one where stability depends on something that is more the antithesis of strategy than its apotheosis—on threats that things will get out of hand, that we might act irrationally, that possibly through inadvertence we could set in motion a process that in its development and conclusion would be beyond human control and comprehension.
This worked because of the unavoidable confusion that would attend a major breakdown in East-West relations in Europe. So to the extent that those who had responsibility for unleashing nuclear arsenals lived by the motto that if they ever had to do so they would have failed, they could claim to have succeeded. The conclusion, drawn from the reported observation on the Charge of the Light Brigade in 1854, from the first edition was: ‘C’est magnifique … mais ce n’est pas la stratégie.’
With this Fourth Edition there is no reason to revise the earlier assessment of the trends in thinking and the logic of the policies of the Cold War. Assessments of the debates of the past are now coloured by the fact that matters worked out better in practice than this theory might have predicted. Even so, the earlier conclusion that the practice was in some way non-strategic now seems mistaken. The practice was quite compatible with the approach of those described by Michael Howard, in one of the first surveys of the impact of nuclear weapons on strategic thought, as the classical strategists, that is ‘the thinkers who assume that the element of force exists in international relations, that it can and must be intelligently controlled, but that it cannot be totally eliminated.’ Such assumptions flowed naturally from a study of history and the more traditional forms of political theory, and became associated with such figures as Brodie, Kissinger, Aron and Howard himself.1
In the first years of the Cold War nuclear developments and deployments, doctrinal pronouncements and commitments to allies were urgent and intense. The policies adopted at this time were crucial in defining the terms of the Cold War and facilitating alliance formation. This meant that by the early 1950s many of the great strategic issues of the post-war era had already been settled. With the Cold War framework in place the most pressing issues involved how best to cope with the pace of technological change and the complexity of the issues that each new development seemed to throw up. Classical strategy was barely adequate to this challenge; a different sort of theory and analysis was required, from which the classical strategists could only borrow. This realisation ushered in the ‘golden age’ of nuclear strategy, which is normally considered to have lasted until the mid-1960s,2 and was marked by the growing prominence of RAND analysts and alumni such as Kahn, Schelling and Wohlstetter. Although these pioneers maintained their links with classical strategy, many of their followers did not. Rather they became specialists whose starting point was the extreme qualities of nuclear weapons rather than the timeless qualities of international politics.
As the conceptual framework of the ‘strategy of stable conflict’ became ingrained into official thought, scores of specialists arrived to fill in the details, monitor its application and bother themselves with some of the unresolvable tensions at its heart. This specialism required familiarity with the technical properties of the weaponry, targeting, the organisation, command and control of operations, doctrine, vulnerabilities to surprise attack and defences, arms control and so on. It led to studies full of calculations and graphs and matrices—computing the kill probabilities of missiles of a given range and accuracy against point targets with a certain hardness, considering the prospect for successful defence against cruise missiles, discussing the relative merits of Equivalent Megatonnage and Counter Military Potential as measures of military power, comparing the surviving missile throw-weight and bomber payload after the initial nuclear exchange, identifying mathematically the level at which destruction is both assured and unacceptable.
Although this knowledge by itself could not be considered ‘strategic’ it seemed to be a pre-requisite for the formation of a strategic view. As Bernard Brodie observed at the start of the nuclear age, ‘everything about the atomic bomb is overshadowed by the twin facts that it exists and its destructive power is fantastically great’.3 The third fact, which came later, was that it came into the possession of more than one state. All this meant that for most objectives nuclear use appeared wholly disproportionate, and carried enormous risks of devastating retaliation.
The efforts of nuclear specialists were devoted to rescuing nuclear means from these facts. To limit destructive potential they encouraged the development of low-yield/low collateral-damage weapons which in themselves were not so destructive, and of strike plans which avoided centres of population. The trouble with this approach was that either the number of weapons required to destroy important targets ensured that the level of destruction remained high or else the weapons became indistinguishable in practice from conventional weapons. In the event the development of super-precise conventional weapons removed the need to think about nuclear weapons as the only means of destroying certain types of targets. The strategic purpose of limited nuclear strikes was therefore more about demonstrating how it might get even worse through further escalation, thereby giving the adversary pause for thought. Yet escalation would add confusion and uncertainty to the suffering and anger. It became hard to develop confidence in limitations on nuclear use and almost impossible to rely upon them.
Any hope of eliminating the adversary’s capacity for retaliation required first strike capabilities, the only limited strike which could make any military sense. All other strikes relied on paralyzing or disrupting or in some way intimidating the enemy decision-makers. In the end, however, they could not physically prevent retaliation. The physical requirements of a true first strike remained demanding, almost impossibly so, and even the more hawkish strategists tended to discount it as a genuine option. President Reagan’s personal odyssey through the countervailing strategy, Star Wars and Reykjavik could be taken as an eloquent statement of this inability to come up with a strategy based on nuclear threats that could ensure a conclusion other than mutual and massive destruction. The only safe statement was that the first nuclear salvos would probably—if not inevitably—create a disastrous situation for all concerned. This led to the large, simple but important conclusion that such situations must be avoided.4 In September 1987 Reagan and Gorbachev affirmed their ‘solemn conviction that a nuclear war cannot be won and must never be fought’.5
It was difficult to attach any rationality whatsoever to the initiation of a chain of events that could well end in the utter devastation of one’s own society (even assuming indifference to the fate of the enemy society). No means of controlling events sure of avoiding the worst outcome were developed. Any propositions about how particular tactics exploiting specified capabilities might just work involved speculation and conjecture, drawing on suppositions about human nature and specific personalities, and past experience of diplomacy and non-nuclear war. At issue was how to influence the thought processes of political leaders, current and still to come, in circumstances so horrific and in the face of pressures so extreme that they could barely be comprehended. It was impossible to know how political leaders would react to the news of even only a few nuclear weapons exploding on their territory. The response could be a reckless fury, lethargic submission, craven cowardice, or a firm and resolute action. Much discussion assumed that retaliation would be almost automatic to any attack. It served deterrence for that to be believed. But should the moment come it was not clear what actual retaliation would achieve.6 An American President, on learning that Soviet forces were advancing through Western Europe, might be impressed most by the danger to his country’s population or else by a sense of obligation in honouring alliance commitments.
During the Cold War, the nuclear specialists accepted that their task was to help avoid a war and not just come up with tactics for its conduct. They remained, however, firmly focused on the superpower nuclear relationship as if it was self-contained, a sort of gyroscope at the heart of the international political system. So long as this gyroscope could reach equilibrium then the total system would enjoy stability. The requirements in this area ran from compounding the disincentives to a first strike to attending to the dangers of accidental launch to the crisis-managing capabilities of political and military elites. However, there was always an alternative argument that there was an inherent stability to the nuclear system, so that it was more likely to be put at risk by political turbulence in the wider system than by its own internal tensions. From this perspective, the specialists had over-complicated the problem by exaggerating the importance of nuclear variables. The analytical quality of their work was often high but their political grasp could leave much to be desired. The interaction between nuclear weapons and strategies and political developments was more complex, often as much about managing alliance cohesion as superpower antagonism.
If it was the case that the anti-status quo power was the Soviet Union and that NATO depended more on the nuclear threat, then a strategic crisis might have been expected as it became apparent that the specialists were unable to provide even a compelling operational analysis of how nuclear weapons might be used in the face of likely retaliation. Yet awareness that there could never be an enduring technical fix to the problem of stability did not encourage a Soviet challenge to the status quo. The opposite happened, though for reasons largely connected with the distorted economic and political development of communist societies. This posed no problem to classical strategists sensitive to the wider political context and ready to judge all military instruments by reference to other means—economic, diplomatic, and political—as well as to the ends they were supposed to serve. However important nuclear weapons were to national strategy they could never be the whole story. Viewing nuclear weapons in isolation, or assuming that they provided a satisfactory vantage point to discuss strategy as a whole, distorted strategic studies.
So, for the classical strategist, strictly speaking there really was no such thing as nuclear strategy. That was not to say that nuclear weapons were irrelevant to strategy—alliances were held together, adversaries contained and crises managed throughout the Cold War in part by reference to them. But the critical reference here was to the big picture, the crude representations of the ultimate horror. It was always possible that this horror could be mitigated and that nuclear war would turn out to be not quite as bad as feared, or that one side might emerge proud in victory, but these possibilities could never be rated high, and even the least destructive of nuclear wars would involve a level of grief and mayhem that would go off the scales of human suffering. Perhaps international affairs would not have followed a particularly different path without nuclear weapons. John Mueller, argued that war was demonstrably horrible enough already without ‘the bomb’ to encourage restraint.7 Yet the weapons made a difference, in offering horrifying images of what a future war might entail—what a Harvard team called in 1983 the ‘crystal ball effect’. Studies of particular crises demonstrated this effect in operation.8
While the specialists’ efforts to demonstrate how nuclear weapons might solve classical types of strategic problems failed, the activity itself reinforced the view that the possibility of nuclear use could never be quite ruled out, thereby supporting foreign policies based not so much on rational nuclear employment as on the uncertain threat of nuclear war. Any risk of being on the receiving end of a nuclear attack was an enormous constraint. Western and to some extent Soviet strategists became adept at manipulating this risk, with occasional and timely hints of dark possibilities without ever quite spelling out how they would be realised in practice. So long as the weapons were around even a skirmish between the great powers seemed extraordinarily dangerous in its escalatory potential. Strategy required working with and around this danger. The durability of containment encouraged the view that at this level the strategy worked and at a basic and crude level it did. The Emperor Deterrence might have had no clothes, but he was still Emperor. For both sides, from the early 1950s and in the face of thermonuclear nuclear weapons, prudence was the better part of valour. This strategic relationship turned out to be sufficiently robust in its essentials to survive new technologies and doctrines.
After the Cold War nuclear specialists found their role diminished as the demand for their services declined precipitously. This is not to say that there was no requirement for nuclear expertise. In some respects the requirements grew. The practical problems of taking weapons out of service and contracting the nuclear facilities had to be addressed while following the programmes of actual and potential proliferators became a more demanding task, again with a very practical aspect as UN inspectors played a cat-and-mouse game with Saddam Hussein. There was even a return to a familiar issue when the debate about the value of missile defences was revived. Nor were nuclear issues less salient. The fear of global nuclear war declined but instead came anxieties about small but malevolent proliferators, perhaps even sub-state in size, who would be ingenious in their methods while able to work outside the normal rules of strategic discourse. After 11 September 2001 the thought occurred that some dreadful event might devastate much of a great city without it being clear who was responsible. A government would be under enormous pressure to respond on the basis of rumours and innuendo. The main difference was that there was less of a need for the US and its allies to rely on nuclear threats because their superiority in conventional forces appeared to be unassailable. The arsenals were maintained, at levels not that far below those of the Cold War, just in case. The main concern was to ensure that threats against them were not allowed to develop.
What is often forgotten in strategic studies, preoccupied with military capabilities, is that the balance of terror rests upon a particular arrangement of political relations as much as on the quantity and quality of the respective nuclear arsenals. Movements in these political relations could prove far more disturbing to nuclear stability than any movements of purely military factors. The major task for the future must be to address the problems of nuclear arsenals in a world of political change.
The question of ‘What do we do if deterrence fails?’ was largely beside the point—not because failure was inconceivable but because it was extremely unlikely under the prevailing political circumstances. The most unnerving situations were likely to arise in completely different circumstances that would in themselves shape the strategic options available. This had already begun as a result of the proliferation of nuclear weapons into the world’s more unsettled regions. It could not be assumed that this process would always reinforce the status quo and dampen drives towards war. There was always a risk that nuclear restraint would be overwhelmed by political turbulence. The paradox of nuclear weapons remained that they reinforced the autonomy of states while, at the same time, they provided the most profound reflection of their ultimate interdependence. The most optimistic prospect was not that international leaders would band together to agree some grand scheme for general and complete disarmament but that if political relations became even less conflictual, nuclear weapons would be seen to be playing a marginal role in international affairs and less care and attention would be lavished upon them.
Following the great political transformations of the last decade of the twentieth century it was possible to see a way forward to the progressive diminution of the role of nuclear weapons in human affairs. But in this respect the twenty-first century has been a disappointment, still marred by violence and cruelty on a substantial scale, while Cold War-era nuclear policies and practices have not been abandoned. Meanwhile, expertise, technologies, materials and means of delivery have been diffused, so that weapons, possibly crude and unreliable but still terrifying, might appear in unlikely places. Even if there can be no purely nuclear strategies there remains a continuing need for strategies that take nuclear weapons into account.
For the better part of the Cold War and post-Cold War periods, the field of nuclear specialists was dominated by a number of giants—those ‘wizards’ who were responsible for many of the key conceptual breakthroughs and were also influential in policy-making. These giants of the ‘golden age’ have now largely left the nuclear scene. The current generation of scholars and policymakers are still the students, or students’ students, of the original nuclear strategists, and in many respects are inheritors and innovators rather than visionaries. They have learned the concepts of their forebears and speak and write in much the same language. The specific terminology may change from time to time but the fundamental ideas remain in place and tend to be recycled. This has led to a belief that nuclear strategy has become a moribund subject; a dead-end for aspiring theorists. Moreover, expertise has become increasingly stove-piped into communities of ‘highly specialized functionalists and regionalists’.9 Whereas the field of nuclear strategy might have once been regarded as ‘esoteric’, it has increasingly been perceived as ‘archaic’; hardly a subject for an aspiring defence intellectual when there were more urgent areas in which to make a career—from counter-terrorism to counter-insurgency to the more traditional forms of grand strategy. The combination of the passing of the Cold War-era giants and the dramatic changes in international affairs reinforces this view. Nuclear strategists risk being seen as relics of a bygone age, restricted to the fringes of contemporary defence debates.10
Under these inauspicious circumstances, what are the implications for nuclear strategy? Should we expect original concepts, or have we reached the ‘end’? It has been generally recognized that the introduction of nuclear weapons had a major impact on the international system and strategic thinking. Most states simply could not avoid dealing with the nuclear issue when devising their foreign and defence policies. In this sense, new concepts were inevitable—the changing nature of the system demanded new theorists to make sense of them and to inform public policy. The community of theorists that then emerged generated the intellectual capital on which the field continues to rely. Scientists, philosophers, strategists, policymakers, etc. thought long and hard about the meaning of the nuclear revolution and what this meant for their own countries, as well as what it meant for humanity as a whole. Gradually, however, the centrality of nuclear weapons in international affairs diminished. It became more of a background factor and was rarely in the foreground. For some of the more recent nuclear states, thinking about these weapons has continued to hold a prominent place in their strategic discourse and policy deliberations. In many cases they have adopted the older concepts, or have made the odd variation on a theme to account for their particular circumstances. They have not made any significant conceptual leaps.
As for disarmament and arms control, apart from the occasional burst of activity, as occurred with the Global Zero movement, many of the same debates about the virtues and drawbacks of limitation versus abolition, or nuclear abolition versus comprehensive disarmament, including conventional forces, continue to be heard. Many disarmament advocates remain highly skeptical and suspicious about arms control. For instance, one proposal that argued for taking nuclear weapons off of alert status, pledging policies of no use against non-nuclear weapons states, no first use against nuclear weapons states, reinstating the ABM Treaty, ratifying the CTBT, safeguarding existing nuclear materials, an interim agreement on ceasing modernization, convening negotiations among nuclear armed countries for total nuclear disarmament—was criticized as being too ‘managerial’ and more likely an excuse to ‘retain nuclear weapons indefinitely’.11 Many of the proposals designed to help reach a ‘global zero’ that gained prominence after 2007 could also have been found in the literature decades earlier.
The key problem was less the substance of the proposals than it was how to sell them to governments that seemed to have more incentive to retain the status quo. Any shift in policy required political willpower but too many other more pressing issues and priorities always seemed to hinder deeper nuclear cuts. Despite the debates between those advocates of a ‘utopian’ agenda versus those who wished to pursue more ‘practical’ steps, not even relatively minimalist practical proposals appeared practicable enough to persuade policymakers to adopt them. In the US, for instance, although this country has such overwhelming military power that it has little need to threaten nuclear escalation, it did not prove possible to adopt a ‘no first use’ policy.12 Such a promise could offer no guarantees. The question of whether it must be followed would arise in circumstances in which emotions were running high, with uncertainties and fear at unprecedented levels. Past declarations might well count for little. Such a promise could not be relied upon. Few would find it reassuring during the early stages of a crisis involving nuclear powers to be told that they should not worry because such a declaration had been made. They would worry that the promise might not be kept. As a statement of intent, however, it could still have a purpose, to signal that there was no absolute reliance on first use threats and that every effort would be made to avoid escalation.
Short of a major shock, such as a nuclear accident or actual use of a nuclear weapon, the level of interest in disarmament and arms control among the major powers is unlikely to increase of its own accord sufficient to lead to major adjustments, much less to achieving the goal of a nuclear-free world. Even the advocates of a treaty proclaiming a nuclear weapons ban recognize that it would not have any immediate impact on the nuclear states, and any longer term success would be dependent on subsequent efforts by the non-nuclear states and disarmament activists to keep up the pressure. Thus, overcoming the present inertia will be a massive political challenge and will require a reappraisal of the ways and means to generate and sustain the same sort of positive momentum that might otherwise be made only as a result of a more catastrophic shock.13 The trend was in the opposite direction as the arms control regime that seemed so strong in the years after the Cold War was slowly dismantled. Arms control was no longer a major restraint in policy. Citing cheating by the Russians, the Trump Administration announced its intention to pull out of the INF Treaty. If no effort was made to extend the 2010 New START agreement after 2021 there would no longer by any restraints on the numbers of US and Russian strategic nuclear weapons.
Looking ahead, a number of impediments are likely to hinder further progress in the field even though it is precisely these impediments that should be spurring a great deal of intellectual activity. With the US long-term modernization of its nuclear arsenal, and similar efforts in other major nuclear weapons states, the window for any major policy adjustment for a generation would seem to be closing. And with little prospect of shifting policy in any significant way, there is little incentive for theoreticians to dedicate their efforts towards conceptual innovation in this field. Instead, more emphasis will probably be placed on reverting to the ‘safer ground’ of justifying the status quo or focusing on more ‘lucrative’ emerging fields such as cyberwar. The same also applies to the increasing complexity of nuclear arms control, with the advent of more precise and long-range conventional munitions, missile defences, drones, hypersonic missiles and cyber-weapons—in addition to all the longstanding political and technical impediments—that are almost certain to complicate future negotiations. More theory is needed to account for the impact of these technological developments and yet it is precisely the complexity they have generated which seems to have discouraged serious efforts to engage with the topic both among academics and policymakers. Nor has the more general problem about the permanence, or lack thereof, of Cold War-era norms of international behaviour and the implications for nuclear weapons received as much attention as it probably merits. Social attitudes to nuclear weapons will evolve and old norms will eventually be challenged by a new generation of policymakers putting the nuclear taboo under pressure.14 It may be the case that in the years and decades ahead the old ‘rules of the game’ will no longer apply, which will lead either to attempts to reinvigorate old norms or to create new ones. Alternatively, it may be the case, as some scholars have argued, that the presumed norms have been a myth all along and therefore are an irrelevant consideration.15 In a survey in which a modern version of the Hiroshima scenario was employed, in this case risking the lives of some 2 million Iranians to save the lives of 20,000 American troops, there was a majority ready to contemplate nuclear use.16
A related issue is the degree to which nuclear weapons are still viewed as ‘prestigious’ and have political, diplomatic or military utility relative to their cost. Similarly, whether making nuclear threats, a relatively common occurrence in some countries despite the absence of their use since 1945—that would ostensibly devalue them—continues to serve any useful purpose, or supports or undermines credibility, is another topic that might also receive more attention.17 Of course, nuclear threats can take many forms. When contemplating the value of nuclear threats it is important to distinguish between nuclear states making nuclear threats and non-nuclear states threatening to cross the nuclear threshold. Nuclear threats can serve many purposes and be aimed at many constituencies. Mere mention of a nuclear capability by a senior official can be a reminder of national greatness for a domestic audience, or it may be intended to deter someone else’s action. The threat of a state going nuclear might also produce diplomatic concessions that might not otherwise be obtained. Moreover, the fear of a nuclear state failing can provide added leverage in their demands for economic concessions.
As has already been pointed out, nuclear strategy has the defects of its qualities. At the heart of this is the problem of rationality. Since 1945 an enormous amount of effort has been placed on how to use nuclear weapons to achieve political or military ends, yet they have not been used again in anger. To the extent that means of war could be labelled as ‘rational’ or ‘irrational’, employing conventional weapons was implicitly cast as rational whereas nuclear weapons—regardless of their yield—were stigmatized as irrational. In contrast to rational means being used to achieve rational ends, rational ends could not be justified by using such seemingly irrational means as nuclear weapons, so that ultimately only irrational ends pursued by irrational leaders was likely to lead to their actual use. Notwithstanding the rational use of irrationality—a topic on which nuclear theorists had much to say—the world has fortunately not experienced the irrational use of these irrational means. Barring any changes in international norms, use of nuclear weapons will still be deemed irrational to the extent that no rational ends could be achieved by employing them. Policymakers of all nuclear weapons states have, thus far at least, always been cognizant of the likely stigma of being the first country to use nuclear weapons since Hiroshima and Nagasaki and the uncertain and probably catastrophic consequences of such an act. In this sense they are self-deterred.
In general, the costs and risks of possessing nuclear weapons have long since been internalized by the vast majority of states, so that only a few outlier states may still aspire to acquire a nuclear arsenal. By contrast, most countries remain content with investing in non-nuclear forms of defence. Nevertheless, it remains striking that at this stage of the nuclear age many of the more persistent conflicts of international relations have a prominent nuclear element or have nuclear weapons at their heart. A succession of crises—over Iraq, Iran, North Korea—have been about access to nuclear weapons.
It also remains the case that it has proved easier to find rational, or at least, sufficient reasons to justify the continued possession and modernization of existing nuclear arsenals than to find compelling reasons to eliminate or decrease their size and not seek further technical improvements in the relevant systems. Put another way, rational arguments may be necessary but have not been sufficient to overcome political and bureaucratic lethargy. But lethargy is not strategy, though it does often demand the veneer of strategy. Reasons are needed to explain why the status quo, or some variation of it, must be preserved, regardless of whether or not these reasons stand up to serious scrutiny.
Looking into the future, there will be an ongoing need for a general appreciation of the role of nuclear weapons in international relations, their relative importance compared with other trends in warfare, such as the growing prominence of cyber weapons, and the role they may or may not play in future wars. The same is also true of the interaction of technology and social trends.
The big question remains how will the changing international system impact on nuclear weapons, especially the way different actors will choose to make best use of them. Will they be marginalized in favour of less destructive, more ‘acceptable’ means of waging war and deterring or compelling adversaries? Should the United States become indifferent to the alliance obligations it took on in the 1950s then debates will be triggered among its allies as to whether national security now demands national nuclear programmes. The weakest link in the whole nuclear order remains extended deterrence, the requirement that the United States accepts the risk of nuclear war on behalf of its allies. The fact that these guarantees have stayed in place for over sixty years is impressive, but it is a long time since they have been tested, and their foundations are becoming more fragile. The impact of new technologies will continue to test nuclear strategists but the most important challenges will always be political. Nuclear strategy has not reached a dead end because international relations have not achieved long-term stability.