http://www.thenation.com/disarmament/index.htm

By Jonathan Schell, The Nation

|| Bruce G. Blair ||

The Horizontal Path

During the cold war, the number of nuclear weapons in U.S. and Soviet arsenals first grew exponentially and then, after reaching their high-water mark of some 60,000 to 80,000 warheads in the mid-eighties, began to drop. The success of nuclear arms control negotiations was measured, above all, in numbers of weapons. Nuclear disarmament of this kind may be called numerical, or vertical, disarmament. Abolition, obviously, occurs when the number falls to zero. However, abolition so conceived in a world of nuclear deterrence gave rise to a doctrinal crisis, for if deterrence theoretically placed a ceiling above which more nuclear weapons were useless, it also placed a floor beneath which it would be dangerous to reduce them. If safety depended on possessing nuclear forces, then reducing them too far--not to speak of abolishing them--was bound to seem perilous. In fact, as long as deterrence doctrine guides military policy, abolition is impossible.

      In the doctrinal crisis lies one of the reasons for the immunity that nuclear arsenals have shown to otherwise powerful winds of political change. It also helps explain why there are, among dovish arms controllers, so many proposals for "minimum deterrence" or "deep cuts," in which numerical reductions are allowed to proceed very far but then--at the last minute, so to speak--are pulled up short, leaving perhaps a thousand, or a few hundred, or maybe even as few as a couple of dozen nuclear weapons to serve in the deterrent role. For to go to zero would be to dissolve deterrence and thus dissolve security. It is this fear of excessive reductions that also sets the stage for the most alarmist statements regarding breakout, in which a nation shorn of its retaliatory nuclear force is portrayed as helpless in the face of just a few nuclear weapons. In the land of the disarmed, according to deterrence theorists, the possessor of one nuclear bomb is king.

      If deterrence cannot countenance abolition, then some other strategic or political framework for disarmament seems to be needed. But what is that new framework to be? The current debate can be seen as a many-sided search for an answer to this question. Wilsonian collective security--or even world federalism--is one possible candidate. But as the former world federalist Cranston acknowledged to me, that venerable, noble, quintessentially American idea is in eclipse at present.

      It is into the space between deterrence and full-fledged Wilsonianism that a younger generation of analysts has stepped to offer a new approach to nuclear disarmament that we might call the horizontal path. Although they do not constitute a self-conscious school, their thinking, taken as a whole, contains many of the elements of a new conceptual framework. If vertical disarmament involves lowering the number of weapons in nuclear arsenals, horizontal disarmament involves progressively standing down, dispersing, disassembling or partially dismantling arsenals. Establishing ceilings on nuclear arsenals, abolishing certain classes of weapons and drawing down the number of weapons are steps along the vertical path. "De-alerting" weapons, "de-mating" warheads from delivery vehicles, storing warheads at a distance from delivery vehicles, removing parts from warheads or delivery vehicles (or adding parts that spoil their performance) or adulterating weapons-grade fissile materials are steps along the horizontal path. Vertical disarmament makes a catastrophe, should it ever occur, smaller. Horizontal disarmament makes a catastrophe of any size less likely to occur. The verticalist looks at the size of arsenals. The horizontalist looks at their operation.

      Vertical disarmament has reached its destination when the last nuclear weapon has been destroyed. The final destination of horizontal disarmament is harder to define. Absolute zero along the horizontal path would, technically speaking, consist of the disassembly of every last component of every last nuclear warhead and delivery vehicle, leaving in existence only the bare knowledge of how to rebuild nuclear weapons. This knowledge is, in the last analysis, the bedrock of the nuclear age, which neither disarmament nor any other human act can remove. The perdurability of this underlying knowledge is the irreducible basis for the danger of cheating or breakout in a nuclear-weapon-free world. As the possessionists see it, the country that uses this knowledge to violate an abolition agreement stands to gain an insuperable advantage. The violator, they seem to suggest, would put itself in the position of the United States in August 1945, while those obedient to the treaty would be in the position of Japan at that time. To such pessimism, however, the horizontal disarmer offers an answer: Because that knowledge must remain in the world, it is possessed just as much by those who obey the agreement as by those who violate it. In obeying the agreement, they do not render themselves helpless before their enemies. Both cheater and victim remain in the nuclear age. Those who have remained faithful to the treaty retain the capacity, if all else fails, to reconstitute their own nuclear arsenals. The violator's advantage, therefore, would be at best temporary.

      In actuality, it would not be feasible to disassemble all the devices that could be used as parts of nuclear weapons or delivery vehicles. Many of them are deeply embedded in civilian life or in arsenals of conventional weaponry. If abolition were defined in a treaty in terms of precise levels of disassembly, a great many levels, ranging from de-alerting and de-mating to prohibitions on virtually all nuclear technology, including nuclear power plants, would be imaginable. Yet while each step along the horizontal path would further attenuate the danger of nuclear holocaust, there would never come a point at which, through technical measures alone, nuclear danger finally disappears. Like the possessionist, the horizontal abolitionist knows that nuclear danger has descended upon the world forever and draws conclusions from that fact.

      Vertical and horizontal disarmament are not mutually exclusive. They are, to a certain degree, two angles of vision on the same process. Every weapon that has been "reduced" has also been "disassembled" to one degree or another. (For example, all arms-control agreements so far have pertained solely to delivery vehicles. No warheads have yet been destroyed; they have only been recycled or stored. This does not make these agreements fraudulent. It only means that the "reductions" they secure are of a more horizontal nature than common parlance might suggest.) Likewise, when a nuclear weapon has been (horizontally) dismantled to a certain point, it would not be wrong to say that it has been (vertically) "eliminated." Nevertheless, the vertical and the horizontal approaches to nuclear disarmament are substantively different. It's quite possible, for example, to imagine disarmament agreements that leave a large, horizontally disarmed arsenal (many warheads, all kept under inspection in locations distant from any delivery vehicles), just as it is possible to imagine ones that leave a small, fully constituted arsenal (few warheads, all on full alert).

Bruce Blair

In the early seventies, when Bruce Blair, now a senior fellow at the Brookings Institution, was serving as an Air Force launch control officer for Minuteman nuclear missiles in Montana, he noticed a discrepancy between the declared nuclear policies of the United States and the drills he was being called on to carry out as a military man. Policy of course specified that for the purpose of deterring a Soviet nuclear attack a retaliatory force would be held in readiness, capable of surviving the first strike and then devastating the Soviet Union. Blair, however, found that he was almost never called on to carry out a drill in which he fired off his missiles after sustaining a full-scale Soviet attack. Instead, he was ordered to practice drills in which no Soviet attack had yet occurred. It appeared that either the United States was launching first, pre-emptively, or else on warning of an oncoming attack. This discrepancy between policy and practice prompted him to embark on a path of investigation and discovery that has continued to occupy him to this day, turning him, in the words of The Washington Post, into the country's "leading expert" on nuclear command and control.

      It was important for the character and duration of Blair's investigation that his interest in the nuclear question originated in a missile command post rather than in, say, a graduate seminar. In a field ruled by theorists, he is a rare empiricist. In a world of scholastic--or even, as some have called it, "theological"--deduction, he is one of the few who intrude Baconian induction.

      To begin with, instead of asking what the requirements of deterrence were, Blair asked what it was that the United States was actually doing. On-the-spot experience led him to two concerns that became lasting and fruitful interests. The first was with nuclear "safety," by which he meant not just safeguards against accidents but the integrity of the nuclear force's command and control structure from top to bottom in both the United States and the Soviet Union. The second was with the management and operation of nuclear arsenals.

      In the company of a handful of scholars and writers (including Desmond Ball, William Arkin and John Steinbruner), he began to piece together a previously unavailable picture of the U.S. nuclear arsenal and its deployment. In a backhanded tribute to his work, the Pentagon demanded and won exclusive custody of a top-secret study that he wrote for the Congressional Office of Technology Assessment. His study became, as a Wall Street Journal headline put it, "The Ultimate Secret: A Pentagon Report Its Author Can't See."

      Nuclear practice, he discovered, had departed in fundamental ways from the doctrines that supposedly guided it. In his first book, Strategic Command and Control, which appeared in 1985, he demonstrated that while the United States was in theory prepared to wage "protracted nuclear war" involving repeated, carefully designed, limited strikes interspersed with pauses to allow for diplomatic activity, in operational fact no such capacities existed. Owing to the extreme vulnerability of command and control, it was doubtful that the United States could deliver even the single prompt, massive retaliatory attack that is the sine qua non of deterrence.

      In a second book, The Logic of Accidental Nuclear War, based in part on conversations Blair had in the Soviet Union and its successor, the Russian Federation, he showed that Moscow's command and control system was, if anything, even more vulnerable than Washington's. He then concluded that "retaliation after ride-out was an abstract idea in the theory of stability but not a viable option in the real world." Hence, to those who say that in the cold war "deterrence worked," readers of Blair must answer that, technically speaking, it never quite existed.

      The vulnerability of command and control networks, though rarely mentioned in public debate on nuclear strategy, was well-known to high-level military officers, who quietly sought to take remedial steps. They adopted plans to strike back in the short interval between the moment warning of an attack was received and the moment the first missiles landed. Launch-on-warning, Blair writes, was a sort of halfway house between pre-emption (a form of first strike) and rideout. However, its adoption drastically compressed decision-making time for nuclear retaliation. The estimated time from the liftoff of Russia's intercontinental ballistic missiles to their detonation on U.S. soil is about half an hour, while for submarine-launched ballistic missiles it is a mere ten to twelve minutes. Detection of such an attack could come half a minute after liftoff. After some five to ten minutes a missile attack conference, including the President and other high-level officials, would be called. If the attack originated from a submarine, the conference might come to an untimely end a minute or two later. If it originated from land, twenty or so minutes would remain before detonation, of which about three would be available for presidential decision-making.

      In sum, the increasing accuracy of missiles and the growing explosive power of nuclear weapons have menaced nuclear command and control with "decapitation," in the language of the trade; the threat of decapitation has forced both sides to adopt launch-on-warning policies; and launch-on-warning has perilously shortened the time available for evaluating warnings of attack and deciding what to do in response. When both sides in a confrontation adopt this stance, the "logic" of accidental nuclear war comes into play, increasing, in a crisis, "the chance for military operations to overrun the intentions of the political leadership and cause the unpremeditated use of nuclear weapons."

      Surprisingly, these "technical" perils have actually worsened since the end of the cold war, owing to the steadily increasing accuracy of missiles and other military miracles of the information age. Blair did not suggest that any of this was likely in itself to bring on war between Russia and the United States. But he noted the frightening independence of technical developments from political ones, and he drew a lesson that transcended the cold war. If Russia and the United States, two superpowers, were unable to protect their nuclear forces sufficiently to support a policy of mutual assured destruction, how could smaller, poorer nations that manage to obtain nuclear arsenals (say, Israel and some future nuclear-armed Arab antagonist) be able to accomplish that feat? "An important lesson of the cold war," he writes, "is that when archenemies acquire the ability to deliver nuclear weapons, each will fear command and control decapitation."

      If the principal weakness of deterrence was the vulnerability of command and control and the consequent hairtrigger on which the two superpowers placed their arsenals--a posture that has survived the cold war itself--then the most pressing need in nuclear arms negotiations, Blair believes, is to relax these postures.

      In a recent short book of proposals, Global Zero Alert for Nuclear Forces, he begins by drawing a distinction between the needs of deterrence and the need for nuclear safety. In a word, deterrence requires that nuclear weapons be ready for prompt use while safety requires that their use be made difficult. During the cold war, Blair observes, the needs of deterrence were understandably uppermost in planners' minds and safety was sacrificed. But now "the problem of nuclear security needs to be reframed with safety at the center." "Safety," in Blair's thinking, is a concept that can be expanded almost without limit.

      As conceived at present, arms control tends largely to overlook the issue of safety. The traditional approach to arms control, Blair observes, involves "diminishing numbers of weapons but no slowing of their reaction time." On the other hand, a purely horizontal approach to nuclear disarmament might leave in place large, disassembled arsenals. But in fact, as Blair acknowledges, there is no need to choose between the two paths; the logical course lies in the direction of ever-smaller, ever-safer arsenals.

      After leaving the Air Force, Blair went to Yale, where he put a background in applied mathematics to use studying command and control. I asked him to describe how American plans had developed at the operational level.

      "The whole system was geared to launch-on-warning," he told me. "Could a President override that? Could he stop that? It's an open question. I think it would have been very difficult. Furthermore, there were provisions made to delegate authority down the chain of command into the military sphere in the event of a breakdown in communications, as almost certainly would have occurred."

      "How did you turn to the question of Soviet command and control?"

      "Soviet strategic rocket force officers and general staff officers got copies of my book and were astonished at the detail on command and control and astonished at the conclusions I had drawn. They began to seek me out to talk about these topics. Gorbachev's science adviser set me up with some real experts. Then, of course, glasnost came along and people ready to talk about this stuff came crawling out of the woodwork. It turns out that they had drifted into the same posture of launch-on-warning that we had, and for the same reasons, except that they were about ten years behind us. They explained to me that, sure enough, their President had a three-minute constraint on his decision-making. Launch-on-warning is today the dominant option in Russian strategy. So now I had the other half of the equation fleshed out. The Russians were telling me, ‘This is really crazy--we're both in this posture.'"

      His findings, I commented, seemed to suggest that nuclear arsenals had an internal momentum highly resistant to interference from without.

      "One of the frightening things about the end of the cold war," Blair replied, "is that these dangerous configurations of forces seem to be divorced from the political process and immune to political manipulation. It's also a little disconcerting to think that there has developed such a sharp separation between the purposes of nuclear forces and the immediate threat that they were supposedly designed to counter that you can now hook up these same forces to vague possibilities that lie ahead. This gives rise to what you could call a sort of ‘virtual deterrence'--deterrence of some threat that the future may hold."

      "How would the universal de-alerting of nuclear forces, which you call global zero alert, help?"

      "For one thing, it would bring deployments better into line with policy. It would also help relink nuclear forces with politics. It would be something positive to do in the military sphere that would correspond to the harmony that has developed between Russia and the United States in the political sphere. At a stroke it would remove the dangers of quasi-intentional or intentional nuclear war that stem from the launch-on-warning postures of the two sides. The fears of each side that the other might launch an annihilating bolt from the blue would be almost entirely removed. In addition, given the disintegration of the Russian military, de-alerting offers relief from the danger of unauthorized or accidental or inadvertent use."

      Opponents of de-alerting have argued that it would create an inherently unstable situation in which one side might race back to full alert and launch its arsenals before the other side could catch up. Blair responds, however, that stability could be assured by keeping in place invulnerable de-alerted forces, submarines being the obvious candidates.

      An important benefit of de-alerting forces would be "immediately to bring other nuclear-armed countries into the dialogue," he said. For whereas numerical reductions are unlikely to draw in the lesser nuclear powers until U.S. and Russian arsenals are reduced from their present thousands to hundreds of weapons, proposals for zero alert require their immediate participation. (Russia and the United States might not fully de-alert their forces, for example, unless China, France and Britain do likewise.) "Finally," Blair went on, "de-alerting can happen almost instantly. We don't want to move toward zero in twenty-five years, with the last missile on hairtrigger alert--and, if present strategic thinking prevails, that's what we're looking at. There is, in fact, a significant precedent for de-alerting. In 1991, Bush decided to de-alert all bombers, 450 Minuteman II missiles and the missiles in ten Poseidon submarines. Gorbachev followed suit by deactivating 500 land-based rockets and six submarines. The moves took only a few days. If you or I had proposed these steps, we would have been laughed out of town. But Bush did it and Gorbachev did it."

      "What, concretely, was done?"

      "The crews in the launch fields had their launch keys taken away from them."

      "Like keys you put in a door?"

      "Yes, keys that you put in, and have to turn, as in the ignition of a car. Take away the key and you can't drive the car. They did something else. They went into the missile silo and put a pin into the motor of the rocket--into its ignition. An analogy might be taking the spark plugs out of a car, or inserting some plastic in the gap of the plug so there cannot be ignition. In order to reverse that step, maintenance crews have to go back out to all the silos and pull the pins. Actually, it takes a long time, so a delay is built into the process. These steps were not verified."

      "What else could you do?"

      "There are all sorts of variations on the theme. A next big step would be to separate warheads from delivery systems, and perhaps put them into storage under international monitoring. Steps beyond that would be to dismantle components of bombs and delivery systems. Ultimately you would be left with blueprints."

      Blair has, in fact, drawn on his store of knowledge of nuclear operations to work out concrete proposals for de-alerting each of the legs of the nuclear strategic "triad" of airborne, sea-based and land-based forces. Although the arrangements are technical and necessarily complex, Blair is confident that the job is doable. "The question," he says, "is how to configure forces so that they are survivable and reconstitutable but not readily available--certainly not for quick launch. The idea is to extend the time to reconstitute--from days to weeks to months to years. I think it is very practical, very feasible, if only the military and the agencies of government would put their minds to it."

      A zero-alert plan, if it went deep enough, Blair has written, "is tantamount to nuclear disarmament." If it were accomplished, something that truly deserves the name of safety would have been attained in the nuclear age.