Nuclear weapons proliferation in South Asia has been a major international concern because of the region’s violent history and because of India and Pakistan’s refusal to accede to the international nonproliferation regime. The debate over South Asian proliferation shifted after 1998 from the question of whether India and Pakistan would acquire nuclear weapons to the question of what effects their nuclearization was likely to have on regional security. Two camps have dominated this post-test debate. The optimist camp argues that given nuclear weapons’ ability to make any war between India and Pakistan catastrophically costly, nuclear proliferation should lower the likelihood of regional conflict. The pessimist camp argues that political, technological, and especially organizational pathologies will make proliferation dangerous, potentially leading to deterrence failure and war despite nuclear weapons’ deterrent effects.
In this chapter, we offer our own views of nuclear weapons proliferation. Ganguly argues that nuclear weapons, contrary to much existing scholarship, can actually stabilize conflict-ridden relationships. Most nuclear pessimists underscore the dangers of organizational pathologies, and highlight the long history of Indo-Pakistani tensions and the existence of ongoing territorial disputes to argue that the introduction of nuclear weapons into the region would inevitably enhance the possibilities of conflict. These considerations, while seemingly compelling, are neither especially relevant to the South Asian context nor likely to render the region more war prone. Instead, careful examination shows that in the South Asian case, nuclear weapons have actually had quite a different effect. They have helped to stabilize an otherwise volatile region by making the potential costs of large-scale war unacceptably high.
Kapur offers a pessimistic argument regarding the effects of nuclear weapons proliferation in South Asia. Kapur’s argument is not based on the organizational factors that dominate leading pessimist analyses of nuclear proliferation. He argues, rather, on rationalist grounds. Kapur’s “strategic pessimism” maintains that states’ relative military capabilities and territorial preferences can conspire to encourage destabilizing behavior on the part of new nuclear states. Specifically, states that are both weak relative to their principal adversary and dissatisfied with existing territorial arrangements may decide that they can challenge the territorial status quo without facing the danger of catastrophic defeat at the hands of their stronger adversary. And they may believe that ensuing nuclearized crises can attract international attention useful to their cause. Such behavior would occur not because of organizational or other pathologies, but because of states’ rational calculation of their politico-military interests. Strategic pessimism thus challenges rational deterrence theory on its own ground showing how nuclear proliferation can increase the likelihood of risky behavior, crisis, and war, even between wholly rational states.
ŠUMIT GANGULY: THE OPTIMISTIC VIEW
A common set of problems afflicts most pessimistic analyses of the South Asian nuclear gyre. They tend to uncritically rely on deductive models and pay inadequate attention to the particular features of the political landscape of South Asia. As a consequence, these analyses, while theoretically sophisticated, have serious empirical shortcomings. For example, the organizational pathologies that Sagan identifies may well have been pertinent to the large, complex, and tightly coupled nuclear forces of the United States and the Soviet Union. But the nuclear forces of India and Pakistan are small, are likely to remain so, and are not tightly coupled. Consequently, the organizational pathologies that afflicted the superpower relationship need not be replicated in South Asia.
Proliferation pessimists also highlight the war-proneness of military regimes because of the putative offensive bias of most military organizations.1 Neither the Soviet Union nor the People’s Republic of China was a military regime. However, they were regimes with professedly revolutionary ideologies and hardly averse to the use of force in international politics.2 Nevertheless, they displayed exemplary restraint when involved in an acute international crisis. For example, the Soviet Union and the People’s Republic of China were embroiled in a territorial dispute along the Ussuri River in the late 1960s. At that time, both states possessed nuclear arsenals, although the Chinese arsenal was in its incipient stages.3 This territorial dispute erupted into war in March 1969 and resulted in considerable loss of life.4 However, it did not escalate into full-scale war and the conflict was brought to a close through bilateral negotiations.
The Ussuri River clashes and their resolution demonstrate that two revolutionary states with risk-prone political leaderships can nevertheless exercise restraint and prevent the onset of a conflict spiral. More to the point, the Soviets who had significantly greater nuclear capabilities, chose not to carry out a preemptive strike on Chinese nuclear forces.5 Admittedly, the Ussuri River conflict constitutes a single case. Nevertheless, it is emblematic of the restraint that was evinced in the midst of a significant border conflict. Consequently, it does not appear unreasonable to argue that similar restraint will ensue in the Indo-Pakistani context. Ultimately, minimally rational national leaders are, above all, interested in national survival. Consequently, they recognize the dramatic destructive powers of nuclear weapons and conclude that no possible political goal can be accomplished through their use, especially when their adversaries possess similar capabilities. Such a realization in turn induces substantial caution when in the midst of a crisis for fear of inadvertent or uncontrolled escalation to the nuclear level. Consequently, governments have every incentive to circumscribe the scope and dimensions of the conflict. In effect, this analysis reaffirms the central logic of rational deterrence theory: namely, that the sheer destructive potential of nuclear weapons forces even risk-prone decision makers to avoid provoking or coercing an adversary in a fashion that could induce it to consider resorting to the use of nuclear weapons.6
My argument’s contribution, however, lies not in this deductive logic, which was pioneered in the work of Thomas Schelling and developed in that of Kenneth Waltz.7 Instead its principal contribution lies in demonstrating that in this particular case the arguments of nuclear pessimists have been proven to be uniformly wanting. Decision makers in the South Asian region have come to fully grasp the awesome destructive power of nuclear weapons; they have learnt from repeated crises of the inherent dangers of an escalatory spiral and accordingly have taken fitful steps to try and reduce the nuclear danger in the region. Whether similar outcomes would obtain in other regions remains an open question. Leaders in other regions may prove to be more risk-prone; their military organizations may have a greater proclivity for the adoption of offensive doctrines; and their control over their weaponry may prove to be less robust.8 Careful study of the South Asian case, however, suggests that these dangers are relatively low and so the robustness of nuclear deterrence is not at question.
SOUTH ASIA’S LONG PEACE
An important debate exists about the causes of the “long peace” that characterized Cold War Europe. The mutual possession of nuclear weapons and the horrific consequences of their possible use, according to one school of thought, can explain the long peace.9 The other school contends that the explanation must be sought elsewhere. Its proponents believe that the terrifying memories of the two world wars effectively inoculated the population of the region to recoil from the prospect of a major war.10
The memories of war do not play a significant role in shaping the strategic behavior of political elites or mass publics in South Asia. All the Indo-Pakistani wars involved considerable mutual military restraint and produced relatively low levels of casualties.11 However, in the South Asian context, the memories of the partition of the British Indian empire in 1947 have profoundly shaped the views of more than one generation of leaders in both states.12 Specifically, Pakistan’s rulers, both civilian and military, have harbored an irredentist claim to Kahsmir, the only Muslim-majority state in the Indian union. Even the successful secession of East Pakistan failed to undermine this claim. For Pakistan’s leadership, Kashmir remains “the unfinished business of partition.”13
Despite Pakistan’s unwillingness to accept the territorial status quo in South Asia, there has been no major war in South Asia since 1971. Indeed, between 1971 and 1989, the region enjoyed its own period of long peace. This peace was based on overwhelming Indian conventional superiority and relative political quiescence in the disputed state of J&K.14 Despite India’s continuing military superiority, the onset of an indigenous ethnoreligious insurgency in December 1989 in Jammu and Kashmir contributed to renewed Indo-Pakistani tensions. Pakistan’s decision makers, sensing an important opportunity to exploit India’s self-inflicted wounds in Kashmir, devised a strategy to transform the insurgency into a well-funded, carefully orchestrated, religiously motivated extortion racket.15 During this period, there was at least one major war scare, in 1990, as bellicose rhetoric on both sides over the question of Kashmir precipitated a crisis. Nevertheless, despite the stakes involved and the high level of ensuing tensions, war was successfully averted. The incipient nuclear capabilities of the two sides played a substantial role in preventing the outbreak of war.16 By the end of the decade of the 1990s India had managed to restore a modicum of order, if not law in Kashmir. Indeed it can be argued that it was the very success of India’s counterinsurgency strategy that promoted Pakistan’s decision makers to pursue a “limited probe” in the Kargil region of Kashmir in 1999.17 In this war the overt possession of nuclear weapons on both sides played a critical role in preventing an escalation or an expansion of the conflict.18
Thus Indian conventional superiority and the absence of any significant political turmoil within Kashmir ensured peace between 1971 and 1989. And after the outbreak of the insurgency, the incipient nuclear capabilities of the two sides induced considerable caution and prevented an outbreak of war despite a bloody, ongoing insurgency in Kashmir in which Pakistan became deeply involved. Multiple crises subsequently wreaked havoc in Indo-Pakistani relations since their mutual acquisition of nuclear weapons especially in 1999 and 2001–2002. 19 But despite intense tensions, none of these crises have culminated in full-scale war. Decision makers in both countries have steadily and increasingly realized that the initiation of a major conventional conflict could, under a number of possible scenarios, tempt one side to consider the use of nuclear weapons. Consequently, both sides have exhibited considerable restraint and have chosen to eschew horizontal escalation and not to violate certain tacit thresholds.
Since the outbreak of the Kargil war, Indian military strategists have been struggling to forge a military doctrine and the requisite capabilities that might enable them to respond to future Pakistani provocation without triggering an escalation to full-scale war. It is far from clear that they will be able to formulate a viable doctrine that could effectively contain the possibilities of an escalatory spiral following the initiation of even limited attacks on Pakistani territory.20 On the other hand, while Pakistan may deem the risks of abetting terror in India both controllable and calculable, it will also be loath to initiate a significant conventional conflict with India. Initiating a conventional conflict of such magnitude could invite substantial Indian retaliation and might push the two states toward a nuclear conflict.21 Pakistan’s decision makers may well be risk-prone; however, in their long history of conflict with India they have not engaged in fundamentally irrational behavior. They have been guilty of flawed judgment, have taken refuge in “false optimism,” and have displayed strategic myopia.22 None of these behavioral traits, however, are unique to Pakistani military and civilian officials.
Furthermore, there is evidence that the politico-military leadership does recognize the significance of the nuclear revolution and the unique properties of nuclear weapons.23 Consequently, they are likely to see their nuclear capabilities as a viable deterrent against India’s present and future conventional superiority. Barring a complete breakdown of the Pakistani state there is little reason to believe that nuclear deterrence will not remain robust in this bilateral relationship and help avert a full-scale war.
S. PAUL KAPUR: STRATEGIC PESSIMISM
The dominant school of proliferation pessimism argues that shortcomings within the organizations that manage nuclear weapons will make nuclear proliferation dangerous. For example, military services will adopt standard operating procedures that increase the likelihood of accident or will indulge officers’ proclivity for offensive, destabilizing strategies. At first blush, this organizational argument appears to offer a fundamental challenge to the claims of proliferation optimism. Organizational pessimism expects proliferators to behave in ways that optimists do not predict, destabilizing rather than pacifying the international environment. Close examination reveals, however, that pessimist arguments in fact accept important optimistic beliefs. Specifically, pessimists do not challenge the optimists’ claim that in a nuclear environment, strategic calculation should lead states to behave cautiously, thereby increasing international stability. Rather, pessimists show that organizational problems may short-circuit strategic policy formulation, preventing proliferators from adopting the cautious, stabilizing courses of action that they would otherwise embrace. Thus, in the leading pessimistic view, newly nuclear states will not behave dangerously because rational calculation or strategic policy formulation might encourage destabilizing behavior. Rather, dangerous behavior will occur because organizational pathologies interfere with strategic decision-making. Thus proliferation pessimism, in its dominant, organizational form, cedes important ground to the optimist camp. It does not challenge the optimists’ fundamental claim that rational proliferators should be cautious proliferators
This is in no way to deny the significance of organizational pessimism’s insights. Such pessimism points up important dangers that are likely to result from the spread of nuclear weapons. My claim is simply that despite its strengths the existing pessimist literature does not challenge proliferation optimism’s core logic. In order to pose a more fundamental challenge to the optimists, a pessimistic approach would have to show that the acquisition of nuclear weapons could encourage destabilizing behavior even if states calculate rationally and behave strategically. I offer such an approach, arguing that even if they devise policy in a largely rational manner, newly nuclear states may decide not to behave cautiously. Instead, under certain circumstances, nuclear weapons can create strong incentives for rational states to adopt aggressive, extremely risky policies. Thus the spread of nuclear weapons can destabilize the international security environment even apart from the organizational logic that has so far driven the arguments of proliferation pessimists. In some cases, rational proliferators can be dangerous proliferators.
How could the acquisition of nuclear weapons create incentives for a state to behave in a destabilizing manner? Suppose that a state was dissatisfied with the location of its borders and militarily weak relative to its primary adversary. In this situation, the state’s leaders would like to alter its territorial boundaries but would fear that doing so could trigger retaliation by the stronger adversary. Such retaliation could result in catastrophic defeat, involving a significant loss of territory or even of sovereignty. Thus the leaders of the weak, dissatisfied state would probably have to live with territorial boundaries that they viewed as undesirable.
Nuclear weapons would change this situation in two important ways. First, nuclear weapons would afford the weaker state a shield against its adversary’s superior conventional military capabilities. If the strong state ever threatened the weak state with catastrophic military defeat, the weak state could respond with a nuclear attack. The strong state could still retaliate in the event of provocation by its weaker opponent. But the danger of a nuclear response would constrain the strong state, making it much less likely to launch a full-scale conventional attack against its adversary. This would not change the actual balance of power between the two sides. Nuclear danger, however, would limit the strong state’s willingness to use its military capabilities. And thus nuclear weapons would nullify a good deal of the strong state’s conventional military advantage. These developments could embolden the weak state to behave in ways that were previously too dangerous. Before acquiring nuclear weapons, the weak state had to fear that attempts to alter territorial boundaries might result in catastrophic defeat. Now, however, the weak state can directly challenge territorial boundaries, encouraged by the knowledge that its opponent is unlikely to employ the full extent of its military power in response.
Second, nuclear weapons could create diplomatic incentives for a weak, dissatisfied state to engage in destabilizing behavior. A nuclear conflict would have catastrophic human and economic effects;24 it would also break the nuclear taboo in effect since the end of World War II.25 The international community is extremely anxious to avoid a nuclear exchange anywhere in the world. Aggressive conventional military behavior that threatens to create a nuclear crisis can attract international attention, including mediation efforts by outside states. Such third-party intervention can result in a territorial settlement superior to any that the weak state could have secured in purely bilateral negotiations with its stronger adversary. Weak, dissatisfied states therefore may have a diplomatic incentive to engage in aggressive conventional military behavior that provokes their adversaries and creates a danger of nuclear confrontation.
It is important to note that proliferation is likely to be emboldening only where both of the criteria that I have specified are satisfied: The proliferating state must be territorially dissatisfied and militarily weak. If the state were satisfied with existing territorial boundaries, it would have no reason to attempt to change them, with or without nuclear weapons. And if the state were strong relative to its primary adversary, it would not need nuclear weapons to facilitate aggressive behavior. It could forcefully challenge existing boundaries even in a purely conventional military environment. Thus proliferation creates incentives for destabilizing behavior by weak, dissatisfied states, but not by those that are militarily strong, territorially satisfied, or both strong and satisfied.
Such provocative behavior by a newly nuclear state would be dangerous. It would invite forceful responses from the state’s stronger adversary, which would seek to defeat ongoing aggression and to deter such behavior in the future. Even though neither side would wish resulting conflicts to spiral to the nuclear level, such escalation could occur. For example, the stronger state’s retaliatory attack might be more successful than either state anticipated, quickly taking territory and threatening the weak state’s nuclear command and control. In this situation, the weak state could delegate launch authority to field commanders or use its weapons before it lost the ability to do so.26 This risk of inadvertent escalation would threaten to make even small-scale confrontations between new nuclear powers catastrophically costly.
Despite these dangers, however, aggressive behavior by a weak, dissatisfied proliferator would not be irrational. Nor would it necessarily be the result of organizational pathologies that short-circuited strategic decision-making. Rather, destabilizing behavior would be the product of a deliberate strategic calculation. A weak, dissatisfied proliferator would challenge existing territorial arrangements in the belief that its insulation from all-out retaliation, and its ability to attract international attention, would afford it a significant chance of achieving its politico-military goals. This means that nuclear weapons proliferation can be destabilizing quite apart from organizational or other pathologies. The structure of relative military capabilities and territorial preferences within a conflict relationship can also create strong incentives for dangerous behavior.
This is not to argue that all weak, dissatisfied proliferators will behave aggressively. The strength of incentives for aggressive behavior will depend to a large extent on a proliferator’s level of dissatisfaction with existing territorial arrangements. The more dissatisfied a state is with current territorial boundaries, the more likely it is to attempt to change them by force. Thus a proliferator that is only mildly unhappy with existing arrangements may not behave aggressively. The level of a proliferator’s relative military weakness will matter as well. A proliferator that is only slightly weaker than its adversary is less likely to need a shield against catastrophic defeat, and thus less likely to be emboldened by the acquisition of nuclear weapons, than a proliferator that suffers from a significant conventional military disparity. But despite these caveats, my basic point remains: Territorial dissatisfaction, conventional military weakness, and nuclear weapons are a potentially destabilizing combination.
I call my approach to nuclear weapons proliferation “strategic pessimism.” Unlike standard organizational pessimism, it challenges the core logic of proliferation optimism, showing that the danger of nuclear weapons will not necessarily lead rationally calculating states to behave in a cautious manner. Instead, under the right circumstances, nuclear danger can create incentives for states to adopt aggressive strategies, thereby destabilizing ongoing conflict relationships and creating a serious risk of catastrophic escalation.
In chapter 4, we apply our theories to Indo-Pakistani militarized disputes from the late 1980s through 2002. The increased number of militarized disputes during the early phase of nuclearization and even the overt acquisition of nuclear weapons, Ganguly contends, were inextricably related to the exigencies of regional politics. Nuclear weapons, far from exacerbating these tensions, actually helped to constrain and limit these crises. None of these crises were allowed to spiral into full-scale war despite high level of tensions and the significant stakes involved in them. The contrast with the prenuclear era could not be more dramatic. In the prenuclear era, both states made very deliberate (if not always careful) plans for war (see chapter 2). However, as both states came increasingly close to crossing the nuclear Rubicon they evinced a growing recognition of the dangers of nuclear escalation.
Kapur, on the other hand, maintains that strategic pessimism is best able to account for instability in South Asia after India and Pakistan’s acquisition of nuclear weapons. As Pakistan developed a nuclear capability, its leaders decided that they could forcefully challenge territorial boundaries in Kashmir, the main source of Indo-Pakistani tension since partition. To this end they adopted policies ranging from the provision of extensive material support to the anti-Indian Kashmir insurgency, to the outright seizure of territory by the Pakistan Army. The Pakistanis believed that their nuclear capacity would prevent India from launching an all-out conventional attack in retaliation for these provocations. They also hoped that the danger of nuclear escalation in any ensuing crisis would attract international diplomatic intervention in the Kashmir dispute. India responded forcefully to Pakistani challenges, and the result was increased regional violence.