The Annals of Unsolved Crime (37 page)

Just before noon, on November 23, 1963, the CIA station chief in Mexico, Winston Scott, received word that Mexican police were about to arrest Silvia Duran and hold her “incommunicado until she gives all details of Oswald known to her” about Oswald’s dealings with the Cuban embassy. Scott telegraphed Washington headquarters. The CIA’s concern, as Richard Helms later recalled, was that Duran might say under the brutal techniques used by Mexican interrogators that Cuban intelligence officers had made a deal with Oswald. Whether true or not, that would precipitate a crisis. It could also reveal one of the CIA’s
darkest secrets: the CIA’s AMLASH plot to assassinate Castro. Langley cabled Scott in Mexico,

Arrest of Silvia Duran is extremely serious matter which could prejudice U.S. freedom of action on entire question of Cuban responsibility. Request you ensure that her arrest is kept absolutely secret, that no information from her is published or leaked, that all such info is cabled to us, and that the fact of her arrest and her statements are not spread to leftist or disloyal circles in the Mexican Government.

To assist the Mexican interrogation, the CIA then provided a list of questions concerning Oswald’s September visit to Mexico. It included highly provocative queries, such as: “Was the assassination of President Kennedy planned by Fidel Castro and were the final details worked out inside the Cuban Embassy in Mexico?”; “Who were Oswald’s contacts during the period 26 September 1963 to 3 October 1963?”; and “If Castro planned that Oswald assassinate President Kennedy, did the Soviets have any knowledge of these plans?” They made it crystal clear to the Mexican authorities, as they were probably intended to do, how explosive the Oswald situation could be for Mexico. If Duran were to say under Mexican-style hostile interrogation that Oswald had met with Cuban intelligence officers, had discussed his assassination capabilities, and been told that his visa was approved, or provide further evidence that the JFK assassination had been organized in Mexico as a response to CIA assassination plots against Castro, Mexico would find itself in the center of the crisis and under enormous pressure to take action against Cuba. The Soviet Union might even be involved. The implications of the CIA’s pointed questions were evidently taken seriously by Mexican authorities. The interrogation was ended, Duran was quietly released from custody, and Cuban officials at the embassy who might have had contact with Oswald were encouraged to return to Havana.

Lee Harvey Oswald did not get to Cuba, or even to Mexico City to pick up his visa. On November 22, at 1:50 p.m., after a suspect in the Tippit shooting had been reported running into a Texas movie theater, Dallas police arrested Oswald, who was holding a snub-nosed revolver in his hand. Dragged out of the movie house in handcuffs, he was shoved into the back of a patrol car.

By that time, the rifle found at the Texas Book Depository had been traced to his post office box, making him a prime suspect in the JFK assassination. On route to his arraignment on November 24, escorted by a phalanx of Dallas police detectives, he was himself shot in the police headquarters basement at 11:21 a.m. by Jack Ruby, a Dallas bar owner. Oswald never regained consciousness. At 1:07 p.m, he was pronounced dead at Parkland Hospital, the same hospital at which President Kennedy had died barely 48 hours earlier.

Cubela returned to Cuba, where he was presumably fully debriefed by the Cuban intelligence officers controlling him. But the game was not yet over. He again contacted the CIA through messages using secret writing technology, “s/w” in CIA parlance. He asked for a cache of weapons and money. He also kept asking the CIA about assassinating Cuban leaders. On December 6, 1964, Sanchez again met with Cubela in Paris. According to the memorandum of the meeting written by Sanchez, Cubela said that “his solution to the Cuban problem was the only one feasible and that he had to continue trying.” By now, Helms had firmly banned further assassination missions, so Sanchez replied that “the U.S. Government could not and would not in any way become involved or provide assistance in the task he had planned for himself.” Cubela then said that he would proceed without direct CIA help, and Sanchez provided him with a potential ally by arranging for him to meet with Manuel Artime in Spain. Sanchez later explained to the CIA’s inspector general, “The thought was that Artime needed
a man inside and Cubela wanted a silenced weapon, which CIA was unwilling to furnish to him directly. By Sanchez putting the two together, Artime might get his man inside Cuba and Cubela might get his silenced weapon—from Artime.” Artime, who had been a key figure in the Bay of Pigs invasion, now ran one of the CIA-supported anti-Castro networks in Cuba. As it turned out, Cubela’s involvement proved disastrous for Artime’s network in Cuba.

Angleton, the head of counterintelligence, watched in frustration as the deadly dance with Cubela continued, but he had no control over FitzGerald’s SAS. He was principally tasked with advising Helms on Soviet strategy, and FitzGerald’s unit was responsible for Cuban operations.

The end of this long intelligence game came on June 23, 1965, when CIA headquarters cabled the SAS, saying that it had

convincing proof that entire AMLASH group insecure and that further contact with key members of group constitutes menace to CIA operations against Cuba as well as to the security of CIA staff personnel in Western Europe. Under the circumstances headquarters desires that contact with key members of the group or peripherally involved in AMLASH conspiracy be warned of danger implicit in these associations and directed to eliminate contacts ASAP.

Within a month, Cuban security police began rounding up the anti-Castro agents with whom Cubela had been in contact. In March 1966, Cubela was put on trial in Havana, for his 1964–65 “counter-revolutionary activities” with Artime and the CIA, including “having planned the assassination of Fidel Castro.” The evidence produced included the rifle with telescopic sight from Artime. Cubela confessed and cooperated. What caught the CIA’s attention was that no mention was made of the November 22 plot, although since Cubela was now a
cooperating witness, he could have cited the poison pen and his meeting with FitzGerald. To Angleton, this was an indication that Castro, like others involved, wanted to keep the lid sealed on that Pandora’s box. Cubela was convicted, but although such a crime ordinarily would result in execution by a firing squad, Castro personally intervened. He wrote the court asking that Cubela be given a prison sentence rather than death because of his past service to the revolution.

In Washington, D.C., Richard Helms, now director of Central Intelligence, needed to close this sorry chapter. He ordered the CIA’s inspector general to prepare a report on the CIA’s involvement in the assassination attempts against Castro, instructing, when it was completed in 1967: “Retain it in personal, EYES ONLY safekeeping. Destroy the one burn copy retained temporarily by the Inspector General. Destroy all notes and other source materials originated by those participating in the writing of this report.” Even the typewriter ribbons were destroyed. So except for this one document locked away in the safe of the director of Central Intelligence, all evidence of the plot was expunged. Helms then certified for the record that only one copy existed, which had been read “either in whole or in part,” by him and only three other people at the CIA. They were Inspector General J. S. Earman, and the two authors of the report, Inspectors K. E. Greer and S. D. Breckinridge. Since all of these men were sworn to strict secrecy about what they read, these measures were believed to have sealed the case shut. What could not be foreseen at the time was that a Senate investigation a decade later would break the seal.

In 1979, Cubela, who had worked during his confinement as a prison doctor, was allowed to emigrate to Spain. He received financial assistance from the Cuban government to buy a home in Madrid. For him, the dance was over.

Castro continued to rule Cuba until 2007, when, at the age of eighty and in failing health, he turned the reins of power over to his brother Raul. There were no further efforts made by the CIA to assassinate him after the aborted poison pen plot on November 22, 1963.

How do all of these dates connect? The first theory is that Oswald acted alone. This was the theory of the Warren Commission in 1964. To be sure, the seven members of the Commission did not know when they signed this report that a second jackal, sponsored by the CIA, was in play on November 22. In particular, a jackal requesting that same day a weapon to kill Castro that was similar to the one found in the Texas Book Depository. Yet the single-assassin theory is still viable, because all of the elements of the parallel plot could be considered unrelated coincidences.

The second theory is that Castro, or his intelligence service, directed Oswald to kill Kennedy. This was the theory advanced by Thomas Mann, who was U.S. ambassador to Mexico in 1963. Mann told me in an interview that, on the basis of what was reported by the CIA station in Mexico about eye witnesses to Oswald’s activities and intelligence intercepts of conversations, there was an “indictable case” against Castro. Like the Warren Commission, Mann did not know about the CIA’s parallel plot. If he had, he might have better understood why he was abruptly recalled from Mexico in 1964 when he pressed the U.S. government for a further investigation. More recently, Castro’s involvement in the Kennedy assassination has been advanced in two books. The first one,
Brothers in Arms: The Kennedys, the Castros, and the Politics of Murder
, by Gus Russo and Stephen Molton, published in 2008, is based on interviews with Cuban intelligence officers who revealed that Oswald’s activities in Mexico, including previously unknown meetings outside
the embassy, were reported to Havana prior to the assassination. The second book,
Castro’s Secrets: The CIA and Cuba’s Intelligence Machine
, published in 2012, is by Brian Latell, who worked for the CIA at the time of the JFK assassination. It reports that Florentino Aspillaga, who defected from Cuban intelligence in 1987, revealed in his CIA debriefings that on November 22, 1963, just before Kennedy was assassinated, he had been ordered to switch all the intelligence-gathering antennae under his command from monitoring Florida to monitoring Texas. He then was told to report any transmissions of interest to the top command in Havana. Latell suggested that these and other activities of Cuban intelligence, which the CIA only learned about a quarter century after the assassination, point to the possibility that Castro had advance knowledge about the assassination.

My own assessment, after a nearly fifty-year investigation of this enduring mystery, is that Cuban intelligence had influenced, if not directed, Oswald’s actions. If this were an ordinary police investigation, detectives would have motive, means, and opportunity. As for the motive, we now know that Castro had the most powerful of all motives: self-preservation. In dangling his double agent Cubela before the CIA, Castro ascertained not only that the CIA was actively planning to assassinate him, but that, with the involvement of FitzGerald in the president’s speech, the plot appeared to have the backing of JFK himself. As for means, his intelligence service had Lee Harvey Oswald in the palm of its hand. As Marina Oswald testified, Oswald’s purpose in going to Mexico in September 1963 was to arrange a new life for himself in Castro’s Cuba. He was desperate to serve Castro, so desperate that he had considered hijacking a plane to get there. He also said that he could kill Kennedy, as Castro himself disclosed in early 1964 to his associate Jack Childs, who was also an informant for the FBI. So here was a potential assassin, asking to obtain a visa.

The Cuban intelligence service also had opportunity, since President Kennedy often traveled through cities in an open car, and a sniper could be positioned in a high building along the route. Even if the sniper missed his target, the attempt would send Castro’s message that American leaders who attempted to assassinate Cuban leaders were themselves vulnerable.

Finally, we are now aware of a number of overt acts by Cuban officials, including Castro’s warning and threat, the conditional visa issued to Oswald, and the request by Castro’s double-agent, Cubela, for a weapon similar to the weapon that Oswald used in Dallas. We do not know whether Oswald was informed of his conditional visa approval, but, if he was, the knowledge could have provided him both with an incentive to commit the act, and—in his mind at least—an escape route. John J. McCloy had been high commissioner for Germany before serving on the Warren Commission. When I interviewed him on June 7, 1965, he told me he had insisted that the report itself leave the door slightly ajar because the possibility that Oswald was an agent “could not be precluded.” So Chief Justice Warren inserted a sentence into the Warren Report stating that it was not possible to prove a negative proposition, such as that Oswald had no sponsorship. It turned out to be a wise decision in retrospect because, as it now turns out, it cannot be precluded that Oswald was used by Castro’s intelligence service in retaliation for the CIA’s efforts to assassinate Castro. Indeed, in my view, that is the most likely explanation of what happened in Dallas on November 22.

FURTHER INVESTIGATION

To pursue any of these unsolved crimes, I recommend the following books as starting points.

Chapter 1: The Assassination of President Lincoln

Michael W. Kauffnan,
American Brutus: John Wilkes Booth and the Lincoln Conspiracies
(2004), Random House, New York

Chapter 2: The Reichstag Fire

Ian Kershaw,
Hitler 1889–1936
(1999), W. W. Norton & Company, New York

Chapter 3: The Lindbergh Kidnapping

Jim Fisher,
The Lindbergh Case
(1987), Rutgers University Press, Piscataway, New Jersey

Chapter 4: The Assassination of Olof Palme

Jan Bondeson,
Blood on the Snow: The Killing of Olof Palme
(2005), Cornell University Press, Ithaca, New York