Washington DC, November 14, 2008 - Forty-six years ago, a month before the Cuban Missile crisis, Soviet leaders put their strategic forces on their “highest readiness stage since the beginning of the Cold War,” according to a newly declassified internal history of the National Security Agency published today for the first time by the National Security Archive. Possibly responding to President Kennedy’s call for reserves, perhaps worried that the White House had discovered Moscow’s plans to deploy missiles on Cuba, the Kremlin kept forces on alert for 10 days, beginning on September 11, 1962.
The NSA’s signals intelligence (SIGINT) history also discloses that, a month later, on October 15th, the Soviets initiated a “precautionary, preliminary” alert, perhaps because Soviet Premier Nikita Khrushchev feared that U.S. intelligence had discovered the missiles. After President Kennedy’s speech on October 22nd 1962, announcing the “quarantine” (blockade) of Cuba, the Kremlin put military forces, especially air defense forces, on an “extraordinarily high state of alert.” Significantly, “offensive forces avoided assuming the highest readiness stage, as if to insure that Kennedy understood that the USSR would not launch first.”
In response to a declassification request by the National Security Archive, the secretive National Security Agency has declassified large portions of a four-part “top-secret Umbra” study, American Cryptology during the Cold War. Despite major redactions, this history discloses much new information about the agency’s history and the role of SIGINT and communications intelligence (COMINT) during the Cold War. Researched and written by NSA historian Thomas Johnson, the three parts released so far provide a frank assessment of the history of the Agency and its forerunners, warts-and-all.
According to National Security Archive visiting fellow Matthew Aid (author of the forthcoming history The Secret Sentry: The Top Secret History of the National Security Agency), Johnson’s study shows “refreshing openness and honesty, acknowledging both the NSA’s impressive successes and abject failures during the Cold War.” Another striking feature of Johnson’s study is the candor with which it discusses the fractious and damaging relationships between the agencies which make up the U.S. government’s intelligence establishment. Among the successes and failures disclosed by Johnson’s history are:

• After the end of World War II, with Soviet codes still unbreakable, the U.S. Army and Navy SIGINT organizations had relatively little to listen to. Johnson’s history reveals that as of mid-1946, the most productive source available to the U.S. Army SIGINT organization was French communications, which accounted for half of the finished reporting going to intelligence consumers in Washington.
  
• SIGINT coverage of the Soviet Union and the Peoples Republic of China by the Air Force Security Agency (an NSA predecessor) during the early 1950s was so bad that a senior CIA official referred to this period as “the dark ages for communications intelligence.”
  
• The discovery of high-level Soviet spies operating inside the Australian government in 1947 led the U.S. to cut off Australian access to classified U.S. government information, which was not resumed until two years later in 1949. Full SIGINT cooperation with Australia did not resume until 1953; according to Johnson, the Australian-American intelligence rift “had a deleterious affect on early U.S. SIGINT efforts against the Peoples Republic of China."
  
• During the 1950s, relations between senior officials at the CIA and NSA were at times so bad that they impeded cooperation between the two agencies. The CIA deliberately cut NSA out of the famous Berlin Tunnel operation (1954-1956), with NSA’s director, General Ralph Canine, finding out about the operation from the New York Times after the Soviets discovered the Tunnel in April 1956.
  
• By the early 1960s, the NSA was beginning to encounter information overload as more and more intercepted messages were stored in huge warehouses of magnetic tapes. According to Johnson, “the volume of unprocessed … tape was becoming difficult to manage technically and was embarrassing politically.”
  
• The Cuban Missile Crisis of 1962 was a major strategic intelligence failure for NSA. SIGINT provided no warning of the presence of Soviet nuclear-armed intermediate and medium-range ballistic missiles in Cuba prior to their discovery by U-2 reconnaissance aircraft; according to Johnson, this “marked the most significant failure of SIGINT to warn national leaders since World War II.”
  
• In April 1975, as the North Vietnamese military prepared for the final offensive to capture the beleaguered South Vietnamese capital of Saigon, ambassador Graham Martin refused to believe SIGINT reporting which clearly indicated that the offensive was about to commence, arguing that the intercepts were a “deception.” He believed that North Vietnamese wanted a coalition government, not military victory. The offensive began on April 26, 1975. Three days later, Saigon fell.
  
• Even though the 1970s was a period of lower budgets and dramatic personnel reductions for NSA, it regained some degree of access to Soviet encrypted communications during the late 1970s. A sentence that the Agency did not delete hints at this and other major cryptanalytic successes “Even with decreased money, cryptology was yielding the best information that it had produced since World War II.”
  
• Ten days before the Soviet Union invaded Afghanistan on 28 December 1979, U.S. intelligence agencies provided “specific warning” of the invasion. The post-mortems evaluating intelligence estimates of the Soviet invasion “were unanimous in describing [them] as an intelligence success.”
  
• During the 1960s and early 1970s, the NSA officials who ran the Agency’s domestic watch-list/eavesdropping program (Minaret) disguised the origin of their reports because they “seemed to understand that the operation was disreputable if not outright illegal.”
  

The NSA released the first three parts of American Cryptology during the Cold War in response to a mandatory review request filed by the National Security Archive. The excisions are currently under appeal, both at the NSA as well as the Interagency Security Classification Appeals Panel. With book four of the history recently completed, the Archive has also requested its declassification.
The National Security Agency during the Cold War
Commentary by Matthew M. Aid
Dr. Thomas R. Johnson’s four-part top secret codeword history of the National Security Agency, American Cryptology during the Cold War, 1945-1989), three parts of which have been released to date, is a unique and invaluable study for readers interested in the history of U.S. intelligence during the Cold War or for those who are simply interested in the role of the secretive National Security Agency in the U.S. government.
U.S. intelligence agencies have produced numerous single-volume histories, usually published at the unclassified level and meant for public distribution. In all but a few cases, these histories tend to be tendentious and emphasize all positive aspects of their agency’s accomplishments while ignoring all of the mistakes and miscues endemic to U.S. intelligence history. For example, most unclassified histories produced by the Central Intelligence Agency’s Center for the Study of Intelligence (CSI) have focused on those historical episodes or intelligence collection systems, which were unqualified successes, such as the histories of the U-2 reconnaissance aircraft and the recently declassified internal history of the 1954-1956 Berlin Tunnel Project. (Note 1) But the CIA has consistently refused to declassified any of its histories on the less successful episodes in the Agency’s history, such as the CIA’s spectacular failures attempting to conduct clandestine intelligence gathering (or human intelligence, HUMINT) and covert action operations inside the USSR and Eastern Europe during the 1940s and 1950s, or within the Peoples Republic of China during the 1950s and 1960s. (Note 2) Sadly, the Agency’s reluctance to release these and other critical internal histories was reinforced by the negative publicity that the Agency garnered when it released in 2007 the so-called “Family Jewels” documents concerning CIA improper or illegal activities prior to 1973.
One of the things that make Dr. Johnson’s history unique among official histories is its refreshing openness and honesty, acknowledging both the NSA’s impressive successes (Johnson states that “No other intelligence source had the revolutionary impact of SIGINT” (Book I, p. 1) and abject failures during the Cold War. For example, the Johnson history frankly acknowledges that one of the single greatest impediments to an effective U.S. national signals intelligence (SIGINT) effort during the Cold was a lack of cooperation and unity of effort within the U.S. intelligence community. This fractious relationship dates back to before the beginning of World War II, when the U.S. Army and Navy SIGINT organizations refused to cooperate with one another (see Book I, pp. 3-7). Internecine warfare amongst the three military services stalled the creation in May 1949 of America’s first unified SIGINT agency, the Armed Forces Security Agency (AFSA), then left it bereft of any meaningful power or influence. Not surprisingly, AFSA was an abject failure as an institution, and was disbanded three and one-half years late in October 1952 in favor of a new and truly unified SIGINT organization, the National Security Agency (NSA) (Book I, pp. 23-35).

Early Cold War
Johnson’s history provides some tantalizing glimpses, for example, into NSA’s cryptologic successes and failures against the Soviet Union during the Cold War (Book I, pp. 157-194). The report describes the genesis of NSA’s attack on Soviet codes and ciphers during World War II, including details about the Anglo-American solution of the Venona one-time pad ciphers used by the KGB during World War II. Unfortunately, so far NSA has refused to declassify the portion of Dr. Johnson’s history pertaining to “Black Friday,” when Army intelligence discovered the across-the-board change of Soviet codes and ciphers in October 1948. This effectively wiped out all Anglo-American cryptanalytic access to Moscow’s high-level communications. The history is also silent about the fact that it took NSA almost thirty years before it was able to solve Soviet high-level cipher systems again. Nevertheless, NSA has released some discussion, albeit in a highly redacted form, of its successes in monitoring Soviet missile tests, finding Soviet ICBM launch sites, tracking Soviet submarines, and the critical SIGINT support that NSA provided to the CIA’s U-2 overflights of the Soviet Union between 1956 and 1960.
While U.S. Army and Navy SIGINT specialists were trying to crack Soviet codes after the end of World War II, they had relatively little to listen to. Johnson’s history reveals that as of mid-1946, the most productive source available to the U.S. Army SIGINT organization was French communications, which accounted for half of the finished reporting going to intelligence consumers in Washington. (Book I, p. 10)
The history shows that AFSA’s performance during the Korean War (1950-1953) was marked by occasional successes and a series of shocking failures (Book I, pp. 43-56). The Agency provided no warning that North Korea intended to invade South Korea in June 1950 because it was paying no attention to Korea. AFSA provided vitally important intelligence during the Battle of the Pusan Perimeter (August-September 1950), but the intelligence it generated in October-November 1950 indicating that Communist China intended to intervene militarily in the Korea was ignored or badly misinterpreted by senior U.S. government officials and military officials, including General Douglas MacArthur. The report shows that for the first year of the war, AFSA experienced considerable success breaking North Korean ciphers, but curiously NSA refused to declassify the fact that throughout the entire Korean War AFSA’s cryptanalysts were unable to break any significant Chinese military cipher systems.
The situation changed dramatically in the summer of 1951 when AFSA lost its access to North Korean traffic; the enemy changed all its codes and ciphers to unbreakable one-time pad cipher systems. This meant that for the remaining two years of the Korean War AFSA was forced to depend on low-level voice intercept and traffic analysis for virtually everything that it knew about the Chinese and North Korean militaries. And finally, the Johnson history reveals (but the reader has to read between the lines) that the U.S. refused to assist the South Korean military in forming its own SIGINT service because of security concerns.
Chinese communications systems, as well as Soviet, remained a tough problem for the AFSA. The AFSA’s SIGINT coverage of the Soviet Union and the Peoples Republic of China during the early 1950s was so bad that a senior CIA official referred to this period as “the dark ages for communications intelligence.” (Book I, p. 61) In any event, NSA was not doing well with the information that it had. According to Johnson’s account, the Agency’s early SIGINT reporting to consumers left much to be desired, with a 1953 report stating that NSA intelligence product was “generally so cluttered with qualifying expressions as to virtually preclude their use by a consumer.” (Book I, p. 70)
As NSA became more and more technically proficient, it created a problem that was even more serious than anodyne reporting. By the early 1960s, the Agency was experiencing an information overload as it stored more and more intercepted messages in huge warehouses of magnetic tapes. According to Johnson, “the volume of unprocessed … tape was becoming difficult to manage technically and was embarrassing politically.” (Book II, p. 373) This was a problem that would bedevil the NSA for decades.
Cuban Missile Crisis
NSA’s performance during the Cuban Missile Crisis (Book II, pp. 317-332) was superior, especially in the important area of tracking the movements of Soviet merchant ships carrying Soviet troops, weapons and equipment to Cuba. Nevertheless, the Johnson history reveals that SIGINT picked up no indication that the Soviets had placed offensive ballistic missiles in Cuba prior to their being discovered by a CIA U-2 reconnaissance aircraft in October 1962. This failure had significant implications because since the first days of the Eisenhower administration, the U.S. intelligence community had depended on NSA for 90 percent of their intelligence information warning of a Soviet strategic threat to the U.S. Dr. Johnson concludes that “SIGINT warning, so highly touted during the Eisenhower administration, failed in Cuba.”
Despite the NSA’s failure, it kept the White House and the Pentagon informed of Soviet military activities. Disclosed for the first time in the NSA account is that U.S. intelligence tracked the readiness condition of Soviet air defense and strategic forces during the Crisis. What has remained secret for years is that Soviet forces went on high alert three times during September and October 1962. The first was on 11 September 1962, when for ten days “Soviet forces went into their highest readiness stage since the beginning of the Cold War,” perhaps because the Soviets believed that U.S. intelligence had learned about the missile deployments. Especially telling is that also on 11 September, the Kremlin publicized its apprehension that President Kennedy’s request to Congress for stand-by authority to call up reservists foreshadowed an attack on Cuba, which the Soviets said was grounds for war. (Note 3) Another alert of a more “precautionary, preliminary” nature began on 15 October, perhaps also because Khrushchev supposed that the missiles had been discovered. Finally, after Kennedy’s speech, Soviet forces went on an “extraordinarily high state of alert,” with the emphasis on air defense forces. Significantly, “offensive forces avoided assuming the highest readiness stage, as if to insure that Kennedy understood that the USSR would not launch first.” (Book II, p. 331)
Johnson’s account of the Missile Crisis illuminates a failure of intelligence cooperation, which is a major theme in his study. On October 23, 1962, with the Cuban Missile Crisis at its height, the Director of Naval Intelligence failed to inform the White House or Secretary of Defense Robert McNamara that new High-Frequency Direction Finding (HFDF) data showed that many of the Soviet merchant ships bound for Cuba had already stopped dead in the water or had turned back for Russia. The Secretary of Defense discovered this huge mistake the next day, with Johnson noting that “McNamara was furious, and he subjected Admiral Anderson, the Chief of Naval Operations, to an abusive tirade. So many years have passed that it is impossible to determine why the Navy held up information that seemed critical to the president’s decisions.” (Book II, p. 329)
SIGINT at the White House
An important development that began during the 1960s was the growing use of SIGINT at the White House. Over the objections of the CIA and State Department, President Lyndon Johnson insisted that NSA transmit SIGINT directly to the White House Situation Room. Johnson may have been the “most avid consumer of intelligence ever to occupy the White House” and he read SIGINT constantly to support his decisions during the Vietnam War. SIGINT would continue to be available to White House officials in the years that followed, although they would use it in different ways. For example, Richard Nixon was not interested in reading intelligence reporting and his security assistant Henry Kissinger was not as “experienced” with SIGINT as his predecessor Walt Rostow had been. The Johnson history is critical of the way that Kissinger handled SIGINT: his reports to Nixon would “subsume [SIGINT] into a mishmash of sources” and not highlight it as Johnson’s advisers had. (Book II, pp. 353-354 and 486)
In spite of a declining ability to generate high-level intelligence about what was going on inside the USSR and China, codebreaking efforts against targets in the Third World improved dramatically (Book II, pp. 425-475). For example, SIGINT provided important intelligence information prior to and during the 1967 Middle East War. Moreover, SIGINT helped U.S. intelligence monitor developments in the Warsaw Pact. During the summer of 1968, SIGINT reporting coming out of NSA clearly showed that growing numbers of Soviet and Warsaw Pact troops were being deployed along the borders of Czechoslovakia. The invasion did not take the White House by surprise (it had “strategic warning”), but the CIA did not provide advance warning because CIA analysts refused to accept the possibility that the Soviets would invade the country (although a minority believed otherwise (Note 4)). (Book II, pp. 454-461). Interestingly, SIGINT also picked up what some U.S. analysts saw as possible Soviet move against Romania, which led President Johnson to make a public warning to Moscow. (Book II, p. 462)
The growing NSA presence greatly increased the vulnerability of its operators to physical harm. In June 1967, Israeli aircraft and torpedo boats attacked the NSA spy ship USS Liberty, killing 34 crewmembers, including 25 military and civilian cryptologists. Dr. Johnson concludes that based on the available SIGINT that the Israelis did not know that the ship they were attacking was a U.S. Navy ship. In January 1968, the North Koreans seized the U.S. Navy spy ship USS Pueblo, which Johnson correctly describes as an intelligence disaster of unparalleled proportions. Then in January 1969, a North Korean MiG-21 fighter shot down a U.S. Navy EC-121 SIGINT aircraft, killing all 31 crewmembers, including nine military cryptologists.
Johnson devotes a significant amount of space to NSA’s involvement in the Vietnam War, beginning with the arrival of the first American SIGINT personnel in 1961 and concluding with the fall of Saigon in 1975 (Book II, pp. 495-584). One of the major revelations stemming from Dr. Johnson’s review is that SIGINT was the principal driver of the U.S. military’s counterinsurgency operations in South Vietnam throughout the war, with most of the Army’s major search-and-destroy missions being predicated on intelligence derived from SIGINT. (Book II, pp. 534-538). On the controversial August 1964 Tonkin Gulf incidents, which served as the predicate for America’s entry into the Vietnam War, Dr. Johnson concludes that: “The White House had started a war on the basis of unconfirmed (and later-to-be-determined probably invalid) information.” (Book II, p. 522) (Note 5)
NSA’s ability to locate North Vietnamese and Viet Cong troop concentrations in South Vietnam and Laos and track their movements via SIGINT is discussed in detail, which led to a number of major battlefield successes during the war. Dr. Johnson reveals that SIGINT was able to provide significant advance warning of virtually all NVA/VC offensives from 1966 until the end of the war in 1975 (p. 539); SIGINT was the only reliable source of intelligence concerning the number of North Vietnamese troops coming down the Ho Chi Minh Trail from the fall of 1967 onwards (Book II, pp. 539-540); and NSA provided 55% of all targeting information for U.S. bombers during the Vietnam War (p. 583).
The history also confirms that NSA provided advanced warning of the North Vietnamese Tet Offensive in January 1968, but President Johnson and the CIA in Washington, along with General William Westmoreland in Saigon, appear to have discounted NSA reporting that the NVA and VC were about to launch a nationwide offensive. Instead, they held that the upcoming offensive would focus on U.S. forces further north. (pp. 562-563). The Johnson history also reveals that U.S. military SIGINT units in Vietnam were heavily dependent on South Vietnamese translators to intercept and process enemy radio traffic, but, as in the Korean War, the U.S. government, at NSA’s urging, forbade giving SIGINT-derived intelligence to the South Vietnamese government or military (pp. 509-510).
Another intelligence failure occurred at the end of the Vietnam conflict. In April 1975, as the North Vietnamese military prepared for the final offensive to capture the beleaguered South Vietnamese capital of Saigon, the U.S. ambassador in Saigon, Graham Martin, refused to believe NSA SIGINT reporting indicating that the North Vietnamese offensive was about to begin. Believing that the North Vietnamese wanted a coalition government, not a military victory, Martin argued that the intercepts were a “NVA deception.” The NVA offensive began on April 26, 1975. Three days later, Saigon fell. (Book III, p. 9)
The 1970s
The NSA’s declassification staff heavily redacted Book III covering the turbulent decade of the 1970s. Nevertheless, one can find a number of interesting tidbits sandwiched among the deletions. For example, the brief but interesting section on NSA’s domestic eavesdropping programs (Shamrock and Minaret), includes a telling quote on p. 85: “Years later the NSA lawyer who first looked at the procedural aspects stated that the people involved [in Minaret] seemed to understand that the operation was disreputable if not outright illegal.” Dr. Johnson also details NSA’s fractious relations with the various congressional committees established during the mid-1970s to investigate abuses by the U.S. intelligence community, which involved unprecedented public testimony by NSA officials before the Church Committee (although the Committee gave them two days of rehearsal!). Johnson also covers the effort to bring domestic intercepts more squarely under the rule of law, with the passing of the Foreign Intelligence Surveillance Act (FISA) in 1978 (Book III, pp. 94, 106-107).
Despite the fact that the 1970s was a period of lower budgets and dramatic personnel reductions at Fort Meade, the Agency finally regained some degree of access to Soviet encrypted communications during the Carter administration in the late 1970s. This and a number of other major cryptanalytic successes are hinted at in a sentence that NSA did not delete from Dr. Johnson’s text, which stated: “Even with decreased money, cryptology was yielding the best information that it had produced since World War II.” (Book III, p. vii). Also surviving the security review is some of the discussion of the Soviet invasion of Afghanistan. According to Johnson, “Generalized warnings [of an invasion] had begun in September, and specific warnings preceded the operation by at least ten days.” Thus, intelligence post-mortems “were unanimous in describing it as intelligence success.” (p. 254) (Note 6)
Inter-Agency Rivalries
As indicated earlier, the Johnson history shows in stark detail that NSA’s relations with the U.S. military and the U.S. intelligence community was often troubled. For example, the coverage of the NSA’s relationship with the U.S. Air Force during the early 1950s provides a history of nearly constant internecine warfare over whether NSA had the authority to control the USAF’s SIGINT activities (Book I, pp. 80-83). The Johnson study also shows how the National Reconnaissance Office (NRO), along with the CIA, marginalized the NSA when it sent up SIGINT satellites. Despite NSA’s responsibilities, it had very little management or tasking responsibility for the first generation of SIGINT satellites sent into orbit by the NRO in the early 1960s. According to Johnson, “NSA was still a minor player. It had very few cleared people, and its only responsibility was to process and report ELINT data.” (Book II, p. 405)
Most revealing are details (hidden amidst a blizzard of deletions by the CIA) concerning NSA’s series of no-holds barred bureaucratic turf battles with the CIA and its predecessors that began immediately after the end of World War II and continued without interruption right up until the 9/11 terrorist attacks in September 2001 (Book I, pp. 86-107; Book II, pp. 341-343). The Berlin Tunnel episode provides a starting example. The CIA deliberately cut NSA out of the famous Berlin Tunnel operation (1954-1956), with NSA’s director, General Ralph Canine, finding out about the operation from the New York Times after the Soviets discovered the Tunnel in April 1956. This episode only served to further upset the already rancorous CIA-NSA relationship, with the Johnson history noting that General Canine “was understandably upset when he found out that he had been bypassed and left in the dark.” (Book I, p. 106) Even a NSA-CIA “peace treaty” signed in 1977, details of which are deleted from the history, failed to put an end to the seemingly never-ending battles between the two agencies (Book III, pp. 224-231). This lack of unity of effort plagued NSA’s SIGINT collection and analytic efforts throughout and after the Cold War, to the ultimate detriment of U.S. national security.
On one of Washington’s most important intelligence relationships, the Anglo-American connection, the history includes a brief but revealing discussion of the interactions of NSA’s predecessors with some of its English-speaking SIGINT partners, and the difficulties they experienced in fashioning a postwar SIGINT alliance that focused on a new target – the USSR. (Book I, pp. 13-19) The late 1945 Gouzenko spy scandal delayed U.S. acceptance of Canada as a full-fledged cryptologic partner until Ottawa signed a separate SIGINT-sharing agreement (the CANUSA Agreement) in the fall of 1949. The discovery of high-level Soviet spies operating inside the Australian government in 1947 led the U.S. to cut off Australian access to classified U.S. government information for two years. Full U.S.-Australia SIGINT cooperation did not resume until 1953, with Johnson noting that the rift “had a deleterious affect on early U.S. SIGINT efforts against the Peoples Republic of China.”(Book I, pp. 16-19)
Unlike the CIA, where the Agency’s censors religiously delete all information concerning the size of the Agency’s staff and budget, considerable detail concerning the organization and manpower strength of NSA are revealed for the first time in Dr. Johnson’s history, confirming that the Agency for most of the Cold War was the single largest and most expensive component of the U.S. intelligence community (Book I, pp. 63-67; Book II, pp. 293-294). NSA reached its historic peak strength in 1969, with 93,067 military and civilian cryptologists working for the Agency and the three military service cryptologic agencies that were subordinate to NSA. (Book II, p. 293).
While NSA’s decision to release the Johnson history is commendable, the many and frequently lengthy excisions gives pause for thought. No doubt, much about the NSA’s history necessarily remains secret, but the copious excisions make one suspect that the security reviewers have gone too far. For example, after 60 years, NSA should be able to release something about “Black Friday.” Moreover, excisions about the Berlin Tunnel project (Book I, 104-105) fly in the face of the CIA’s declassification of its internal history of that well-known operation. (Note 7) Another futile excision concerns verification problems raised by the SALT (Strategic Arms Limitation Talks) II Agreement. (Book III, pp. 202-206, and pp. 219 ff) No doubt, much of the deleted information concerns the impact of the 1979 Iranian revolution and the loss of SIGINT facilities which played a major role in tracking Soviet ICBM tests, especially intercepting missile telemetry. Given that the memoirs of President Jimmy Carter and former Secretary of State Cyrus Vance included frank discussions of the Iranian listening posts, it should be relatively easy for the NSA to declassify some information on SALT II verification. Not all of the excisions, however, are NSA’s responsibility. The many “OGA” [Other Government Agency] excisions are evidence of the CIA’s security review.
The first three parts of American Cryptology during the Cold War were released to the National Security Archive through a mandatory review request. To challenge the significant excisions, the Archive filed appeals, which led NSA to release more information from book II and III. The Archive has appealed the many excisions that remain with the Interagency Security Classification Appeals Panel. NSA has not yet complete work on the Archive’s appeal of book I, apparently because of delays at the Central Intelligence Agency. With book 4 recently completed, the Archive has requested the NSA to release it as well.
Matthew Aid, a visiting scholar at the National Security Archive, has written extensively on the history of the National Security Agency and U.S. SIGINT programs for Intelligence and National Security, among other scholarly journals. His book, The Secret Sentry: The Top Secret History of the National Security Agency will be published by Bloomsbury in early 2009. He thanks John Prados and Jeff Richelson for their comments.
Read the Documents
Document 1: Thomas R. Johnson, American Cryptology during the Cold War, 1945-1989: Book I: The Struggle for Centralization, 1945-1960 (National Security Agency: Center for Cryptological History, 1995), Top Secret Umbra, Excised copy, pp. i-xvii and 1-155
Document 2: Thomas R. Johnson, American Cryptology during the Cold War, 1945-1989: Book I: The Struggle for Centralization, 1945-1960, pp. 157-287
Document 3: Thomas R. Johnson, American Cryptology during the Cold War, 1945-1989: Book II: Centralization Wins, 1960-1972 (National Security Agency: Center for Cryptological History, 1995), Top Secret Umbra, Excised copy, pp. 289-494
Document 4: Thomas R. Johnson, American Cryptology during the Cold War, 1945-1989: Book II: Centralization Wins, 1960-1972, pp. 495-652
Document 5: Thomas R. Johnson, American Cryptology during the Cold War, 1945-1989: Book III: Retrenchment and Reform, 1972-1980 (National Security Agency: Center for Cryptological History, 1998), Top Secret Umbra, Excised copy, pp. i-ix, and 1-116
Document 6: Thomas R. Johnson, American Cryptology during the Cold War, 1945-1989: Book III: Retrenchment and Reform, 1972-1980, pp. 117-262

Notes
1. A significant exception is Harold Ford’s CIA and the Vietnam Policymakers: Three Episodes, 1962-1968 (CIA History Staff: Center for the Study of Intelligence, 1998) which reviews the intelligence controversies, warts and all, even the role of DCI John McCone in “distort[ing]” the findings of a National Intelligence Estimate so that it had a less pessimistic take on the degree of progress in South Vietnam.

2. For example, so far the CIA has refused to declassify most of the contents of a major history, "Office of Policy Coordination, 1948-1952," although a pending appeal before the Interagency Security Classification Appeals Panel could lead to a reversal.
3. “Kennedy Assailed,” by Seymour Topping, The New York Times, 12 September 1962.
4. Jeffrey Richelson, A Century of Spies: Intelligence in the Twentieth Century (New York: Oxford University Press, 1994), p. 327.
5. It is worth noting that there are interesting differences between Johnson’s account of the Gulf of Tonkin incident (as well as the Tet offensive) and that offered by NSA historian Robert Hanyok in “Skunks, Bogies, Silent Hounds, and the Flying Fish: The Gulf of Tonkin Mystery, 2-4 August 1964,” at http://www.nsa.gov/vietnam/releases/relea00012.pdf, and in Spartans in Darkness: American SIGINT and the Indochina War, 1945-1975 (National Security Agency: Center for Cryptological History. 2002), at http://www.fas.org/irp/nsa/spartans/index.html. For example, while Hanyok argues that NSA officials manipulated intelligence on the August 4 events in the Tonkin Gulf, Johnson suggests that the analysts misinterpreted North Vietnamese messages.

6. According to a CIA study by Douglas MacEachin’s study, Afghanistan was a collection success but an analytical failure; see Predicting the Soviet Invasion of Afghanistan: The Intelligence Community's Record (Central Intelligence Agency: Center for the Study of Intelligence, 2002).
7. Central Intelligence Agency, “Clandestine Services History: The Berlin Tunnel Operation, 1952-1956,” 24 June 1968, at http://www.foia.cia.gov/browse_docs.asp (The Agency has released several versions of this history; this links to the most complete release). For the inconsistencies, see Federation of American Scientists, Secrecy News, at http://ftp.fas.org/sgp/news/secrecy/2007/12/120607.html