Is Nuclear Deterrence Ethical and Legal?

Nor has the anti-nuclear movement subsided. Even before Russia’s full-scale invasion of Ukraine, international moves to prohibit the possession and the threat of nuclear use as well as actual use have gathered pace.

After a variety of reports into the devastating short and long-term effects of a nuclear war in 2013 the UN General Assembly set up a working group on how to negotiate a world without nuclear weapons. Under the auspices of the International Campaign to Abolish Nuclear Weapons (ICAN), which received a Nobel Peace Prize for its troubles, the terms of a Treaty on the Prohibition of Nuclear Weapons (TPNW) was agreed in 2017. Those who sign up agree not to develop, test, produce, acquire, possess, stockpile, use or threaten to use nuclear weapons. They also agree not to assist others with these prohibited activities. Some 70 states have ratified the treaty although none of these are states that possess nuclear weapons or depend on deterrence for their security.

The elimination of all nuclear weapons would mean we could stop worrying about ‘the bomb’, although no doubt we would find something else to worry about. But for now, given the worsening security situation, abolition is an even more remote prospect than it seemed in 2017. Nonetheless ICAN, and other moves to challenge the legitimacy of nuclear arsenals and question the morality of plans for their use, leave the nuclear weapons states with some explaining to do. How can they justify security policies based on threats of mass destruction?

One way to address the issue is to follow international humanitarian law, which seeks to limit the harmful effects of armed conflict. For nuclear weapons to be compliant ways would have to be found of using them that would avoid mass civilian casualties, presumably by directing them against military targets in the numbers necessary to destroy those targets, but no more.

This would  represent a significant departure from the thinking that dominated the question of how to manage the superpower nuclear relationship during the Cold War. Then the tendency was to focus on the disastrous consequences of a nuclear war to ensure that one did not take place. This included denying the possibility that one could be kept limited. The reasoning was that states knowing that aggressive action risked terrible retaliation would be cautious. That caution would give time for diplomacy to resolve the conflict before it escalated into nuclear exchanges.  A nuclear war contained and focused, with civilian targets avoided, sounds less scary but for that reason would be more likely to occur. The more these terrible weapons conformed to the laws of war it would be potentially easier to use them. This raises the issue of whether it is better to loosen or keep the close association of nuclear war with mass death.

Early last week I was fortunate to be able to attend a workshop on law, ethics and nuclear weapons organized by Professors Janina Dill of Oxford and Scott Sagan of Stanford which addressed this topic.  What follows is not a report on the workshop – the individual papers will eventually be published – but my own thoughts prompted by a fascinating discussion.

The Laws of War and Air Power
International humanitarian law is founded on ethical values, notably the Christian ‘Just War’ criteria, which describe when it is right to go to war and how wars should be fought. The basic principles are that violence can only be used to accomplish legitimate military purposes, that attacks must only be directed at combatants and not civilians, and even incidental threats to civilian life must be avoided if they are likely to be excessive in relation to the anticipated military advantage. At all times care must be taken when mounting military operations to spare the civilian population, civilians and civilian objects.

Nuclear weapons were first introduced at the end of the Second World War, during which international law had been progressively disregarded and the distinction between combatants and non-combatants steadily eroded. Helpless civilians had died in many different ways, from extermination camps to mass air raids. Arguably the culmination of this process was the atom bombing of Hiroshima and Nagasaki, though the fire-bombing of Tokyo in March 1945 remained the deadliest air raid of the war.

Efforts to use air power solely against military or military-related targets, from weapons factories to railway marshalling yards, had led to heavy losses in aircraft because of the vulnerability to air defenses as a result of flying low and during the day to spot specific targets. It was easier to target enemy morale by bombing cities, even if it was unclear how that would bring forward the end of the war. After the Luftwaffe set the precedents with the blitz the RAF needed little encouragement to follow. As Basil Liddell Hart observed: ‘inaccuracy of bomb aim led to inhumanity of war-aim.’

In 1949, after the war, when the Geneva Conventions were revised the Americans and British were adamant that there could be no limits on air power. Conventional air raids against cities continued in subsequent wars, notably in Korea and Vietnam. By the early 1970s, however, it became much harder to dismiss the idea of limits as discrimination in the use of air power became possible with the introduction of highly accurate ‘smart’ bombs. These removed the excuse that even with a focus on military-related targets it was likely that large numbers of civilians would be killed not intentionally but as ‘collateral’.

There was now an element of choice. It was possible to confine attacks solely to military targets without harming civilians; equally, of course, if the aim was to maximize harm to civilians that could also be done as Russia has demonstrated in Syria and Ukraine. In 1977, the 1949 Geneva Conventions were revised. Included in the extensive Additional Protocol 1 were articles outlawing indiscriminate attacks on civilian populations, and the destruction of food, water, and other materials needed for survival. This also included using technologies whose scope of destruction could not be limited. The US and UK accepted that this applied to air power but still insisted that ‘the new rules introduced by the Protocol are not intended to have any effect on and do not regulate or prohibit the use of nuclear weapons,’ though there was nothing in the wording to support this claim. In 1996 the International Court of Justice gave an advisory opinion. It was not very helpful:

the threat or use of nuclear weapons would generally be contrary to the rules of international law applicable in armed conflict, and in particular the principles and rules of humanitarian law; However, in view of the current state of international law, and of the elements of fact at its disposal, the Court cannot conclude definitively whether the threat or use of nuclear weapons would be lawful or unlawful in an extreme circumstance of self-defense, in which the very survival of a State would be at stake.

The ICJ also encouraged the nuclear weapons states to pursue disarmament.

Nuclear Weapons: Do Different Considerations Apply?
Could nuclear weapons be used in ways consistent with international humanitarian law? As early as the 1950s there had been efforts to see whether nuclear weapons could be used as if they were conventional weapons. ‘Tactical’ nuclear weapons were developed, including mines and artillery shells as well as low-yield bombs, and presented as similar in character to conventional weapons, only more powerful. The claim was that their effects could somehow be confined to the battlefield. This was never convincing. The essential feature of nuclear weapons was their destructiveness, with radiation effects in addition to those of the blast and fire.

If low yield weapons were used on any scale, then the cumulative effects would be horrendous, especially when fallout was taken into account. The risk of introducing these weapons into an ongoing battle in the middle of Europe was that it would not bring the fight to a quick end but that it would lead to nuclear escalation far earlier than would otherwise have been the case. This is what is meant by ‘lowering the nuclear threshold.’

In the late 1970s there were proposals to replace some of these aging battlefield weapons with what was described as an ‘enhanced radiation reduced blast’ weapon, designed to disable tank crews, otherwise known as the ‘neutron bomb’. This was denounced by the anti-nuclear movement as another abomination, the ultimate ‘capitalist’ bomb as it would kill people while leaving property intact. More seriously they warned of any blurring of the boundaries between conventional and nuclear weapons. Nuclear weapons had to be kept in a separate category.

Though small yield weapons became part of the superpower arsenals the arrival of thermonuclear ‘hydrogen’ weapons, with yields equivalent to millions of tons of TNT, each capable of obliterating single cities, also in the 1950s, made the association of nuclear weapons with mass death ever stronger. In the 1960s it became even stronger still when the US administration began to worry about how to stabilize the arms race and prevent what was described as the ‘reciprocal fear of surprise attack.’

The nuclear relationship had stabilized because neither superpower could gain a decisive advantage in the nuclear arms race. The result was the condition known as ‘mutual assured destruction.’ But this depended on the first strike/ second strike distinction.  A ‘counterforce’ first strike would represent an attempt to win a nuclear war by taking out the adversary’s means of retaliation, so leaving it defenseless. If the adversary had a second-strike capability that meant it could absorb a first strike and then retaliate to inflict an ‘unacceptable’ level of damage. This was normally described as ‘counter-value’ targeting, measured in terms of the proportions of the population killed and industrial capacity destroyed. The assumption was that if both sides faced unacceptable damage neither would have an incentive to rush to a nuclear strike.

The point was not that this a desirable targeting philosophy. Its relevance was as a source of caution, reinforcing mutual deterrence. If this was a fact of life from which neither superpower could escape then it was an argument for restraint, both when it came to fighting wars and also weapons development. In particular, and controversially, this logic led to an argument for abandoning attempts to defend against missile strikes. A successful defense could support a first strike by thwarting a retaliatory second strike. In practice a successful defense would be difficult because of the ease with which the enemy could mount an overwhelming attack with yet more offensive weapons along with chaff and decoys. The main effect of a defensive program therefore would be to fuel the arms race with more offensive programs, leading to more uncertainty and instability.

This turned all the traditional just war/international law considerations on their head leading to a paradoxical logic: ‘Offence good; defense bad. Killing cities good; killing weapons bad.’ However perverse this logic, with its apparent endorsement of mass killing, it became the new orthodoxy. It influenced the design of both arms control treaties and force structures. The underlying view was that it was for the best if both sides had survivable forces that could withstand a first strike. Submarines patrolling the oceans with their own ballistic missiles represented the most reliable form of safety from first strikes.

The legacy of MAD thinking remains profound. It can be seen in the presumptions that the ability to attack an adversary’s nuclear capabilities risks instability, and that any escalation to nuclear use will almost certainly end with Armageddon. Discussions of limited nuclear wars still sound delusional.

Nonetheless, just as the development of precision guidance, improved sensors, and more sophisticated command and control systems, made it possible to think about using conventional weapons in a variety of ways, there were attempts to do the same with nuclear weapons. The discomfort with the idea that the President might be stuck only with mass destruction and counter-value targeting, encouraged a drive in the US to develop alternative options.

Then when the end of the Cold War came there seemed to be less need to worry about these things. With improved political relations the heat was off: the question of targeting could be left quite open. With the major powers cashing in their ‘peace dividends’ deterrence seemed less of an issue.

Against this backdrop it became harder to keep the question of nuclear use separate from questions of international humanitarian law. If more discriminate targeting of conventional weapons was possible then was that not also the case with nuclear weapons? In 2013 the Obama administration announced that in the future all nuclear plans must be ‘consistent with the fundamental principles of the Law of Armed Conflict.’ He also wanted to sustain a policy of deterrence. Much of the Oxford conference was preoccupied with what that might mean in practice.

Michael Quinlan
One person who tried to engage with this question while in a position to do something about it was Michael Quinlan, someone who got mentioned during the conference. He provided most of the intellectual heft behind the UK’s nuclear deterrence, while a senior official at the MoD (not the one I was referring to earlier). He died in 2009, not long after publishing a typically thoughtful book on the topic. He was one of those civil servants who cared deeply about policy and wrote about it with great lucidity. He was also a deeply religious man, educated by Jesuits, whose Catholicism influenced his thought. After his retirement he wrote a book on just war with someone who shared both his faith and interest, General Sir Charles Guthrie, a former Chief of Defense Staff. His moral sensibility led him to oppose the 2003 war against Iraq.

Quinlan was an energetic proponent of nuclear deterrence. His starting point was that nuclear weapons were in a category of their own. They could not be considered as equivalent to conventional weapons. Their arrival

carried the potential of warfare past a boundary at which many previous concepts and categories of appraisal—both military and political—ceased to apply, or even to have meaning.

When combined with ‘the worldwide delivery capability of modern missiles and the diversity and elusiveness of missile platforms’ they made available ‘what is for practical purposes infinite destructive power, unstoppable and inexhaustible at any humanly-relevant levels.’

Quinlan recognized that there could be possible ‘subdivisions of the spectrum of force,’ including in the move from conventional to nuclear weapons, but he was doubtful about ‘abstractions like thresholds and firebreaks.’ The key point was that ‘no conceptual boundary could be wholly dependable amid the stresses of major war.’ How non-nuclear war was fought would affect both the likelihood and character of a nuclear war.

He accepted that deterrence depends on the risk of actual nuclear use, although not a certainty that they would definitely be used. It was impossible to know how a set of political leaders would respond to their own moments of truth.

The concept of deterrence … cannot exist solely in the present - it inevitably contains a reference forward to future action, however contingent. The reference need not entail automaticity, or even firm intention linked to defined hypotheses; it need entail no more than a refusal to rule out all possibility of use; but it cannot entail less than that.

Quinlan also argued that this required real preparations by real people, who could not be expected to be part of a charade. He doubted whether the

commitment of many thousands of individuals — often in career-long and very demanding tasks—could be durably sustained in the known absence of any policy for use or for contingency planning which they could regard as seriously intended.

How then did he reconcile this acceptance of the need for nuclear plans with his religious beliefs? He addressed this issue when the case was being made both within government and then to the wider public for a new generation of nuclear submarines and Trident missiles. The solution was to shift targeting from a focus on Soviet cities to what was described as the ‘sources of Soviet state power.’

This idea of attacking leadership targets in a so-called ‘decapitation’ attack, had been discussed in the US in the 1970s. The ability to take out the ‘national command authority’ might at least give the elite pause for thought before they set in motion events that could end with their own demise. But if the enemy government no longer existed who would call an end to the war? And if the elite believed that they could protect themselves by hiding in underground bunkers might they feel more confident that they could escape the consequences of their actions?

Directing attacks in this way therefore might get round the problem of intent but in practice, given the number and location of the relevant targets, the effect would still be to kill large numbers of Russians. And as Quinlan believed that deterrence depended on the infinite possibilities of destruction then he could not argue that a more limited strike would serve the purposes of deterrence. Focusing on the prime intent of the targeting did not preclude a secondary intent that still involved mass death. So this option did not really resolve the ethical dilemma. After he retired from public service Quinlan was actively engaged in public debates and was always prepared to make a robust case for deterrence, but this was always the issue that caused him the most difficulty.

Since the time that Quinlan was grappling with these issue as an official more options for limited strikes have been developed, including for the UK. If the moment came decision-makers, if they had survived an enemy attack, might insist on full retaliation or decide that it is all quite pointless as the harm has already been done and they have no wish to add to the suffering. It is also doubtful that they will be worrying at this stage about international law. 

The argument that there is no automaticity in nuclear use should deterrence fail therefore still holds. Deterrence can still work so long as there is a risk, which can be quite small, that nuclear weapons might be used. Over these past two years Vladimir Putin has managed to exude a sense of menace and danger sufficient in practice to deter Western government from intervening more directly on behalf of Ukraine without ever quite spelling out what he would do should the moment come..

In some highly stressed and extreme situation, perhaps in the aftermath of a strike against their country, political leaders might find ways of using these weapons against purely military targets and in a limited way. Then they could claim that were staying on the right side of international law and morality, to the extent that they cared in such a situation. But the more they dwell on this possibility in advance they risk reducing deterrence because the danger might not seem so bad to an adversary or even worse, because counterforce capabilities threaten their second-strike capabilities.

There are really no good targeting options for nuclear weapons. All pose awful choices. The only way to escape these dilemmas would be to get out of the nuclear age by eliminating the weapons. There may come at a time when political conditions make that possible. But for now those conditions are remote. Too many countries believe in their strategic value of nuclear arsenals. Making the case for complete disarmament may follow a moral imperative but it can also be a way of ducking all the awkward questions inherent in living in the nuclear age.

A more realistic alternative is to accept that the nuclear weapons states will continue to wish to deter threats to their very existence and that may involve threatening the potential aggressor with something equally terrible. This does not preclude developing options for limited use against military targets. If the last resort is reached, and deterrence has failed completely, restraint would be preferable to adding to the sum total of terror and misery, although limited use would appear as an odd compromise. Either anger and a desire for retribution would lead to a massive response, or despair and paralysis might lead to no response at all.

The problem is not with the idea that nuclear weapons could be used legally and morally, with discrimination and proportionality; only with any suggestion that if this idea is pursued the potential for escalation to the extremes will be removed. If we want political leaders to stay risk averse at times of crisis then it is best not to break the close association of nuclear weapons with mass death. Too much confidence in limited nuclear use could make it more tempting in a crisis or when things are going badly in a conventional war. In practice I suspect the association with mutual destruction, which is so deeply embedded in popular discourse, will prove to be very durable. And maybe that is just as well.

To state the obvious the nuclear situation will become more manageable and tolerable when great power relations are relaxed. When and if current tensions ease it would be wise to look for ways to reduce even more the risks of a nuclear calamity. If the weapons cannot be completely eliminated, however, then neither can the risks of the worst imaginable outcomes. Little can be gained by pretending otherwise.

Lawrence Freedman is Emeritus Professor of War Studies at King’s College London. The article originally appeared on Lawrence Freedman’s Substack “Comment is Freed.”