The Objection That Never Arrived

Amar Murray

On 13 June 1961, less than two months after the failed landing at the Bay of Pigs, the Cuba Study Group that President Kennedy had set up under General Maxwell Taylor sent him its third memorandum. It contains a small observation that is easy to pass over on first reading. The Joint Chiefs of Staff had initially preferred a different plan from the one that was carried out. When they accepted the revised plan, they gave others the impression that they approved of it, and their original preference apparently never reached the senior civilian officials.

I dwell on this observation because it overturns a question we tend to ask after every institutional disaster. We usually ask: was there no one in the room to object? The documents now available show that there were objectors, and that some of them stated their objection in the President's presence or wrote to him directly. The question worth examining is a different one: what happens to an objection in the distance between the mouth that speaks it and the hand that signs the decision?

I should say what I mean by "arriving." An objection can pass through three stages. It can exist. It can reach the person who decides. And it can enter a procedure that obliges others to discuss it, answer it and record how it was resolved. At the Bay of Pigs, some objections never went beyond the first stage, and some reached the second: Fulbright spoke in the President's presence, and Schlesinger wrote to him because he had been asked. So the title does not claim that the information failed to reach the President. What did not arrive, in the sense I use here, is the third stage. By "the records examined here" I mean a limited set: the four State Department documents cited below and the Kennedy Library's account of the operation. This essay is a test of that set, not a survey of every minute of every meeting, and within that set I found no objection that reached the third stage. Every later statement in this essay that something is absent is a statement about these records only. That third stage is a standard I am proposing, not a rule the Kennedy administration had adopted and then broken; I use it as a lens for reading the records, not as a charge against officials for failing to meet it. The Taylor group's own conclusions, written within weeks of the failure, meet this standard in part, in their call for documentation, coordination and follow-up. The published conclusions say nothing about answering objections; the group's procedural recommendations are not reproduced in the published edition, so I cannot say whether they went further. Arriving does not mean prevailing: a recorded answer may reject the objection. Nor does the standard apply to every objection; I set out below which objections should qualify. The point is not to give every objector a veto, but to oblige the decision-maker to explain himself in a form that can be reviewed later.

Written objections, an unchanged decision

On 31 March 1961, Under Secretary of State Chester Bowles wrote to Secretary of State Dean Rusk opposing the operation. He reported what appeared to him to be agreement among those most familiar with it: that the chances of success were "not greater than one out of three." He did not send the memorandum to the President. He suggested that Rusk, if he agreed, carry the objection to the President himself, and added that he thought Rusk's voice would be decisive.

On the evening of 4 April, the President met at the State Department with the Joint Chiefs and others involved in planning the operation. Senator Fulbright, who had been invited, spoke against the plan, and General Gray's summary of the meeting records that Secretary Rusk opposed it as well.

The next day Arthur Schlesinger, Special Assistant to the President, wrote to Kennedy. He noted that the President had asked his opinion after the meeting, and he stated plainly that he was against the operation. He warned that the United States would be held responsible for it and that its standing would depend on its success.

Then came 17 April. Some 1,400 Cuban exiles landed on the island's southern coast. Two days later it was over: almost 1,200 members of the brigade surrendered, and more than a hundred of them were killed.

I do not claim that these objectors were right in everything they said, nor that listening to them would certainly have prevented the disaster. Nor do I claim that what I am examining explains the military defeat. The Taylor report itself lists several causes, from political constraints on the operation to problems with air support and operational assumptions. My subject is narrower: how the institution handled objections while it was making its decision. On that, the documents examined here are clear on one point. The objection existed, in writing and in speech, and some of it was voiced in the President's presence. The decision went ahead nonetheless, and in the records examined here I found no documented requirement that these objections receive a formal answer before approval. That is a finding about these records, not proof that no such mechanism existed anywhere in the government.

Three ways an objection can lose its force

Reading these papers together, I see three ways in which an objection can lose its institutional force. The objections were not of one kind: the Joint Chiefs had an operational preference for another plan, Bowles combined an estimate of the odds with an argument about legal obligations, Fulbright made a political case in the President's presence, and Schlesinger wrote a personal memorandum. They reached different people with different authority. The first two ways are supported by the documents, to different degrees; the third is an analytical question that one case raises, not a path these records prove. Some of what follows is established by the documents; some is interpretation I propose. I will keep the two apart.

The first is that objection dissolves into acceptance. What is established is that the Joint Chiefs accepted the revised plan and that their acceptance was read as approval. Why they accepted it, the document does not say. My hypothesis is that acceptance was what was expected of them at that moment, but that interpretation needs wider evidence. In the Islamic legal tradition there is an old maxim, codified in the Ottoman Mecelle and later taken into modern Arab civil codes such as Jordan's: "No statement is attributed to one who is silent, but silence where there is a need to speak is itself a declaration." The maxim does not say that silence is consent in every case; it says when silence may count, and the Jordanian code itself confines silence as acceptance to particular circumstances, such as a prior course of dealing between the parties. Nor do I use it as a legal judgement on what happened; the Joint Chiefs did not stay silent, they accepted a plan, which is an act. I use it as a comparison. Jurists needed a rule to say when silence may be read as an expression of will at all. The institution in Washington read the acceptance of its experts as approval, and the records consulted here contain no documented step requiring an explicit distinction between approval and acquiescence: do you agree, or are you simply going along?

The second is that objection travels through a personal channel that binds no one. Bowles wrote to his immediate superior and left it to him to decide whether to carry the objection forward. Schlesinger's view did reach the President himself, because the President asked for it. Both objections reached a person. But the Taylor report notes that the Joint Chiefs' understanding of key elements of the plan came from oral briefings in the absence of written documents, and it concludes that there was a need for a formalized procedure with adequate record-keeping of decisions. That conclusion raises a question rather than settling it: was the documented procedure adequate to an operation of this scale, and did anything in it place objections like these on the agenda, so that no decision was taken until they had been answered? The records examined here do not show such a step; they do not prove that none existed.

The third is a question rather than a finding: what becomes of an objection that comes from outside the circle that is held accountable? Fulbright was a senator who had been invited to the meeting and bore no executive responsibility for the operation. I have no evidence that his view was set aside because of his position, so I do not count his case as an example of an objection disabled. But his position raises the question: where does an objection like his sit in the decision procedure, and who is obliged to answer it?

These cases show that personal willingness to voice dissent was not sufficient; they do not show that fear or the cost of dissent played no part elsewhere. The objectors included an under secretary, a special assistant to the President, a senator and the Secretary of State. What the records examined here suggest is that objection had no visible address in the procedure they document: I found in them no minute that required a response, and no point at which the decision paused until it had been settled.

What the institution itself concluded

What stands out is that the Taylor group, looking back at the failure, did not conclude that braver officials or a humbler President were needed. It pointed to procedural machinery: planning, it wrote, "would have been improved if there had been an initial statement of governmental policy" defining the mission and its limits, and "there was a need for a formalized procedure for interdepartmental coordination and follow-up with adequate record-keeping of decisions." That is a call for documentation and follow-up, not an endorsement of the rule I propose below; the convergence is partial, and I do not cite it as confirmation. The group's own recommendations on procedure, which followed in a separate memorandum, are omitted from the published edition, and I do not speculate about what they contained.

I take this to be the most important point in the memorandum, for all its cold technical language. A lack of documentation does not make correction impossible, but it makes it harder, because what was never written down is hard to ask about later. The memorandum also shows that one of the plan's central assumptions, that the force could turn to guerrilla warfare if the landing failed, was not in fact available under the conditions that arose. I find nothing in the documents before me to show that an assumption of this size was defended in writing before those who doubted it, though absence of evidence does not by itself prove that this never happened.

Here this essay connects with what I wrote in "Recognition That Changes Nothing." There I asked about the institution after an error: what becomes possible once it has acknowledged it? Here I ask about the institution before the error: where does an objection go once it has been spoken? Both questions concern accountability, one before the decision and the other after the mistake. An institution that gives objection no address before a decision may find itself, after the disaster, acknowledging an error against which warnings were already written down in its own files.

Two objections to this reading

The first is serious and comes from academic research. In social psychology this disaster became the best-known example of what Irving Janis called "groupthink": a cohesive group suppressing doubt to preserve its unity. In 1998, however, a quarter of a century after Janis first proposed the idea, Roderick Kramer returned to the declassified documents, oral histories and memoirs, and concluded that concern for group cohesion had played a weaker role than Janis believed, and that Kennedy weighed his options chiefly in terms of political consequences and the risk to his reputation. If the decision was political at its core, what is the point of talking about the path of objection?

This objection can be read in two directions. If Kennedy heard the objections, weighed them, and then chose to proceed on a conscious political calculation, the problem may lie in priorities rather than in the absence of procedure, and no procedure can oblige a president to change his mind. I accept this, and I limit my claim accordingly. I do not claim that recording the objection and answering it would have changed the decision. I claim that it would have changed the terms of the weighing. When a decision-maker must write down why he is proceeding despite a specific objection, ignoring it becomes a documented decision that can be reviewed and for which he can be held to account, rather than a voice that fades in the room. And heavy political calculation, such as the cost of abandoning a plan prepared under Eisenhower and approved by a new president in his second month, is precisely what leaves an objection that obliges no one to respond in a weak position.

The second objection is practical. If decision-makers had to answer every objection before acting, they would never act. That is true if the point is taken without limits. I am not speaking of every opinion voiced in a corridor. I mean an objection from someone charged with assessing or carrying out the operation, one that touches an assumption on which the operation rests, such as the Joint Chiefs' preference or the assumption of a fallback to guerrilla warfare. The minimum for such a procedure is threefold: that the objection be recorded; that the decision-maker write a response, even if it is only "we know, and we are proceeding anyway for these reasons"; and that a body be named to review the record after the operation. Without the third, the procedure becomes a compliance form, filled in and forgotten. Such a procedure is not free. It takes time, and secrecy and military urgency press against it. So I would limit the obligation to objections that touch a central assumption of the operation, its legality, or the risk of an irreversible failure, and allow emergency exceptions provided the exception itself is recorded. Among the costs of the operation were some 1,200 members of the brigade taken prisoner and more than a hundred of them killed. I do not claim that a recorded answer would have prevented this, but its absence made it hard for anyone to be asked afterwards: why did you go ahead despite these warnings?

The quality of a procedure is not the same as the legitimacy of a decision. A decision can be taken through a careful, well-documented procedure and still be morally or legally illegitimate. Bowles's memorandum itself did not stop at the odds of success; Bowles argued that the operation would breach the non-intervention obligations the United States had assumed under the charter of the Organization of American States. A rigorous procedure can improve accountability without justifying the substance of a decision; it can even make a wrongful act more efficient. What I discuss here is only the first question, and I do not treat it as a substitute for the second.

The question that remains

What concerns me about the Bay of Pigs is not that a president made a mistake; presidents make mistakes. It is that serious objections and warnings had been raised inside the government before the decision, by people who said so in the President's presence and wrote them down over their own names and dates, and that the records examined here show no procedure that obliged anyone to answer them before the decision was taken. I cannot show that the missing third stage caused the failure; that remains an institutional hypothesis, not a historical finding. But institutions that take comfort in having objectors confuse the existence of objection with its arrival.

So I suggest that anyone who works inside an institution ask, after every major decision and before asking whether it was right: the objection that was raised before this decision, where is it now? In a minute that carries a response to it, or only in the memory of the person who raised it?

Sources

1. Foreign Relations of the United States, 1961–1963, Vol. X, Cuba, 1961–1962, Document 233: Memorandum No. 3 From the Cuba Study Group to President Kennedy, June 13, 1961 ("Conclusions of the Cuban Study Group"), paras. 1(h) and 1(j). Office of the Historian, U.S. Department of State. https://history.state.gov/historicaldocuments/frus1961-63v10/d233; see also Document 234: Memorandum No. 4, June 13, 1961 (Recommendations 1–5 not printed). https://history.state.gov/historicaldocuments/frus1961-63v10/d234

2. Ibid., Document 75: Memorandum From the Under Secretary of State (Bowles) to Secretary of State Rusk, March 31, 1961. https://history.state.gov/historicaldocuments/frus1961-63v10/d75

3. Ibid., Document 80: Editorial Note (meeting of April 4, 1961, at the Department of State, and General Gray's summary). https://history.state.gov/historicaldocuments/frus1961-63v10/d80

4. Ibid., Document 81: Memorandum From the President's Special Assistant (Schlesinger) to President Kennedy, April 5, 1961. https://history.state.gov/historicaldocuments/frus1961-63v10/d81

5. John F. Kennedy Presidential Library and Museum, "The Bay of Pigs." https://www.jfklibrary.org/learn/about-jfk/jfk-in-history/the-bay-of-pigs (figures for surrendered and killed refer to members of Brigade 2506).

6. Irving L. Janis, Victims of Groupthink (Boston: Houghton Mifflin, 1972).

7. Roderick M. Kramer, "Revisiting the Bay of Pigs and Vietnam Decisions 25 Years Later: How Well Has the Groupthink Hypothesis Stood the Test of Time?", Organizational Behavior and Human Decision Processes 73, no. 2–3 (1998): 236–271. Abstract: https://www.gsb.stanford.edu/faculty-research/publications/revisiting-bay-pigs-vietnam-decisions-25-years-later-how-well-has

8. Mecelle (Majallat al-Ahkam al-Adliyya), Article 67. Jordanian Civil Code, Law No. 43 of 1976, Article 95(1) and (2). https://sjd.gov.jo/EchoBusV3.0/SystemAssets/PDFs/AR/AppliedLegislations/qanonmadani2ordoni.pdf

9. Amar Murray, "Recognition That Changes Nothing." https://amarmurray.com/writing/recognition-that-changes-nothing/

Note: Quotations from the documents are paraphrased except where marked with quotation marks.