The Silence in Washington and London

The author argued that the hundreds of thousands of conversations recorded between May 1955 and April 1956 constituted a body of intelligence whose volume spoke for itself, and that further investigation into the possibility of Soviet manipulation would serve no purpose except to disturb a matter that had already been resolved.

With Blake removed from public view, the intelligence services were left alone with their internal divisions and the silent question of the tunnel’s value.

A memorandum circulated within the CIA’s counterintelligence staff in the months after the trial advised against reopening that question. The document did not specify who had resolved it, or when. Its tone was bureaucratic, its recommendation practical: let the matter rest. The phrase “at the present time” carried no deadline. It implied no future moment when the question would be taken up again. The line was, in its way, the most honest in the document, an acknowledgment that the present time, for institutional purposes, could be extended indefinitely.

The memorandum’s author wrote under constraints that had nothing to do with the tunnel’s technical performance. Blake’s trial at the Old Bailey had concluded in June 1961 with a sentence of forty-two years, the longest ever imposed by a British court for espionage. The proceedings, held in camera, had produced a sanitized narrative: one man, motivated by ideological conviction, had betrayed his service and his country. The prosecution presented no evidence about the tunnel. The court heard nothing about the KGB’s prior knowledge of Operation Gold, nothing about the possibility that Soviet military communications had been shaped for Western consumption, nothing about the eleven months of uninterrupted recording that had preceded the tunnel’s theatrical discovery. The trial’s structure ensured that the public record contained only individual betrayal. The institutional record, classified and internal, would have to absorb everything else.

The pressure to avoid a full accounting came from both sides of the Atlantic. In Washington, the CIA’s Berlin Operations Base had been the tunnel’s operational home. William King Harvey, who had directed the operation from December 1953, had built his reputation on it. Harvey’s colleagues in the Soviet Division and the technical staff who had designed the tap and managed the recording chain had invested years of work in the project. The tunnel represented, by every measure available to them, a triumph of engineering and operational daring.

A quarter-mile of shaft driven through saturated soil beneath the sector boundary from a warehouse in Rudow with an unconventionally deep basement. Three trunk cables tapped without detection near Altglienicke. Recording equipment that ran continuously for nearly a year. The facts existed in requisition forms, engineering reports, and the daily logs of the tunnel’s maintenance schedule. No subsequent revelation could make the shaft shorter or the tap less precise.

In London, the pressures were different but pointed in the same direction. The Secret Intelligence Service had co-designed the operation. Initial planning meetings had been held at No. 2 Carlton Gardens, from which the West German government was excluded because of the compromised state of their intelligence apparatus. The resulting agreement was that the United States would supply the engineering and the personnel, while the British would contribute intelligence expertise and target selection. Peter Lunn, the MI6 officer who had overseen the British side, had staked professional credibility on the operation’s value. Blake had been Lunn’s subordinate in Berlin, a case officer assigned to recruit Soviet and East German targets. The fact that Blake had reported the tunnel’s existence to the KGB before construction began meant that Lunn had vouched for a man who was actively betraying the operation Lunn had helped design.

The counterintelligence officers in both services saw the situation differently. Their concern was not the tunnel’s engineering but its intelligence product. If the KGB had known about the tap from the outset, the question was not whether the Soviets had allowed the tunnel to operate but what they had done with it. Two possibilities presented themselves, and they were not mutually exclusive.

The first was that Moscow had let the tunnel run solely to protect George Blake. Blake was a high-value penetration agent inside the Secret Intelligence Service, a man who had already demonstrated his willingness to betray operations, colleagues, and his country.

Exposing the tunnel to neutralize it would have meant exposing Blake, since the tunnel’s existence was known only to a small circle of officers and the KGB’s control of that circle was, by 1956, incomplete. Protecting Blake required letting the tunnel produce. Under this reading, the conversations recorded between May 1955 and April 1956 were genuine Soviet military communications, intercepted because the Soviets chose not to shut down the tap, not because they were feeding false information through it. The intelligence would be compromised in the sense that the Soviets knew it was being collected, but the content itself would be authentic.

The second possibility was darker. It held that the KGB had used the tunnel, not merely tolerated it.

Under this reading, Soviet military officers whose telephone lines ran through the tapped cables near Altglienicke had been instructed, or had their conversations structured, to convey specific impressions to the West. The tunnel’s output would then be not intercepted intelligence but engineered intelligence, shaped at its source to produce desired conclusions in Washington and London. The technical sophistication of the collection system would have made this easier, not harder. The tap was invisible to its operators. The recording chain ran without human intervention. The analysts who transcribed and translated the conversations in Washington had no way to assess whether the speakers on the other end of the line were speaking freely or performing for a microphone they knew was there. The system’s technical perfection had eliminated the very checks that might have detected manipulation.

The collection method was so advanced, so far beyond anything the Soviets could plausibly have anticipated, that its product acquired an aura of authenticity no one thought to question. The recording equipment did not lie. The cables were real. The voices were real. The question was whether the words those voices spoke were their own.

The counterintelligence argument found its strongest voice in officers who had no investment in the tunnel’s success. Within the CIA’s counterintelligence staff, led by James Angleton, the Blake case was not an anomaly but a pattern. Angleton’s office had spent years tracking Soviet penetrations of Western services, and Blake’s confession confirmed what Angleton had long suspected: that the KGB’s penetration of allied intelligence was deeper, more systematic, and more patient than anyone outside counterintelligence was willing to acknowledge. The tunnel, from Angleton’s perspective, was a compromised operation that had been allowed to look like a triumph. The distinction mattered. If the KGB had controlled the tunnel’s output, then every assessment based on that output, every intelligence report, every briefing to policymakers, every strategic judgment about Soviet military intentions in Central Europe, was potentially suspect.

The advocates for the tunnel’s value pushed back with numbers. The operation had produced hundreds of thousands of recorded conversations. The sheer volume, they argued, made systematic manipulation implausible. The KGB would have had to orchestrate a vast proportion of Soviet military telephone traffic in East Germany, coordinating the speech of hundreds of officers across multiple installations, to have shaped the tunnel’s output in any meaningful way. The logistics of such an operation would have been staggering. The recordings contained routine administrative calls, logistical chatter, personal conversations, and the mundane noise of military bureaucracy. No deception operation could have generated that texture at that scale. The content, they insisted, was too various, too granular, too boring to have been fabricated.

The counterintelligence officers were not persuaded. They pointed out that volume was not the same as authenticity. A deception operation did not need to control every conversation. It needed only to control the conversations that mattered, the ones containing orders, assessments, strategic intentions, the intelligence that climbed the chain to presidents and prime ministers. The mundane traffic could be left alone. The KGB’s task was not to fabricate the entire output but to ensure that the significant fraction of it, the fraction that reached analysts’ desks and policymakers’ briefings, carried the messages Moscow wanted the West to receive. This required controlling a small number of speakers, not a vast bureaucracy. The argument from volume, the counterintelligence officers said, proved nothing except that the tunnel had been busy.

The debate did not produce a showdown. It produced a stalemate, and the stalemate was managed not by resolution but by deferment.

In Washington, the CIA’s leadership under Allen Dulles and, after November 1961, under John McCone, chose not to commission a full post-mortem on the tunnel’s intelligence product. The reasons were institutional.

A thorough investigation would have required examining every significant conversation recorded during the tunnel’s operational life, assessing each for the possibility of Soviet manipulation, and comparing the tunnel’s intelligence against other sources: interrogations of defectors, agent reports, signals intelligence from other platforms. The effort would have been enormous, the conclusions uncertain, and the political consequences severe. If the investigation concluded that the KGB had manipulated the tunnel’s output, the CIA would have to acknowledge that a substantial body of intelligence had informed Western assessments of Soviet military intentions for years. The credibility damage would extend beyond the tunnel to the agency’s broader intelligence product.

If the investigation concluded that the tunnel’s output was authentic, the counterintelligence officers would not accept the finding. They would argue that the absence of evidence of manipulation was itself evidence of the manipulation’s sophistication. The investigation would settle nothing and cost everything.

Dulles, who had approved the tunnel and championed it, had no incentive to reopen the question. His tenure at the CIA was ending under the shadow of the Bay of Pigs and the escalating crisis in Cuba. On November 30, 1961, President Kennedy authorized Operation Mongoose, also known as the Cuban Project, an extensive campaign of covert operations against Cuba. The CIA’s institutional attention shifted south. The tunnel, already a year in the past, receded further. McCone, arriving as director, inherited an agency in crisis and had no appetite for a protracted internal fight over an operation that had been exposed five years earlier and whose primary intelligence value had been consumed before his arrival.

In London, the dynamics were similar. The Secret Intelligence Service had its own institutional reasons to avoid a full accounting. The tunnel had been a joint operation, and any finding that the KGB had manipulated its output would implicate both services. The British contribution had been intelligence expertise and target selection, the identification of which cables to tap and which conversations to prioritize. If those targets had been compromised from the start, the British role in the operation would be revealed as worse than useless. It would have been the mechanism through which the KGB channeled its deception into Western hands.

The chief of the Secret Intelligence Service, Dick White, had been in post since 1956. He had not been involved in the tunnel’s planning or execution. He had, however, inherited its legacy and the Blake case. White’s approach was characteristic: he favored quiet internal review over public spectacle. The Blake trial had been spectacle enough. A formal inquiry into the tunnel’s intelligence value would reopen wounds that the trial had, from the service’s perspective, closed. Blake was in prison. The operation was exposed. The alliance with the CIA had survived the breach. Those facts mattered to the institution. Whether the KGB had manipulated the tunnel’s output was a question for historians, not for a service that needed to rebuild its credibility and continue operating.

Within the CIA, a small team reviewed the tunnel’s intelligence product against the known facts of Blake’s betrayal. The review confirmed that Blake had informed the KGB of the tunnel’s plan before construction began. It confirmed that the KGB had allowed the tunnel to operate for nearly eleven months before staging its discovery. It confirmed that the staged discovery, announced with great publicity by the East German government in April 1956, had been orchestrated to conceal the KGB’s prior knowledge. These findings were not in dispute. What the review did not determine, and what it was not asked to determine, was whether the KGB had used the tunnel’s operational life to feed manipulated intelligence to the West.

The review’s terms of reference, as drafted by the counterintelligence staff, included this question. The terms of reference, as approved by senior management, did not. The question was removed before the review began. The removal was not explained in the file. It did not need to be. The institutional logic was self-evident: a question that could not be answered with confidence was a question that should not be asked in an official forum, because the absence of a confident answer would be interpreted as evidence of failure.

Within the Secret Intelligence Service, the parallel review was even more circumscribed. The British inquiry focused on Blake’s career, his recruitment by the KGB in Korea, his subsequent postings, and the operations he had compromised. The tunnel was listed among those operations, but the listing was procedural. The inquiry did not examine the tunnel’s intelligence product. It did not engage with the question of Soviet manipulation. It treated the tunnel as one entry in a catalogue of betrayals, assessed for the damage Blake had caused, not for the deception the KGB may have conducted. The distinction was deliberate. Assessing damage was a backward-looking exercise that assigned blame to a convicted spy. Assessing deception was a forward-looking exercise that questioned the competence of the services themselves.

The stalemate hardened into policy. Both services adopted positions that were technically defensible and practically useless. The CIA’s official position, reflected in internal memoranda and the declassified history of the operation, was that the tunnel had produced valuable intelligence and that Blake’s betrayal, while damaging, did not necessarily invalidate the intelligence collected. The word “necessarily” carried the full weight of the evasion. It acknowledged the possibility of manipulation without accepting it. It left the question open while implying that it had been answered in the tunnel’s favor.

The Secret Intelligence Service’s position was less articulated but functionally identical. The service did not produce a formal assessment of the tunnel’s intelligence value after Blake’s trial. It did not publicly or internally contest the CIA’s characterization. The absence of a British position was itself a position, a calculated silence that avoided committing the service to a judgment it could not safely make. If the tunnel’s output was authentic, silence cost nothing. If it was manipulated, silence protected the service from the consequences of having endorsed it.

The institutional silence served multiple functions simultaneously. It protected the reputations of officers who had championed the tunnel and whose careers had been built on its success. It avoided paralyzing recriminations within and between the two services. It preserved the operational relationship between the CIA and the Secret Intelligence Service at a moment when that relationship was under strain from the Blake case and broader revelations of Soviet penetration. It buried the most uncomfortable questions in classified files that few would read and fewer would act upon.

The silence also served a function that neither service acknowledged explicitly. It preserved the tunnel’s place in history as both a technical achievement and an intelligence success, with the caveat of Blake’s betrayal appended as a footnote rather than integrated as a central fact. The tunnel could remain in the institutional memory as a brilliant engineering coup that was compromised by a single mole. It did not have to become what the counterintelligence officers believed it was: a monument to deception, a channel through which the KGB had conducted a strategic influence operation against the West for nearly a year, using the West’s own technology and ambition against it.

The counter-explanation had strong support within both services. It held that the tunnel had yielded genuine intelligence, and that the damage to its worth came only after Blake’s confession revealed the KGB’s prior knowledge. The explanation was simpler. It required no assumption of Soviet capabilities beyond what was already known. It was consistent with the KGB’s demonstrated pattern of protecting high-level penetrations. Blake was a significant asset. The KGB had invested years in his cultivation and deployment. Exposing the tunnel to shut it down would have been a rational choice only if the intelligence the tunnel was collecting was more valuable than Blake’s continued service. Given that Blake was positioned inside the Secret Intelligence Service with access to a wide range of operations, the calculation was not close. Protecting Blake was worth more than neutralizing one intercept operation, however impressive its engineering.

This reading had the further advantage of being consistent with the KGB’s behavior after the tunnel’s discovery. The East German announcement in April 1956 was staged to maximize propaganda value while concealing the KGB’s prior knowledge. The Soviets presented the tunnel as a Western aggression discovered through vigilance, not as an operation they had monitored from its inception. This deception, if it was one, was designed to protect Blake. If the KGB’s motive had been solely to protect Blake, then the tunnel’s intelligence product was authentic, compromised only in the sense that the Soviets knew it was being collected, not in the sense that they had shaped its content.

The counterintelligence officers did not dispute this reading. They acknowledged its plausibility. But they argued that plausibility was not proof, and that the KGB’s motive for protecting Blake did not exclude the possibility of simultaneous manipulation. The two objectives were not mutually exclusive. The KGB could protect Blake and use the tunnel. The question was not whether the KGB had a motive to protect Blake. The question was whether the KGB had restricted itself to that single motive when the tunnel offered an additional opportunity so elegant and so damaging that abandoning it would have been irrational.

The advocates for the tunnel’s value responded that the KGB’s ability to manipulate Soviet military telephone traffic was unproven. The tapped cables carried conversations between Soviet military headquarters in Zossen, near Berlin, and the Soviet embassy in East Berlin, as well as conversations between East German and Soviet officials. Orchestrating the content of these conversations would have required the KGB to influence the speech of senior military officers, some of whom communicated with Moscow directly. The risk of detection, of a military officer noticing that his conversations were being scripted or shaped, was significant. The KGB was powerful, but its power within the Soviet military had limits.

The counterintelligence officers countered that the KGB did not need to script conversations. It needed only to ensure that certain officers were briefed, before they picked up the telephone, that their lines were compromised. An officer who knew his calls were being listened to by the enemy would naturally adjust his speech. He would avoid saying anything that contradicted the impression Moscow wanted to convey. He would emphasize the points Moscow wanted emphasized. He would, in effect, become his own censor, and the censorship would be invisible because it would be indistinguishable from prudence. The KGB’s task was not to fabricate content but to manage the context in which content was produced. The distinction was fine, and it was precisely the distinction that the tunnel’s collection system was incapable of detecting.

The debate reached no conclusion because it could reach no conclusion. The evidence available to both sides was the same evidence: the recordings, the transcripts, the analytical reports, and the fact of Blake’s betrayal. The recordings could not testify to their own authenticity. The transcripts could not reveal whether the speakers had been briefed. The analytical reports could not distinguish between genuine intelligence and intelligence shaped at its source. The fact of Blake’s betrayal established that the KGB had known about the tunnel but could not, by itself, establish what the KGB had done with that knowledge. The question was unanswerable with the evidence available, and both services chose to live with the unanswerable rather than invest the resources and accept the risks that a more aggressive investigation would have required.

By 1963, the internal reviews had concluded, the memoranda had been filed, and the question of the tunnel’s intelligence value had been formally set aside. The CIA’s declassified history of the operation, completed in the late 1960s, described the tunnel as a technical success and acknowledged Blake’s betrayal as a factor in its exposure. It did not address the possibility of Soviet manipulation of the tunnel’s output. The history’s silence on this point was not an oversight. It reflected the institutional position that had crystallized over the preceding two years: the question would not be asked because it could not be answered, and it could not be answered because it would not be asked.

The files went into storage. The recordings, on thousands of reels of magnetic wire and tape, were archived. The transcripts, in English translation, were bound and catalogued. The engineering reports, the requisition forms, the daily maintenance logs, all the paper generated by a quarter-mile of tunnel and eleven months of continuous operation, were boxed and indexed. The counterintelligence officers’ concerns were documented in memoranda that were themselves classified and filed. The institutional record was complete in the sense that it contained everything. It answered nothing.

The men who had built the tunnel and analyzed its product were left with the silence. Harvey had left Berlin before the tunnel’s discovery and had moved on to other assignments, carrying the operation as a line on his record whose value no one would officially assess. The technical staff who had maintained the recording equipment, the analysts who had transcribed the conversations, the translators who had rendered Russian and German into English, all of them had performed their duties with competence and dedication. They had produced work of measurable quality. They had no way of knowing whether the work was worth the effort, and no one in authority would tell them. The official verdict was that there was no verdict. The institutional silence left the tunnel’s builders and analysts without an official finding, forcing them to construct their own.