During the early 1960s the intercontinental ballistic missile (ICBM) introduced the world to the possibility of instant total war. Thirty years later, no nation has yet fired any nuclear missile at a real target. Orthodox history holds that a succession of defensive nuclear doctrines and strategies -- from "massive retaliation" to "mutual assured destruction" -- worked, almost seamlessly, to deter Soviet aggression against the United States and to prevent the use of nuclear weapons.
The possibility of U.S. aggression in nuclear conflict is seldom considered. And why should it be? Virtually nothing in the public record suggests that high U.S. authorities ever contemplated a first strike against the Soviet Union, except in response to a Soviet invasion of Western Europe, or that they doubted the deterrent power of Soviet nuclear forces. The main documented exception was the Air Force Chief of Staff in the early 1960s, Curtis LeMay, a seemingly idiosyncratic case.
But beginning in 1957 the U.S. military did prepare plans for a preemptive nuclear strike against the U.S.S.R., based on our growing lead in land-based missiles. And top military and intelligence leaders presented an assessment of those plans to President John F. Kennedy in July of 1961. At that time, some high Air Force and CIA leaders apparently believed that a window of outright ballistic missile superiority, perhaps sufficient for a successful first strike, would be open in late 1963.
The document reproduced opposite is published here for the first time. It describes a meeting of the National Security Council on July 20, 1961. At that meeting, the document shows, the chairman of the Joint Chiefs of Staff, the director of the CIA, and others presented plans for a surprise attack. They answered some questions from Kennedy about timing and effects, and promised further information. The meeting recessed under a presidential injunction of secrecy that has not been broken until now.
The Real Missile Gap
In 1960, claims of a "missile gap" favoring the Soviets had given the Democrats a critical election theme, and many millions of Americans entered the Sixties feeling intensely vulnerable to the new Soviet ICBM threat. But as Richard Reeves has recently written, intelligence based on satellites launched in August of 1960 soon challenged the campaign assessment and public view. The United States had beaten the USSR to an operational ICBM and enjoyed a clear, and growing, numerical advantage. We were far ahead, and our military planners knew it.
Kennedy was quickly convinced of this truth, which was further confirmed as new satellites brought back new information. Later in 1961, a National Intelligence Estimate came through showing only four Soviet ICBMs in place, all of them on low alert at a test site called Plesetsk. By fall, Defense Undersecretary Roswell Gilpatric would acknowledge in a public speech that U.S. forces (with 185 ICBMs and over 3,400 deliverable nuclear bombs at that time) were vastly superior to those of the Russians.
It was in this context, of an increasing nuclear edge based on a runaway lead in land-based missiles, that Kennedy faced his first nuclear-tinged crisis, which erupted over Berlin in July of 1961.
The Berlin Crisis
The July 20th meeting took place under conditions of unusual tension. Only three months before, Kennedy had suffered the failure of the Bay of Pigs invasion and his loss of confidence in both the CIA and the Joint Chiefs. One month before, he had been shaken by his Vienna confrontation with Nikita Khrushchev. Now, the Soviets were threatening to turn control of access to West Berlin over to the East Germans, and to conclude a separate peace treaty with that satellite state.
At the crucial National Security Council discussion of the brewing Berlin crisis on July 13, Secretary of State Dean Rusk had opposed negotiations with the Soviets until the last moment. As Arthur Schlesinger, Jr., then a special assistant, later summarized for the President, adviser Dean Acheson had prepared a paper arguing that
we are in a fateful test of wills, that our major task is to demonstrate our unalterable determination, and that Krushchev will be deterred only by a US readiness to go to nuclear war rather than to abandon the status quo. On this theory, negotiation is harmful until the crisis is well developed; then it is useful only for propaganda purposes . . . 1
Kennedy favored negotiations over conflict. While not directly challenging Acheson, he encouraged Schlesinger to produce an unsigned memo critical of Acheson's stance.
Schlesinger advised caution. In a passage especially pertinent to the larger issue, he wrote:
The [Acheson] paper hinges on our willingness to face nuclear war. But this option is undefined. Before you are asked to make the decision to go to nuclear war, you are entitled to know concretely what nuclear war is likely to mean. The Pentagon should be required to make an analysis of the possible levels and implications of nuclear warfare and the possible gradations of our own nuclear response.2
It is possible (though we do not know) that the decision to bring the Net Evaluation to Kennedy occurred in response to the raising of these concerns. At any rate, the meeting occurred.
The Burris Memorandum
The memorandum reproduced here was written for Vice President Lyndon Johnson, who did not attend the meeting, by Colonel Howard Burris, his military aide. Declassified only in June of 1993, it has not previously received any public attention so far as we have been able to determine.
The first paragraph introduces General Hickey and his group, the Net Evaluation Subcommittee. Although the subcommittee report is described as "annual," this would be the first one given to President Kennedy and his advisors, and it is not clear whether President Eisenhower received such reports in person. General Lyman Lemnitzer, chairman of the Joint Chiefs, stepped in to explain the "assumption" of the 1961 report: "a surprise attack in late 1963, preceded by a period of heightened tensions." The question arises: A surprise attack by whom on whom?
The following paragraphs answer the question. The second paragraph reports that after hearing the presentations, President Kennedy asked the presenters "if there had ever been made an assessment of damage results to the U.S.S.R. which would be incurred by a preemptive attack." Kennedy also asked for an effectiveness trend since "these studies have been made since 1957." Lemnitzer responded that he would later answer both of the President's questions in private.
Paragraph three records Kennedy asking a hypothetical question: what would happen if we launched a strike in the winter of 1962? Allen Dulles of the CIA responded that "the attack would be much less effective since there would be considerably fewer missiles involved." Lemnitzer then cautioned against putting too much faith in the findings since the assumptions might be faulty. The discussion thus provides a time-frame. December of 1962 was too early for an attack because the U.S. would have too few missiles; by December of 1963 there would likely be sufficient numbers.
Paragraph four reports one more Kennedy question: how much time would "citizens" need to remain in shelters following an attack? The President receives a qualified estimate of two weeks from a member of the subcommittee. The group was clearly talking about U.S. citizens protecting themselves from the globe-encircling fallout following a U.S. nuclear attack on the U.S.S.R.
Paragraph five adds to the intensity of the document with Kennedy's directive "that no member in attendance disclose even the subject of the meeting."
Other Accounts of the Meeting
So far as we know, the official record of this meeting remains secret. The excellent Foreign Relations of the United States, volume XIV, "Berlin Crisis 1961-1962," published in late 1993, though replete with memoranda detailing the nuclear aspects of the Berlin confrontation, makes no mention of it. The only official reference we know of is the agenda for the National Security Council issued on July 18, 1961, declassified in 1977, which reads, simply "The Net Evaluation Subcommittee (NSC 5816; N.S. Action No. 2223) . . . Presentation of the report by the Chairman of the Subcommittee." (The most detailed discussion of the Net Evaluation Subcommittee we have found is in Desmond Ball's Politics and Force Levels, which identifies the larger task of the subcommittee as the preparation of revised targeting plans.)
On the other hand, the fact of a meeting, and Kennedy's personal reaction to it, has been reported. The President was displeased. But no account yet published has told what he was displeased about.
For example, Arthur Schlesinger's Robert Kennedy and His Times gives this account:
. . . Kennedy received the Net Evaluation, an annual doomsday briefing analyzing the chances of nuclear war. An Air Force General presented it, said Roswell Gilpatric, the deputy secretary of defense, "as though it were for a kindergarten class . . . Finally Kennedy got up and walked right out in the middle of it, and that was the end of it. We never had another one."3
McGeorge Bundy evidently refers to the same meeting in this passage:
In the summer of 1961 [Kennedy] went through a formal briefing on the net assessment of a general nuclear war between the two superpowers, and he expressed his own reaction to Dean Rusk as they walked from the cabinet room to the Oval Office for a private meeting on other subjects: "And we call ourselves the human race."4
(Dean Rusk's memoirs repeat Kennedy's remark, though they place the meeting "shortly after our assuming office." Richard Reeves does not mention the July meeting, and attributes Kennedy's remark to a later briefing in September, 1961.)
Numerous other apparent accounts of the meeting exist, though they do not refer to it by name or date. All agree on Kennedy's reaction. But none reveal what was actually discussed. Theodore Sorenson's Kennedy, published only four years later, presents an understandably benign version:
That briefing confirmed, however, the harsh facts [Kennedy] already knew: (1) that neither the Soviet Union nor the United States could 'win' a nuclear war in any rational sense of the word; (2) that, except to deter an all-out Soviet attack, our threat of 'massive retaliation' to every Communist move was no longer credible, now that it invited our own destruction; and (3) that a policy of 'pre-emptive first strike' or 'preventive war' was no longer open to either side, inasmuch as even a surprise missile attack would trigger, before those missiles reached their targets, a devastating retaliation that neither country could risk or accept.5
Unfortunately, the critical third point was not yet true. As former Assistant Secretary of State Roger Hilsman wrote in 1967:
As the intelligence community looked at their estimates in 1958, 1959, and 1960, and even through the first half of 1961, they saw a missile gap developing that would come to a peak about 1963.6
What Hilsman does not say explicitly is that the estimated missile gap was in America's favor. The Soviets had virtually no operational ICBMs in 1961, a fact known to American intelligence at least by the end of 1960. And it appears the Russians did not solve their fundamental technical problem, namely building a hydrogen bomb small enough to be carried by a missile of manageable size, until years later.7
Dean Rusk describes the meeting as an "awesome experience" in his memoirs, As I Saw It, published in 1990.
President Kennedy clearly understood what nuclear war meant and was appalled by it. In our many talks together, he never worried about the threat of assassination, but he occasionally brooded over whether it would be his fate to push the nuclear button... If any of us had doubts, that 1961 briefing convinced us that a nuclear war must never be fought. Consequently, throughout the Kennedy and Johnson years we worked to establish a stable deterrent. . .8
What Rusk does not say is that the problem of a "stable deterrent" in 1961 did not lie in an insufficiency of American missiles. It lay, rather, in the need for the Soviets to develop sufficient effective ICBM (and submarine) forces, to deter us. That is an ugly but unavoidable fact. Rusk goes on, a page later, with comments that appear almost anguished, and for which his own account of the meeting gives no apparent rationale:
. . . the United States has never renounced possible first use of nuclear weapons. I personally think that the United States is committed to a second strike only, after we have received nuclear weapons on our own soil. Under no circumstances would I have participated in an order to launch a first strike, with the possible exception of a massive conventional attack on Western Europe.9
The July 25 Speeck on Berlin
Nuclear conflict was very much in the air that week. Another document of the time indicates the directions Kennedy's nuclear thinking was actually taking -- quite the Cold Warrior, but at the same time far removed from pre-emptive strikes and the inflexible all-out attack envisioned by the Joint Chiefs. This is a paper entitled "Nuclear Strategy in the Berlin Crisis," by the economist Thomas C. Schelling, which was sent to Hyannis Port over the weekend of July 21, 1961 and which, as Bundy noted, made a "deep impression" on the President. In it Schelling presented arguments for a capability, which did not then exist, to wage limited nuclear war:
. . . the role of nuclears in Europe should not be to win a grand nuclear campaign, but to pose a higher level of risk to the enemy. The important thing in limited nuclear war is to impress the Soviet leadership with the risk of general war -- a war that may occur whether we or they intend it or not....We should plan for a war of nerve, of demonstration, and of bargaining, not of tactical target destruction.10
Schelling also advocated centralization of the control of weapons in the hands of the President so as to
permit deliberate, discriminating, selective use for dangerous nuclear bargaining. This means preventing any use, by anyone, not specifically authorized as part of the nuclear bargaining plan...This is a controlled strategic exchange.11
Schelling's paper thus called attention to a key concern: the diffuse character of nuclear command and control in 1961 did not assure that the President in fact enjoyed the full authority over the bomb which most Americans assumed to be the case. Establishing such control became a priority for Kennedy in the months that followed.
The cumulative impact of this diverse advice can be seen in Kennedy's televised address to the nation on July 25, 1961. "We cannot and will not permit the Communists to drive us out of Berlin, either gradually or by force" Yet Kennedy also stressed the dangers: "miscommunication could rain down more devastation in several hours than has been wrought in all the wars of human history." He asked for increased military appropriations and called out 150,000 reserve personnel. But he did not engage the Soviets. The Berlin Wall was allowed to remain intact when constructed in August of 1961, a symbolic column of soldiers was sent through to West Berlin, and a fallout shelter program was undertaken in the United States.
With the Burris memorandum, the reasoning behind the fallout shelter program now begins to fall into place. As a civil defense measure against a Soviet nuclear attack, the flimsy cinderblock shelters Americans were told to build were absurd. But they could indeed protect those in them, for a couple of weeks, from radiation drifting thousands of miles after a U.S. pre-emptive strike on the Soviet Union. It is known that Kennedy later regretted this program.
Down the Road: 1962 and 1963
The U.S. was far ahead in the arms race. Yet the military continued to press for a rapid build-up of strategic missiles. Curtis LeMay had asked for at least 2400 Minutemen; Thomas Powers of the Strategic Air Command had asked for 10,000. All were to be unleashed in a single paroxysm of mass annihilation, know as SIOP, the Single Integrated Operating Plan.
SIOP was a recipe for blowing up the world, whether in a first or a second strike. As McGeorge Bundy wrote to the President on July 7, 1961:
...All agree that the current strategic war plan is dangerously rigid and, if continued without amendment, may leave you with very little choice as to how you face the moment of thermonuclear truth. We believe that you may want to raise this question with Bob McNamara in order to have a prompt review and new orders if necessary. In essence, the current plan calls for shooting off everything we have in one shot, and is so constructed as to make any more flexible course very difficult.12
During that summer of 1961, the Defense Secretary ordered an overhaul of SIOP carried out by RAND analysts (including Daniel Ellsberg) and quickly approved by the JCS. Kennedy and Defense Secretary Robert McNamara eventually imposed a limit of 1,000 Minuteman missiles, angering the Chiefs. Kennedy also launched efforts to gain operational control of the nuclear force, then far from being securely concentrated in the President's hands.
The Burris memorandum may help to explain both the military's drive for a vast U.S. nuclear build-up, despite the fact that America was already far ahead, and the resistance from JFK and McNamara. The Net Evaluation Subcommittee had offered the Pentagon, the CIA, and President Kennedy a glimpse of the opportunity that lay ahead in the winter of 1963: U.S. nuclear superiority so complete that a first strike might be successful. But it also alerted Kennedy to a danger. American nuclear superiority might then be so complete, that rogue elements from the military and intelligence forces, seeking to precipitate an American first strike, might not feel deterred by fear of Soviet retaliation. What was the dispute over the numbers of land-based ICBMs really about? To be sure, at some level it involved the sufficiency of deterrence. But there may also have been an even graver concern: the offensive capabilities of the nuclear force, at a time when the President could not be sure of his control over the nuclear button.
By October of 1962, the U.S. nuclear lead remained strong, though perhaps not yet airtight, given the number of Soviet bombers and the risks to Europe. Twenty years later, Anthony Cordesman described the picture:
During the Cuban missile crisis of October 1962, the US had approximately 1500 B-47s and 500 B-52s, and had already deployed over 200 of its first generation of ICBMs. In marked contrast, the Soviet strategic missile threat consisted of a few token ICBM deployments whose unreliability was so great that it was uncertain exactly whom they threatened. Soviet long range bomber forces consisted only of 100 Tu-Bears and 35 May Bison, whose range and flight characteristics forced them to fly at medium and high altitudes, and which made them extremely vulnerable to US fighters and surface-to-air missiles.13
Kennedy resisted strong pressures to test this advantage in October of 1962, as he might have had to do, had he agreed to launch bombing raids on the Cuban missile installations. Nikita Khrushchev's memoirs, published in 1970, tell of graphic fears expressed by Robert Kennedy to the Russian ambassador, Anatoly Dobrynin at the peak of the crisis:
Even though the President himself is very much against starting a war over Cuba, an irreversible chain of events could occur against his will... If the situation continues for much longer, the President is not sure that the military will not overthrow him and seize power. The American military could get out of control.14
Not even the American editors of Khrushchev's memoirs took these remarks seriously at the time they were first published. A rare editorial note reads: "Obviously this is Khrushchev's own version of what was reported to him. There is no evidence that the President was acting out of fear of a military take-over."
Looking down the road, the Net Evaluation calendar of 1961 implied that the period from Cuba to Dallas and just after was, perhaps, critical to the survival of the world. Had tensions escalated or been aroused in some violent way in late 1963, the President might have faced an excruciating choice -- to strike first, or to give up "victory" during the last brief moment in all history when it could conceivably have been won.
We cannot say whether Kennedy believed the Net Evaluation calendar, or indeed, perhaps equally serious, whether he believed that others in the government might believe it. We do know that the last year of his life saw repeated initiatives to settle conflicts and reduce tensions: the normalization of Berlin, the withdrawal of missiles from Turkey, the no-invasion pledge on Cuba and the effort, only partially effective, to end to the covert campaign (OP/MONGOOSE) against Castro, the test-ban treaty, and -- though the point is disputed -- the order in October 1963 to begin a phased withdrawal from Vietnam. By November of 1963, the potential for "heightened tensions" leading to uncontrollable pressures to strike first had indeed been reduced. And, some time later, the Soviet Rocket Forces did evidently shut the window. From that point, the world probably became a good deal more secure. But exactly when this happened is not clear.
And Lyndon Baines Johnson, the recipient of Burris's note, was still uneasy on the point when he assumed office on November 22, 1963, amid swirling rumors connecting Lee Harvey Oswald, falsely as we now know, to the KGB. David Wise, then bureau chief of the New York Herald Tribune, reports hearing Johnson tell in late 1963 of recruiting Earl Warren to head the Warren Commission in the following terms:
. . . when Warren came to the White House, [LBJ] told the Chief Justice he knew he had been a first lieutenant in World War I, and he knew Earl Warren would walk across the Atlantic Ocean to save the lives of three Americans, and possibly a hundred million lives were at stake here . . . 15 15. Wise, p. 292.
Whose lives, exactly?
One meeting, even in the White House, does not establish that first-strike was in fact the nuclear policy of the United States. Kennedy's recorded response, moreover, indicates his personal determination, shared by his civilian advisers, that it never become so. But we do know, from Howard Burris's notes, that a first strike plan had authors close to the decision center. How close, in the end, did they get? Civilian control of nuclear forces was no sure thing in 1961. Was it secure when the window opened, if it did, in 1963? Kennedy's actions and Johnson's eerie remark are consistent with the possibility that the calendar and risks of a first-strike window remained in the minds of both men as late as November, 1963 and possibly in Johnson's mind for a good deal longer.
In any event, the fact that first-strike planning got as far as it did raises grave questions about the history of the Cold War. Much more needs to be known: about nuclear decision-making under Eisenhower and Nixon, about the events of late 1963, about later technical developments such as MIRV and Star Wars. Surely it is now time to declassify all records on this and related history.
1. Foreign Relations, XIV, 173. Complete sources
are given at the end of this article.
2. Foreign Relations, XIV, p. 173.
3. Schlesinger, p. 483.
4.Bundy, p. 354.
5. Sorenson, p. 513.
6. Hilsman, p. 162.
7. See Sorenson, p. 524; Bobbitt, p. 61.
8. Rusk, p. 246-7.
9. Rusk, p. 248.
10. Foreign Relations, XIV, p. 170.
11. Foreign Relations, XIV, p. 172.
12. Quoted in Kaplan, p. 297.
13. Cordesman, p. 7, cited in Bobbitt.
14. Khrushchev p. 497.
Desmond Ball, Politics and Force Levels: The Strategic Missile Program of the Kennedy Administration, University of California Press, 1980.
John Baylis and John Garnett, eds. The Makers of Nuclear Strategy, St. Martin's Press, 1991.
Philip Bobbitt, Democracy and Deterrence, St. Martin's Press, 1988.
Philip Bobbitt, Lawrence Freedman and Gregory Treverton, U.S. Nuclear Strategy, New York University Press, 1989.
McGeorge Bundy, Danger and Survival: Choices About the Bomb in the First Fifty Years, Random House, 1988.
Grant Burns, The Nuclear Present, Scarecrow Press, 1992.
Anthony Cordesman, Deterrence in the '80s, International Institute for Strategic Studies, 1982.
Foreign Relations of the United States 1961-1963, Vol XIV, Berlin Crisis 1961-1962, United States Department of State, 1993.
Lawrence Freedman, U.S. Intelligence and the Soviet Strategic Threat, Princeton University Press, 1986.
David Goldfisher, The Best Defense: Policy Alternatives for Nuclear Security from the 1950's to the 1990's, Cornell University Press, 1993.
Fred M. Kaplan, The Wizards of Armageddon, Stanford University Press, 1991.
Roger Hilsman, To Move a Nation: The Politics of Foreign Policy in the Administration of John F. Kennedy, Doubleday, 1967.
Lyndon B. Johnson, The Vantage Point: Perspectives of the Presidency 1963-1969, Holt, Rinehart and Winston 1971.
Nikita S. Khrushchev, Khrushchev Remembers, Little Brown, 1970.
John M. Newman, JFK and Vietnam, Warner Books, 1992.
Richard Reeves, President Kennedy, Simon & Schuster, 1993.
Dean Rusk, As I Saw It, WW Norton, 1990.
Arthur Schlesinger, Jr. Robert Kennedy and His Times, Ballantine, 1978.
David Schwartzman, Games of Chicken: Four Decades of U.S. Nuclear Policy, Praeger, 1988.
Theodore C. Sorenson, Kennedy, Harper and Row, 1965.
David Wise, The Politics of Lying, Random House, 1973.