U.S.-Russian Nuclear Arms Control Negotiations—A Short History

An accomplished negotiator puts nuclear arms control in perspective—what it has achieved, where it has failed and what it can do for our future security.


Brian Hubble

In my line of work, you have to have a long memory. Periods of success in negotiations are followed by droughts, because of politics, military upheaval, arms buildups—yes, sometimes the weapons have to be built before they can be reduced—or a sense of complacency: “We have arms control treaties in place; let’s just focus on implementing them.” In those cases, new thinking and new negotiations may slow or even stop. Yet, the national security interest of the United States continues to drive the necessity for nuclear arms control.

The calculation of our own national security interest must always be front and center when we consider a nuclear negotiation. Sometimes arms control is touted as an absolute good, one that should be pursued for its own sake. We do have international obligations in this realm, most prominently the commitment under Article VI of the Non-Proliferation Treaty to reduce and eliminate nuclear weapons until we reach zero. This commitment is shared by the other NPT nuclear weapon states—France, the U.K., Russia and China; and sometimes it gets a boost, as it did when President Barack Obama strongly reiterated U.S. intent to proceed on the path to zero nuclear weapons during his speech in Prague in April 2009, the first major foreign policy speech of his presidency.

That international obligation is important, but still we must consider first and foremost our own national security interest. I think about that interest as follows: Nuclear arms control is the only way that we can attain stable and predictable deployments of these most fearsome weapons, and it is the only way that we can assure that we won’t be bankrupted by nuclear arms racing. These points are especially important now, as we contemplate a world where China has more nuclear weapons and more missiles with which to deliver them.

China now has many fewer nuclear weapons than the United States and Russia, and it has not yet shown an interest in coming to the table to negotiate constraints on them. It is constrained by its doctrine, which has held that China will not strike first with nuclear weapons and will only maintain enough secure nuclear weapons to ensure a second strike can take place if another country strikes China first. In the Chinese view, this doctrinal approach forges a kind of insurance policy for the international community. However, since China has now started to build more kinds of nuclear delivery systems, including long-range submarine-launched ballistic missiles, there is real concern that its doctrine may be changing.

So all of us need to think about the long arc of nuclear arms control—what it has accomplished, where it has failed and what it can do for our future security. In looking at the history, this article pulls the different strands from one period into the next, but does not delve into the details of any particular agreement. Nuclear arms control experts may take exception to this surface skimming, but I think it makes sense as food for thought: to remind us all how we determined the value of nuclear arms control in the first place, and how we have sustained it over time. Now we have to consider what makes sense for the future.

From Hiroshima to the Cuban Missile Crisis

The early history of nuclear arms control was wedded to the closing days of World War II: Hiroshima and Nagasaki had taken place; the United States had won the race to acquire nuclear weapons. To its credit, U.S. leadership immediately grasped that efforts should be made to control this new weapon of mass destruction and, if possible, share the benefits of the atom—nuclear energy—internationally. Secretary of State Dean Acheson joined with David Lilienthal, chairman of the Tennessee Valley Authority (responsible for fissile material production), and four other prominent figures to prepare what became the Acheson-Lilienthal Report on the International Control of Atomic Energy. Its goal was to ensure that the United Nations would control nuclear resources and ensure that they were only used for peaceful purposes. Those countries acquiring nuclear weapons technology would give it up; and once U.N. controls over their programs were in place, the United States would relinquish its arsenal.

Bernard Baruch was the U.S. negotiator who presented this proposal to the U.N. Security Council in January 1946. It was already evident that the Soviet Union was unlikely to cooperate, so Baruch modified the plan in several ways, importantly seeking to prevent the UNSC veto from being used in this setting. The Soviets presented their own competing Gromyko Plan, which called for the immediate prohibition of nuclear weapons and would have caused the United States to give up its arsenal immediately. These competing plans were debated until December 1946, when the Baruch Plan was put to a vote before the Security Council. Ten of the 12 members voted in favor, but the USSR and Poland abstained. The measure was not passed, so the first major international effort at nuclear arms control failed.

As the 1950s unfolded, both the United States and Soviet Union continued to test more and more powerful weapons, racing to acquire the hydrogen bomb.

As the 1950s unfolded, both the United States and Soviet Union continued to test more and more powerful weapons, racing to acquire the hydrogen bomb. The first Soviet test was at Semipalatinsk in 1955, and the first U.S. test was at Bikini Atoll in 1956. Both continued to build nuclear warheads, so that by the mid-1960s, the United States had an arsenal of approximately 32,000 warheads, and the Soviets, according to the account of former Minister of Atomic Energy Viktor Mikhailov in the Sept. 26, 1993 New York Times, had more than 40,000.

Thus, the stage was set for a major nuclear crisis in the Cold War years, when the United States and Soviet Union were constantly confronting each other: whether on the diplomatic front in the United Nations, on the borders between NATO and the Warsaw Pact, in Berlin, or in regional wars and insurgencies across Eurasia and into Africa and Latin America. The fulcrum for communist revolution in Latin America, of course, was Cuba. I am not going to recount the details of the October 1962 Cuban Missile Crisis here; Graham Allison’s Essence of Decision is still the classic analysis (see 2nd ed., Longman, 1999). New analyses were also undertaken once the old Soviet archives opened up and Russian participants started interacting with their U.S. counterparts at the time of the 40th anniversary. A very good wrap-up of this work appears on the National Security Archive website (see nsarchive2.gwu.edu). Suffice it to say, we came close to nuclear war.

The Cuban Missile Crisis deeply shook the leaders on both sides, and so it provided the first impetus to pursue true nuclear arms control. President John F. Kennedy’s American University commencement speech in June 1963 was a U.S. watershed: He declared an immediate moratorium on U.S. nuclear tests in the atmosphere, to be maintained as long as others did not test, and announced an agreement with Soviet leader Nikita Khrushchev and British Prime Minister Harold Macmillan to begin negotiation of a Comprehensive Test Ban Treaty. By August 1963, a mere two months later, an atmospheric test ban had been negotiated and signed: the Limited Nuclear Test Ban Treaty. As an aside, it is interesting that there was an environmental impetus to these negotiations that made them popular both among the U.S. public and internationally. People were realizing that strontium-90 from atmospheric testing was getting into the food supply—most crucially, into children’s milk.

NATO and the NPT

Now fast-forward to the mid-1960s, when a lot was going on. First, beginning in 1965, the Non-Proliferation Treaty was under negotiation. This involved tough bargaining about the behavior of those states that had already tested nuclear weapons; they turned out, eventually, to be the five permanent members of the U.N. Security Council: the United States, USSR, China, U.K. and France. The rest of the states also drove a hard bargain, eventually ending up with a three-pillared construction for the treaty: all would work to prevent nuclear proliferation; all would cooperate to share the benefits of the peaceful atom; and all would pursue nuclear disarmament. The disarmament pillar was particularly directed at the nuclear weapons states: they would work steadily to eliminate nuclear weapons while the other countries would eschew them. It was the grand bargain of the NPT inscribed in its Article VI.

Among those who had tested nuclear weapons, the bargaining was particularly sharp between the United States and the Soviet Union, because they had tested the most and had deployed by far the biggest arsenals, which is still the case today. It also brought in the NATO Alliance, which had been suffering its own version of an existential threat. In 1967 France withdrew from the military command structure of NATO and threw its headquarters—civilian and military—out of Paris. This, in my view, is the most difficult crisis that the Alliance has weathered, and it led to some deep soul-searching on the part of the allies, led by Belgian Foreign Minister Pierre Harmel. He produced a short and succinct report that called for détente with the Soviets while continuing to pursue firm deterrence measures. This basic conclusion of the Harmel Report set the stage for NATO to participate fully in arms control policymaking and negotiations with the USSR.

Where the NPT was concerned, the Soviets were trying to destroy a bête noire that had preoccupied them since World War II—the specter of a German nuclear weapons program. It is easy to forget, but when the NPT was being negotiated, a number of European states were pursuing their own nuclear weapons—not only Germany, but countries as diverse as Sweden and Switzerland. The Soviets were intent on ensuring that the Germans never got their own nuclear arsenal. They therefore agreed to the notion that certain NATO countries in Europe would have nuclear weapons on their territories, but those weapons would remain in full control of the United States. For the Soviets, the NPT, which was opened for signature in 1968, was the instrument by which Germany would remain a non-nuclear weapon state, and for that reason Moscow accepted U.S. nuclear weapons on the territory of some NATO countries.

I note this because for the past few years the Russians have been complaining that the United States is “violating” the NPT by deploying nuclear weapons under its control on NATO allied territory. However, the NPT negotiating record clearly shows that their Soviet predecessors agreed to these arrangements. It was worth it to them to keep nuclear weapons out of the hands of the Germans.

Glassboro: Toward the First Détente

The Glassboro Summit is an important but little-remembered moment in arms control history that took place June 23-25, 1967, at Glassboro State College in New Jersey. Now called Rowan University, the site was chosen because of its proximity to New York City, where Soviet Premier Alexei Kosygin was addressing the United Nations over the Middle East crisis—the Six-Day War had occurred just a few weeks earlier. Tensions were also high over the Vietnam War. Kosygin wrote to President Lyndon Johnson, and the two agreed to meet.

It was the first time that the United States presented to Soviet leaders the proposition that it is important to limit strategic ballistic missile defenses as well as strategic nuclear offensive weapon systems. It is a simple argument: If strategic strike offensive missiles are limited and ballistic missile defense systems continue to improve technologically and expand operationally, then over time, the defense systems will begin to undermine the strategic offensive deterrent of one party or the other. The Soviet leaders were baffled: How could limiting defenses ever be a good thing? Kosygin and his colleagues were no doubt confounded because Soviet military doctrine and strategy, including nuclear doctrine, were strictly the purview of the Soviet military leadership. It was doubtless the first time that the Communist Party leadership had ever heard anything in detail about the nuclear offense-defense relationship.

But by the time President Richard Nixon met in Moscow with General Secretary Leonid Brezhnev in May 1972, the Soviets were convinced of the need to limit defensive as well as offensive systems. Nixon and Brezhnev signed the Anti-Ballistic Missile Treaty, which limited each side to 100 defensive launchers in two sites each; they also signed the Strategic Arms Limitation Treaty (SALT I), which was called an interim treaty because it simply froze the number of launchers then deployed. This is significant because the same theme comes up again and again in the history of U.S.-Russian arms control policy: namely, the delicacy of the offense-defense balance and the importance of its maintenance to strategic stability.

A treaty that is being hollowed out from the inside is no longer in the U.S. national security interest, which must be the litmus test for any nuclear arms control treaty.

Fast-forward now to 1979 and the completion of SALT II, the first treaty to seek to limit strategic offensive arms. It never entered into force because of the Soviet invasion of Afghanistan in December of that year—but President Jimmy Carter was already facing an uphill battle in getting the advice and consent of the Senate to its ratification. The reason? U.S. hawks and skeptics were sharply criticizing what they called the Soviet breakout potential—the advent of MIRV technology. MIRV stands for multiple independently targetable reentry vehicles, which allow a country to deploy multiple warheads on top of individual missiles. Because the Soviets were deploying heavy missiles—the SS-18 and SS-19 intercontinental ballistic missiles (ICBMs)—they had more capability to carry warheads and deliver them. This was the famous “throw-weight” debate of the 1970s and 1980s. It was feared that they had enormous potential to deploy and deliver many more warheads than the United States could, thus upsetting the strategic balance.

Of course, two can play at this game; and within a short time the United States was also deploying very capable MIRVs on its ground-based systems, the ICBMs, but more so on its submarine-based systems, the sea-launched ballistic missiles (SLBMs). The United States maintained much quieter submarines in that era, and was able to deliver more accurate strikes from sea-based platforms than the Soviets. The Soviets thus had cause to consider what would happen should the United States choose to deploy an unlimited number of highly accurate warheads at sea, where they could not be easily tracked and targeted.

MIRV technology, in my view, became the real impetus for the two sides to agree in the 1980s to Strategic Arms Reduction Talks. It had proved futile to try to limit strategic systems; they had to be reduced, and reduced in such a way that each side could be certain that the other side was not able to out-deploy it in warhead numbers.

Destabilizing Developments and the INF Treaty

The other potentially destabilizing development in the late 1970s and early 1980s was the advent of ground-launched intermediate-range missiles, both ballistic and cruise missiles. (Intermediate range is considered to be between 500 and 5,500 kilometers.) These missiles were destabilizing because they either had a fast flight time to target (the ballistic systems), or were stealthy flyers (the cruise systems, which were able to fly below radar coverage). In both cases, they did not give leaders time to make nuclear launch decisions. Thus, in theory they could be used for a “decapitating” first strike, destroying the command and control potential of the other country and leaving it helpless to launch a response strike.

When the Soviets began to deploy their SS-20 missiles in 1976, it got everybody’s attention not only in Washington, but among the NATO allies in Europe: Could the Soviet Union now attack and destroy Berlin or Paris or London without warning? Would this threat alone “decouple” NATO Europe from the United States—i.e., would the United States ever be willing to respond to such an attack on a NATO country by launching its intercontinental systems and bringing down a response strike on U.S. territory? Would it not be more likely to let NATO go?

These are the debates that raged at NATO and among NATO capitals during the late 1970s and early 1980s. They led to one of the most significant decisions ever taken at NATO—the dual-track decision to deploy intermediate-range ground-launched missiles such as the Pershing-2 in Europe, and to push the USSR to begin to negotiate. This is the most significant period during which we built up weapons in order to bring the other side to the negotiating table. The decision was very controversial, although it was in line with the Harmel approach—to be firm on deterrence and defense but also ready to negotiate. In the end, it brought many Europeans out into the streets to protest; but it also worked.

The Intermediate-Range Nuclear Forces Treaty (INF) signed by President Ronald Reagan and USSR General Secretary Mikhail Gorbachev in 1987 was a global ban on such missiles in the hands of either the Americans or the Soviets. The treaty worked because the Soviets came to realize that, once the Pershing-2 and ground-launched cruise missiles (GLCMs) were deployed in NATO Europe, they faced the threat of a no-warning strike on critical command and control targets in Moscow. The decapitation threat had come home to them. It also worked because the United States insisted, and the Soviet Union finally agreed, that on-site inspections and other detailed verification measures were needed to ensure compliance with the treaty. This was a long-sought breakthrough in nuclear arms control.

The treaty enshrining this great arms control victory remained in place for more than three decades, until the Donald Trump administration withdrew from it in August 2019. It is worth noting, however, that per the treaty’s provisions the on-site inspection regime had ended in May 2001, 10 years after all of the INF missiles had been eliminated. As verification expert John Russell noted at the time, “The treaty has now come of age and must survive the rest of its indefinite duration without the security of regular on-site inspections” (VERTIC Briefing Paper 01/02, August 2001). That proved to be a tall order: With no on-site inspections, the treaty was vulnerable to violation.

We became aware after 2010 that the Russians were developing a ground-launched intermediate-range missile in violation of the INF Treaty, the 9M-729 (SSC-8 in NATO parlance). I raised it more than 20 times with my Russian counterparts during the period between 2013 and 2016, when I left the State Department; but the Russians always failed to acknowledge the existence of the missile. When the Trump administration engaged with them, they acknowledged the missile, but said it was not a ground-launched intermediate-range system. However, we were able to prove not only to ourselves, but also to our allies, that the missile is indeed in violation of the INF Treaty, and so all NATO allies and the U.S. allies in Asia joined the United States in calling the Russians out. The United States determined Russia to be in material breach of the treaty, which means that Russia is violating the treaty in a way that defeats its object and purpose.

My bottom line regarding this difficult decision is that the United States had good reason to withdraw from the INF Treaty, and it had the support of U.S. allies. A treaty that is being hollowed out from the inside is no longer in the U.S. national security interest, which must be the litmus test for any nuclear arms control treaty.

On Strategic Arms Reduction

Finally, it is important to get some perspective on strategic arms reduction—the first Strategic Arms Reduction Treaty (START), the Moscow Treaty (the Strategic Offensive Reductions Treaty, or SORT) and the New START Treaty. I worked on both START and New START, in 1990 and 1991 as a lowly State Department representative and in 2009 and 2010 as chief negotiator. The basic recipe for the success of both treaties has been that both sides have used them to reduce and eliminate strategic nuclear delivery vehicles and launchers, and to take warheads out of deployment. We have been certain of that because of the monitoring and verification provisions of both treaties—the on-site inspections, yes, but also the use of unhampered national technical means of verification (e.g., national satellites, radar), exchanges of data and telemetry information, notifications, and demonstrations and exhibitions, which help when compliance problems arise. This is Ronald Reagan’s “trust but verify” adage in action.

START, which entered into force in 1994, brought the number of deployed warheads down from 12,000 to 6,000 on each side. SORT, which entered into force in 2003 and was implemented while START continued in force, brought the numbers of deployed warheads down to approximately 2,200. New START, which entered into force in 2011, brought the numbers down to 1,550 on each side. So there has been real strategic nuclear arms reduction through this series of treaties.

The need of the hour is to reflect on what the long arc of our experience has taught us in the nuclear arms control arena, and think about better treaties for the future.

Note, however, that these treaties focus on the elimination of delivery vehicles (e.g., missiles) and launchers (e.g., submarine tubes) because they can easily be seen by national technical means and counted as they are destroyed. Once the warheads are off the delivery vehicles they go into storage and so become “non-deployed,” no longer counted under treaty limits. The holy grail for future nuclear arms treaties is to figure out how to eliminate warheads and verify that process, since neither the United States nor Russia so far has been willing to have foreign inspectors poke around in sensitive nuclear warhead facilities.

I do think we are now on the cusp of grasping that holy grail, and I want to make a plea for continued innovation in the arms control arena. We decided to innovate in the New START Treaty by not pursuing the counting rule approach that had been used in START. We had determined those counting rules on the basis of the maximum number of warheads with which a given missile had been tested. The heavy Russian SS-18 ICBMs, for example, were designated under the counting rule to carry 10 warheads each. In New START we went beyond the counting rule approach to actually confirm the number of objects declared to be on the front end of a missile—both nuclear warheads and non-nuclear objects (e.g., missile defense decoys). We do this through on-site inspections that are focused on the missile reentry vehicles— their front ends—where we actually determine which are objects that are non-nuclear. Therefore, we have a better picture of how many nuclear warheads the Russians are actually deploying.

This is the kind of innovation that will help us to begin to reduce and eliminate nuclear warheads, and we need to continue to develop these kinds of tools so that we can grasp this holy grail in the not-too-distant future. It is precisely where the Trump administration wants to go in its efforts to seek limits on nonstrategic nuclear warheads, which are usually held in storage and not operationally deployed on a day-to-day basis. I am convinced we can do it.

The Way Ahead

The need of the hour is to reflect on what the long arc of our experience has taught us in the nuclear arms control arena, and think about better treaties for the future.

First, we have learned how to do verification better over time. The on-site inspection regimes of today could not have been imagined when Nixon and Brezhnev signed the SALT I Treaty in 1972. We are now, as I described above, getting to the point when we can begin to control and limit warheads, because we can imagine how to inspect the process. We need to think through how we would develop new warhead verification regimes.

Second, national technical means (NTM), the satellites and radars that are controlled by governments, have gained in sophistication and coverage over time. That they should not be interfered with during treaty implementation is a well-accepted principle of arms control practice. How NTM should be developed and used in future treaties is now ripe for consideration. We should consider how the new tools on offer, such as the commercial satellite networks, can be fitted into the processes and procedures that we have honed over 50 years of experience with nuclear arms limitation and reduction.

Third, we understand now how to structure treaties to ensure that they actually achieve limitations on and elimination of nuclear weapons systems—missiles, bombers, submarines. Our procedures for conversion or elimination of these systems are well understood; we know what worked and what did not work in the past. Can some of that experience be adjusted to the elimination of nuclear warheads, or do we have to think completely outside the box? Luckily, there has been a wealth of good work at our national laboratories and in the nongovernmental community on this topic.

I am not at all pessimistic about this future, despite the challenges it holds. Certain tensions, such as over the offense-defense relationship, are not going to go away and will have to be dealt with. Likewise, when more countries, first of all China, become invited to the arms control table, the negotiations become more complicated. To begin with, Beijing will have to be convinced that its interests are served by joining in the negotiations. Finally, the debate within the U.S. political system as to whether or not arms control negotiations serve our national security interest will always be a factor.

That calculation, in my view, must be made in every treaty setting and throughout a treaty’s lifetime. When the Russians violated the INF Treaty to the point that it was being hollowed out, it was time for the United States to leave. While New START provides us with predictability about the Russian strategic force structure and prevents Moscow from building up its nuclear weapons, it is clearly in our interest to stay. We must be clear-eyed when nuclear arms control is serving us well, but not shy away from admitting when it fails us.

I will end where I began: Nuclear arms control is the only way that we can attain stable and predictable deployments of these most fearsome weapons, and it is the only way that we can ensure we won’t be bankrupted by nuclear arms racing.

Rose Gottemoeller is the Frank E. and Arthur W. Payne Distinguished Lecturer at the Freeman Spogli Institute for International Studies and Center for Security and Cooperation at Stanford University.

Before joining Stanford, Gottemoeller was the Deputy Secretary General of NATO from 2016 to 2019. Prior to NATO, she served for nearly five years as the under secretary for arms control and international security at the U.S. Department of State. While assistant secretary of State for arms control, verification and compliance in 2009 and 2010, she was the chief U.S. negotiator of the New Strategic Arms Reduction Treaty (New START) with the Russian Federation.

Prior to government service, she was a senior associate with the Carnegie Endowment for International Peace, with joint appointments to the nonproliferation and Russia programs. She served as the director of the Carnegie Moscow Center from 2006 to 2008, and is currently a nonresident fellow in Carnegie’s Nuclear Policy Program. She is also a research fellow at the Hoover Institution.