The disarmament movement represented one stream of thought that took off in the aftermath of the Cold War. It pushed hard to deny claims that nuclear deterrence added value, insisting instead that it only added to international distrust and threatened catastrophe. An alternative stream of thought noted that a nuclear free world might work to the advantage of the Western powers but understood from the start that this was one reason why it might never happen. Members of this school never doubted the utility of nuclear weapons as a source of security, especially for those who might otherwise feel themselves to be at a strategic disadvantage.
In an article that gained considerable attention, largely for its resolute refusal to share the general mood of optimism that surrounded the events of 1989, John Mearsheimer assumed that Germany would become a nuclear power. Then, as the Soviet Union collapsed, he explained why it might make sense for Ukraine to hold on to its nuclear bequest.1 In the event Germany made an explicit renunciation of the nuclear option at the time of the country’s unification in 1990, while Japan, the other defeated power of 1945, continued to insist that it had closed off this option. Nor in the end did Kiev agree that the nuclear component of Ukraine’s Soviet inheritance provided a natural and even commendable way of affirming a new-found statehood. Along with Belarus and Kazakhstan, Ukraine eased out of its nuclear status. As it gained its independence from the USSR, Ukraine adopted a non-nuclear policy. The idea that a state with nuclear weapons would choose to give them up, especially when its neighbour was a nuclear state with historic claims on its territory, was anathema to many realists. One of his critics claimed that when asked in 1992, ‘What would happen if Ukraine were to give up nuclear weapons?’ Mearsheimer responded, ‘That would be a tremendous blow to realist theory.’2
Why then did Ukraine not act in the way realists predicted? The dilemma Ukrainian policymakers faced at the time was stark. Having inherited some 2600 tactical nuclear warheads, 176 ICBMs and 44 nuclear-capable bombers—the world’s third largest nuclear arsenal—what policy options were available? Though the possession of a nuclear capability is often discussed without further reference to the details about what such a capability consists of, the practicalities associated with maintaining a nuclear arsenal were a central feature of Ukraine’s strategic debate. There were also the international political considerations. Disarming would impress Western nations which would provide diplomatic recognition, desperately needed aid and trade, and integration with Europe. By contrast, not to disarm risked turning Ukraine into a ‘rogue state’, thereby increasing the risks of isolation and outside intervention. Domestically, the 1986 Chernobyl explosion generated both popular anti-nuclear sentiments as well as distrust over the safety of the nuclear infrastructure on Ukrainian territory. Another major factor was the practical difficulties of independently sustaining a nuclear force that was not tied to Russia, due to reliance on spare parts, maintenance, early warning and command and control. From this perspective, Ukrainian political independence from Russia meant not being a nuclear power. Ukraine did not possess the safeguard and launch codes to gain operational control over the ICBMs, though it probably could have broken these codes given sufficient time. Economic considerations also played a major role. Upon its independence, Ukraine was facing a major economic crisis and therefore the costs associated with maintaining a large nuclear arsenal, including developing new infrastructure, figured prominently in the Ukrainian calculus. Indeed, the costs would be so prohibitive simply to maintain weapons whose service life would eventually expire and could not be replaced, that there would be little money to fund a conventional army. Moreover, Ukraine was dependent on Russia for energy supplies. To the extent that nuclear ownership could serve to gain additional economic concessions for Ukraine, this was counterbalanced by the risks inherent in being perceived to be blackmailing the West. Russia made it clear how strongly it would respond if Ukraine attempted to become an independent nuclear power. As a face-saving device it was offered a statement—the 1994 Budapest Memorandum—in which the US, UK and Russia guaranteed its security in return for it giving up any nuclear aspirations.3
As Ukraine had a purely defensive foreign policy, the only realistic war scenarios in which nuclear use could be contemplated were ones in which Ukraine was invaded by Russia. In light of the hostilities that erupted between the two countries in 2014, beginning with the annexation of Crimea, much criticism was heaped on Ukraine’s earlier decision to disarm. However, even in the early 1990s, when such scenarios positing Russian aggression were discussed, nuclear deterrence was still viewed as insufficient. It was believed that in such a situation, any Ukrainian nuclear use to resist an attack was almost certain to result in unacceptable devastation of its territory. It was therefore not regarded as a credible deterrent. Moreover, due to Ukraine’s close historic and societal links with Russia, it was hard to conceive of any scenario in which nuclear use would be seriously considered. Consequently, a combination of strategic, diplomatic, financial and technical issues made the costs of retaining a nuclear option far outweigh any conceivable security benefits.4
Ukraine was not the only case where the possession of nuclear weapons was judged to constitute a source of insecurity rather than security. In March 1993 it was revealed that South Africa had acquired a small nuclear arsenal of six weapons but also that it had been recently dismantled. Pretoria’s relatively brief flirtation with nuclear strategy can be traced at least as far back as the mid-to-late 1970s. Then a perceived Communist threat led the leadership to seriously consider the advantages of acquiring nuclear weapons. They also needed to decide whether a demonstration of nuclear capability was sufficient or did South Africa need to proceed to the development of deliverable nuclear weapons. In 1979, a high-level committee recommended the latter option in order to acquire a ‘credible nuclear deterrent’.5 However, apart from this general policy choice, little guidance was provided about the actual numbers and types of nuclear weapons that would constitute a credible deterrent. To the extent that a South African nuclear strategy was worked out, this seems to have been mainly limited to discussions that occurred in the early 1980s within a defence industry working group that consulted with senior government officials and outside experts. This group recommended a graduated series of three steps. In the absence of a significant Soviet or Soviet-backed military threat, South Africa would adopt a policy of nuclear ambiguity. If a military threat did emerge, it would then seek to covertly use its nuclear weapons as a diplomatic means of pressuring allies to intervene. This intermediate option seems to have been derived from the ‘lesson’ South African officials learned from Israeli diplomacy during the Yom Kippur War, during which Israel had reportedly used its nuclear capability to extract military assistance from the US.6 Should allied intervention not be forthcoming, South Africa would then move towards overt disclosure and deterrence. Depending on the precise circumstances, this last step could involve a public declaration of its nuclear capability, conducting a nuclear test, or threatening to use the nuclear weapons on the battlefield. Concerns were expressed, however, that if deterrence failed, and South Africa was obliged to employ nuclear weapons locally, that the Soviet Union would retaliate with nuclear weapons which would mean the end of the country.
Later the policy option of developing thermonuclear weapons was rejected on the grounds that South Africa would never use such a weapon as it would be suicidal. At best, any nuclear use would have been limited to retaliation in response to a chemical or nuclear attack on South Africa from a state or non-state actor. Though such a possibility was extremely remote, due to the way that the Soviet and Chinese threat had been exaggerated within the South African security establishment it had some plausibility.7 Regardless, possession of a nuclear capability was viewed as most effective if it could be used to solicit intervention from allies to help resolve a major security crisis. By the late 1980s an arsenal of half a dozen atomic devices had been constructed.8
Support for a nuclear capability did not emanate from the South African military, which was unenthusiastic about it due to the country’s limited resources and the more pressing need to purchase conventional armaments.9 With the withdrawal of the Soviet and Cuban presence in sub-Saharan Africa during this period, any strategic rationale for retaining a limited nuclear arsenal evaporated. Whether this lack of external threat would have been sufficient to lead South Africa to dismantle its nuclear capability is impossible to say. Other factors probably played a more important role, namely the pressure from the US as well as concerns that as South Africa transitioned from apartheid to majority rule the African National Congress would eventually acquire nuclear weapons, and that this had to be avoided.10
Many other states, such as Japan, Sweden and Egypt, that had long been identified as potential proliferators chose to exercise restraint.11 These cases as well as the special cases of Ukraine and South Africa, indicated the importance of normative factors in international politics, as well as why governments might quite ‘realistically’ avoid acting in a provocative manner at a sensitive time. In particular, proliferation in the former Soviet territory—a region calming down after decades of confrontation—was not necessarily the best test of realism. Mearsheimer viewed the acquisition of nuclear weapons as a tempting option for threatened states with the requisite capabilities, although he did not claim that it would necessarily be stabilising.
The most relaxed take on proliferation came from Kenneth Waltz. Although there was tension between the pessimistic view of humanity underpinning neo-realism and the optimism his position implied,12 he believed that nuclear deterrence had made a substantial and positive difference and that this could be replicated even in the more turbulent parts of the world. Having challenged the universal consensus opposed to nuclear proliferation in the 1970s he saw no reason to change his views with the end of the Cold War. He argued his corner in a vigorous debate with Scott Sagan13 insisting that the international system was still largely shaped by its principal actors who were in the grip of an ‘indestructible’ balance of terror. Despite their best efforts the United States and the Soviet Union/Russia had not been able to escape the logic of this balance and the caution it required in their foreign policies. Because a nuclear war on any scale for any actor would be catastrophic the threat of such a war would always exercise a powerful deterrent effect. It was patronising and ethnocentric to assume that lesser powers would not be able to cope with their nuclear responsibilities or fail to act in their blatant self-interest. The more confident could argue that there was no reason in principle why other groups of states could not set up systems of mutual deterrence comparable to the one which was credited with keeping the peace in Europe during the Cold War period.
Mainstream strategists did not so readily accept that the West’s record of successful deterrence pointed a way forward for others. This was largely because of doubts surrounding the general political stability of other countries and regions. There was some intuitive logic in the proposition that at some point the introduction of nuclear weapons would accelerate instability rather than reinforce stability, and few were inclined to experiment in these matters. There could be no guarantees that even areas of stability would remain so. The position of the former Soviet Union in the early 1990s underlined this point.
Sagan drew attention to the close calls in the nuclear history of the United States—the calls for preventive war, the neglect of critical vulnerabilities, the breakdowns in organizational procedures that could have led to unauthorized detonations. Because of this parallel narrative full of hair-raising stories of false alerts, straying aircraft and temperamental technologies, he would not accept Waltz’s confidence in the capacity of many potential proliferators to build bombs that could be stored safely in peace-time and kept under control at times of crisis and war. All things being equal, Waltz had the stronger argument but Sagan encouraged the critical thought that one can never rely on all things being equal. The prospect of nuclear war may render political leaders cautious, but during the Cold War there was quite a learning process before mutual deterrence appeared at all stable. It took some time before all the reassuring paraphernalia of hot-lines and confidence-building measures was in place. Sagan stressed the extent to which governmental organizations can become predisposed to foolishness, fail to learn from the experience of others (on such matters as nuclear safety and the design of stable force structures) and can simply screw up. Arguably, it was precisely because of these dangers that mutual deterrence worked at all, because otherwise it might be tempting to rely on rational decision-making to ensure that all threats remained bluffs.
Those who were dubious about the inherent stability of mutual deterrence in the US-Soviet relationship saw no reason to assume that other nuclear relationships would be stable, especially in the absence of secure second-strike capabilities. The corollary of the claim that nuclear bipolarity led to stability was that multipolarity would lead to volatility. There could be many variations in alliance configurations, affecting the size and capability of the nuclear arsenals of possible adversaries. The logic of a multipolar system, suggested James Wirtz, would be to add to uncertainty and lead states to rely less on alliances and more on their own increasingly large arsenals to meet their security needs.14 Either states would compensate for this risk in their political relations or else they would succumb to arms races and crisis instability.
The new challenges to thinking about nuclear strategy, therefore, were a function of the changing character of international relations in the aftermath of the Cold War, along with wider questions concerning the changing character of the modern state, the balance between great and small powers and the role of military measures as against economic and diplomatic in the conduct of international affairs. The fluid conditions of the 1990s, and the possible arrival of new nuclear powers, promised a test of Waltz’s theories, as well as a slew of neo-realist claims regarding the continued primacy of great powers in international politics and the role of nuclear weapons as the ultimate guarantee of stability.15
The new situation complicated the established debates on nuclear strategy. Fred Iklé noted that it ‘ought to have made it easier to develop strategies that would complement or, where necessary, substitute for deterrence.’ Yet, he complained, ‘the weapons, arsenals and intellectual mindset that constitute the Cold War’s enormous detritus have obstructed the search for new policies.’16 He and others argued that a second nuclear age had begun which required fresh thinking.17 While the first had been dominated by a focused bipolar confrontation the second involved a range of lesser but still significant dangers. ‘We have slain a large dragon’ remarked James Woolsey, former Director of Central Intelligence, but ‘we live now in a jungle filled with a bewildering variety of poisonous snakes. And in many ways the dragon was easier to keep track of.’18 It was hard to be optimistic that some of the venom would not be nuclear. Much depended on whether these states were aiming to find their own balances of terror with local adversaries or whether their objective was to undermine western security.
By the 1990s, nuclear proliferation was apt to be a symptom of instability as much as its cause. The old arguments for nuclear power as a symbol of modernization or as an alternative energy source to oil had faded, while claims that international prestige depended on a nuclear arsenal had been largely undermined by the success stories of non-nuclear Japan and Germany. The anti-proliferation norm, reinforced by the Non-Proliferation Treaty, remained firmly in place even though the promised best efforts by the established nuclear powers to disarm themselves had produced few agreements. There were other reasons for optimism. In addition to the restraint shown by Ukraine, Belarus, Kazakhstan and South Africa, Argentina and Brazil also moved away from acquiring a nuclear weapons capability.19 It was evident that if states had nuclear aspirations they would need to overcome many non-trivial technical obstacles as well as what had become a formidable political consensus opposed to any new entrants into the nuclear club. The established nuclear powers saw nothing inconsistent with maintaining their own nuclear arsenals, albeit these days at lower levels than before, and denying others the same right. They had the clout to enforce what might be seen as double standards: experience confirmed a view that states who wanted access to western markets and capital could not ‘isolate nuclear policy from these other questions of economic development’.20 New nuclear states were more likely to come from the more unsettled parts of the world. Proliferation was coming to be associated with a sort of international underclass—‘pariah’ states feeling under threat and uncertain of any outside support or else run by militarist regimes. The likely proliferators were countries that had been troublesome for some decades: Iraq, Iran and North Korea. In addition, the long-standing conflict between India and Pakistan would also generate nuclear tensions, and constitute one of the regions where a nuclear war was likely to erupt. And beyond the threat posed by states, there was a nagging fear that terrorist groups would get hold of their own nuclear devices. Unknown until the early 2000s the Pakistani scientist A Q Khan had his own organisation actively promoting proliferation. Early in his career he had been employed by the Uranium Enrichment Consortium (URENCO) in the Netherlands. He used the information and the contacts acquired there to help his own country build nuclear weapons. Then he used his network to assist would-be nuclear powers to acquire basic plans and infrastructure, circumventing export controls and using complex payment systems to confuse foreign intelligence services.21
Although new nuclear powers made their impact in the international system as it developed after the end of the Cold War the key programmes had been set in motion decades earlier. There had been discussions in India, for example, about the benefits and drawbacks of acquiring a nuclear capability from the moment it gained independence from Britain in 1947. In the aftermath of Hiroshima, Mahatma Gandhi rejected nuclear weapons as being immoral. Though continuing to emphasize non-violence, he was not against the idea of using force to defend India. Nehru was similarly hostile to nuclear weapons in public and advocated a nuclear-free world, yet in private he was more pragmatic and unwilling to close off the possibility of acquiring a nuclear capability.22 Indian leaders continued to toy with the idea of developing a nuclear hedge capability, but it wasn’t until the Chinese nuclear test in 1964, occurring just two years after the 1962 Sino-Indian war, that they chose to re-evaluate their defence requirements.
The nuclear physicist Homi Bhabha argued that far from having a prohibitive cost, a nuclear deterrent would represent the cheapest form of defence, although he almost certainly understated the likely cost.23 India had reason to fear an asymmetrical situation, in which China possessed nuclear weapons and India did not. Beijing could then blackmail Delhi without provoking outside intervention. Simply by possessing a nuclear arsenal China could potentially achieve political objectives without war; India would be unwilling to risk a major escalation, especially so long as Beijing kept its aims limited.24 Yet even after the Chinese test there was no immediate decision to acquire a nuclear capability. The government hesitated before abandoning India’s traditional non-nuclear stance. The preferred option was to remain non-nuclear but seek protection by asking other nuclear powers to provide an extended deterrent to counter China. The failure to obtain this extended deterrent was a major factor in the decision to proceed with India’s nuclear program.25
Another key factor in the decision was the question about whether or not to sign the Nuclear Non-Proliferation Treaty. Advocates of a nuclear arsenal, such as the strategist K. Subrahmanyam, rejected any attempt to lock India into a non-nuclear posture. Subrahmanyam talked about a ‘nuclear apartheid’ and referred to the other nuclear powers as the ‘ayatollahs of disarmament’. He argued that possessing a nuclear capability would gain India prestige and allow it to deal with China on a more equitable basis. Having accepted the principle of India possessing a nuclear arsenal, the question arose about what sort of nuclear arsenal it should acquire. Subrahmanyam made the case that a limited arsenal would suffice.26 The final factor that tipped the balance was the 1971 war with Pakistan which led to the creation of Bangladesh. The strong show of support made by the United States on behalf of Pakistan persuaded Prime Minister Indira Gandhi to make a move. She decided to demonstrate the capability while playing down its military implications. In 1974 this was achieved by what was claimed to be a ‘peaceful’ underground nuclear explosion. The government hoped to use the claim that this had a non-military purpose to avoid international condemnation while still making a point about what it might achieve in the future, and also demonstrating that not only developed countries could acquire the technical capacity.27
Despite proving its capability to build nuclear weapons, India refrained from producing them, preferring instead to rely on a policy of non-weaponized deterrence. Until the late 1980s the intense secrecy and civilian constraints imposed on the military’s nuclear roles and responsibilities was a cause of significant frustration for India’s defence policymakers, especially as it became apparent that Pakistan was acquiring its own nuclear capability.
The most prominent military figure to address India’s nuclear strategy during this period was General K. Sundarji, who rose to head the Indian Army during 1986–1988. An admirer of Waltz, Sundarji supported his idea that for deterrence ‘more is not better if less is adequate’. He believed that conventional wars fought by countries without nuclear weapons were more likely than by countries that possessed nuclear weapons. Though well-versed in nuclear matters, Sundarji was appalled at the lack of informed discussion about nuclear matters within the Indian military. In his fictional Blind Men of Hindoostan: Indo-Pak Nuclear War, an Indian general complains ‘The doctrine of uncertainty or ambiguity is intended to keep your potential adversaries unsure of the situation. It certainly does not mean keeping your own top policymakers unsure’.28
Similar to Subrahmanyam, Sundarji advocated minimum deterrence, with nuclear weapons targeted against Pakistani and Chinese cities. He worried about the role of nuclear weapons in land combat, especially in the context of a war with a nuclear-armed Pakistan. In this situation, India’s conventional forces would be at a considerable disadvantage as Pakistani forces could concentrate whereas Indian forces would have to remain dispersed to avoid presenting a valuable nuclear target. Therefore, if one side possessed nuclear weapons, the mere possession was enough to oblige the adversary to operate at a conventional disadvantage. Sundarji’s conclusion was that only nuclear weapons can deter nuclear weapons. To be able to maximise its conventional firepower, it was essential that India have its own nuclear arsenal. In 1985, Sundarji contributed to a high-level report that attempted to define India’s nuclear policy and its associated requirements. The report argued for a minimum deterrent as well as a no-first use policy. Although the report was not immediately acted upon, as the Indian leadership were not yet ready for such a step, its recommendations influenced later Indian policy. In the meantime, both Indian and Pakistani politicians increasingly recognized the risks of conventional strikes targeted against each other’s nuclear facilities. To eliminate this risk, both sides agreed in 1985 not to attack them.29
The next phase of India’s nuclear development was weaponization, albeit without an open declaration. One reason for this shift was the lack of response to a speech by Prime Minister Rajiv Gandhi in 1988 to the UN General Assembly promoting universal nuclear disarmament. Nor was there any change to India’s ambiguous deterrence strategy. It was not until the mid-1990s that India developed an air-delivered nuclear capability, and even then little consideration was given to how it could be utilized most effectively. Thinking about the unthinkable remained confined for the most part to a small handful of politicians and scientists who were not well-versed in the operational requirements for nuclear deterrence.30
In May 1998 India tested nuclear weapons and so became an openly declared nuclear weapons state. On several previous occasions, nuclear tests had been considered, but ultimately rejected, largely due to the likely negative diplomatic and economic consequences. The state of Indo-Pakistan relations was one factor, and the Indians also made specific mention of China when justifying their test. The timing could also be explained by reference to political factors, including the demands of national morale, as much as to any immediate military need.31
Irrespective of the reasons for choosing that particular moment to conduct the tests, openly acknowledging India’s nuclear status led to a more wide-ranging discussion of nuclear strategy than before. In August 1999, a draft doctrine was produced that defined India’s policy as one of ‘credible minimum deterrence’. Emphasizing Delhi’s commitment to no first use, the new doctrine referred to India’s nuclear deterrent as being able to inflict ‘unacceptable’ damage on an aggressor. Similar to earlier Indian thinking about nuclear strategy, there was no emphasis on nuclear warfighting; targets would be cities in Pakistan and China. A revised version of the draft doctrine was formally approved in 2003. There were two noteworthy additions. Reference was made to nuclear retaliation in the event ‘Indian forces anywhere’ were attacked with nuclear weapons. This implied that if Pakistan attacked Indian forces inside Pakistan itself, that this could lead to a ‘massive’ retaliation—an implication that raised credibility concerns. The second addition, which called into question India’s no first use stance, was a reference to nuclear retaliation against a biological or chemical weapons attack.32
The official nuclear doctrine was highly contested within India’s strategic community. The debate pitted those advocating retaining the status quo against advocates of an expansion of the size and reach of India’s nuclear forces. Among the most debated issues was India’s insistence on no first use, and the related problems of survivability and maintaining a credible minimum deterrent, particularly given the doctrine’s reference to a ‘massive’ retaliatory capability. Many Indian commentators on strategic affairs, including senior policymakers, questioned the utility of a no first use policy. Successive Indian governments rejected these arguments and consistently reaffirmed this commitment.33 Pakistan’s development of a tactical nuclear capability posed additional problems about India’s willingness to engage in escalation, with some analysts arguing for de-emphasising ‘massive’ in favour of ‘punitive’ retaliation, and possibly developing Indian conventional capabilities to target Pakistan’s shorter range missiles.
In 1999, the year after both sides had tested their nuclear weapons for the first time, Pakistani troops, pretending to be tribal mujahedeen, intruded into the Kargil section of Kashmir, on India’s side of the Line of Control. Their aim was to force Indian troops out of that area. It took many weeks before the Indians were able to get their army to Kargil and then push the intruders out. Eventually the Pakistanis withdrew. At the time it was judged that they had undertaken a risky operation that had been rejected in the past because they had been emboldened by the nuclear tests, and so assumed that India would not take a chance on escalation by turning this into a major war. The failure in Kargil led to a military coup, with General Musharraf in charge. He continued to support insurgents in Kashmir. In December 2001 a group he had long supported mounted an attack on the Indian Parliament, which almost led to a massacre of Indian politicians. The Indians prepared for war and with further deadly incidents in Kashmir got forces in a position to mount an offensive into Pakistan. With active American and British diplomacy the crisis was defused, with Pakistani promises to end support for the militants and India pulling back.
As a result of these crises, ‘limited war’ became a prominent theme of India’s strategic discourse. Defence Minister George Fernandes explained: ‘The issue is not that war has been made obsolete by nuclear weapons, and that covert war by proxy is the only option, but that conventional war remained feasible, but with definite limitations if escalation across the nuclear threshold is to be avoided’.34 The initial delay in deploying India’s main strike corps in the aftermath of the 2001 parliament attack, combined with the problems inherent in ensuring that any future conflict with Pakistan remained non-nuclear, led Indian defence thinkers to promote the concept of ‘Cold Start’.35 Cold Start would involve creating a capacity for the Indian military to quickly mobilize its forces along Pakistan’s border to undertake limited retaliatory attacks before the international community could intervene. As the Indian units involved would be small—divisions instead of corps—and relatively dispersed, they would make less attractive nuclear targets. Since the concept was first discussed there has been considerable debate about both the intricacies of Cold Start, as well as the extent to which the Indian military has embraced it.36 India’s relatively restrained reaction to the Mumbai terrorist attack in 2008 seemed to indicate it still had few military options to retaliate that would not run some risk of uncontrolled escalation. Regardless of the debate about the theory versus practice of Cold Start, the concept was taken very seriously by Pakistan as it had important implications for its nuclear posture.
In addition to the nuclear challenges posed by Pakistan, the rapid rise of China also drove changes to India’s nuclear strategy and procurement choices. For instance, the ‘China threat’ was used to justify New Delhi’s commitment to building longer-range land and sea-based missiles, thereby giving it more credible second-strike options to threaten Beijing. This justification was questioned on the grounds that China, like India, was guided by a no first use nuclear doctrine and therefore unlikely to initiate an attack. In addition the border disputes between the two countries should be politically manageable. Moreover, it was argued that India’s existing delivery systems already provided it with an assured retaliation capability against many Chinese population centres, while developing enhanced capabilities would create an unnecessary risk of provoking an arms race. Nevertheless, Delhi saw psychological dangers if they were seen to fall behind China in the nuclear field, thus mimicking the reasoning of the superpowers during the Cold War where the fear of nuclear inferiority was often a sufficiently persuasive argument to justify developing the next generation of weapons.37
Whilst the precise origins of Pakistan’s nuclear programme remain contested among scholars, it is generally understood that its shock military defeat in 1971, which cost it the eastern half of the country (now Bangladesh), was a major factor. India’s 1974 detonation added to the sense of urgency. The potential need for a deterrent was recognized much earlier. In March 1965 Foreign Minister Zulfikar Ali Bhutto famously remarked to the Manchester Guardian that if India developed an atomic bomb, ‘Then we should have to eat grass and get one, or buy one, of our own’.38 This remark was provoked by the Indian debate about acquiring a nuclear capability that followed China’s test the previous October. Yet Bhutto also downplayed the usefulness of nuclear weapons by invoking the analogy of Vietnam. Several years later he wrote, ‘the lesson of that war is that a people armed can resist any aggressor … Pakistan’s best deterrent would be a national militia … to support the standing forces in the event of total war’.39 Beyond arguments about deterrence, becoming a nuclear power was seen as a matter of national morale and pride. Pakistan needed to be identified as an equal with India. Fear of asymmetry in the aftermath of India’s 1974 test thus sparked Pakistan’s ‘Manhattan-style’ effort to produce a bomb. Bhutto later wrote from his jail cell, ‘The Christian, Jewish and Hindu civilizations have this capability. The communist powers also possess it. Only the Islamic civilization was without it, but that position was about to change’.40
In the mid-1980s, Pakistan reached the status of a nuclear threshold state and employed nuclear threats to deter Indian military attacks, most notably in response to the large-scale Indian military exercise Brasstacks in 1986–1987. During the 1989–1990 Kashmir crisis, Pakistan implicitly threatened to use nuclear weapons if Indian forces crossed the Line of Control. However, these threats came at a time when Pakistan’s ability to use a nuclear weapon against India was questionable, not only due to the lack of a test, but Pakistan’s military had not seriously considered how it might employ a nuclear weapon. Regardless, India’s restraint during these crises led Pakistani leaders to become confident that nuclear threats could succeed as a deterrent and diplomatic bargaining tool.
The decision to abandon the policy of ambiguity was only taken after India’s 1998 nuclear tests. Prior to this the fear of economic sanctions had induced caution, but after the Indian tests the fear of a perceived nuclear imbalance was greater. Again, as with India, Pakistan’s military began to institutionalize its nuclear planning only after its tests. The Strategic Plans Division was created to consider how to operationalise a nuclear capability, as well as to define the sort of strategy and numbers and types of weapons needed to deter India. Pakistan initially adopted a principle of ‘minimum credible deterrence’. This was essentially a massive retaliation strategy intended to inflict unacceptable damage on several Indian cities. Pakistan also rejected the concept of ‘no first use’ as its deterrent was aimed at countering both an Indian conventional invasion as well as any Indian nuclear threat.
During both the 1999 Kargil War and the 2001–2002 Twin Peaks Crisis, India refrained from large-scale military action against Pakistan. In the latter case, Pakistani leader General Pervez Musharraf claimed that he had ‘personally conveyed messages to Prime Minister Vajpayee through every international leader who came to Pakistan, that if Indian troops moved a single step across the international border or Line of Control, they should not expect a conventional war from Pakistan’. Indian leaders claimed that they were not deterred by this prospect, but that war was prevented by Pakistani promises to control Kashmiri militants. Though Musharraf appeared relaxed about the prospect of war as the 2001–2 crisis was developing, as if the nuclear relationship meant Pakistan could engage in limited actions against India without excessive risk, once it became apparent that India was prepared to invade despite the possibility that the war could go nuclear, he looked for a way out. He then became notably more circumspect, speaking of war as an ‘expensive hobby’, noting the dangers of miscalculation, provocative rhetoric and ‘brinkmanship’.41
It was in response to this constraint on their military response to earlier crises, real or perceived, that India began developing the doctrine of Cold Start. This sparked a reaction in Pakistan. As the Cold Start concept was intended to force Pakistan into a submissive position during a crisis by undertaking military operations below its nuclear threshold, thereby undermining the deterrent value of Pakistan’s nuclear arsenal, Islamabad took its nuclear strategy to the next step. In response to Cold Start, Pakistan replaced its ‘minimum credible’ deterrent with a ‘full spectrum’ deterrent capability in which tactical nuclear weapons would be employed to negate the prospect of a tactical or operational level Indian incursion. The tables were now turned on India. If Pakistan used tactical nuclear weapons to counter a small-scale Indian conventional attack, India would hardly be likely to escalate to a strategic attack on Pakistan.42
The ability of India and Pakistan to ride out a number of crises without full-scale war, let alone nuclear escalation, gives some support to Waltz’s thesis. Students of the various crises point to the importance of the nuclear factor in encouraging caution, although the state of the conventional balance and international diplomacy were also relevant.43 In this case neither side could be confident in a first strike capability, thus leaving open the question of whether a shift in the balance between the two might add an extra element of uncertainty at a time that a crisis had developed as a result of unresolved political issues.44