"Wouldn't it be better done at working level?" asked F.J., but Hollis' jaw was set firm, and although Arthur tried to move him, it was clearly a waste of time.
"I have heard the arguments. My decision is made," he snapped, glowering at Arthur across the table.
Hollis left for the United States almost immediately, where he briefed John McCone, the new Director of the CIA following the removal of Allan Dulles after the Bay of Pigs, and Hoover. Shortly afterward Arthur followed on to brief the Bureau and the Agency at working level. He got a rough reception. The Americans simply failed to understand how a case could be left in such an inconclusive state. Here, allegedly, was one of the most dangerous spies of the twentieth century, recently retired from one of the prime counterespionage posts in the West, and yet he had not even been interrogated. The whole affair smacked, to them, of the kind of incompetence demonstrated by MI5 in 1951, and in a sense they were absolutely right.
Hollis returned, determined to resolve the case. He ordered a new review to be written by Ronnie Symonds, and Symonds was specifically instructed not to communicate or cooperate with either Arthur or me in the research and drafting of the new report.
When the Mitchell case was handed over to Symonds, I returned to the Directorate of Science, where I was informed that Willis had made a change in procedures. He felt the Directorate need no longer involve itself in GCHQ's affairs, and wanted me to relinquish all contacts with the organization. I was incensed, I knew that unless MI5 hunted and chivvied for facilities and cooperation from GCHQ, things would soon slip back to the desperate state that existed before 1955. Few officers inside MI5 had any real idea of what could be done for them by GCHQ, and, equally, few GCHQ people bothered to think what they could do for each other, a job which I felt was vital for the Directorate to continue. But Willis could not be shifted. He wanted me to leave Counterclan, and join the bureaucrats. It was the final straw. I went to see Hollis, and told him that I could no longer continue to work in the Directorate. I told him I wanted to join D Branch if possible, or else return to A Branch. The Mitchell case gave me a taste of research, and I knew that the position as head of D3 was still vacant. To my surprise, Hollis offered me a transfer to D3 immediately. There was just a small caveat. He wanted me to return to the Directorate to finish one final special project for Willis, before taking up my new post in January 1964.
Willis' special project turned out to be one of the most important, and controversial, pieces of work I ever did for MI5. He wanted me to conduct a comprehensive review, to my knowledge the only one that was ever done inside British Intelligence, of every scrap of intelligence provided by yet another defector to appear in the West in the early 1960s - Oleg Penkovsky.
Penkovsky was, at the time, the jewel in MI6's crown. He was a senior GRU officer who spied in place for MI6 and the CIA during 1961 and 1962, providing massive quantities of intelligence about Soviet military capabilities and intentions. It was hailed on both sides of the Atlantic as the most successful penetration of Soviet Intelligence since World War II. Penkovsky alerted the West to the presence of Soviet missiles in Cuba, and his information about the Soviet nuclear arsenal shaped the American approach to the subsequent Cuban missile crisis. He also provided the evidence for the identification of the Russian missiles in Cuba. But in late 1962 Penkovsky and a British businessman, Greville Wynne, who was his cutout to MI6, were both arrested by the KGB, and put on trial. Wynne was given a long prison term (although he was eventually exchanged for Gordon Lonsdale and the Krogers) and Penkovsky, apparently, was shot.
I had been involved in the Penkovsky case during the time it was running. Penkovsky visited London on a number of occasions, as a member of a Soviet trade delegation, and had a series of clandestine debriefings with MI6 and CIA officers in the Mount Royal Hotel. At the time Hugh Winterborn was absent for a prolonged period through ill-health, and I was Acting A2, and was asked by MI6 to provide the technical coverage for the London Penkovsky operations. I arranged for continuous Watcher coverage of him and for the sophisticated microphoning system needed to capture every drop of intelligence that spilled out of him during the tense all-night sessions with his controllers.
The Penkovsky case ran counter to everything which was alleged about the penetration of MI5. Arthur and I often discussed this during the Mitchell case. If there was a high-level penetration, then Penkovsky had to be a plant, because news of him was known to the handful of senior suspects, including Mitchell, from a relatively early stage. When I was arranging the Mount Royal operation, Hollis asked me for the name of the agent MI6 were meeting, and I gave it to him. Cumming also asked, but since he was not on the MI6 indoctrination list, I refused to give it to him. This provoked a furious row, and Cumming accused me of becoming too big for my boots. He seemed to resent the fact that I did not consider myself in his debt for the role he played in hiring me into the Service.
Penkovsky seemed to fit into the most far-reaching of the allegations made by Golitsin. Golitsin said that in December 1958 Khrushchev transferred the head of the KGB, General Serov, to run the GRU. His replacement in the KGB was Alexander Shelepin. Shelepin was a much more subtle, flexible man than Serov, who was an old-style Beria henchman, a "nuts and bolts" man. The problem set to Shelepin was that Khrushchev and the Politburo had come to the conclusion that an all-out war with the West was not on. Khrushchev wanted to know how Russia could win without doing this. Shelepin took six months to survey the problem. He then called a large conference in Moscow of all the senior KGB officers the world over and discussed ways in which KGB methods could be modernized. Shelepin, according to Golitsin, boasted that the KGB had so many sources at its disposal in the West that he favored returning to the methods of the OGPU and the "Trust" as a means of masking the real nature of Soviet strategic intentions.
As a result of the Shelepin conference, Department D of the First Chief Directorate of the KGB (responsible for all overseas operations) was formed, a new department charged with planning deception or disinformation exercises on a strategic scale. Department D was put under the control of 1. 1, Agayants, an old, much respected KGB officer. In 1959, Golitsin said, he approached a friend who worked in this new Department to see if he could get a job there. The friend confided in him that Department D was planning a major disinformation operation using the GRU, but that it could not be implemented for some time because the GRU was penetrated by the CIA and this must be eliminated first. This penetration was almost certainly Colonel Popov, a high-ranking GRU official who spied in place for the CIA before being captured, tortured, and shot in 1959.
In fact, Golitsin never went back, as by then he was planning his defection, so he never learned any more details about the planned disinformation plan, other than the fact that it was basically a technical exercise, and involved all resources available to the First Chief Directorate. When Golitsin reached the West he began to speculate that the Sino-Soviet split was the Department D plan, and that it was a ploy designed to mislead the West. Some of Golitsin's admirers, like Arthur, believed (and continue to believe) this analysis, but although I was, during this early period, one of Golitsin's fervent supporters in the Anglo-American intelligence community, it has always seemed to me that the Penkovsky operation is a far better fit for the type of task Department D was set up for, than the inherently unlikely Sino-Soviet hypothesis.