1998 Congressional Hearings
Special Weapons
Nuclear, Chemical, Biological and Missile


Senior Fellow, Foreign Policy Studies

The Brookings Institution

Testimony Before the Subcommittee on Strategic Forces

Senate Armed Services Committee

March 31, 1998

Mr. Chairman, I am honored to give testimony on strategic nuclear policy, and I congratulate this committee for its wisdom and foresight in holding this hearing. Few people outside this room realize that the strategic landscape is changing fast and that we need to set a new course now.

A fundamental change of direction is necessary for one simple reason: Russia is weak. The premise of our current policy, carried over from the Cold War, is that we must contain and deter a strong and aggressive adversary. But Russia possesses neither the will nor the ability to pursue imperial designs. We need to recognize that the nuclear threat to American security today stems from Russian weakness, not strength, and that deterring a cold-blooded deliberate attack is not our biggest challenge. Reasonable requirements of deterrence are easily met at this stage in U.S.-Russian relations. Far fewer U.S. strategic weapons than currently deployed would provide ample deterrence under any plausible conditions of U.S.-Russian tension.

The nuclear risks we face instead stem from the political, economic, and military weakness of Russia, a decline highlighted in the military arena by the following:

* First, Russian strategic forces and command centers have become very vulnerable, posing a potential risk of unnerving Russia in a crisis. Russia's weak conventional forces can no longer perform the traditional defense mission of protecting Russian territory. Into this vacuum has rushed a growing Russian reliance on nuclear weapons, and on their first use early into a conventional conflict. To make matters worse, the nuclear forces themselves are less secure. Budget shortages among other problems prevent Russia from dispersing its weapons into the sanctuaries of the oceans and forests. In their present configuration, Russian forces could not ride out an attack. Russia today in fact faces far stronger pressures and incentives to "use or lose" its strategic nuclear arsenal than at any time since the early 1960s.

* Second, aging Russian nuclear forces are heading into a tailspin that, coupled with an economic depression that severely limits the production of new weapons, could leave Russia with fewer than 1,000 weapons at the end of 2007 and as few as 500 weapons at the end of 2012. START II and III ceilings are thus unrealistic for Russia, and if Russia slides into numerical inferiority, then our relationship will suffer. The missile and bomber "gaps" and "windows of vulnerability" that once afflicted our relations could re-surface with a vengeance, both in bilateral politics and in Russian domestic politics. A growing disparity would likely buoy Russian nationalists who would revive nuclear tensions and reintroduce Cold-war style politics into our relations.

* Third, while Russia relies more on nuclear weapons and on launching them on warning, its nuclear control is steadily deteriorating in physical, organizational, and human terms. All the trends pertinent to the functioning of Russian's nuclear command and early warning system are negative, and I strongly doubt whether it can endure the stress and strain indefinitely. The susceptibility of Russian nuclear forces to accidental, unauthorized or mistaken launch has been growing since the end of the Cold War.

These adverse developments call into serious dispute the wisdom of our current policy in which deterrence is the predominant concern. Deterrence is not the problem. Restructuring our posture and forces to fortify deterrence would misdiagnose the threat; in fact this response would only aggravate the dangers described earlier. The right approach would be to revise our strategic policy so that it strikes a better balance between deterrence, reassurance, and operational safety. Our biggest challenges are to reduce the "use or lose" pressure on the Russian strategic arsenal, and to prevent its accidental, unauthorized or mistaken launch.

To meet these challenges, Russia and the United States should take their nuclear weapons off alert, so that they could not be launched on short notice. By extending the time needed to prepare nuclear weapons for launch, "de-alerting" on a reciprocal basis would:

* Take Russian nuclear weapons off their dangerous hair-trigger posture, thereby reducing their susceptibility to inadvertent or illicit use. If Russian forces required many hours, days, or longer to get ready for launch, then we would enjoy a larger margin of safety against many scenarios, ranging from the temporary loss of legitimate civilian control over Russian weapons, to the generation of false alarms in their early warning system.

* Rectify the current imbalance in strategic capabilities which undermines crisis stability. America's security depends on a calm Russian finger on the nuclear trigger, and this could be ensured if the U.S. threat of sudden attack were removed by lengthening the fuse on our forces from the current minutes to many hours or longer.

* Smooth over the emerging wrinkle of sharp numerical inequality favoring the United States, a development that could de-rail arms control and sour relations. Given Russia's nosedive, if we intend to preserve the principle of numerical equality, then we need to fast forward START, lower the ceiling to 1,000 weapons or below, and immediately bring Britain, France, and China into the negotiations. Alternatively, de-alerting could help compensate for Russia's slide into numerical inferiority by greatly reducing the operational significance of this imbalance.

* Strike a better balance between deterrence and operational safety. With the Cold War over, nuclear safety should be the centerpiece of our nuclear agenda. Keeping thousands of nuclear weapons poised for immediate launch is an inherent danger whose alleviation carries overriding priority regardless of the strength or weakness of Russian nuclear control at this time. De-alerting hedges against normal accidents and human foibles. For instance, might the year 2000 computer defect produce a glitch that affects the operational safety of Russian nuclear missiles? De-alerting would help ensure that it is impossible for electronic or computer malfunctions to trigger a nuclear accident.

* Strike a better balance between deterrence and reassurance. While nuclear deterrence is resilient enough, Russia desperately needs to be reassured that its security is resilient. That is not easily given or received in view of the dramatic decline of Russia's military and its geostrategic vulnerability along its western, southern, and eastern borders. Standing down the nuclear arsenals would scarcely fortify Russian confidence in its ability to handle conventional attacks (or prevent internal disintegration), nor wean Russia of its overreliance on nuclear weapons for security. Nonetheless, it would represent a reassuring gesture on our part that would at least alleviate the stark vulnerability of Russian nuclear forces. And it would lend impetus to the general aim of recognizing that Russian insecurity is a problem for the U.S. and the world as well as Russia.

De-alerting also gives us leverage over nuclear dangers lurking elsewhere in the world. Among the biggest concerns are the following:

* China is developing a modern strategic arsenal and the likelihood is growing that China will greatly increase the launch readiness of its strategic missile force.

* Others countries such as India appear to be considering putting nuclear weapons on higher alert status, a trend that would undermine crisis stability and increase the risk of an inadvertent launch.

* The proliferation of weapons of mass destruction continues to be a source of frustration and threat to the United States and the world.

De-alerting addresses these concerns in the following ways:

* It would preserve, for the foreseeable future, the flexibility of the President in the matter of nuclear options. Nuclear weapons that have been taken off alert and effectively mothballed still exist. If a national emergency requires it, the weapons could be re-alerted in small or large numbers depending on the circumstances. This latitude for nuclear action clearly contrasts sharply with the absence of any scope for nuclear maneuvering if nuclear weapons are abolished. For many people who fear that abolition would deprive U.S. leaders of an option that might be warranted under some hypothetical circumstances -- ranging from a rogue nation brandishing weapons of mass destruction to a Russian nuclear break-out -- but who also believe that nuclear weapons maintained on hair-trigger alert pose unacceptable risks, the de-alerting alternative holds strong appeal. It offers a better alternative to both abolition and the status quo.

* It would project, over the long run, the ultimate elimination of nuclear weapons, which in turn would strengthen American diplomacy in the area of non-proliferation. Standing down the weapons and lengthening the fuse on their possible use takes a long stride not only toward downgrading their importance, but also toward demonstrating our nation's interest in their eventual abolition. De-alerting measures could be devised that would take months to reverse. This timetable is fairly close to the several months needed to build a nuclear bomb from scratch on a crash basis in the absence of an existing nuclear weapons complex. (This estimate of several months assumes that our government had access to commercial nuclear power reactors but nothing else except expertise in nuclear physics.) By gradually adopting more elaborate and extensive measures that make re-alerting more time-consuming, the U.S. government would ascend to higher moral ground in its quest to dissuade other nations from going nuclear.

* It would help create an international norm of operational safety that would apply universally, to all nuclear states, making it taboo for any nation to keep nuclear forces in a launch-ready configuration. If the United States and Russia take all their weapons off alert, then pressure, and possibly sanctions, could be imposed on other countries that move toward adopting hair-trigger alert practices. This taboo could have positive effect on nuclear stability in the Middle East, Asia, and South Asia.

* It would create momentum for establishing an exact accounting and monitoring arrangement for nuclear warheads, which would also enhance their security against theft and diversion to rogue states or terrorists. De-alerting by removing warheads from missiles and aircraft imposes new requirements for verifying the status of the warhead stockpiles. The location and numbers of warheads in the inventories must become more transparent in order to ensure that any re-alerting of forces would be detected in a timely fashion. Although the missiles and bombers themselves can be checked to verify that they remain off alert and devoid of warheads, additional confidence would accrue if the warheads themselves can be monitored. International monitoring of the weapons in storage would serve the additional purposes of strengthening safeguards against theft and creating confidence over time that the world may move toward the elimination of all nuclear weapons with less fear that covert stockpiles exist.

How should we go about de-alerting? Any blueprint for taking weapons off alert should meet three criteria. De-alerting measures should: (1) truly extend the launch preparation time (in contrast to cosmetic and symbolic measures such as those adopted to fulfill the 1994 de-targeting agreement, (2) provide for adequate verification, and (3) preserve the invulnerability of a core deterrent force.

Let me briefly describe a hypothetical U.S. posture that passes these tests:

1. Immediately download to storage the warheads of the MX missiles (which will, in any event, be retired under START II).

2. Disable all 500 Minuteman III missiles by having their safety switches pinned open (as was done for the Minuteman IIs almost overnight in 1991). Then these missiles should be further immobilized as long as Russia reciprocates and immobilizes its silo-based forces. The options for this include removing the warheads, guidance batteries, or mechanisms for opening the silo lids. For a U.S. START III force of 300 Minuteman missiles, we should consider removing the warheads for all 300 missiles and placing them in nearby silos -- one warhead to a silo -- that would become available after retiring 200 Minuteman and 50 Peacekeeper missiles.

3. Remove to storage the warheads on the four Trident submarines that are to be retired under START III and reduce the number of warheads on each remaining submarine missile from eight to four.

4. Take the W88 warheads off the Trident II missiles, place them in storage and replace them with lower-yield W76 warheads.

5. Allow Russia to verify these actions by using some of their annual inspections permitted by the START I treaty, and accept a greater number of inspections if Russia will also do so.

6. Reduce from two-thirds to one-third the fraction of submarines at sea in peacetime. Those at sea would adopt a low level of alert. Their missile guidance sets would be kept off the missiles but onboard the submarines, and seals would be put on the missiles to allow for eventual inspection proving the missiles never went on high alert. (It would take the crews about three days to reinstall all 24 guidance sets on a Trident submarine's missiles.) Lastly, the submarines would patrol many days out of range of Russian targets by, for instance, creating patrol areas in the southern hemisphere. The submarines could disclose their location one at a time on a periodic basis to verify their status.

I consider this de-alerted posture to be a sufficient deterrent under all plausible conditions of U.S.-Russian tension. Five U.S. submarines carrying 480 warheads would remain undetectable at sea, and the immobilized Minuteman III missiles could be destroyed only by a massive Russian attack on many hundreds of silos. This U.S. force would be fully de-alerted and could not be fired quickly, and yet it would provide a secure deterrent force whose survivability would not depend on timely warning of a Russian violation of its de-alerting commitments. Thus Russian re-alerting would offer no significant advantage to Russia nor would it create strong pressures or incentives for the U.S. to bring its forces to a launch-ready configuration. It should be emphasized that the potential for Russian re-alerting in the future will decrease sharply because of the tailspin in delivery systems in store over the next decade. If a break-out problem exists, it will be a much larger headache for Russia than for the United States.

This hypothetical U.S. posture would reassure Russia that its deterrent forces would no longer have to depend on quick launch for their survival. Russia could eliminate the hair trigger on thousands of its weapons, and move to de-alert them systematically in parallel with the United States.

With the chairman's permission, I would submit for the record a list of reciprocal Russian steps for de-alerting its strategic arsenal.

In closing, I wish to add that de-alerting would align our nuclear postures with the end of the Cold War, a long overdue adjustment. By ending the hair-trigger alert practices they carried over from the Cold War, the former adversaries can finally expunge the darkest suspicion from their political relationship and lay the groundwork for a truly productive partnership. Lastly, de-alerting also offers a remedy for a host of other vexing nuclear problems that beset the American government and people. Its virtues should weigh heavily in crafting our nuclear policy toward Russia and the world.