By Robert S. McNamara (Long Road to Reykjavik)

In the mid-1960s we had irrefutable evidence that the Soviets were deploying an antiballistic-missile system around Moscow–a system to defend their capital against our long-range missiles. We made the reasonable–but perhaps incorrect–assumption that they would deploy the system across the entire Soviet Union. Why would anyone put a system around one city and nowhere else? Were a nationwide Soviet ABM system to be put in place, it would require that we make major changes in our force levels.

The Congress believed that the proper response to a full-fledged Soviet antiballistic-missile network was for the U.S. to deploy its own countrywide ABM system. The Army had been working on such systems since the late 1950s, first the Nike-Zeus and later the Nike-X. In 1966, therefore, the Congress authorized and appropriated $167.9 million for production of a Nike system (when fully deployed, the weapons would probably have cost a total of $30 billion). President Johnson and I believed the system would provide little if any protection either to our population or our weapons. We refused to spend the funds that Congress had appropriated.

On Dec. 6, 1966, Deputy Secretary of Defense Cyrus Vance, the Joint Chiefs of Staff and I went to Austin to meet with the President and Walt Rostow, special assistant to the President for national security affairs. Our purpose was to review with the President the defense budget for the fiscal year 1968, which was to be presented to the Congress in February 1967. Among the items to be considered was the recommendation of the Chiefs that the budget request include funds for production of an antiballistic-missile system. I explained to the President that the Chiefs had recommended the action, but that Cy and I strongly opposed it.

The President called on each of the five Chiefs in turn, and each one of them urged approval of the ABM program. Walt Rostow sided with the Chiefs. This was an extraordinarily difficult moment for President Johnson. I never hesitated to disagree with a unanimous recommendation of the Joint Chiefs if I felt it was the wrong decision. In this case, however, Congress had already passed a law authorizing production of the ABM system. To continue to refuse to proceed in the direction that had been supported by the Congress, and to do so in the face of a unanimous recommendation by the Chiefs, put the President in an almost untenable position.

At that point I said to the President, “The Chiefs’ recommendation is wrong; it’s absolutely wrong. The proper response to a Soviet ABM system is not the deployment of an admittedly ‘leaky’ U.S. defense. The proper response is action that will ensure that we maintain our deterrent capability in the face of the Soviet defense. What the Chiefs are recommending has nothing to do with maintaining that deterrent. If our deterrent force–our offensive missiles and bombers–was of the proper size before the Soviets deployed their defenses, it must now be expanded to ensure that the same number of weapons will land on Soviet targets, after taking account of the attrition the U.S. missile force will suffer as it passes through the Soviet defenses. So for the U.S. to deploy an ABM defense is the wrong response to the Soviet action. But since we are in this bind, why don’t we do this: put a small amount of money in the budget for ABM procurement, but state in the budget, and in my written report to the Congress, that none of those funds will be spent and no decision will be made to deploy an ABM system until after we make every possible effort to negotiate an agreement with the Soviets, which will prohibit deployment of defenses by either side and will limit offensive forces as well.”

The President seized on this proposal as a way out of a very difficult position.

I informed Dean Rusk, the Secretary of State, of the President’s decision. He immediately approached the Soviets, seeking to initiate negotiation of an ABM treaty. They refused to participate even in preliminary discussions of such an agreement.

In June 1967 the Soviet Premier, Aleksei Kosygin, came to New York City to visit the United Nations. After some difficulty, it was arranged for the Premier and President Johnson to meet on June 23 at Glassboro, NJ.–Glassboro is halfway between New York and Washington–to discuss the question of ABM deployment. At lunch in New Jersey on that June day, the President, the Premier and a group of their associates were sitting around a small oval table. It was clear the President was becoming frustrated by Kosygin’s failure to see the U.S. point of view on ABM defenses. Finally, the President turned to me and asked me to explain our position.

I said, “Mr. Prime Minister, you must understand that the proper U.S. response to your Soviet ABM system is an expansion of our offensive force. If we had the right number of offensive weapons to maintain a deterrent before you put your defenses in, then, to maintain the same degree of deterrence in the face of your defense, we must strengthen our offense. Deployment of a Soviet ABM system will lead to an escalation of the arms race. That’s not good for either one of us.”

Kosygin was furious. The blood rushed to his face, he pounded on the table, and he said, “Defense is moral; offense is immoral!” That was essentially the end of the discussion. The Soviet Union was by no means ready at that time to discuss an agreement banning defensive systems.

Following our return to Washington, there was unanimous agreement among the Chiefs, the President and me that we must initiate action to expand our offensive forces. The cheapest way to do that was to develop MIRVs. By placing more than one warhead on each missile, the U.S. could increase the number of warheads far more cheaply than by building more missiles. But we recognized this was a very dangerous step–if the Soviets followed our lead, as we must assume they would, it would lead to a dramatic increase in the offensive forces of each side. We therefore concluded that we would proceed with the development of MIRVs, but we would make no decision to deploy them until we had explored fully the possibilities of negotiating an agreement to prohibit defenses. If such a treaty was negotiated, the MIRV program would be scrapped.

When I left office in February 1968, the Soviets were still moving ahead with the ABM system, and our MIRV program was acquiring a strong constituency in this country. What followed is a matter of record. The U.S. and the U.S.S.R. did agree on the ABM treaty in 1972, but it was decided, nonetheless, to proceed with deployment of MIRVs. The number of U.S. missile warheads increased from 1,800 in 1970 to 6,100 in 1975; during this period, the Soviets, who were behind us in the development of MIRV technology, expanded their warheads by only 900, from 1,600 in 1970 to 2,500 in 1975.

The explosive growth in the number of Soviet missile warheads–which is pointed to by those who say, “We build, they build; we stop, they build”–occurred between 1975 and 1980. During that period the Soviets more than doubled the number of their missile warheads, from 2,500 to 5,500. It was in part a delayed response to our MIRV decision of five years before. And it still left us in 1980 with a substantial numerical superiority: 7,300 to 5,500.

Since 1980 the Soviet buildup has continued. Today the number of warheads (missiles plus bombs) in the U.S. and Soviet strategic forces is approximately equal. For 35 years, from 1950 to 1985, the Soviets lived with numerical inferiority, although “parity” in operational terms has existed since at least October 1962. (During a recent visit to the Soviet Union, I was asked by several political and scientific leaders to define nuclear parity. I replied that parity exists when each side is deterred from initiating a strategic strike by the recognition that such an attack would be followed by a retaliatory strike that would inflict unacceptable damage on the attacker. I went on to say: “I will surprise you by stating that I believe parity existed in October 1962, at the time of the Cuban missile crisis. The U.S. then had approximately 5,000 strategic warheads, compared with the Soviets’ 300. Despite an advantage of 17 to 1 in our favor, President Kennedy and I were deterred from even considering a nuclear attack on the U.S.S.R. by the knowledge that although such a strike would destroy the Soviet Union, tens of their weapons would survive to be launched against the U.S. These would kill millions of Americans. No responsible political leader would expose his nation to such a catastrophe.”)

It is essential to understand the action-reaction dynamic and to take it into account in formulating arms-control and defense policies. We must understand that every action stimulates a reaction in an endless cycle. Already, the cost of our failure to do so has been the development of ridiculously large arsenals and missed opportunities to negotiate agreements to reduce them.