Deep State

by Marc Ambinder

  “Then we are in trouble,” responded Kissinger.

  From the outset, skeptics doubted President Ford’s rationale for appointing the Rockefeller Commission. It was either defensive posturing in a post-Watergate environment, designed to mitigate political damage (the investigation was limited to CIA operations on U.S. soil), or something more insidious was at play. In 1975, New York magazine reported that many observers believed that “Ford may have moved in order to fend off accusations of a more serious kind against the CIA—even more serious than domestic snooping in contravention of the agency’s charter.”6 The suspicion, of course, was assassination. At any rate, Congress didn’t waste time waiting for the executive branch to investigate itself. Senate leadership granted Frank Church a committee with full authority to investigate the whole of the intelligence community and its activities both foreign and domestic.

  The hearings were brutal for the intelligence community. When Church’s final report was released in April 1976, few had trouble predicting its conclusion: “Domestic intelligence activity has threatened and undermined the Constitutional rights of Americans to free speech, association and privacy. It has done so primarily because the Constitutional system for checking abuse of power has not been applied.”7

  The tragedy of the final report of the Church Committee is that it was right: the intelligence community was in dire need of reform and legal guidance. But the committee’s gleeful partisanship undermined an otherwise worthy goal. It was the first time the nation—indeed, the world—was given access to the machinery of tradecraft, and some evenhandedness was merited. Many people didn’t like what they saw, which was the point, but was to some degree an injustice.

  The government had been spying on citizens for quite some time. During World War II, all telegrams sent to and from the United States were screened by the Office of Censorship and its chief client, the FBI. The program collected intelligence on persons of interest and potential threats to national security. With the end of hostilities came the end of censorship, and consequently an immediate cessation of telegram cable intelligence. That wouldn’t do at all.

  On August 18, 1945, the Army Signal Security Agency (SSA) sent representatives to “make the necessary contacts with the heads of the Commercial Communications Companies in New York, secure their approval of the interception of all Governmental traffic entering the United States, leaving the United States, or transiting the United States, and make the necessary arrangements for this photographic intercept work.”8

  International Telephone and Telegraph (ITT) “very definitely and finally refused” to play any part in the obviously illegal program.9 Officials found a warmer reception with Western Union Telegraph Company, which agreed to cooperate under the condition that the attorney general sign off on the project. The SSA representatives then returned to ITT. In a meeting with a vice president, the SSA offered the veiled threat that “his company would not desire to be the only non-cooperative company on this project.”10 ITT relented, under the same proviso as Western Union. RCA was equally amenable, but again, only with authorization from the attorney general.11 As the CIA Center for the Study of Intelligence would later report:

  Two very evident fears existed in the minds of the heads of each of these communications companies. One was the fear of the illegality of the procedure according to present FCC regulations. In spite of the fact that favorable opinions have been received from the Judge Advocate General of the Navy and the Judge Advocate General of the Army, it was feared that these opinions would not hold in civil court and, as a consequence, the companies would not be protected. If a favorable opinion is handed down by the Attorney General, this fear will be completely allayed, and cooperation may be expected for the complete intercept coverage of this material. The second fear uppermost in the minds of these executives is the fear of the ACA which is the communications union. This union has reported on many occasions minor infractions of FCC regulations and it is feared that a major infraction, such as the proposed intercept coverage, if disclosed by the Union, might cause severe repercussions.12

  There is no evidence to suggest that either the president or the attorney general were ever briefed on the project, but not long after the SSA men visited the telegraph companies—and in spite of adamant internal resistance from each company’s lawyers—Operation Shamrock went active. (Decades later, Louis Tordella, deputy director of the National Security Agency, would admit that “he did not know if any subsequent president or attorney general had ever been briefed on it.”)13

  One problem remained: physically transferring thousands of cables in secret. William Sidney Sparks, the traffic manager for RCA and a lieutenant colonel in the U.S. Army Signal Corps Reserve, worked closely with the SSA to find a solution. He swatted down ill-conceived schemes by his government counterparts on the grounds that “everybody and his brother would know just exactly what we were doing and why.”14

  According to James Bamford, author of The Puzzle Palace, “He told the officers that probably the most secure and efficient way to handle the problem would be to turn over to the agency all traffic entering, leaving, or transiting the company.” The SSA couldn’t believe their luck. Sparks initially stipulated that the SSA (soon renamed the Army Security Agency, or ASA) would receive only “header” information stating the origin and destination of each telegram. That policy soon fell by the wayside, and the agency began collecting hard copies of cables in their entirety.

  For his part, Sidney Sparks understood that his actions were illegal, but also that the United States was staring down the barrel of a new kind of war. “I knew in my own mind that the Cold War was heating up at the time,” he said, adding, “I was under no illusion at all that any responsible Government has to monitor, to some degree, the traffic of the other [foreign] Government agencies as far as it can get hold of them.”15 His superiors, as well as executives at Western Union and ITT, were equally cognizant of the criminal activities to which they were party and would remain terminally paranoid. General Dwight D. Eisenhower, then the Army chief of staff, sought to allay their fears, as would Secretary of Defense James Forrestal. Neither man, however, would prove particularly persuasive on this point. Shamrock was a military program; of course the defense secretary approved of industry participation. Meanwhile, neither the attorney general nor the president of the United States would ever directly convey any legal cover—or even any direct knowledge—of the program. Communications executives wondered, indeed, if Operation Shamrock ever reached their desks.

  The ASA would eventually be absorbed by the Armed Forces Security Agency (AFSA), which would become the National Security Agency. The NSA thus inherited Shamrock and maintained tens of thousands of files on U.S. citizens.∗ As a practical matter, the AFSA and its successor acted as an information broker to the FBI and CIA. The intelligence agencies were even internally referred to as “customers.” Initially, each agency (and a number of others) set up desks at Arlington Hall in Virginia, the nerve center of the AFSA. (The site was originally Arlington Hall Junior College for Women, a nonprofit girls’ school seized by the Army Signal Intelligence Service in 1942 under the War Powers Act.16 The fledgling NSA would later install itself in abandoned Army barracks at Fort Meade, Maryland.) These liaisons from every segment of the intelligence community rifled daily through the nation’s cable traffic, forwarding useful data to their respective headquarters.17 During the years that followed, each agency would submit watch lists of “persons of interest” for the NSA to be on the lookout for. In addition, NSA agents combed data in search of trigger words.18

  Although Shamrock ostensibly searched only traffic originating and terminating on foreign soil, the project expanded to monitor perceived radicals susceptible to foreign influence.19 For most Americans, this probably crossed the line between what was possibly illegal but benign and an absolute outrage.

  The intelligence community, ever thirsty for more information and already operating outside of the law, would push the N
SA as far as the agency would permit. Watch lists eventually became blanket requests. As Frank Raven, a former NSA official, later observed, “When J. Edgar Hoover gives you a requirement for complete surveillance of all Quakers in the United States, and when Richard Nixon is a Quaker and he’s president of the United States, it gets pretty funny.” (At the time, Hoover suspected that the Quakers were sending food to Southeast Asia. As for the FBI’s involvement in illegalities, Hoover had requested and received permission for such activities from the attorney general in 1941.20 This would suffice as legal cover for the Bureau.)

  The NSA would also, in turn, act as a client of the FBI and the CIA. While the NSA had negotiated deals with the communications giants in New York, the Bureau dominated the Washington cable circuit and provided daily intelligence to the NSA. Furthermore, the Bureau, which had long mastered the art of infiltrating offices and installing listening devices, provided these services to the NSA, which was at its essence a stationary global listening post lacking an agile force on the ground. These so-called black bag missions saved the NSA time, money, and manpower. Deciphering encrypted calls and cables from foreign embassies in Washington might take the NSA months, if not years. The FBI’s practiced special agents could plant a bug overnight.21

  Meanwhile, the CIA performed similar operations on foreign embassies overseas, filching ciphers and codebooks. And when the NSA found itself in need of office space in New York City to process the massive collection of cable traffic, it approached the CIA for a “safe house.” For seven years, the Company obliged the NSA’s real estate needs.22

  Operation Shamrock would run for thirty years, and at its height would collect 150,000 messages a month, illegally.23

  On March 4, 1977, Robert Keuch, deputy assistant attorney general in the Carter administration, received a comprehensive summary of the illegalities of the intelligence community as gathered by Congress, and the names of those who should be prosecuted as a result.

  The problem, according to the memorandum, was that domestic surveillance and intelligence collection was an evolutionary process and began with presidential authorization. Presidential national security power, it explained, “did not spring full grown from one source, such as the Constitution; rather, it started with an idea and grew steadily over the better part of four decades.”24

  Interestingly, as far back as the 1940s, J. Edgar Hoover sought secondary approval from the attorney general, writing that it would be “highly desirable that some definite decisions be made by the Department of Justice relative to the legality of the [wiretapping activity].” (Despite the aggressive campaign against Hoover since his death, even this “prosecutive summary” specifically notes that such a request for Justice Department guidance was “not really unusual in light of Director Hoover’s strong dislike for wiretapping.”)25

  By the time the NSA had been created, surveillance authority of the executive branch was largely unrestricted. Indeed, with regard to communications intelligence, National Security Council Intelligence Directive 9 specifically stated:

  The special nature of Communications Intelligence activities requires that they be treated in all respects as being outside the framework of other or general intelligence activities. Orders, directives, policies, or recommendations of any authority of the Executive Branch relating to the collection, production, security, handling, dissemination, or utilization of intelligence, and/or classified material, shall not be applicable to Communications Intelligence activities, unless specifically so stated and issued by competent departmental or agency authority represented on the Board . . . . Other National Security Council Intelligence Directives to the Director of Central Intelligence and related implementing directives issued by the Director of Central Intelligence shall be construed as non-applicable to Communications Intelligence activities . . . unless the National Security Council has made its directive specifically applicable to Communications Intelligence.26

  Over the decades, the Department of Justice “had repeatedly sought (and invited) legislation from Congress which would both permit wiretapping and allow the use of the results or fruits of such surveillance at trial, but Congress, however, declined to act.” In 1968, Congress would enact wiretap legislation—Title III of the Omnibus Crime Control and Safe Streets Act. Notably, however, a section of the act “expressly exempted the President’s power from the coverage of the provisions of Title III.”27

  When you wonder why so many Americans doubt the trustworthiness of intelligence agencies, understand that from their respective inceptions, the FBI, the NSA, and telecoms would in fact violate aspects of the law with respect to wiretapping. That said, prosecution was difficult, if not impossible, because (as stated in the memorandum):

  1. Prior Presidents and Attorneys General had notice of and, in at least once case, appeared to approve the operation;

  2. Two Secretaries of Defense had tried to give the companies immunity;

  3. Clause one of [section] 605 permits companies to disclose information “upon demand of lawful authority”;

  4. There was no divulgence outside the Executive Branch, so there was no divulgence within the meaning of [section] 605;

  5. A use which benefits the Government is not the type of “use” contemplated by the statute;

  6. It is not illegal to “ask” a company to give out copies of cables. If the company complies, it may be violating the statute but the recipient would not; and

  7. The putative defendants acted in good faith, and they lacked the necessary intent to prove a violation of the law.

  Further, “as it is clear from a review of an evolution of the President’s power from its inception, the true scope of the President’s power (with which the Bureau and the Agency were familiar) was unknown.”

  Congress did not escape scrutiny; by funding the initiatives, it had at least some notice of their activities, although very few members were equipped to understand them—a systemic imbalance that intelligence agencies continue to exploit. As for domestic surveillance by the CIA, again, Congress was not blameless. Again, according to the 1977 memorandum:

  In July of 1973, William Colby testified before the Senate Armed Services Committee on his nomination to become DCI. In response to a question specifically addressed to whether CIA was then engaged in assisting law enforcement agencies in addition to the FBI, Colby replied in the affirmative . . . . Since there was little doubt that at least some of CIA’s information was governed by electronic surveillance, the Agency regards the lack of congressional objection as tacit approval of such dissemination.

  Furthermore, National Security Council Intelligence Directive 5 delegated the CIA national security responsibilities abroad, and Title 50 U.S. Code Section 403 grants the Agency authority to perform “such additional services of common concern as the National Security Council determines can be more effectively accomplished centrally.”

  Perhaps the most important finding of the “prosecutive summary” is that in not a single instance was electronic intelligence used “for personal or partisan political purposes. The participants in every questionable operation, however oblivious or unmindful, appear to have acted under at least some colorable semblance of authority in what they conscientiously deemed to be the best interests of the United States.”

  The memorandum closes, “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.”

  No one went to jail. In many ways, the hearings of 1975—the “Year of Intelligence,” as Director Colby dubbed it—in fact emboldened the executive by infusing partisanship into the issue. Congressional oversight, already hapless at best, would further allow a certain permissiveness to intelligence activities depending on which party held the gavel. President Ford would fire Colby as director of central intelligence. In his autobiography, Colby wrote, “I believe I was fired because of the way I went about d
ealing with the C.I.A.’s crisis. My approach, pragmatically and philosophically, was in conflict with that of the President and his principal advisors.” Colby’s approach was sunlight—to cooperate with investigations “and try to educate the Congress, press, and public, as well as I could, about American intelligence, its importance, its successes and its failings.”28

  That simply would not do. The message was received, and thirty years later a new set of old problems would confront the intelligence community, the White House, and Congress.

  Covert operations have inspired more acrimony between the legislative and executive branches than almost any other issue. Since the Year of Intelligence, Congress and the White House have furiously debated what information, exactly, Congress has the right to, and under what circumstances. It is an intragovernmental mirror of the wider secrecy debate.

  In the earliest days of the Cold War, Congress showed little interest in the operational details of U.S. intelligence. There simply wasn’t an appetite to know the nation’s dirty secrets. As Leverett Saltonstall, senator from Massachusetts, explained in 1956, “It is not a question of reluctance on the part of CIA officials to speak to us. Instead, it is a question of our reluctance, if you will, to seek information and knowledge on subjects which I personally, as a member of Congress and as a citizen, would rather not have.”29

  Congressional attitudes toward executive power and the national security bureaucracy hardened in the wake of Watergate and revelations of controversial CIA actions at home and abroad. Standing over the festering remains of the Nixon administration, Congress had a gladiator’s temperament. It asserted itself on the issue of covert actions in 1974 through the Hughes-Ryan Act, an amendment to the Foreign Assistance Act of 1961. Passed in 1974, Hughes-Ryan prevented the CIA from spending funds on covert activities unrelated to intelligence gathering unless the president “finds that each such operation is important to the national security of the United States and reports, in a timely fashion, a description and scope of such operation” to Congress.30 The definition of “timely fashion” would be a matter of debate for decades. But at the time there was an understanding that briefings need not necessarily take place before a covert action had begun.31