Michael KreponTriangular Nuclear Competitions

A serious competition between two nuclear-armed rivals is very hard to stabilize. When one rival increases its nuclear capability, the other does, too. Then both rivals feel less secure – even when they possess secure retaliatory capabilities. It’s even harder to stabilize a triangular nuclear competition. Isosceles triangles don’t exist in the nuclear business, and three unequal sides do not make for stable geometry.

Triangular competitions are never static. Gregory Koblentz characterizes three-sided competitions as “trilemmas.” Like two-party competitions, they can only be stabilized when disputes are resolved or set aside, direct trade increases, and rivals tacitly agree to restrain their nuclear capabilities.

Stabilization requires roughly balanced strategic modernization programs, conventional capabilities and national trajectories. These conditions were absent during the Cold War. The triangular competition among United States, the Soviet Union and China was particularly unstable because it involved shifting allegiances. Moscow and Beijing colluded at first, and then became bitter rivals, even engaging in a border clash. Once Beijing acquired a minimal deterrent, it dropped out of the nuclear competition, focusing instead on domestic and economic priorities. Today’s triangular competition among the United States, China, and Russia is also unstable. Russia is helping China to compete, even though Moscow understands that Beijing will pose as much of a strategic concern in the future as the United States.

The triangular nuclear competition among China, India and Pakistan is inherently unstable, with features that were not present during the Cold War. The Chinese and Indian legs of the triangle are growing taller, but unevenly. Pakistan’s leg is shrinking despite the growth of its nuclear arsenal, because of weak social and economic indicators. Pakistan measures its strategic requirements against India, while India measures against both its nuclear-armed neighbors. Even if Pakistan were to drop out of the nuclear competition, which is unlikely, India will continue to measure itself against China. China and Pakistan are becoming closer, while Washington gravitates increasingly toward New Delhi. Now add border disputes and violent extremist groups in Pakistan to ongoing nuclear modernization programs, disparate conventional military capabilities and national trajectories.

Nuclear weapons do not stabilize this geometry. Border disputes are certainly not dampened by off-setting nuclear capabilities. In two cases – India and Pakistan as well as the Soviet Union and China – border clashes occurred after the weaker rival acquired a usable nuclear deterrent. On the other hand, border disputes do not necessarily accelerate nuclear competitions. China did not ramp up its nuclear capabilities because of its border dispute with the Soviet Union; nor has the nuclear dynamic between India and China been affected by their unresolved border. Pakistan, in contrast, has upped its nuclear weapon requirements to deal with a border dispute with a rival than enjoys convention military advantages.

If domestic political compulsions do not permit the resolution of border disputes, the most promising way to stabilize a triangular competition is through direct trade and tacit agreements. The most important tacit agreement available to China and India would be to end aggressive patrolling along their disputed border. The most important tacit agreement for India and Pakistan would be to refrain from inserting or supporting violent extremists in Kashmir and Baluchistan. Tacit agreements not to play with fire in these disaffected regions would be necessary but insufficient to reduce deterrence instability. For example, a quiet Line of Control dividing Kashmir will not reduce the risk of conflict if violent extremists based in Pakistan attack iconic Indian targets elsewhere. To guard against this possibility, the intelligence cooperation between India and Pakistan – agreed in principle but poorly implemented in practice – could help defuse nuclear-tinged crises and military clashes.

Tacit agreements are also possible with respect to nuclear weapon-related programs. All three states are on course to increase their nuclear arsenals. Over the next decade, China and India could decide to place more than one warhead atop single missiles and to field ballistic missile defenses. These capabilities will be hard and expensive for Pakistan to acquire. Increases in deterrence instability will grow proportionately along with the extent to which Beijing and New Delhi decide to embrace multiple warhead missiles and missile defenses. Improved missile accuracy and multiple warheads could lead to increased targeting lists that take on a war-fighting character.

A tacit agreement between Beijing and New Delhi not to field missile defenses, or to deploy them only for narrow missions, could serve useful purposes. Tacit agreements to forego nuclear war-fighting capabilities and to adhere to well-established, non-offensive Chinese and Indian nuclear postures could also dampen deterrence instability amidst strategic modernization programs.

China and India have ample resources for the growth of their nuclear capabilities. Pakistan does not. The wisest choice of the weakest competitor, as the Soviet Union and China demonstrated in different ways during the Cold War, is not to engage in a nuclear competition. Pakistan is on a different course, however, because of prior investment decisions. Even as Pakistan’s nuclear capabilities grow, it will fall further and further behind an India led by governments that are more inclined to compete.

However many nuclear weapons Pakistan has, deterrence stability will be elusive unless Pakistan and India improve relations. China and India have a modicum of deterrence stability, despite their growing arsenals, improved conventional capabilities and economic dynamism because they have set aside their territorial dispute while increasing direct trade and investment. With two strong, risk-taking leaders, they might even be able to address their border dispute. After decades of deferring a settlement, this would come as a surprise.

In contrast, there is little evidence that India and Pakistan will try to resolve the Kashmir dispute, or that spoilers would accept an agreement, even if one could be reached. Absent a Pakistani strategy to adopt Beijing’s approach toward New Delhi, India and Pakistan will face conditions of significant deterrence instability in the years ahead.

Note to readers: A shorter version of this essay appeared in Dawn, a Pakistani daily, on March 3rd.


  1. Jonah Speaks (History)

    In certain statements above, I am not sure whether “stability” refers to arms race stability or deterrence stability. Arms races are simply a waste of resources, better spent on economic growth and development. Deterrence instability can lead to risk of nuclear war, which would be an existential tragedy. In the near-term future, the nuclear war risk is highest between India and Pakistan; lowest between India or Pakistan and China.

    Deterrence stability is a military/technological relationship between nuclear and conventional forces that can reduce risk of nuclear war between potential enemies, but is mostly irrelevant between allies and non-enemies (e.g., U.S., France, Great Britain, India, Canada — it makes no difference whether U.S. could “beat” all four). It should not be confused with the risk of nuclear war, which depends on both deterrence stability and the level of tension or dispute between countries.

    • krepon (History)

      Parse all you want, but arms racing is destabilizing and deterrence stability between serious rivals that could engage in combat is a myth.

    • George William Herbert (History)

      Michael, there is a widespread belief, some evidence, and a lot of models that something akin to deterrence stability exists, although there’s also a lot of information that it’s less stable than thought. As an engineer trained in a field where the things we built (ships) are metastable (i.e., can turn over and be stable upside down, again), I understand the risks of stability with more than one equilibrium. That’s different than unstable.

      I do think that a deep rethinking and reframing of stability and deterrence would be timely.

    • Jonah Speaks (History)

      Expanding on George’s analogy, some ships are more prone to being capsized than others, but no ship is invulnerable to being capsized. Similarly, nuclear deterrence can never be perfectly stable, but some deterrence relationships are more stable than others.

      Between nuclear-armed nations that insist on a high level of tension, the risk of nuclear war can never be zero, but the odds can be impacted by how the nuclear competition proceeds. Land-based MIRVs and tactical nuclear weapons increase the odds of nuclear war and do not provide commensurate benefits. Given that the odds of nuclear war are not zero, reducing tensions and making peace are also of vital importance.

    • krepon (History)

      Shameless plug: Stimson is publishing a collection of essays next month, Deterrence Instability and Nuclear Weapons in South Asia. My essay tries to identify conditions for deterrence stability. Hint: nuclear weapons hurt more than they help.

  2. George William Herbert (History)

    From a stability point of view, I do not think that MIRV ever stabilized anything. It’s a warfighting and economics win; your expensive missile is 2x as expensive but hits more than 2x as many targets. But that assumes you’ve started fighting a nuclear war, in which case pre-war economics considerations are a bit silly. In a crisis, MIRV weapons are attractive preemption targets because of their relative vulnerability per unit offensive power.

    We know why China has been interested in possibly a few, but *very* few MIRV units, along with very few units in total of their strategic nuclear forces. That restraint was admirable and stabilizing. India would do well to think on that, and Pakistan.

  3. Maybach57 (History)

    I believe that the India-China relationship – especially pertaining to atomic weapons is rather stable.
    1) Neither country is growing its atomic arsenal at a pace that suggests an arms race.
    2) Both countries have taken a no first-use policy. This is significant even though China’s conventional forces are much bigger. The same is not true for the India-Pakistan equation.
    3) Both countries have much to lose in case of even a conventional war in terms of disruption to their economies.

    There will be border skirmishes on a regular basis. China has regularly entered territories that are disputed or that it wants to make disputed. Island reclamations in the South China Sea is a case in point. Actions within the Nine Dash Line are another. So, despite the occasional, but regular border incursions, I believe this is a stable equilibrium. It is also to be noted that China has settled its border disputes with Russia quickly, but has not done so against smaller neighbors.

    The India-Pakistan relationship, however, is completely different. What is most surprising is Michael’s comment about sharing intelligence between India and Pakistan. It is pretty clear that Pakistan is not merely unreliable as an intelligence partner (it was not even informed on the Bin Laden raid) and there is a lot of evidence that the ISI has actively participated in terrorist actions against India. The 2008 Indian embassy attack in Kabul involvement has been cited a number of times. Several other incidents including 2014 embassy attack at Heart and 2008 Mumbai terrorist attack have led Indian officials to suspect ISI. Under these circumstances, it would be naïve to think that if India were to share intelligence with Pakistan, any meaningful action will be taken. Just look at the mountain of evidence given following the Mumbai attack. If anything, Pakistan has a vested interest to protect the terrorists in order to shield its involvement.

    As for the India-Pakistan military clashes, they are exclusively in Kashmir – not in the borders of Gujarat, Rajasthan or Punjab. Why? The border firings are really cover provided by Pakistani military for terrorists to cross over to India. Events at Kargil were a more open and brazen provocation.

    Dangers with Pakistani nuclear weapons are twofold:
    1) An escalation that follows another major terrorist attack in India launched by militants based in Pakistan.
    2) Equally dangerous is the possibility that a nuclear bomb or even nuclear material get into the hands of a terrorist organization. Given the successful raids that terrorists have made in Pakistani military bases, this cannot be ruled out. If this were to happen, then not just India, but the western world is likely to be affected.

  4. syed muhammad ali (History)

    The dangers identified by Mayback57 at the end of his post are both peculiar and interesting. The first one assumes that it is in India’s best interest to attack a nuclear-armed Pakistan instead of dealing with the world’s largest insurgency which, despite the immense hue and cry being raised against ISIS, engulfs 223 districts of nuclear armed India. Second, civilian contractors operate over 100 power reactors in the US, the state which received the most destructive terrorist attacks this century on 9/11, which led to the loss of over 3000 lives within a couple of hours. In addition, India operates over 20 civilian reactors, despite 20 of its total 28 states having an expanding and active Maoist insurgency. Its nuclear safety record is also less than impressive. On March 4, 1987, the core of the Fast Breeder Test Reactor at Kalpakkam ruptured, on September 10, 1989 Tarapur nuclear power station released radioactive iodine 700 times more than normal levels, on March 31, 1993, a major fire incident damaged and almost led to a melt down of the Narora Nuclear Power station and on Feb 2, 1995 the Rajhastan Atomic Power station leaked Helium and heavy water into the river. So what lessons an objective observer should draw from these facts. Instead of a holier than thou attitude and a perpetual blame game, US, China, India and Pakistan should collectively work on addressing the root causes not merely the temporary and exorbitant management of the negative consequences, which lead to arms build up, including nuclear weapons. The core issue in this quadrangular strategic matrix is that the US intends to strategically contain China and Pakistan’s nuclear weapons but help prop up India, which is neither acceptable for China and Pakistan nor a realistic or rationale for regional or global peace. The negligence towards the world’s largest active and expanding Maoist insurgency inside India and the large quantity of fissile material being imported, developed, used, transported and stored in many of the unstable Indian states and the huge focus on the small nuclear program of Pakistan is both interesting and peculiar but at the same time seems deliberate and disproportionate. A better collective understanding of the relationship between politics and technology, for common good of the entire humanity, deserves empirical not normative approach.