Electronic Assassinations Newsletter

Issue #2 New Discoveries in the Recently Released Assassination Files



Peter Dale Scott

This piece was originally published in: Deep Politics II: Essays on Oswald, Mexico and Cuba (Essay completed April, 1994)

Right-Wing Conspiratorial Pressures on the CIA

In the preceding two chapters I have argued that, beginning some two or three months before the assassination, events attributed to Oswald were systematically misrepresented in CIA files. These misrepresentations appear to have been part of an intelligence operation, whether one run by the CIA or possibly some other agency.

However these misrepresentations need not necessarily have been conscious preparations for the "lone assassin" phase-two account of the Kennedy assassination. One can imagine an alternative version of events, in which some or all of the authors of the misrepresentations are not themselves part of a complex assassination conspiracy (involving a "phase one" story about Oswald and Kostikov), but the victims of such a conspiracy.

This alternative version supposes a force outside the CIA, but knowledgeable about CIA operations and procedures, and possibly represented within its ranks. In such a situation someone could embarrass the CIA into evasive procedures, delays, and even falsifications.

Let us pursue the hypothesis that the CIA had mounted a counterintelligence operation involving Oswald, or the Soviet Embassy in Mexico City, or the Cuban Embassy there. And let us return to the distinction raised by the authors of the Lopez Report, that Oswald visited the Cuban and Soviet Embassies on September 27, but that the man who identified himself as Lee Oswald on October I (and allegedly "spoke with consul whom he believed be... Kostikov") was someone else, an impostor. (1)

If so, the second man may well have been part of a plot, launched outside the CIA, to implicate Oswald as the patsy in the assassination. If Oswald was part of a different, authorized CIA operation, then the evasive behavior of Egerter, Roman, et al. would be understandable. The standard CIA procedure of reporting such Embassy contacts to the FBI would have put the authors of the October 10 messages in a bind; they did not want the Oswald-Kostikov link to be investigated, because in the resulting "flap" the authorized Oswald operation would be blown.

There are indications that through the immediate post-assassination period the CIA continued to be subjected to embarrassing pressures from "phase one" advocates outside, but close to, the Agency. A long CIA memorandum of 11 December 1963 welcomed the announcement by the New York Times one day earlier that the FBI had found Oswald to be categorically the lone assassin, and not the agent of any foreign government. The memo continued:

These disclosures presumably eliminate the possibility of further confrontations with Mr. Robert Slusser. In the event that Mr. Slusser continues to insist that the President was murdered by the Soviet secret police, the following additional negative indications and observations may be of some value. (2)

The memo continued for three and a half single-spaced pages to argue against the KGB "phase one" hypothesis, suggesting by its thoroughness that the confrontations with Mr. Slusser had been taken seriously.

A published authority on Soviet affairs, Robert Slusser was almost put into a position to lend credibility to his hypothesis. Early FBI reports about Lee Harvey Oswald's brother Robert indicate that at one point Mr. Slusser was about to be hired to write Marina's story. Eventually, after what looks like intrigue, the contract went instead to Priscilla Johnson (later Priscilla Johnson McMillan). Her book, long delayed in its appearance, corroborated the FBI's and Warren Commission's "phase two" finding that Oswald acted alone. (3)

Other right-wing sources, often explicitly hostile to the CIA, kept alive the phase-one specter of a link between Oswald and either Soviet or Cuban intelligence. From as early as December 1963, the CIA itself was blamed by such sources, either implicitly or explicitly, for its part in the President's murder. John Martino, an active plotter against Castro with a mob background, surfaced one such story in December 1963. blaming the President's death on Castro's response to a plot between Kennedy and the Soviets to have Castro replaced in Cuba by Huber Matos, a former Castro ally now detained in a Cuban jail. (4)

This alleged plot was a veiled allusion to the AMTRUNK plan mounted by the CIA and Robert Kennedy in 1963 to use old allies of Matos to overthrow Castro. (5) But Martino the source is perhaps more interesting than his story. In 1963 he had been receiving support for his own anti-Castro operations from Julien Sourwine, Chief Counsel for the Senate Internal Security Subcommittee. (6) It was this Committee, the reader may recall, whose phase-one interests were Johnson's reason (or pretext) for setting up the Warren Commission.

There was no shortage of such allegations, though they often came back to the same sources. Early in 1964, Colonel Philip J. Corso, a veteran of Army Intelligence and old foe of the CIA, told his friend Julien Sourwine (who in turn told the FBI) that Oswald was tied to a Communist ring inside the CIA. (7) Congressman Michael Feighan, for whom Corso worked as an aide, was reported to be "very curious concerning an intelligence report to the effect that there is a special school for assassins near Minsk, Russia, where Oswald worked in a factory." (8) An article in the journal of the John Birch Society, whose author Revilo Oliver later cited sources among veterans of army intelligence and the FBI, also argued that Kennedy's murder "was part of a Communist plot engineered with the help of the Central Intelligence Agency," and cited the "fake 'revolt'" plotted by Kennedy and Khrushchev to replace Castro with a crypto- Communist "'agrarian reformer.'" (9)

Thus the right-wing pressures which forced the Warren Commission into being continued to play on it throughout its existence. And insofar as one can detect a common source for all these stories, that source would appear to be not only outside the CIA but extremely hostile to it.

The word "outside" here can however be misleading. Every one of the allegations here summarized drew on inside information. For example John Martino reported in his December 1963 article that Oswald had tried to penetrate the anti-Castro Cuban group JURE: this claim was not generally known at the time but it was later corroborated by Silvia Odio's account of her meeting with Oswald in September 1963. Furthermore the Kennedys and the CIA had a plan (AMTRUNK) to oust Castro, which would have used, among others, the forces of JURE. The plan was still on-going in 1964, and thus extremely sensitive. Above all it planned to install a new government which would be free from mob influence, a detail which was sufficient to incur the hostility of mob allies like John Martino.

These so-called "outsiders" knew enough about the ways of government, and specifically the CIA, to embarrass it into cover-up. It seems likely therefore that somewhere they had their spies inside government, and possibly inside the CIA.

The Most Likely Manipulator: David Atlee Phillips

So far this discussion has focused on those "phase one" stories linking Oswald to Soviet or Cuban intelligence which at the time existed uniquely in government files, and which for a while the U.S. Government took seriously. We have not yet mentioned the veritable blizzard of similar stories which reached the FBI and CIA from external sources after the assassination. After November 24 there were still more "phase one" stories attributing a similar role to Jack Ruby. And to all these anti-Communist stories denouncing the KGB and Cuba one must add those stories with an opposite political spin, linking Oswald and/or Ruby to right-wing Texas millionaires, oilmen, anti-Castro Cubans, the mob, or the right-wing terrorist Minutemen. Most of these leads did not check out.

There were so many such false leads that one might be easily tempted to write them all off as meaningless "noise." However House Committee researcher Dan Hardway chose to look closely at all the stories that came out of Mexico City and Miami connecting Oswald with Soviet or Castro intelligence. According to his colleague Gaeton Fonzi, "Hardway's research had indicated that most of the individuals originating the reports" were assets of the Mexico City Station's Chief of Covert Action and Cuban Operations, David Phillips. (10)

Hardway had the opportunity to quiz Phillips about this at an informal Committee interview, with Fonzi present. Hardway was armed at the interview with documentation from the Agency to dispute Phillips' claim that these assets had been run by other CIA agents. After the session, Hardway told Fonzi,

I'm firmly convinced now that he ran the red-herring, disinformation aspects of the plot. The thing that got him so nervous was when I started mentioning all the anti-Castro Cubans who were in reports filed with the FBI for the Warren Commission and every one of them had a tie I could trace back to him. (11)

To date I have been unable to contact Dan Hardway, although another good source has confirmed that he did conduct this research. It is also clear that a number of the "phase one" stories linking Oswald to Cuba did come from a single milieu of anti-Castro Cubans in Miami close to, and in some cases supported by, the CIA's JM/WAVE station there. David Phillips does therefore seem a likely candidate to have coordinated the stories coming out of Mexico City and Miami. For in the second half of 1963 he was cross-posted to both stations, as Chief of Cuban Operations in Mexico City, and as Chief of Psychological Operations (i.e. propaganda) in Miami. (In fact it is possible that David Phillips held down three posts in 1963, and was doubling also as a member of the Special Affairs Staff Counterintelligence (SAS/CI) staff.)

A small intelligence-backed "press agency," the Agencia de Informaciones Periodisticas (A.I.P.), was a source for one recurring Oswald story, that he had worked on behalf of Cuban intelligence in the Miami area. (The A.I.P. attracted notice again during the wave of Chilean-financed Cuban terrorism of the mid-1970s, involving many Cuban exile veterans of the JM/WAVE operations, when the A.I.P. was revealed to be an agency by then financed by the Chilean intelligence service DINA). (12) The story was traced by the FBI to Fernando Fernandez Capada of the A.I.P., who told it to Jim Buchanan, a close ally of Frank Sturgis; the story was later publicized by Frank Sturgis and John Martino. (13)

Another A.I.P. story, traced to Dr. Fernando Carrandi, spoke of Ruby's travel to Cuba. Those involved in circulating this story included Salvador Lew, p.r. agent for the CIA-backed Comandos Mambises, and Paul Bethel, described by Fonzi as "a close friend of David Atlee Phillips." (14) Yet another Oswald-Cuban intelligence story involved Miguel "Cuco" de Leon, senior adviser to Manuel Artime in the JM/WAVE-backed Operation Second Naval Guerrilla. (15)

Any evidence for linking Phillips to these intelligence-tinged stories has not yet been made public. We have however Phillips' own statements that he was involved in the transmission of both of the key "phase one" allegations promoted in CIA cables, the Kostikov story of October, and the Alvarado story of November 25.

As mentioned above, it would appear that Phillips' claim to have signed off on the Kostikov cable of October 8 is simply not true. Phillips claimed this in sworn testimony, as part of his effort to rationalize the delay of one week in transmitting the intercepted conversation of October 1. (16) Phillips' admitted role in the transmission of the Alvarado story, that Oswald was paid money in the Mexico City Cuban Consulate to kill Kennedy, is however corroborated by the documentary record. Here too there is a difference between Phillips own account and the cables however. In his autobiography Phillips describes the story he heard from Alvarado's lips as a lie easily seen through, indeed as a "transparent operation." (17) In the cables sent after his interviews with Alvarado, however, the tone is quite different. There we hear that "This officer was impressed by Alvarado ... wealth of detail Alvarado gives is striking." (18) One cable described Alvarado as a "quiet, very serious person, who speaks with conviction;" another, the next day, called him "completely cooperative." (19)

Most revealing was the description of Alvarado as a "well-known Nicaraguan Communist underground member," whereas in fact (as he himself revealed later the same day) he was a penetration agent of the right-wing Somoza Government of Nicaragua. (20) (This revelation was quickly confirmed by CIA cables from Managua and Headquarters). (21)

Winston Scott, Ambassador Thomas Mann, and the Mexican DFS

Assuredly Phillips was not alone in backing the Alvarado story at the time. Ambassador Thomas Mann, together with Station Chief Win Scott and FBI Legal Attache Clark Anderson, sent a Flash cable on November 26 suggesting that Silvia Duran should be rearrested in order to corroborate it:

We suggest that the Nicaraguan be put at the disposition of President Lopez Mateos on condition that Lopez Mateos will agree to order rearrest and interrogate again Silvia Tirado de Duran along following lines: A. Confront Silvia Duran again with Nicaraguan and have Nicaraguan inform her of details of his statement to us. B. Tell Silvia Duran that she is only living non-Cuban who knows full story and hence she is in same position as Oswald was prior to his assassination; her only chance for survival is to come clean with whole story and to cooperate completely.... Given apparent character of Silvia Duran there would appear to be good chance of her cracking when confronted with details of reported deal between Oswald, Azcue, Mirabal [the two Cuban consuls] and Duran and the unknown Cuban negro [described by Alvarado]. If she did break under interrogation - and we suggest Mexicans should be asked to go all out in seeing that she does - we and Mexicans would have needed corroboration of statement of the Nicaraguan. (22)

Mann on his own went on to recommend the arrest of three Cuban members of the Cuban consulate, and later to argue forcefully that Castro was the "kind of person who would avenge himself " by assassinations. (23)

These cables were in defiant opposition to the cooler approach in Washington. Headquarters had already tried to oppose the original arrest of Duran, rightly fearing that the arrest (and interrogation by the Mexican secret police, or DFS) "could jeopardize U.S. freedom of action on the whole question of Cuban responsibility." (24) Headquarters replied again to the new Duran cable, warning the Station Chief that the Ambassador was pushing the case too hard, and his proposals could lead to an international "flap" with the Cubans. (25)

Headquarters were absolutely right in their concern that the Mexican DFS were out to "prove" an international conspiracy involving Oswald with Cuba. Silvia Duran later confirmed that in their interrogations of her:

all the time they tell me that I was a Communist ... and they insisted that I was a very important person for ... the Cuban Government and that I was the link for the International Communists - the Cuban Communists, the Mexican Communists and the American Communists, and that we were going to kill Kennedy, and I was the link. For them I was very important. (26)

In its performance however, the DFS was almost certainly (as Edwin Lopez has since corroborated to me) tightly controlled by the CIA Station. The DFS was part of the Mexican Ministry of the Interior, or Gobemacion; its Minister, Gustavo Diaz Ordiz, was a CIA asset, and also a close friend of Station Chief Win Scott (the best man at Scott's third wedding), as well as of Ambassador Mann - and Lyndon Johnson. (27) Details of Duran's interrogation suggest that the DFS, seeking to prove her conspiratorial involvement, was being fed clues by the Americans. (28)

Given the predisposition of the DFS to find a conspiracy, a fact known even in Washington, and given the well-known brutality of DFS interrogation methods (which included torture), it is particularly revealing that Mann and Scott would recommend asking the DFS "to go all out in seeing that she [Duran] ... break under interrogation." Circumstances suggest that the documentary record here is incomplete, in at least two respects:

1) Contrary to the records we now have, Duran had already been tortured, and had already "confessed" to a sexual involvement with Oswald, since expunged from the record.

2) Mann's apparently reckless defiance of official instructions against the arrest of Duran was probably based on unofficial guidance from a very high level in Washington.

A Suppressed "Phase One" Story: Oswald's Alleged Sexual Liaison

Alvarado introduced a sex angle into his fantastic story about seeing Oswald be paid $6,500 to kill someone. He spoke of a "pretty girl" in the Consulate (whose manners reminded him of a "prostitute") who had given Oswald an embrace and also a home address "where he could find her." (29)

In 1967, transmitting an agent's report of an interview with a source who knew Duran, Win Scott commented:

The fact that Silvia DURAN had sexual intercourse with Lee Harvey Oswald on several occasions when the latter was in Mexico City is probably new, but adds little to the OSWALD case. The Mexican police did not report the extent of the DURAN-OSWALD relationship to this Station. (30)

Scott's choice of words ("fact," "extent") is indicative of earlier events involving Duran that have not hitherto been publicly reported.

In fact Scott had both misrepresented what the informant apparently said (reporting sexual relations with Oswald, but not "on several occasions"), and suppressed its most important revelation, that she had been tortured by the DFS until she "admitted that she had had an affair with Oswald:"

[Long redaction] XXXX continued that Silvia Duran informed XXXX that she had first met Oswald when he applied for a visa and gone out with him several times since she liked him from the start. She admitted that she had sexual relations with him but insisted that she had no idea of his plans. When the news of the assassination broke she stated that she was immediately taken into custody by the Mexican police and interrogated thoroughly and beaten until she admitted that she had had an affair with Oswald. (31)

It is noteworthy that Scott, far from rebutting the torture allegation, apparently accepted it as a fact, and one not worth commenting on.

The Lopez Report, in transmitting this interview, commented that "Silvia Duran admitted that the Mexican police had questioned her on this point but denied that she had had an affair with Oswald." (32) This account is confirmed by its cited source, Silvia Duran's interview of June 6, 1978.

Cornwell: Did the officers from the Securidad Department ever suggest to you during the questioning that they had information that you and Oswald had been lovers?

Tirado [Duran]: Yes, and also that we were Communists and that we were planning the Revolution and uh, a lot of false things. (33)

Curious as to why Ms. Duran had not been asked about the torture, I contacted Edwin Lopez, who had translated at the interview. He confirmed that, off the record, Ms. Duran had said that she was tortured badly, and that indeed in recalling this she had broken down and wept. She had however declined to say anything about the torture on the record because, as a citizen and resident of Mexico, she feared reprisal.

One hesitates now to make any revelation that would put Ms. Duran at risk. The issue however is an important one. According to the account which Scott accepted as "fact," she was not only tortured on the matter of the liaison, but coerced into admitting it. If Scott's blase comment is true (this "adds little to the OSWALD case"), then the accounts of her confession have probably been altered, to convert a suppressed "phase one" story of a sexual liaison into the innocuous "phase two" version published by the Warren Commission. (34) Lending credence to this hypothesis is the known fact that the published version was censored and rewritten (by the CIA, according to the Lopez Report) on at least one other point, Duran's original description of Oswald as "blonde and short." (35)

Were Mann and Scott Backed in Their Defiance of Official Instructions?

All this lends dramatic urgency to the question of whether or not Scott and Mann were "acting alone" in their defiant recommendation, against earlier official instructions from Headquarters, that Duran be rearrested by the DFS, and coerced into corroborating the Alvarado story.

One interpretation of the known facts is to postulate a real division within the Administration, between "phase one" enthusiasts like Scott and Mann (who wanted to ask the DFS "to go all out") and "phase two" pragmatists like Karamessines, who struggled in vain to prevent the arrest and rearrest from taking place.

The chief problem with this analysis is that Scott and Mann drew no disapprobation for their course of action. Scott remained in his post as Chief of Station until his retirement six years later. Mann, far from being rebuked, was swiftly promoted by the new President, Lyndon Johnson, on December 14, 1963, to become the new Undersecretary of State for Latin American Affairs. (Mann's promotion was the more dramatic because unexpected; he had earlier announced, under Kennedy, his plans to retire at the end of the year). (36)

A second interpretation of the facts is that beneath the apparent contest of opposing forces, "phase one" and "phase two," a higher authority was manipulating the Alvarado story, backed as it was by Scott and Mann, towards the desired "phase two" outcome of the Warren Commission and Report. I truly do not know whether or not such a higher authority existed. If it did, however, it almost certainly involved Lyndon Johnson.

Lyndon Johnson was a close personal friend of the soon-to-be-elected Mexican President Gustavo Diaz Ordaz, who has been described as the most right-wing (and pro-American) President since Miguel Aleman in the early 1950s. Just as Diaz Ordaz maintained tight control over the DFS (along with his good friend Win Scott), so Johnson was the friend and hope of those in the CIA who thought that Kennedy had been wrong to dismiss Allen Dulles after the Bay of Pigs fiasco.

As Johnson barely spoke Spanish, he relied in his meetings with Diaz Ordaz on the translating ability of a fellow Texan, Thomas Mann. As Mann later told author Dick Russell,

Lyndon Johnson had lines into Mexico that I knew nothing about. He was an amazing man. He didn't speak Spanish, but he was a good friend of [Gustavo] Diaz Ordaz, who became President of Mexico. He used to come down and see Johnson at the ranch several times, and Johnson would have me down to translate. (37)

(For what it is worth, former KGB Colonel Oleg Nechiporenko reports that his DFS contact told him that "many in the DFS felt that Lyndon Johnson was responsible" for the assassination). (38)

Johnson was not particularly close to the CIA as an Agency. His lack of interest in intelligence estimates has been cited as a reason for the resignation of CIA Director John McCone in 1965, and Johnson's replacement of him by an inept outsider, Admiral William Raborn. (39) In 1966 Johnson did however give the CIA its first Director who was also a career officer, Richard Helms. Helms had been close to Dulles since their days together in Germany with OSS. (40) (Helms later revealed that Johnson had explained to him in 1965 that Raborn was a "temporary measure," and that Johnson would appoint Helms when he had proved himself as Deputy Director). (41)

What remains unknown is the extent of the new President's knowledge of the "phase one" rumors which, as he informed Earl Warren, were "floating around." If he had any intimate knowledge of either the Kostikov story or the Alvarado story, he must have known that a true investigation of the case would have to be at arms length from the CIA. Instead Johnson named Allen Dulles to the Warren Commission. Dulles' strategic location was to play an important role in the CIA cover-up that ensued. If there was a conspiracy to ensure such a cover-up, then the naming of Dulles to his new post was almost certainly part of it. (42)

However important the personal connection between Johnson, Diaz Ordaz and Thomas Mann, it could never, however, have explained the strange falsifications of CIA messages that occurred at CIA Headquarters. To explain that phenomenon we must look inside the Agency itself.

Such a program of falsification and subsequent cover-up could have been coordinated, I shall suggest, by those who were closest to former Director Allen Dulles.

The Dulles-Angleton-Hunt-Phillips "Agency--Within-the-Agency"

In 1963 the "responsible" press, the New York Times and the Washington Post never commented critically on Johnson's choice of Allen Dulles, the most important official fired by John F. Kennedy, to serve on the Commission investigating the President's murder.

Even though one would never expect them to play this critical role, they should have, for Allen Dulles was perhaps the Kennedys' most powerful enemy in the U.S., arguably more powerful even than the new President. Dulles had resented his being made to take the blame for the Bay of Pigs fiasco: "He thought other people should be resigning before he did, and made it clear that he was thinking of one person in particular, Robert Kennedy." (43)

Before the assassination, Dulles had fought back in the media, leaking his resentment against the Kennedys to the sympathetic ears of Charles J. V. Murphy of Fortune magazine, part of Henry Luce's Time-Life empire. Murphy's pro-Dulles apologia, "Cuba: The Record Set Straight," was simultaneously a piece lobbying for escalated U.S. involvement in Indochina, just before Kennedy's first major Vietnam decision. (44) In this counter-attack, Dulles had Agency support. Dulles asked to have one of his CIA proteges, E. Howard Hunt, go over Murphy's article in detail; and Hunt was accordingly instructed to do so. (45)

If Hunt was close to Dulles, he was even closer to his own protege, David Atlee Phillips. In fact it was probably through Hunt that Phillips became "an active player in a small clique within the CIA hierarchy who were almost autonomous in their operational capabilities," an OSS brotherhood of whom Allen Dulles, inside the Agency or out, was the acknowledged leader. (46) What merits further investigation is that members of this brotherhood played key roles on both sides of the Oswald "phase one"-"phase two" dialectic.

The key to Dulles' "agency-within-the-Agency," as Aarons and Loftus have called it, was the power Dulles had conferred on his close friend Jim Angleton. (47) As Counterintelligence Chief Angleton was authorized to spy on the rest of the CIA, and maintain a CI network of assets in other branches. The close connection between Dulles and Angleton endured well beyond Dulles' departure from the Agency. (48)

One sign of in-house CIA intrigue over the assassination is that those responsible for falsifying the Oswald-Kostikov story were not punished, or even distanced from the investigation of Kennedy's murder. On the contrary, John Scelso of WH/3, the Mexico desk, and Birch D. O'Neal, the head of CI/SIG, both involved in the falsified October messages from Headquarters, were assigned after the assassination to key roles in the CIA investigation and resulting liaison with the FBI. (49)

The man responsible for these assignments was Deputy Director of Plans Richard Helms, another Dulles loyalist and OSS brother. It is not clear that Helms' role was conspiratorial. On the contrary, while Scelso may have encouraged the proliferation of "phase one" Oswald stories, Helms appears to have constrained them. (50) What remains to be explored is whether these two apparently opposing efforts were actually part of a single coordinated scenario.

Helms's assignment of Scelso and O'Neal to the investigation made the same kind of sense as Johnson's putting Dulles on the Warren Commission. On the Commission, it is generally conceded, Dulles actively covered up the CIA involvement in the CIA-Mafia plots against Castro. (51) The House Committee, in an Appendix to its Report, concluded that Helms himself, "though the main contact with the Commission, apparently did not inform it of the CIA plots to assassinate Castro," and found a further "indication that his testimony before the Commission was misleading." (52)

Helms and Angleton designated Angleton's Chief of Research and Analysis, Ray Rocca, to be the CIA's point of contact with the Commission. (53) Angleton clearly hoped by doing so to prevent a number of highly relevant counterintelligence operations from being exposed, such as the CIA's illegal HT/LINGUAL mail-opening program (overseen by Birch D. O'Neal), and the photographic and electronic surveillance of the Soviet and Cuban Embassies in Mexico City. (54)

Angleton also visited Dulles on instructions from Helms, in order to learn and prepare for the questions which Dulles thought the Commission might put to the CIA. (55) Angleton's consistent approach was. in Rocca's words, "to wait out the Commission." (56) One might have expected as much from the man who would later tell the Church Committee, "It is inconceivable that a secret intelligence arm of the Government has to comply with all the overt orders of the Govemment." (57)

Having observed how closely the Dulles-Helms-Angleton network controlled the Warren Commission after the assassination, one is moved to ask about certain pre-assassination personnel movements, presumably authorized by Helms, which affected the Oswald-Kostikov story. One crucial move was the recall in 1963 of Tennant Bagley from Berne to Langley, where he was rapidly promoted to chief of the Counterintelligence Branch of the Soviet Division" (C/SR/CI). (58) This promotion came in time for him to suppress mention of Kostikov in the October messages, and then sound the assassination alarm about Kostikov on November 23.

Another move at this time was the temporary duty assignment of David Phillips, the Chief of Cuban Operations and Covert Action at the Mexico City Station, to Washington and Miami, "from at least late September to October 9, 1963." (59) In view of allegations about Hunt's Counterintelligence activities at this time (see below), it is relevant that while in Washington Phillips appears to have been attached to the Counterintelligence Staff of Fitzgerald's Special Affairs Staff devoted to anti-Castro operations.

Then there is the much disputed question of whether, as Tad Szulc has alleged, Howard Hunt was assigned to temporary duty in Mexico City for the period of August and September 1963, at the time of Oswald's alleged visit there. (60) Both Hunt and the CIA have strongly denied this claim. It is however supported by the sworn testimony of David Phillips in a libel suit, that he had seen Hunt in Mexico City at the time Hunt denied being there. (61) In a 1973 House Watergate Hearing, Hunt testified how a retired CIA agent "had during the Cuban operation been my inside man in the Embassy when I was outside in Mexico operating as part of the Cuban task force." (62)

An even more dramatic allegation, also strongly disputed, is that Hunt was in Dallas on November 22, 1963, at the time of the assassination. According to reporter Joseph Trento, a secret CIA memo of 1966, said to have been initialed by Angleton and Helms, emphasized the importance of keeping Hunt's presence there a secret, and suggested a cover story to provide Hunt with an alibi. (63) According to author Dick Russell, Trento later told him that Angleton himself was the source of the story, and arranged for a copy of the internal CIA memo to be delivered to him, as well as the House Committee. (64) If this is true, Angleton's role is sinister, and apparently part of a cover-up, whether the memo is real (and Hunt was in Dallas), or whether it was disinformation (and Hunt was not).

Trento told Russell he understood from Angleton that Hunt was in Dallas because "of a serious counterintelligence problem with the [CIA] Cubans," some of whom were known to be "penetrated by Castro's intelligence." (65) Far-fetched as an explanation to justify Hunt's presence in Dallas, it would make sense of his temporary detachment to Mexico City, where a number of JURE Cubans, suspected by Hunt and Angleton for their left-leaning politics, were preparing to take part in a Bobby Kennedy-backed operation against Castro. It would indeed have been characteristic of Angleton to use a CIA officer like Hunt, not nominally part of the Counterintelligence Staff, to spy on left-leaning CIA-sanctioned operations. And Hunt's animosity against the Cuban Manuel Ray of JURE, conceded by Hunt himself in his memoir Give Us This Day, was well-known throughout the Agency.

The CIA itself has said that Hunt's title at this time was Chief of Covert Operations for the Domestic Operations Division headed by Dulles' old friend Tracy Barnes. (66) Szulc however has wtitten that Hunt was asked to assist Dulles in writing a book, The Craft of Intelligence, that Dulles wrote following his involuntary retirement in 1961. (67) Just how long it took to complete the book is not clear; it was however published in 1963. Certainly the book would have given Hunt the opportunity to spend many long hours (presumably on Company time) with Dulles, his former boss.

A third person who would presumably have been present would have been Howard E. Roman, Dulles' close friend and alleged collaborator on the book. (68) Another member of the OSS "Old Boy" brotherhood in the CIA, Roman resigned in 1962 before taking up the book-writing job with Dulles (and possibly Hunt). Roman went on to write a total of two books (and two more edited volumes) with Dulles. In that capacity he was with Dulles at the moment, on November 22, 1963, when Dulles heard of the President's murder. (69)

Roman's post-war career had been with Soviet matters, but I know nothing to connect him officially with the Lee Harvey Oswald files. The same cannot however be said of his wife, Jane Roman. A CIA official herself, it was Jane Roman who, as noted earlier, was the releasing officer on the falsified CIA cable to Mexico City on October 1963.

Conclusion: The "Phase One" Stories Affected History

Assuredly the new President was not prevaricating, or being over-cautious, when he spoke to Chief Justice Warren of the risk of war. "Phase one stories" were not just street rumors, they were being promoted energetically and almost conspiratorially, at at least the Ambassadorial level.

We need to insist that the promotion of such stories, per se, does nothing to link the proponents to the assassination. It is hardly surprising that opponents of Castro within the Government, along with anti-Castro Cubans in Miami, should seize this opportunity to reverse what they saw, rightly or wrongly, as the Kennedy policy of prolonged inaction.

With the pre-assassination Kostikov story, on the other hand, we can be more specific. Unlike the Alvarado and other false stories, the Kostikov story was never exploited to achieve a policy change. It remained a secret in government files, and those who spoke publicly of KGB involvement never referred to it. To say that the falsifications of the October 1963 CIA messages had something to do with the plot to kill the President does not tell us anything about the motives of those falsifying the cables. As said above, they may have been illegal conspirators, or they may have been responding to a potential embarrassment created for them by these conspirators.

One can reach one simple conclusion about these two alternative ways of reading the facts: The public has both the right and the need to know which of these alternatives is the true one.

The first person one would have wanted to interrogate under oath about these falsifications, and about other falsehoods in his own earlier testimony, would have been David Atlee Phillips. Mr. Phillips unfortunately has since died, as have Win Scott and other relevant witnesses. This only adds to the urgency of securing testimony under oath from those who survive.


1. Lopez Report, 242-50; MEXI 6453 of 8 October 1963. According to the full script, the name of Kostikov was actually raised by "Oswald"'s interlocutor, the Soviet Embassy guard Obyedkov, rather than by Oswald (Lopez Report, 79).

2. Memo of 11 December 1963 to Chief/Soviet Russia from Neil Huntley, C/SRI. "Additional Notes and Comments on the Oswald Case;" CIA Document #376-154.

3. Priscilla Johnson McMillan, Marina and Lee (New York: Harper & Row, 1977).

4. John Martino, "Cuba and the Kennedy Assassination," Human Events, December 21, 1963, 3.

5. The chief of these allies were the JURE leaders Manuel Ray, Ramon Barquin, and Napoleon Becquer (10 AH 137; cf New York Times, April 19, 1962).

6. Scott, Deep Politics, 116; 11 AH 65.

7. Scott, Deep Politics, 215. See Chapter VII.

8. The Cross and the Flag, March 1964, 31. The CIA investigated the rumor of a Minsk assassination school early on, and pronounced it to be untrue.

9. Scott, Deep Politics, 215; Revilo Oliver, American Opinion, March 1964; 15 WH 710.

10. Fonzi, 292.

11. Fonzi, 292-93.

12. Magnus Linklater et al., The Nazi Legacy (New York: Holt, Rinehart, and Winston, 1984), 278.

13. WCD 1020.1-6, 16; 26 WH 424-25; Scott, Deep Politics, 338.

14. WCD 916.2-3. Bethel was also involved in another false story of hit teams dispatched by Castro to kill Kennedy: see WCD 893.1-7; 22 WH 864, Scott, Crime and Cover-Up, 60-61.

15. WCD 770.7-9; Scott, Crime and Cover-Up, 18, 20.

16. Lopez Report, 127-28; Fonzi, 293. After interrogating Phillips informally on this issue, Hardway told Fonzi that, "based on the research he had done tracking the routing of the cables and the lack of credible answers about them from Phillips, he believed there was a strong possibility the cables were created after the fact" (Fonzi, 293).

17. David Phillips, The Night Watch (New York: Atheneum, 1977), 141-42; quoted in Scott, Deep Politics, 122.

18. MEXI 7104 of 27 November 1963; CIA Document #174-616.

19. Mexico City cable of 26 November 1963 (MEXI 7067?), retransmitted as DIR 85199 of 27 November, WCD 1000B.4; WCD 1000C.2.

20. Memo of 26 November, WCD 1000A; MEXI 7083 of 26 November.

21. MANAGUA cable of 26 November, 262237Z; DIR 85196 of 27 November 1963.

22. MEXI 7072 of 26 November 1963; CIA Document #128-590.

23. MEXI 7104 of 27 November 1963; MEXI cable of November 28 1963, Anthony Summers, Conspiracy, 441. Cf. 3 AH 569.

24. DIR 84916 of 23 November; Lopez Report, 185-86; Schweiker-Hart Report, 25.

25. DIR 85371 of 28 November 1963; Lopez Report, 187; Schweiker-Hart Report, 29.

26. 3 AH 91; Cf. 3 AH 86. Note that the DFS exempted the Soviets from their hypothetical conspiracy, as did Ambassador Mann (Summers, 441).

27. Scott, Deep Politics, 123; Philip Agee, Inside the Company: CIA Diary (Harmondsworth, Middlesex: Penguin, 1975), 274-75; Dick Russell, 454, 457-58. In late 1963 Diaz Ordaz was on leave as the Presidential candidate of the ruling PRI; but his replacement as Acting Minister, Luis Echeverria, was also a CIA asset on the CIA payroll.

28. For example, the Americans knew that Duran's name and the telephone number of the Cuban Consulate, 11-28-47, were in Oswald's address book (16 WH 54). Duran told the House committee that the DFS "asked me I don't know how many times, the way that I used to give my name and telephone number and they made me write and they take the paper out and then again, they ask me, how do you do this, and I write it down, and I give the paper. I think I do this five or six times" (3 AH 102).

29. MEXI 7067(?) of 26 November 1963; WCD 1000B.

30. Dispatch HMMA-32243 of 13 June 1967 from COS, Mexico City, to Chief, Western Hemisphere Division; CIA Document #1094-965.

31. TX-1937 of 26 May 1967, CIA Document #1084-965, reporting interview of informant in safehouse on 25 May, 1967. In 1964 the Station had also heard the allegation of an Oswald-Duran liaison from a dubious witness, Elena Garro, with strong DFS connections (Lopez Report, 207, 220; 3 AH 302).

32. Lopez Report, 254.

33. 3 AH 86.

34. 24 WH 565.

35. Lopez Report, 190; Scott, "The Lopez Report," 6.

36. Scott, Deep Politics, 94.

37. Russell, 454.

38. Oleg Nechiporenko, Passport to Assassination (New York: Birch Lane/Carol Publishing, 1993), 181.

39. John Ranelagh, The Agency (New York: Simon and Schuster, 1986), 422-23. Raborn is remembered in the Agency as the man who asked "Who's this fellow Oligarchy?' and who thought that "KUWAIT" was a CIA cryptonym.

40. Burton Hersh, The Old Boys (New York: Scribner's, 1992). 160-61.

41. Ranelagh, 448.

42. Johnson did invoke the threat of the Senate Internal Security Subcommittee as a major part of his case for a Warren Commission. As a man with years of Washington experience, he must have known of the on-going collaboration between Eastland and Sourwine of the Subcommittee with elements inside the CIA.

43. Leonard Mosley, Dulles (New York: Dial Press/James Wade, 1978), 473. Although Dulles had offered to resign at the moment of failure, the offer had been refused. He declined to offer his resignation again, after being rebuked in a secret in-house CIA review. Thus he was fired, and without prior warning (ibid.).

44. Charles J. V. Murphy, "Cuba: The Record Set Straight." Fortune, September 1961. Discussion in Paul W. Blackstock, The Strategy of Subversion (Chicago: Quadrangle, 1964), 250; Peter Dale Scott, The War Conspiracy (New York: Bobbs Merrill, 1972), 19-20.

45. Tad Szulc, Compulsive Spy: The Strange Career of E. Howard Hunt (New York: Viking, 1974), 95; E. Howard Hunt, Under Cover (New York: Berkley, 1974), 216.

46. Fonzi, 331, 346n. Cf. Scott, Deep Politics, 54, 67, 322. Phillips had not served in OSS; his mentor Hunt had.

47. Mark Aarons and John Loftus, Unholy Trinity (New York: St. Martin's Press, 1991), 260,

48. David Wise, Molehunt (New York: Random House, 1992), 39; Scott, Deep Politics, 67.

49. 11 AH 57, 476, 485.

50. Scelso's role is hard to assess. On November 23, 1963, when ordered by Assistant Deputy Director of Plans Karamessines to tell the Mexico City CIA Station to stop the arrest of Silvia Duran, Scelso entered a memo for the record, which said in part, "We phoned as ordered, against my wishes, and Scelso wrote a FLASH cable which we did not then send" (TX-1240 of 23 November 1963; C/WH/3 memo for record; emphasis added). On the other hand, he soon afterward prepared a summary report for Helms which was transmitted to President Johnson. "This report stated that Oswald probably was a lone assassin who had no visible ties to Soviet or Cuban intelligence though such ties could not be excluded from consideration" (11 AH 477).

51. Mosley, 477-78; Arthur M. Schlesinger, Robert Kennedy and His Times, 536, 663.

52. 11 AH 58.

53. 11 AH 47, 477-79.

54. 4 AH 215; 11 AH 476, 479, 491; AR 205.

55. 4 AH 232-35.

56. 4 AH 215, 232.

57. Hersh, Old Boys, 317; citing Seymour Hersh, New York Times Magazine, June 25, 1978.

58. Mangold, 170.

59. Lopez Report, 128.

60. Szulc, Compulsive Spy, 96, 99.

61. Mark Lane, Plausible Denial (New York: Thunder's Mouth, 1991), 193. Hunt had made the sworn statement, "I was not in Mexico between the years 1961 and 1970." Phillips testified under oath that he had seen Hunt in Mexico City "sometime between September of 1961 and March of 1965," adding that he "must have seen him once or twice" somewhere in Mexico prior to November 22, 1963.

62. Nedzi Hearing, 518 (June 28, 1973).

63. Wilmington Sunday News Journal, August 20, 1978; reprinted in Lane, Plausible Denial, 152-55; Dick Russell, 474-75. In his book Lane claims that former CIA official Victor Marchetti had told him about this memo prior to the Trento story, citing Marine intelligence Colonel William Corson as his source (Lane, 134-35). Corson was close to Trento. The two eventually were co-authors of the book Widows.

64. Ibid.

65. Ibid.

66. San Francisco Chronicle, December 31, 1974, 1.

67. Szulc, 95.

68. Mosley, 475-77.

69. Mosley, 477.

Return to Table of Contents

Return to Home Page