Code Warriors
Page 36
The President’s Foreign Intelligence Advisory Board had warned as early as the mid-1960s of NSA’s tendency to become “ingrown and defensive”; a more comprehensive review by the consulting firm Arthur D. Little in the late 1970s found that NSA’s management had become “paranoid,” “untrustworthy,” and “uncooperative,” divided into fiefdoms under the control of entrenched senior managers each defending their own institutional interests rather than responding to the mission of the agency as a whole.6
The revolving door of military flag officers serving as director for a four-year term certainly had its own problems; all were outsiders to the agency, and the kind of background, personality, and temperament that it took to make it to three-star general or admiral did not always have much to do with the technical, management, and Washington political skills required to run an exceedingly complex organization like NSA. Tordella’s firm grip of the technical management of the agency was both a buffer against martinets like Admiral Frost, who tried to run NSA like a carrier battle group, and a welcome relief to technically unsavvy officeholders like Noel Gayler, a vice admiral who served as director from 1969 to 1972 and often seemed frankly bewildered why he had been selected for the job and was happy to leave the running of the place to his civilian deputies. But either way it was a fundamentally dysfunctional situation. The PFIAB meanwhile complained repeatedly about the organizational “anarchy” arising out of the divided loyalties of NSA’s military directors, who were often reluctant to challenge the wishes of the Joint Chiefs, notably when it came to asserting NSA control of the ELINT mission. Despite a series of clear directives giving NSA the authority to assume command of the fragmented and highly duplicative ELINT programs of the services, no NSA director ever seemed willing to rock that boat.7
Of course, it was hardly a new phenomenon in the annals of government to find a bureaucracy run by the bureaucrats rather than by the political appointee nominally in charge, but NSA’s growing inside political influence and privileged relationship with the White House further insulated the agency from even the normal checks of management accountability and government oversight. NSA had always shrugged off efforts to make it more accountable to its governing board or the director of central intelligence, who by statute had responsibility for coordinating the work of all government intelligence agencies, including setting their budgets. A 1958 directive by the National Security Council, approved by President Eisenhower over NSA’s objections, had abolished the U.S. Communications Intelligence Board, declared NSA to be a part of the intelligence community, and placed it under a new, unified U.S. Intelligence Board (USIB), headed by the CIA director. But that had had no discernible effect on NSA’s increasing independence.8 A blistering CIA memo in the mid-1970s on the “CIA/NSA Relationship” pointed out just how much its rival agency had been able to become answerable to no one but itself as a result of its favored status and a budget that had “doubled, tripled and quadrupled”:
An organization which began with a serious inferiority complex gradually developed a feeling that it has “a corner on the market” in terms of intelligence information fit to print.
This new feeling of importance by NSA manifested itself in various ways such as the installation of a direct communications link over CIA objections between Ft. Meade and the White House and the issuance of the SIGINT Summary, a SIGINT current intelligence publication designed to compete with the then Central Intelligence Bulletin. CIA also objected, to no avail, to the SIGINT Summary because it contained then as now gists and summaries of what NSA analysts consider to be “hot” items of information which were in the process of being published in individual translation or report form, but for which NSA wanted to get credit in the eyes of top level intelligence recipients….NSA reps let it be known in numerous ways that there was little or no need for “middlemen” such as CIA, DIA, etc. to chew, digest and regurgitate perfectly good SIGINT data and provide it to the real intelligence consumers such as the President, the Secretary of State and the NSC Staff….
The increasingly aggressive, determined and sometimes overbearing policy on NSA’s part…have resulted almost by default in the emergence of NSA in a Community role in which the tail too often wags the dog.9
Hoarding information was also a venerable element of the bureaucratic power game, but NSA could trump even CIA in that contest, given the special handling that SIGINT had always been accorded, and increasingly NSA withheld details in its reports that it considered “technical information” on sources, intercept locations, and cryptographic systems. A later CIA director, Admiral Stansfield Turner, who headed that agency from 1977 to 1981, complained in his memoirs that aside from dangerously skewing the intelligence picture by overemphasizing SIGINT at the cost of other sources, NSA was mainly trying to “get credit for the scoop….There is a fine line here, but there is no question in my mind that the NSA regularly and deliberately draws that line to make itself look good rather than to protect secrets.”10
The chief of the Situation Room when Nixon entered the White House was David McManis, the first time an NSA official had held the position, and a majority of the intelligence coming in was from the SIGINT system. Continuing where Lyndon Johnson left off, Nixon’s national security adviser, Henry Kissinger, ordered that any intercepts mentioning the president or himself be delivered to him and to no one else in government: all were to be marked NODIS—Not for Distribution. Gayler and Tordella were only too happy to oblige, seeing a further opportunity to bolster NSA’s standing.11
More cravenly, or venally, the two top NSA officials also backed an extremely secret, and undoubtedly illegal, plan drawn up by the Nixon White House to authorize a vastly expanded monitoring of Nixon’s domestic critics by the intelligence agencies. The Huston plan, as it was known, was eventually withdrawn in the face of strenuous opposition by FBI director J. Edgar Hoover (who had no moral objections to spying on left-wing and antiwar groups, but insisted that it was the FBI’s job). But even after the dropping of the Huston plan, NSA obligingly added hundreds of names submitted by Nixon administration officials to its secret “watch list” of domestic surveillance targets.12
Nixon and Kissinger’s efforts to turn the intelligence agencies into a tool of the White House greatly exacerbated the dysfunctional relationships throughout the intelligence system, forcing military and diplomatic officials to engage in palace intrigues of their own just to maintain their routine access to NSA intelligence reports. Kissinger tried to cut out Secretary of Defense Melvin Laird and Secretary of State William P. Rogers from receiving anything but the most “innocuous” intercepts, as Laird’s military assistant described the material that arrived in the secretary’s office via official channels. Laird shrewdly countered by ensuring his own back channel. As he later told the journalist Seymour Hersh, he called Vice Admiral Gayler and the Army three-star officer he named to direct the Defense Intelligence Agency, Lieutenant General Donald V. Bennett, into his office: “I told them they’d better be loyal to me. If they were, they’d get four stars after four years. And goddamn it, they were loyal.” Laird often met privately with Gayler two or three times a week to find out what NSA was giving the White House. Both Gayler and Bennett got their fourth stars.13
NSA’s extremely cozy relationship with the Nixon White House and the end runs that it forced the secretary of defense and others to make to maintain the normal flow of intelligence was the result of NSA’s own assiduous efforts to promote its standing in the corridors of Washington power. “It was not good for SIGINT,” acknowledged an NSA internal history, “and it was deadly for the presidency.” But while it lasted it remained extremely good for NSA.14
—
For a while it looked as if NSA might even weather the post-Watergate storms that ripped away the covers which had long kept American intelligence operations out of public sight and free from constitutional scrutiny. “The CIA had always operated under minimal Congressional oversight,” observed John Lewis Gaddis. “The assumption h
ad been that the nation’s representatives neither needed nor wanted to know what the Agency was doing. That attitude had survived the U-2 and Bay of Pigs incidents, the onset and escalation of the Vietnam War…but it did not survive Watergate.”15
The revelation that Nixon had ordered CIA to tap phones and open mail of “left-wing radicals” and antiwar protestors and that former CIA officers were part of the White House “Plumbers Unit,” which undertook a series of burglaries, surveillance operations, and wiretaps against domestic “enemies” on Nixon’s direct orders, brought CIA under unprecedented scrutiny in the press, and then in Congress, after the Plumbers were caught breaking into Democratic National Committee headquarters in the Watergate office building in Washington during the 1972 presidential election campaign and the whole story started to unravel. By law CIA, like NSA, was restricted to foreign intelligence activities and could not legally operate within the United States. Nixon had swept those concerns aside (insisting to interviewer David Frost even after his resignation, “Well, when the president does it, that means it is not illegal”). In 1973, during the Watergate investigations that would lead to Nixon’s resignation the following year in the face of a certain House vote for impeachment, the CIA’s director of operations, William Colby, ordered a complete internal review of the agency’s past illegal activities. Parts of the resulting 693-page report—known a bit flippantly within the agency as “the family jewels”—were leaked to Hersh, who wrote a long article in the New York Times in December 1974 revealing CIA spying on Nixon’s domestic critics. In an effort to make a clean sweep, Colby, who became the agency’s director in the ensuing shakeup, immediately confirmed the accuracy of the article and pledged full cooperation with Senate, House, and presidential commissions that were quickly appointed to conduct a thorough, and unprecedentedly public, investigation of U.S. intelligence operations.16
NSA had largely succeeded in keeping its fingerprints off the Nixon White House’s more serious illegal operations; and the special status signals intelligence had long enjoyed as a secret within a secret in the U.S. intelligence system stymied the initial efforts by congressional investigators to learn much of anything about its activities, legal or otherwise. L. Britt Snider was a thirty-year-old lawyer for the Senate Select Committee on Intelligence, chaired by Democrat Frank Church of Idaho, and along with another staff member was given the job to tackle NSA. “They must have done something,” the committee’s staff director kept telling them. Snider began by asking the Congressional Research Service for everything in the public record referring to NSA. The researchers came up with a one-paragraph description of the agency from the Government Organization Manual and “a patently erroneous piece from Rolling Stone magazine.”
Hoping to locate some insider whistle-blowers, Snider and his colleague, Peter Fenn, tracked down and went to see a handful of NSA retirees living in the Washington area; the most egregious abuses they had to report “were complaints about how NSA allocated its parking spaces among employees and a few cases of time and attendance fraud.” But it was clear from the interviews that they were up against another impenetrable barrier, the strict compartmentalization that limited the flow of information within NSA: even if NSA had been involved in improper activities, knowledge of the fact was sure to be restricted to a tiny circle.
Trying the direct approach, Snider and Fenn then decided to simply ask for a meeting with the director. That elicited NSA’s trademark dazzle-the-natives treatment: VIP visitor parking spots at the front door, “broadly smiling handlers” waiting to whisk them to the director’s top-floor suite, and an earnest promise from Lieutenant General Lew Allen Jr. himself, the agency’s director since 1973, of full cooperation—but nothing, even after the weeks of personal briefings by NSA officials that followed, that identified “a single avenue that appeared promising from an investigative standpoint,” Snider wrote.
Their first real break came in May 1975 when the committee received a copy of the CIA family jewels report, which on close reading contained two small references to NSA. One alluded to a CIA-supplied office in New York City used by NSA to copy telegrams. The other mentioned NSA’s involvement in monitoring the communications of members of the antiwar movement. But it was only three months later, after another New York Times story appeared under the headline “National Security Agency Reported Eavesdropping on Most Private Cables” that NSA agreed to provide a full account to the congressional investigators of its long-running arrangement with the cable companies to copy every international telegram. Pressed by Snider on how the program began and who had approved it, the NSA briefer said the only person alive who knew the whole story was “Dr. Tordella,” recently retired as deputy director.
Once he saw how much Snider already knew, Tordella proved surprisingly cooperative when Snider showed up at his home in suburban Maryland. A single manager, who reported directly to Tordella, was responsible for the program, code-named Shamrock. A courier would each day collect microfilm copies or punched paper tape of international telegrams that passed through the offices of the three major cable companies in New York, Washington, and San Francisco; in the early 1960s, when the companies switched to magnetic tape, NSA set up an office in New York to duplicate the tapes and keep a copy for itself. Tordella said the collection program “just ran on” ever since its beginnings in World War II “without a great deal of attention from anyone,” and actually “wasn’t producing very much of value.”17
That was probably true enough, and it appeared that for the most part the only cables normally retained for further study were those that an initial automatic processing identified as having been sent by foreign embassies or that appeared to be enciphered. Still, the existence of such a “vacuum cleaner” surveillance program, as the press quickly called it, was a grave embarrassment, almost certainly a violation of the law, and NSA vociferously objected to the committee’s exposing any further details about it even though Allen assured the investigators he had ordered the program ended that May. The Church Committee nonetheless released in November 1975 a full report on NSA’s bulk cable collection practices. Given that the agency since the 1950s had been running a “New Shamrock” program that directly tapped teleprinter and other communication links of sixty to seventy foreign embassies to obtain the same information with far less trouble, the major damage done by the loss and exposure of the old Shamrock was to the agency’s reputation.18
Far more troubling was the other program that Snider and Fenn stumbled on, code-named Minaret. What had begun in 1962 as a so-called watch list of Americans traveling to Cuba whom NSA was asked to monitor was expanded in the mid-1960s to include suspected narcotics traffickers, and then in 1967, on Lyndon Johnson’s orders, members of the antiwar movement, who Johnson believed were being aided by foreign powers. Under the increasingly paranoid Nixon administration the watch list exploded, and NSA was soon up to its elbows spying directly for the White House on the communications of more than sixteen hundred American citizens who had done nothing more than arouse Nixon’s or Kissinger’s suspicions or dislike. Among them were the civil rights leaders Martin Luther King Jr. and Whitney Young, the heavyweight boxing champion and Vietnam War opponent Muhammad Ali, the New York Times columnist Tom Wicker and the humorist Art Buchwald, U.S. senators Frank Church and Howard Baker, and a few names of NSA’s own choosing thrown in for good measure, including the author and historian David Kahn, whose groundbreaking 1967 book on the history of cryptology, The Codebreakers, had caused the agency considerable panic. (Lieutenant General Marshall “Pat” Carter, the NSA director at the time, seriously put forth a number of other suggestions to stop publication of the book or discredit Kahn, including breaking into his home to steal the manuscript, planting disparaging reviews, or hiring him for a government job so that he would be liable to criminal penalties under the Espionage Act if he revealed classified information. Although NSA officials and the USIB spent “innumerable hours” considering those ideas, none in the end wer
e carried out, other than placing Kahn on the watch list.)19
As an NSA lawyer who later examined the program concluded, NSA staff responsible for distributing the Minaret reports were clearly aware that the entire operation was “disreputable if not outright illegal,” and took extraordinary pains to conceal what they were doing. The reports bore no trace of which agency they had come from, contained none of the usual sequential serial numbers assigned to NSA reporting, and were disguised to look like agent reports rather than communications intercepts.
None of the names of Americans on the Minaret watch list came out during the investigations, however, and NSA got off rather easy compared to CIA, which had to endure a litany of sordid revelations about foreign assassinations, coup plots, drug experiments on unwitting subjects, and other illegal and immoral activities. General Allen made an unprecedented appearance as an NSA director testifying in public before Congress, explaining the origins of Minaret and the difficulties of drawing an absolute line between foreign intelligence and domestic law enforcement, and revealing that he had personally shut down the program two years earlier. He was widely praised for his performance, which clearly had done the agency more good than harm. Allen, a man of brilliant intellect—he had a PhD in physics, had done nuclear weapons research at Los Alamos in the 1950s, and would go on after his term as NSA director to become Air Force chief of staff and then director of the Caltech Jet Propulsion Laboratory—was also, in Snider’s words, “a man of impeccable integrity,” and he steered NSA through a storm that might have sunk it.
A 1977 Justice Department review concluded that while both Shamrock and Minaret involved criminal violations of wiretap laws, a prosecution was unlikely to succeed and recommended against taking legal action against NSA officials.* The ultimate problem was the “ill-defined power” NSA and CIA had been granted from their inception: “If the intelligence agencies possessed too much discretionary authority with too little accountability, that would seem to be a 35-year failing of Presidents and the Congress rather than the agencies or their personnel,” the Justice Department concluded. Nonetheless, one of Allen’s decisive actions following Tordella’s retirement was to ensure that no NSA deputy director ever again wielded such untrammeled power.20