Nuclear Weapons --- by BRIAN HALL

This is an unedited, longer version of "Overkill is Not Dead" which was published March 15, 1998 in the New York Times Magazine

President Clinton, when touting the accomplishments of his administration in the realm of reducing the dangers of nuclear war, has been fond of leading off with "detargetting": the statement that, following an agreement between himself and Russian President Boris Yeltsin in May 1994, the two countries' weapons are no longer pointed at each other, but at the ocean. Perhaps Clinton will have the decency to stop saying this, now that recent news reports have made widely known what informed people knew all along: the statement is deliberately misleading. Or perhaps--a more disturbing but regrettably conceivable construction--Clinton just didn't know what he was talking about.

With all the wartime targets still in the computers, retargetting takes seconds, and in fact doesn't add a single second to the launch procedure because targets have always had to be loaded into warheads before firing (under different strike options, the weapons have different targets). So let's keep it straight: the United States and Russia, eight years after the end of the Cold War, could fire off thousands of nuclear warheads at each other within one to three minutes of your reading this sentence, destroying our two societies and killing tens of millions of people. Since Russia and the United States are no longer ideologically opposed, have no territorial ambitions on each other, and maintain a political and military relationship whose mix of cooperation and mutual wariness resembles that between the United States and France, this is a bizarre situation, to say the least. Try thinking of it this way: the only big thorn in the side of US-Russian relations now is the fact that each country points nuclear weapons at the other. This is a political problem, for which the "prudent" military solution is to keep pointing those same nuclear weapons.

The "detargetting" canard is emblematic of the Clinton administration's great reluctance to put forward even modest initiatives toward ending nuclear brinksmanship and its even greater reluctance to appear lacking in initiative. This stands in contrast to the Bush administration, which achieved a fair amount in four years: the signing of START I, which for the first time cut strategic nuclear arsenals, to 6500 warheads on each side; the parallel unilateral withdrawals, first by Bush and then by Gorbachev, of tactical nuclear weapons stored on surface ships and attack submarines, the pulling back of Soviet short-range missiles into Russia proper, and the removal of all but a handful of U.S. tactical nuclear bombs from Europe; the standing down from alert of United States strategic bombers and the early deactivation of 450 Minutemen II missiles slated to go under START I; and the signing of the START II treaty, which commits the United States and Russia to cutting their deployed strategic arsenals to 3500 weapons each.

START II was signed on January 3, 1993, and Clinton was sworn in as president seventeen days later. Since that time, although there have been accomplishments in global security such as the indefinite extension of the Non-Proliferation Treaty and the signing of a Comprehensive Test Ban Treaty, on the core issue of this surreal nuclear superpower standoff, progress has ground to a halt. The past year and a half has seen the accretion of a chorus of distinguished voices--generals and admirals, including the former head of our nuclear Strategic Command, General George Lee Butler, the National Academy of Science, former defense secretary William Perry, former Secretary of State George Shultz, Jimmy Carter, Sam Nunn, and more than 100 heads of state from around the world--assailing the lack of progress and suggesting various ways to move forward. Yet the Clinton administration seems determined to do nothing until START II is ratified in the Russian Duma, where it has been languishing for two and a half years.

Our window of opportunity may close. US-Russian relations have soured in the past four years, with the stalled progress on nuclear weapons increasing the suspicion about intentions on both sides. There are also proliferation concerns. The bedrock on which the Non-Proliferation Treaty is built is the obligation on the part of the nuclear powers to negotiate "in good faith" toward deep reductions in arsenals and the eventual abolition of nuclear weapons. The United States' periodic statements about its need to retain a nuclear deterrent into the foreseeable future, and worse, its public justifications of nuclear weapons as a deterrent against chemical and biological weapons undermine non-proliferation at a time when the illicit acquisition of weapon's grade material is probably easier than at any time during the Cold War.

Moreover, the hair-trigger posture itself could result in a catastrophic accident. The hair trigger has been around for about 25 years--think of the 30-odd years of nuclear power before Chernobyl--and during that time several alarming incidents have occurred. When a computer chip failed in 1980, news of an all-out Soviet missile attack flashed over the consoles at NORAD, the United States' command post in Colorado for assessing nuclear attacks. The commander at NORAD fretted for eight minutes, as conflicting reports came in, over whether this could possibly be a real attack, and since the regulations call for him to decide within three minutes, he was subsequently fired. As recently as 1995, a U.S. research rocket sent up off the coast of Norway to study the aurora borealis so alarmed the Russians that for the first time in their history, they activated the nuclear suitcase that accompanies the president. They were a couple of minutes from a launch decision when they determined that the rocket was not headed toward Russia. (By the way, the Russian's early warning system and radar continue to erode, and one of our sea-launched ballistic missiles could reach Moscow in fifteen minutes.)

In contrast to the public boasting about "detargetting," a Presidential Decision Directive issued last November that provided new guidance---the first since 1981--for the employment of United States nuclear weapons was supposed to be secret, and was only reluctantly discussed in general terms with the press after word leaked out. Sources in the administration pooh-poohed the attention, saying there was little in the guidance that was actually new. And they are exactly right. The Clinton administration has finally put its official stamp on nuclear doctrine, and all it does is reaffirm President Ronald Reagan's guidance from the apogee of the Cold War, except in two particulars. One, although we can still plan for fighting a protracted nuclear war, we no longer have to be sure of winning it. In other words, we have moved from a delusional Cold War guidance to a merely impractical Cold War guidance. And two, for the first time presidential guidance has codified the utility of nuclear weapons as deterrents against other weapons of mass destruction, such as chemical and biological weapons. In an era in which the United States, with its overwhelming conventional capabilities, has no conceivable strategic vulnerability other than the possession of nuclear weapons by other countries, it is a foolish capitulation to the nuclear weapons establishment--the weapons labs, the manufacturers, the nuclear arm of the military-- to be thinking up new ways to legitimize the possession and use of their product.


History suggests that increased public criticism--something so far missing from the chorus of voices mentioned above--will be necessary to spur change. But to press for change, it helps to understand the terms of the debate, and have an inkling of how we got into our present predicament. That the public is confused is no surprise, since every administration since 1945 has worked to keep it that way.

If we assume that ten percent of our ICBMs are undergoing maintenance on any given day, and credit a report from inside Strategic Command that four ballistic missile submarines--out of the nine or ten at sea--are on continual "hard alert," we could fire as many as 2600 warheads at Russia any time we felt like it. These warheads have a combined explosive power of a little over 600 megatons (600,000,000 tons of TNT) or, to use a popular unit of measure, 46,154 Hiroshimas. This is the "overkill" that we think we know about, and even the estwhile nuclear weapons officer Jimmy Carter, in his first meeting with the Joint Chiefs of Staff as president-elect, asked why the United States needed more than two hundred missiles to deter the Soviet Union. As Janne E. Nolan describes the encounter in her excellent history of operational planning, Guardians of the Arsenal, "His advisers, especially Brzezinski, were profoundly embarrassed at this wanton display of naïveté." [nolan, p. 129]

To measure "overkill" by Hiroshimas, or for that matter, Moscows, is not to understand what our military perceives our weapons to be for (and it is the military's opinion that counts). By the same token, that other constant of the nuclear age that we all think we know, Mutual Assured Destruction, or MAD--which connotes a bilateral defensive posture with forces that can survive attack and inflict annihilating retaliation--is a myth: it has never been the United States' operational doctrine, nor was it the Soviet Union's, nor is it Russia's today.

Far from being "unthinkable," the problem with nuclear war is that it is nothing but thinkable. What it is not, is doable, nor is it--on a deep level that even many warplanners admit to--quite believable. So hyperrationality reigns, unconstrained by experience or cautionary common sense. And like most exercises in weightless sophistry, you can reason your way with equal ease to opposite sides of almost every nuclear proposition. That is why all nuclear think-tankers, gadflies, and targeteers talk about "the nuclear priesthood" and the "theology of deterrence." In the last analysis, reason isn't enough. You've got to have faith.

A core conundrum is the notion of credible deterrence. The principal justification for nuclear weapons is that they deter aggression. Therefore, if your deterrence isn't "credible"--that is, if the actual use of your weapons would be a disaster for you, as it is in MAD--then your deterrence is undermined. But to make your deterrence credible, you have to enhance your practical ability to act if deterrence fails, and yet any success at enhancing the practical utility of nuclear weapons . . . undermines deterrence!

This is not so much of a problem if all you want from nuclear weapons is "minimum" deterrence, which is the idea that you will only use your weapons when national survival is at stake. Under those circumstances, your deterrence is sufficiently credible, because you reserve it until you have nothing, or at least little, left to lose. But from early on, the United States committed itself to "extended deterrence," when it settled on nuclear weapons as a cheap alternative to the conventional arms build-up in Europe that otherwise would have been necessary to counter the presumed threat of the Red Army. Even when the United States enjoyed a vast nuclear superiority in the early 60s, the Berlin crisis immediately made obvious the weak credibility of extended deterrence. The Pentagon assured President Kennedy that the United States could mount an all-out first strike against Soviet nuclear forces that would miss only a few bombers, thus ensuring that American casualties would be kept to less than ten million. [nolan, pp 66-67] In other words, for the modest price of ten times the American casualties from all of World War II, we could save the western half of a city that we spent World War II trying to bomb into oblivion.

Trying to grapple with the incredibility of extended detterence, then-Secretary of Defense Robert McNamara came up with the idea of "flexible response," which meant the ability to engage in limited nuclear operations that would not (one hoped) escalate into a general exchange. This required developing small sets of targets and the ability to hit them precisely so that somehow, through the panic and fog of a nuclear attack, our limited intentions would be clear, and the Soviets wouldn't let go with everything they had.

As Nolan points out, for the military men at Strategic Air Command, or SAC, which since 1952 has been the real operational planner of nuclear war (until it was reorganized to include the Navy nuclear force and was renamed Strategic Command, or STRATCOM, in 1992), the idea of withholding most of your force, thereby making it vulnerable to destruction by the enemy before it could be used, was just the kind of armchair idiocy you could expect from civilians.

But if the civilians' faith was that you could contain a limited nuclear war, the military's faith has been that nuclear war is like conventional war, only a hell of a lot bigger. This is no surprise, since the military has no experience with nuclear war. (In the famous words of civilian strategist Herman Kahn, retorting to then-head of SAC General Thomas Power's complaints about his meddling, "I've fought just as many nuclear wars as you have, General.")

The military has always devoted itself, to the extent the accuracy of its weapons allowed, to "counterforce" nuclear targetting--aiming for the other side's weapons and military apparatus, rather than its cities--not for moral reasons but because in conventional wars it is wasteful to expend your munitions on things that aren't firing back at you. The popular idea, nurtured by presidential bromides through the years, that nuclear weapons exist only to insure that they never be used, is not a military idea. The point is not that the military wants to fight a nuclear war--of course it doesn't--but if deterrence ever failed, the military's job is to fight wars, and that means figuring out how to use the weapons at its disposal. If you are trying to limit damage to the United States--a normal goal in any conventional war--you try to destroy the other side's weapons and war-supporting industries. ("Damage limitation" is enshrined in presidential guidance and is, of course, directly contradictory to Mutual Assured Destruction.)

But two fundamental gaps in the logic are never addressed. One, the more capable your counterforce ability, the more it looks to the other side like a first-strike force--the more it is a first-strike force--which naturally undermines deterrence. And second, even if you do strike first, since the other side is ready to launch its weapons on warning--the obvious answer to your first strike capability--you're not going to destroy the weapons that threaten you, anyway. The result, ironically, is the Mutual Assured Destruction--or at least Mutual Assumed Destruction--that our doctrine has never endorsed, and that our war-fighting stance has dedicated itself to avoiding. But if we have MAD, we have it in an unstable form, with temptations on both sides, in a deepening political crisis, to launch their weapons first, "to use 'em or lose 'em," as the parlance goes.

The one constant in this welter of contradictions has been an upward pressure on the number, power, and accuracy of nuclear weapons. Two hundred warheads might be plenty for annihilating Russian society, but they're hopelessly inadequate if the other side has, say, 500 missiles in hardened silos and you need to aim two warheads at each silo to ensure its destruction. Meanwhile, every administration's attempt to make the unworkable idea of "flexible response" work, has not touched this core counterforce "requirement," while it has added layers of other needs--multiple target sets, faster retargeting, redundant command and control, reserve forces, and so on. The military may doubt that these frills make much sense, but bureaucracies and the industrial complexes that serve them do not miss opportunities to expand their resources.


The real issue is not whether the military or civilians in the Pentagon or the National Security Council plan our nuclear war, but the lack of accountability. The Single Integrated Operational Plan, or SIOP, which is our blueprint for nuclear operations against Russia, is so top-secret that it has its own classification, SIOP-ESI, which stands for "extremely sensitive information." Very few people, perhaps a few dozen at a time in the military, the Pentagon, and the White House know what's in the SIOP beyond the level of broad categories of targets. This inner circle hasn't even included most presidents, who have more pressing claims on their attention than some eventuality that they, like everybody else, have trouble believing will ever happen.

The secrecy, coupled with the extreme complexity of choreographing the destruction of hardened, buried, hidden, or mobile forces by missiles and bombers that cannot get in each other's way, has resulted in virtually no one having a workable grasp of the war plans except the targetters themselves, who are thus effectively immune from criticism. As the current head of Strategic Command, General Eugene Habiger, said to me on the subject of the open letter published in December 1996 by 72 generals and admirals from around the world calling for deeper cuts in strategic forces and a commitment to the eventual abolition of nuclear weapons, "If you look at the list . . . there's only one or two or maybe half a dozen who really understand what's going on." In other words, they--and I, and needless to say, you--don't belong in this debate. In other words, the planners are saying, "Trust us."

But one of the signers of the open letter was General George Lee Butler, who as a former head of Strategic Command understands very well what's going on, and his revelations about some of the details of nuclear planning confirm that our war planners deserve no more blind trust than any other part of the government, which is to say, none.

An articulate and impassioned man, Butler already had his doubts about the operational feasibility of the U.S. nuclear war plan when he took over what was then Strategic Air Command in early 1991. In his first months at SAC, he personally undertook a painstaking review of the million lines of computer code that constitute the SIOP. For the first time, he saw in detail what happens when broad presidential guidance is translated into actual weapons aimed at actual targets, what he calls "climbing down the ladder of abstraction." He was appalled at what he found at the bottom rung.

For example, of the 12,500 targets in the SIOP at that time, one of them was slated to be hit by 69 consecutive nuclear weapons. It seems superfluous to say that this is crazy, but it is important to understand how the planning process could result in such a figure. At the level of a presidential directive, a document of a thousand words or so, you will have the reasonable-sounding requirement--if you're thinking about war-fighting at all--to, say, target the political and military leadership. That guidance goes to the Office of the Secretary of Defense, which in a 15- or 20-page document called a NUWEP (for "nuclear weapons employment policy") adds some detail: for example, what sorts of leadership facilities should be targeted. The NUWEP then goes to the Joint Strategic Target Planning Staff of the Joint Chiefs of Staff, which in hundreds of pages of a document called Annex C to the Joint Strategic Capabilities Plan lists specific facilities to be struck and damage requirements to be met. Annex C then goes to STRATCOM, where the targetting staff figures out which weapons, and how many, to apply to each target to meet the required level of damage.

Butler won't say which target those 69 weapons were aimed at, but a reasonable guess would be one of the deeply buried complexes for Russia's military and political leadership during wartime. Let's say Chekhov, which is located about 60 miles southeast of Moscow. Let's assume that the damage requirement in Annex C called for a 90 to 95% confidence in the destruction of this command post, which is a typical figure for that class of target. When I mentioned Butler's 69 weapons to Dr. Bruce Blair, a former Minuteman missileer and acknowledged expert on the operational aspects of nuclear warfighting now at the Brookings Institution, he found in his notes a statement by a high official at SAC in the late 1980s that the highest kill probability for the United States' best weapon against deeply buried, sprawling, hardened command posts was less than 5% (how they calculate this is a whole other matter, but the short answer is, they guess). Blair got out a calculator, assumed a kill probability of 4% for one weapon, and started multiplying. To attain a 50% confidence in destroying the target required 17 weapons. When Blair got up to 69 weapons, the "kill probability" had reached 94%.

As you can see, then, it is all perfectly rational. And perfectly plannable, too, since what is a mere 69 nuclear weapons out of 12,000?

To take another example: I asked General Butler how low in importance the targets went in the SIOP that he had inherited in 1991. His reply is worth quoting at length:

"Let's take communication sites, for example. The most likely way to cripple a communication site is to strike its antenna. If the antenna is in the Soviet Union, and this is a communications yard that serves a key command post, then you must take out the antennas. Now what is an antenna? It's an ungainly spire of structural steel. How would you do that if this was not a nuclear war? You'd send in an airplane with a couple of 500 lb bombs, you might even send in a special forces team to just put a dynamite charge on one leg and topple the thing. What do you when you're a nuclear war planner? You target a nuclear weapon against it. A nuclear weapon measured in--the halfmegaton range? Whatever it takes! And so part of trying to understand what I call the grotesque excesses of the nuclear age is to appreciate that what might have been a seemingly reasonable war plan in the Second World War, where on any given day a thousand or more bombers might be flying across the English channel from England to Germany to rain down tens of thousands of--what?--500 pound conventional bombs, in the nuclear era you had very much that same thing, a wave of thousands of airplanes, missiles, tens of thousands of weapons, except they were nuclear. And they were going to targets, which, taken out of this apocalyptic context, you would say, 'You're going to strike a radio antenna with a--a half megaton bomb? What are you thinking about? '"


But that was the old SIOP. During Butler's tenure as Commander in Chief of SAC, and later of Strategic Command, he, Secretary of Defense Richard Cheney and Chairman of the Joint Chiefs of Staff General Colin Powell worked together to reduce the target set from 12,500 to a little over 3000. And five years have passed since that effort, during which, presumably, targetting changes have continued. So perhaps, one likes to think, today's SIOP manages Word War III a bit more efficiently.

The most important thing to say on this score is, very few people know, and you have probably never heard of any of them. It is highly doubtful that Secretary of Defense William Cohen, for example, knows today's SIOP down to a level of detail adequate to judge its merits. Butler tells an anecdote about Secretary of Defense Richard Cheney, who in a briefing by the Joint Chiefs on the war plan surprised everyone by asking what the planners' criteria were for choosing bridges as targets. This was much farther down "the ladder of abstraction" than anyone had expected a Secretary of Defense to go, and in fact none of the military briefers in the room knew the answer to his question. This is the sort of thing an obscure lieutenant colonel on the Joint Strategic Target Planning Staff has to decide, when interpreting the guidance to "destroy 80% of the Soviet transport system." The answer Cheney eventually got was that there was a single criterion: the length of the bridge. Not its function as a node, not its proximity to a military base, nor to a command post, nor to a weapons storage area.

Any discussion of the current SIOP is necessarily speculative. It must draw on the work of outside experts who have devoted years of their lives to ferreting out facts from confidential sources and leaked documents, and some of what I have written here, probably, is wrong. But one should keep in mind the often overlooked, extraordinary reality that this, the most consequential set of plans in our democracy, is utterly beyond the reach of public scrutiny.

First, little comfort can be derived from the reduction of the target set to some number in the 3000-range. (The exact number is classified.) Many of the 9000-plus targets that were cut were in eastern Europe, or were units of the Red Army, which has largely dissolved, or were industrial plants that have stopped producing anything of value, even without the excuse of a nuclear weapon having fallen on them. To a great extent, Russia has reduced our target set for us. Other reductions have been possible because we've been improving the accuracy of our missiles. The closer you can get to the target the higher the damage expectency provided by each warhead. STRATCOM is also engaged in something called "nodal analysis," in which the targetters try to figure out which spokes of a system don't have to be hit as long as they hit the hub. (In the bridge anecdote above, they didn't have to bother with this, because they had enough weapons for as many bridges as they wanted.) Some damage requirements have probably been lowered, from 90% to 70%, or perhaps for some classes of targets such as war-supporting industries, to 50%.

What has almost certainly been retained is as confident and complete a strike as we can marshal against the single aspect of Russia's military power that we still fear, its nuclear forces. As the independent nuclear expert and author William Arkin puts it, "In essence, STRATCOM has been able to manipulate--or they would say 'guide'--reductions to retain the core, which is the ability to take out all strategic Russian forces and command and control, i.e., a first strike."

This counterforce strike is called Major Attack Option 1, or MAO-1. (The four Major Attack Options in the SIOP are nested. MAO-2 is counterforce plus "other military targets," such as warhead storage sites and secondary airfields; MAO-3 adds leadership; and MAO-4 adds economic targets, which through "nodal" analysis have been reduced from hundreds of factories to--probably--an emphasis on energy production. It is interesting to note that even at the level of MAO-4, cities are not explicitly targeted.) MAO-1 is intended to spare the political and at least some of the military leadership--to enable intra-war negotiating--and to avoid, as much as is possible, the cities. Under the current plan, the number of targets in MAO-1 is in the 800 to 1200 range, or about a third of the total: hardened missile silos, the Russian submarine bases, primary airfields, critical command and control facilities.

In the Hall of Mirrors that is deterrence theory, one can always argue that rationalizing the SIOP is a bad idea, because making it more reasonable only encourages the notion that it is actually executable. Defenders of the status quo at STRATCOM and the Pentagon try to have it both ways. They argue that the deep cuts we and the Russians have already made in our strategic arsenals, plus "nodal" analaysis, plus the increased ability of our missiles to really hit other missiles, signify real progress away from the bad old SIOPs that called for dropping 40 warheads on Kiev (a figure General Colin Powell cites in his autobiography). But when challenged with the proposition that "surgical nuclear strike" is an oxymoron, they retreat into the formulation that nuclear war, of course, would always be unimaginably horrific, and that is the whole point, because the name of the game is deterrence.

It is the second argument that is surely right. The very notion of a pure counterforce strike seems untenable when you consider its effect on Moscow. This is perhaps the most sensitive element of the SIOP, because it goes to the heart of an old debate about how many weapons we need. If our thousands of very accurate counterforce weapons still kill millions of Russians and make their country a radioactive wasteland, why not just target their cities, which takes far fewer weapons? As seems always to be the case with SIOPs, General Butler can't speak with real authority about the current one, but he says, "If the guidance still requires, as I think it does, a devastating attack on the command and control of forces, Moscow still has to be the target of an astounding number of nuclear weapons. For one thing, there's the ABM ring around Moscow." A targetting expert who worked on formulating a future SIOP at START II levels recalls that the START I-level SIOP--which he assumes we are still using--called for some 60 warheads to strike Moscow and its environs in the "pure counterforce" strike. On the other hand, William Arkin surmises that STRATCOM has moved much farther in the direction of restraint regarding Moscow, and posits a low number of high-altitude bursts to disrupt communications, resulting in immediate deaths in the hundred-thousand range, rather than the millions.

Obviously, these are guesses. But a statement General Butler made to me suggests that the disparity among them is moot: "One of the exercises I asked my staff to go through was to remove all the weapons directly targetted on Moscow . . . and just calculate radiation levels in the city by looking at the strikes upwind of the city in various climatological scenarios. The result was exactly as I would have predicted: the city was rendered uninhabitable for generations." (As recently as 1992, in response to a study proposing a serious look at city-targetting as a way to reduce force levels, Paul Nitze wrote, "I do not believe such a retaliation would be morally acceptable.")

This is why critics of the START process point out that even if START II is ratified, and even if START III is negotiated and ratified, and implementation goes according to schedule (which assumes many things, not least of which is that Boris Yeltsin not only will not die, but will remain president of Russia past his constitutional limit) then in January 2008, after another decade of peaceful and reasonable cooperation with the Russians, we will still have 2000 to 2500 warheads aimed at their country in a "counterforce" posture that represents, at best, slow radiological slaughter of millions rather than fast, blast-pressure slaughter of millions. Our "glide path," as the Clinton administration likes to call it--actually quite a good metaphor, since it implies no impetus from our government--is from over-over-overkill down to overkill. It is a "reduction" that masks stasis.

Arkin goes further: "We actually have a greater capacity to destroy Russian nuclear forces than we did ten years ago." Bruce Blair agrees: "The situation is more unstable than it's ever been. The Russians are very vulnerable." This is because Russian nuclear forces have not been immune to their country's economic plight. The Russians apparently cannot afford to keep more than one or two of their missile submarines at sea at any one time, and all of their rail-mobile ICBMs are in garrison, as are all but one or two of their road-mobile ICBMs. The Russian bomber force appears to have degraded almost to insignificance. Thus, while the United States at all times has about 1600 nuclear warheads at sea, where they are invulnerable (because subs cannot be located) the Russians have only two to three hundred warheads that are similarly protected.

Now it may seem obvious that having 200, or 100, or even 50 invulnerable weapons should keep the Russians from worrying too much. If we nuke them, after all, they can destroy our fifty largest cities--quite a deterrent, you would think. But that's the old Jimmy Carter navet_ again. When the United States was perceived to be in an analogously, although much less critically, vulnerable position in the early 80s, our military found plenty to worry about, and it got Congress and the Reagan administration worried, too, and we embarked on a massive build-up of forces. This was the so-called "window of vulnerability," and the reasoning went like this: The Soviets were thought to have acquired enough accurate, multiple-warhead missiles to be able to destroy our entire land-based ICBM force with only a portion of their strategic forces (killing, say, a few hundred thousand Americans). If that were to occur, we would lack sufficient accuracy and power, with only our submarines and bombers remaining, to effectively target the Soviet military forces, and thus our only recourse would be to retaliate against Soviet cities. But that would spur a retaliation against our own cities, resulting in scores of millions of additional American deaths. Thus, the "credibility" of any such retalitation from the U.S. would be weak, and the Soviet Union could dictate terms to its effectively disarmed foe. (Two things to note: one, this should give some idea of just how many weapons you will think you need once you start constructing these nested boxes of "credibility"; and two, it is worth pondering the irony that it was our own land-based nuclear weapons that made us "vulnerable" in this scenario.)

The point is not what our intentions are, but what some Russians believe our intentions are, or what some can argue our intentions are in order to score political points off Yeltsin for being weak on defense (an aspect of democracy that we know well). And if we were to return to a period of political tension with Russia, the Russians' fears about our abilities should very much be our fears, too, because in this mirror-image game of using 'em or losing 'em, the weaker a partner gets, the more dangerous he is. Furthermore, it's a downward spiral: our realization of Russia's desperation would add to our temptation to launch first. In other words, whatever "intentions" we may have regarding our first-strike force are irrelevant. An operational posture exerts its own logic.


There are currently a number of cogent suggestions for how to move forward more quickly and meaningfully.

Over the past 14 months, de-alerting United States and Russian strategic nuclear forces has been endorsed by the National Academy of Sciences, the Canberra Commission, Sam Nunn, George Schultz, William Perry, General Butler, and others. The attraction of de-alerting is that it could remove within a matter of months--if the US and Russia would only do it--many of the dangers of the strategic nuclear standoff, without having to wait for the deep cuts in the arsenals that at the current rate are going to take more than a decade. De-alerting would accomplish what Clinton's "detargetting" pretended to: it would significantly increase the time necessary to prepare and launch a missile, so that computer chip failures and research rockets wouldn't have the potential for destroying the northern hemisphere. (The bomber leg of the U.S. strategic triad was unilaterally taken off alert in 1991 by George Bush, when the bombs and cruise missiles were removed from the planes and stored separately.)

De-alerting proposals have taken a number of forms. Bruce Blair, who has worked hard in this field, proposed with two colleagues in the November 1997 Scientific American that the U.S. unilaterally remove the warheads from its MX missiles (which are due to be retired under START II anyway) and pin open the safety switches of the Minuteman III force, a procedure that would take a couple of hours to reverse for each missile. If the Russians reciprocated, Blair envisions an ideal eventual arrangement, under which warheads would be removed from all land-based missiles and perhaps stored in the partially-filled silos of missiles that had been cut from the force. In this posture, both the warheads and the missiles would be secure in hardened silos, so it would take two attacking warheads to "kill" one missile-warhead pair (this is stabilizing, since the attacker is disarming faster than the victim), and it would be easy to verify compliance since re-mating warheads and missiles would be highly visible.

This is only one idea among several--some addressing the more complex problems of the subs--that are broadly agreed to be technically feasible, and sufficiently verifiable by remote cameras, electronic seals, or on-site inspections that are no more intrusive than we already have under START I. A task force in the Pentagon is looking at various de-alerting options, but there are indications both from the Office of the Secretary of Defense and from the National Security Council that no overtures to the Russians are contemplated until after the Duma ratifies START II. The lack of urgency can be inferred from the statement to me by one of the officials looking at de-alerting: "I don't buy, if you will, the picture that we are on a 'dangerous hair-trigger' today." The Commander in Chief of Strategic Command General Habiger also spoke reassuringly about the state of the Russian nuclear forces, when he returned from a trip to Russia in November of last year.

Consider, however, the incongruity: General Habiger can visit Russia and be shown, as he said in a press conference at the Pentagon, "a great deal about specifically their strategic rocket forces from their command and control to allowing me the first, as I understand it, non-Russian to ever go into a nuclear weapons storage area and to see how they keep their nuclear weapons secure and safe"--as Russia's head of Strategic Rocket Forces has previous visited STRATCOM. General Habiger can also announce that he and his counterpart General Yakovlev have agreed to "establish a shadow program where we would take the equivalent of a wing commander, and squadron commander, a flight commander, and a missile crew member from one of his missile bases to come to the United States and shadow their respective counterparts for a one-week period--meetings, fitness center, dining facilities, everything--and then he would reciprocate with a team from my command." And yet not only is Habiger returning to the same apparatus of mutual annihilation that everyone on both sides wishes would disappear, but while the missileers are shadowing each other and sharing meals, we don't have a direct link between NORAD and its counterpart outside Moscow for sharing global missile launch date, so that when the research rockets take off and the fifteen-minute countdowns begin there's less of a chance for confusion. Senator Sam Nunn suggested such a link back in the 1980s, and Bruce Blair, who has talked it up for years, finds no particular objection to it in the administration. It has just been, he says, "crowded off the plate by other issues."

Meanwhile, the CIA is disagreeing with Habiger's assessment that the Russian Strategic Rocket force is safe. A recurrent computer malfunction in the Russian system has lately been switching part of the force to combat mode automatically. Many of their early-warning radar installations no longer work, and they are missing two slots in their satellite warning system. Senator Nunn, widely respected by the military and no alarmist, tells me, "I worry a great deal about the erosion of Russia's defense satellites. This is a serious national security problem. I got General Ellis [head of SAC from 198- to 198-] to do a whole assessment of our ability and the Soviets' ability to detect the origin of a missile attack. That report is classified, but I can say this, the US ability wasn't at a desirable level, and Russia's was a lot worse. And since then we've gotten better and the Russian's have gotten still worse."


There are also various proposals for the United States to jump-start START, either by unilateral action or bilateral proposal. The Clinton administration is resisting all of them, insisting that it will and should do nothing until the Russian Duma ratifies START II (the US Senate ratified it in January 1996).

It is relevant to consider why START II is bogged down in the Duma in the first place. During the START II negotiations in 1992, Boris Yeltsin proposed a ceiling on deployed strategic weapons of 2000 on each side. This was during General Lee Butler's tenure at Strategic Command, and he vividly remembers a telephone call he received from a top official at the Pentagon: "He said, 'Lee, we're coming down to the final cuts here on this. Yeltsin has put the number 2000 on the table.' I said, 'Good, take it.' 'Well, a lot of people seem to think that's too low.' I said, 'What do you mean, too low? Have we forgotten what the threat here is?' He said, 'Well, our number is 3500.' I said, 'Excuse me? Why would we want to keep 1500 more warheads than they themselves are willing to voluntarily give up?' He said, 'Well, we're not familiar with numbers that low.' I said, 'I've run war plans down to numbers as low as 500, I've done it from 6000 down to 500, in 500-increments . . . You want me to devastate Russia with 2000 weapons? I'll be back this afternoon, show you how to do it.' I didn't prevail."

One reason for U.S. reluctance to agree in 1992 to the number we now say we want for START III (but which we can't reach because we're stuck on START II) was that we were in the middle of an extremely expensive program for the Trident submarines that involved refitting them with the more accurate D-5 missiles and higher-yield warheads, to increase our ability to "kill" Russian missile silos. Butler says, "When you go to those levels [2000] you start trading off systems. That always triggers interservice rivalry. There is an enormous amount of money at stake . . . If you stay at 3500, you have to go ahead and backfit all of the D-5s on the submarines. We're talking tens of billions of dollars You're talking about the future, the pride, the standing of the SSBN [ballistic missile submarine] force."

Colin Powell, who was Chairman of the Joint Chiefs of Staff in 1992, has a somewhat different view: "We wanted 3500 because, targetting-wise, it was what we felt we needed." (One would think the opinion of the head of STRATCOM on this issue would carry the day.) But Powell agrees with Butler on the role that money played: "I don't think it [staying at 3500] did any damage to our modernizing programs, either. That was not the major factor, but still, you spend a lot of money on a new program . . ." He left the sentence unfinished.

The U.S. insistence on a higher ceiling turned out to be a time bomb. The START II agreement centers on the banning of land-based missiles with more than one warhead, or MIRV (Multiple Independently-Targetable Re-entry Vehicles). MIRVed missiles are a terrible idea, because they make silo-based missile parity impossible. If the two sides each have, say, 100 silo-based missiles with three warheads apiece, neither side is safe from a first strike because an attacker can launch 80 missiles with 240 warheads, lose 40 of them in the arctic ocean, miss somewhat with 100 more, wipe out all the other side's missiles with the remaining 100 warheads, and still have 60 warheads left at home for pulverizing a few cities if there's any backtalk. MIRVs were invented by the United States--as nearly all technical innovations in the nuclear arms race have been--but subsequently the Soviet Union relied on them more than we did. Because of the high number of Russian MIRVed missiles, if they want to de-MIRV and still have 3500 deployed warheads, they actually have to build several hundred new single-warhead missiles, and they haven't got the money for it. Meanwhile, their whole force is aging, and officials on both sides agree that they'll come down to a level of 2000-2500 within the next ten years purely through attrition, regardless of what happens with START II or III.

Officials in the Clinton administration and at STRATCOM say that no one could have foreseen in 1992 that the Russian economy would deteriorate so fast, but that leaves unaddressed the question of why Yeltsin offered 2000 in the first place. It also ignores the fact that within a few weeks of the signing of START II by Bush and Yeltsin, a delegation of American officials in Moscow to sell the accord to the Russian military found strong opposition. It is not all that hard to see their point of view: the START II treaty bans the U.S. MX missile, of which we have 50, and the Russians' ss-18s and ss-19s, of which they have 340. It reduces our Minuteman III missiles to a single warhead apiece, but since we have made it clear that we intend to hold on to a few thousand undeployed warheads as a "hedge" against future conflict with Russia, we could "break out" some time in the future by re-MIRVing our Minutemen. The Russians have no comparable opportunity for "breakout." In Helsinki in May 1997, Clinton and Yeltsin agreed to measures that ease the financial problems for the Russians by stretching out the timetable for START II (in other words, having first arguably slowed down the process by insisting on a ceiling of 3500, we have fixed the resultant problems by slowing down the process some more), but the political repercussions of having fashioned a treaty widely perceived in Russia to be unfair linger on, and probably contribute to the Duma's foot-dragging on ratification.

But when it comes to souring the US-Russian strategic relationship, the details of START II are as nothing compared with the Clinton administration's bid to expand NATO. This of course, is the centerpiece of Clinton's foreign policy, and it is an enormous blunder. If NATO has a post-Cold War role, it is as the world's only multinational military structure with some operational experience, not as a containment device for an empire that no longer exists. To invite Poland, the Czech Republic, and Hungary to join undercuts the former justification while seeming to endorse the latter. And since NATO still has a nuclear capability this means moving the nuclear threat, at least theoretically, closer to the Russian borders. As Sam Nunn, an opponent of NATO expansion, said to me, "I know the way our military would react. If you reverse the situation, and had Russian tactical [nuclear] air coming close to our borders, I think we'd change our nuclear planning." And for what? It's hard to think of a reason, except to give a president who's just not much interested in foreign policy a "centerpiece" that sounds better in stump speeches than keeping US troops in Bosnia.

The administration argues that we've defused this problem through the Russia-NATO Founding Act, which establishes a mechanism for Russian consultation (although, we hasten to add, not a veto) on NATO decisions. Furthermore, Assistant Secretary of Defense Frank Miller reasons that for the Duma to stall on START II because of NATO expansion is "counterintuitive." He explains, "If the U.S. could stay at START I levels indefinitely, and Russians can't afford to stay even at START II levels, the last thing you want to do, if you believe NATO expansion is a threat, is allow an imbalance to grow in the strategic arsenals of the two sides. You want to ratify START II and lock the [US] in." But then he goes on to concede, "OK, politics is politics."

But exactly. We have to get used to the fact that we are dealing with a democracy now in Russia, and the same sort of domestic power struggles occur there as here, in which logic matters much less than perceptions that are easily manipulated by legislators appealing to the emotions of particular constituencies. How could that not be a familiar process to us? Imagine, again, the reverse: Russia expanding a military alliance toward our borders while the Senate debates an arms treaty that can easily be made to look unfair to the United States, its imbalances to be redressed in a future treaty that Russia promises but refuses to formally negotiate. Ratification would be unthinkable without the president expending an enormous amount of political capital. Officials of the Clinton administration are perhaps right when they say Yeltsin could probably win Duma ratification if he fought hard enough for it, but like any president he has to choose his battles. Why are we making it difficult for him?

Former CIA chief Stansfield Turner, in a recent book, proposes a unilateral "cut"--in effect, a de-alerting arrangement--in U.S. strategic forces of 1000 missiles, to send a signal of a desire to break us and the Russians out of this box. General Butler opines, "Why don't we just pull back START II, change the number to 2000 from 3500 and get on with it?" The de-alerting enthusiasts argue that a serious proposal from the U.S. along any of their lines would help. In a recent statement, 117 world leaders, including Mikhail Gorbachev and Jimmy Carter--and even Paul Nitze--call for "immediate U.S./Russian negotiations toward further, deep reductions of their nuclear arsenals, irrespective of START II ratification."

The Clinton administration, however, continues to insist that it will do nothing further until the Duma ratifies START II. Officials call this "keeping pressure" on the Russians, or "holding their feet to the fire." But its effect is to hand the Communists and nationalists of the Duma a veto over United States strategic nuclear policy. Whenever you happen to ask, officials are expressing optimism about action "soon." The START II treaty was sent to the Duma on June 20, 1995.


A third issue is the call for the United States to declare that it will not use nuclear weapons first in a conflict, and for it to throw its weight behind an effort to negotiate a global treaty of no first use. The support for this initiative is broad: the generals' and admirals' letter of December 1996, the National Academy of Sciences, the Canberra Commission, the world leaders statement of February 1998, and a wide array of experienced voices on nuclear issues.

The idea of no first use goes to the heart of our vision for the future of nuclear weapons. It is important to remember that although the eventual abolition of nuclear weapons is not at the moment a foreseeable goal, it is a treaty obligation on the United States. The pledge in Article VI of the Nonproliferation Treaty by the nuclear powers to negotiate "in good faith" toward abolition is a linchpin in the logic of the treaty, which would hardly be viable as an instrument permanently dividing the world into nuclear haves and have-nots. And as the nuclear scientist and co-founder of the Pugwash Conference Joseph Rotblatt recently argued in the Nation, a no-first-use treaty "prepares the way to go to zero, for if each nuclear power possesses nuclear weapons only to deter other powers' nuclear weapons, and all were to agree to eliminate them, then no one would have any reason any longer to retain them." A declaration of no first use is something the Clinton administration could announce tomorrow, if if wanted to. Needless to say, it doesn't want to.

Why not? Through most of the Cold War, we refused to pledge against first use because the west's conventional forces in Europe were no match for the Red Army. That, of course, is no longer the case, but we have shifted our rationale toward the deterrence of the use of chemical and biological weapons, mostly by "rogue leaders" such as Saddam Hussein. The phrase now commonly in use for these types of weapons--"weapons of mass destruction"--strikes some as a deliberate effort to blur the distinction between nuclear, biological, and chemical weapons, so that the threat to go nuclear doesn't look so much like a threat of "first use"--it would just be one weapon of mass destruction deterring the use of another. In a way, this is sophistry, for there is no evidence that either biological or chemical weapons can compare with nuclear weapons for the breadth and devastation of their effects. But in another way, the conflation of these three types of weapons serves to underscore the basic illegitimacy of nuclear weapons. Most people would probably agree that it would be politically (not to mention morally) untenable for the President of the United States to hint to Hussein that if he hit our troops with nerve gas, we would lay down anthrax on Baghdad. (The conflation also undermines the old "genie-out-of-the-bottle" argument that, since you can't "disinvent" nuclear weapons, you can't ban them. You could as well argue that the United States should be stockpiling thousands of chemical and biological warheads, since we can't "disinvent" them, either. Technologies can and have been "banned"--by which is meant, effectively marginalized--when a consensus is reached that they are morally repugnant and illegitimate. The United States does not argue that land mines cannot be banned: it just argues that it needs them.)

Even though what-if arguments are inherently shaky, it has apparently become conventional wisdom that President George Bush's 1991 letter to Saddam Hussein hinting at U.S. willingness to retaliate with nuclear weapons is what kept Hussein from using biological or chemical weapons in the Gulf War. (It is rarely recalled that although Bush warned in the same terms against lighting the Kuwaiti oilfields, Hussein did it anyway.) General Butler convincingly argues for the incredibility of Bush's threat. He imagines what the president's response would have been to a general who arrived at the Oval Office with a confirmation of a chemical attack on a U.S. batallion and a request for authorization for nuclear release: : "Let me get this straight--we're the only nation in the world ever to have used an atomic weapon, in 1945, we have led a global movement in every year since to prevent the spread, and to insure non-use, and now you're telling me you want me to employ one? That's point number one. Point number two, are you telling me that with 500,000 troops and the all of the conventional power at our disposal, we can't find an avenue of retaliation that doesn't include nuclear weapons. Number three, what's the target? Is it Baghdad? I mean, you're not just asking me to drop it in the desert somewhere are you? And if it's Baghdad, how many people are going to die as a result of this? Four, where does the cloud of radiation go? Does it stop at the border somehow? Five, what happens to the coalition, what happens to our image in the area? We immediately become a pariah. The coalition immediately shatters, and we become a mass murderer of Arabs. Six, what if the worst thing happens, and it succeeds? Now what message have we sent? Proliferation is dead. There would be a mass scramble."

Defenders of "calculated ambiguity," as these coded threats are often called, shrug off such argumentation, however, logical. They say that in the last analysis, now matter how ridiculous your threat, your opponent cannot be 100% sure that you aren't nuts enough actually to carry it out, and that tiny, unreasonable doubt encourages a salutary caution. This is called "existential deterrence," and of the various deterrences. I personally find it the most plausible. I do think that there is a small, but genuine, utility to nuclear weapons, and that it resides here. But we are in the hall of mirrors, remember: is it in anyone's interest, not least of all ours, to be advertising this utility, which after all is a better argument for a small nation than a large one? Why shouldn't Iran, which has been a victim of Iraqi gas attacks, argue the same?


The Clinton administration has successfully lobbied for the indefinite extension of the Nuclear Nonproliferation Treaty--and over the past few weeks has undermined Article VI again by once again pointedly refused to rule out the first use of a nuclear weapon againt Saddam Hussein.

The administration has supported the completion of a Comprehensive Test Ban Treaty: and plans to spend almost five billion dollars a year--more than the average annual cost in constant dollars for building up the arsenal during the Cold War--on a "Stockpile Stewardship and Maintenance Program" that is a rich man's virtual nuclear testing regime, enabling us, in the words of the Department of Energy's own documents, to pursue "structural enhancements" to nuclear weapons if needed because of "changes in military requirements." (But not, of course, to develop "new weapons.") Also, to maintain expertise. (Nuclear weapons can't be disinvented, but it seems to be hellishly expensive to remember how to make them.)

Since 1995, we have developed a new earth penetrating nuclear bomb, the B61-11, better than our old 8-megaton B53 (fondly called "Crowdpleaser") for destroying buried Russian command posts or chemical facilities in Libya.

We have reiterated that Libya would not be subject to nuclear attack since it is a member of the African Nuclear Weapons Free Zone--a treaty that we have signed--then mumbled something about a "principle of international law called 'belligerent response,'" that empowers us, if some country bothers us enough, to ignore the treaties we have signed.

We are moving ahead with research on theater missile defenses that technical experts agree will be indistinguishable in capabilities from strategic missile defenses, in violation of the ABM treaty. But actually, we are not violating the ABM treaty, because we got the Russians to agree in Helsinki [New York?] that each side could decide for itself whether what it was doing violated the treaty.

We are upgrading our computers at STRATCOM to allow faster retargetting, so that with our improved satellites we can locate Russian mobile ICBMs better and strike them faster.

We are still manufacturing the D-5 missile, and are deploying it on our subs in place of the less-accurate C-4, to give us better "hard-target-kill" capability against Russian silos.

When we retire the MX, we will transfer the MX warheads to our Minutemen, upping the yield from 170 kilotons to 300 kilotons.

The RC-35 reconnaisance plane pilots tell Bruce Blair that their overflights of Russia--which they still do regularly, plotting out the best invasion routes for our bombers--have become boring, because the Russian air defenses are in such a degraded state that the whole thing is a cakewalk.

Boris Yeltsin, speaking to the Duma, talks about wanting to cut Russia's strategic force to 1000 warheads. Speaking in Stockholm, he announces a unilateral cut of the Russian strategic force by one third. A US diplomat is quoted in the New York Times as saying, "We don't take what Yeltsin says all that seriously."

The United States is pulling ever farther away from the rest of the world. We are in a triumphalist mood. We are "winning" the see-saw deterrent balance, and when the other side rises high enough, something will come sliding down on us, and it only needs to be a pebble. A single crude nuclear bomb--even a mere radiation dispersal device--set off in New York or Washington DC, would be the worst terrorist attack in US history, and have grave consequences for the openness of our society. We of all the nations have the best reason to fear that nuclear pebble, because nothing else can touch us. As the Indian General [name?] observed, "The conclusion to be drawn from the Gulf War is that if you are going to go up against the United States, you had better have a nuclear weapon." Why we are leading this dance instead of calling it a night, I can't imagine.

Presidential power in the area of national security is huge. Presidential directives on nuclear operations don't have to be what they often have been: post facto stamps of approval for changes already underway operationally--a form of camp-following, instead of guidance. Clinton could change the terms of the debate tomorrow. But he and his administration project the image of elephantine helplessness. Not a giant, tied down by little nations, but a knight fallen from his horse, pinned to the ground by his own oversize armor.

General Butler: "I have come to the conclusion that very early on the dynamics of the Cold War developed a momentum and a direction and a force of their own. That generated a set of attitudes and prompted the building of institutions and belief systems that were simply beyond the capability of any single person or agency to manage. That should be a great cautionary tale for any power today that imagines it wants to be nuclear weapons capable. Because you are putting in motion a set of circumstances that, you will wake up some day and find that you are not in charge of."

Return to Chip's page

Return to Introduction to IR syllabus

last revised 4/29/98