Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM Venona: Soviet Espionage and the American Response, 1939-1957 Preface The muggy Washington summer of 1948 grew even hotter when news media reported that a "blonde spy queen" three years earlier had given federal investigators convincing evidence of widespread Soviet espionage in America during World War II. In a few days the world learned her name--Elizabeth Bentley-- and heard her and another ex-Communist agent, Whittaker Chambers, repeat their charges before Congress. Republican congressmen and candidates cited the stories as further evidence of the Roosevelt and Truman administrations' softness toward Communism and neglect of national security. Outraged officials both in and out of government, as well as Democrats fearing a campaign issue that would sink President Truman's apparently foundering re-election chances, insisted that Bentley and Chambers were peddling hearsay and innuendo. Almost lost in the furor was one isolated recollection of Bentley's that ultimately would provide a clue to the truth behind the charges and denials. Bentley, according to press reports, had told a federal grand jury that an aide to President Roosevelt had learned during the war that American intelligence was on the verge of breaking "the Russian secret code." The aide, said Bentley, had passed this nugget to his Soviet contact. (1) For almost 30 years this fragmentary anecdote remained virtually all that the public would hear about one of the Cold War's greatest intelligence coups. Bentley's charges, and the debates they fueled, typified the American experience with intelligence and related "internal security" issues in the era of totalitarianism and total war. For roughly 60 years the Western democracies struggled to preserve civil liberties and due process while ascertaining the extent of clandestine penetrations by the intelligence services of fascist and Communist regimes. At midcentury the Soviet Union's main strength was "human" intelligence--the collection of information through agents with access to foreign secrets. Washington's forte was "signals" intelligence--the procurement and analysis of coded foreign messages. At the beginning of the Cold War strength met strength in a struggle that still reverberates 50 years later. The tale of this struggle is the Venona story. The term "Venona" served as an arbitrary codeword stamped on a relatively small number of documents in order to limit access to a particular cryptanalytic breakthrough. This achievement enabled Western counterintelligence specialists to read portions of more than 2,900 Soviet diplomatic telegrams sent between 1940 and 1948. The encipherment of these telegrams shared a common flaw that left them vulnerable to cryptanalysis. It was that flawærather than any commonality of dates, origins, or subject matteræthat made the messages a unique and discrete body of documents. American and allied services spent almost four decades deciphering the original texts and then puzzling over their meanings. By the time this effort was formally closed in 1980, the codeword "Venona" meant, to a handful of witting US Intelligence Community officers, the entire program of crypt-analytic and exploitation activities based on the messages. Espionage in America http://www.cia.gov/csi/books/venona/preface.htm Page 1 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM The United States made a tempting espionage target for allies and adversaries alike in the 1940s. Berlin, Tokyo, and Moscow all wanted to discover Washington's strategic plans and the progress being made in American factories and laboratories. Axis spies fared poorly in North America, however, in part because allied civilian and military counterintelligence services rolled up Axis nets and agents early in the war. Soviet intelligence fared much better. Indeed, the tensions and crises in East-West relations in the 1940s and 1950s unfolded along patterns determined in no small part by the success of Soviet intelligence officers, and, belatedly, by the growing ability of Western services to counter Moscow's espionage campaign. Several advantages helped Soviet intelligence succeed where the Axis services failed. First, Soviet intelligence services, in many respects, were stronger than their Axis counterparts, with better leadership and more resources. Second, operating conditions in America were easier for several reasons, the foremost being the fact that the Soviet Union was an ally and therefore was able to post large numbers of officials on American soil in various liaison capacities. Beyond this, many Americans regarded their Russian allies as comrades-in-arms who should be helped with material as well as rhetorical support. There were some instances of American citizens volunteering actual secrets to Soviets during the war, and Soviet officials in the United States sometimes enjoyed considerable hospitality and access. Finally, Soviet intelligence benefited directly and indirectly from the activities and infrastructure of the Communist Party of the United States (CPUSA). Moscow collected secrets in the United States through overlapping organizations. The Communist International (better known as the Comintern) monitored the CPUSA and supervised the Party's clandestine apparatus. The CPUSA had reached the peak of its strength and limited influence in American life in the late 1930s, when the Great Depression and Stalin's opposition to Hitler and Mussolini convinced thousands of native-born Americans that capitalism was doomed and that the socialist experiment in Russia represented the world's only reliable bulwark against fascism. (2) Party leaders and some trusted members gathered political and industrial information, most of which probably made its way to Soviet intelligence services. (3) Although the CPUSA lost perhaps a third of its members after the Hitler-Stalin pact of August 1939 temporarily made the USSR a junior partner in Nazi aggression, a committed core remained in the ranks. (4) In addition, military intelligence (GRU) officers based in Soviet consular posts worked with the Comintern and with Party-controlled agents in US Government agencies and private industry. (5) Another Soviet organization, the NKVD (later to become the KGB)-- Moscow's secret police and quasi- civilian intelligence service--had its own foreign intelligence arm. The service had long busied itself with internal repression and with foreign counterintelligence that helped guide covert action against émigré Russian political factions. In other types of foreign operations it had been overshadowed by the GRU, at least in the United States. (6) This relationship would be reversed during the war. (7) The KGB and GRU ran parallel "legal" and "illegal" intelligence networks in the United States. One set of operations was run by intelligence officers working under legal (usually diplomatic) cover in the "residencies" located clandestinely in Soviet diplomatic missions, trade, and media organizations--for example, the Amtorg Trading Corporation, the Soviet Purchasing Commission, and the Tass news agency. (8) Other KGB and GRU networks, in contrast, had no apparent connections to Soviet establishments and were run by "illegals"--Soviet intelligence officers usually living under false identities. In addition, some GRU and KGB agents were themselves CPUSA officers whose clandestine activities were known, to a greater or lesser extent, to the CPUSA leadership and the Comintern. KGB officer Jacob Golos (covername SOUND), for instance, ran the Soviet-sponsored World Tourists corporation in New York and allegedly also served on the central control commission of the CPUSA. His lover, Elizabeth Bentley (covernames SMART http://www.cia.gov/csi/books/venona/preface.htm Page 2 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM GIRL and MYRNA), had moved from the open Party into underground work in the late 1930s. She helped Golos service various agents and run the World Tourists business. Stalin wanted his intelligence officers in America to collect information in four main areas. He directed Pavel M. Fitin, the 34-year-old chief of the KGB's First Directorate, to seek American intelligence concerning Hitler's plans for the war in Russia; secret war aims of London and Washington, particularly with regard to planning for a second front in Europe; any indications that the Western allies might cut a separate peace with Hitler; and, finally, American scientific and technological progress, particularly in developing an atomic weapon. (9) Soviet espionage operatives in the United States during World War II funneled information to Moscow through a handful of professional intelligence officers who sent reports to the Center and relayed orders and questions from the Center to agents in the field. Operations in America were led by experienced hands such as Vassili M. Zarubin (covername MAKSIM), who served as rezident in New York and later in Washington, and Iskhak A. Akhmerov (covernames MER and ALBERT), the senior illegal. Some Soviet case officers, however, were raw recruits recently brought into the services in order to fill out ranks depleted during Stalin's purges of the late 1930s.(10) For many of these officers, America was their first overseas post. Elizabeth Bentley described her initial meeting with "John" (Anatoli A. Yatskov, Venona covername ALEKSEI), who turned out to be a thin, pale, blond, young man of about my height, who was dressed in badly fitting clothes of obviously European make. . . . He had that half-starved look so characteristic of new Soviet arrivals, his English was so meager I had difficulty in understanding him, and he displayed an astounding ignorance of American life. "John," despite his unpromising debut, would play a key role in Soviet espionage against the atomic bomb.(11) During the latter part of the war, the KGB gradually took over assets and networks originally established by the GRU and the Comintern (particularly after Stalin dissolved the latter body in May 1943).(12) A general re-division of labor among the Soviet services appears to have given political tasks to the KGB, while focusing the GRU more on military collection; both collected scientific and technical data. In addition, agents hitherto run in cooperation with the CPUSA were turned over to direct Soviet control. This streamlining effort faced daunting operational security challenges; Bentley and others who had worked with the Soviets had learned far more than they needed to about other agents and operations--and resented the change in direction. A Slow Response The US Government had grown concerned about reputed fascist and Communist subversion in the late 1930s. The war in Europe--and the Nazi-Soviet nonaggression pact of 1939--gave J. Edgar Hoover's Federal Bureau of Investigation an opportunity to move against individuals and organizations suspected of working with the Russians. Tipped by a State Department probe that had uncovered American and foreign Communists traveling on fraudulent American passports, FBI Special Agents in 1939 raided the facilities of several organizations linked to the CPUSA and found sufficient evidence to arrest General Secretary Earl Browder on charges of passport fraud.(13) In 1940, leads developed by British and Canadian investigators in the Woolwich Arsenal spy case pointed the Bureau toward the senior KGB officer in America, New York http://www.cia.gov/csi/books/venona/preface.htm Page 3 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM rezident Gaik Ovakimian (covername GENNADI), whom the FBI arrested in May 1941 for violating the Foreign Agents Registration Act.(14) Information on Soviet intelligence contacts and methods obtained by the Bureau in these early investigations would prove valuable during and after the war. American authorities, nonetheless, did not act as decisively as they might have at the time. At least three defectors from Soviet intelligence were in the United States (Alexander Orlov, Walter Krivitsky, and Whittaker Chambers) and have given the Bureau relatively current information, had they been questioned in depth--and well protected--by federal officials.(15) In addition, the German invasion of the USSR in June 1941 virtually reversed American attitudes toward the Soviets. The State Department quickly reached an understanding with Moscow that allowed Ovakimian to leave the country. Similarly, President Roosevelt commuted Browder's sentence in May 1942 in the interest of inter-allied relations. The FBI remained watchful, but the raids and prosecutions ceased. Any foreign intelligence service needs secure communications channels between its headquarters and its officers abroad. Although Soviet intelligence services had clandestine radio transmitters in diplomatic missions located in several American cities, these apparently were to be used only in emergencies.(16) In consequence, KGB and GRU stations cabled their important messages over commercial telegraph lines and sent bulky reports and documents--including most of the information acquired by agents--in diplomatic pouches. As a new European war loomed in 1939, the US Army had begun collecting enciphered Soviet telegrams, and soon thousands of cables were piling up in the offices of the Army's Signals Security Agency (SSA). A June 1942 agreement with the Navy and FBI gave the Army exclusive responsibility for analysis of foreign diplomatic and military ciphers, and the Army consequently had general responsibility for studying diplomatic traffic. SSA analysts, based at Arlington Hall in Northern Virginia, soon began to explore whether the collected Soviet diplomatic messages might be readable. The Army and Navy had sporadically studied Soviet codes and ciphers over the preceding decade, but with little success.(17) Decrypted 1942 cables between the Japanese Army's general staff and its military attaches in Berlin and Helsinki showed that Finland's excellent cryptanalysts had made progress on some Soviet military ciphers, had determined the characteristics for sorting the still unsolvable diplomatic messages, and were sharing results with the Japanese.(18) This information was probably the immediate inspiration for SSA's examination of the Soviet telegrams. On 1 February 1943, SSA created a small program to examine the encrypted Soviet telegrams on orders from Col. Carter Clarke, Chief of the Special Branch of the Army's Military Intelligence Service (MID). Clarke seemed particularly concerned that Moscow and Berlin might negotiate a separate peace, and wanted to be able to warn his superiors of such a development. Gene Grabeel, a Virginia schoolteacher who had recently arrived at Arlington Hall as part of its large civilian contingent, began the effort to read the Soviet diplomatic messages (and would continue working on the project for the next 36 years). She and others assigned to the project in 1943 spent months sorting stored and incoming telegrams by communications circuits. They gradually expanded their knowledge of the characteristics that separated the messages into the groups that the Finnish cryptanalysts had investigated. Five separate cryptographic systems became apparent. More than half the telegrams belonged to a system that analysts dubbed "Trade" because it carried the messages of the Amtorg Trading Corporation and the Soviet Purchasing Commission--most of which concerned the transfer of Lend Lease materiel to the USSR. The other four systems were used by the KGB, GRU, Naval GRU, and Foreign Ministry, but these users would not be fully identified until the mid-1940s. Moscow had already learned from well-placed agents that both enemies and allies were trying to read its http://www.cia.gov/csi/books/venona/preface.htm Page 4 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM diplomatic cables. Finnish troops found scorched codebooks and cryptographic materials in the USSR's Petsamo consulate in June 1941, and before the end of that year a Soviet agent in Berlin reported that the Germans were trying to exploit a Russian codebook acquired from their Finnish allies.(19) These developments in themselves were not alarming to Moscow, because the security of messages enciphered by one-time pads lies in the cipher and not in the codes per se (see inset). In any event, the compromised KGB codebook was not replaced until late 1943.(20) Another important piece of information came from British intelligence officer and Soviet agent H.A.R. "Kim" Philby in 1944, when he told the KGB that British cryptanalysts had turned their attention to Soviet ciphers.(21) Philby probably reported nothing at that time about American efforts against the Soviet messages. (US analysts did not begin to collaborate with their British counterparts on Soviet communications in general until about August 1945.) Nevertheless, senior KGB officials may have become worried when White House aide Lauchlin Currie apparently told Soviet contacts (possibly in spring 1944) that the Americans were about to break a Soviet code. Currie had access to signals intelligence at the White House and could have heard overoptimistic rumors that Arlington Hall would soon be reading Soviet messages. Currie's tip probably was too vague to have alarmed Soviet cryptographers, but it might have worried higher-ups in Moscow. Indeed, the only change observed in the characteristics of the Soviet messages around that time appeared to be a cosmetic correction implemented to please higher authority. On 1 May 1944, KGB code clerks began using a new message starting-point indicator for telegrams--a change that ironically would make work easier for Arlington Hall crypt-analysts.(22) What Made Venona Possible? The messages broken by the Venona program were both coded and enciphered. When a code is enciphered with a one-time pad, the cryptographer who designed the system expects the encipherment to provide absolute security--even if an adversary somehow obtains an underlying codebook or debriefs a defecting code clerk (such as Igor Gouzenko). A flaw in the encipherment, however, can leave such messages vulnerable to analysis even in the absence of a codebook. Such was the case for the Soviet diplomatic systems from which the Venona translations came. Arlington Hall's Venona breakthrough in 1943-46 was a purely analytic accomplishment, achieved without the benefit of either Soviet codebooks or plain-text copies of original messages. The 1944-46 messages--which yielded the early translations and the bulk of all translations--were recovered over a period of years by Arlington Hall cryptanalysts and decoded from a "codebook" that crypto- linguist Meredith Gardner reconstructed by using classic codebreaking techniques. A Soviet code clerk preparing a message first reduced its text into numeric code groups drawn from a codebook (a kind of dictionary in which the words and common phrases correspond to four-digit numbers). After encoding the plain text with numeric code groups, the clerk would obscure the code groups by adding them, digit by digit, to a string of random digits. This second series of digits, called "additive" or "key," was known to both the sender and receiver because it was printed on the pages of a "one-time pad." One-time pads were periodically pouched to Soviet consular missions in sealed packets. The pad pages--with 60 five- digit additive groups per page--were used in order, always starting with the group in the upper lefthand corner (the pad-page number to be used was more or less concealed somewhere on the face of the message). Code clerks in different Soviet missions used up these packets at varying rates, depending on the volume of messages to be enciphered or deciphered. http://www.cia.gov/csi/books/venona/preface.htm Page 5 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM The security of such an encipherment-decipherment system depends on both the randomness (that is, unpredictability) of the "key" on the one-time pad pages and the uniqueness of the one-time pad sets held by the sender and the receiver. Different Soviet organizations used their own codes, changing them every few years (probably more to improve vocabulary and convenience than to enhance security). The flaw in the Soviet messages resulted from the manufacturers' duplication of one-time pad pages, rather than from a malfunctioning random-number generator or extensive re-use of pages by code clerks. For a few months in early 1942, a time of great strain on the Soviet regime, the KGB's cryptographic center in the Soviet Union for some unknown reason printed duplicate copies of the "key" on more than 35,000 pages of additive and then assembled and bound these in one-time pads. Arlington Hall's Lt. Richard Hallock analyzed Soviet "Trade" messages in autumn 1943, producing evidence of extensive use of duplicate key pages (often with different page numbers) assembled in separate one-time pad books. Thus, two sets of the ostensibly unique one-time pad-page sets were manufactured. Despite the opinion that a single duplication was insufficient for solution, Hallock and his colleagues continued to attack the Trade messages and made considerable progress in understanding the cryptographic basis of the diplomatic systems. From Hallock's original discovery, additional analysis yielded techniques for finding duplicate pages separated in time and among different users. The duplicate pages began showing up in messages in mid-1942 and were still occurring in one circuit as late as June 1948. Nevertheless, most of the duplicate pages were used between 1942 and 1944--years of rapid expansion of Soviet diplomatic communications. We do not know how and when the Soviets discovered the flaw, but we believe Moscow learned of it through agents William W. Weisband and Kim Philby. By the time the Soviets saw the consequences of the manufacturing flaw in the late 1940s, however, most of the duplicate one-time pad pages had already been used. The set of potentially exploitable messages thus was bound by the production of the duplicate pages and the West's ability to spot duplicate uses. Finding duplicates, however, only made the messages potentially readable; indeed, some messages and passages remained unexploitable even after 37 years of effort. Cecil James Phillips National Security Agency In November 1944 Arlington Hall analysts solved this new starting-point indicator problem and soon expanded their effort against the "Trade" messages to encompass a second, still unidentified Soviet system. This was the breakthrough that eventually made the cables readable. The method revealed hundreds of instances in which individual pages of additive digits from KGB one-time pads were duplicated by key used for Trade messages. Over a period of years, cryptanalysts were able to determine the one-time pad additive values for significant parts of hundreds of enciphered telegrams, leaving the coded texts vulnerable to crypto-linguist codebuilders trying to recover the meanings of the four-digit words and phrases.(23) American authorities inferred during World War II that the Soviets were engaged in espionage, but as yet there was little coordination among the various counterintelligence organizations. A June 1939 Presidential directive gave the FBI responsibility for domestic counterintelligence with the Office of Naval Intelligence (ONI) and the War Department's Military Intelligence Division (better known as the G-2).(24) The three organizations comprised a body known as the Interdepartmental Intelligence Conference (IIC) and stood together to protect their monopoly on domestic counterintelligence work from other agencies, particularly the new Office of Strategic Services (OSS) and its activist chief, Maj. Gen. William J. Donovan. OSS http://www.cia.gov/csi/books/venona/preface.htm Page 6 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM eventually developed a capable counterintelligence apparatus of its own in Europe--the X-2 Branch--but it had no authority to operate on American soil. In keeping with the limited extent of interagency cooperation, American counterintelligence organizations made uneven progress in integrating signals intelligence leads during the war. OSS and the FBI separately launched their own cryptologic intelligence operations, but these were short-lived.(25) (Neither OSS nor the FBI, incidentally, was shown the Army's "MAGIC" intercepts of wartime foreign diplomatic messages.) The X-2 Branch of OSS had been created to provide British intelligence services with a point of contact in OSS for sharing certain sensitive "ULTRA" reports derived from decrypted German military and intelligence communications.(26) Unfortunately for OSS, however, X-2 monitored the agents of Axis--not allie-- services. It had little influence over security clearances for OSS personnel (some of whom indeed spied for the Soviets).(27) Coordination was little better between the military's cryptologic services, which in any event had left domestic security largely to the FBI. The Army and Navy signals intelligence organizations barely cooperated with one another, jealously guarding their reports and their access to President Roosevelt. Outside of the Oval Office there was no collation and analysis of the totality of the intelligence information- -let alone the counter- intelligence leads--collected by the US Government. Hoover's FBI monitored the CPUSA during the war but did not always share its leads with other agencies. In April 1943, FBI agents began to collect solid information on current KGB activities and personnel. New York consular officer Vassili M. Zarubin (a KGB general operating under the alias Zubilin) called on senior CPUSA officer Steve Nelson in Berkeley, California. Unbeknownst to both, the FBI had been watching Nelson for weeks. Zarubin's conversation made it obvious to Bureau eavesdroppers that he was an important KGB officer (although they could not yet know that he was Gaik Ovakimian's replacement as senior KGB rezident in America).(28) The FBI watched Zarubin from that day until he left the United States in 1944 (although he occasionally gave his trailers the slip), and Bureau agents catalogued hundreds of contacts and leads developed by this operation.(29) More leads dropped into the Bureau's mailbox in August 1943, in the form of an anonymous letter drafted on a Russian typewriter and mailed in Washington, DC. This extraordinary note--the author's identity still is uncertain--denounced Zarubin and 10 other KGB officers in North America, along with two of their assets.(30) Special Agents quickly concluded that the letter was genuine and largely accurate, although they gave little credence to its claim that the Soviets were passing secrets to Japan. The FBI subsequently increased surveillance of persons named in the letter and even doubled two agents recruited by one of them, KGB officer Andrei Shevchenko.(31) Nevertheless, the FBI did apparently not pass copies of the anonymous letter to other agencies until after World War II, nor did Special Agents try to recruit Soviet officers named by its author. The Atomic Era US Government agencies ran a wartime security system that was porous for Soviet agents and yet opaque for American counterintelligence agencies charged with protecting secrets. FBI Director Hoover allegedly knew nothing of the super-secret Manhattan Project before Steve Nelson inadvertently informed him in the spring of 1943. High-level political and strategic motivations in Washington also hampered US efforts against Soviet espionage. President Roosevelt wanted to strengthen a distrustful Stalin in his fight against Hitler, and his lieutenants had no desire to antagonize Moscow by suppressing the CPUSA or publicly http://www.cia.gov/csi/books/venona/preface.htm Page 7 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM probing rumors that members of the Party had infiltrated government agencies.(32) Hoover, for his part, kept a close eye on the CPUSA but did not, at least before 1945, try to convince the White House that Soviet officials in the United States were actively engaged in espionage. Donovan's OSS also encountered the administration's reluctance to antagonize Moscow when OSS officers bought unidentified Soviet cryptographic documents from emigre Finnish cryptanalysts in late 1944. Secretary of State Edward P. Stettinius insisted that the papers be given back to the Russians, and Donovan promptly obeyed a White House order to return them to the Soviet Embassy.(33) The intrigues surrounding the development of the atomic bomb both symbolized and helped widen the growing breach between the Soviet Union and its Western allies in 1945. Washington and London jointly built the bomb but said nothing about their work to Moscow. Stalin's clandestine sources, however, obtained detailed political, military, and diplomatic reports on his allies' strategic planning and war aims.(34) He knew of the bomb project long before the new President Truman finally divulged it to him in July 1945. The KGB effort against the Manhattan Project (codenamed ENORMOUS) represented a shift in collection emphasis. Moscow hitherto had regarded the United States primarily as a source of information useful in the war against Germany; now America became in Russian eyes a rival and even a threat to the Soviet Union itself. Soviet agents penetrated the Manhattan Project at several points. At the Los Alamos facility alone, at least four agents reported through couriers such as Lona Cohen to the Soviet consulate in New York, where a KGB sub-residency under a young engineer named Leonid R. Kvasnikov (covername ANTON) coordinated operations and dispatched intelligence to Moscow.(35) US perceptions of the Soviets began shifting after the war had been won. Two defections in autumn 1945 galvanized US counterintelligence. Igor Gouzenko, a GRU code clerk in the USSR's Ottawa Embassy, revealed to Canadian authorities that the Soviets had indeed penetrated the Manhattan Project and other agencies.(36) A few weeks later, Elizabeth Bentley gave the FBI details about spies in the State and Treasury Department, OSS, the Pentagon, and even the White House. Both Bentley's and Gouzenko's accounts dovetailed with the story that Time magazine editor and former GRU agent Whittaker Chambers had told FBI agents in 1942 and again, in detail, in May 1945.(37) By mid-November, the White House knew the outlines of the defectors' stories and had heard of their accusations against dozens of US Government employees, including high officials such as White House aide Lauchlin Currie, OSS executive assistant Duncan Lee, and Assistant Secretary of the Treasury Harry Dexter White.(38) A Canadian Government White Paper on the Gouzenko affair in July 1946 confirmed the gist of press speculation about Soviet wartime espionage and gave the Western public its first official account of the extent of the problem. This confirmation of the essential truth behind the rumors diminished public tolerance for Communism at home and abroad. Truman became convinced of the need for a government- wide tightening of security, but he had no intention of condoning witch-hunts for allegedly disloyal Democratic officials or blanket accusations against federal workers and Roosevelt's New Deal. Little could be done, for the time being, against the individuals named by Gouzenko or Bentley--apart from corroborating their reports and limiting the suspects' access to sensitive information--until Western governments could gather evidence that would stand up in court. Domestic politics, however, prompted the White House to act. Republicans campaigning in the 1946 Congressional elections accused Democrats of ignoring Communist infiltration and disloyalty; the charge helped the GOP regain control of Congress for the first time since 1931. Truman's response was motivated in part by his own political considerations. Hoping to deter free-ranging Congressional probes and harsh Republican-drafted loyalty legislation, he signed Executive Order 9835, which institutionalized the wartime loyalty regime. The executive order mandated loyalty boards in all federal agencies and defined employee http://www.cia.gov/csi/books/venona/preface.htm Page 8 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM disloyalty to include membership in groups judged subversive by the Attorney General.(39) During this period, Bentley gave the FBI details that opened a hitherto unnoticed window on the networks run by "illegals"--Soviet citizens abroad under false identities who worked for the KGB or GRU in apparent isolation from official Soviet consular missions. Special Agents fanned out across the country to investigate Bentley's leads and to monitor persons whom she had named, and for about a year the FBI entertained hopes of "doubling" her against the KGB.(40) The "Gregory" case--as the investigations prompted by Bentley were known inside the government-- produced many leads but led to no espionage prosecutions. FBI agents could not use evidence gathered by wiretaps in court, and they were unable to catch suspected spies in the act of compromising official secrets.(41) Meanwhile, Soviet agents and intelligence officers almost certainly surmised the existence of a serious leak. They took precautions even before a federal grand jury, meeting in 1947, probed Bentley's allegations and called as witnesses dozens of individuals named in her testimony. At roughly the same time, the renamed Army Security Agency (ASA--formerly the Signals Security Agency) developed evidence that would soon corroborate Bentley's testimony and the 1943 anonymous letter. After the war, the "Russian Section" at Arlington Hall expanded. Work on diplomatic messages benefited from additional technical personnel and new analysts--among them Samuel Chew, who had focused on Japan, and linguist Meredith Gardner, who had worked on both German and Japanese messages. Chew had considerable success at defining the underlying structure of the coded Russian texts. Gardner and his colleagues began analytically reconstructing the KGB codebooks. Late in 1946, Gardner broke the codebook's "spell table" for encoding English letters. With the solution of this spell table, ASA could read significant portions of messages that included English names and phrases. Gardner soon found himself reading a 1944 message listing prominent atomic scientists, including several with the Manhattan Project.(42) Gardner henceforth made rapid progress, reading dozens of messages sent between Moscow and New York in 1944 and 1945. By May 1947 he had read one that implied the Soviets ran an asset with access to sensitive information from the War Department General Staff.(43) It became apparent to Gardner that he was reading KGB messages showing massive Soviet espionage in the United States. Another problem soon arose--that of determining how and to whom to disseminate the extraordinary information Gardner was developing. ASA's reporting procedures did not seem appropriate because the decrypted messages could not even be paraphrased for Arlington Hall's regular intelligence customers without divulging their source. At this point, ASA knew nothing about the federal grand jury impaneled in Manhattan to probe the espionage and disloyalty charges leveled by Bentley and other defectors from Soviet intelligence, so no one in the US Government was aware that evidence against the Soviets was suddenly developing on two adjacent tracks. Gardner took matters into his own hands in the summer of 1947, drafting "Special Report #1," which went to a handful of senior ASA officials. One item in it about an unidentified Soviet asset would later prove fateful: LIB?? (Lieb?) or possibly LIBERAL: was ANTENKO [later understood as ANTENNA] until 29 Sept. 1944. Occurs 6 times, 22 October--20 December 1944. Message of 27 November speaks of his wife ETHEL, 29 years old married (?) 5 years, ". . . . . . . husband's work and the role of METR(O) and NIL." [Spelling and punctuation in original](44) http://www.cia.gov/csi/books/venona/preface.htm Page 9 of 23 Venona: Soviet Espionage and the American Response, 1939-1957 10/10/2005 07:58 AM Cooperation Expands Deputy G-2 Carter Clarke read Special Report #1 and in late August or early September 1948 asked FBI liaison officer S. Wesley Reynolds for a list of KGB and GRU covernames. Clarke's hint that ASA had broken a KGB code piqued the interest of the Bureau, which at that time was questioning former Soviet agents living in the United States concerning the allegations of Bentley and others and information gathered from surveillance of Soviet officials during the war. The Bureau quickly sent ASA a list of some 200 names, and, although few of them appeared in the translated messages, the long cooperation later known as the Venona program had begun.(45) This cooperative spirit between cryptanalysts and investigators endures to the present day. Full inter-agency cooperation, however, was still several years away. President Truman, unhappy about the mass of unanalyzed reports that the departments daily sent to the White House, had insisted in 1945 on greater coordination of intelligence information. His new Central Intelligence Group was intended to solve this problem, but it started out slowly. American intelligence agencies on the whole did not do a good job of presenting counterintelligence analyses to the President and his aides. FBI Director Hoover, for instance, frequently sent to the Truman White House allegations of Communist plotting and Soviet espionage. It is not clear how much of this information actually reached the President, however, or how seriously it was regarded by White House aides. Despite the Truman administration's sustained but piecemeal restructuring of the Intelligence Community, the division of labor in counter- intelligence functions remained much as it had been set early in World War II. The new National Security Council preserved the FBI's and armed services' monopoly of domestic counterintelligence in NSC-17/4 and 17/6 in 1949.(46) The agencies outside this monopoly were expected to provide information but were not invited to join operations involving domestic security. The new Central Intelligence Agency (CIA), the closest institutional successor to OSS and CIG and, consequently, the inheritor of OSS's dismal security reputation, saw little of the information Gardner and his colleagues were developing. CIA counterintelligence officers, however, now had wider access to signals intelligence than had their predecessors in X-2, and they briefly joined the Army and Navy in a Joint Counterintelligence Information Center (JCIC) to exploit current signals intelligence leads, using X-2's wartime employment of ULTRA as its model.(47) The JCIC received Special Report #1 at roughly the same time Colonel Clarke notified the FBI, but the Bureau never joined the JCIC or sought its assistance with the Soviet translations. When the JCIC inquired about additional Special Reports in early 1949, Clarke apparently instructed his subordinates not to provide anything. The early American effort to use the information from the Soviet messages thus remained understaffed and highly compartmented, and exploitation opportunities were almost certainly lost in consequence. For several years the major investigative burden remained with the FBI, which assigned the most important inter-agency liaison work to a single Special Agent, Robert Lamphere. "I stood in the vestibule of the enemy's house, having entered by stealth," Lamphere recalled in his memoir of the investigations.(48) Lamphere began sharing liaison duty with Wesley Reynolds in the spring of 1948. That October he had a private meeting with Meredith Gardner and began full-time liaison on the project. It was Lamphere's tenacity that taught the FBI how to use the translations against Soviet espionage. Through him the Bureau received a steady flow of translations and re-translations, as well as Gardner's insights about the "tradecraft" of Soviet spying. Gardner and his colleagues, in return, received collateral evidence, identifications, and additional leads.(49) The process was essentially a slow comparison of evidence for and against various competing hypotheses, with the knowledge gained in many cases being greater than the sum of its parts. http://www.cia.gov/csi/books/venona/preface.htm Page 10 of 23