Scholars attribute conventional violence in a nuclear South Asia to a phenomenon known as the “stability/instability paradox.” According to this paradox, the risk of nuclear war makes it unlikely that conventional confict will escalate to the nuclear level, thereby making conventional confict more likely. Although this phenomenon encouraged U.S.-Soviet violence during the Cold War, it does not explain the dynamics of the ongoing confict between India and Pakistan. Recent violence has seen Pakistan or its proxies launching limited attacks on Indian territory, and India refusing to retaliate in kind. The stability/instability paradox would not predict such behavior. A low probability of conventional war escalating to the nuclear level would reduce the ability of Pakistan's nuclear weapons to deter an Indian conventional attack. Because Pakistan is conventionally weaker than India, this would discourage Pakistani aggression and encourage robust Indian conventional retaliation against Pakistani provocations. Pakistani boldness and Indian restraint have actually resulted from instability in the strategic environment. A full-scale Indo-Pakistani conventional confict would create a significant risk of nuclear escalation. This danger enables Pakistan to launch limited attacks on India while deterring allout Indian conventional retaliation and attracting international attention to the two countries' dispute over Kashmir. Unlike in Cold War Europe, in contemporary South Asia nuclear danger facilitates, rather than impedes, conventional confict.

Current U.S. nuclear strategy identifies new nuclear counterforce missions as a means of impeding the proliferation of weapons of mass destruction. The strategy appears to overvalue these counterforce missions. U.S. conventional weapons can destroy most targets that can be destroyed with nuclear weapons; only moderately deep and precisely located targets can be destroyed only by nuclear weapons. In addition, the benefits of nuclear counterforce-which could include deterrence, damage limitation, and the continued ability of the United States to pursue its foreign policy objectives-are relatively small, because the United States possesses large nuclear forces and highly effective conventional forces. Finally, nuclear counterforce would bring a variety of costs, including an increased probability of accidental war and unnecessary preemptive attacks in a severe crisis. Nevertheless, the case for nuclear counterforce is stronger than during the Cold War, when the enormous size and redundancy of U.S. and Soviet forces rendered counterforce useless. When facing a small nuclear force, the United States may decide to use counterforce to limit damage. Although complex trade-offs are involved, if there are critical targets that can be destroyed only with nuclear weapons, then under a narrow set of conditions the benefits of planning for damage limitation might exceed the dangers. The United States must not, however, rely on nuclear counterforce to support a more assertive foreign policy; doing so would unjustifiably increase the probability of nuclear war.