SIX

TYRANNICIDE

THE TWENTY-SEVEN-YEAR-OLD DGI AGENT PRESENTED A FALSE Cuban diplomatic passport in the name of Vicente Ramírez López. His true name, as the CIA would soon discover, was Vladimir Rodríguez Lahera. He was the first person ever to defect from the heart of Cuban foreign intelligence.¹

Rodríguez Lahera did not speak a word of English when he stepped off the Cubana Airline flight in Halifax, Nova Scotia, on April 21, 1964. He nervously scanned the tarmac for a Canadian government official he could approach to signal that he desperately sought political asylum. There would be very little time, and probably just this one chance, to persuade the authorities his life would be in danger if he were forced back on the plane.

Halifax, about midway on the long North Atlantic haul between Havana and Prague, was a refueling stop for Cuban government airliners. Transiting passengers from communist nations had sprinted to freedom there before. But escape would be risky. Rodríguez Lahera knew from his work at DGI headquarters that several of his countrymen on board, including the stewardesses, were agents or informants of state security trained to watch for defectors.²

His life was on the line. If he were snared by Cuban agents before he could plead his case, he would be guilty of treason, doubly or triply compounded. He would be branded a thief as well as a traitor, as he carried a small trove of secret DGI documents jammed into a briefcase and a two-inch wad of money stolen from official accounts. Among his meager possessions were 127 American twenty-dollar bills and a smaller amount in colones, the currency of El Salvador. Summary execution in a dank island prison would be his fate for this combination of crimes if he were thrust back into Cuban hands.³

Once processed in Halifax, he was relocated to Ottawa, where he made clear that he hoped to move as soon as possible to the United States. He had relatives in New York and Miami and wanted to share with American authorities all he knew about Cuban intelligence and subversive operations. Like Aspillaga and most other Cuban defectors who would follow, Rodríguez Lahera had learned to despise Castro and the system he created.

From a poor, rural family in Oriente Province, at the eastern end of the island, Rodríguez Lahera had eleven years of schooling. Letters to his wife Luisa, whom he left behind in Havana, demonstrated that he was smart and creative. Some of that correspondence is included in his once highly sensitive CIA files, which were declassified in the late 1990s. They reveal in remarkably fine-textured detail everything he knew about Cuban intelligence as well the dangerous work he performed for the CIA after he switched sides. There is no other intelligence defector of any nationality whose operational history has been so completely bared.

According to a CIA performance appraisal, he was “personable, intelligent, and foresighted.” Lahera had “proven himself self-confident and courageous . . . quick thinking, and had performed excellently in all tasks.” A Cuban American friend from that era remembers him as “very sociable, not a bragger.” He had been toughened in the Castros’ revolutionary movement, which he joined in 1957 as a twenty-year-old. Rising from private to lieutenant, he first engaged in urban sabotage and propaganda operations, served a sentence in a Batista prison, and then went to the sierra to fight against the dictator as a guerrilla. There he joined Huber Matos, one of the revolution’s most popular commanders.⁴

But Matos ran afoul of Fidel just nine months after their victory when he criticized the growing influence of communist apparatchiks in the new government. Convicted by a kangaroo court, the ascetic Matos was sentenced to twenty years in prison. Rodríguez Lahera’s estrangement had its roots in those events. His hero had fought on Castro’s side, but for a free and democratic Cuba. The old dictatorship was gone, but a new one was forming. Matos’s only offense had been to speak the truth about Fidel’s duplicity.⁵

“Laddie,” as Rodríguez Lahera was soon nicknamed by his new American friends, had served as a DGI staff officer for only about nine months when he defected. But he was observant and curious, had a retentive memory, and, not long into his brief intelligence career, was already planning on eventually assisting the Americans. CIA debriefers were amazed at how much detailed knowledge he brought. About 320 finished intelligence reports about the inner workings of Cuban intelligence were issued based on what he told them. Information of operational value was even more bountiful.

Until Aspillaga’s defection twenty-three years later, Laddie was the CIA’s most valuable Cuban acquisition by far. I never met him, but as a young CIA analyst working Cuba, I was familiar with much of his groundbreaking reporting and relied on it as gospel. Ray Rocca, Jim Angleton’s deputy in the Counterintelligence Staff, was right when he exuberantly described Rodríguez Lahera as “an operational gold mine.”⁶

The mother lode of Cuban secrets he brought was for CIA eyes only, and once in Ottawa, he did not have to wait long. Harold “Hal” Swenson, a tall and athletic CIA officer with leading-man looks and a loose, confident gait, was on a flight out of Washington almost as soon as word of the defection reached headquarters. Brooklyn-born and street smart, the forty-nine-year-old Swenson was chief of the large SAS counterintelligence unit.

He had tallied almost a quarter century of such work, beginning as an FBI agent before Pearl Harbor. One of his first assignments, according to his daughter Sally, was to search for suspected secret Japanese air bases in Baja California, in northwestern Mexico. Fluent in Spanish, he was later assigned by the Bureau to a remote area of Argentina before he joined the Marine Corps when the United States entered World War II.⁷

His first meeting with the Cuban was on the night of April 23. Swenson’s preliminary report, communicated to Langley the next day, was positive but not effusive. A lawyer with a degree from Fordham, he was cautious, a doubter trained to sniff out deception and worry about worst-case possibilities. Loath to elevate expectations too high at Langley, initially he could not be sure he was dealing with a genuine defector rather than a Cuban dangle. Swenson was always more alert to that possibility than Des FitzGerald or anyone in SAS.⁸

In cable language shorthand, he told headquarters the Cuban was “eager go US and cooperate” and that “local authorities willing permit him do so.”⁹ On May 1, still in Ottawa awaiting immigration clearances, Swenson securely sent twenty-three reels of tape-recorded interviews to headquarters. Once transcribed, they formed the foundation of the massive Rodríguez Lahera archive. Thousands of pages of administrative, operational, and information reports and cables to and from Agency field stations would bloat the files devoted to this prize defector. From a starting point of profound ignorance of Cuban intelligence personnel, structures, and workings, suddenly CIA was awash in inside information.¹⁰

Swenson later told the Church Committee that the defection had been a watershed. Until then “[w]e didn’t know there was a DGI; our knowledge was quite fragmentary.” Even among Cubans on the island, Laddie told Swenson, “very few people know anything about it or are even aware of its existence.” He said Cuba’s hidden spy service was generally referred to by code, as “M,” rather than its true name.¹¹

Indeed, it was shrouded so well that the Agency had not previously confirmed its identity. Langley had no doubt that Cuba was running sophisticated intelligence operations of many kinds against American interests. But even after Rodríguez Lahera described the DGI to Swenson, the closest anyone could come to pinpointing its pedigree was that it had been established “sometime in 1961.”

Most of its operating methods had also been beyond the Agency’s ken. Before leaving Ottawa for a CIA safe house in suburban Virginia, Laddie began explaining the missions, internal structure, leadership, and unique tradecraft of Cuban intelligence. He told how agents communicated and traveled. He identified dozens of his former colleagues—although not always by true name because they all used pseudonyms, even among themselves. (His was Victor.) He knew of the DGI’s then-limited capabilities for forging and fabricating passports and other documents and shared a wealth of details about Cuban subversive operations.¹²

Laddie gave the Agency detailed descriptions and floor plans of the previously unknown DGI headquarters that occupied an entire walled city block in Marianao, a close-in Havana suburb. CIA learned for the first time of the innards of the three operational components of its adversary service. The Legal and Illegal Departments were patterned closely on KGB models, but the National Liberation Department was a Cuban innovation.

Swenson—known in the declassified CIA records by his pseudonym, Joseph Langosch, and to Laddie as Mr. Safely—was certain he was dealing with a genuine and truthful defector. The best ones, no matter their nationality, are able to assuage doubts within a few hours. It helped that Rodríguez Lahera had never been a member of the Cuban Communist Party or engaged in operations that imperiled Americans. And he had brought revealing DGI documents.

All the same, he was polygraphed, some years before the Cubans had developed the skills to defeat the lie detector. Thirty-six questions were put to him, including one designed to confirm that he was not a double agent: “Do you have any secret missions unknown to us?” By all measures he answered truthfully. Swenson wrote the CIA security director “there is no question in my mind” that he is a “bona fide defector or that he has furnished us with accurate and valuable information.” After that, there would never be questions from any quarter about his reliability.¹³

Rodríguez Lahera’s swift transformation from a Cuban to an American intelligence operative soon became official. An oral contract was concluded, presumably sealed with a handshake. He began earning $300 a week and was provided a modestly furnished apartment.¹⁴

Around the same time, his cryptonym—AMMUG—was randomly pulled out of a database. Occasionally, however, the choice of a crypt carried an obvious, even mischievous, connotation, perhaps for the amusement of Agency officers who alone are familiar with them. Fidel, for example, was AMTHUG; Raúl Castro, AMLOUT; and Che Guevara, AMQUACK. The DGI came to be known as the sonorous AMAPOLA, the Spanish word for “poppy.” There were hundreds of others in the AM series during the early 1960s. That two-letter prefix referred to something or someone relevant to Cuban operations.¹⁵

In August 1964, using his true name, the CIA reassigned Rodríguez Lahera to JMWAVE. He remained on the CIA payroll for a few more years, assisting, among other duties, in recruitment operations against former colleagues. Even then, years before Aspillaga’s experience in London, it was dangerous work against armed and edgy DGI operatives. Laddie received an Agency commendation for the “courage” demonstrated in his travels, a well-calculated boost to his sometimes-sagging morale.¹⁶

For most operatives, defection opens a lifelong wound. Nearly all, regardless of nationality or background, present dismaying and continuing problems for the intelligence services that receive them. Adjustment issues, feelings of guilt and estrangement, and ultimately a sense of abandonment contribute to emotional distress. For those who leave family members behind, the anguish can be debilitating. Rodríguez Lahera was no exception. After nearly three months of debriefings, when there was nothing more to be squeezed out of him, he complained of being underutilized and neglected. He had made only a few friends, remaining closest to Hal Swenson, always a calm, reassuring father figure who counseled and encouraged the Cuban during his most painful periods.

He was “amicably terminated” from service on November 30, 1967, and received generous separation bonuses. The Agency even took a number of complicated steps to return to him the $2,540 he had stolen from the DGI. Creative finance officers, surely enjoying the challenge of legitimizing the theft, reimbursed him for the federal taxes due on that “income” as well as accrued interest. He moved to Miami and found work there as a driver with a trucking firm. An American citizen by then, he seems to have left his two-chapter intelligence career permanently behind, blending inconspicuously into the Cuban American community. According to reports I have not been able to verify, Laddie died sometime later, still a relatively young man.

HIS DEATH MAY HAVE BEEN from natural causes. But the mysterious demise of his friend and Agency comrade Miguel Roche, AMNIP–1, leads directly to the top of the Castro regime. Roche was a teenage veteran of the anti-Batista urban underground, a distant relative of Redbeard Piñeiro who recruited him for police and intelligence work and later protected him when he ran afoul of fanatical revolutionary ideologues.

At the age of twenty-one, Roche was promoted to lieutenant and put in charge of internal security in Camagüey Province. Later he managed a government-controlled import company used for intelligence purposes and may have had responsibility for local gambling operations. Quick with calculations and financial transactions, he easily played at black-market currency manipulations to the advantage of the new regime. Had he remained in the Cuban service, he probably would have risen to its top ranks, becoming a DGI colonel or general.¹⁷

Gradually disaffected, however, he took asylum in the Ecuadorian embassy in Havana in December 1961 and later gained safe passage to Brazil. He was contacted there almost immediately by Hal Swenson and brought to the United States. Dangerous overseas assignments for CIA soon became his specialty. In mid-June 1964, Swenson wrote for the record: “I explained to MUG that in the immediate future I would like him to work with Roche in identifying members of Cuban intelligence . . . and I explained it might be necessary . . . to make trips outside the US to contact persons of operational interest who might be recruited or defected.”

Laddie and Roche became close friends, both residing in the northern Virginia suburbs of Washington, near CIA headquarters, and often traveling abroad together on joint operations. They were an impressive pair. Miguel, from a welloff commercial family that lost everything to the revolution, was the dominant, buoyantly self-confident senior partner. Five-foot-ten-inches tall and weighing over two hundred pounds, he towered over the diminutive Laddie.

Clever, charming, and gregariously persuasive, Roche was a born street operative and was soon cutting to the DGI’s quick. His widow told me that during the next several years he traveled abroad as often as twice a month carrying out Agency missions. She remembers that he went to Europe as well as Canada, Africa, Mexico, and South America. Declassified records confirm that at least one successful recruitment resulted. An agent, known as CAPRICE 1, was pitched by Roche in 1964 and formally enlisted several years later. Judging from how aggressively the Cuban regime responded to Roche’s efforts, there were probably other successes too.¹⁸

In June, Roche traveled to Mexico City and Santiago, Chile, to meet DGI officers considered good recruitment prospects. The hoped-for rendezvous in Mexico never materialized. But soon after Roche huddled in a hotel bar in Santiago with two Cubans, one a known intelligence officer. It was a tense, high-stakes face-off, with recruitment pitches flying in both directions. Interior Minister Ramiro Valdés and other officials in Havana had “a high regard for him,” Roche was told, and they “deeply regretted his defection.” The Cubans wanted him back in the DGI fold as a double agent . . . or so they claimed. But, in a sign of what was to come half a dozen years later, he was also crudely threatened. The CIA station in Santiago cabled headquarters that Roche was warned: “Cuban intelligence has long arms and could kill him anyplace, anytime if he played false.”¹⁹

The meeting had apparently been authorized and scripted by the senior leadership in Havana. Cuban figurehead president Osvaldo Dorticós told a CIA asset, “Oh yes, we know all about ‘El Cojo’ Roche’s [recruitment] attempt . . . on behalf of the Americans. . . . We know all about his activity.” ( Cojo means “cripple” or “lame”; Roche walked with a limp after being thrown through the windshield of a car when working for Cuban intelligence in Havana.) The highlevel interest meant that his work for the CIA had become so fruitful that he was a marked man.²⁰

Undeterred, Roche remained operationally productive and brazen until his death. But with each approach he made to a Cuban, he put himself in greater peril. A thick dossier cataloging his efforts became the source of increasing outrage in Havana. He and Laddie had crossed two Rubicons: They had defected to the “imperialist” enemy with valuable information, and they went on to taunt and subvert their former service. By unforgiving fidelista diktat, they were committing capital crimes. Roche’s were compounded many times over.

He received threatening letters at his home indicating Cuban authorship. He was warned: “We know what you are doing . . . so one way or another we will take care of you.” His widow still has one of the letters. Roche was only thirty-one years old and in robust good health when he collapsed in front of his home in Falls Church, Virginia, in September 1970. He was dead by the time of arrival at a local hospital. Earlier that day, his widow told me, accompanied by an FBI agent, he had had an operational lunch meeting at a Mexican restaurant with one or more mysterious Cubans. It was just hours before he died. His widow was kept in the dark about whom Roche met, and why, but to this day she has no doubt that her husband was poisoned by Cuban intelligence. Autopsy reports she shared with me suggest that she is right.

Roche’s blood tested negative for alcohol. His cardiovascular system checked out as normal; he had never been diagnosed with heart or circulatory problems. “He had been perfectly healthy until the day he died,” his widow says. The toxicology report stated that 90 percent of the hemoglobin in his body had been contaminated. Cyanide, I am told, can have that effect. Extreme inflammation was found throughout the cells of his respiratory system. The examining pathologist, a medical doctor, told Roche’s widow that he believed the damage “was caused by some unknown poison.”

There was no police investigation, and CIA was unable to do anything despite powerful suspicions of foul play. Perhaps, however, Roche’s untimely death persuaded the Agency that future star Cuban defectors would have to be provided new identities in order to live more safely beyond the reach of Castro’s assassins. The dueling work that Rodríguez Lahera and Roche performed when trying to suborn DGI agents was much more dangerous than was thought at the time.

RODRÍGUEZ LAHERA’S GREATEST CONTRIBUTIONS were in exposing Cuban subversive operations. With almost no background or training other than his own guerrilla experience, he had been assigned to run the El Salvador desk at DGI headquarters.

Governed by a tenacious military-plutocratic elite, that little Central American country was a much lower priority for Fidel and Piñeiro than Venezuela or Guatemala, where guerrilla insurrections were taking root. Revolution in El Salvador was not as promising; the Cubans thought, correctly, that an upheaval there was still years in the future. Still, small Marxist cells and knots of Cuban-inspired youths were eager to take up arms.

Before his defection, Laddie kept busy supporting them. He worked with guerrilla combat trainees, illegals, and members of the Communist Party who clandestinely made their way to Havana. Their mentor and principal facilitator, he quickly became the DGI’s resident country expert. His knowledge of espionage tradecraft extended to the workings of the Mexico City Center, which served as the conduit and clearinghouse for Central Americans seeking passage to Cuba.

His defection was a boon for the counterinsurgency experts at Langley. He identified about 120 aspiring guerrillas and DGI illegals of a half dozen nationalities, all by name. He knew seven of them well enough to describe quirks and vulnerabilities that made them easier prey for CIA recruitment or, failing that, nullification as viable Cuban assets. In mid-June 1964, it was noted in an “operational target analysis” that “we have already undertaken action.” The kind of action was not specified.²¹

Partly on the basis of Rodríguez Lahera’s reporting, the United States was able to grind down, all but annihilate nascent insurgencies in a few countries. The most drastically thwarted were Cuban subversive efforts in El Salvador and Nicaragua. Laddie passed on the names and descriptions of about thirty-five Salvadorans working with the DGI, most of them graduates of guerrilla training camps. He also identified twenty-three trainees from Nicaragua. Those exposed constituted nearly the entire trained revolutionary cadre in the two countries; it would be another fifteen years before Cuban-supported insurgents could regroup and fire up viable insurrections.

One of Rodríguez Lahera’s most valuable tips concerned a brooding young Salvadoran firebrand. Roque Dalton, author of more than a dozen volumes of prose and poetry, is still recognized as a talented Latin American literary figure and martyr for revolutionary causes. He is the most honored poet ever to emerge from El Salvador.

Dalton was recruited in Cuba as a DGI agent in October 1963 and codenamed “Montenegro.” Rodríguez Lahera, his case officer, arranged for training in secret writing, radio transmission and reception, and other clandestine tradecraft. Like so many of his generation of Latin Americans, Dalton also completed courses in guerrilla warfare.²²

He had studied at universities in Chile, Mexico, and El Salvador. Twenty-eight when Rodríguez Lahera began describing him to the CIA, he was an attractive figure with an authentic legend that even Piñeiro himself could not have invented. According to CIA records, he had “a long thin face, black eyes, two gold teeth, and black hair parted on the left side.” Oddly, the description failed to mention his most distinguishing physical characteristic, his long Pinocchio nose.

Dalton must have been one of the most promising of the DGI’s Central American recruits: fit, young, charismatic, and admired by Salvadoran youths and intellectuals. When he was singled out, he was surely thought of as just the kind of romantic hero who could lead a Cuban-style guerrilla revolt in the volcanic mountains of his country. He might even prove to be a successful Salvadoran Che Guevara.²³

But Dalton dismayed the Cubans. It turned out he had been poorly motivated from the start. “He has a weakness for women and the easy life, and a generally weak nature,” according to the CIA vulnerability analysis. Dalton “is very intelligent,” Rodríguez Lahera told the Agency, “but never showed a real desire to learn during his Cuban training.”

He had been given $600 to purchase a sophisticated radio set when he returned covertly to El Salvador and instructed to transmit coded messages on the second Monday of each month. The frequencies he was to use were recorded on seven scraps of microfilm concealed in a hollow in the heel of his shoe. Despite the elaborate training and the DGI’s high hopes, the poet never communicated. Swenson reported in his first debriefing dispatch from Ottawa that once back in his country, Dalton “took a little trip for himself, lived high, and spent the money.”²⁴

Of all the foreign revolutionaries who had come to his attention, Rodríguez Lahera considered Dalton the “leading candidate for recruitment by us,” according to a CIA report. It was speculated that “subject would not be a too difficult target to hit,” meaning to approach and recruit as a double agent against the Cubans.²⁵

But why would Dalton work more diligently for CIA than he had for the Cubans? That obvious concern was noted in the records: “his failure to perform for the DGI . . . might mitigate against our use of him.” But on balance, it did not matter. The concern, after all, was motivation. He was like all the other DGI agents and illegals over the years who got paid only for expenses and perhaps an occasional small honorarium. A substantial flow of American dollars might be sufficient inducement for him to play a double game.

Soon the Agency concluded that Dalton’s membership on the central committee of El Salvador’s Communist Party was reason enough to keep him in its sights; “it makes him an operational target for that reason alone.” A headquarters memo in November 1964 expressed the range of what the CIA hoped to achieve: “Our object is to double, defect, or nullify Dalton.” Indeed, after he was arrested and imprisoned in San Salvador earlier that year, Rodríguez Lahera and a CIA officer—probably Swenson—interrogated the poet, determined to turn him against the Cubans. He “was much shaken” and “begged” to be released, promising to abandon politics, according to a CIA field report cabled to headquarters. Yet, even in the presence of his former DGI case officer, Dalton was adamant in denying any association with Cuban intelligence. But this is where the declassified record runs dry. If he was eventually recruited, if Dalton was persuaded to cooperate in any way with the Agency, nothing of it has been revealed, and most likely it never will be.²⁶

On the surface, he was a model revolutionary. For a number of years after his time in prison he lived in Havana and Prague, and most likely remained on the DGI’s rolls despite his failings as an agent. Most of his writings were published in Havana by the Casa de las Americas, the government’s publishing and cultural arm that has always been entwined with intelligence.

In late 1974 he returned yet again to El Salvador to take up arms against the government and penned some of his most didactic underground verse. A few lines from his poem “The Violence Here” are typical:

Violence will not only be the midwife

of History in El Salvador.

It will also be the mother of a child-people.²⁷

Dalton was often a focus of flattering Cuban propaganda, awarded literary prizes and featured at international symposia. His public posture remained purposefully militant. For all appearances he was a true believer in fidelista causes, particularly violent revolutionary warfare in his own country.

Yet, in the hall of mirrors of international espionage, the obvious often is a mirage. He actually may have had reasons to cooperate somehow with CIA. His father was an American who emigrated to El Salvador. Three of his brothers were American citizens, one of whom is said to have fought against Japan in the Pacific during World War II. And the circumstances surrounding Roque’s death in May 1975, when he was just days short of his fortieth birthday, have kept speculation alive that he in fact had some kind of CIA connection.

Back in El Salvador, he joined an emerging guerrilla movement. But soon he was in strident conflict with its leaders. They staged a secret revolutionary trial, finding him guilty of insubordination or treason. Dalton was executed the same day in San Salvador, and his body was never found. He had strayed far beyond his literary talents, a victim of the fantasy that his twin muses of poetry and revolutionary warfare could happily cohabit. He was another of the many doomed acolytes of the 1960s revolutionary fevers, a willing victim of Fidel’s morbid incantations to revolutionary violence.

Those who imposed the death sentence spread the word that the poet had to be killed because he was a CIA agent. That may have been contrived as a sure way of justifying their actions; or they may have been convinced it was true. I am not aware that the killers ever elaborated on the charge or provided evidence. Certainly nothing ever leaked from CIA vaults about Dalton that would have compromised him. And the archived records where I found descriptions of CIA operational interest in him were not released until 1998.²⁸

Ironically, there is a good chance that the killers were victims themselves—of a masterful deception operation. If so, the most likely perpetrator would have been the brutal Salvadoran military dictatorship. The generals and army intelligence had ample motive to plant false evidence linking Dalton to CIA. They would have calculated that it could hobble the insurgency, further dividing the guerrillas and putting one of their most appealing leaders in mortal jeopardy. In 1960s and 1970s Latin America, almost anything was permissible in guerrilla and counterinsurgency warfare.

And yet there is another possibility: The Cubans may have been responsible for Dalton’s execution. He was a mercurial dilettante, had abused DGI tutelage, and may have run afoul of Fidel and Piñeiro. If Cuban intelligence concluded Roque had been secretly involved with CIA, nothing would have been more certain to result in a death sentence. As a trained DGI illegal, he would have been marked just as surely as the traitors Aspillaga and Roche were.²⁹

The possibility is strengthened by the subsequent execution—on Fidel’s orders, it appears—of another prominent Salvadoran revolutionary. Salvador Cayetano Carpio, “Marcial” to his brethren, the “Ho Chi Minh of Central America” as he liked to boast, was the most intransigent hard-liner among the guerrillas. He was blamed in 1984 for the gruesome ice-pick assassination of “Ana Maria,” the nom de guerre of his Salvadoran guerrilla deputy and rival. Shortly after attending her funeral in Managua, Carpio was also found dead, of a gunshot wound, in the Nicaraguan capital where the Marxist Sandinista regime was in power. His death was explained away as a suicide.³⁰

But Juan Antonio Rodríguez Menier, one of the 1987 DGI defectors, believes differently. He is sure it was Fidel who issued the death warrant. Ana Maria’s murder “was investigated by top Cuban officials” in Managua, he says. According to other sources, Piñeiro was among those investigating officials. Rodríguez Menier adds that Fidel “then ordered that the man responsible be executed.”³¹

Carpio had been dangerously at odds with the Cubans over political and military strategy. And he objected to their aggressive meddling. Just months before the two murders, he is reliably reported to have repudiated Fidel’s bullying, telling him “to go to hell.” That is not the kind of rebuke the vain Cuban leader suffers lightly. Jorge Castañeda, the former Mexican foreign minister and an expert on the Latin American radical left, wrote that Carpio “had been fighting the Cubans . . . for months.”³²

Fidel was accustomed to exercising patrimonial influence over the guerrillas and to getting his way. Notably, in 1979, he brought leaders of their competing factions to Havana, where, with cajolery and promises of munitions, he forged the insurgent alliance that became known as the Farabundo Martí National Liberation Front, or FMLN.

Joaquín Villalobos, leader of one of the factions, was there. He told me how two representatives of each of the guerrilla groups were met at the Havana airport by Cuban intelligence officers and taken separately in Mercedes limousines to guesthouses. Together, all of them later met with Castro in the presidential palace with Piñeiro attending. “Carpio was the problem; he challenged Fidel; he’d been drinking,” Villalobos told me during a meeting in November 2011 in Toronto. Another Salvadoran guerrilla commander later recalled that “had it not been for the Cubans, especially Fidel Castro, it would not have been possible to attain unity.”³³

After building up their capabilities, on January 10, 1981, just ten days before Ronald Reagan’s inauguration in Washington, Fidel presumed to act as the guerrillas’ commander in chief. From a DGI war room in Havana, he relayed orders to their fighting units as they launched what was optimistically called the “final offensive” to topple the military government. A former American intelligence officer with direct knowledge of the Cuban role told me, “They [the Cubans] were running the whole thing, receiving intercepts on Salvadoran military movements and operations, and communicating with the rebels. They were highly confident the FMLN would be victorious.”

Plans had been in the works for over a year with as many as five hundred Salvadorans being trained and readied in Cuba. They received hundreds of tons of arms, channeled through Nicaragua. Seasoned Cuban advisors were on the ground with them. Fidel’s audacious plan was to present the new American administration with a revolutionary fait accompli. The fighting over several days was savage, but the government’s military forces beat back the onslaught.³⁴

Apart from Rodríguez Menier’s assertion, I can find no evidence linking Fidel to Carpio’s murder. But everything else the defector has revealed appears to be reliable. His story about Carpio’s death is consistent with Castro’s violent behavior since his university days and with the seigniorial privilege he wielded over the Salvadoran rebels. Assassinations and attempts against enemies, rivals, and traitors were as characteristic of his governing style as speechmaking and grandstanding. During his decades in power, Fidel authorized the executions of a long list of offenders, including two of his intelligence chieftains who died under mysterious circumstances, probably on his orders.

Redbeard Piñeiro allegedly was one of them. A well-connected former Cuban government official who worked with the spymaster for many years claims that Piñeiro was murdered in 1998 on the orders of either Fidel or Raúl Castro.

Like so many others in the brothers’ entourages over the decades, Piñeiro had strayed from their rigid orthodoxies and had been shunted aside. But “he was impossible to retire,” my source told me. He knew too much; he was writing a book, and he made the mistake of telling others about it. His whole career—more than forty years—had been in intelligence and intrigue. Everything he had done was as Fidel or Raúl’s proxy in covert operations in dozens of countries on several continents. What else could he have been writing about?

The day after his death, security personnel searched his home in Havana, “as if he were a dissident” or a conspirator, and “kept all kinds of papers,” according to the former Cuban official who now resides in Florida. He also told me that Piñeiro’s home “was surely bugged” and that he had been talking too freely.

Redbeard’s bodyguard, who doubled as his driver, was also certain his boss had been murdered. Piñeiro was said to have fainted at the wheel of his car, resulting in a single-vehicle accident on the streets of Havana. He survived with minor injuries and was taken to a government executive hospital for observation. A day before the accident, the transportation ministry had instructed his driver to take some time off: Piñeiro would drive himself. My source told me the distraught driver lamented openly: “They knew. They knew.” The regime said that Redbeard died of a heart attack in his hospital bed. The story is reminiscent of brutal scenes in the Godfather movies that Aspillaga told me were shown in his day for DGI training purposes.

ASSASSINATION OPERATIONS HAD ALWAYS BEEN Fidel’s personal bailiwick. None could be conducted that he did not authorize and help plan. The means for carrying out this most sinister of secret Cuban capabilities were always decentralized and rigidly compartmentalized. In the 1950s and early 1960s, the KGB’s infamous Department 13 killed traitors in what the Soviets called “wet operations.” But when I asked Aspillaga if Cuba had created anything similar, he laughed: “The Russians in that respect are more vulgar.” It was not scruples that concerned Fidel but the need for airtight deniability.

The Cubans used DGI-controlled illegals, surrogates of other nationalities, as executioners. They carried out some of the most sensitive missions overseas, especially against high-visibility, well-protected targets. Death squads drawn from Latin American terrorist and revolutionary groups beholden to Cuba could be relied on, deniability compounded by degrees of separation. Carefully screened, the foreign assassins were trained at secret Cuban bases, learning to kill in gangland-style hits, elaborately orchestrated paramilitary operations, commando strikes, and sly poisonings.

In the most sensitive operations, when even greater deniability was desired, Fidel did rely on carefully screened Cubans. In the 1970s and 1980s, according to Aspillaga, a super-secret four-man squad of assassins reported exclusively to Castro. With few exceptions, including Minister of Interior José Abrahantes, who succeeded Valdes in 1985, no one else was aware of these professional hit men, Tiny told me. It was dangerous for any Cuban to share the commander in chief’s most incriminating secrets. No wonder perhaps that, like Redbeard, Abrahantes also died under mysterious circumstances. His death received just a few memorial lines in the official media. In our meetings, Aspillaga described two of Fidel’s secret assassins. One he knew in the 1980s was nicknamed “El Chiquitico,” the Little One. Another was familiar to him only as “El Chamaco,” the Kid. In one of our recorded interviews, Aspillaga said of Fidel, “When he chooses someone, he takes his personality and dominates you . . . he controls you mentally. That’s what he did to those four assassins.” They had been molded and brainwashed, Aspillaga believed, into blindly loyal killing machines.

I asked him for examples of their handiwork.

Fidel, he said, “had generals in Bolivia who were involved in Che’s death killed.” CIA analysts had come to that conclusion years before Aspillaga defected. Four Bolivians—two generals, an army captain, and a peasant—who had materially contributed to Che’s demise were assassinated, for all appearances, by death squads. Another general, René Barrientos, the popular president of Bolivia when Che was hunted down, died himself a year and a half later in an unexplained helicopter crash.

In the late 1960s, we CIA desk analysts knew nothing about Castro’s personal team of assassins and, frankly, little about his compulsion for lethal revenge. But the number and pattern of the killings of the Bolivians, Fidel’s obvious motive, and the professionalism of the executions all suggested official Cuban involvement. These were not the kinds of mysterious deaths that could have been explained away as heart attacks, suicides, or accidents. We had no doubt that the Bolivians had been murdered by killers intent on avenging Che.

The first to die after Barrientos was Honorato Rojas, a subsistence farmer in the backlands where Che’s insurgency had struggled for a toehold. At first Rojas assisted a band of guerrillas commanded by one of Guevara’s lieutenants, agreeing to guide them through the tangled terrain. But a Bolivian army officer persuaded him to betray the strange, bedraggled intruders, most of them Cubans. On August 31, 1967, Rojas led the guerrillas straight into a killing ambush at the confluence of two swift rivers. A half dozen of Guevara’s dwindling band were killed instantly, and others were captured. Che’s lover, a German Argentine woman known by her war name, “Tania,” was among them. It was one of the decisive skirmishes in the lopsided Bolivian conflict and was followed five weeks later by Che’s capture and execution.³⁵

Rojas’s betrayal was key to the failure of the entire revolutionary endeavor; the ambush he arranged eliminated a third of Che’s force. For Fidel, the Bolivian farmer was appallingly reminiscent of Eutímio Guerra, the Cuban peasant who had betrayed him to Batista’s army, one of the “big three” traitors he would publicly condemn in 1987. In July 1969, Rojas paid the ultimate price for his treachery, as Guerra had a dozen years earlier. The luckless peasant was gunned down by unknown assailants claiming to be members of a Bolivian revolutionary front.

The next target was Roberto Quintanilla, a Bolivian army intelligence officer who played a role in Che’s failure. He was murdered in Germany in 1971. The best known victim was General Joaquín Zenteno, commander of the army division that pursued Che. Zenteno was shot in Paris in May 1976 while serving as his country’s ambassador. The previously unknown Che Guevara Command claimed responsibility; it was never heard from again. Two weeks later another general, Juan José Torres, a top Bolivian staff officer who had ratified the order for Che to be executed, was murdered by an Argentine death squad. All the cases quickly went cold.³⁶

General Zenteno was doubly anathema to Fidel. Assisting him in his hunt for Che were two Cuban exile contract CIA operatives, both veterans of the earlier clandestine wars across the Florida Straits. They were well known to Cuban intelligence. In his memoirs, Félix Rodríguez admitted participating in an assassination plot against Fidel in 1961, and he believes he was targeted for death by Castro after Che’s execution. Gustavo Villoldo, the second Cuban exile advisor to General Zenteno, also published memoirs, and told me that he was targeted for death on three different occasions by Cuban operatives, most recently in 2003 during a visit to Bolivia.”³⁷

There is no evidence of Cuban government participation in the murders of the Bolivians. It is possible that all the death squads acted independently of Havana, each incident entirely separate from the other. Latin American and European terrorists were enchanted with the iconic Che, wanted revenge, and could have acted on their own. But, on balance, Cuban responsibility for at least four of the cases seems likely.

Arranging for the executions of defectors, traitors, worthy enemies, and even an occasional foreign general was commonplace in Fidel’s nearly fifty-year career in office. Targeting serving and former heads of state was a more daring undertaking.

Since the Middle Ages and early Renaissance period, regicide practiced by a national leader against enemy monarchs and chief executives has been rare. Bill Harvey, one of the CIA’s legendary operatives—who twice in the early 1960s plotted Castro’s death with Mafia gangsters—told the Church Committee that even Soviet dictator Joseph Stalin must have eschewed such risky behavior: “I cannot personally document or testify to a single Soviet assassination of a leader of a foreign state.”³⁸

But through most of his years in power, Fidel played by his own vengeful rules. At least four sitting or former presidents of Latin American countries were the targets of meticulously planned Cuban “black” operations. Probably other such operations left no traces.

Knowledgeable exile sources have told me that Fidel for years had his predecessor, Batista, marked for execution. The old dictator, living in exile in Portugal and Spain, was the target in 1973 of an elaborately rehearsed Cuban plot. Fidel had compelling motive. In the mid-1950s he had spent nearly two years in one of Batista’s prisons. The dictator had authorized untold numbers of executions of Castro’s supporters in the late 1950s. After the triumph of the revolution, hundreds of his henchmen were put up against execution walls in retaliation. Had he not escaped in time, Batista assuredly would have stood before a firing squad himself.³⁹

Fidel’s plan was not to assassinate him but to snatch, or kidnap, him alive. It would be a Cuban version of the justice meted out to Nazi mass murderer Adolf Eichmann, who was kidnapped by Israeli intelligence in Argentina and convicted in a show trial in Jerusalem in 1961. Cuban commandos and DGI operatives were ready to seize Batista from the walled compound near Lisbon where he lived or when he ventured out. He would be drugged, smuggled to Havana—probably on a Cuban merchant vessel—displayed and humiliated before a revolutionary tribunal, and then executed.

I learned of this previously untold conspiracy from a ranking DGI defector. For his safety and that of his family members, I will call him Francisco Compostela. Now living in the United States under an assumed identity, he learned of the Lisbon plot from another senior DGI officer with knowledge of what was in the works. Compostela told me, “The plan was ready to be implemented. We had a squad of illegals set up in a safe house, ready to seize Batista and take him to Cuba . . . or assassinate him if the plot could not be fulfilled. It was elaborately planned.” Ironically, Batista died of natural causes during a vacation at a Spanish resort town in August 1973 shortly before the operation was to take place.

Fidel always denied any intent to seek vengeance against his predecessor. During a fundraising speech in New York in November 1955, a year before he set sail from Mexico to begin his insurgency, he told an audience of Cuban exiles: “We do not practice tyrannicide.”⁴⁰

He meant Batista, of course. And then, quoting a hero of Cuba’s nineteenth-century independence wars, he elaborated: “The man who exposes himself to death and can kill his opponent on the battlefield does not use the treacherous and disgraceful means of assassination.”

The “battlefield” reference is key to Fidel’s thinking. That is where he believes he has spent his entire adult life. As he tells it, his personal saga has been a protracted life-and-death conflict with powerful enemies bent on destroying him and his work. His preference was always to seize the initiative against them, to go on the offensive, and to give no quarter. Tyrannicide, therefore, was merely one of the many forms of homicide he believes to be justifiable—in self-defense, in warfare, but more precisely in revolutionary struggle. When an enemy poses a grave threat, assassination is justifiable—not only in what he would characterize as a just war but also against a foreign leader he considered heinous.

The savage Dominican dictator Rafael Trujillo was an early example. He was a genuine tyrant from almost any perspective, including John F. Kennedy’s. Trujillo authorized the torture and merciless killing of his opponents. The grudge Fidel held, however, was due to Trujillo’s sponsorship of a clumsy coup attempt against him in August 1959. Castro even then—his first summer in power—was running double agents, one of whom kept him informed of Trujillo’s conspiracy.⁴¹

Rumors of Fidel’s desire for revenge against his Caribbean nemesis circulated for years. But until recently, I was not aware of persuasive evidence. The new information comes from Dariel Alarcón Ramírez, still known by his nom de guerre, “Benigno.” A faithful foot soldier who fought with Che in Bolivia, he was one of a handful who survived the debacle, managing, after tortuous peregrinations, to get safely back to Cuba. Later, though, like so many other experienced DGI officers, he defected. He has firsthand knowledge of an attempt on Trujillo’s life.

Benigno recently told me: “I know of direct Cuban participation in an assassination attempt against Trujillo in 1959.” A prominent veteran of the Castros’ insurgency, a comandante named Derminio Escalona, was allegedly the leader of the failed attempt. Benigno also confirmed what Aspillaga told me about Fidel targeting the errant Bolivians.⁴²

For Castro, however, there were no more deserving objects of his wrath than two of modern Latin America’s most reviled dictators. Also both generals, Anastasio Somoza, the durable Nicaraguan dictator, and Augusto Pinochet, the Chilean president from 1973 until 1989, were for years high on Fidel’s most wanted list.

Somoza, commander of Nicaragua’s National Guard before inheriting the presidency in 1967, had done much to earn Fidel’s wrath. Working with the CIA, he had provided training facilities and an air base for the Bay of Pigs brigade in 1961. Two years later he allowed Manuel Artime’s “autonomous” exile group to train and launch sabotage attacks on the island from a base on Nicaragua’s Caribbean coast. As noted, Fidel knew all about it from a Cuban double agent. Somoza’s was the kind of mercenary belligerence that Castro cannot forgive.

The DGI mounted the first serious attempt against the dictator in 1964. But it was not until sixteen years later that a perfectly executed commando operation succeeded in assassinating the former Nicaraguan leader. The armored car in which he was being chauffeured in the streets of Asunción, Paraguay, was incinerated in a coolly calibrated bazooka attack on September 17, 1980.

Jorge Masetti, one of Piñeiro’s boys—his godson, in fact—has written about it. Masetti was the son and namesake of a fallen Argentine guerrilla leader who had been close to Che. Following in his father’s footsteps, the younger Masetti was for years a roving DGI warrior and operative. After defecting in 1990, he described Somoza’s murder. It was a precision attack, conceived, planned, and practiced to perfection at a secret base in Cuba.⁴³

The executioner “knelt in the middle of the street,” according to Masetti. “His shot hit the mark dead center, but the projectile was a dud. And then, amid the ensuing crossfire . . . he calmly reloaded and made the second shot that killed Somoza. The guerrillas then hastily withdrew according to plan.” Masetti knew them; they were a group of Argentine terrorists, DGI illegals. “They were my companeros,” he wrote.”⁴⁴

With Somoza gone, Pinochet rose to the top of Fidel’s demonology. Leader of the September 1973 coup that overthrew fervid Cuban ally Salvador Allende, the Chilean president would prove less vulnerable than the exiled Somoza. There may have been other failed attempts, but the one that came closest to success occurred in September 1986.

It was a paramilitary operation similar to the one against Somoza, conducted at the curve of a road in the outskirts of the capital of Santiago with an arsenal of heavy weapons. Two Cuban defectors—former top DGI operative José Maragon and Lázaro Betancourt, a commando and sharpshooter—know details of the meticulously planned attack. They told me the guiding Cuban hand was common knowledge in their intelligence circles.

Betancourt was familiar with the failed attempt because it was used as a case study in his commando training. His instructor had prepared the Chilean terrorists who conducted the assault. They were members of the Manuel Rodríguez Patriotic Front, one of the South American terrorist groups the DGI used for special operations that could not easily be traced back to Cuba.

During the explosives phase of his course, Betancourt was posted at Punto Cero, a secret military base in Guanabo, east of Havana. (It has the same name as Fidel’s family compound at Jaimanitas but is entirely distinct.) Betancourt told me:

They explained to us the assassination attempt on Pinochet and why it had failed. It was because the Chilean terrorists changed the attack plan at the last minute. They should have placed the explosives in the path of Pinochet’s motorcade and then used firearms to finish him off. But they reversed it, and fired first. My instructor said that if they had done it right it would have been successful.

No Cubans participated. But the planning and training had all been done at the Cuban base. Cuban Special Troops delivered the Vietnam-era American weapons used—aboard a vessel of the Cuban fishing fleet—to an isolated spot on Chile’s northern Pacific coast.

The Guardian newspaper in London described the assault as “dramatically cinematic in its execution.” Pinochet’s heavy armored vehicle came under a rain of machine-gun fire and was jolted by at least one grenade explosion. Reportedly bazookas and rocket launchers were also used. The dictator, accompanied by his young grandson, was slightly wounded but went on to serve another three years in office. Five of his bodyguards were killed and eleven others were wounded. According to Maragon, “the escape plan worked very well.” All the attackers managed to flee safely back to Cuba.⁴⁵