for example, could seek to justify under this rubric a nuclear strike on Iraq if the latter attacked US troops with chemicals. The 1977 Geneva Protocol I prohibits certain types of reprisal (for example against civilians). But the US is not a party and the UK made an interpretative declaration when ratifying clearly intended to maintain a right of reprisal. Against this it is argued that Protocol I now has the status of customary international law and the UK reservation is of no effect. These matters are highly controversial and it would be unwise to draw conclusions from them.

More substantial difficulties are of a practical nature. A ballistic missile can easily be traced back to its source. (American satellites were able to track in real time all the Iraqi SCUDS fired in the 1991 Gulf war). And the explosion of a nuclear warhead is unambiguous. But the same is far from true of a biological or chemical attack. As the British House of Commons Defence Committee has pointed out in a recent report:

• ‘A radiological release from a dirty bomb would, initially, have many of the characteristics of a conventional explosion. It might be some time before radiological contamination became apparent. An attack with a chemical weapon might be identified by the smell of an agent, if any. Otherwise the appearance of symptoms might be the first evidence. .. A biological attack might be rapidly identified – if for example a visible agent was used, such as the anthrax spores sent through the post in the United States. On the other hand it might only become apparent over time in the same way as a naturally occurring epidemic would be identified’.
  (House of Commons Defence Committee. Defence and Security in the UK. Sixth report of Session 2001-02, Vol. I, HC 518-I, 24th July 2002)
  

This highlights the possible difficulty of tracing an attack of this kind back to its country of origin and of identifying its nature beyond doubt. It is also far from certain that a chemical or biological attack against armed forces well prepared for it would have major military implications. It follows that a nuclear response to such attack might well be disqualified both on grounds of proportion (too devastating) and discrimination (wrong target). If a city or a whole population had been attacked with chemical or biological weapons then the moral and legal justification for a nuclear response might be stronger but the practical difficulties would remain.

The British Government has given elusive answers on this question. In 1998, when asked specifically what the Government’s policy was with regard to nuclear retaliation in the case of aggressor states contemplating the use of chemical or biological weapons their spokesman replied;

• ‘The use of chemical or biological weapons by any state would be a grave breach of international law. A state which chose to use chemical or biological weapons against the United Kingdom should expect us to exercise our right of self defence and to make a proportionate response’.
  (Lord Hoyle in a Written Parliamentary Answer to Lord Judd, WA 224, House of Lords Official Report, 29th October 1998)
  

Later the Defence Secretary widened the scope of this reply to bring in the possibility of retaliating against attacks on British forces as well as territory :

• ‘Let me make … clear the long standing British government policy that if our forces – if our people – were threatened by weapons of mass destruction we would reserve the right to use appropriate proportionate responses which might … in extreme circumstances include the use of nuclear weapons’.
  (Rt. Hon. Geoff Hoon MP, Secretary of Sate for Defence, speaking on the Jonathan Dimbleby programme, ITV, 24th March 2002)
  

Most recently he has seemed slightly to resile from this position:

• ‘We would be prepared to use nuclear weapons only in extreme circumstances of self defence. We would not use our weapons, whether conventional or nuclear, contrary to international law. A policy on no first use of nuclear weapons would be incompatible with our and NATO’s doctrine of deterrence, nor would it further disarmament objectives. We have made it clear, as have our NATO allies, that the circumstances in which any use of nuclear weapons would be contemplated are extremely remote. Our overall strategy is to ensure uncertainty in the mind of any aggressor about the exact nature of our response and thus to maintain effective deterrence’.
  (House of Commons Official Report, 11th July 2002: Column 1133W)
  

There is indeed a pragmatic case to be made for a policy of constructive ambiguity regarding the circumstances in which a country would use its nuclear arsenal. Arguably, as Geoff Hoon implies, this maximises its deterrent effect. But it is slightly out of kilter with his other principle of adherence to International Law. Given these doubts, what can be said about the military implementation of such an NFU? It is much less a promise not to use nuclear weapons save in retaliation as it is a warning that in certain other circumstances, deliberately left vague, they might be used. One possible implication could be to emphasise the need for ‘sub-strategic’ nuclear weapons in that country’s armoury. This idea, originating in France, has been recently adopted by the British. Their Strategic Defence Review defined the sub-strategic concept as providing ‘an option for a limited strike that would not automatically lead to a full scale nuclear exchange’ (The Strategic Defence Review, The Stationery Office London, Cm 3999, July 1998, page 18, paragraph 63). This option is being provided for, within the total of 48 warheads to be carried by each Trident submarine, presumably by loading some missiles with single warheads, and possibly with reduced yields produced by detonating only the warhead primary - boosted or un-boosted as the circumstances may require.

A similar evolution can be seen in the United States. There also, government spokesmen have been at pains to emphasise that if the country or its forces were attacked with weapons of mass destruction, nuclear retaliation could not be ruled out. Thus General Richard Meyers, Chairman of the Joint Chiefs of Staff, in speaking of the Pentagon’s recent Nuclear Posture Review (of which extracts are available on www.globalsecurity.org/wmd/library/policy/dod/npr.htm) explained that this merely preserves for the president the options he might need ‘in case this country or our friends and allies were attacked with weapons of mass destruction, be they nuclear, biological, chemical or for that matter high explosives’. (The Economist, March 16th 2002). By way of implementation the Review suggests several nuclear weapon options that might provide important advantages for the nation’s deterrence posture: modifications to existing weapons to provide additional yield flexibility in the stockpile; improved earth penetrating weapons to counter the increased use by potential adversaries of hardened and deeply buried facilities; and warheads that reduce collateral damage. These are all variants on the theme of increased usability of nuclear weapons, sub-strategically. Possible immediate contingencies are given as an Iraqi attack on Israel or its neighbours, a North Korean attack on South Korea, or a military confrontation over the status of Taiwan. In every such case ‘first use’ is clearly implied. So all this, in a round-about way, could be described as implementation of a policy of NFU against WMD. Its abandonment would logically remove the need for new varieties of nuclear warhead and confine the task of the weapons laboratories to stockpile stewardship only.


Britain and France

It remains to discuss some particular cases relating to NFU policies in the narrow sense: that is, in response to nuclear attack only. Britain would be in an ideal position to make such a declaration. Her deterrent force (one submarine permanently on station) comes as close to invulnerability as it is possible to be. While the machinery of government and the means of communicating with the submarine would clearly be vulnerable to a nuclear strike, the boat herself would be immune. Moreover it is hard to see in what circumstances and against what country the UK would ever go it alone in attacking with her nuclear missiles except in direct retaliation. Other than in the NATO context no British government has ever attempted to suggest one. As discussed above the option of a nuclear strike as a response to (or even to pre-empt) an attack on the UK or British forces with germs or chemicals is extremely dubious in law and all but impossible to visualise as a practical proposition. In order to implement an NFU declaration no adjustment whatever would be needed to her nuclear or conventional forces – save possibly to eliminate the need for a sub-strategic nuclear option. But any such declaration is out of the question save in the context of a change of policy by NATO (see below). The same is true, mutatis mutandis, of France. Her nuclear deterrent force still includes a force of some sixty Mirage medium bombers with stand-off missiles and twenty-eight Super Etendard naval strike aircraft. These are clearly more vulnerable to a nuclear ‘first strike’ than submarines, but France also possesses the latter, roughly equivalent to the British Trident force. Her deterrence policy also encompasses responding to chemical or biological attack if the country’s vital interests were at stake. France is traditionally far more independent of NATO than Britain and more fiercely possessive of her own independent nuclear force de frappe. It is hard to see France volunteering for a formal NFU posture unless NATO took the lead, a move she would be likely to oppose. Again no change would be needed in her force posture since her strike aircraft are dual-capable.

Israel

Israeli’s nuclear deterrent, while of long standing, is a great ‘unmentionable’. Thus a recent Adelphi paper entitled ‘Continuity and Change in Israeli Defence Policy’ makes no mention of the word ‘nuclear’ whatever, save in describing the Iraqi reactor destroyed by Israel in 1981. (Mark A. Heller, Adelphi Paper No. 335, IISS, 2000). The nearest he comes is saying that ‘deterrence’ might be strengthened by developing a secure second-strike capability, perhaps built around submarine launched cruise missiles. ‘The Military Balance 2001-2002’ (IISS October 2001) says bleakly: ‘Israel is believed to have a nuclear capability with up to 100 warheads. Delivery means could include aircraft (F-151), Jericho I missiles (range up to 500 km.) and Jericho II missiles (estimated range 1,500-2000km.)’. Israel is one of the only four nations of the world not party to the Nuclear Non-Proliferation Treaty NPT (with India, Pakistan and Cuba). Hence she is in breach of no treaty. But the irony is not lost on her neighbours that whereas Iraq and Iran (both parties to the NPT, as described above) figure in the American ‘axis of evil’ category, no censure is directed at Israel for her covert nuclear armoury. Since she will not admit to possession, no question arises of an NFU declaration. No doubt these weapons are now mainly seen as counters to a missile attack by neighbours. During the Gulf War of 1991 Iraq launched 42 SCUD missiles against Israel, and although only one person was killed as a direct result, passive defence measures caused widespread disruption. These missiles carried conventional explosive warheads, but the spectre of biological or chemical attack has particular resonance for such a small country. The dispersed and concealed nature of these threats makes them all but immune to pre-emptive attack. Retaliation against targets of high value to the perpetrators seems the likeliest response. In three previous wars (1948-49, 1967 and 1973) Israel faced an existential threat from a combination of Arab armies, greatly outnumbering her forces numerically and threatening all her land frontiers (the assailants being Egypt, Jordan, Syria and Iraq). While these attacks were handsomely seen off by Israeli forces, and political developments since then have diminished the threat, the possibility of needing a nuclear strike from the last ditch can never be wished away until there can be a comprehensive regional settlement securing Israel’s long term future. But if we take the purely theoretical case of Israel embracing an NFU policy, one obvious implication would be to put the deterrent under water, as Mark Heller indicated.


South Asia

As already mentioned both India and China have given undertakings never to use nuclear weapons first under any circumstances. This mutual NFU posture may be adding some small measure of stability to their relationship. It is worth asking if anything further would be gained if Pakistan were to do likewise. Pakistan’s forces, both nuclear and conventional, are heavily outnumbered by India’s. As regards the nuclear threat, having no secure second-strike capability she might well feel the need to strike first. India and Pakistan have been to war three times during the half-century or so of their independent existence, twice over Kashmir. The current state of tension between them is plainly very dangerous, and it has been argued that Pakistan would not have chanced her arm in this way had she not enjoyed the comfort of a nuclear last resort. But war has not broken out and at the time of writing (in August 2002) there has been some relaxation of tension. The Pakistani government has made some show of throttling back the infiltration of Islamic fighters into Indian-held Jammu and Kashmir. Intense diplomatic pressure has been applied by outside powers, notably the USA. But the possibility of the war becoming nuclear may have had had some restraining influence this time round. If so, it will have been because any serious Indian incursion, say to suppress guerrilla training camps in Pakistani-held Kashmir or in Pakistan itself, runs the risk of spilling over into a general assault on Pakistan’s road and rail links or military infrastructure. Pakistani generals have made it clear that major incursions of this kind would be met with a nuclear response. Indian generals have said that they know better than to provoke Pakistan to this dangerous extent. But one of the wars between the two countries led to the dismemberment of Pakistan as it then was. Given India’s superiority of forces it would be a rash Pakistani leader who relied too heavily on Indian self-restraint if it came to fighting a major war. An NFU declaration by Pakistan, to the extent that it was believed, could prompt India to take greater risks, which is exactly what is not wanted. If it also led Pakistan to devote more of its meagre resources to nuclear and conventional military forces and further to militarise the country and the regime, this would be a very bad bargain indeed. So far from being an effective confidence building measure it could be the reverse.

NATO

The position of NATO is quite different. It now has a very comfortable superiority of force over any possible Russian-led coalition, nor is there the slightest sign of danger from that quarter. The current NATO doctrine of ‘flexible response’ is a hangover from the Cold War and Mr. Hoon’s justification of it (page 9 above) makes no sense in the East-West context. The need to defend against rogue states using gas or disease has been discussed above and scarcely amounts to a convincing justification for retaining a nuclear first-use option. Moreover a NFU declaration by NATO could have other consequences. It would do away with any rationale for the tactical nuclear weapons still maintained by the US on bases in NATO Europe. If these were sent home no one would be a whit less secure. It would also provide the West with some leverage on the Russians in persuading them to do away with their much larger inventory of tactical nuclear weapons. This has been a long-standing Western objective, towards which no noticeable progress has been made in recent years. In this sense implementation of a NFU declaration could have beneficial consequences. But the forward basing of weapons has always been justified as providing a tangible expression of America’s commitment to Europe, and normally welcomed as such on the European side. The last thing needed at the moment is for the Europeans to make an issue of this; there are causes of dissension enough. Any move in this direction could only come from the Americans. If it did, then the European allies should welcome it.

Conclusion

Measures to implement an NFU declaration might, in theory, include a build-up of conventional forces to compensate but there is no instance where this is known to have happened. Some rearrangement of the size and posture of nuclear forces could be expected. These effects may have been seen where NFU declarations have been made (i.e. by China, India and during the Cold War the USSR). Nor they have been beneficial, quite the reverse in fact. An NFU declaration by a power clearly inferior to its adversary in conventional terms (e.g. Pakistan) could be positively destabilising. An NFU policy applied to Weapons of Mass Destruction, thus justifying the first use of nuclear weapons in response to a chemical or biological attack, would be legally contentious, of doubtful practical value and could open a Pandora’s box by making nuclear weapons, at least in theory, more usable. One benign consequence, which arguably could flow from a NFU policy, would be the phasing out of tactical or sub-strategic nuclear capabilities. If NATO were to take this course, which would in no way diminish its security in the context of collective defence, it would allow American nuclear weapons to be finally removed from continental Europe with possible pressure on Russia to do likewise. This is one clear instance where beneficial consequences could flow from an NFU declaration.