The Missiles of November, December, January, February The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

Coleman, David G.
Journal of Cold War Studies, Volume 9, Number 3, Summer 2007, pp. 5-48 (Article)
Published by The MIT Press

For additional information about this article

http://muse.jhu.edu/journals/cws/summary/v009/9.3coleman.html

Access Provided by Northside College Preparatory High School at 02/10/11 5:44PM GMT

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement Coleman

The Missiles of November, December, January, February . . . The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement
David G. Coleman

n 17 November 1962 the director of the U.S. Central Intelligence Agency (CIA), John McCone, met in a secure room at Dulles Airport with President John F. Kennedy and former president Dwight D. Eisenhower and warned them that the United States still had a missiles in Cuba problem. McCones statement came less than three weeks after the leaders of the Communist Party of the Soviet Union (CPSU) had backed down in the Cuban missile crisis and agreed to remove newly installed Soviet medium-range ballistic missiles (MRBMs) from Cuba and ship them back to the USSRa settlement laid out in an exchange of letters between Kennedy and the Soviet leader Nikita Khrushchev. But McCone told Kennedy and Eisenhower that at least four kinds of Soviet short-range missiles remained in Cuba, including hundreds of surface-to-surface, surface-to-air, and air-to-surface missiles.1 A number of these were so-called dual-use missiles, capable of delivering both conventional and nuclear warheads. Although they posed no direct threat to the continental United States and could not reach any American cities, they were able to strike the U.S. naval base at Guantnamo and U.S. ships close to the coast of Cuba. Moreover, if Kennedy at some point ordered an invasion of the islandan option that was still on the tableU.S. troops might well be
1. John A. McCone, Memorandum of Conversation between President Kennedy and Former President Eisenhower, 17 November 1962, in U.S. Department of State, Foreign Relations of the United States, 19611963, Vol. XI, p. 479 (hereinafter referred to as FRUS, with appropriate year and volume numbers). Submarine-launched missiles were also included in the Anadyr buildup. See William Burr and Thomas Blanton, eds., The Submarines of October: U.S. and Soviet Naval Encounters during the Cuban Missile Crisis, in National Security Archive Electronic Brieng Book No. 75, 31 October 2002; and Peter A. Huchthausen, October Fury (Hoboken, NJ: Wiley, 2002).

Journal of Cold War Studies Vol. 9, No. 3, Summer 2007, pp. 548 2007 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology

Coleman

entering a nuclear battleeld, which in turn could escalate to global nuclear war. The missiles and remaining troops also posed a political problem. The continued presence of these military forces allowed the administrations critics to accuse the White House of turning a blind eye to the establishment of a Soviet base on Americas doorstep.2 This article examines how the Kennedy administration assessed the risk posed by the Soviet short-range missiles in Cuba and the associated combat troops, particularly during the post-crisis settlement period. The issue had a strong domestic political subtext that played out long after the famous fortnight of late October ended. Missiles in Cuba had been a topic of discussion well before the dramatic events of October 1962the intelligence le for 1959 reporting claims by refugees that Soviet missiles were present in Cuba was reportedly four inches thick3and it dragged on well past the famous thirteen days. Many studies assume a nal resolution to the crisis that did not actually exist either in October 1962 or afterward.4 Khrushchevs dramatic retreat on 28 October alleviated the sense of imminent peril but left many critical issues unresolved. The November crisis that ensued played out differently from the scenario envisaged by Khrushchev.5 Rather than negotiating from a position of strength to compel the United States to accept Soviet demands on a number of key issuesthe Berlin problem, nuclear disarmament, and dtenteKhrushchev tried desperately to salvage something from his aborted venture in the Caribbean. In the process, the Soviet Union became
2. The phrase comes from Senator Kenneth Keatings statement on the oor of the U.S. Senate, 31 August 1962, transcribed in U.S. Congress, Congressional Record, 87th Cong., 2nd Sess., 31 August 1962, p. 18361. 3. John Prados, The Soviet Estimate: U.S. Intelligence Analysis and Russian Military Strength (New York: Dial, 1982), p. 133. 4. For some of the literature dealing with the period before and after the thirteen days, see Stephen G. Rabe, After the Missiles of October: John F. Kennedy and Cuba, November 1962 to November 1963, Presidential Studies Quarterly, Vol. 30, No. 4 (December 2000), pp. 714726; James G. Blight and Philip Brenner, Sad and Luminous Days: Cubas Struggle with the Superpowers after the Missile Crisis (Lanham, MD: Rowman & Littleeld, 2002); Philip Brenner, Thirteen Months: Cubas Perspective on the Missile Crisis, in James A. Nathan, ed., The Cuban Missile Crisis Revisited (New York: St. Martins Press, 1992), pp. 187218; Raymond L. Garthoff, Reections on the Cuban Missile Crisis, rev. ed. (Washington, DC: Brookings Institution, 1989); Aleksandr Fursenko and Timothy Naftali, One Hell of a Gamble: Khrushchev, Castro, and Kennedy, 19581964 (New York: W. W. Norton, 1997); Lawrence Freedman, Kennedys Wars: Berlin, Cuba, Laos, and Vietnam (New York: Oxford University Press, 2000), pp. 225237; James G. Hershberg, The United States, Brazil, and the Cuban Missile Crisis, 1962 (Part 2), Journal of Cold War Studies, Vol. 6, No. 3 (Summer 2004), pp. 567; Barton J. Bernstein, Bombers, Inspection and the No Invasion Pledge, Foreign Service Journal, Vol. 56, No. 7 (July 1979), pp. 812; James G. Hershberg, Before The Missiles of October: Did Kennedy Plan a Military Strike against Cuba? Diplomatic History, Vol. 14, No. 2 (Spring 1990), pp. 163198; and Jeremy Pressman, September Statements, October Missiles, November Elections: Domestic Politics, Foreign Policy Making, and the Cuban Missile Crisis, Security Studies, Vol. 10, No. 3 (Spring 2001), pp. 80 114. 5. Aleksandr Fursenko and Timothy Naftali, Khrushchevs Cold War: The Inside Story of an American Adversary (New York: W. W. Norton, 2006), esp. pp. 440441.

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

embroiled in diplomatic wrangling with the United States, Cuba, and the United Nations over demands for the entry of weapons inspectors into Cuba, the withdrawal of long-range nuclear bombers, and the formal codication of Kennedys pledge not to invade the island. Much to Washingtons frustration, Fidel Castro remained in power and refused to allow international weapons inspectors into Cuba to verify that the MRBMs and bombers had left and were not being reintroduced. Thousands of Soviet combat troops and technicians remained in Cuba, as did much of the sophisticated weaponry they had brought with them. The U.S. Joint Chiefs of Staff (JCS) continued to prepare to invade Cuba if the diplomatic settlement fell apart. Military readiness for such an operation peaked on 15 November.6 Even the lifting of the U.S. naval blockade of Cuba on 20 November was not the end of the crisis. Much of this unnished business, as Kennedy called it in February 1963, was never fully resolved. Cuba remained a preoccupation within the U.S. intelligence community and among U.S. policymakers, politicians, and journalists well into 1963. Secretary of State Dean Rusk warned the Senate Foreign Relations Committee in January 1963 that the Cuban crisis is in no sense over. It continues to be a highly dangerous situation and could are up in a number of contingencies.7 On another occasion secretly recorded by Kennedys White House recording system, Rusk wondered aloud whether the Soviet Union and Cuba, by promising to withdraw the MRBMs, might have orchestrated a gigantic hoax of which history has had no parallel.8 CIA Director John McCone, a Republican, issued similar warnings directly to lawmakers on Capitol Hill, fueling an already highly charged political debate. In short, the Cuban missile crisis did not simply evaporate on 28 October. As McCones warning to Kennedy and Eisenhower in mid-November implied, even the most fundamental of issuesthe presence of Soviet missiles in Cubacontinued to dog the Kennedy administration for months. In addition to persistent rumors that the Soviet Union was still hiding MRBMs on the island, several thousand Soviet troops armed with sophisticated weaponry remained in Cuba. Their weaponry included several kinds of short-range missiles, some of which were notable primarily for their nuclear capability. The most contentious of these were the so-called Luna missiles (known as Frogs in
6. Department of Defense Operations during the Cuban Crisis, n.d., in Cuba, Subjects, Cuban Testimony, Defense and Military Responses.Box 55, National Security File (NSF), John F. Kennedy Library (JFKL). 7. U.S. Congress, Senate, Committee on Foreign Relations, Executive Sessions of the Senate Foreign Relations Committee (Historical Series), Vol. 15, 88th Cong., 1st Sess., 1963 (Washington, D.C.: U.S. Government Printing Ofce, 1987), p. 15 (hereinafter referred to as SFRC, ESSFRC, 1963). 8. 16th Meeting of the ExComm, 1 November 1962, in David Coleman and Timothy Naftali, eds., The Presidential Recordings: John F. Kennedy, Vol. 4 (New York: W. W. Norton, forthcoming). The audio recording is on Tape 47, Presidential Recordings Collection, Presidents Ofce Files (POF), JFKL.

Coleman

the West), which were short-range, battleeld, surface-to-surface weapons similar to the American-made Honest Johns. The Kennedy administrations handling of the longer-term aspects of the Cuban missile crisis settlement, both internationally and domestically, reveals much about its policies toward Cuba and the Soviet Union and about its assessment of security risks. An analysis of this matter sheds new light on the Cuban missile crisis not only by correcting misperceptions in some recent commentaries but also by revealing that the much publicized aspect of Soviet tactical nuclear weapons in Cuba was, for the Kennedy administration, both a known risk for some of the acute phase of the crisis and an acceptable risk during the settlement period. The analysis provided here has the further benet of elucidating how policy was made in the Kennedy administration. Scholars have long disagreed about the inuence of domestic politics before and during the Cuban missile crisis. Some have argued that a rebreak of sorts existed between international policy and domestic politicsthat Kennedys Cuba policy was made solely on the basis of what was best for the United States, without any regard to domestic politics. Revisionists have contended that domestic politics offered a negative, or at least cynical, motivation for decisions, pushing Kennedy to favor unnecessary confrontation over diplomacy.9 The bulk of these studies have focused on the thirteen days of peak crisis. Extending the discussion to the postcrisis period recasts the debate. The evidence from this period indicates that domestic political considerations were a fundamental factor in Kennedys decision-making and apparently induced him to take a slightly harder line in the post-crisis negotiations with the Soviet Union than he otherwise might have. But the evidence also suggests that Kennedy was more willing than some of his advisers and many Congressional critics to accept a degree of permanent military risk in Cuba.

The Making of a Controversy

The second phase of Operation Anadyrthe codename for the Soviet military buildup in Cuba in the summer and fall of 1962added tactical nuclear missiles and other sophisticated weapons to the Soviet forces stationed in Cuba in the fall of 1962. The disclosure that tactical nuclear arms were present in Cuba has been described as probably the single most important revelation about the crisis since the new sources began to emerge after the end of
9. See Richard Ned Lebow, Domestic Politics and the Cuban Missile Crisis: The Traditional and Revisionist Interpretations Reevaluated, Diplomatic History, Vol. 14, No. 4 (Fall 1990), pp. 471492.

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

the Cold War.10 Robert McNamara has described the potential use of the Luna missiles in Cuba as the most dangerous element of the entire episode.11 To date, the story of those weapons in Cuba has been derived mostly from Russian sources.12 Much of the existing literature reects uncertainty about when the Kennedy administration learned about these forces, what it knew, andmost importantwhat it decided to do about them. The prevailing view has been that the tactical nuclear missiles were, for the Kennedy administration, an unknown risk of the crisis because U.S. policymakers were seriously considering invasion plans of Cuba ignorant of the risk that they would be confronted with a nuclear battleeld.13 Newly available materials suggest that this prevailing view requires some elaboration and revision. The controversy about tactical nuclear missiles in Cuba was sparked at an important international history conference in Havana in January 1992 marking the thirtieth anniversary of the crisis. That conference brought together a number of former ofcials from the United States and the former Soviet Union and leading gures in Cuba, as well as some scholars. At that conference, General Anatoli Gribkov, who had been one of the Soviet military ofcers responsible for coordinating Soviet deployments in Cuba in 1962, startled the audience when he seemed to claim that Khrushchev had predelegated authority to the local Soviet commander in Cuba, General Issa Pliev, to use the dozens of tactical nuclear weapons he had on hand in the event of a U.S. invasion of the island.14 Most observers at the conference were
10. James G. Hershberg, The Crisis Years, 19581963, in Odd Arne Westad, ed., Reviewing the Cold War: Approaches, Interpretations, Theory (London: Frank Cass, 2000), p. 311. 11. James G. Blight, Bruce J. Allyn, and David A. Welch, Cuba on the Brink: Castro, the Missile Crisis, and the Soviet Collapse (New York: Pantheon, 1993), p. 249. 12. Svetlana Savranskaya, Tactical Nuclear Weapons in Cuba: New Evidence, Cold War International History Project Bulletin, Issue No. 14/15 (Winter 2003Spring 2004), pp. 385398; Mark Kramer, Tactical Nuclear Weapons, Soviet Command Authority, and the Cuban Missile Crisis, Cold War International History Project Bulletin, Issue No. 3 (Fall 1993), pp. 4041; Mark Kramer, The Cuban Missile Crisis and Nuclear Proliferation, Security Studies, Vol. 5, No. 1 (Autumn 1995), pp. 171 179; Raymond Garthoff, New Evidence on the Cuban Missile Crisis: Khrushchev, Nuclear Weapons, and the Cuban Missile Crisis, Cold War International History Project Bulletin, Issue No. 11 (Winter 1998), pp. 251254; and Aleksandr Fursenko and Timothy Naftali, The Pitsunda Decision: Khrushchev and Nuclear Weapons, Cold War International History Project Bulletin, Issue No. 10 (March 1998), pp. 223225. 13. Savranskaya, Tactical Nuclear Weapons in Cuba, p. 385. See also Max Frankel, High Noon in the Cold War: Kennedy, Khrushchev, and the Cuban Missile Crisis (New York: Ballantine Books, 2004); and James G. Hershberg, The United States, Brazil, and the Cuban Missile Crisis, 1962 (Part 2), Journal of Cold War Studies, Vol. 6, No. 3 (Summer 2004), p. 11. For exceptions, see Michael C. Desch, That Deep Mud in Cuba: The Strategic Threat and U.S. Planning for a Conventional Response during the Cuban Missile Crisis, Security Studies, Vol. 1, No. 2 (Winter 1991), pp. 317351; and Freedman, Kennedys Wars, pp. 175176. 14. John Newhouse, A Reporter at Large (Cuba), The New Yorker, 27 April 1992, pp. 5257; Arthur M. Schlesinger, Jr., Four Days with Fidel: A Havana Diary, The New York Review of Books, Vol. 39, No. 6 (26 March 1992), pp. 710; Raymond L. Garthoff, The Havana Conference on the Cuban

Coleman

reportedly shocked and alarmed to learn that Khrushchev not only had sent tactical nuclear weapons to Cuba but had also shared his grip on the nuclear trigger. News organizations jumped on the revelations. In ABCs primetime news broadcast, the anchorman Peter Jennings broke the news to millions of viewers that battleeld missiles armed with nuclear warheads . . . were deployed in coastal areas of Cuba during the missile crisis and that Soviet commanders were ready to use them against U.S. troops.15 Vigorously challenged by historians, the predelegation claim was later modied and corrected. [F]ar from having unlimited authority to use tactical nuclear missiles as they saw t, Mark Kramer wrote in correcting the error of the predelegation claim, Soviet commanders were in fact categorically forbidden to use such weapons under any circumstances without explicit orders from Moscow.16 Subsequent work by Kramer, Aleksandr Fursenko, and Timothy Naftali further claried the issue: although Pliev had the technical capability to launch tactical nuclear weapons if American forces invaded Cuba and may well have been tempted to exercise that capability if backed into a use it or lose it corner, the document authorizing him to do so remained unsigned, undelivered, and locked securely in a vault in Moscow.17 Responding to Gribkovs original claim about the delegation of command authority, Robert McNamara, secretary of defense in the Kennedy and Johnson administrations, stated that the Kennedy administration did not know that tactical nuclear weapons were in Cuba. He claimed that if the administration had known about those weapons, such knowledge might have inuenced the crisis in important ways: He speculated that the Executive Committee of the National Security Council (ExComm) might have given different advice to the president and that the president might have made different decisions. We never in the world believed he had tactical nuclear weapons, McNamara said on one occasion a few years later.18 Raymond
Missile Crisis, Cold War International History Project Bulletin, Vol. 1 (Spring 1992), pp. 24; Blight, Allyn, and Welch, Cuba on the Brink; and Anatoli I. Gribkov and William Y. Smith, Operation Anadyr: U.S. and Soviet Generals Recount the Cuban Missile Crisis (Chicago: Edition Q, 1994). 15. Kramer, Tactical Nuclear Weapons, pp. 4041. See also Hershberg, The Crisis Years, 1958 1963, pp. 311314; Garthoff, The Havana Conference on the Cuban Missile Crisis; Schlesinger, Four Days with Fidel; as well as the exchange between Schlesinger and Kramer in The New York Review of Books, Vol. 39, No. 10 (28 May 1992), pp. 6061; Kramer, The Cuban Missile Crisis and Nuclear Proliferation, pp. 171179; and James G. Blight and David A. Welch, The Cuban Missile Crisis: On Historical Judgment and Inference: A Reply to Mark Kramer, Security Studies, Vol. 5, No. 4 (Summer 1996), pp. 172182. 16. Kramer, Tactical Nuclear Weapons, p. 40. 17. Ibid.; Kramer, The Cuban Missile Crisis and Nuclear Proliferation, pp. 171179; Fursenko and Naftali, One Hell of a Gamble, p. 243; and Fursenko and Naftali, Khrushchevs Cold War, p. 473. 18. Transcript, 13 Days: An Insiders Perspective, 1 October 2002, in JFKL. See also, McNamaras introductory comments to the October 2002 conference in Havana, Forty Years after 13 Days, Arms Control Today, Vol. 32, No. 9 (November 2002), pp. 1115; Robert McNamara with Brian Van De-

10

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

Garthoff, a State Department ofcial in 1962, acknowledged that the United States knew about some of the nuclear-capable battleeld missiles in Cuba, but he argued that no one had seriously believed that the weapons might be armed with nuclear warheads. Hence, in his view, the presence of tactical nuclear weapons in Cuba at the time of the crisis was completely unexpected.19 On another occasion, Garthoff argued that [h]ad it been known that there were about one hundred tactical nuclear weapons in Cuba uncertainties over whether they all had later been removed would have seriously plagued the settlement of the crisis, since it would have been very difcult if not impossible to verify that none remained.20 Also commenting before Gribkovs predelegation claims were claried and amended, George Ball, who was under secretary of state in the Kennedy and Johnson administrations, wrote that none of us on the so-called ExComm (the group that advised President Kennedy during the crisis) knew that . . . the Soviets had also sent a number of short-range missiles to support their own military contingents in Cuba.21 More recently, the late Arthur M. Schlesinger, Jr., a former adviser to President Kennedy, praised Kennedys restraint during the crisis because only decades later did we discover that the Soviet forces in Cuba had tactical nuclear weapons and orders to use them to repel a U.S. invasion.22 Those statements are problematic. Some old and much new evidence suggests that the 1992 revelations about Soviet tactical nuclear weapons in Cuba should not have stirred as much surprise as they did. In a book published in 1967, Roger Hilsman, who had been director of the State Departments Bureau of Intelligence and Research in 1962, wrote that the second phase of the Soviet military buildup in Cuba in the spring and summer of 1962 included four battle groups of special ground troops armed with tactical nuclear weapons.23 Hilsman claried three decades later that no concrete
Mark, In Retrospect: The Tragedy and Lessons of Vietnam (New York: Times, 1995), pp. 340341; and Blight, Allyn, and Welch, Cuba on the Brink, pp. 249, 352. See also Robert Weisbrot, Maximum Danger: Kennedy, the Missiles, and the Crisis of American Condence (Chicago: Ivan R. Dee, 2001), pp. 256 n. 40. 19. Garthoff, The Havana Conference on the Cuban Missile Crisis, p. 2; and Raymond L. Garthoff, U.S. Intelligence in the Cuban Missile Crisis, in James G. Blight and David A. Welch, eds., Intelligence and the Cuban Missile Crisis (London: Frank Cass, 1998), pp. 1863. See also General William Y. Smiths comments quoted in Blight, Allyn, and Welch, Cuba on the Brink, p. 261. 20. Garthoff, U.S. Intelligence in the Cuban Missile Crisis, pp. 29, 31. 21. George Ball, Present after the Creation, The New York Review of Books, Vol. 39, No. 21 (17 December 1992), p. 7. 22. Arthur M. Schlesinger, Jr., Bushs Thousand Days, The Washington Post, 29 April 2006, p. A17. Apparently, Schlesinger was unaware that his characterization here was outdated and did not take account of the important clarications and corrections that had followed Gribkovs initial statement. 23. Roger Hilsman, To Move a Nation: The Politics of Foreign Policy in the Administration of John F. Kennedy (New York: Doubleday, 1967), p. 159. See also, Hershberg, The Crisis Years, 19581963, pp. 311312.

11

Coleman

evidence had been available at the time that nuclear warheads were in fact located in Cuba.24 In a book transcribing the recordings President Kennedy had made secretly in the White House during the missile crisis, Ernest May and Philip Zelikow noted that a recording of a meeting on 26 October 1962 showed that Kennedy and his senior advisers had been briefed on the presence of Luna missile launchers.25 Other evidence, such as declassied documents, secret White House recordings, congressional testimony, newspaper reports, and news conferences, indicate that the issue subsequently was discussed on several occasions. The evidence now available does not show that U.S. ofcials conclusively knew that nuclear warheads were already present in Cuba, but it does make clear that they explicitly assumed that such warheads had been sent there. In any case, the striking thing that emerges from this evidence is the extent to which the administration discussed the Luna missiles in late 1962 and early 1963 both privately and publicly. In the months after the peak of the crisis, policymakers and military leaders debated whether it was an acceptable risk for those weapons and Soviet troops to stay in Cuba, even as the U.S. government was insisting that some other types of dual-use weapons be removed. Intelligence ofcials consistently warned that they could not provide denitive proof that nuclear warheads of any kind were present inor, more signicantly, not present inCuba. In the absence of hard evidence, intelligence analysts urged policymakers to assume that nuclear warheads were present in Cuba, a warning that many took to heart. Rather than becoming a sticking point in the settlement negotiations, the presence of Frog missiles and Soviet combat troops was tolerated by the Kennedy administration, which itself did not want to pay the price of formalizing its no-invasion pledge. In short, the presence of tactical nuclear missiles in Cuba was a known risk during the peak of the crisis and was regarded as an acceptable risk during the post-crisis settlement period.

Operation Anadyr and the Short-Range Missiles

When Khrushchev authorized Operation Anadyr on 24 May 1962, it led to a marked acceleration of the buildup of Soviet military forces in Cuba and
24. Roger Hilsman, The Cuban Missile Crisis: The Struggle over Policy (New York: Praeger, 1996), 116 117; and Hershberg, The Crisis Years, 19581963, pp. 311312. 25. Ernest R. May and Philip D. Zelikow, eds., The Kennedy Tapes: Inside the White House during the Cuban Missile Crisis (Cambridge, MA: Harvard University Press, 1997), p. 475. See also Ernest May and Philip Zelikow, eds., The Presidential Recordings: John F. Kennedy, 3 Vols., (New York: W. W. Norton, 2001), Vol. 3, p. 327; Graham Allison and Philip Zelikow, Essence of Decision: Explaining the Cuban Missile Crisis, 2nd ed. (New York: Longman, 1999); and Ernest R. May and Philip D. Zelikow,

12

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

added new categories of weapons to the deployment that went signicantly beyond earlier military supplies to other Third World allies such as Iraq, Indonesia, Egypt, and Syria.26 In addition to the MRBMs that sparked so much controversy, Anadyr ultimately included at least three types of short-range missiles, the most contentious of which were the surface-to-surface missiles. The weapons singled out at the time by policymakers and later by historians as the most controversial were the nuclear-capable missiles known to Warsaw Pact ofcials as Luna and to U.S. ofcials as Frogs (Free Rocket Over Ground). The Frogs were unguided tactical weapons designed for battleeld use. Depending on the model, they were capable of delivering a nuclear warhead with a yield of two to twenty kilotons over a distance of up to 40 kilometers. With launchers mounted on a tracked amphibious tank chassis, the Frogs were highly mobile nuclear artillery, and their solid-propellant rocket motors made them particularly suited to long-term, remote deployments.27 Lunas were considered unremarkable as conventional weapons because they were inaccurate, but when armed with a two-kiloton nuclear warhead they became potent battleeld weapons. During the Cold War, the Soviet Union transferred different types of Lunas to a number of friends and allies, including Algeria, Egypt, Iraq, Kuwait, Libya, North Korea, Syria, South Yemen, and several countries in Eastern Europe.28 In the Cuban case in 1962, the Lunas nuclear capability attracted the attention of U.S. analysts and policymakers from the start. After U.S. intelligence and military ofcials detected the Lunas, they singled them out for special mention in reports and nearly always described them as analogous to the U.S. Armys Honest Johns, a weapon notable for its nuclear capabilities.29 Available evidence suggests that the Soviet Union secretly sent 36 Luna rockets to Cuba, twelve of which were slated to be equipped with special (nuclear) warheads.30 Khrushchev authorized sending the Luna warheads in a revision to the original Anadyr plan on 7 Sepeds., The Kennedy Tapes: Inside the White House during the Cuban Missile Crisis, concise ed. (New York: W. W. Norton, 2002), p. 290. 26. The Anadyr operation has been covered in detail elsewhere. See particularly Gribkov and Smith, Operation Anadyr; Fursenko and Naftali, One Hell of a Gamble, pp. 166197; and Allison and Zelikow, Essence of Decision. 27. Ofce of Naval Intelligence, Intelligence Brief, ONI Review, Vol. 18, No. 1 (January 1963), p. 559. 28. Aaron Karp, Ballistic Missiles in the Third World, International Security, Vol. 9, No. 3 (Winter 1984/1985), p. 176. 29. Mary T. Cagle, History of the Basic (M31) Honest John Rocket System, 19501964, U.S. Army Historical Monograph Project Number AMC 7 M, Part I (1964); and John P. Rose, The Evolution of U.S. Army Nuclear Doctrine, 19451980 (Boulder, CO: Westview, 1980). 30. One recent study has identied the Luna deployment as consisting of twelve nuclear Frog-5 and 24 conventional Frog-3s. Those designations differ from the ones used by U.S. military intelligence specialists in 19621963. See Norman Polmar and John F. Gresham, DEFCON-2: Standing on the Brink of Nuclear War during the Cuban Missile Crisis (Hoboken, NJ: John Wiley, 2006), p. 312.

13

Coleman

tember, the same day that Kennedy announced his request for authority from Congress to call up 150,000 reserves. The nuclear warheads for the Lunas arrived in Cuba in October on the Indigirka.31 Detecting and identifying the Luna missiles proved a challenge for U.S. intelligence. As of late 1962, U.S. defense and intelligence analysts had spotted four variants of Frogs, which they numbered one through four after identifying them visually by the shape of the nose cone.32 They believed that the Frog-1, -2, and -4 variants could carry either conventional or nuclear payloads, whereas the Frog-3 variant was capable only of nuclear re (see Table 1). The Frog system comprised several components: the delivery vehicle itself, the transporter, the launcher, and the warhead. Because transporters and launchers were easiest to photograph and identify by virtue of their physical size, the U.S. analysts based their assessments on photographs of those two parts of the Frog system. Initially, the Soviet forces parked the Frogs in the open, but when low-level U.S. surveillance ights began on 23 October, Soviet troops began hiding and camouaging the weapons.33 Missiles were generally covered or stored in hangars, large tents, or caves and were therefore hard to identify from U.S. surveillance ights. The warheads were even more difcult to identify directly because they were small, easily hidden in bunkers, buildings, or caves, and tightly guarded by Soviet soldiers. Intelligence analysts therefore had to rely on circumstantial evidence such as the construction of dome-shaped bunkers and unusual security arrangements. Other Soviet short-range, surface-to-surface missiles were also present in Cuba, the rst of which to be deployed and detected by American surveillance were cruise missiles. U.S. intelligence ultimately counted a total of 150 cruise missiles in Cuba but never fully appreciated that these included two distinct types, one of which was nuclear-capable.34 The rst were Sopka coastal defense missiles. Resembling small jet airplanes, they were known to the West as SSC-2B (Samlet) missiles, which had a range of 65 to 180 kilometers and were armed with one-ton conventional warheads.35 The second kind of cruise
31. Fursenko and Naftali, One Hell of a Gamble, p. 213. 32. In the mid-1960s, an updated version with a longer range, designated the Frog-7 by NATO, became the standard and may have been shipped to Cuba to replace the obsolete models. See International Institute of Strategic Studies, The Military Balance, 19741975 (London: IISS, 1974). Since the 1960s, a number of other countries have acquired long-range Frog-5s and Frog-7s, including North Korea. See W. Seth Carus, Long Range Rocket Artillery in the Third World, Janes Intelligence Review, Vol. 9, No. 10 (October 1991), pp. 471477. 33. Testimony by John Hughes of the U.S. Defense Intelligence Agency (DIA) before the Committee on Armed Services of the U.S. House of Representatives, 30 January 1963, U.S. Congress, House of Representatives, Committee on Armed Services, Armed Services Papers 196364, 88th Cong., 2nd Sess., 1964, p. 254. 34. Garthoff, U.S. Intelligence in the Cuban Missile Crisis, p. 29. 35. CIA/DIA, Cuba 1962: Khrushchevs Miscalculated Risk, 13 February 1964, in Box 35, NSF,

14

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

Table 1. U.S. Intelligence Data on Frog Missiles, c. 1962 Max Range 15 NM 11 NM 13 NM 26 NM Range of nuclear warhead yields 20-1100 KT 5-335 KT 5-335 KT Time required to re after arrival at prepared site 1530 minutes 1530 minutes 1530 minutes (estimated) 1530 minutes (estimated)

System FROG 1 FROG 2 FROG 3 FROG 4

CEP 380750 meters 270540 meters 450900 meters 6001500 meters

1. The warhead yields are the capabilities given the respective weapons based on technical characteristics. The probability is that the warhead yield will be less than the maximum indicated. 2. The FROG 1 and 2 have both a nuclear and high explosive capability, while it is believed the FROG 3 has only a nuclear capability. The FROG 4 is possibly designed solely for use with high explosive, but could use the FROG 3 warhead. Source: Cuba: Subjects, Intelligence Material, 10/1/6211/12/62, Box 51, NSF, JFKL. The table is reproduced intact (including the two notes in the original)

missile was known in Soviet parlance as FKRs (frontovye krylatye rakety). These had a range of 145 to 175 kilometers and were armed with nuclear warheads of 5 to 12 kilotons. Their combat role in Cuba was primarily for protecting beachheads and targeting the U.S. naval base at Guantnamo Bay. U.S. intelligence analysts never identied the FKRs, apparently counting them as Sopka missiles. In the 7 September 1962 revision to the Anadyr plan, Khrushchev approved the Soviet Defense Ministrys proposal to add 80 nuclear warheads for the FKRs.36 By far the most numerous of the short-range missiles were the sophisticated surface-to-air missiles (SAMs) known to Soviet commanders as V-75 Volkhovs and to the North Atlantic Treaty Organization (NATO) as SA-2 Guidelines. By mid-October 1962, approximately 500 V-75 missiles with 144 launchers were distributed among 24 operational SAM sites in Cuba. These missiles had a lateral (or slant) range of 40 to 50 kilometers and an effective vertical range of more than 24 kilometers. Supported by a state-of-theart radar network and carrying 240-kilogram high-explosive warheads, they posed a serious threat to aircraftas was dramatically demonstrated when a V-75 brought down the U-2 piloted by Francis Gary Powers over Soviet terriLyndon Baines Johnson Library (LBJL); and Roger Hilsman to Dean Rusk, Review of Recent Developments in Cuba, 23 January 1963, in Cuba, Subjects, Intelligence Materials 1/63, Box 51, NSF, JFKL. 36. Gribkov and Smith, Operation Anadyr, p. 4; Fursenko and Naftali, One Hell of a Gamble, p. 188; Fursenko and Naftali, The Pitsunda Decision, pp. 223225; and Garthoff, New Evidence on the Cuban Missile Crisis, pp. 251254.

15

Coleman

tory in May 1960but were of little if any use against anything on the ground or at sea.37 The technology behind the V-75 systems was state-of-theart, in some respects more advanced than the Hawk missiles that the United States had recently begun supplying to Israel. The V-75 missiles conceivably could also be armed with nuclear warheads, although U.S. analysts considered such a scenario unlikely and Soviet air defense forces were never known to have deployed nuclear-tipped V-75s.38 The Soviet Union also had air-to-air missiles and anti-ship missiles in Cuba. Soviet military commanders sent as many as sixteen Komar-class guided-missile patrol boats to Cuba. These vessels were similar to motor torpedo boats, but instead of torpedoes they carried two short-range cruise missiles tted with conventional warheads. The boats were small and fast, but their limited range restricted them to inshore operations, and they were best used against ships and amphibious attacks.39 For air combat, Soviet commanders sent 100 MiG ghters to Cuba. Of these, 42 were advanced MiG-21 (Fishbed) aircraft, a supersonic ghter that could be used for both interception and ground attack. The MiG-21 had a combat radius of more than 550 kilometers and was typically armed with a variety of cannons, infrared air-toair missiles, and air-to-surface rockets.40 U.S. analysts concluded that the MiG-21s in Cuba probably were capable of carrying nuclear weapons but that arming them with nuclear bombs would severely limit their effective range (to around 320 kilometers) and hinder their navigation systems. As Kennedys national security adviser, McGeorge Bundy, put it, such tradeoffs meant that nuclear weapons on the MiG-21s were not a likely conguration.41

Offensive vs. Defensive

During a press conference on 13 September 1962, the president was asked about the threat posed by the military buildup in Cuba, and his answer set forth a dichotomy that framed the issue throughout the crisis and after. He said that defensive weapons, such as those that existed in Cuba at the time
37. CIA, The Crisis: U.S.S.R./Cuba, 3 November 1962, in Countries: Cuba, CIA Memoranda, 11/1/6211/3/62, Box 46. NSF, JFKL. 38. For a detailed discussion of the SA-2 systems specications and capabilities, see CIA/DIA, Cuba 1962: Khrushchevs Miscalculated Risk, 13 February 1964, in Box 35, NSF, LBJL. 39. The Trend toward Modernization in the Soviet Navy, ONI Review, Vol. 17, No. 2 (February 1962), pp. 3952, Box 1292; Folder: ONI Review Vol. 1718, Naval Historical Ofce, Washington Navy Yard; CIA/DIA, Cuba 1962, 13 February 1964,. 40. Trevor Cliffe, Military Technology and the European Balance, Adelphi Paper No. 89 (London: International Institute for Strategic Studies, 1972), p. 46; and CIA/DIA, Cuba 1962, 13 February 1964. 41. Naftali and Zelikow, The Presidential Recordings, Vol. 2, p. 21.

16

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

of his statement, would be tolerated, whereas offensive weapons would not. Under pressure from Congress, Kennedy characterized the increasingly strident calls for invasion of the island as irresponsible loose talk that only encouraged Castros claims of American aggressive intentions.42 He said that existing shipments of weaponry to Cuba did not constitute a serious threat to any other part of this hemisphere. Nevertheless, he warned that if the buildup crossed the threshold from defensive to offensive, or if the U.S. naval base at Guantnamo was threatened, the gravest issues would arise.43 The problem was in dening that threshold. Senator Kenneth Keating, a Republican from New York who was one of the most tenacious critics of the administrations Cuba policy, denied that any such distinction existed: Who is to say whether a weapon is offensive or defensive? It depends on the direction which it is aimed.44 U.S. analysts, especially after the failed Bay of Pigs invasion, recognized that Cuba might deploy missiles for its own defense. When the 21 members of the Organization of American States met at Punta del Este, Uruguay, in January 1962, U.S. ofcials matter-of-factly acknowledged the possibility that bombers, naval craft, and possibly short-range tactical guided missiles might eventually be delivered to Cuba.45 By the time of Kennedys September statement, some short-range missiles had already been detected and publicly acknowledged. On 29 August, two weeks before Kennedys press conference, a U-2 reconnaissance plane provided the rst conclusive evidence that Soviet shortrange missiles were already deployed in Cuba. When analysts from the CIAs National Photographic Interpretation Center (NPIC) pored over the 928 photographs from the U-2s six-minute ight over the island, they discovered eight separate installations of V-75 SAMs.46 Kennedy instructed the acting CIA director, General Marshall Carter, to put that information in the box and nail it shut.47
42. See Hershberg, Before The Missiles of October. 43. Public Papers of the Presidents: John F. Kennedy, 1962 (Washington, DC: U.S. Government Printing Ofce, 1963), p. 517. 44. Paterson and Brophy, October Missiles and November Elections, p. 95. 45. That possibility was acknowledged in two statements prepared on 10 January and 5 February 1962 in connection with the Punta del Este conference. Neither statement was used publicly, but the January version was released to a number of reporters in Washington. The documents are quoted and referred to in a Memorandum from Joseph W. Neubert to Edwin Martin, 17 October 1962, in Box 1, CountriesCuba 1962, Box 1, Roger Hilsman Papers, JFKL. 46. Editorial Note, in FRUS, 19611963, Vol. XI, p. 29. The earliest physical evidence that the Soviet Union was sending SAMs to Cuba came from photography of Soviet ships in early August. The photographs did not show actual SA-2 missiles, but they did reveal the types of construction and transport equipment associated with Soviet deployments of SAMs in Indonesia in 1961. 47. Lyman B. Kirkpatrick, Action Generated by DCI Cables Concerning Cuban Low-Level Photography and Offensive Weapons, in McAuliffe, ed., CIA Documents on the Cuban Missile Crisis, p. 97.

17

Coleman

The ofcial position was that the SAMs posed no threat to the United States, but some ofcials, particularly CIA Director John McCone, were worried about the new weapons. McCone argued that the SAMs might be there to provide cover for something more important, perhaps MRBMs. Theyre not putting [the SAMs] there to protect the [sugar] cane-cutters, he is reported to have said.48 Senator Keating claimed publicly that he had obtained reliable evidence of the presence of Soviet MRBMs in Cuba.49 Nevertheless, the consensus of ofcial opinion through September and early October remained that the existing armaments in Cuba were defensive in nature and therefore posed no serious threat to other countries in the region. The tipping point, agreed McGeorge Bundy and other senior advisers at a meeting on 4 September, would be the deployment of surface-to-surface missiles or nuclear warheads.50 A National Intelligence Estimate (NIE) dated 19 September concluded that in fact the United States could tolerate a signicant amount: to wit, that weapons of a more offensive character: e.g., light bombers, submarines, and additional types of short-range surface-to-surface missiles (SSMs) could still fall within the defensive category.51 The NIE deemed it unlikely that the Soviet Union would be so provocative as to cross the threshold from a defensive to an offensive buildup, a judgment that proved famously off the mark.52 Unconrmed reports of the presence of surface-to-surface cruise missiles had been circulating in the intelligence community since early August. The rst photographic conrmation came from U-2 surveillance ights on 19 August and another ight on 5 September. NPIC analysts identied a missile complex at the port of Banes on the eastern end of the island but could not immediately determine the types of missiles. The president, McNamara, and Rusk were informed of this nding on 6 September. Over the next two weeks, attempts to identify the missiles still were unsuccessful. Despite public rumors of cruise missiles with a range of more than 300 kilometers, analysts judged
48. Dino A. Brugioni, Eyeball to Eyeball: The Inside Story of the Cuban Missile Crisis (New York: Random House, 1990), p. 105. 49. On Keatings sources, see Max Holland, A Luce Connection: Senator Keating, William Pawley, and the Cuban Missile Crisis, Journal of Cold War Studies, Vol. 1, No. 3 (Fall 1999), pp. 139167. 50. Naftali and Zelikow, eds., Presidential Recordings, Vol. 1, pp. 2122. The key gures in this exchange (Bundy, Rusk, and McNamara) did not specify whether they were referring to long range or tactical surface-to-surface missiles. 51. Special National Intelligence Estimate (SNIE) 85-3-62, The Military Buildup in Cuba, 19 September 1962, in Countries, Latin America, Cuba, CubaIntelligence Reports 196263, Vol. 2, Box 37, NSF, LBJL. 52. SNIE 80-62, The Threat to U.S. Security Interests in the Caribbean Area, 17 January 1962; NIE 85-62, The Situation and Prospects in Cuba, 21 March 1962; NIE 85-2-62, The Situation and Prospects in Cuba, 1 August 1962; and SNIE 85-3-62, The Military Build-Up in Cuba, 19 September 1962.

18

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

that the weapons most likely were short-range, land-based, coastal defense missiles with a range of only around 4050 kilometers.53 This assessment was later rened. A consensus emerged within the intelligence community that the missiles observed at several coastal sites were cruise missiles with a range of 50 to 65 kilometers, probably derived from the AS-1 Kennel-type air-tosurface anti-shipping missile armed with a 1,000-kilogram conventional warhead.54 U.S. intelligence analysts ultimately counted 150 coastal defense missiles. As of February 1963, they had pinpointed four operational sites, each with eight to ten missiles, at Siguanea, Santa Cruz del Norte, Banes, and Campo Florida. The remaining missiles remained in crates at several locations throughout Cuba. U.S. analysts estimated that the Soviet Union planned to deploy the remaining coastal defense missiles at up to another fteen sites. With the benet of hindsight (and, more importantly, information from former Soviet sources) it now appears that U.S. intelligence counted the total number of cruise missiles in Cuba fairly accurately but failed to recognize that the missiles were of two different types.55 By comparing information gleaned from Russian sources with declassied U.S. intelligence data, it appears that the weapons photographed at the four operational sites were the Sopka missiles and that the other missiles, initially observed in crates, were the FKRs. It is now known that 80 nuclear warheads were in Cuba at the time, but partly because of the failure to identify the types of missiles accurately, U.S. analysts never fully realized that nuclear-capable cruise missiles were in Cuba. Under Secretary of State George Ball was the rst to disclose outside the administration the news that surface-to-surface cruise missiles were present in Cuba when he told a congressional committee on 3 October that U.S. intelligence agencies had detected three to four coastal defense missile sites and sixteen Komar class guided-missile patrol boats. Available intelligence, which he
53. CIA Ofce of Current Intelligence, Analysis of the Suspect Missile Site at Banes, Cuba, 13 September 1962, in McAuliffe, ed., CIA Documents on the Cuban Missile Crisis, pp. 7173; and SNIE 853-62, The Military Buildup in Cuba, 19 September 1962. 54. CIA, Cruise Missile Deployment in Cuba, 20 February 1963, in Cuba, Subjects, Intelligence Materials 2/63, NSF, JFKL; Joint Evaluation of Soviet Missile Threat in Cuba, 19 October 1962, in McAuliffe, ed., CIA Documents on the Cuban Missile Crisis, p. 207; The Threat of Soviet Naval Air to Western Task Forces, ONI Review, Vol. 17 (September 1962), pp. 294295; and Soviet Forces in Cuba, 5 February 1963, in Meetings & Memoranda, Executive Committee Meetings, Meetings 38 42, 1/25/633/29/63, Box 316, NSF, JFKL. 55. In January 1965, U.S. intelligence analysts rst recognized that surface-to-surface cruise missiles were present in Cuba. These may have been the FKRs, but the evidence is patchy. During a military parade in Havana on 2 January 1965, the Cubans paraded three cruise missiles that were slightly different from the usual Kennel variant. The parade announcer identied them as new land-to-land missiles, but U.S. intelligence struggled to identify them positively. See Current Intelligence Digest, 2 January 1965, Country, Cuba, Intelligence Vol. II 1/654/65, NSF, LBJL; and Directorate of Intelligence, Ofce of Current Intelligence, Weekly Cuban Summary, 13 January 1965, in Countries, Cuba, CIA Daily and Weekly Summaries, Vol.1, Box 36, NSF, LBJL.

19

Coleman

characterized as quite complete and very good and very hard, indicated that no missiles on the island had a range beyond 50 kilometers.56 His statement was regarded as routine at the time, but after the crisis it became controversial in the bitter political ght to untangle what the administration knew and when it knew it. Less than two weeks after Balls statement, it was rendered obsolete. The discovery of MRBMs in Cuba, rst reported to Kennedy on the morning of 16 October, escalated tensions markedly and injected a new sense of urgency into the debate over the distinction between offensive and defensive weapons. Throughout the ensuing crisis and its aftermath, Soviet ofcials persistently denied that the MRBMs were offensive weapons, and Khrushchev adopted the formulation the weapons you call offensive when addressing the issue. The same forms of weapons can have different interpretations, he wrote to Kennedy on 26 October. Our conceptions are different on this score, or rather, we have different denitions for these or those military means.57 His argument was more than just political doublespeak. Soviet military doctrine used the terms tactical and strategic differently from the way they were used by Western governments. The Soviet Union emphasized the role of the weapons, whereas Western military planners stressed technical capabilities such as range and explosive yield.58 Regardless, as far as Washington was concerned, the discovery of MRBMs in Cuba crossed the threshold from defensive to offensive as Kennedy had dened them in his statement on 13 September.59 The MRBMs had a range of nearly 1,800 kilometers and could reach Washington, DC, and about half of the continental United States within minutes of launch. The construction of bases for intermediate-range ballistic missiles (IRBMs) with a range of 4,000 kilometers indicated that those missiles were on their way, although none actually arrived.
56. Roger Hilsman to George Ball, 2 October 1962, in Tabs CG, Countries, Cuba, Subjects, Testimony, Under Secretary Ball, Tabs C-G NSF, JFKL. In early February 1963, this earlier brieng was revived by members of the Mahon committee, particularly Representative Liscomb, when they suspected serious discrepancies between Balls testimony and McCones. See, for instance, Michael V. Forrestal to McGeorge Bundy, Congressional Testimony on Cuba, 18 February 1963, in Cuba, Subjects, Testimony, 5/7/622/27/63, Box 61, NSF, JFKL; and Thomas L. Hughes to George Ball, 14 February 1963, in Cuba, Subjects, Testimony, 5/7/622/27/63, Box 61, NSF, JFKL. 57. See, for example, Khrushchev to Kennedy, 26 October 1962, in FRUS, 19611963, Vol. XI, pp. 235236. Privately, Khrushchev himself had referred to those same weapons as offensive. See Fursenko and Naftali, Khrushchevs Cold War, p. 477. 58. Lawrence Freedman, The Evolution of Nuclear Strategy, 3rd ed. (Houndmills, UK: Palgrave Macmillan, 2003), pp. 105106. 59. For a narrative of the thirteen days of crisis, see May and Zelikow, eds., The Kennedy Tapes; Fursenko and Naftali, One Hell of a Gamble; Allison and Zelikow, Essence of Decision; and Laurence Chang and Peter Kornbluh, eds., The Cuban Missile Crisis, 1962: A National Security Archive Documents Reader, rev. ed. (New York: The New Press, 1998).

20

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

During the rst ten days of deliberations, the ExComm knew that V-75s (SA-2s), Komar class torpedo boats, MiG-21 jets with air-to-air missiles, coastal-defense cruise missiles, and Il-28 bombers were deployed in Cuba. Up to that point, however, no ofcial mention had been made of the Lunas (Frogs), a weapon notable primarily as a tactical nuclear weapon. A CIA document distributed to senior policymakers on 19 October warned that refugee reports indicated that Luna rockets were in Cuba, but NPIC analysts scouring the high-altitude photography initially could nd no evidence to back up that claim.60 But after the rst waves of low-level surveillance ights, which began on 23 October, U.S. analysts could view greater detail of smaller patches of territory. A ight on 25 October photographed a Luna launcher at a large camp at Remedios, a site in central Cuba being prepped to receive IRBMs. Kennedy and his advisers were informed the next day.61 Neither the presidential recordings nor the ofcial minutes of the ExComm meetings reveal any immediate follow-up discussion regarding the Lunas or the implications of their presence during the peak period of crisis prior to 28 October. The news did, however, have a direct effect on military planning. The original plans for possible military action against Cuba did not envisage the use of nuclear weapons. But when the low-level ights conrmed the presence of Lunas, the JCS and the commander-in-chief of the Atlantic Theater (CINCLANT), Admiral Robert L. Dennison, recognized that they needed to be ready for the possibility that if the president ordered an invasion U.S. troops could confront tactical nuclear weapons. On the morning of 28 October, the JCS decided that Dennison should revise the main military plan for the invasion of Cuba (OPLAN-316) and make recommendations on whether tactical nuclear weapons (air and ground) should be included in the arsenal of our invasion and supporting forces.62 As the Army later explained it in recommending language for McNamaras report to Congress, the revi-

60. Joint Evaluation of Soviet Missile Threat in Cuba, 19 October 1962, in McAuliffe, p. 208. 61. Zelikow and May, eds., The Presidential Recordings, Vol. 3, p. 327; Supplement 7 to Joint Evaluation of Soviet Missile Threat in Cuba, 27 October 1962, in McAuliffe, ed., CIA Documents on the Cuban Missile Crisis, p. 325; Hilsman, To Move a Nation, p. 215; JCS, Chronology of JCS Decisions Concerning the Cuban Crisis, pp. 48, 79, Compilation of High Level Exchange in Cuba Crisis, in Box 16, Maxwell Taylor Papers, National Defense University, Washington, DC; and U.S. Congress, Senate, Committee on Armed Services (SCAS), Preparedness Investigating Subcommittee (PIS), The Cuban Military Buildup: Interim Report, 88th Cong., 1st Sess., 9 May 1963. Members of the Berlin Ambassadorial Group, which had British, French, and U.S. representatives, were given a brieng on 29 October that covered the Frog missiles. See Sir David Ormsby-Gore to Foreign Ofce, 29 October 1962, in PREM 11/3691, The National Archives (TNA) of the United Kingdom. 62. Marine Corps Emergency Actions Center, Summary of Items of Signicant Interest, 29 October 1962, Item No. CC01664, in Cuban Missile Crisis Collection, National Security Archive; and JCS, Chronology of JCS Decisions Concerning the Cuban Crisis, 21 December 1962, p. 52.

21

Coleman

sions of the invasion plans were sparked by the discovery of the Luna/Frog launchers:
when a determination was made that nuclear capable FROG (Free Rocket Over Ground) missiles were present in Cuba, CINCLANT expressed his concern, and proposed to have tactical nuclear weapons readily available in his invasion force. These weapons would be used only in retaliation for the employment of nuclear weapons against US forces.63

Dennison told the JCS that he would modify his plans accordingly, but the Joint Chiefs and McNamara ultimately authorized only part of Dennisons request, allowing him to equip the invasion force with nuclear-capable delivery systems such as Honest John missiles and eight-inch howitzers. They did not authorize him to include nuclear warheads without explicit JCS approval. In addition, the JCS authorized the pre-positioning of some tactical nuclear weapons: Honest John rockets were deployed to Florida without their nuclear warheads, and batteries of Nike-Hercules nuclear surface-to-air missiles were also sent to Florida.64 Some sources indicate that U.S. Pershing missiles were sent to Key West with nuclear warheads, although hard evidence to support this claim remains elusive.65 No documentary evidence has come to light explaining the thinking behind the decision to withhold authority to use nuclear weapons. Available documents from the period record the decision without the rationale.66 Decades later, Robert McNamara said that he and President Kennedy explicitly ruled out arming the prospective invasion force with tactical nuclear weapons but that no one should believe that, had American troops been attacked with nuclear weapons, the United States would have refrained from a nuclear response.67 General William Smith, who was special assistant to JCS Chairman General Maxwell Taylor and played a central role in U.S. invasion planning, said at a 1992 conference that Dennisons request was turned down because there was no evidence any [Soviet] nuclear weapons were there.68 If his recollection is accurate, it is unclear whether the decision amounted to an extraor63. John A. Heintges to Special Assistant to the Assistant to the Secretary, 29 December 1962, Item CC02795, in Cuban Missile Crisis Collection, National Security Archive. 64. Brieng Report for General Taylor, 10 November 1962, in Box 6, Record Group (RG) 218, U.S. National Archives (NA); and Brieng Report for General Taylor, 11 November 1962, in Box 6, RG 218, NA. 65. Brugioni, Eyeball to Eyeball, pp. 296297. 66. John A. Heintges to Special Assistant to the Secretary of Defense, 29 December 1962, Item No. CC02795, Cuban Missile Crisis Collection, National Security Archive. No record exists of the JCS meeting. If a record of this meeting ever existed, it probably was destroyed in August 1974. 67. McNamara, In Retrospect, p. 341. 68. Blight, Allyn, and Welch, Cuba on the Brink, p. 261.

22

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

dinary order from McNamara or the White House or simply a reafrmation of standing orders. The prevailing rules of engagement for air defense of Florida stated explicitly that nuclear weapons would not be used.69 Whatever the specic rationale, Dennisons request was denied. To compensate, CINCLANT planners tried to reduce reaction times and increase the numbers of troops in order to reduce the vulnerability of the invading force to Soviet tactical nuclear weapons.70

The Missiles Weve Had on Our Minds

After Khrushchevs retreat on 28 October, the rst MRBM transporters began leaving Cuba on 5 November. By 10 November, U.S. surveillance had counted 33 MRBM transporters on eight ships heading back to the Soviet Union, and the Kennedy administration had decided to take Soviet leaders at their word that they had deployed a total of 42 MRBMs in Cuba and that all the missiles had been removed.71 Many issues remained unresolved, however. The focus shifted to other Soviet weapons based in Cuba and whether the administration should push for the status quo ante or accept some of the military buildup. The early shipments leaving Cuba did not include V-75s, cruise missiles, Lunas, or Il-28s. The urry of letters between Kennedy and Khrushchev during the crisis had often been vague and not always consistent about the types of weapons covered. The Kennedy administrations quarantine proclamation of 23 October had listed surface-to-surface missiles; bomber aircraft; bombs, air-to-surface rockets and guided missiles; warheads for any of the above weapons; mechanical or electronic equipment to support or operate the above items; and any other classes of materiel hereafter designated by the Secretary of Defense for the purpose of effectuating this Proclamation. In the subsequent letters to Khrushchev, Kennedy returned to the offensive/defensive dichotomy, repeatedly demanding that the Soviet Union remove its offensive weapons from Cuba. Khrushchev, for his part, repeatedly disputed that label even for the long-range missiles and bombers. In his note of 28 October, Khrushchev
69. Memorandum for the Chairman, Joint Chiefs of Staff, Rules of Engagement, n.d., in Box 6, Maxwell Taylor Papers, RG 218, NA. 70. Desch, That Deep Mud in Cuba, p. 333; and JCS, Chronology of JCS Decisions Concerning the Cuban Crisis, p. 65. 71. U.S. intelligence analysts had photographic conrmation of only 33 MRBMs in Cuba and could nd no compelling reason not to believe Soviet and Cuban ofcials when they said that 42 MRBMs had been in Cuba and that all of them had been removed. See Ofce of Naval Intelligence, The Missiles Leave Cuba, ONI Review, Vol. 17, No. 12 (December 1962), p. 511; and Ofce of Naval Intelligence, Intelligence Briefs, ONI Review, Vol. 18, No. 1 (January 1963), p. 557.

23

Coleman

had agreed to remove the arms which you described as offensive, a phrase that U.S. policymakers initially construed as referring to the list in the quarantine proclamation of 23 October.72 Afterward, however, Soviet forces in Cuba continued uncrating Il-28 bombers, constructing barracks for the combat regiments, continuing work on the cruise missile sites, and showing signs that they were still planning to establish submarine bases in Cuba. Administration ofcials realized that their interpretation of the agreement was markedly different from Moscows. President Kennedy ultimately settled for less than what he had set out in his 23 October proclamation. On 2 November, U.S. Ambassador to the United Nations Adlai Stevenson handed to Soviet Deputy Premier Anastas Mikoyan a list of specic weapons to be removed. The JCS had played an inuential role in devising the list, which reected the Chiefs maximal approach. The list, the Joint Chiefs said, must include, as a minimum, surfaceto-surface missiles, bomber or attack aircraft, other combat aircraft that can be considered as nuclear weapons carriers including the MIG-21s, Komar class PT boats, nuclear warheads of any kind, nuclear storage sites and any other nuclear delivery systems.73 When Khrushchev objected to the list in a note delivered to the White House three days later, Kennedy was prepared to compromise. He convened his senior advisers to come up with a revised list. Revising the list forced U.S. policymakers to think carefully about what they considered a real threat and what they could tolerate if need be. Kennedy and his aides sensed that if they pushed too hard, they might push Khrushchev beyond the breaking point, alienate their allies, and reignite the crisis. But they also worried that if they were too cautious, they would end up with an unnecessary threat in the Caribbean and all the military and political consequences it would entail. Faced with this dilemma, U.S. policymakers reduced their assessment of the threat to a binary equation, emulating what Kennedy had done in his statement of 13 September: defensive weapons were tolerable, offensive weapons were not. But they had not yet reached a consensus about the weapons that fell into each category. No one in Washington questioned that the MRBMs and IRBMs fell squarely within the offensive category, but views about other kinds of weapons were more divergent. The Il-28s, for example, were the subject of extensive internal debate. The president was reluctant to let the settlement get hung up on the bombers and had to be convinced by several of his senior advisers that, as McNamara put it, its absolutely essential that the IL-28s are part of the deal. I dont
72. Khrushchev to Kennedy, 28 October 1962, in FRUS, 19611963, Vol. XI, p. 279. 73. Colonel A. B. Parsons (USAF), Talking Paper for the Chairman, JCS, for the JCS Meeting, 30 Oct, 30 October 1962, in Box 6, Maxwell Taylor Papers, RG 218, NA.

24

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

think we could live with the American public if they werent. Ultimately, Kennedy was persuaded to insist on their removal.74 What was deliberately omitted from the revised list was as signicant as what was included. The short-range surface-to-surface missiles fell within a gray area in threat assessments. Those weapons could not reach U.S. territory but could be used against U.S. forces at Guantnamo or U.S. forces that might be ordered to invade Cuba. The primary concern, however, was that if the Soviet Union continued to strengthen its garrison in Cuba with SAMs, Lunas, and other ground weaponry, the deployments would provide cover for the rapid reintroduction of strategic weapons.75 Fears also abounded that warheads could be delivered even more quickly by submarine or by air.76 In addition, some ofcials were concerned that allowing any nuclear-capable forces to stay in Cuba, even battleeld ones, would set a bad precedent and create political difculties. The White House statement of 4 September had implicitly accepted that the short-range missile systems that had been publicly identiedSAMs, Komar torpedo boats, and MiGswere all defensive weapons systems. The statement said nothing about the Lunas because they had not yet been found, although they presumably fell within the broad category of surface-to-surface missiles.77 In early November, as the ExComm received increasingly detailed information about the Lunas and associated forces, high-level debate about the short-range missiles resumed. The key decisions were taken at a secretly recorded ExComm meeting on the evening of 7 November as the group made nal revisions to the list of prohibited weapons that Adlai Stevenson would hand to Vasilii Kuznetsov in New York to bring the negotiations to a close. Under Secretary of State U. Alexis Johnson and Assistant Secretary of Defense Paul Nitze specically raised the issue of whether the Lunas/Frogs should be included on the list. The FROGs clearly arent offensive weapons in the way in which we meant the term, McNamara responded. Theyve never been in
74. Bernstein, Bombers, Inspection and the No Invasion Pledge, pp. 812. 75. 19th Meeting of the ExComm, 3 November 1962, in Coleman and Naftali, eds., Presidential Recordings, Vol. 4. The audio recording is on Tape 49, Presidential Recordings Collection, POF, JFKL. 76. U.S. ofcials debated whether the Soviet Union could reintroduce nuclear warheads into Cuba quickly and secretly. One possibility they considered was that Soviet submarines might deliver the warheads. Another possibility was delivery by air, especially after the Soviet airline Aeroot began regular weekly civilian service from Moscow to Havana via Murmansk on 7 January 1963. See Roger Hilsman to Dean Rusk, Review of Recent Developments in Cuba, 23 January 1963, in Cuba, Subjects, Intelligence Materials, 1/63,NSF, JFKL. 77. The full list included in the proclamation was: Surface-to-surface missiles; bomber aircraft; bombs, air-to-surface rockets and guided missiles; warheads for any of the above weapons; mechanical or electronic equipment to support or operate the above items; and any other classes of materiel hereafter designated by the Secretary of Defense for the purpose of effectuating this Proclamation. Proclamation 3504, p. 487.

25

Coleman

our list. When Nitze objected that the Frogs were surface-to-surface missiles, a category that had been included in the original list of prohibited weapons, McNamara responded, No, but we didnt mean the FROGs. Bundy agreed with him, arguing that the short-range surface-to-surface missiles, whether the Frogs or Komar torpedo boats, were tolerable under the circumstances: If we put another ratchet on these ground-to-ground missiles after our letter of last night, he really will think were just applying the squeeze. McNamara agreed: Seems to me we just have to concentrate on these IL-28s and we confuse the issue by dragging in a lot of questionable items.78 Harlan Clevelands notes from the meeting indicate that the nal decision was to insist on the Il28s but to let the other items go.79 The ExComm instructed Stevenson and McCloy to let Kuznetsov know that the administration was still concerned about the Il-28s and nuclear warheads but was willing to be relaxed and exible about the other weapons.80 Even though President Kennedy (who was present at the meeting) did not participate directly in that exchange, the available evidence strongly suggests that he was closely following the issue of the remaining ground forces and accepted the logic. He recognized that the Soviet Union would have little incentive to remove the remaining troops unless the United States offered a formal UN-supervised guarantee that it would not invade Cuba. Such a guarantee, Kennedy believed, was too high a price to pay because it would psychologically be a source of strength to Castro and a source of difculty for us, here.81 U.S. allies also seemed reluctant to support a further ratcheting up of pressure over the lesser weapons. When Sherman Kent, the head of the CIAs Board of National Estimates, provided a special brieng to NATO allies on 20 November, the participants pressed him on whether the short-range missiles would be dened as offensive.82 During an ExComm meeting on 29 November, Kennedy explained why he was prepared to go along tacitly with the
78. 22nd Meeting of the ExComm, 7 November 1962, in Coleman and Naftali, eds., Presidential Recordings, Vol. 4. The audio recording is on Tape 53A, Presidential Recordings Collection, POF, JFKL. See also 14th Meeting of the ExComm, 31 October 1962, in Coleman and Naftali, eds., Presidential Recordings, Vol. 4. The audio recording of that meeting is on Tape 46, Presidential Recordings Collection, POF, JFKL. 79. Harlan Clevelands Meeting Notes, 7 November 1962, in Cuba 11/7/62, Box 76, Harlan Cleveland Papers, JFKL. 80. Harlan Cleveland, Draft Telegram to Stevenson and McCloy, 8 November 1962, in CubaCuban CrisisH.C.s BookTop Secret, 11/62, Box 78, Harlan Cleveland Papers, JFKL. According to Clevelands handwritten note on the draft, he conveyed the message by telephone. 81. 31st Meeting of the ExComm, 29 November 1962, in David Coleman and Marc Selverstone, eds., The Presidential Recordings: John F. Kennedy, Vol. 5 (New York: W. W. Norton, forthcoming). The audio recording is on Tape 65, Presidential Recordings Collection, POF, JFKL. 82. Thomas K. Finletter to Rusk, 26 November 1962, in Cuba-Cables11/24/6211/28/62, Box 42, NSF, JFKL.

26

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

continued presence of Soviet troops. I think without the assurance, Kennedy told his advisers,
the chances of his [Khrushchev] taking out those guard units and these FROGs and all the restwhich he indicated or hinted that hed probably be doing it in due timeI would think he wouldnt take them out. Because hell say we havent given the assurance, therefore weve got to maintain our defensive strength to protect Cuba. So I would think that we have to really make a choice. My feeling would be that we would be better off to have those units in and not have the guarantee.

Bundy agreed that the Lunas were not that important and that a formal guarantee about Cuba would be a bargaining point that we dont buy. Thats what I think, Kennedy said. I think were all in agreement that we would much rather have all this stuff there, SAM sites, FROGs, and all the rest, than to be really locked in any kind of a guarantee that would be difcult to get out of. McNamara pointed out that they have patrol craft missiles, the missiles on the Komars. And they have FROGs, surface-to-surface missiles. There are ve types of missiles, and I would just guess off-hand, although we have no real basis to make such an estimate, that they have at least a thousand in total in these ve categories. Interestingly, the main concern was not about the military implications of those missiles but about the lack of public understanding that the missiles were not strategic threats. As Bundy put it, the issue was not a security problem but a public relations problem: It really would blanket a lot of this chatter if we could get clear that there are a lot of missiles which are not the missiles weve had on our mind.83 The practical outcome of the discussion was that despite lingering uncertainty about the presence of warheads, the ExComm had decided that the United States should not pay any further price for the removal of the thousands of Soviet combat troops, the 24 to 36 Lunas that intelligence agencies estimated were in Cuba, the 500 SAMs, the MiG-21 ghters, or the 150 coastal defense missiles. Having made that decision, the administration had to defend it on several fronts, confronting tensions between diplomatic expediency, military imperatives, and domestic politics. The JCS remained uncomfortable about the prospect that Soviet troops and their Lunas would remain in Cuba. The Joint Chiefs did not deny that such short-range weapons were primarily defensive forces, but the Chiefs were
83. The ofcial memorandum of this conversation recorded that the President commented that the Russians wont take out their ground forces until we give a no-invasion assurance. It is better for us to have the Soviet units in Cuba than to give a formal no-invasion assurance. Memorandum of Conversation, 29 November 1962, in FRUS, 19611963, Vol. XI, p. 541. See also Allan Evans to Roger Hilsman, 29 November 1962, in CubaExComm Notes 10/6212/62, Box 1, Roger Hilsman Papers, JFKL.

27

Coleman

responsible for developing contingency plans to counter Cuban and Soviet defenses in the event that the president ordered military action against Cuba. The JCS continued to plan for this contingency; and the U.S. military buildup that accompanied these plans reached a maximum state of readiness on 15 November.84 The JCS also had to provide for the defense of the U.S. naval base at Guantnamo. Short-range missiles hidden in the mountains near Guantnamo and pointed at the base had attracted the Joint Chiefs attention. Subsequent information suggests that they were right to be concerned: those missiles were probably nuclear-armed FKRs.85 The Lunas also remained a priority for the Joint Chiefs. They repeatedly raised the issue when they met with the president from November 1962 to February 1963, arguing that the weapons were a threat to Guantnamo and posed a risk that the Cubans could export them to Communist subversives elsewhere in Latin America.86 To gather more information about the weapons that were still in Cuba, the JCS persistently called for the resumption of lowlevel surveillance ights. General Taylor argued on 15 January, There are still very interesting, important points which were losing touch with, particularly the heavy ground equipment, whether or not they have indeed taken out the nuclear-capable FROGs.87 But the president and most of his advisers (Rusk was a notable exception) considered low-level ights too risky under the circumstances. No good options were available if another U-2 were shot down over Cuba. The request to resume low-level surveillance ights was therefore denied. Underlying all this discussion was the inevitable uncertainty about the presence of nuclear warheads. The Soviet Union had offered assurances that all nuclear warheads had been removed well before they actually were. When the British ambassador to the Soviet Union, Sir Frank Roberts, paid a farewell call to Khrushchev at the Kremlin on 12 November, the Soviet leader assured him that all nuclear warheads were gone from Cuba.88 During meetings in New York on 13, 15, and 18 November, Vasilii Kuznetsov gave further assurances that no nuclear weapons whatsoever were any longer on the territory of
84. JCS to Kennedy, 16 November 1962, in FRUS, 19611963, Vol. XI, pp. 472474. 85. Raymond L. Garthoff, New Evidence on the Cuban Missile Crisis, p. 252; and Talking Paper for the Joint Chiefs of Staff, for the Meeting with the President, 16 November 1962, in 091 Cuba (1962) Cuba CrisisMisc.Gen. Taylor File, Box 6, Maxwell Taylor Papers, RG 218, NA. 86. Talking Paper for the Joint Chiefs of Staff, for the Meeting with the President, 16 November 1962; Meeting with the JCS, 16 November 1962, Coleman and Naftali, eds., Presidential Recordings, Vol. 4 (the audio recording is on Tape 60, Presidential Recordings Collection, POF, JFKL); and Meeting with the JCS, 15 January 1963, in Coleman and Selverstone, eds., Presidential Recordings, Vol. 5 (the audio recording is on Tape 69, Presidential Recordings Collection, POF, JFKL). 87. Meeting with the JCS, 15 January 1963. 88. Roberts to Foreign Ofce, 12 November 1962, in PREM 11/3996: 83827, PRO, TNA.

28

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

Cuba.89 On 20 November, Khrushchev personally offered another assurance to Kennedy that [a]ll the nuclear weapons have been taken away from Cuba.90 None of those statements was true. The tactical nuclear warheads did not actually leave Cuba until the following month. In fact, the Soviet commander in Cuba, Army-General Issa Pliev, did not receive orders to return all of the warheads until 20 November. The decision not to leave any tactical warheads in Cuba was a policy reversal. The available evidence from Russian archives suggests that the plan had originally been to hand over the Luna rockets to the Cubans.91 U.S. military intelligence suspected as much. Such a step would have been in keeping with the recent policy of supplying Warsaw Pact countries with tactical weapon carriers and launchers. The Soviet Union was providing launchers and missiles to the East European countries but was keeping the warheads under tight Soviet control. In times of heightened alert, special Soviet commando units would deliver the warheads to the missiles and supervise their deployment and use.92 The Soviet Defense Ministry had also apparently been operating under the assumption that the Lunas and other battleeld weapons would be handed over to the Cubans.93 In early November, Soviet Defense Minister Rodion Malinovksii informed General Pliev in Havana that the Lunas and coastal defense missiles with conventional payloads would likely remain in Cuba, but that no nal decision on the nuclear warheads for these weapons had been made.94 The factor that apparently changed Khrushchevs mind was Castros seemingly erratic behavior in late October through mid-November. Khrushchev had dispatched Mikoyan to Havana to reason with Castro and encourage him to relax his hard line on the settlement of the crisis. Khrushchevs unilateral decision to withdraw the missiles had infuriated the Cuban leader.
89. A. Akalovksy to Brubeck, Memorandum of Conversation between Stevenson, McCloy, and Kuznetsov, 13 November 1962, in Cuba 11/14/62, Box 77, Cleveland Papers, JFKL; Stevenson to Rusk, 15 November 1962, in FRUS, 19611963, Vol. XI, p. 464; Bundy to ExComm, 16 November 1962, in FRUS, 19611963, Vol. XI, p. 468; and McCloy to Rusk, 19 November 1962, in FRUS, 19611963, Vol. XI, pp. 48687. 90. Khrushchev to Kennedy, 20 November 1962, in FRUS, 19611963, Vol. XI, p. 496; and Public Papers of the Presidents, John F. Kennedy, 1962, pp. 830838. 91. Fursenko and Naftali, One Hell of a Gamble; and Fursenko and Naftali, Khrushchevs Cold War. 92. Matthias Uhl, Nuclear Warhead Delivery Systems for the Warsaw Pact, 196165: Documents from the Russian State Archives of Economics and the German Federal Military Archives of the Reorganization and Modernization of the Armed Services of the Soviet Bloc, Parallel History Project on NATO and the Warsaw Pact Paper, September 2002; Matthias Uhl, Storming On to Paris: The 1961 Buria Exercise and the Planned Solution of the Berlin Crisis, in Vojtech Mastny, Sven G. Holtsmark, and Andreas Wenger, eds., War Plans and Alliances in the Cold War: Threat Perceptions in the East and West (London: Routledge, 2006), pp. 5862. 93. Fursenko and Naftali, Khrushchevs Cold War, p. 472. 94. Malinovskii to Pliev, early November 1962, translated and reproduced in Savranskaya, Tactical Nuclear Weapons in Cuba, pp. 385398.

29

Coleman

Castro later claimed that if Khrushchev had come to Havana instead of Mikoyan, I would have boxed his ears.95 Even with Mikoyan, the discussions had been difcult, and Mikoyan told Khrushchev that Castros famed revolutionary passion was becoming worryingly apocalyptic. Mikoyan had tried to reassure the Cuban leader that Moscow would provide the means for Cuba to defend itself but would not embark on nuclear war with the United States. Castros response was combative: Cuba cannot be conquered, it can only be destroyed.96 The ery rhetoric of Cuban Deputy Premier Che Guevara did not help. Talking to reporters on 28 November, in the rst interview given by a Cuban leader since the crisis began, Guevara boasted that not only would Cubans ght to the end if the United States attacked, but if the rockets had remained, we would have used them all and directed them against the very heart of the United States including New York.97 When Khrushchev learned on 15 November that Castro had issued orders to shoot down U.S. surveillance planes, he made the decision to bring all the tactical nuclear warheads back to the Soviet Union, although some Luna rockets and FKRs would remain without nuclear payloads. The instructions were sent to Pliev ve days later.98 Unaware of this order, the Cubans remained hopeful that the islands defense still included a nuclear dimension. Soviet ofcials intercepted a communication from the Cuban foreign minister, Ral Roa, to the Cuban representative at the United Nations, Carlos Lechuga, on 20 November asserting that we still have tactical atomic weapons, which must be kept.99 Decades later, Castro claimed that he had known about the Luna nuclear warheads in Cuba and had been condent that they would be used.100 The warheads did not actually leave Cuba until sometime in December, but the precise date and the ship on which they left remain in dispute. Some sources claim that the warheads left Havana on 25 December.101 Others say that they left on 1 or 4 December.102 U.S. ofcials were unaware of Khrush95. United Press International, Would Have Boxed Ks Ears, Castro Says, The Washington Post, 22 March 1963, p. A14. 96. Blight and Brenner, Sad and Luminous Days, p. 80. See also Vladislav M. Zubok, Dismayed by the Actions of the Soviet Union: Mikoyans Talks with Castro and the Cuban Leadership, November 1962, Cold War International History Project Bulletin, Issue No. 5 (Spring 1995), pp. 59, 8992; Mikoyans Mission to Havana: Cuba-Soviet Negotiations, November 1962, Cold War International History Project Bulletin, Issue No. 5 (Spring 1995), pp. 93109, 159. 97. JCS Situation Report 7633, 30 November 1962, in Cuba, Cables 11/30/6212/31/62 [Folder 2 of 2], Box 43, NSF, JFKL. 98. Fursenko and Naftali, Khrushchevs Cold War, p. 503. 99. Fursenko and Naftali, One Hell of a Gamble, p. 311. 100. Blight, Allyn, and Welch, Cuba on the Brink, p. 249. 101. Fursenko and Naftali, One Hell of a Gamble, p. 315. 102. Garthoff, U.S. Intelligence in the Cuban Missile Crisis, p. 61 n. 52; A. I. Gribkov, ed., On the Brink of the Nuclear Abyss, trans. by Svetlana Savranskaya (Moscow: Gregory Page, 1998).

30

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

chevs order and the movement of the warheads, although U.S. aerial surveillance ights photographed ten Luna missile transporters on a pier at Mariel on 26 December. Those transporters were apparently loaded onto the Kislovodsk and shipped to the USSR on 5 January. After this shipment, U.S. intelligence continued to observe Luna transporters and launchers at Soviet encampments in Cuba.103

The Season of Post-Mortems

The post-mortem season is in full swing, McGeorge Bundy observed on 30 October.104 The diplomatic settlement of the crisis played out against a backdrop of intense partisanship in domestic politics. Several Republicans had anticipated using the Cuba issue to attack the Democrats and the administration in the lead-up to the 1962 midterm election. Although the Cuba issue seems to have had little impact on how people actually voted, Republican hawks felt they had been cheated at the eleventh hour.105 As a result, they resumed their attacks in November 1962 and kept up the campaign for the next several months. They denounced what they claimed was awed intelligence and poor decision-making and also accused the administration of distorting intelligence to t an election-year political agenda and letting political concerns create a photo gap in surveillance of the military buildup in Cuba.106 As late as April 1963, some political critics cited the continued presence of Soviet troops and their nuclear-capable armaments in Cuba as evidence that the administration was appeasing Soviet leaders by allowing them to keep forces in Cuba that could potentially inict Hiroshima-like devastation upon Florida. The administration was under re from congressional critics, the press, and Cuban migrs agitating for a U.S. invasion of Cuba to oust Castro. From November 1962 through February 1963, Cuba was one of the thorniest issues confronting the administration.
103. JCS, Chronology of JCS Decisions Concerning the Cuban Crisis, pp. 79, 82, 8889, 91; Hilsman to Rusk, Review of Recent Developments in Cuba, 23 January 1963, in Cuba, Subjects, Intelligence Materials, 1/63, NSF, JFKL. In March 1964, an intelligence source indicated that there were nuclear warheads of some type in Cuba in October 1962, but they were removed. See Richard Helms to the Director of Central Intelligence, The Soviet Military Presence in Cuba, 17 March 1964, in Country Files, Latin America, Cuba, Intelligence ,Vol. 1, Box 24, NSF, LBJL. 104. 13th Meeting of the ExComm, 30 October 1962, Tape 44, Presidential Recordings Collection, POF, JFKL. 105. Thomas G. Paterson and William J. Brophy, October Missiles and November Elections: The Cuban Missile Crisis and American Politics, 1962, Journal of American History, Vol. 73, No. 1 (June 1986), pp. 87119. 106. On the photo gap, see Max Holland, The Photo Gap That Delayed Discovery of Missiles in Cuba, Studies in Intelligence, Vol. 49, No. 4 (2005), pp. 1530.

31

Coleman

With the press eager to thwart the administrations control of information, disclosures from Cuban refugees were published that, though often inaccurate, kept the administration on the defensive. Cuban migr groups in Miami, encouraged by Senators Barry Goldwater, Kenneth Keating, and Strom Thurmond, and Representative Donald Bruce, spread rumors that the Soviet Union had sent at least 80 long-range missiles to Cubanot the 42 Khrushchev had claimed and a gure the administration had acceptedand that dozens of long-range nuclear missiles were being hidden underground in the caves that riddled Cuba.107 Kennedy had anticipated the problem but was unable to prevent it. On 29 October he told his advisers that in the absence of detailed photographic evidence of the missiles leaving Cuba, everybodys going to be running around next week saying in the press Well, how do you know theyve left?108 Within days his prediction was borne out. Newspapers published detailed maps showing where the weapons were supposedly being stored, accompanied by streams of reports based on information attributed to reliable sources recently arrived from Cuba, usually a euphemism for what Arthur Schlesinger later characterized as anti-Castro zealots.109 Proving beyond all doubt that these refugee reports were wrong and that missiles were not in Cuba was an impossible task. The notion that Soviet ofcials were unlikely to lie about the number of missiles they had sent and the number they had removed was not at all convincing after the recent dramatic demonstrations of Soviet duplicity which the administration had trumpeted for its own purposes. Even the CIA and the Defense Intelligence Agency (DIA) cautioned that Soviet statements regarding nuclear warheads cannot be taken at face value.110 Some press reports proved remarkably accurate. As Khrushchev and Kennedy reached agreement on the withdrawal of the Il-28 bombers and the lifting of the quarantine on 20 November, correspondent Marguerite Higginss account of Lunas in Cuba appeared on the front page of The New York Herald-Tribune under the headline: More Cuba Missiles: We Reveal Castro Has A[tomic]-Artillery. In her article, Higgins described the capabilities of the Lunas and said that although the United States had no strong evidence of either the presence or absence of nuclear warheads in Cuba, the presence of such missiles is taken to indicate that the warheads that give them their punch
107. Report: Cuba Hides Missiles Underground, The Chicago Daily Tribune, 7 November 1962, p. A12. 108. 12th Meeting of the ExComm, 29 October 1962, in Coleman and Naftali, eds., Presidential Recordings, Vol. 4. 109. Arthur M. Schlesinger, Jr., Robert Kennedy and His Times (Boston: Mariner, 2002), p. 539. 110. CIA/DIA, Cuba 1962: Khrushchevs Miscalculated Risk, in Box 35, NSF, LBJL.

32

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

are probably on the island, too.111 Higgins did not reveal her sources, but Soviet documents conrm that her supposition on the presence of warheads was correct. At the time, however, the report was overshadowed by Kennedys press conference later that day announcing the Soviet agreement to withdraw the Il-28s and the lifting of the U.S. naval quarantine. Even after the deal of 20 November, Kennedy continued to face domestic political criticism, which built through January and early February 1963. Amid partisan attacks and internal disagreements that threatened to spill out into the open, Kennedy regained the initiative by authorizing a publicity campaign that included the disclosure of an unprecedented amount of detailed intelligence. This campaign was ultimately successful in silencing the critics and rescuing the generally favorable impression of Kennedys leadership that many people retained after the missile crisis. The presidents own performance during the crisis resonated well with the American public. A Gallup poll in the wake of the crisis showed Kennedys approval rating at 74 percent.112 But the surprise outbreak of the crisis was widely seen as the result of a devastating intelligence failure. Many commentators questioned the competence of U.S. intelligence agencies and of the Kennedy administration, especially after the Bay of Pigs asco. Additional reports about new and unidentied arms shipments to Cuba and indications of round-the-clock work at Cuban military sites were unsettling. To make matters worse, the administration could not give a rm answer to such seemingly simple questions as how many Soviet troops had been in Cuba at the height of the crisis and how many remained. Having announced that the number of Soviet troops in Cuba peaked at around 17,000 during the crisis, ofcials had a hard time explaining why the gure remained unchanged for months after the crisis even after several announcements that thousands of Soviet troops had been withdrawn.113 The controversy became politically charged. During the summer and fall of 1962, polls indicated that Cuba was Kennedys most vulnerable point on foreign policy, something Republicans had hoped to exploit in the run-up to the November 1962 midterm elections, especially to take the focus away from Kennedys preferred topic, Medicare.114 On 13 September, Bundy told the
111. Marguerite Higgins, More Cuba Missiles: We Reveal Castro Has A-Artillery, New York Herald Tribune, 20 November 1962, p. 1. 112. Time, Vol. 80, No. 24, 14 December 1962, p. 20. 113. Post-crisis estimates retroactively judged that some 22,000 Soviet troops had been in Cuba during the crisis. The actual number was later conrmed to be more than 42,000. 114. Paterson and Brophy, October Missiles and November Elections, p. 94; and Freedman, Kennedys Wars, p. 161.

33

Coleman

president that [t]he Congressional head of steam on this [Cuba] is the most serious that we have had.115 During the crisis itself, even the administrations harshest critics had rallied behind the ag. Kennedy will have the 100 percent backing of every American regardless of party, Senator Keating had said.116 After the crisis, this public consensus dissipated almost immediately. Having been robbed of what seemed like a national security trump card before voters went to the polls, critics staged a series of congressional inquiries into a range of issues involving Cuba. The Kennedy administration generally stuck to a tight ofcial script, which was notably candid about the forces remaining on the island.117 The accusations from critics varied but generally involved the notion that the administration was being too soft on Cuba. Keating and others accused the administration of negligence in reducing aerial surveillance in September and early October and alleged that the White House had deliberately created its own October surprise to steal the 6 November election; that the administration was being duped by Khrushchev and Castro, and that long-range missiles were being hidden in caves in Cuba.118 Of these accusations, the last proved the most persistent, frustrating, and difcult to rebut largely because proving the negativethat missiles were not being hidden in caveswas impossible. One of the most signicant operational ndings of internal postmortems of intelligence was an upgraded assessment of the Soviet combat troops in Cuba, which were revealed to be more tightly organized, more numerous, and better armed than had previously been recognized.119 This was a politically sensitive nding. As the circumstances leading to the crisis unfolded in the summer and fall of 1962, the White House had insisted that Soviet personnel in Cuba were technicians, whereas Republican critics had called them troops.120 After the crisis, intelligence analysts determined that four Soviet combat groups were spread across the island, each of which resembled Soviet motorized rie regiments. Three were stationed near former strategic missile bases (Santiago de las Vegas and Artemisa in western Cuba and Remedios in central Cuba), and the fourth was near the site of an important Il-28 and
115. Paterson and Brophy, October Missiles and November Elections, p. 96. 116. The Backdown, Time, Vol. 80, No. 18 (2 November 1962), p. 16. 117. See, for instance, CFP-DCSOPS-6, Cuba-Threat and Army Plans, 7 December 1962, in GEOG G Cuba 370.2 U.S. Forces, 228.01, U.S. Army Center for Military History. 118. See Holland, The Photo Gap That Delayed Discovery of Missiles in Cuba, pp. 1530. 119. SNIE 95-4-62, Castros Subversive Capabilities in Latin America, 9 November 1962, in 85: Cuba, NIEs, Box 9, NSF, LBJL. The Stennis Committee identied this late identication of the Soviet ground combat forces as an intelligence aw. See SCAS/PIS, The Cuban Military Buildup, p. 2. 120. Paterson and Brophy, October Missiles and November Elections, p. 95.

34

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

MiG aireld (Holguin in eastern Cuba). U.S. intelligence analysts estimated that each group consisted of roughly 1,250 men and had its own support structure along with an impressive array of weapons: assault guns, tanks, Snapper anti-tank missiles, and Lunas.121 The signicance of this last item was not lost on intelligence analysts. As the Ofce of Naval Intelligence put it: Analysis of recent photographic coverage of four major camps in Cuba suggests that they contain highly mobile composite Soviet Army ground combat forces of regimental size, with a possible nuclear capability.122 The composition of these forces did not conform precisely to known Soviet formations, but seemed to reect innovations observed in recent deployments in East Germany aimed at modernizing Soviet forces by creating reinforced regimental formations that could undertake independent operations.123 Analysts concluded that the original mission of these forces was to provide local ground defense against either a U.S. invasion or Cuban sabotage to protect the MRBM and IRBM sites. But even as the MRBM and IRBM sites were being bulldozed, the construction of barracks suggested that Soviet leaders had no intention of withdrawing those troops.124 On a number of occasions through January and February, U.S. ofcials testifying before congressional committees laid out detailed information about the remaining short-range missiles and the remaining Soviet combat troops, and again said that they expected Moscow to withdraw the troops. The ofcials were candid about nuclear issues. They drew analogies between Lunas/Frogs and Honest John missiles and referred to the Soviet armaments as advanced-type tactical rocket launchers, anti-personnel weapons with a
121. Soviet Forces in Cuba: Annex A, 5 February 1963, in Cuba, Subjects, Intelligence Materials, 2/63, Box 51, NSF, JFKL; DIA, Round-up of Signicant World Events, 23 November 1962, Item No. CC02562, in Cuban Missile Crisis Collection, National Security Archive; and Ofce of Naval Intelligence, Intelligence Briefs, ONI Review, Vol. 18, No. 1 (January 1963), p. 557. 122. Ofce of Naval Intelligence, Intelligence Briefs, ONI Review, Vol. 18, No. 1 (January 1963), p. 556. 123. NIE 11-14-62, Capabilities of the Soviet Theater Forces, 5 December 1962, in 1162 USSR, Box 2, NSF-NIEs, LBJL. 124. CIA, Phasing of the Soviet Military Deployment to Cuba, 28 November 1962, in Cuba, Subjects, CIA Memos, 11/20/6211/29/62, Box 46A, NSF, JFKL; CIA, The Crisis: U.S.S.R./Cuba, 29 November 1962, in Cuba, Subjects, CIA Memos, 11/20/6211/29/62, Box 46A, NSF, JFKL; CIA, Soviet Presence in Cuba, 7 December 1962, in Cuba, Subjects, CIA Memoranda, 12/1/623/26/ 63, Box 47, NSF, JFKL; CIA, The U.S.S.R.s Intentions with Respect to its Military Presence in Cuba, 13 December 1962, in Cuba, Subjects, CIA Memos, 11/20/6211/29/62, Box 47, NSF, JFKL; and CIA, The Cuban Situation, 16 December 1962, in Cuba, Subjects, CIA Memos, 11/20/ 6211/29/62, Box 46A, NSF, JFKL; and CIA, The Cuban Situation, 18 December 1962, in Cuba, Subjects, CIA Memos, 11/20/6211/29/62,NSF, JFKL. This assessment continued through the 1960s based on such evidence as the arrival in Cuba of some wives of Soviet personnel. CIA, Weekly Cuban Summary, 3 February 1965, in CIA Daily and Weekly Summaries , Vol. 1 (12/642/65), Box 36, Cuba, Country, NSF, LBJL.

35

Coleman

range of about 25 miles.125 Senator Thurmond cited the presence of nuclear-tipped FROG missiles among many examples of what he charged were the Kennedy administrations failed policies in Cuba, but few others in Congress seemed interested in following up on the conrmed presence of Soviet tactical missiles in Cuba.126 Instead, committee members focused on the possible presence of long-range strategic missiles in Cuba and the arming of MiG-21s with nuclear weaponsa less likely but more sensational scenario. That latter prospect, warned Frank Lausche (D-Ohio), raised the possibility that MiG jets could be used against Florida just as we dropped atomic bombs on Hiroshima.127 The political pressure reached a crisis point for the administration in early February 1963. Word had leaked out that McCone, a Republican, had been warning since September that the Soviet Union might be installing MRBMs in Cuba. Congressional Republicans jumped on apparent inconsistencies between what McCone was saying and what other members of the administration with more dovish reputations were saying. Kennedy and the ExComm decided to end the speculation once and for all. Perhaps drawing lessons from the missile gap myth of three years earlier, which several senior administration ofcials, including Kennedy himself, had helped to perpetuate, the president endorsed McNamaras proposal for greater transparency by laying out in a public brieng what had been withdrawn from Cuba and what was still there.128 This tactical maneuver was not without its risks. As John Norris of The Washington Post observed immediately after McNamaras briefing, he may have helped convince millions of Americans unversed in the complexities of the situation that there is a powerful Russian bastion 90 miles off the US coast. By highlighting some of the other sophisticated, nuclearcapable weaponry still in Cuba (the MiG-21s and the armored battalions with their Luna missiles), the brieng could blur in the public mind the distinction between offensive and defensive weapons.129 Despite this risk, McNamara went ahead with the plan on 6 February by staging an elaborate, nationally televised special brieng on Cuba from the State Departments auditorium.
125. McCones Testimony to the House Defense Appropriations Subcommittee, 6 February 1963, in Box 26, McNamara Papers, RG 200, NA. 126. Strom Thurmond, Weekly Newsletter, 1 February 1963, in Background Materials III 2/7/63, Box 58, POF, JFKL. 127. U.S. Congress, Senate, Committee on Foreign Relations, Hearing, 16 January 1963, in SFRC, ESSFRC, 1963, p. 66. 128. ExComm Meeting, 5 February 1963, in Coleman and Selverstone, eds., Presidential Recordings, Vol. 5. The audio recording is on Tape 72A, in Presidential Recordings Collection, POF, JFKL. 129. John G. Norris, Administration Case on Cuba Not Air-Tight, The Washington Post, 8 February 1963, p. A11.

36

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

As the New York Times accurately characterized it the next day, the performance was a hard sell effort designed to head off domestic political criticism and to prevent Republican critics from hijacking the national security agenda.130 The nature of the brieng was, in many ways, unprecedented. It included dozens of surveillance photographs and detailed information about the capabilities of the military equipment and forces in Cuba.131 McNamara and John Hughes from the Defense Intelligence Agency went into a level of detail that had previously been reserved only for classied briengs to congressional committees. McNamara said publicly that the issue was so important that he was to disclose the information even at the risk of degrading our intelligence collection capabilities.132 Nevertheless, the CIA, probably piqued at the starring role being played by the Pentagons new intelligence agency, criticized the DIA for disclosing too much about intelligence sources and methods and for being sloppy with factual information.133 McNamara and Hughes addressed in detail the issue of the Soviet combat forces that remained. They provided specic numbers of Soviet and Cuban weaponry along with graphic high- and low-level surveillance photos. They showed photographs of Luna missile launchers, coastal defense missile installations, and most of the other weapons systems for which they had veriable surveillance photographs. In the question and answer period at the conclusion of the press brieng, a reporter asked McNamara, Do the FROG missiles, some of which are still in Cuba, have nuclear capability? McNamara responded, The FROGS are almost certainly capable of nuclear and nonnuclear re.134 He provided no further elaboration or qualication before moving on to the next question. In a press conference the following day, Kennedy addressed the issue of the thousands of Soviet combat troops who remained in Cuba and said that he was still trying to draw from Khrushchev an assurance that these forces would be withdrawn in due course. We do not view that threat lightly, he warned. At the same time, he expressed frustration with the growing chorus of congressional and press criticism and urged
130. Tom Wicker, McNamara Insists Offensive Arms Are Out of Cuba, The New York Times, 7 February 1963, p. 1. 131. The full, verbatim transcript of the brieng is available in Cuba, 196263, Box 26, McNamara Papers, RG 200, NA. 132. Wicker, McNamara Insists Offensive Arms Are Out of Cuba, p. 1. 133. CIA, Disclosures of U.S. Intelligence Methodology, 29 March 1963, in Cuba, Subjects, Intelligence Materials 3/63, NSF, JFKL; and CIA, Additional Points in Secretary McNamaras Presentation Not Directly Discussed by Mr. McCone, n.d., in Cuba, Subjects, Testimony, 5/7/622/27/63, Box 61, NSF, JFKL. 134. Transcript of the McNamara/Hughes Press Brieng, 6 February 1963, in Cuba 196263, Box 26, McNamara Papers, RG 200, NA.

37

Coleman

members of those groups to keep a sense of proportion of what we are talking about.135 Ultimately, the administrations new tactic of transparency proved remarkably successful, but not before critics mounted another attack. Five leading congressional critics of the administrations Cuba policyKeating, Thurmond, Rep. William C. Cramer, Rep. H. R. Gross, and Rep. Armistead Selden, Jr.responded to McNamaras brieng by staging their own televised news conference accusing McNamara of deliberately playing down the implications of the continuing Soviet military presence in Cuba. They said that even the so-called defensive forces remaining could be a menace by enabling the Soviet Union to bring in offensive weapons furtively. Each legislator stressed, however, that he was not advocating direct military action.136 In April, with Soviet troops still in Cuba, former Vice President Richard Nixon attacked the administration for allowing a Soviet beachhead in Cuba and giving the Soviet Union squatters rights in our own backyard. This allegation echoed Keatings pre-crisis charges that the White House had adopted a look-the-other-way policy regarding the Soviet military buildup in Cuba.137 The administration faced two nal political challenges. The rst was to explain the apparent inconsistencies between the public statements of McNamara and McCone. One of the key points of contention was the signicance of the remaining Soviet ground forces. McNamara was considerably more optimistic than McCone. The CIA took issue with McNamaras public suggestion that the deployment of Luna rockets was diminishing in size: We think Mr. McNamara is basing his lessening on a very slender reed.138 Bundy met with McNamara, McCone, and Rusk at the White House on 19 February to try to smooth over the differences and establish a set of guidelines for discussing the Cuba issue. According to Bundys summary of the meeting, they agreed that apparent discrepancies regarding some of the most provocative military forces (MiGs, submarine bases, and Lunas) should
135. Brieng Paper for the Presidents Press Conference, 20 November 1962, in Presidential Press Conf (TV) 20 Nov 62 [Folder 2 of 2], Box 119, Pierre Salinger Papers, JFKL; and Brieng Paper for the Presidents Press Conference, 12 December 1962, in Presidential Press Conf. 12 Nov 62 [Folder 1 of 2], Box 119, Pierre Salinger Papers, JFKL. 136. John G. Norris, Hill Critics Ask Tough Cuba Stand, The Washington Post, 9 February 1963, p. A1. 137. Excerpts from Address from Nixon Calling for a Stronger U.S. Foreign Policy, The New York Times, 21 April 1963, p. 62; Max Frankel, Nixon Proposes Freedom Policy to Conquer Reds, The New York Times, 21 April 1963, p. 1; and Schlesinger, Robert Kennedy and His Times, pp. 539540. 138. CIA, Removal of FROGs, n.d., in Cuba, Subjects, Testimony, 5/7/622/27/63, Box 61, NSF, JFKL.

38

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

be explained as differences in personal opinion rather than any fundamental disagreement.139 The second challenge was to defuse criticism from Senator Keating and others. To assuage Keating, McCone agreed to meet with the senator so that Keating would have the opportunity to convey the information he claimed to have.140 When they met, Keating insisted that more missiles and troops remained in Cuba than the administration had acknowledged.141 Keating also told McCone that nuclear warheads for Lunas were being hidden in caves, a charge that government experts rebutted by pointing out that such an arrangement would be unduly hazardous and inconvenient when more suitable storage bunkers were available.142 Although the Cuba issue did not completely disappear, this wave of criticism proved to be the last sustained congressional attack on the Kennedy administrations handling of the missile crisis. Despite Congressman Seldens efforts to keep the Cuba debate alive by having the House Foreign Affairs Subcommittee on Latin America hold hearings, Congress generally moved on to other issues.143 The Stennis Committee, the most inuential of the congressional inquiries into Cuba-related issues, issued a mildly critical report in early May 1963 before moving on to the what now? question.144

The Nuclear Presumption

Soviet archival materials and other sources that have become available since the end of the Cold War have revealed that in addition to the warheads for the MRBMs, twelve nuclear warheads for the Luna rockets, six nuclear gravity bombs for the Il-28 bombers, 80 nuclear warheads for the FKR cruise mis139. Bundy to Rusk, McNamara, and McCone, 19 February 1963, in FRUS, 1961-11963, Vol. 11XI, p. 705. 140. Associated Press, Keating Reveals His Data on Cuba, The New York Times, 9 February 1963, p. 4; and William Moore, CIA Gets Cuba Brieng, The Chicago Daily Tribune, 9 February 1963, p. W1. 141. U.S. intelligence had conrmed through radio intercepts that a shipment containing a small amount of radioactive isotopes had arrived in Cuba aboard the Michurinsk in mid-January. See CIA, Comments on Statements Made by Senator Keating on February 8, 20 February 1963, in Cuba, Subjects, Senator Kenneth B. Keating, Statements, 11/624/63, NSF, JFKL. 142. Ibid. 143. Frederick G. Dutton to McGeorge Bundy and Larry OBrien, 12 February 1962, in Cuba, Subjects, Testimony, 5/7/622/27/63, Box 61, NSF, JFKL. 144. Memorandum of Telephone Conversation between George Ball and Cong. Mahon, 11 February 1963, 10 a.m., in Cuba 1/5/6310/31/63, Box 3, George Ball Papers, JFKL; and SCAS/PIS, The Cuban Military Buildup.

39

Coleman

siles, and four nuclear mines were present in Cuba.145 Although U.S. intelligence ofcials did not know for sure that the Soviet Union had delivered nuclear warheads, they never argued that the Soviet Union had not delivered them. U.S. policymakers did know, as of 26 October, that in addition to the MRBMs, the Soviet Union had deployed nuclear-capable battleeld rockets in Cuba. The crucial question remains: in the minds of Washingtons policymakers, did nuclear-capable mean nuclear? The controversy in recent years over whether the Lunas and other tactical nuclear-capable missiles were a known risk of the Cuban missile crisis implies in part that the U.S. intelligence community failed to detect the tactical nuclear missiles in Cuba. U.S. intelligence during the crisis was certainly imperfect: The CIA underestimated by about half the number of Soviet troops in Cuba, and the agency detected only 33 of the 42 MRBMs in Cuba. Moreover, the much-contested photo gap in the fall of 1962arguably more a political failure than an intelligence oneimposed limits on the photographic surveillance that might have detected the long-range missiles sooner.146 But apart from the misidentication of the crated FKRs, intelligence reporting on the tactical nuclear weapons proved relatively accurate within the known and acknowledged technical limits of the time. U.S. intelligence ofcials openly acknowledged that they could not prove one way or the other whether nuclear warheads of any kind were present in Cuba. This admission did not signal a failure of diligence or imagination; rather, the existing technical capabilities did not permit certainty on this matter. Some circumstantial indicatorsthe construction of nuclear storage bunkers, suspicious shipping, the presence of warhead transporters, Soviet statements, and even radiation readings taken from helicopters hovering over departing shipsdid exist, but all of these were, as the CIA put it, so ambiguous and inconclusive that it is not possible to reach a judgment based on factual information.147 Proving the negative that nuclear warheads were not in Cuba was even more difcult, as the CIA openly acknowledged in the realtime intelligence it provided to policymakers during the crisis and after. Faced with that uncertainty, policymakers, analysts, and military planners were obliged to assume that nuclear warheads were in Cuba.148 The CIA
145. Gribkov and Smith, Operation Anadyr, p. 5. 146. Garthoff, U.S. Intelligence in the Cuban Missile Crisis, pp. 1863. 147. CIA/DIA, Cuba 1962: Khrushchevs Miscalculated Risk, 13 February 1964, in Box 35, NSF, LBJL. See also Brugioni, Eyeball to Eyeball, pp. 538548. 148. Joint Evaluation of Soviet Missile Threat in Cuba, 19 October 1962, in McAuliffe, CIA Documents on the Cuban Missile Crisis, p. 205; and Supplement 8 to Joint Evaluation of Soviet Missile Threat in Cuba, 28 October 1962, in Countries, Latin America, CubaIntelligence Reports 1962 63, Vol. 2, Box 37, NSF, LBJL. For a contrary view, see Ray S. Cline, Commentary: The Cuban Missile Crisis, Foreign Affairs, Vol. 67, No. 2 (Fall 1989), pp. 134138.

40

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

and DIA warned repeatedly that even though they had no hard evidence that nuclear warheads were actually in Cuba, the only prudent course was to assume that they were.149 Policymakers accepted, internalized, and applied this principle to more than just the MRBMs. At the rst ExComm meeting after the MRBMs had been discovered, McNamara cautioned about the MiG-21s: If there are nuclear warheads associated with the launchers, you must assume there will be nuclear warheads associated with aircraft.150 Similar presumptions were expressed about the Il-28 bombers.151 During a 12 November ExComm discussion about whether pushing for some kind of formal agreement or statement about the warheads was feasible, Bundy remarked: But the question about warheads is whether he [Khrushchev] wontyou know, whether you need to bargain on that. You cant prove it in either case. You get a certain additional hazard for him in leaving any behind if he tells you he hasnt.152 In a closed session before the Senate Foreign Relations Committee in mid-January 1963, Dean Rusk explained:
We were intrigued by the fact that we never found a nose cone [or the MRBMs], never found an identiable nuclear weapon in Cuba. We have to assume that they were there, but they are not too difcult to conceal, maybe ve or six feet long. We never, however, actually identied them. It is possible that the quarantine intercepted the later shipments of this total effort, and that this might have included the nose cones, but we just dont know. We never found them.153

The following week, Rusk told the same committee that he could not say for sure whether nuclear warheads were still in Cuba.154 McNamara provided even more detail and partly contradicted Rusks statements in his 6 February 1963 televised brieng. Asked whether nuclear weapons had been identied or detected in Cuba, McNamara answered: The movement of nuclear warheads into Cuba I believe occurred. I believe we observed it in certain vehicles and we observed the movement of those vehicles out of Cuba, and we traced the shipment of those vehicles on ships back into the home waters of the So149. Joint Evaluation of Soviet Missile Threat in Cuba, 19 October 1962, in Countries, Latin America, CubaIntelligence Reports 196263, Vol. 2, Box 37, NSF, LBJL; and Supplement 8 to Joint Evaluation of Soviet Missile Threat in Cuba. 150. Naftali and Zelikow, eds., Presidential Recordings, Vol. 2, p. 410. 151. 22nd Meeting of the ExComm, 7 November 1962, in Coleman and Naftali, eds., Presidential Recordings, Vol. 4. The audio recording is on Tape 53A, Presidential Recordings Collection, POF, JFKL. 152. 25th Meeting of the ExComm, 12 November 1962, in Coleman and Naftali, eds., Presidential Recordings, Vol. 4. The audio recording is on Tape 56, Presidential Recordings Collection, POF, JFKL. 153. Rusks Testimony to the Senate Foreign Relations Committee, 11 January 1963, in SFRC, ESSFRC, 1963, p. 8. 154. Rusks Testimony to the Senate Foreign Relations Committee, 16 January 1963, in SFRC, ESSFRC, 1963, pp. 6667.

41

Coleman

viet Union.155 The level of condence implied in McNamaras answer was certainly atypical. Few other ofcials during or after the crisis were willing to say whether warheads had been in Cuba, and almost no one was willing to discuss the sources and methods of intelligence gathering in such specic terms. The intelligence basis for McNamaras statement (assuming it is accurate) remains unclear. More typically, ofcials hedged their statements, as Richard Helms, the CIAs director of operations, did in February 1963 when he told a congressional committee that although the CIA had found no evidence that Luna nuclear warheads were present in Cuba, he could not rule out that they were there.156 Raymond Garthoff has argued that although the potential nuclear role of the dual-capable IL-28 aircraft and Luna rockets had been known, few if any in Washington really believed there were tactical nuclear warheads in Cuba. He added: There was no evidence, and no presumption, that [the tactical missiles] were armed with nuclear weapons.157 Similarly, Robert McNamara said in October 2002 he was surprised to learn that Khrushchev not only had them [tactical nuclear weapons] there, but after we forced . . . him to agree to take out the missiles and the bombers, he didnt take out the tactical nuclear weapons. We didnt ask for it. We didnt know they were there.158 Although it is impossible to say for certain what McNamara and others may or may not have believed at the time, new evidence from declassied documents and presidential recordings suggests that in the case of the Lunas U.S. ofcials always made a strong nuclear presumption. Unlike with the coastal defense missiles, which U.S. analysts never accurately identied, the CIA had an accurate understanding of the capabilities and limitationsof the Lunas and also their role in Soviet military doctrine. U.S. ofcials knew that the Lunas, MiG-21s, and V-75s they had seen in Cuba were capable of delivering nuclear warheads.159 But analysts and policymakers all believed it was highly unlikely that the MiG-21s and SA-2s would be used for nuclear delivery. With the Lunas, however, their assumption evidently was the opposite. Given numerous opportunities to contest the claim that the Frogs in Cuba could be armed with nuclear warheads, U.S. ofcials never did. Nor did they say that such deployments were unlikely. The admin155. Transcript of Robert McNamaras Special Cuba Brieng, 6 February 1963, in Cuba, Subjects, Testimony, Secretary McNamara, 2/8/632/19/63, Box 63, NSF, JFKL. 156. Richard Helms Testimony to the House Defense Appropriations Subcommittee, February 1963, p. 144, Cuba 196263, Box 36, McNamara Papers, RG 200, NA. 157. Garthoff, The Havana Conference on the Cuban Missile Crisis, p. 2; and Garthoff, U.S. Intelligence in the Cuban Missile Crisis, pp. 29, 31. 158. Transcript, 13 Days: An Insiders Perspective, 1 October 2002, in JFKL. 159. A notable exception is discussed below.

42

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

istration apparently made no effort to counter Marguerite Higginss frontpage story alleging that Lunas and their nuclear warheads were still in Cuba. Nor did U.S. ofcials attempt to correct or even tone down the tendency of the press and Congress to describe the Lunas as nuclear-capable. Behind closed doors and in public, senior ofcials routinely described the Lunas as analogous to the U.S. Armys Honest John, a missile known primarily as a rst-generation tactical nuclear weapon. The available evidence shows that in only one instance did a U.S. ofcial categorically state that the Lunas in Cuba were not armed with nuclear warheads, and his assessment was awed. A memorandum dated 29 October 1962 addressed to George Ball and signed by U. Alexis Johnson but drafted by Raymond Garthoff discussed the coastal defense missiles, the Lunas, and the Komar torpedo boats. The memorandum judged the Komar torpedo boats the most susceptible of offensive employment and recommended that the U.S. fallback position should be to drop the demand for these three weapons systems to be removed from Cuba. For reasons that are unclear, the document inaccurately claimed that none of these three systems, incidentally, has a nuclear delivery capability.160 Policymakers had good reason to suspect that the Lunas might be armed with nuclear warheads. Lunas were so inaccurate that they had little utility as conventional weapons in combat.161 Because they followed an unguided ballistic trajectory and lacked in-ight targeting capabilities, they could not be relied on to hit a specic target. Missile accuracy is measured in terms of Circular Error Probable (CEP), a gure calculated by taking the radius of a circle centered on the target in which 50 percent of the missiles red are expected to land. Depending on the variety of Luna, U.S. analysts in 1962 judged that its CEP ranged from 270 to 1,500 meters (see Table 1). By comparison, Honest John missiles had a CEP of around 180 meters and Sopkas had a CEP of around 50 meters. With the relatively small blast zone of a conventional high explosive warhead, measured in tens of meters at most, this inaccuracy rendered Lunas of little value as conventional weapons. Indeed, they proved to be so when used with non-nuclear warheads in later conicts. But if equipped with nuclear warheads, they would pose a severe hazard to any invasion force.162
160. Alexis Johnson to George Ball, 29 October 1962, in Garthoff, Reections on the Cuban Missile Crisis, p. 212. 161. See also Allison and Zelikow, Essence of Decision, p. 246 n. 20. 162. Aaron Karp, Ballistic Missiles in the Third World, International Security, Vol. 9, No. 3 (Winter 1984/85), p. 169; and Polmar and Gresham, DEFCON-2, p. 312. For a description of the effects of a Luna rocket armed with a nuclear warhead, see Fursenko and Naftali, One Hell of a Gamble, p. 242.

43

Coleman

U.S. intelligence analysts also knew, at least in general terms, the role of Lunas and other battleeld nuclear weapons in Soviet military doctrine. Although the extent of what has been called the nuclear romanticism of Soviet military planning was not fully documented until decades later, Soviet military ofcers had made no secret of their increasing emphasis on nuclear forces in their war planning.163 Like the U.S. Armys Honest John rockets, with which the Lunas were so often compared, the dening characteristic of the Lunasand the very reason for their existence in the rst placewas that they introduced a battleeld nuclear capability to ground forces, thereby lling a gap between long-range strategic missiles and traditional battleeld artillery forces.164 Their strengths could be most effective against concentrations of troops on the moveas would have been the case with a U.S. invasion of Cuba.165 Perhaps the most intriguing evidence of all regarding what the Kennedy administration knew about the tactical nuclear warheads in Cuba was an exchange between President Kennedy, Marine Commandant David Shoup, and Chief of Naval Operations Admiral George Anderson on 29 October 1962, secretly captured by Kennedys White House recording system.
David Shoup: The question isthe $64 question is, whether they would use the tactical nuclear weapons President Kennedy: Nuclear weapons? Shoup: at that point, because they would deal bloody hell with Guantnamo, of course. If nuclear weapons start down there, Id say were at nuclear couldnt afford to let them do that. I mean theyre not [unclear]. President Kennedy: But my guess is, well, everybody sort of gures that, in extremis, that everybody would use nuclear weapons. The decision to use any
163. Vojtech Mastny and Malcolm Byrne, eds., A Cardboard Castle? An Inside History of the Warsaw Pact, 19551991 (Budapest: Central European University Press, 2005), pp. 2027; Uhl, Storming On to Paris, pp. 4671; and Petr Luk, War Plans from Stalin to Brezhnev: The Czechoslovak Pivot, in Mastny, Holtsmark, and Wenger, eds., War Plans and Alliances in the Cold War, pp. 7294. See also Raymond L. Garthoff, Soviet Strategy in the Nuclear Age, Rev. Ed. (New York: Praeger, 1962), p. 112. 164. Stephen M. Meyer, Soviet Theatre Nuclear Forces: Part II: Capabilities and Implications, Adelphi Paper No. 188 (London: IISS, 1983), p. 15; Stockholm International Peace Research Institute, Tactical Nuclear Weapons: European Perspectives (London: Taylor & Francis, 1978), pp. 134, 15152; Uhl, Storming On to Paris, pp. 5354; Cagle, History of the Basic (M31) Honest John Rocket System; and Robert S. Norris, Steven M. Kosiak, and Stephen I. Schwartz, Deploying the Bomb, in Stephen I. Schwartz, ed., Atomic Audit: The Costs and Consequences of U.S. Nuclear Weapons since 1940 (Washington, DC: Brookings Institution Press, 1998), p. 155. 165. U.S. intelligence at the time did not know that the rst Lunas delivered to the Warsaw Pact were deployed with Soviet forces in the German Democratic Republic, a deployment that was conducted simultaneously with Operation Anadyr; it began in May 1962 and was completed on 30 September 1962. During the Cuban missile crisis, those tactical nuclear missiles in East Germany apparently were placed on alert. See Uhl, Nuclear Warhead Delivery Systems for the Warsaw Pact; and Uhl, Storming On to Paris, p. 61.

44

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

kind of a nuclear weapon, even the tactical ones, presents such a risk of it getting out of control so quickly, that theres a . . .166

Although replete with unnished thoughts and allusions, the exchange suggests that Kennedy and the JCS took seriously the problem of tactical nuclear weapons in Cuba and were thinking hard about the grave risks those weapons would pose.

Conclusion
The risk of nuclear war during the Cuban missile crisis is inherently difcult to measure.167 The 1992 controversy about the presence of tactical nuclear weapons in Cuba in October 1962 convinced some that a crisis already widely regarded as the most perilous moment of the Cold War was even more dangerous than they had imagined. At rst ush, the revelations seemed to add a new element that had been missing from previous narratives of the missile crisis: When President Kennedy and his advisers were contemplating military action against Cuba, including a possible ground invasion, they did so without knowing for sure that Soviet tactical nuclear weapons were present in Cuba. The direct threat posed by the tactical nuclear missiles was far less than that posed by the MRBMs and IRBMs, which, unlike the tactical weapons, could strike targets in the continental United States. The real risk was that any use of tactical nuclear weapons in Cuba could have led to uncontrollable escalation culminating in global thermonuclear war. Roger Hilsman wrote in 1967: It is awesome to contemplate the situation of American ground and air forces attacking Soviet nuclear missiles poised on their pads and defended by Soviet ground combat forces equipped with tactical atomic weapons.168 Decades later, in the wake of the 1992 conference, George Ball reected with alarm that we might well have set off a nuclear exchange if the military options under discussion at the time had been adopted.169 However, the subsequent clarication that Khrushchev had not in fact delegated authority to use nuclear weapons to local Soviet commanders in Cuba meant that the danger of nuclear war during the crisis was not as high as
166. Off-the-Record Meeting on the Military Situation in Cuba, 29 October 1962, in Coleman and Naftali, eds., Presidential Recordings, Vol. 4. The audio recording of this meeting is on Tape 43, in Presidential Recordings Collection, POF, JFKL. 167. On assessing the nuclear risk of the MRBMs, see Marc Trachtenberg, The Inuence of Nuclear Weapons in the Cuban Missile Crisis, International Security, Vol. 10, No. 1 (Summer 1985), pp. 137163. 168. Hilsman, To Move a Nation, p. 227. 169. Ibid.

45

Coleman

some had momentarily feared.170 Furthermore, the Luna nuclear warheads were apparently stored more than 450 kilometers away from the delivery vehicles themselves.171 Contrary to hyperbolic claims that followed General Gribkovs initial remarks, Soviet tactical nuclear weapons were not on hairtrigger alert during the missile crisis. Nor could they have been put on alert at any subsequent stage unless Soviet command-and-control procedures had broken down or had been altered by leaders in Moscow. No hard evidence has come to light indicating whether the tactical nuclear missiles had a deterrent effect on President Kennedy. The available evidence suggests that the detection of Luna deployments in Cuba did not fundamentally change the advice the president received. Presumably referring to the strategic missiles, Kennedy told some of his senior advisers at a December 1962 defense budget discussion that what they [the USSR] had in Cuba alone would have been a substantial deterrent to me, a remarkable confession for a Cold War president to make even in private.172 Kennedy was neither oblivious nor cavalier when it came to the risks of tactical nuclear weapons, nor did he subscribe to the view expressed by Dwight Eisenhower almost a decade earlierwhen tactical nuclear weapons were just starting to be deployed widely by both sidesthat the bomb should be regarded as just another weapon. Eisenhower had famously declared: I see no reason why [nuclear weapons] shouldnt be used just exactly as you would use a bullet or anything else.173 Kennedy in 1959 rejected this view, arguing that small atomic weapons suffer from much the same handicaps as large atomic weapons insofar as no country could afford to come under even a limited nuclear attack without responding in kind. If we use them, the Russians use them, Kennedy argued, and inevitably, the use of small nuclear armaments will lead to larger and larger nuclear armaments on both sides, until the world-wide holocaust has begun.174 This position was in keeping with his oft-expressed skepticism that escalation of conict could be reliably controlled.175
170. Among those confused was Schlesinger, Four Days with Fidel, pp. 710. 171. Kramer, The Cuban Missile Crisis and Nuclear Proliferation, pp. 173174; Mark Kramer, Letter to the Editors, The New York Review of Books, Vol. 39, No. 10, 28 May 1992, p. 47; Blight and Welch, The Cuban Missile Crisis: On Historical Judgment and Inference, p. 174; and Fursenko and Naftali, Khrushchevs Cold War, p. 469. 172. David Coleman, Camelots Nuclear Conscience, Bulletin of the Atomic Scientists, Vol. 62, No. 3, (May/June 2006), p. 42. 173. Eisenhower Press Conference, 16 March 1954, in Public Papers of the Presidents: Dwight D. Eisenhower, 1954 (Washington, DC: U.S. Government Printing Ofce), p. 302. 174. Kennedy Speech at Lake Charles, Louisiana, 16 October 1959, in Allen Nevins, ed., The Strategy of Peace (New York: Harper, 1960), p. 185. 175. See Francis Gavin, The Myth of Flexible Response: American Strategy in Europe during the 1960s, International History Review, Vol. 23, No. 4 (December 2001), pp. 847875; and Lawrence

46

The Problem of Acceptable Risk in the Cuban Missile Crisis Settlement

Having reached the conclusion that Soviet troops in Cuba armed with dual-use tactical missiles was an acceptable risk, Kennedy did not push the Soviet Union harder to remove the Lunas and Soviet combat troops. The record suggests that after 28 October his policy was driven by his effort to deate the pressure from the political right and his more hawkish advisers, who insisted that he take a harder line, while also pushing ahead with his own desire to forge an accommodation with Khrushchev that would bring the crisis to a conclusion, a pattern that also seemed to characterize the secret deal to remove the Jupiter missiles in Turkey.176 U.S. intelligence analysts never fully lost sight of the Soviet troops or the Lunas even as these deployments faded as a priority after February 1963. Throughout the 1960sand presumably beyond, although much intelligence material remains classied for later periodsreports continued to mention dozens of nuclear-capable battleeld rockets, some 500600 surface-toair missiles, 150 coastal defense missiles, and a wide range of other advanced weaponry that remained in the Cuban arsenal.177 From time to time, hints came that Cubas defense was guaranteed by nuclear weapons, but the evidence was always ambiguous about the source of that protection (i.e., whether from tactical nuclear weapons or the deterrent value of Soviet strategic weapons).178 When Castro was interviewed by New York Times columnist C. L.
Freedman, The Evolution of Nuclear Strategy, 3rd ed. (Houndmills, UK: Palgrave Macmillan, 2003), pp. 196211. 176. Some observers have nevertheless argued that Kennedy missed important opportunities for even further accommodation with Cuba and the USSR. See, for instance, Thomas G. Paterson, Fixation with Cuba: The Bay of Pigs, Missile Crisis, and Covert War against Castro, in Thomas G. Paterson, ed., Kennedys Quest for Victory: American Foreign Policy, 19611963 (New York: Oxford University Press, 1989), p. 151. 177. For a sample of the reports circulated among U.S. policymakers that cataloged these weapons in considerable detail, see CIA, Soviet Forces in Cuba: Annex A, 5 February 1963; CIA, Current Status of Soviet Forces in Cuba, 25 March 1963, in Cuba, Subjects, CIA Memoranda, 12/1/63 3/26/63, Box 47, NSF, JFKL; DIA, Current Intelligence and Indications Center, n.d. (May 1963), in Countries, Cuba, Security 1963, Box 115, POF, JFKL; CIA, Soviet Forces in Cuba, 7 May 1963, in Cuba, Subjects, Intelligence Materials 5/63, Box 51A, NSF, JFKL; CIA, Extent of Soviet Presence in Cuba, 9 May 1963, in Cuba, Subjects, Intelligence Materials 5/63 Box 51A, NSF, JFKL; NIE 85-64 Situation and Prospects in Cuba, 5 August 1964, in Country Files, Latin America, Cuba, Intelligence Vol.1, Box 24, NSF, LBJL; CIA, Intelligence Handbook on Cuba, 1 January 1965, in Country, Cuba, Intelligence Vol. II, 1/654/65,, Box 24, NSF, LBJL; Directorate of Intelligence, Ofce of Current Intelligence, Weekly Cuban Summary, 24 February 1965, in Cuba, CIA Daily and Weekly Summaries Vol.2, 2/655/65, Box 36, NSF, LBJL; Brown to Director of CIA and Director of DIA, Joint CIA-DIA Team Report on the Military Posture in Cuba as it Pertains to Strategic Weapons, 19 April 1966, in Country, Latin America, Cuba, Strategic Missiles in Cuba, Box 20, NSF, LBJL; CIA, Current Intelligence Weekly Special Report, 30 September 1966, in LBJL, NSF, Box 19, Country, Latin America, Cuba, Bowdler, Vol. 2, Box 19, NSF, LBJL; and CFP-DCSOPS-6, Cuba-Threat and Army Plans, 7 December 1962. 178. Nikita Khrushchev, Excerpts from Khrushchevs Talk Pledging Aid to Cuba, The New York Times, 10 July 1960, p. 2; Tad Szulc, Cubans Cheered by Soviet Action, The New York Times, 10 July 1960, p. 3; Democrats Assail Khrushchev Threat, The New York Times, 10 July 1960, 3; and

47

Coleman

Sulzberger in November and December 1964, he insisted that [t]he presence here of both tactical and strategic weapons in the fall of 1962 was justied for Cubas defense.179 After February 1963, Kennedy never paid a political price for his decision to relax his demand for all Soviet combat troops to be withdrawn from Cuba. But a decade and a half later, one of his successors, Jimmy Carter, did pay such a price. After the early 1960s, the remaining Soviet troops had disappeared from public view. They suddenly reappeared in October 1979 when news broke that the CIA had discovered in Cuba a brigade of about 2,600 Soviet combat troops. This disclosure by Senator Frank Church, a Democrat involved in a tough reelection ght, created a political restorm for the Carter administration.180 At the height of that crisis, McGeorge Bundy wrote an oped article in The New York Times claiming responsibility for the presence of Soviet troops in Cuba and explaining that the seeds of the controversy lay in decisions taken in the aftermath of the missile crisis. Bundy wrote that after the offensive weapons were removed, the ongoing presence of Soviet troops in Cuba ceased to be a priority for the administration and the intelligence community. He noted that the Kennedy and Johnson administrations had known about the remaining Soviet ground forces, debated the nature of the risk, and decided that a few thousand Soviet troops in Cuba posed no meaningful threat to the United States or other countries in the region.181

Acknowledgments
The author would like to thank William Burr, Melvyn Lefer, Timothy Naftali, Marc Selverstone, Marc Trachtenberg, Philip Zelikow, and the anonymous Journal of Cold War Studies reviewers for their helpful suggestions and comments on earlier drafts of this article.

Thomas L. Hughes to SecState, 15 May 1964, in Country, Cuba, INR Reports 11/636/65, NSF, LBJL. 179. C. L. Sulzberger, Castro Asserts Cubans Control Soviet Missiles, The New York Times, 8 November 1964, p. 1. 180. David Newsom, The Soviet Brigade in Cuba: A Study in Political Diplomacy (Bloomington: Indiana University Press, 1987). 181. Bundy also pointed out that the focus of the Kennedy administrations deliberations was naturally on offensive weapons. He further wrote that [o]ffensive was clearly understood to mean nuclear-capable. McGeorge Bundy, The Brigades My Fault, The New York Times, 23 October 1979, p. A23.

48