One of the least lamented casualties of the Soviet Union's demise was the arcane specialty of nuclear doctrine. Those who had wasted their youth studying the ins and outs of nuclear deterrence -- the peculiar logic of nuclear war and the kinds of policies and weapons that might make it more or less likely -- went the way of the blacksmith in the age of Henry Ford. Their skill was like conversational Latin: Its time had come and gone.

Or so it seemed. Now, thanks to India and Pakistan, deterrence theory is back. And it is needed. Every newspaper and commentator in the country is saying gravely that now that India and Pakistan have acquired nuclear weapons, the subcontinent is an area of great instability. What exactly does that mean?

After all, India and Pakistan have been at each other's throats for 50 years. They fought three wars. They routinely exchange rifle and artillery fire in Kashmir. Yes, the situation is more dangerous today. The stakes are obviously higher. Any war could be fought not just with conventional weapons but nuclear ones. But that does not necessarily mean that the subcontinent is more unstable.

Indeed, nuclear weapons can actually be stabilizing. The United States and the Soviet Union might well have gone to war at some point in the last 50 years had the specter of nuclear annihilation not hung over both countries. In the same way, India and Pakistan might be less likely to go to war if that means not just the loss of a few soldiers on the frontier, but the possible annihilation of one's major cities. This is true, however, only over the long run. In the short run, the nukes are destabilizing -- but not for the reasons being advanced in the papers.

The subcontinent in the near term will be an area of great instability not just because, obviously, Pakistan and India are new at the nuclear game and thus will be prone to miscalculation, but because each side has so few nuclear weapons. This sounds odd. It is odd. But it reflects the central paradox of deterrence theory: Generally speaking, the more nukes the better. There is safety in numbers.

Now, conventional wisdom holds that the way to nuclear safety is to reduce numbers. That was the fuel for the frenzied nuclear disarmament movements of the early '80s and for the American obsession with arms control today. But conventional wisdom is wrong. When the numbers get very low, the nuclear balance becomes unstable.

Consider India and Pakistan. At the moment, each has a very small number of nuclear weapons and an equally small number of missiles on which those weapons might be delivered. In a time of crisis -- say, fighting breaks out over Kashmir -- this presents each side with the opportunity to destroy the other's entire nuclear arsenal in one fell swoop at the beginning of the conflict. Small numbers make for a small target. And a small target is a tempting target. It places a premium on preemption. It rewards striking first.

Conversely, each side fears that if it does not strike first, its nuclear arsenal could be wiped out in a first strike, leaving it naked to nuclear blackmail or to further nuclear attack. Because the side that destroys the other side's arsenal will have some of its nukes left over to intimidate the other's population. That intimidation could be enough to tip the scales of any war or even to induce the aggressed-against to a quick surrender. Very low numbers thus encourage a "use it or lose it" mentality. They promote the single most important contributor to nuclear instability: the temptation to preemption.

Consider the U.S.-Soviet example. Apart from the few days of the Cuban missile crisis, the most unstable period was the late 1940s and early 1950s, when the United States and the Soviet Union were just developing their nuclear arsenals. There was even talk in Washington in the early '50s of destroying the Soviet arsenal before it could be developed. That kind of talk disappeared when the Soviets had built enough rockets, airplanes, and subs -- enough redundancy -- to make a first strike futile.

The essence of nuclear stability is the existence on both sides of a retaliatory or "second strike" capacity. If the other guy has such a capacity, it is crazy to launch a preemptive nuclear attack. No matter how massive or how accurate it is, it cannot succeed in wiping out all of his weapons. There would be enough left over for him to retaliate massively. A first strike would thus bring on the incineration of your own homeland. Result? You don't attack first. Nobody attacks first. And if nobody attacks first, there can be no nuclear war. QED.

That is how nuclear stability is established. Inconveniently for the doe-eyed arms controllers, this tends to happen at high levels of weaponry. More precisely, it is not the numbers that are decisive, but how they are distributed and how invulnerable they are to preemption. Thus the most stabilizing factor in the nuclear equation between the United States and the Soviet Union was the submarine forces. Their nuclear weapons could never be found with any accuracy by the enemy. They could thus never be preemptively destroyed. If either side were foolish enough to attack the other, there would always be the submarines to bring Armageddon on the aggressor. Huge, dispersed, mobile, and hardened arsenals of land-based and air-borne nukes had the same stabilizing effect.

This explains why the nuclear-freeze hysteria of the early '80s was so pointless. It promoted not a panacea, not even a palliative, but a nullity. It also explains why the nuclear arms reduction agreements that so mesmerize the Clinton administration today are largely irrelevant. Yes, they are worth pursuing for their marginal economic savings and for reducing the stock of stuff that, if poorly tended, might be prone to accident or theft. These treaties do practically nothing, however, to enhance strategic stability.

Even worse is the movement today, led by such luminaries as Gen. Lee Butler, former head of the U.S. Strategic Air Command, to bring about (through gradual arms reduction) the total denuclearization of the United States. This idea, trumpeted for its boldness, is simply crazy.

Why? Because the nuclear genie is out of the bottle. There is no way to undo the knowledge of how to make the weapons. Many regimes can potentially make them. The only problem is acquiring the materials and the brains to assemble the devices. As Saddam has shown, a determined tyrant can do this under even the most stringent inspection regime. Remember: Before the Gulf War, Saddam was a card-carrying, paid-up, cooperating member of the Nuclear Non-Proliferation Treaty. He was regularly inspected by the International Atomic Energy Agency. And under the nose of that agency, he built not one but two clandestine nuclear programs. Were the United States to engage in the folly of total denuclearization, it would surely wake up one day looking down the barrel of some nuclear-armed bad actor -- Iraq or North Korea or Iran or who knows what other rogue state of the future. Never in history would a Great Power have voluntarily put itself at such pointless risk.

After all, the ultimate instability -- and vulnerability -- occurs when one side has nukes and the other doesn't. Imagine what the world would have been like in the late 1940s if Stalin had acquired the bomb before we did. (Even worse, imagine if Hitler had.) Stalin might not have used the bomb. But the very fact that he could might have intimidated us into surrendering large parts of Europe, or even more.

The world was lucky that the first nuclear power was as benign a nation as the United States. It allowed us to go through the birth of the nuclear era in the most stable way: unilateral possession by a nonaggressive power. We used the bomb to end a war, not to start -- or win by threat -- new ones. Once the Soviets acquired theirs, however, a period of severe nuclear instability began.

That is where the India-Pakistan balance is now -- about where we were in 1950. And it is U.S. policy today to try to rush in and get both sides to forswear the nukes: no further testing, no weaponization. This would be very nice. It is also very unlikely.

Indeed, it is likely that even if both sides agree, one side or the other will cheat. And it is certain that even if neither side does in fact cheat, both sides will surely suspect each other of doing just that. And when you suspect the other side of cheating, you suspect that it may have a first-strike capacity -- which would spur you to cheat as well and develop a clandestine countervailing arsenal. Indeed, that is precisely what has happened between India and Pakistan -- competing clandestine nuclear programs -- to bring us to the current crisis.

The period of nuclear instability on the subcontinent is beginning. And paradoxically, the situation will only begin to stabilize when both countries have deployed enough nukes -- spread out in enough areas -- that neither side can be sure of a successful first strike. It sounds perverse to say it, but the fact is, now that the race is on, nuclear stability will only come to the Indian subcontinent when the respective nuclear arsenals have grown larger and more mature.

It is not a happy prospect. But it is reality. Welcome once again -- just when you thought it was over -- to the unthinkable world of nuclear deterrence.

Charles Krauthammer is a contributing editor to THE WEEKLY STANDARD.

Next Page