What can we know, why did it start, and how did it end?1
Robert Jervis
As is true for many events, the more we know about the Cuban missile crisis the more puzzling some aspects of it become. So much has been written about it that rather than trying to provide complete coverage, I will cover topics that have either been under-explored or remain in dispute: the extent and role of uncertainty and surprise in the crisis; the particularly political nature of the disputes over the major issues; Khrushchevâs motives; how the blockade brought pressure to bear on both sides; and the place of threats and promises in resolving the crisis, especially the role of the removal of the Jupiter missiles from Turkey, which was more complicated and subtle than is normally portrayed. I will close by pointing out five ways in which the crisis was typical of Cold War interactions.
Knowledge and uncertainty
Before and during the crisis, the leading actors had different degrees of knowledge, ignorance, and misinformation about what was happening, but all were surprised by how it unfolded. Most obviously, the US was taken by surprise, which was Khrushchevâs intention (although in retrospect, seeking surprise may have been a mistake). But when a U-2 flight revealed the secret and the US reacted, it was Khrushchev and Castro who were surprised. These surprises were not only reciprocal, but in a sense the second caused the first. âWe missed the Soviet decision to put missiles into Cuba because we could not believe that Khrushchev could make such a mistake,â declared the leading American intelligence analyst who had been responsible for earlier estimates that had confidently predicted that the USSR would not deploy missiles.2 Although self-serving, the statement is essentially correct. If Khrushchev had known how strongly â or dangerously â the US would react, he would not have proceeded. Even if the crisis did bring some gains, the risk was not worth it. For the Americans as well, the risks were perhaps not worth the gains, or even the losses avoided. On the day that he learned what Khrushchev had done, Kennedy told his colleagues: âLast month I said we werenât going to [allow it]. Last month I should have said we donât care.â3 No doubt he was joking, but like every good joke this reveals an element of truth. At the very least, had Kennedy understood that Khrushchev was so reckless and so highly motivated (leaving aside for the moment the content of the motivation), he surely would have behaved differently, although exactly what he would have done is unclear.
Not only the start of the crisis but its course took everyone by surprise. At no point could anyone be confident of what would unfold within the next 24 hours, and that uncertainty drove the felt need to end the crisis as soon as possible. Indeed, events moved much more quickly than Kennedy, and probably Khrushchev, had expected, and the formerâs initial speech talked about the need for âself-sacrifice and self-disciplineâ over a period of several months.
Of course not all was unexpected, and each side had some inkling of what the other would do. If the US had been completely confident that the Soviets would not put missiles in Cuba, it would not have collected intelligence reports or staged U-2 flights.4 If Khrushchev had been confident that the US would accept the emplacement of missiles, he would not have acted in secret. Most importantly, throughout the crisis, both Kennedy and Khrushchev were confident that the other did not intend to start a nuclear war (which brings up the question of what they did fear, which I will discuss later).
Although Kennedy and Khrushchev often acted boldly â and even in retrospect it is not clear what would have been cautious â they acknowledged the uncertainties. It is striking how much the record is filled with statements to the effect that how the other will react is crucial, but is also unknown. While the participants had hunches, the very fact that they had already been taken by surprise gave them unusual humility. This must have induced great psychological tension because only rarely did someone on either side claim to have what game theorists call a dominant strategy â i.e. one that would be best no matter how the other played the game. Thus the Americans had to debate whether the Soviets would react more strongly to bombing the missile sites or to boarding their ships, and whether Khrushchev would be willing to stand by his first conciliatory letter of Friday night and settle for a no-invasion pledge or whether it was necessary to promise to remove the Jupiter missiles from Turkey. Such debates are typical, but what is less so is that the participants were rarely dogmatic in their assertions, frequently changed their minds, and did not hesitate to acknowledge uncertainty. The phrase âI donât knowâ appears with great frequency. As Kennedy told the ExComm when the blockade was about to take effect, âwhat we are doing is throwing down a card on the table in a game which we donât know the ending ofâ.5 When he told Congressional leaders that his initial decision for a blockade was based on his belief that attacking the missiles would be much more dangerous, he admitted, âNow, who knows that? ⌠We just tried to make good judgments about a matter about which everyone is uncertain.â6
As they took their steps, or even more, contemplating using greater violence, they admitted they were looking into a void. Even the self-confident McGeorge Bundy said that âafter weâve done a violent thing we, none of us, know where it will goâ.7 The Deputy Director of Central Intelligence, Marshall Carter, spoke for many when he said that an attack on the missiles sites âjust frightens the hell out of me as to what goes beyondâŚ. This isnât the end; this is the beginning, I thinkâ.8 Whether escalation would occur or not could not be foreseen, and its perceived likelihood was a crucial factor separating those who were more inclined to favour using force (whether it be an airstrike or an invasion) because they thought that Khrushchev was at such a military disadvantage that he would have to acquiesce, from those who believed that he would feel great pressures to respond militarily in some way and would probably do so.9
The uncertainty loomed largest and most frightening when the increase in pressure or use of force was being contemplated, but it inhibited diplomatic initiatives as well because once launched, no one could be sure of their result. Would concessions lead to further demands? Would allies become demoralized âor, conversely, would they see an America that was standing firm as unduly reckless? Kennedy famously said that an American attack on Cuba would be âone hell of a gambleâ and Soviet ambassador Anatoly Dobrynin referred to the president as âa hot-tempered gamblerâ, but the pejorative connotations were tied to an understanding that anything they did was a gamble.10
Uncertainty is normal in international politics. But here the decision-makers, at least in the American side, were openly at sea, and although available Soviet and Cuban records are much less complete, their great thirst for every scrap of information indicates that their leaders also knew how little they knew. The situation was unprecedented, and the fact that each side was taken by surprise destabilized everyoneâs expectations and made it hard for anyone to feel that he understood the other side or could predict what it would do.11 If major beliefs about the other side had just been shown to be wildly incorrect, what other ideas needed to be modified or discarded? On what basis could either side now estimate how the adversary would respond? For the US the problem was especially acute because the two established interlocutors (Ambassador Anatoly Dobrynin and Georgi Bolshakov, the intelligence agent who provided a back-channel) had been exposed as uninformed or duplicitous.
Kennedyâs openness and willingness to acknowledge uncertainty undoubtedly brought out these characteristics in his colleagues, but more than they Kennedy realized that while they could not predict the future, it was important to understand the past in order to resolve the crisis. Throughout, and especially during the meeting the first evening of the crisis, he pressed for answers as to why Khrushchev had deployed the missiles. He never got much of an answer, and perhaps he gave up too soon.12 But he realized that because the established assumptions about Khrushchevâs perceived self-interest, calculations, and view of the world had just been disconfirmed, it was important to put them on a more secure footing as a prelude to taking action. In fact, the ExCommâs refusal to delve into Khrushchevâs motives (understandable, perhaps, in light of the need to rapidly establish a policy) reduced membersâ sensitivity to some of the diplomatic tools the US could deploy, most obviously a pledge not to invade Cuba.
The uncertainty discussed so far refers to the behaviour of others. While this was central, two other forms were important as well. One was uncertainty of a more factual sort. Most obviously, while Kennedy and his colleagues knew quite a bit about what the Soviets had done, they did not know everything â as they well understood. They realized that they could not be certain about the extent of the Soviet deployment or whether nuclear warheads had arrived (and, if so, whether they had been mated to the missiles). Even more hidden were the activities of Soviet submarines, which posed a menace to the warships that might stop and search Soviet vessels. But the Americans were not uncertain enough: they never thought that the submarines might be armed with nuclear torpedoes or worried that they had vastly underestimated Soviet ground forces in Cuba (and, until late in the crisis, that these forces included tactical nuclear weapons) and that the resistance to an invasion would be much greater than they calculated.13 More uncertainty surrounded the American estimates of how many airstrikes it would take to wipe out the Soviet missiles, and indeed whether all of them could be destroyed before they could be launched. What was crucial to the decision to opt for a blockade and to the sense that if that failed airstrikes would have to be combined with an invasion was the estimate that even a large strike might leave some missiles untouched. This knowledge of the inability to confidently predict the physical, let alone the political, effects of bombing played a large role in turning the tide against an airstrike, and at least some subsequent analysis indicates that the American leaders may have overestimated the difficulties of an attack on the Soviet missiles and exaggerated the ease and speed with which they could be moved.14
Ever since the publication of James Fearonâs path-breaking âRationalist Explanations for Warâ, political scientists have returned to the bargaining problems â and opportunities â caused by the well-known fact that incentives to misrepresent mean that adversaries cannot be certain of each otherâs intentions and resolve.15 Fearonâs basic point is that states have âprivate informationâ about their resolve, but often lack credible means to convey it, and the actors in 1962 were aware of how hard it was to judge what others would do. But it is a mistake to believe that states always know their own resolve.16 In fact neither Kennedy nor Khrushchev appears to have known much more about the risks he was willing to run than he knew about the otherâs tolerance for danger. One advisor hinted at this when he said that the blockade gave the Soviets âa couple of days while they make up their own minds what their intentions areâ.17 Indeed, resolve came and went, and moved sideways. Khrushchev first was ready to remove the missiles in return for a no-invasion pledge and then a few hours later decided to try for more; Kennedy and his colleagues had decided to retaliate if a reconnaissance aircraft was shot down, but then thought better of it in the event. And while we can speculate about what either Kennedy or K...