"I actually have a pretty good collection of Arms Control Today, which I have read throughout my career. It's one of the few really serious publications on arms control issues."
– Gary Samore
Former White House Coordinator for Arms Control and WMD Terrorism
Strategic Policy

The Future of U.S. Nuclear Weapons Policy

William F. Burns, former director of the Arms Control and Disarmament Agency (ACDA), spoke at the annual dinner of the Arms Control Association (ACA) October 14 on "The Future of U.S. Nuclear Weapons Policy." A member of the ACA Board of Directors, Burns was director of a similarly titled study by the National Academy of Sciences Committee on International Security and Arms Control (CISAC) released in mid 1997. (See ACT, May 1997.)

Burns, a retired U.S. Army major general, has been involved with U.S. nuclear weapons and U.S. nuclear policy since the mid 1950s, when he was commissioned in the field artillery and assigned to one of the first nuclear armed artillery battalions in Europe. Burns was appointed in 1981 as the Joint Chiefs of Staff representative to the intermediate range nuclear forces negotiations, a position he held until 1986 when he was appointed principal deputy assistant secretary of state for political military affairs. Burns retired from the Army in 1987 when he became ACDA's ninth director.

In 1992, Burns served as the first U.S. special envoy to the denuclearization negotiations with the states of the former Soviet Union, and in 1993 signed the U.S. Russian agreement on the U.S. purchase from Russia of highly enriched uranium derived from dismantled Soviet nuclear weapons. Burns now resides in Carlisle, Pennsylvania, where he is a distinguished fellow at the Army War College. The following is an edited version of his remarks.


Ladies and gentlemen, I am delighted to be here tonight for a number of reasons. First of all, I think the study we did is important, and it's an audience like this, which by better understanding the study and what motivated us, can help us to disseminate the ideas it contained. Secondly, this audience is filled with old friends, and although I would certainly not want to try to identify all of them here, let me just identify two. First, my mentor, Paul Nitze. If it had not been for Paul, I'd be a retired one star right now probably trying to teach someplace. I consider Paul to be the father of arms control in this generation. The other individual, the individual without whom we would not have had a CISAC study is Jo Husbands, the staff director of CISAC.

What I'd like to do this evening is not just talk about the CISAC study, but talk in more general terms about assessing the progress and future of arms control. I became involved in arms control many, many years ago, but I didn't realize it. I didn't realize that the military, whether you're a lieutenant or a captain or a major or a general, are certainly your primary arms controllers. When I first became involved with nuclear weapons in the mid 1950s, the Army didn't know very much about nuclear weapons but we saluted the flag and took them. We were given no particular assets with our nuclear weapons, beyond what an 8 inch howitzer battalion had.

The first nuclear weapon I saw was a training round. It contained only Uranium 238, not Uranium 235 [U 235] and was not capable of nuclear detonation. It was in the back of an M 109 van tied down with a frayed piece of hemp, and a corporal in the van was going to tell us how to put five components together and make an 8 inch projectile. After about three hours and after many false starts on his part, with the manual we were able to more or less put the projectile together. After this we were certified as being able to put together "a nuclear round," and two weeks later we deployed to the United States Army Europe.

In Europe, we suddenly found ourselves proud owners of not only a training round but several of those olive green colored nuclear rounds. For the next three years my battalion was custodian of these rounds. It was an interesting situation because there was very little guidance compared to the guidance in later years. This was before the days of Permissive Action Links or very elaborate release systems. Release was basically tied to the command chain. I had no doubt in my military mind that if my battalion commander said "Shoot," we would shoot, and if my battalion commander said "Don't shoot," we would not shoot. It was as simple as that.

The safest place to secure nuclear rounds was in the basement of our battalion headquarters building. The building had been the former Gestapo headquarters of the town and it had cells behind steel and barred doors, making a very secure arrangement.

What strikes me about this particular weapon—a gun assembly type weapon that is very primitive by today's standards—is that it's the kind of weapon that a terrorist organization or a small, economically deprived nation state could build with a reasonably adequate machine shop; 75 to 80 kilograms of stolen U 235 or maybe plutonium; and a couple of individuals who had degrees, perhaps even a graduate degree, in physics. All you need is a system that allows you to take two sub critical masses of uranium and jam them together long enough so they become super critical. What concerns me today is that the technology of 50 years ago could be used to build a weapon that could end up in the hands of a fairly small terrorist group.

Have we made progress in the last few years? I certainly think we have. Let me give you a report card on where I think we stand.

In terms of bilateral nuclear weapons reductions, that is, between the United States and Russia involving delivery systems with ranges of over 500 kilometers, I'd give us a B+. We have the 1988 Intermediate Range Nuclear Forces [INF] Treaty, which most people forget about today but has been very, very successful. Although the INF Treaty is of indefinite duration, the on site inspections conducted under the treaty will conclude in 2001. With all the talk about START II and START III, it is important to remember that we do have a START I agreement and we are dismantling weapons at a pace somewhere near the capacity of both sides. So, I'd give us a B+ in that area.

In bilateral reductions in delivery systems with ranges below 500 kilometers, I'd give us only a D. When Presidents Bush and Gorbachev announced in 1991 unilateral reductions in non strategic weapons—weapons in Europe—I thought it was a great idea. I've rethought that since. I think it was a terrible idea because in the process we really lost count of non strategic weapons, which are the weapons that can most easily be proliferated.

As for the accountability of nuclear material in Russia, I'd give us a C+. I had great hopes in 1993 that the agreements we had signed under Nunn Lugar would give us a clear handle on these materials, and it has helped. But we've been very slow in the last few years in moving to build a facility there to store the material, to develop a system of accountability to which the Russians agreed, and to encourage greater openness with regard to this nuclear material in the hands the Russians.

In verification procedures I'd give us a B. I think this grade mostly accrues to the On Site Inspection Agency [OSIA], which has done a tremendous job in establishing routine procedures for the verification of all sorts of things pertaining to nuclear weapons. OSIA has been able to respond with great alacrity to the kinds of unknowns that have developed over the last several years.

With regard to the inclusion of third parties in the control of nuclear weapons, I would give only a C. We haven't done very much with the Chinese, although some would argue that that's a very difficult task. Well, it was a very difficult task in 1981 to begin to convince the Soviets that "zero zero" was a reasonable option in the INF talks. We could exert ourselves a little bit more in dealing with the Chinese. We also should be exerting ourselves more with the British and the French to make sure that, if we can make substantial reductions in the U.S. and Russian arsenals, the British and the French are prepared to join that process.

The final area to which I would give a grade is surveillance and defeat of non state actors such as terrorist groups bent on acquiring nuclear weapons. I would say that we've got a solid D or D+ here. One of the problems the United States government has is that it often sloganizes itself into meaninglessness. Remember counter terrorism in the 1970s? We were going to solve the problem of terrorism in the world. We were also going to solve the problem of insurgency in the 1960s. And I just watched on television a travelogue on Vietnam, where we didn't make out so well. I hope that's not the case in our counter proliferation efforts, where we're at the beginning stage, not the end stage.

The greatest threat to our security today is not the threat of nuclear weapons in the hands of other nation states, necessarily, but the threat posed by the proliferation of these weapons into the hands of states that may not show the same restraint that the nuclear weapon states have over the last 50 years.


The CISAC Study

Let me talk a bit about the CISAC study, which is entitled "The Future of U.S. Nuclear Weapons Policy." It was a difficult undertaking because we had a very broad range of views in our committee. I think the strength of our study is that after literally two and a half years of arguments—face to face or via fax and e mail—we were able to reach a fairly strong consensus and there were fairly strong arguments supporting that consensus.

The study recommends significant changes to U.S. nuclear weapons policies to reflect the realities of the post Cold War world. The central recommendation is that the core mission of U.S. nuclear weapons should be to deter the use of nuclear weapons by others. Reflecting this limited role, the study calls for a regime of progressive constraints on U.S. and Russian forces to levels of 2,000 strategic warheads and then to levels of 1,000 total warheads each, with concurrent steps to reduce alert levels and reorient nuclear doctrine away from rapid, massive responses. It further concludes that, given the right security environment, verification procedures and participation of other declared nuclear powers, reductions to a few hundred nuclear warheads would assure the core deterrence mission. Finally, the study examined a number of paths by which the possession of nuclear weapons might be safely prohibited even though their absolute permanent elimination might not be guaranteed.

I assume most of you have read the study, and I'm not going to bore you with a line by line recital of all of the study's many conclusions and recommendations. Let me explain to you what the CISAC study does not say or does not do.

The CISAC study does not minimize the continuing value of nuclear weapons in deterring the escalation of conflicts into unlimited war among major powers. In fact, it suggests, quite the contrary, that nuclear weapons, even after they are reduced to very low levels, or even prohibited, will continue to deter conventional war among major powers simply because these weapons can always be resurrected. The study does not advocate a unilateral commitment to eliminate nuclear weapons now. We are not ban the bomb advocates in the sense that we say that the primary necessity is to eliminate nuclear weapons. We recognize that nuclear weapons have served a necessary and essential role for the last 50 years and that to eliminate nuclear weapons right now is not possible. We do not advocate unilateral reductions by the United States. We do not ignore the threat of other weapons of mass destruction. We do not advocate de alerting because of perceived concerns that U.S. nuclear weapons are unsafe or dangerous in any absolute sense. We do not advocate the prohibition of all ballistic missile defenses. And we do not advocate the abandonment of target planning against potential nuclear adversaries.

What the CISAC study does do is provide tools to continue the process of reducing nuclear weapons, both strategic and non strategic, in the hands of declared nuclear weapon states. It also provides motivation for the non declared weapon states to reassess their security needs, and provides the United States with timely methods of monitoring the stockpiles of others.

The CISAC response to the present nuclear situation suggests the following: There are risks in the world today because of nuclear weapons. They're not the same risks that existed when the United States and the Soviet Union faced off against each other, but they're still risks. The risks now tend to be associated with proliferation and the dangers of erroneous or unauthorized or accidental use of nuclear weapons.

The CISAC study really argues for a two pronged approach to address these dangers. First, the reduction of nuclear weapons to the lowest possible levels, whatever those levels might be. And second, a concomitant regime of progressive constraints which will ensure that nuclear weapons, as they are being reduced, are less likely to be used. I group these constraints into several kinds of initiatives.

First of all, there are the policy initiatives. We argue that it's time for the United States to adopt a no first use policy because the first use policy has more or less outlived its usefulness. But, we argue that if we are to adopt a no first use policy, the United States must maintain strong conventional forces so that it can meet its commitments and, if you will, execute its will without resort to nuclear weapons.

I don't think many of us really realize, unless we've done a lot of traveling outside the United States, what the effects of the Gulf War have been on the rest of the world. They give us much more credit than we deserve for our conventional capability. They believe that the M 1 tank will cut through anything anybody has, and that the F 16 fighter and the various Stealth fighters and bombers can do what they want and there's no stopping them. They believe that the U.S. cruise missiles are absolutely systems against which there is no defense.

And ladies and gentlemen, I'd like to keep the rest of the world thinking that. But, that means that we need to maintain a strong conventional defense.

We believe that we must maintain a strong system of strategic warning. I don't mean warning of a missile attack, but warning through intelligence channels, so we know well in advance, years in advance, perhaps, of any other nation that might be considering a clandestine buildup. We believe strongly that we should maintain the ABM Treaty as it currently exists with regard to a nationwide strategic defense, but we should also ensure that our forces deployed in the field are protected against missile attack.

With regard to operational initiatives, we believe that mutual de alerting should be employed as soon as possible to expand the amount of time—from minutes to hours to days, perhaps to weeks—that decision makers have in the event of a nuclear crisis. We also believe that to maintain the deterrent credibility of our forces, the readiness of individual nuclear elements must remain high.

With regard to technical initiatives, we have a number of challenges ahead of us. If we are to bring down the numbers of nuclear weapons that we have, and we believe that a level below 1,000 is possible, then it's essential that this thousand include all nuclear warheads—deployed, non deployed and so forth. Right now we don't have a system or a theory or an approach that will allow us to count, with great accuracy, nuclear warheads. So we believe that verification of compliance in such a regime will require us to develop a technical system for counting warheads.

We also need to develop systems for securing warheads. If a de alerting measure requires you to remove warheads and place them in a common storage area, obviously under national control, then how do you secure these warheads to the satisfaction of the other side? Sandia National Laboratory has done some very interesting work along these lines in cooperative monitoring, and it seems to me that that kind of work must be extended.

Also important from a technical point of view is how you extend this regime to third parties. In one sense it's fairly simple for the United States and Russia to come up with verification measures. But we found out with the Conventional Armed Forces in Europe [CFE] Treaty how difficult it is to extend these measures to multilateral agreements. I found out in 1988 how difficult it was to take the bilateral INF agreement and extend it to our allies where these systems were based. In 1988, we had not really thought too much about how we were going to deal with the verification provisions of our own agreement. I remember sitting in a meeting in the State Department when it was suggested that we should talk to the basing country allies about inspections. So we called a meeting and within about 10 minutes it was obvious that we were not prepared to deal with allies' concerns.

I remember one representative who said, "Do you realize that you're prepared to sign an agreement with the Soviet Union that authorizes you and the Soviets to do away with our customs regulations and bring anyone into our country without our permission?" And I said, "Well, we haven't really thought about it that quite that way." Nine months later we concluded negotiations with five basing countries that would enable us to execute the verification provisions of the INF Treaty. We cannot afford to make this miscalculation again.

That convinced me that we should not take it as a given that at some time in the future other nuclear powers are suddenly going to say to the United States: "You've done such a great job in this. We're just anxious to sign your treaty as another partner." It's not going to happen. We must take that into consideration right now.

There are other technical problems as well. For instance, let's say that at some point we abandon our nuclear triad. What does that mean? How far do you shift into what we call "adaptive targeting" and away from the present single integrated operational plan [SIOP] targeting? What are the effects as you get down to very low numbers, say, in the hundreds of warheads? One of the issues we raised in our study was the fact that we will probably never be able to eliminate nuclear weapons, at least in a verifiable way. But perhaps nuclear weapons can be prohibited in terms of their use. We haven't really thought enough about that, and that's an area that we suggest needs to be explored.

I think it's important that we learn from our own experience. Let's make sure that we understand the technical problems that proliferators face, and how we dealt with them in a less sophisticated time and how we might deal with them in the future. By profiting from what we learn, perhaps we can cope better with issues of "loose nukes" in the 21st century as the declared nuclear powers continue to reduce their stockpiles. I am optimistic that we can do this and do it well.

The CISAC study, obviously, is not a "be all" and an "end all"; it's a beginning. I encourage and charge others to think more deeply than we were able about some of the problems that we raised. But I see a general sea change beginning. Some of the things that we talk about in the study would have been anathema five years ago, certainly 10 years ago, in government circles. Today, in my private conversations with individuals in government, the anathema is not holding quite as strongly.

So, it seems to me that now is the time and now is the opportunity, for us as individuals and groups we might represent here tonight, to move forward and seize the opportunities as they present themselves, to ensure that the Pandora's box which was opened in 1945 is at least managed carefully, if the lid cannot be shut completely. Thank you very much.

The Next Step in Strategic Arms Control

September 1997

By Spurgeon M. Keeny, Jr.

Following up on the Helsinki summit, a package of agreements was signed in New York September 26 that addresses the Russian Duma's concerns about the fairness of START II and the Senate's concerns about the impact of the ABM Treaty on theater missile defense (TMD) systems. It remains to be seen whether these linked measures will persuade the Duma to ratify START II and amendments to the ABM Treaty and the Senate to ratify amendments to both treaties. Although aspects of the agreements are disappointing from an arms control perspective, the overall package deserves support if it allows START II to enter into force and preserves the ABM Treaty, which are both essential to progress in further strategic arms reductions.

The agreements relating to START II have been crafted to respond to the Duma's complaints that the originally mandated elimination of land based, multiple warhead missiles would have forced Russia to build, at great expense, a large number of new, single warhead missiles in order to maintain parity with the United States at the 3,000 to 3,500 warhead level by 2003. This fundamental objection has been answered by extending the treaty's implementation period five years to the end of 2007, and by the Helsinki commitment to negotiate a START III treaty, after START II enters into force, with a new ceiling of 2,000 to 2,500 warheads at the end of 2007.

The delay in implementing START II, however, postpones the actual destruction of the powerful land based, multiple warhead missiles until the end of 2007. This provides Russia an insurance policy against collapse of the reduction process or U.S. repudiation of the ABM Treaty. To prevent this delay from undercutting the major accomplishment of START II—the elimination of the destabilizing Russian SS 18 and SS 24 and U.S. MX ICBMs—these missiles will have to be "deactivated" by the end of 2003. Together, these new provisions give Russia an alternative to a costly buildup of missiles, and the United States can secure removal of the most dangerous component of Russia's operational forces on roughly the START II time schedule. To underscore the linkage, Russian Foreign Minister Primakov made clear in an exchange of formal letters that the deactivation would not take place unless START III is completed "well in advance" of that deadline.

The agreements sustain the ABM treaty's central objective of severely limiting national missile defense (NMD) systems, but kick down the road final resolution of the demarcation line between future high performance TMD and NMD systems. Except for space based interceptors, anything called a TMD system is permitted—even though it might have significant capabilities against strategic warheads—provided only that it is not tested against a target traveling more than 5 kilometers per second. Each party can also judge whether its future TMD systems are treaty compliant—hardly a desirable precedent for arms control. It remains to be seen whether this flexible interpretation will simultaneously satisfy Senate boosters of ballistic missile defense who oppose any constraints on TMD or NMD systems and Duma critics who see Russian deterrence threatened by highly capable TMD systems. The treaty can accommodate this relaxed approach, provided neither side aggressively exploits its highly permissive provisions—the United States by undertaking large deployments of highly capable TMD systems and Russia by arming its TMD systems with nuclear warheads. At the signing, Primakov underscored that the agreements do not fully resolve the interpretation of the treaty.

In the Duma, the START II ratification debate will occur under the cloud of NATO expansion, which, despite the successful negotiation of the NATO Russian Founding Act, is seen across the Russian political spectrum as a provocative security threat. Unfortunately, Senate hearings on NATO expansion will parallel the Duma's consideration of START II. In this connection, any administration commitments on future expansion of NATO to include states of the former Soviet Union, which President Yeltsin has labeled as unacceptable, will seriously diminish prospects for the Duma's approval of START II.

The Duma now has before it a reasonable response to its concerns that offers a weakened Russia continued nuclear parity with the world's one remaining superpower. If the Duma seizes the opportunity, the Senate can reduce by 75 percent the Russian strategic nuclear arsenal by acting positively on START II amendments and START III. To complete the package, the two legislatures will also have to accept the controversial compromise amendments to the ABM Treaty, which meet the perceived U.S. need for advanced TMD systems while retaining the treaty's central objective which is fundamental to both U.S. and Russian interests. If the Duma or the Senate choose to reject the package, the prospects for START II and further nuclear reductions are indeed bleak.

Following up on the Helsinki summit, a package of agreements was signed in New York September 26 that...

Bulgaria, Slovakia Still Hold SS 23s

The United States continues to press Bulgaria and Slovakia to destroy their small inventories of SS 23 ballistic missiles, but resistance from both governments has stalled U.S. efforts to eliminate one of the Cold War's remaining vestiges. Both countries apparently have rejected a low level July demarche from Washington seeking destruction of the SS 23s, which are banned under the 1987 U.S. Soviet Intermediate Range Nuclear Forces (INF) Treaty. The missiles are the only systems remaining from the 72 SS 23s the former Soviet Union transferred to Bulgaria, Czechoslovakia and East Germany (24 to each country) two years prior to signing the treaty.

The United States has already achieved a primary goal of its ongoing initiative¾the destruction of the connecting sections for the missiles that would enable them to carry nuclear warheads¾and Germany and the Czech Republic have destroyed their SS 23s. But Bulgaria and Slovakia have cited financial, environmental and national security concerns as reasons they cannot eliminate the 500 to 1,000 kilometer range, solid fuel systems believed to be operational and number fewer than 10 in each country. The United States has said it is prepared to assist in the destruction of the aging missiles.

The Soviet Union claimed the transfers, which the United States did not learn of until 1990, were not a violation of the INF Treaty because they occurred before signature and the three former Warsaw Pact countries were not parties to the accord. However, the United States maintained that Moscow had acted in bad faith during the treaty negotiations because all SS 23s were to have been declared and destroyed.

The Clinton administration is seeking the destruction of the remaining missiles to fulfill the objectives of the INF Treaty and the Missile Technology Control Regime (MTCR), which Bulgaria and Slovakia have agreed to adhere to unilaterally. The United States is asking a number of countries to eliminate their so called MTCR "Category I" missiles (which can carry a 500 kilogram payload at least 300 kilometers) because of their inherent capability to deliver weapons of mass destruction.

QDR Supports Nuclear Status Quo, Adds Billions More to NMD Program


Craig Cerniello

ON MAY 19, Secretary of Defense William Cohen submitted to Congress the Quadrennial Defense Review (QDR), a six-month study that examined all aspects of U.S. defense strategy and requirements through 2015. The review, which was conducted in consultation with Congress and approved by President Bill Clinton, will "serve as the overall strategic planning document" of the Defense Department. While the QDR made no substantial changes to U.S. strategic nuclear forces and posture, it added $2.3 billion to the Clinton administration's national missile defense (NMD) efforts in order to preserve the current schedule, and delayed deployment of the Army's Theater High Altitude Area Defense (THAAD) system by two years.

In a reaffirmation of current policy, the QDR noted that "We are committed to reducing our nuclear forces to START II levels once the treaty is ratified by the Russian Duma and then immediately negotiating further reductions consistent with the START III framework." In addition, the QDR concluded that the United States would maintain its strategic nuclear forces at START I levels until Russia has ratified START II, a measure that is consistent with the START II resolution of ratification approved by the Senate in January 1996. (See ACT, February 1996.) Challenging this conclusion, the independent National Defense Panel—which was created by Congress to review the findings of the QDR—argued that "the move to START II force levels should proceed even if the Duma fails to act on START II this year."

Under START I, the United States is expected to deploy a total of 6,000 "treaty-accountable" warheads on the following systems: 50 MX ICBMs, 500 Minuteman III ICBMs, 18 Trident ballistic missile submarines, 71 B-52H bombers and 21 B-2 bombers. (Due to START I counting rules, this force will comprise approximately 8,000 actual deployed warheads). Appearing before the Senate Armed Services Committee on May 20, Cohen estimated that it will cost the United States approximately $64 million to maintain START I force levels in fiscal year (FY) 1998 and $1 billion per year thereafter. The majority of the increase reflects the costs associated with refueling the four additional Trident submarines that will be maintained under START I.

General John Shalikashvili, chairman of the Joint Chiefs of Staff, told the committee that, in light of this significant potential cost, he would support adjustments to the current congressional prohibition on U.S. strategic nuclear force reductions below START I levels. Shalikashvili said the Joint Chiefs believe "it would be good if we could have the freedom to discuss with [Congress] alternatives that would, on the one hand, meet our security needs, [and] on the other hand not undermine the process of putting the requisite pressure on the Duma to ratify START II. We believe there is a middle way that we can find that will accomplish that."


Missile Defense Issues

As part of its comprehensive analysis, the QDR also evaluated U.S. ballistic missile defense policy and programs. In particular, the review reaffirmed the Clinton administration's "three-plus-three" program, which calls for the development of the initial elements of an NMD system by 2000, at which time the United States will assess the long-range ballistic missile threat to its territory and be in a position to deploy such a system by 2003 if necessary. If no decision is taken to deploy, the United States will continue development efforts while maintaining a three-year deployment capability.

Nevertheless, the QDR determined that the program as currently funded would not enable the United States to meet the 2000 deadline for making a possible NMD deployment decision. Therefore, the review added $2.3 billion to the program over the next five years, but cautioned that even with this additional funding it will "remain a program with very high schedule and technical risk."

With respect to the administration's theater missile defense (TMD) program, the QDR pushed back the deployment date for THAAD from 2004 (which had just been announced in December 1996) again to 2006. This delay was necessary because THAAD has failed in all four of its intercept attempts over the past year and a half. In addition, the QDR provided funding through 1999 for the Medium Extended Air Defense System (MEADS), which is being developed by the United States, Germany and Italy for NATO deployment (previously MEADS was funded through FY 1998).

The QDR also reaffirmed the administration's commitment to other TMD programs: the Army's Patriot Advanced Capability3 (PAC3) system, an improvement over the system deployed during the Gulf War; the Navy's Area Defense ("lower-tier") system; the Navy's Theater-Wide Defense ("upper-tier") system; and the Air Force's Airborne Laser program, which is in the early stages of development.

The Future of U.S. Nuclear Weapons Policy


Executive Summary of the National Academy of Sciences Report

The National Academy of Sciences (NAS) released a major policy report on June 17, entitled The Future of U.S. Nuclear Weapons Policy , recommending substantial changes in U.S. nuclear weapons policies and forces to bring them into conformity with the post-Cold War environment. The report's central recommendation is that the use of nuclear weapons be limited to a core mission of deterring the use of nuclear weapons by others.

The study calls for a program of progressive constraints to reduce U.S. and Russian nuclear arsenals to 1,000 total warheads each and then, if security conditions permit, to a few hundred warheads, provided adequate verification procedures and transparency measures have been implemented. Parallel steps are also necessary to reduce high alert levels and to substitute much more selective targeting than now incorporated in present war plans. While the path to the prohibition of all nuclear weapons is not yet clear, the report examines the conditions that would have to exist for this to be acceptable and suggests various ways it might be achieved.

The report was prepared by the NAS Committee on International Security and Arms Control (CISAC), a group of distinguished scientists, retired senior military officers and experts policy analysts, most of whom have been closely associated with various aspects of nuclear security affairs. (See p. below.)



The debate about appropriate purposes and policies for U.S. nuclear weapons has been under way since the beginning of the nuclear age. With the end of the Cold War, the debate entered a new phase, propelled by the post-Cold War transformations of the international political landscape and the altered foreign policy challenges and opportunities that these changes are bringing about. This report—based on an exhaustive reexamination of the issues addressed in the committee's 1991 report on The Future of the U.S.Soviet Nuclear Relationship—describes the state to which U.S. and Russian nuclear forces and policies have evolved since the Cold War ended, the reasons why further evolution is desirable, and the shape of a regime of progressive constraints responsive to these reasons. It concludes with a discussion of the conditions and means under which, in the longer term, it could become desirable and feasible to prohibit the possession of nuclear weapons altogether.



The first Strategic Arms Reduction Treaty (START I), signed in 1991 as the Cold War was ending and now being implemented by both the United States and Russia, will reduce the number of strategic nuclear warheads deployed by the two countries from 13,000 and 11,000, respectively, to about 8,000 each. START II, signed in 1993 and ratified by the United States in early 1996 but not yet (as of this writing) ratified by Russia, would further limit the number of deployed strategic warheads to 3,000 to 3,500 on each side. At the Helsinki summit in March 1997 Presidents Clinton and Yeltsin agreed to seek a START III treaty with a level of 2,000 to 2,500 deployed strategic nuclear warheads. Unilateral initiatives since the Cold War began winding down have also reduced very substantially the numbers of deployed non-strategic warheads, especially on the U.S. side. In addition, nuclear testing has ended, the United States and Russia have agreed not to target their missiles against each other on a day-to-day basis, and production of weapons-grade fissile material has stopped in the United States and is expected to stop soon in Russia.

These actions have unambiguously halted and reversed the bilateral nuclear competition that was the most conspicuous characteristic of the Cold War's military confrontation but, unfortunately, have not sufficiently altered the physical threat that these weapons pose. The reduced forces could still inflict catastrophic damage on the societies they target or could target, and the thousands of non-deployed and non-strategic nuclear warheads not addressed by the START process and likely to be retained without further agreements will pose substantial risks of breakout, theft, or unauthorized use. In addition, the United States and its NATO allies retain their Cold War "weapons of last resort" doctrine that allows the first use of nuclear weapons if deemed necessary to cope with non-nuclear attacks, and Russia has announced that it is abandoning the Soviet Union's no-first-use pledge in order to adopt a position similar to NATO's.

The basic structure of plans for using nuclear weapons appears largely unchanged from the situation during the Cold War, with both sides apparently continuing to emphasize early and large counterforce strikes and both remaining capable, despite reductions in numbers and alert levels, of rapidly bringing their nuclear forces to full readiness for use. As a result, the dangers of initiation of nuclear war by error (e.g., based on false warning of attack) or by accident (e.g., by a technical failure) remain unacceptably high. (On the Russian side, the dangers of erroneous, accidental, or unauthorized nuclear weapons use may be even higher than during the Cold War because of subsequent deterioration of the military and internal-security infrastructure and of morale.) The continuing competitive assumptions underlying some official discussions of the U.S. Russian nuclear relationship, when coupled with the postures of the forces and the potential for destabilizing deployments of ballistic missile defenses, pose the risk that the arms control fabric woven during the Cold War and immediately thereafter could unravel.

In addition, continued actions by the United States and Russia to reduce their nuclear arsenals—and to reduce the roles assigned to those arsenals—are needed to help bring the other declared and undeclared nuclear weapons states into the arms reduction process and to strengthen the global nonproliferation regime. The effectiveness of that regime depends on the full support and cooperation of a large number of non-nuclear weapons states in the maintenance of a vigorous International Atomic Energy Agency with the inspection powers and resources to do its job, in the implementation of effective controls on the transfer of sensitive technologies and in the creation of transparency conditions conducive to building confidence that proliferation is not taking place. The degree of commitment of the non-nuclear weapons states to these crucial collective efforts will surely depend at least in part on impressions about whether the nuclear weapons states are working seriously on the arms reduction part of the global nonproliferation bargain.



During the Cold War, nuclear deterrence was the bedrock of U.S. strategy for preventing both nuclear war and major conventional war because a more effective alternative was not apparent: the adversarial U.S.Soviet relationship made it seem imprudent to rely on good intentions to preclude nuclear attack or massive conventional assault; the character of nuclear weapons and the diverse means for delivering them meant that attempts to defend the United States or its allies against nuclear attacks on their populations could be overcome with much less effort than would have to be invested in the defenses; highly survivable basing modes for significant parts of each side's nuclear forces made it impractical to execute a disarming first strike even if conflict seemed imminent; and concern about the powerful conventional forces of the Warsaw Treaty Organization (and, in Asia, those of China and North Korea) motivated the United States and its allies to adopt a first-use-if-necessary nuclear posture to deter large-scale conventional attacks.

But nuclear deterrence itself was (and is) burdened with an array of dilemmas and dangers. For example, deterrence is likely to succeed only if there are credible plans for what to do if it fails, but constructing such plans is exceedingly difficult, and attempts to make the threat of nuclear retaliation credible can be seen as aggressive advantage seeking by the other side. This raises tensions, stimulates arms races, or increases the chance of nuclear war from crisis instability or accident. In addition, the assertion by some countries of a need and a right to have a nuclear deterrent may encourage additional countries to assert the same need and right, leading to further nuclear proliferation.

This committee has concluded that the dilemmas and dangers of nuclear deterrence as practiced by the United States in the past can and should be alleviated in the post-Cold War security environment by confining such deterrence to the core function of deterring nuclear attack, or coercion by threat of nuclear attack, against the United States or its allies. That is, the United States would no longer threaten to respond with nuclear weapons against conventional, chemical, or biological attacks. Given adequate conventional forces, the active and conspicuous role given to nuclear weapons during the Cold War can be greatly reduced without significant adverse effect on the probability of major war or on this country's ability to deal effectively with regional conflicts where its vital interests and those of its allies are at stake. The committee believes that Russia and the other nuclear weapons states can be persuaded to reach a comparable conclusion.

In all likelihood the United States will consider it necessary to continue to rely on the core function of nuclear deterrence as long as nuclear weapons continue to exist in the possession of states that might consider using them against this country or its allies. The committee assumes that some—although it is hoped not all—other nuclear weapons states will similarly consider it necessary to retain some nuclear weapons for "core deterrence." But the size and scope of the efforts deemed necessary by the United States and others to fulfill the core function presumably will shrink in parallel with what the committee hopes is the declining plausibility that any state would consider mounting a nuclear attack on anyone. Moreover, there are strong reasons to make every effort to hasten the arrival of international conditions in which threats of nuclear attack are simply no longer thinkable, so that the practice of deterrence with all its dilemmas and dangers would no longer be necessary.

As long as nuclear weapons exist, this very existence will exert a deterrent effect— existential deterrence—against unrestricted conventional war among the major powers, since it will be recognized that, in a world with nuclear weapons, such conflicts might well lead to their use, with intolerable destruction as the result. Indeed, even the existence of the idea of nuclear weapons—more specifically, the ability of many states to make them—is enough to create an existential deterrent effect against large-scale conflicts of all kinds. That is not to say that this effect would necessarily always be sufficient to prevent conflict in the future, any more than it has always been in the past. But it could provide a part of the assurance required, in an international system much different than today's, that all-out wars are unlikely to occur.



If only the core function of nuclear weapons retains validity, fundamental changes in the nuclear force structures and operational practices of the major nuclear powers become both possible and desirable. Accordingly, the committee has concluded that the United States should pursue a two-part program of change in its nuclear weapons policies.

  • The first part of the program is a near and midterm set of force reductions—together with accompanying changes in nuclear operations and declaratory policies and with measures to increase the security of nuclear weapons and fissile materials worldwide—to diminish further confrontational and potentially destabilizing aspects of force postures, to reduce the risks of erroneous, unauthorized, or accidental nuclear-weapons use, and to help curb the threat of further nuclear proliferation. In their early phases these measures are largely bilateral ones between the United States and Russia, and close cooperation between the two countries is essential for success.
  • The second part of the program is a long-term effort to foster international conditions in which the possession of nuclear weapons would no longer be seen as necessary or legitimate for the preservation of national and global security.
Nuclear force reductions and changes in nuclear operations would increase U.S. and global security in important ways.
  • First, reducing U.S. and Russian nuclear forces and revising operations for the mission of fulfilling only the core function will decrease the continuing risk of accidental, erroneous, or unauthorized use of nuclear weapons for several reasons. Smaller arsenals will be easier to safeguard and protect from accident, theft, and unauthorized use, not only by virtue of reduced numbers of weapons to monitor at a smaller number of sites but also by permitting retention of only those weapons with the most modern safety and security features. Reducing alert rates, decreasing capacities to use nuclear weapons quickly and with little warning, abandoning plans for the rapid use of nuclear weapons, and deploying cooperative measures to assure states that forces are not being readied for attack should reduce the probability and consequences of erroneous nuclear weapons use—for example, on false warning of attack. (Of course it is extremely important to take care that reductions in deployed nuclear warheads—and dismantlement of the warheads made surplus as a result—do not lead to countervailing increases in the dangers of theft and unauthorized use as a consequence of inattention to the challenges of safe storage of these weapons and the nuclear materials removed from them.)
  • Second, further reductions will bolster the nuclear nonproliferation regime. U.S.Russian nuclear arms reductions will not in themselves dissuade a state bent on acquiring nuclear weapons; today's undeclared nuclear powers and would-be proliferators are driven above all by regional security concerns. In such cases, the denial of material and technical resources and a combination of political and economic incentives and disincentives provide the greatest leverage. But U.S. and Russian progress in arms reductions helps shore up global support for anti-proliferation measures; and lack of such progress can strengthen the influence of those arguing for nuclear weapons acquisition in countries where this is under internal debate.
  • Third, continued actions by the United States and Russia to reduce their nuclear arsenals—and the roles and missions assigned to those arsenals—will help persuade the other declared and undeclared nuclear weapons states to join the arms control process. At planned START II levels, for example, under which it is estimated that the United States and Russia each would retain a total of about 10,000 nuclear warheads, deployed and in reserve, the other nuclear powers have little motivation to submit their much smaller arsenals to any form of control.


The program that the committee recommends would shift the focus of U.S. nuclear policy. While preserving the core function of deterring nuclear aggression, nuclear forces would be reduced, their roles would be more narrowly defined, and increased emphasis would be placed on achieving higher standards of operational safety.

Building on past nuclear arms control agreements and the anticipated START III agreement, future bilateral U.S.Russian negotiations should center on specific means to achieve these goals. The first step is to encourage the Russian Duma's ratification of START II by beginning now to discuss a START III agreement limiting the number of deployed strategic warheads to about 2,000 on each side.

Establishing progressive constraints on nuclear operations is equally urgent; additional efforts should be pursued in parallel with, but not linked to, discussions of a START III agreement. Such constraints would include programs to reduce alert levels further and progressively to reorient nuclear doctrine away from the requirement to plan for rapid, massive response. Limits on ballistic missile defenses consistent with the Anti-Ballistic Missile (ABM) Treaty would be maintained.

A continuing high priority effort is also needed to improve the protection of nuclear weapons and fissile materials in Russia. Joint U.S.Russian work along these lines, which has been going on since 1991 under the Nunn-Lugar Cooperative Threat Reduction Program, complements and strengthens arms reductions and other changes in nuclear policies. (Because this committee and other NRC committees have recently offered detailed analysis and recommendations on this subject in other reports, the committee does not treat it in detail here.)

During the Cold War, reducing the risk of a surprise attack appeared to be more important than the risks generated by maintaining nuclear forces in a continuous state of alert. With the end of that era, the opposite view is now more credible. This has important implications for U.S. nuclear policy and calls for dramatically reduced alert levels. Elimination of continuous-alert practices should be pursued as a principal goal in parallel with, but not linked to, START III. As a related confidence-building measure, the United States and Russia should adopt cooperative practices to assure each other that they are not preparing for a nuclear attack.

With the Cold War over, planning to retaliate massively against a nuclear attack is not the appropriate basis for making responsible decisions regarding the actual use of nuclear weapons. Operational doctrine regarding the magnitude and timing of any actual retaliation in response to a nuclear attack should be revised. The United States should adopt a strategy that would permit much more selective targeting options and that would be based neither on predetermined prompt attacks on counterforce targets nor on automatic destruction of cities. The presumption instead would be that nuclear weapons, if they were ever to be used, would be employed against targets that would be designated in response to immediate circumstances—in the smallest possible numbers. Some changes in this direction have begun, but the move to a more flexible planning system should be accelerated.

Together, positive and negative security assurances and guarantees have been a useful policy tool to ensure that friends and allies of the United States are not penalized by foregoing nuclear weapons. The United States could do more, however, to make negative security assurances and guarantees serve nonproliferation interests. Most important, the United States should adopt no-first-use of nuclear weapons as its declaratory policy at an early date. Changing to a no-first-use policy will, of course, require consultation with allies to reassure them that the United States will meet, by non-nuclear means, its obligations to come to their aid in the event of a non-nuclear attack on them.

Efforts to ban nuclear weapons from specific regions or environments strengthened nonproliferation in the past and helped to limit the perceived utility of nuclear weapons. The United States should continue to support these agreements and sign them without reservations that undermine their basic purpose, consistent with the unequivocal no-first-use policy recommended above. A new nuclear weapon free zone in Central Europe would, for example, offer immediate security advantages to Russia as well as NATO.

The committee has concluded that the changed international security environment makes possible further reductions in nuclear armaments. After the reductions envisioned in a START III accord, reduction to about 1,000 total warheads each for the United States and Russia would be a logical next step. (All nuclear warheads—regardless of type, function, stage of assembly, associated delivery system, or basing mode—would then be included in the negotiated limits.) A force of this size could effectively maintain the core function against the most challenging potential U.S. adversaries under any credible circumstances. This reduction process must ensure stability at each rung of the ladder, requiring survivable nuclear forces not at risk from a first strike.

Verifying limits on total nuclear warheads is substantially more difficult than verifying limits on their delivery vehicles. Verifying numbers of non-deployed and non-strategic warheads, in particular, would require transparency measures regarding the production, storage, and dismantling of nuclear warheads, as well as a mechanism for exchanging and verifying information about the location and status of warheads. Since nuclear weapons can be small and portable and not easily detectable by technical means, however, a regime that would provide high confidence of locating a small number of hidden warheads would be extremely difficult to achieve. Even an imperfect verification regime would greatly reduce the uncertainties in present U.S. estimates of the number of Russian warheads.

Fulfilling the goals of current arms control initiatives and successfully providing for much deeper reductions will also require improved standards of accounting, transparency, and physical security for fissile materials. Efforts to control fissile materials must address not only the problems presented by military stockpiles but also by civilian use of such materials, in particular plutonium produced by reprocessing. A fissile material cutoff would be a significant nonproliferation measure and should continue to be strongly supported by the United States.

The ABM treaty will continue to play a crucial role in a world in which the numbers of nuclear weapons are drastically reduced and the role of nuclear weapons is restricted to the core function. Maintaining and enhancing its integrity in light of changes in offensive nuclear capabilities will require periodic evaluation. The focus of the U.S. ballistic missile defense research and development program should be to field a mobile system capable of defending relatively small areas against projected theater ballistic missile threats, which the committee believes will remain limited to a range of roughly 1,000 kilometers for some time.

The achievement of U.S.Russian reductions to a mutually agreed level of about 1,000 total warheads each should not represent the final level for nuclear arms reductions. There will still be powerful reasons to continue down to a level of a few hundred nuclear warheads on each side, with the other three declared nuclear powers at lower levels, or with no remaining nuclear forces.

The small numbers of nuclear weapons presumably now held by the undeclared nuclear states—India, Israel, and Pakistan—would become a key issue when the United States and Russia, as well as the other declared nuclear powers, consider reductions to very small numbers of warheads. High priority should be given to diplomatic strategies tailored to the security perceptions of each state in order to freeze or reduce and, if possible, eliminate these undeclared programs in parallel with the reduction programs of the nuclear powers.

The committee's analysis does not assume a fundamental change in the nature of international relations in order to achieve these low levels of nuclear arms. It does assume unprecedented cooperation and transparency among all classes of nuclear powers on the specific issue of nuclear arms reductions. A few hundred nuclear weapons would be sufficient to deter nuclear attack through their potential to destroy essential elements of the society of any possible attacker. These remaining nuclear forces would have to be survivable and their command-and-control structure adequately redundant and robust; and widespread and effective national ballistic missile defenses must be absent. The operational posture of the much smaller forces must be designed for deliberate response rather than reaction in a matter of minutes.



The end of the Cold War has created conditions that open the possibility for serious consideration of proposals to prohibit the possession of nuclear weapons. It is not clear today how or when this could be achieved; what is clear is that comprehensive nuclear disarmament should be undertaken only in circumstances such that, on balance, it would enhance the security of the United States and the rest of the world.

The committee uses the word "prohibit" rather than "eliminate" or "abolish" because the world can never truly be free from the potential reappearance of nuclear weapons and their effects on international politics. Even the most effective verification system that can be envisioned would not produce complete confidence that a small number of nuclear weapons had not been hidden or fabricated in secret. More fundamentally, the knowledge of how to build nuclear weapons cannot be erased from the human mind. Even if every nuclear warhead were destroyed, the current nuclear weapons states, and a growing number of other technologically advanced states, would be able to build nuclear weapons within a few months or few years of a national decision to do so.

A durable prohibition on nuclear weapons would have three main benefits:

  • It would virtually eliminate the possibility of use—whether authorized and deliberate or not—of nuclear weapons by states now possessing them. Viewed in light of the possibility of reconstitution of such arsenals in a crisis, prohibition can be seen as extending the dealerting measures recommended in the near-term part of the program—that is, increasing the time required to ready nuclear weapons for use from hours or days to months or years.
  • It would reduce the likelihood that additional states will acquire nuclear weapons. Although the Nuclear Nonproliferation Treaty currently enjoys almost universal adherence, the nuclear weapons states cannot be confident of maintaining indefinitely a regime in which they proclaim nuclear weapons essential to their security while denying all others the right to possess them.
  • It would deal decisively with the uncertain moral and legal status of nuclear weapons, as underlined by the recent advisory opinion of the International Court of Justice.
Nuclear disarmament poses risks as well as benefits, however:
  • The prohibition on nuclear weapons might break down via cheating or overt withdrawal from the disarmament regime. To reduce these risks, a disarmament regime would have to be built within a larger international security system that would be capable not only of deterring or punishing the acquisition or use of nuclear weapons but also of responding to major aggression.
  • Comprehensive nuclear disarmament could remove the moderating effect that nuclear weapons appear to have had on the behavior of states. The nuclear era represents the longest period without war among the major powers since the emergence of the modern nation state in the sixteenth century. Thus, it is argued that, if the major powers believed the risk of nuclear war had been eliminated, they might initiate or intensify conflicts that might otherwise have been avoided or limited. But there have been, and continue to be, profound changes in the structure of the international order that are acting to reduce the probability of major war independent of nuclear deterrence. Moreover, even if all nuclear weapons were eliminated, the inherent capacities to rebuild them could act as a deterrent to the outbreak of major wars.
If the preconditions for agreed prohibition of nuclear weapons are met, however, the committee believes that a path to eventual prohibition can be found. One possible path for managing the transition to comprehensive nuclear disarmament would involve having an international agency assume joint or full custody of the arsenals remaining during the transition to prohibition. Alternatively, nations might find it preferable to bypass the intermediate step involving an international agency and proceed directly to negotiations to prohibit nuclear weapons either globally in a single agreement or in steps involving successive expansions in the number and geographical scope of nuclear weapon free zones.

It will not be easy to achieve the conditions necessary to make a durable global prohibition on the possession of nuclear weapons both desirable and feasible. Complete nuclear disarmament will require continued evolution of the international system toward collective action, transparency, and the rule of law; a comprehensive system of verification, which itself will require an unprecedented degree of cooperation and transparency; and safeguards to protect against the possibility of cheating or rapid breakout. As difficult as this may seem today, the process of reducing national nuclear arsenals to a few hundred warheads would lay much of the necessary groundwork. For example, the stringent verification requirements of an agreement on very low levels of nuclear weapons and fissile materials might by then have led to some new or expanded international agency with vigorous powers of inspection. The committee has concluded that the potential benefits of a global prohibition of nuclear weapons are so attractive relative to the attendant risks that increased attention is now warranted to studying and fostering the conditions that would have to be met to make prohibition desirable and feasible.

In any case, the regime of progressive constraints constituting the committee's proposed near to midterm program makes good sense in its own right—as a prescription for reducing nuclear dangers without adverse impact on other U.S. security interests—regardless of one's view of the desirability and feasibility of ultimately moving to prohibition.


CISAC Study Panel

William F. Burns, study chair, Major General (U.S. Army, Retired)

John P. Holdren, committee chair, Professor, Center for Science and International Affairs, John F. Kennedy School of Government, Harvard University

John D. Steinbruner, committee vice chair, Senior Fellow, Foreign Policy Studies Program, The Brookings Institution

George Lee Butler, Vice President, Peter Kiewit Sons, Inc.

Paul M. Doty, Director Emeritus, Center for Science and International Affairs, Harvard University

Steve Fetter, Associate Professor, School of Public Affairs, University of Maryland, College Park

Alexander H. Flax, President Emeritus, Institute for Defense Analyses and Senior Fellow, National Academy of Engineering

Richard L. Garwin, Fellow Emeritus, Thomas H. Watson Research Center, IBM Corporation

Rose Gottemoeller, Deputy Director, International Institute for Strategic Studies

Spurgeon M. Keeny, Jr., President, Arms Control Association

Matthew Meselson, Professor, Department of Molecular Biology and Cellular Biology, Harvard University

Wolfgang K. H. Panofsky, Professor and Director Emeritus, Stanford Linear Accelerator Center, Stanford University

C. Kumar N. Patel, Vice Chancellor—Research, University of California at Los Angeles

Jonathan D. Pollack, Senior Advisor for International Policy, The RAND Corporation

Robert H. Wertheim, Rear Admiral (U.S. Navy, Retired)

Jo L. Husbands, study director, Committee on International Security and Arms Control, National Academy of Sciences

Copies of The Future of U.S. Nuclear Weapons Policy are available for $15.00 (plus $4 for shipping and handling) from the National Academy of Sciences, 2101 Constitution Avenue, NW, Washington, DC, 20148 or phone 202-334-2138 or 1-800-624-6242.

Retired Generals Re-Ignite Debate Over Abolition of Nuclear Weapons

Please Contact ACA if you would like a copy of this story.


Subscribe to RSS - Strategic Policy