Code Warriors

by Stephen Budiansky

  For decades, standard histories of the air war in Korea attributed the sudden improvement in mid-1951 in the kill ratio achieved by American fighter pilots against Chinese MiG-15 jets to the arrival of the new and more capable American F-86. During the final year of the war U.S. fighters shot down 345 MiGs in air battles with a loss of only 18 F-86s, a kill ratio of 19 to 1. In fact, the real breakthrough had come from pulling together all of the signals intelligence sources in one center so that they could be rapidly correlated and passed on to fighters in the air. “The present top-heavy success of the F-86 against MiG-15s dates almost from the day of the inception of the new integrated [signals intelligence] service,” reported an officer involved in the operation. On one day, a visiting ASA colonel observed the system in action as 15 MiGs were shot down without a single loss by U.S. F-86s. With more enthusiasm than originality, the colonel said it was “just like shooting ducks in a rain barrel,” but it was an unmistakable demonstration of the incredible force multiplier that the signal interception and reporting system had provided: not a single one of the MiGs was tracked on U.S. radar during the course of the battle; all of the information passed to U.S. pilots had come from listening, in real time, to the communications of the enemy controllers and planes.39

  An analysis of ground control traffic in June 1952 concluded that more than 90 percent of MiGs engaged in air operations over Korea were being flown by Russians. That the Soviets had engaged in a shooting war with Americans remained classified Top Secret for a quarter of a century. It, of course, could hardly have been a secret to the Russians: it was only the American people who could not be trusted with such information, in an era when nuclear weapons risked turning any spark into a conflagration. “The two superpowers had found it necessary but also dangerous to be in combat with one another,” John Lewis Gaddis observed. “They tacitly agreed, therefore, to a cover-up.”40

  Armistice talks had begun in July 1951, but the war dragged on for two more years, settling into a grinding stalemate that resembled the trench warfare of World War I and ultimately taking two million Korean, six hundred thousand Chinese, and thirty-seven thousand American lives. Stalin urged dragging out the negotiations as a way to tie down the United States, keep the Truman administration off balance, and give American military prestige a black eye. Only with Stalin’s death in March 1953 did the Soviet regime agree to a cease-fire. The agreement, which went into effect in July, left the two Koreas with almost exactly the same territory they had held when the war began, but with their lands devastated.

  —

  Those who had been in a position to see firsthand the miracles that Ultra intelligence regularly performed in World War II were dismayed at what happened to this vital source during the ensuing half decade of peace. “It has become apparent that during the between-wars interim we have lost, through neglect, disinterest and possibly jealousy much of the effectiveness in intelligence work that we acquired so painfully in World War II,” observed General James Van Fleet, who had commanded a corps in Patton’s Third Army and succeeded Ridgway as commander of the Eighth Army in Korea. “Today, our intelligence operations have not yet approached the standard that we reached in the final year of the last war.”41

  Despite some remarkable successes under difficult conditions, the experience in Korea laid bare the complete unworkability of AFSA and exposed its near-total failure to bring order and central control to the chaos of the sprawling American signals intelligence bureaucracy. By 1952, AFSA plus the three service cryptologic agencies had 32,500 military and civilian personnel and an annual budget of $400 million, but the rivalry and strife that the reorganization was supposed to put an end to had if anything grown worse.42 At the behest of the new CIA director, General Walter Bedell Smith, Truman in late December 1951 ordered a complete review of the tangled communications intelligence structure. Smith had a well-deserved reputation as a take-charge executive, and the fix was in from the start; by the time the Joint Chiefs of Staff even knew about Truman’s order the review committee had already been appointed and its staff, drawn entirely from CIA and the State Department, was in place.

  Headed by a prominent New York attorney, George A. Brownell, the committee swiftly concluded that the creation of AFSA had been “a step backwards” in solving the chronic problems that beset the military-run signals intelligence establishment. Even in achieving the most basic goal of consolidating cryptanalytic processing into a single, centralized organization and eliminating the pointless duplication and interservice rivalry that stubbornly resisted earlier efforts at “coordination,” AFSA had been less than a roaring success. AFSA’s first director, Rear Admiral Earl E. Stone, had been able to bring about what at least had the outward appearance of a merger in the Washington-area Army and Navy headquarters, moving all communications intelligence activities to Arlington Hall and communications security to Nebraska Avenue. But AFSA’s seventy-six-hundred-person staff and $35 million budget remained a small share of the total enterprise, and both the Army and Air Force cryptologic agencies continued to grab important projects for themselves. ASAPAC and USAFSS both duplicated AFSA’s work on Soviet and Chinese codes throughout the Korean War, and simply ignored attempts by AFSA to take charge of field processing within the theater. The Air Force had meanwhile established its headquarters of USAFSS at Brooks Air Force Base in Texas, a not too subtle attempt to escape from the Washington orbit altogether.43

  Control of intercept was more tangled than ever. Joseph Wenger, now a real admiral and deputy director of AFSA, frankly told the Brownell Committee that AFSA made a mockery of the fundamental management principle that “authority must be commensurate with responsibility.” Under the pre-AFSA arrangement, the coordinator of joint operations was at least allowed to tell the services’ intercept stations what signals to monitor in areas of “joint” responsibility, which included diplomatic targets. The director of AFSA was reduced to begging. Admiral Stone conducted an arduous negotiation with the Air Force, ending with an agreement that the service would place under AFSA control all of its “fixed” intercept stations. The Air Force then declared all of its stations a “radio group mobile.” (Most, an Air Force general later admitted, were “as mobile as the Eiffel Tower.”)

  When AFSA organized a small “Field Activity, Far East” to try to bring some order to the three services’ intercept stations supporting AFSA’s work on North Korean, Chinese, and Soviet targets during the Korean War, the Navy grudgingly offered some cramped space at its station in Yokosuka, but all three services resisted following its orders, taking the line (as the Brownell report summarized their attitude) that to do so was “inconsistent with normal command relationships and with the responsibility of each Service to provide combat intelligence for its own operations.” AFSA was powerless to prevent even the most obvious duplication of effort: for over a year the Army and the Air Force both insisted on intercepting Russian and Chinese air communications, and it was not until March 1952, after months of negotiations, that ASA finally agreed to leave the job to the Air Force. The Navy meanwhile flatly refused to put its worldwide network of direction-finding stations—which provided the single most important source of information on the location and movement of Soviet surface ships and submarines—under central control.44

  The worst problem was that although it had the outward appearance of following the model of “unification” that had been behind the establishment of the Department of Defense, the creation of AFSA was actually a bureaucratic Frankenstein that left the individual military services once again calling all the shots. Deliberately or not, the result was a near-perfect exemplar of that wonder of organizational theory, the circular chain of command. The service elements answered to the director of AFSA, who answered to a committee called the Armed Forces Security Agency Council, which answered to the services. Just as in USCIB, the decisions of AFSAC had to be unanimous; as the Brownell Committee noted, the council spent most of its time “safeguarding individual Service autonomies.”
/>
  The internal structure of AFSA similarly ensured that the services had an effective veto over every decision; the director was required to have three vice directors, one from each service, and even at lower management levels every service was represented as a coequal. Hoping to improve relations with the “customers,” Stone allowed analysts from the separate military services (along with CIA, State, and FBI) each to maintain “beachhead” offices at Arlington Hall where they could directly access the product. But the effect was only to further undermine AFSA’s authority and embolden the services’ claim to their “sovereign powers” over COMINT. “Since they felt we couldn’t process the stuff fast enough,” recalled Oliver Kirby, these “roving intelligence representatives” would each grab the same raw data and write separate, usually mutually contradictory reports. “We had to spend inordinate amounts of time trying to figure out how to get something done within the system,” Kirby said.45

  At the same time, the establishment of AFSA, and its new council made up solely of representatives from the three military services, created a complete parallel command structure alongside the existing USCIB arrangement that was meant to bring signals intelligence under the control of the National Security Council, with representation of all interested parties, civilian and military alike. Who AFSA’s director actually reported to, and how it was to serve the needs of State and CIA, was a hopeless muddle.46

  The Brownell Committee delivered its final report in six months, calling for a complete reorganization that would give the director of the agency real power and break the military services’ hold, terminating what it called the failed “experiment” that had put AFSA under the Joint Chiefs’ auspices. On October 24, 1952, Truman issued an order accepting nearly all of the committee’s recommendations, declaring communications intelligence a “national” function, and establishing the National Security Agency to replace AFSA. The Department of Defense was named the “executive agent” to carry out these duties on behalf of the government, but the new agency was no longer directly answerable to the military authorities. The director of central intelligence became the permanent chairman of USCIB, and from now on a majority vote would suffice to make decisions—and that majority was now firmly in civilian hands, with State, CIA, FBI, and the secretary of defense each having two votes to the military services’ one vote apiece. The NSA director was given “operational and technical control” over all communications intelligence “collection and production resources” of the Department of Defense, and was also made a voting member of USCIB, further strengthening his clout.47

  —

  With Soviet high-level communications still unreadable, Meredith Gardner and the Russian group at Arlington Hall continued to work on the 1943 to 1945 one-time-pad messages that were vulnerable owing to the Soviets’ duplicate key use. In October 1949, Kim Philby arrived in Washington to become the British Secret Intelligence Service’s station chief, and within weeks was on the best of terms with Gardner and the FBI agent now working with Arlington Hall to try to identify the Soviet spies named in the messages, Robert Lamphere.48

  Philby, like his fellow Cambridge recruits of the 1930s to the Soviet cause, was a scion of the British upper classes who managed effortlessly to combine a commitment to Marxism with a sense of privileged entitlement that the normal rules of society simply did not apply to people like him. His father had been a colonial administrator in India who exemplified the breed: Hillary St. John Bridger Philby was a world traveler, amateur ornithologist, and well-connected member of the English old boys’ network who in his later years was a renowned Arabic scholar and convert to Islam. He took as his second wife a Baluchistani slave girl presented to him by the first monarch of Saudi Arabia, King Ibn Saud, in gratitude for his service to the kingdom.49

  In obedience to Moscow’s instructions, Kim Philby had feigned sympathy for the fascists in the Spanish Civil War as a newspaper correspondent in the 1930s and had moved from there to ever more responsible positions in the British secret service. Whatever Philby’s original ideological convictions, he came to find the dangerous life of a double agent an intoxicating and addictive drug, matched by a relentless appetite for drunken binges, sexual liaisons, and exercising his duplicitous charm on all around him. “You didn’t just like him, admire him, agree with him,” said Sir Robert Mackenzie, the security officer at the British embassy in Washington, “you worshipped him.” In London during the war he had charmed the hopelessly anglophilic James Jesus Angleton, in 1949 already a key figure in the newly established CIA, and on Philby’s arrival in Washington he and Angleton became bosom friends, meeting for weekly, and then almost daily, very alcoholic lunches at Harvey’s Restaurant in downtown Washington, beginning with bourbon on the rocks, then lobster and wine, then brandy and cigars. At his home at 4100 Nebraska Avenue—just next to the Navy’s signals intelligence center, as it happened—Philby and his wife hosted famously drunken parties for his new CIA and FBI colleagues.50

  One of the first things the FBI brought to the attention of the new British liaison upon his arrival in the fall of 1949 were decrypted NKGB messages from 1944 and 1945 referring to the activities of the Soviet agent HOMER at the British embassy in Washington. There were hundreds of possible candidates who fit the scant information available about HOMER’s true identity, but Philby recognized his fellow spy at once. Donald Maclean had in fact been recruited by Philby at Cambridge, had shrewdly allayed doubts about his political past when interviewing for a job in the Foreign Office by acknowledging his involvement with left-wing organizations as a student and disarmingly admitting that he was trying to shake off his Communist views but had “not yet entirely succeeded,” and had risen swiftly in the ranks to become first secretary in the Washington embassy, where he was posted from 1944 to 1948.51

  That fall Lamphere also read a newly decrypted 1944 NKGB cable from New York to Moscow that contained a summary of a theoretical paper produced by the Manhattan Project on the gaseous diffusion process of enriching uranium. The author of the paper was Klaus Fuchs, a refugee from Nazi Germany who had worked on the British and then the American atomic bomb projects during the war. Fuchs was known to have joined the Communist Party in Germany as a young man, fleeing the Nazis’ mass arrest of political opponents following the 1933 Reichstag fire; what the FBI did not yet know was that since 1941 Fuchs had been volunteering his services to the Soviets as a spy. A week after reading the message containing Fuchs’s theoretical paper, Lamphere was sure he had identified Fuchs himself as the agent CHARLZ who had supplied the document: another cable mentioned that CHARLZ was a British subject whose sister was attending an American university, which fit Fuchs.52

  Now back in Britain, Fuchs was interrogated by an MI5 officer who patiently waited out his halfhearted evasions and after a few weeks extracted a full confession from the scientist on January 24, 1950. Among other things, Fuchs admitted that in the summer of 1945 he had turned over to his contact Harry Gold a comprehensive technical report containing everything he had learned about the design and construction of the bomb. Like the other atomic spies, Fuchs insisted he was motivated by idealistic aims, to aid a heroic ally that was bearing the brunt of the fight against Hitler: he had been furiously insulted when Gold tried to hand him an envelope stuffed with $1,500 in cash supplied by Moscow in payment for his services.

  Later a Los Alamos physicist colleague and close friend, Rudolf Peierls, with whom Fuchs had lodged while working in England, asked him how, as a scientist, he could have swallowed the doctrinaire orthodoxies of Marxism. Peierls was stunned by the “arrogance and naïveté” of Fuchs’s answer. “You must remember what I went through under Nazis,” Fuchs replied. “Besides, it was my intention, when I had helped the Russians to take over everything, to get up and tell them what is wrong with their system.” Peierls’s wife, Genia, who had been something of a mother figure to their young lodger, wrote him a more personal rebuke. Hadn’t he at least thought about the betrayal of his friends he had committe
d, and the harm he had done them? she asked.

  “I didn’t, and that’s the greatest horror I had to face when I looked at myself,” Fuchs wrote back from prison. “I thought I knew what I was doing, and there was this simple thing, obvious to the simplest decent creature, and I didn’t think of it.” He told another friend, “Some people grow up at fifteen, some at thirty-eight. It is more painful at thirty-eight.”53

  Matching together Arlington Hall’s decrypts and FBI files of suspected Communist agents was yielding a slew of more names. Harry Gold was arrested on May 22, 1950. Julius Rosenberg, identified as LIBERAL and ANTENNA, was arrested on July 17, and his wife, Ethel, the following month. Julius Rosenberg and Harry Gold were unquestionably part of the Soviet spy ring that had made contact with Klaus Fuchs and other Los Alamos scientists and passed their reports on the atomic bomb to Moscow, but Ethel’s involvement was limited at most to putting the group in touch with her brother, David Greenglass, an Army technician at Los Alamos.54

  In the panicky atmosphere following the Soviets’ first atomic test and the shock of the Korean War, both Rosenbergs were condemned to death by a federal judge for espionage, a grotesquely barbaric sentence that the Rosenbergs accepted in defiant martyrdom, spurning offers of leniency in return for their cooperation. They went to their deaths in the electric chair at Sing Sing refusing to confess or provide any information on their espionage activities.

  By Kim Philby’s second year in Washington he knew the net was closing around him. It was only a matter of time before Maclean was identified by a newly decoded message, and Philby tried to keep an eye on the progress of the work by paying a personal visit to Meredith Gardner at Arlington Hall; it may have been with the imagination of hindsight, but Gardner would later recall the silent, rapt intensity with which Philby stood observing the work of the Russian section. If Maclean were caught, the trail would almost certainly point to Philby, as their earlier associations would not be hard to turn up. In June 1950, Arlington Hall read a message referring to a valuable agent named STANLEY operating in Britain in 1945: that was Philby himself.55