3 Nuclear Weapons and Deterrence in an Embedded Pluralist Order

This post-Western world order will not look like anything we have had before. It may show some superficial parallels with earlier international orders. The existence of several great powers might suggest the multipolarity of the pre-1945 world, and China and the US as the biggest of these may suggest a return to Cold War bipolarity. But these similarities mislead more than they enlighten. In the modern world we have never had a truly global diffusion of power; nor multiple sources of cultural authority and legitimacy; nor a situation in which no great power is either able or willing to dominate the system; nor such a formidable array of non-state actors. Neither has humankind faced collectively such a daunting array of potentially existential shared threats. How will the issues of nuclear weapons and deterrence play in such a novel context? This section proceeds by examining several longstanding themes within the literature on nuclear weapons and deterrence, and relating them to a post-Western world order defined by embedded pluralism: technology, status, proliferation, deterrence logic, arms control, war and the nuclear taboo. To some extent, nuclear weapons and deterrence have their own logics, most obviously in how new technologies affect military options and relationships. But they also depend significantly on the political context, as illustrated by the differences in behaviour during and after the Cold War.

3.1 Technology

During the Cold War, technological change was a huge driver of both nuclear weapons and deterrence. This occurred because of a powerful interplay between rapid technological advances in both nuclear weapons and delivery and detection systems on the one hand, and intense military rivalry between the United States and the Soviet Union on the other. Both nuclear weapons and their delivery systems were at the bottom and middle parts of the S-curve with huge scope for improvement. Shifts from bombers to ICBMs and SLBMs, from single to multiple warheads, and from accuracies of a few kilometers to a few meters, all made huge differences to the logic of deterrence. From the 1990s, this technology driver quieted down very considerably. This was partly because many of the technologies had matured, reaching the top of the S-curve where further improvements were relatively marginal. This was true for bombers, missiles, the nuclear warheads themselves, and the speed and accuracy of delivery. It was also partly because the ending of superpower rivalry both took the heat out of nuclear arms racing, and resulted in substantial reductions in nuclear arsenals. The main areas in which relevant technologies had not matured were ballistic missile defence (BMD) and anti-submarine warfare (ASW), both notoriously difficult technologies in which transformational breakthroughs were difficult and expensive, but not impossible, to find.

In the absence of any superpower, or even great power, nuclear rivalries during the post-Cold War period, the main interest in the technology variable concerned the various catch-up games that minor nuclear weapon states (NWS) played with each other and with the US. Relatively small and crude nuclear forces such as those of India and Pakistan had to replay many of the technological phases of the early Cold War years in order to get to a condition of mutually assured destruction and secure second strike. Israel had also to do the same in anticipation of not always being the sole NWS in the Middle East.

How are these dynamics likely to carry forward into an embedded pluralist world? Since there will be no superpowers, and no great ideological rivalries, there will probably be no replay of the intense Cold War nuclear and political rivalry. It would be reasonable to expect a continuation of the existing pattern of catch-up by China, and continued interest, perhaps more so, by the US in advancing BMD and ASW. The US has more to lose than to gain from breakthroughs in ASW, but BMD also serves its needs against nuclear threats from the likes of North Korea, and possibly Iran.

There are certainly advocates in both countries for taking this path, but the costs would be very high, and active rivalry will be constrained by the fear that adding yet more debris to the cloud of it already embracing the planet will make earth orbital space unusable by anyone.

The existing pattern of catch-up for the smaller nuclear powers will continue as it has been doing for the past couple of decades, the main question being whether the local rivalries can keep themselves stable as their technological conditions evolve. As in the past, technologically more advanced powers might be able to play stabilising roles by providing intelligence and reassurance at critical times.

3.2 Nuclear Weapons and Great Power Status

Ever since they were invented, nuclear weapons have been important to great (and super-) power status. They were central to the US and Soviet claims to superpower status after the Second World War. Britain, France and China all made acquisition of them a high priority to bolster their claims to great power status and their right to be members of the P5 in the UN Security Council. Japan and Germany were excluded from these ranks in part because they did not have nuclear weapons, and despite the fact that their economies quickly outpaced those of Britain and France, and in the case of Japan, also the Soviet Union.

The available evidence about the ongoing strength of this link mainly hinges on India, which has been a longstanding aspirant to great power recognition. India put down an early marker with its one-off “peaceful” nuclear test in the 1970s, and got serious with its multiple weapons tests in the 1990s. The problem for India was its exclusion from the club of NWS by the non-proliferation treaty (NPT). What is interesting is how, as the US became more interested in having India as a counterweight to China, the NPT became less of an obstacle. At some diplomatic cost the US negotiated a nuclear deal with India which gave de facto recognition to it as a NWS, and therefore in effect acknowledge its great power standing. In a world order of embedded pluralism where wealth and power are more diffused, and cultural exceptionalism and differentiation are more legitimate, there may be other plausible aspirants to great power status, most obviously Brazil, but possibly also Germany and Japan (on which more below). If and when this happens, some version of the precedent set by India may well be applied if the other great powers accept the claims of the aspirant(s) as legitimate. The ongoing importance of nuclear weapons to great power status was shown by the complete boycott by the NWS of a UN Conference in March 2017 that was aimed at banning nuclear weapons.

3.3 The Ongoing Problem of Nuclear Proliferation

This link between nuclear weapons and great power status points towards the longstanding problem of how to control the proliferation of nuclear weapons. As Buzan and Hansen note, concern about nuclear proliferation has been the most consistent element in the literature of International Security Studies. The number of likely plausible aspirants to great power status in the post-Western world order is likely to be small, but unfortunately there are several other motives for some actors, both state and non-state, to want to acquire nuclear weapons capability. There is a technological driver in play here too. Nuclear technology is now as old as television, electronic computers, and jet engines. In both scientific terms, and as an applied technology for power generation, both knowledge and technical capability about nuclear technology, as well as stockpiles of nuclear material, have become increasingly widespread. Many countries that do not have, or want, nuclear weapons, have the capability to build them fairly quickly should the need arise. Japan, Germany, Canada, South Korea, Australia, Sweden, South Africa, Brazil and Argentina are only the top end of this list of states with relatively short lead-times should they decide to build nuclear weapons. Others have both the capability and the desire, most notably North Korea and Iran. While many countries would still have lead-times of many decades if they wanted to make their own nuclear weapons, over the decades since 1945, a widening spread of nuclear capability has created the potential for a cascade of new NWSs.

While the existing club of nuclear great powers might well be willing to accommodate a small increase in its own membership by making an exception to the non-proliferation rule for arriving great powers, it is unlikely to look kindly on the spread of nuclear weapons to lesser states for other reasons. China may have been the exception rather than the rule in its assistance to Pakistan's nuclear weapons capability. In this sense there is likely to be a lot of continuity, with the club of great powers continuing to oppose nuclear proliferation beyond its own membership.The reasons for doing so will remain the same: the self-interested one that nuclear proliferation blurs their own status claims, and the collective interest one that the proliferation of nuclear weapons increases the likelihood that they will be used, and is therefore a danger to everyone. If a legitimate link can be sustained between NWS and great power status, that will help declining great powers like Britain, France and Russia stay in the club.

This combination of political sensitivity and technological spread raises some interesting and difficult questions about what “nuclear proliferation” means. The obvious dividing line is between those states that have nuclear weapons (and have conducted a test to prove that capability), and those that do not. There is, however, a very significant grey zone between these two options which concerns the lead-time. Lead-time can vary from decades (for a state with little industrial nuclear capability) to mere seconds (for a state with a so-called “last-wire” option, which enables it to deny actually possessing nuclear weapons, while in practice being able to cross that line almost immediately). States like Japan that have big nuclear industries, advanced space programmes, and big stocks of fissile material clearly have relatively short lead times. Possessing a short lead-time is an attractive option, gaining some of the status and leverage from actually possessing them, while avoiding some of the political pressures from the great powers. Pursuit of a short lead-time might be thought of as a form of nuclear proliferation, and Iran, and in a different way Japan, are the clearest examples of this strategy. States with short lead-times can pursue strategies of “recessed deterrence”, basically saying: “don't push me around because I can become a NWS very quickly if you do”. A world in which many states have short lead-times is clearly much more vulnerable to a cascade of proliferation than one in which lead-times are generally long.

There are four main motives driving actors other than great powers to acquire nuclear weapons or short lead-times: survival, status seeking, the breakdown of extended deterrence, and terrorism. These motives are not mutually exclusive.

3.3.1 Survival

The survival motive belongs mainly to states and/or regimes that have come to be seen as “outlaws” by one or more of the great powers, or by a powerful neighbour, or by international society as a whole. These states fear that either or both of the state and the ruling regime will be attacked by outsiders and overthrown. They have to be capable of mounting credible autonomous nuclear weapon programs, which somewhat narrows the field. Israel is a clear example of this type that long ago opted to build nuclear weapons, and for a short time, apartheid South Africa was also. The most obvious contemporary candidates are North Korea, Pakistan, and arguably Iran. North Korea is threatened both by the idea of a unified Korea dominated by South Korea, and by the fact that its curious dynastic regime is seen as crazy and dangerous, especially by the US, Japan and South Korea. Pakistan fears absorption by, or subordination to, its much larger neighbour India. Both of these two already have nuclear weapons. Iran is not threatened as a state, but its regime is considered extremist and dangerous by some of its neighbours, and also by the US. It has a considerable nuclear capability, and is widely thought to aspire to being a NWS.

3.3.2 Status Seeking

While the principle of status seeking is different from that of survival, they can easily be found in the same place. Iran, Pakistan and North Korea, for example, all have this motive alongside survival. Iran sees itself as a civilizational state and as the heartland and protector of Shia Islam. In some ways, it thinks of itself as a great power, even though it is highly unlikely ever to be accepted as such by that club. Pakistan is reluctant to accept the idea of inferiority to India, and likes also to think of itself as a leading power within the Islamic world. For North Korea, nuclear weapons enhance the standing of the regime by giving it a palpable achievement to show off. It also gives the regime an effective bargaining chip in international diplomacy. In a more decentred world political order, some regional powers that seek more influence in their regions might also be inspired by status motives. This could be especially so in regions that do not contain their own great power, so leaving the job of regional leader up for grabs. Africa and the Middle East are obvious places that fit this description. Aspirations to regional leadership are no doubt part of what influences Iran's interest in acquiring nuclear weapons, and until he was overthrown also motivated Iraq's Saddam Hussein. In a world with no superpowers and mainly autistic great powers, regions like Africa and the Middle East are likely to be left more to their own devices than they have been for a long time. In that case regional powers that command sufficient resources, such as Turkey, Saudi Arabia, and possibly South Africa, might well think hard about pursuing a nuclear option, either to assert their own status or to counter similar moves by neighbouring regional powers. In the longer run, states with high aspirations but lower capabilities, and thus long lead-times, might also get into this game. Nigeria and Egypt are the obvious cases here.

There is room within this motive for interplay between the global and regional levels. If, for example, Brazil were to take up a nuclear option as part of a bid for great power status, history suggests that Argentina would think very hard about renewing its own nuclear option. This case has some resonance with that of Pakistan and India, with the less powerful state unwilling to acknowledge inferiority to its larger neighbour.

3.3.3 Breakdown of Extended Deterrence

During the Cold War, both superpowers adopted strategies of extended deterrence, in which they took their allies under their nuclear umbrellas. This was partly about consolidating their leadership roles in alliances, and partly about reducing the incentives for nuclear proliferation. Soviet extended deterrence to China broke down during the 1950s, but US extended deterrence to its allies in NATO and Northeast Asia continued after the end of the Cold War. During the 2016 US election campaign, however, candidate Trump called this arrangement into question. He denounced NATO, told Japan to get its own nuclear weapons, generally suggested that he wanted to withdraw the US from many of its international commitments, and strongly supported closer ties between Russia and the US. This rhetoric raised fears of abandonment amongst America's allies, and started them thinking about what to do in such a case. In office, the Trump administration has seemingly moved back from openly withdrawing its commitments to extended deterrence. Yet Trump's words nevertheless corroded the credibility of those commitments, and put the question of a breakdown of extended deterrence back on the table for the post-Western world order.

The main states affected by this are Japan, South Korea, and Germany. For all of these, the loss of extended deterrence, or even a weakening of its credibility, would leave them exposed to survival questions. Germany would face an unconstrained Russia. Japan and South Korea would also face the nuclear-armed crazy state in Pyongyang. For Germany, Japan and up to a point South Korea, status logics might also come into play. If Europe was left on its own, then as the strongest state in the region, Germany would have to confront the leadership issue from which the EU and NATO have for long shielded it. If the US abandoned Northeast Asia, then Japan, and up to a point South Korea, would be in positions analogous to those of Pakistan and Argentina, trying to maintain equal status against an overbearing larger neighbour.

3.3.4 Terrorism

As if the issue of nuclear proliferation to states was not difficult enough, there is also the growing concern about nuclear proliferation to non-state actors, particularly terrorist groups. This is not a new concern, but in a more decentred world, non-state actors, both of the civil and uncivil kind, are also beneficiaries of the diffusion of power. There is plenty of evidence that terrorist groups are interested in how they might acquire a nuclear option. That said, the most widespread view is that it would be quite difficult for such groups to do so. But if they were to succeed, the security dynamics of nuclear weapons and deterrence would be radically transformed. Imagine the political consequences of even a single city anywhere being incinerated by a terrorist nuclear weapon, and the magnitude of this threat becomes clear. It is not so much the physical threat, which would be catastrophic only locally, but not globally threatening to the planet in the way in which a superpower nuclear war would have been. It is more about the political threat. If terrorists had, and used, such weapons, then democracy would become untenable, and the necessary security precautions would mean that we would all be living in police states. This, therefore, is a combination of low probability, but extremely high cost, with which thinkers about nuclear deterrence have long been familiar. That said, however, the non-territorial quality of many terrorist organizations makes the deterrence logic developed during the Cold War difficult to apply to them. The lack of territoriality greatly complicates the problem of retaliation, an issue already very much part of the problems of the global war on terrorism. Given that the existence of extremist non-state actors is highly unlikely to disappear, concern about this type of nuclear proliferation will remain on the international agenda, perhaps moving to the top of the list.

3.4 The More Localised Role of Deterrence

For most of the nuclear era, the main focus of nuclear weapons and deterrence was at the global level. During the Cold War, all attention was on the nuclear rivalry of the two superpowers, and deterrence theory grew up in this high-intensity, high-stakes context. After the Cold War, this global focus shrank because the US had no peer rival. Russia still had significant nuclear capability, but neither the motive nor the capacity to challenge the US. China was rising, but had a long way to go before achieving anything like nuclear parity.

To the extent that they existed at all, regional nuclear rivalries were a side-show. By the end of the Cold War, Argentina and Brazil had put aside their nuclear rivalry, and South Africa had walked away from its small nuclear arsenal. The Middle East remained of concern about whether any challengers to Israel's nuclear monopoly would arise. After the Cold War, both Saddam's Iraq and revolutionary Iran bid for that role, but did not succeed – although Iran has clearly shortened its lead-time quite considerably. During the 1990s, South Asia emerged as the first fully-fledged regional nuclear rivalry, with India and Pakistan deploying nuclear weapons against each other, and India also deploying them against China. Increasingly, also, the rise of China and the steady growth of its military nuclear capability, raised the spectre of a peer rival to the US re-emerging. This rivalry is commonly still seen as a kind of replay of the US-Soviet one: i.e. a matter of global scale competition between an established superpower and a rising one. In the perspective of embedded pluralism, this is a mistaken approach. In this perspective, the US is steadily losing its sole superpower position, and China will never get to be a superpower because the wider diffusion of power and legitimacy will prevent any state from having either the relative dominance of resources, or the legitimacy, to take that role. The coming world order is one of great and regional powers with no superpowers. The US-China rivalry thus becomes more of a spheresof-influence competition in East Asia between two great powers, than a competition for global domination.

If some of the proliferation scenarios suggested above come about, then the postWestern world order would be one in which the issue of nuclear deterrence would be dominated by a set of regional concerns. If the US abandons extended deterrence, then Britain, France and Germany on the one side, and Russia on the other, would have to work out their own nuclear deterrence relationship. The same would be true in Northeast Asia if Japan decided that the US guarantee was no longer reliable and that it needed to provide its own nuclear deterrence against China and North Korea. Given the historical tensions between Japan and the Korean Peninsula, that scenario would become even more complicated if South Korea took a similar decision. Each of these regional deterrence formations would be shaped by the particular technological and political conditions that pertained to each of the regions concerned. Some would be quite advanced technologically (Northeast Asia, South America), while others would be starting from basics (Middle East, Africa). The logic of nuclear deterrence was wellworked out during the Cold War, and seems unlikely to change much. But in a variety of regions, the significance and application of that logic will vary quite significantly with local conditions.

3.5 Arms Control

Arms control became a key feature of nuclear deterrence during the Cold War. Partly it was driven by state logics, with governments wanting to avoid unnecessary risks and costs. This was true of most of the arms control agreements between the US and the Soviet Union. But partly also, and especially in some democracies, it was driven by anti-nuclear sentiment in civil society organized into sometimes large campaigns. Anti-nuclear and “peace” civil society movements were not uniformly spread, being particularly strong in Britain, the US, Scandinavia, and Germany but much less so in France. There was also some pressure from within the UN system, where non-nuclear states sought to constrain the NWS and push for the eventual elimination of nuclear weapons. The NPT and some other superpower agreements resulted from a mix of these drivers, while the BMD agreements were mostly driven by state logics.

Cold War arms control had the peculiar feature that most of the practice, and much of the thinking behind it, was based on an assumption of dyadic nuclear relationships. That made it relatively simple to think about parity, stability and suchlike as the basic ideas underpinning arms control. China was sometimes thought of as part of the superpower nuclear deterrence equation, but was of sufficiently inconsequential nuclear capability that it did not much dent this dyadic assumption. After the Cold War, much of the urgency went out of arms control, and some of the agreements made between the US and the Soviet/Union, have decayed (SALT and START) or been abandoned (BMD). The NPT is somewhere in the middle, weakened both by the deal done with India and the failure of the NWS to fulfil their part of the bargain by getting rid of their nuclear weapons.

In the more regionalised scenario of nuclear deterrence just sketched, it seems reasonable to assume that the incentives for arms control will rise again, but more on a regional than on a global basis. As argued above, non- or anti-proliferation against lesser powers may well remain a strong global aim of the great power club, though it is difficult to predict how well a great power club will function in pursuit of such joint aims. During the Cold War, the US and the Soviet Union were able to agree on pursuing non-proliferation even though they disagreed about almost everything else. It helped that they were an exclusive club of only two members. But that sense of urgency suggests that a great power concert against nuclear proliferation by lesser powers might also be possible with a wider membership of the great power club. Such a concert might be facilitated also by the fact that ideological differences are considerably less wide than during the Cold War because all the great powers are in one way or another now capitalist. Civil society seems unlikely to play a strong role generally, partly because the threat of nuclear obliteration is much lower, and partly because regional circumstances vary a lot. Anti-nuclear movements are not allowed much scope in authoritarian states such as Russia unless governments want to promote them. There are no strong signs of anti-nuclear sentiments in India or Pakistan, but civil society could play a strong role in what decisions about nuclear weapons get made in democracies like Japan, Germany, South Korea, Argentina and Brazil. Pressure from the non-NWS via the UN system seems likely to continue, but not to be any more effective than it has in the past.

Beyond nuclear non-proliferation, much else about arms control is more likely to relate to particular regional arrangements. Local nuclear rivals might well find it in their interest to agree on measures to limit force deployments or enhance transparency in order to pursue a joint interest in survival and the avoidance of unnecessary cost and risk. That said, it is unlikely that the essentially dyadic logic of arms control developed for the Cold War will serve well in many of the likely cases. In the more decentred world ahead of us, nuclear calculations will more often be multi-sided, and therefore intrinsically much more complicated. Amongst the great powers, China and the US will have not only to think about each other, but also about Russia. India already has to think not only about Pakistan, but also China. China might have to think as well about Japan. Russia might have to think about not just the US, but also Europe and China. And so on. If nuclear relationships unfold along these lines, then there will need to be some hard thinking about how arms control can work to stabilise such multi-sided nuclear relationships.

3.6 War

Right from its inception, nuclear deterrence has been mainly a strategy for war prevention. Once nuclear deterrence became mutual, it was a pretty simple calculation that all-out wars between nuclear-armed powers would be pointless exercises in mutual destruction without much prospect of victory in any meaningful sense. There was of course some thinking about nuclear warfighting, mainly under the label of 'limited nuclear war', and no doubt there were some extremists in the US and Soviet militaries, and are now in the US and Chinese militaries, who look at this as a serious option. But for the most part, thinking about limited nuclear war was about how to tighten up the logic of deterrence so as to leave no loopholes. There was an inherent tension between preparing options to fight limited nuclear wars and increasing the probability of accidental war or loss of control over the escalation process, and this seems to be a structural logic that will apply also in the post-Western political order, especially where political fear and hatred are deep and strong. Yet after some decades of increasingly complicated elaborations of logic, deterrence theory, by the end of the Cold War deterrence theory had more or less settled on the idea of existential deterrence: that the main constraint on war was the catastrophic damage that nuclear war would do to all its participants. This threat was so big that even low probabilities did not much affect the logic of avoiding war. This general assumption seems likely to remain strong, making war between states with nuclear weapons an irrational choice under almost all circumstances. This logic, however, may well not apply to the more extreme and nihilistic terrorists.

The powerful hesitation that nuclear weapons place on the resort to war amongst nuclear armed states forges a link to arms control consideration by placing constraints on measures to prevent nuclear proliferation. There have been several cases in which one power considered using military action to take out the nuclear capability of another in a pre-emptive strike: the US against the Soviet Union and Cuba, the Soviet Union against China, Israel against Iran, and probably the US against North Korea. Yet such action has generally not been taken, the reason being a combination of the impossibility of guaranteeing a 100% success, and the understanding that such action would probably only re-create the problem in a more intense way some years down the line. Perhaps the only exception was the invasion of Iraq in 2003, where concern over Saddam's apparent progress towards acquiring nuclear weapons was one of the motives, and Iraq did not yet have nuclear weapons. There is no reason why this logic should not continue to apply in the post-Western world order, especially where countries have already acquired some nuclear weapon capability. If true, that means that non-proliferation measures are largely confined to making it more difficult and expensive for countries to acquire nuclear weapons through the restrictions on trade in the relevant technologies and materials that are part of the non-proliferation regime. But if a state is determined to push ahead towards nuclear capability despite such costs, as shown by North Korea, the apparent calculation is that there is not much that even the great powers can safely do about it.

3.7 The Nuclear Taboo

A case can be made that the prolonged non-use of nuclear weapons has created a moral barrier against any state resorting to them. This kind of thinking is complementary with, and reinforced by, a variety of policies of 'no first-use' of nuclear weapons.Although the record shows that since 1945 various states have thought about, and even threatened, nuclear attacks on others, none has actually done so. The question for the post-Western world order is whether this taboo will continue to be influential? The case that it will rests both on the perhaps soft ground that it has become a strong interstate moral norm, and on the firmer ground that the taboo simply reflects a rational fear of getting involved in a nuclear war. Should Japan and Germany become NWS, then it seems likely that they would be strongly influenced by both the moral the material logics. Long-established NWS such as the US, Russia, China, Britain and France might well also have been to some degree socialised into this moral norm. But at the other end of the spectrum lie extremist terrorist groups that might well not be influenced by either the moral or the material consideration. Such groups have certainly taken an interest in acquiring nuclear weapons. Should they acquire them, it hardly seems likely that the nuclear taboo would play any role in their thinking. Their purpose would be entirely instrumental – to use, or threaten to use, nuclear weapons in whatever way suited their political aims. They might also not be much constrained by the material fear of retaliation if they had no identifiable territorial base against which retaliation might be directed. In between these two ends of the spectrum lie local cases of nuclear deterrence where the depth of fear and hatred between the antagonists might well weaken or nullify at least the moral side of the nuclear taboo. Judging by its unrestrained rhetoric of nuclear threats, North Korea seems to be the extreme case. Israel and Pakistan both fear for their very existence against neighbours they perceive as existentially hostile. Between Israel's existential fears, and Iran's messianic mission it is unlikely that moral restraints would play much role.