The Post-9/11 Surveillance Responses

The Patriot Act, among other measures, expanded the FBI’s ability to collect bank, credit card, and communication billing records and to monitor phone calls and emails. Some of the act’s most controversial surveillance provisions would automatically expire in four years unless reauthorized by Congress, suggesting that legislators would play a direct role in observing and, if need be, checking new surveillance activities.^{Footnote 2} However, as the FBI used its new investigative powers in highly aggressive and expansive ways, the Bush administration took steps to stymie oversight of the bureau’s post-9/11 practices.

After the enactment of the Patriot Act, Justice Department officials repeatedly rejected requests from the House and Senate Judiciary committees for direct access to records documenting how this expanded authority had been used. Even so, Congress made many of these powers permanent in 2006. During the debate over reauthorization, members of Congress differed over whether these broad powers had been or would be abused. Of special concern were national security letters (NSLs)—subpoena-like orders—that allowed agents to seize financial and communication records without judicial approval. To allay doubts, Congress adopted a provision directing the Justice Department’s inspector general to conduct an internal review of “the effectiveness and use, including any improper or illegal use, of national security letters issued by the Department of Justice.” Under this provision, Inspector General Glenn Fine was directed to evaluate the FBI’s use of NSLs between 2003 and 2005 and to submit his conclusions to the appropriate congressional oversight committees.^{Footnote 3}

Fine released two reports based on his office’s investigation of the FBI’s use of NSLs, one in March 2007 and the second in January 2010. His reports disclosed that between 2003 and 2005, the FBI had issued 143,074 NSLs. This total likely underestimated the FBI’s actual usage, first, because it was based on a “judgmental” sample of selected files from only four of the FBI’s fifty-six field offices and, second, because investigators had uncovered “widespread” inaccuracies in the FBI’s database. Indeed, Fine pointed out that a NSL “may contain more than one request for information,” making it exceedingly difficult to determine how much data had been collected and how many individuals had been targeted in total. The inspector general further stated that FBI officials had underreported their agency’s uses of NSLs to Congress, that they had accessed more than 3,500 telephone accounts without following legal procedures, that at times NSLs had been used illegally in place of warrants to obtain sensitive records, that many of the FBI’s “exigent letters” (ostensibly issued in emergency situations) involved nonemergencies, and that 125 FBI reports inaccurately identified the target as a non-U.S. person when the “appropriate memoranda on the investigation indicated that the subject was a U.S. person or a presumed U.S. person.”^{Footnote 4}

FBI records released in response to a series of Freedom of Information Act requests, moreover, seemingly confirmed that FBI investigations had spilled over to monitor political activists and militant Muslims. These targets included antiwar protestors (notably the Catholic Worker movement, the Thomas Merton Center, and the Quakers), environmentalists (Greenpeace), animal rights activists (PETA), Occupy Wall Street demonstrators, prominent Arab Muslims and mosques in Michigan and California, black separatists, a student chapter of the National Lawyers Guild, and even, bizarrely, people carrying almanacs. Apparently, FBI officials worried that almanacs could be used to “assist with target selection and planning” of large-scale, al-Qaeda–style terrorist attacks. Nonetheless, based on his office’s review of these investigations, Fine subsequently concluded that the “evidence did not indicate that the FBI targeted any of these groups on the basis of their First Amendment activities.”^{Footnote 5}

Despite the FBI’s intensive and relatively indiscriminate monitoring of Muslim activists and mosques, the agency failed to anticipate two planned terrorist plots: the successful Boston Marathon bombing and the aborted (owing to the incompetence of the would-be attacker) New York Times Square bombing. Furthermore, many of the publicized plots uncovered by the FBI had only been attempted in the first place because authorities or their recruited informers had provided crucial assistance to the terrorist suspects. In many cases, plotters lacked the know-how to develop plans on their own or relied on material provided by their FBI contact. A survey of 508 FBI terrorist investigations, conducted by a researcher with the assistance of the University of California–Berkeley Graduate School of Journalism, found that “243 had been targeted through an FBI informant, 158 [had] been caught in a FBI terrorist sting, and 49 had encountered an agent provocateur.”^{Footnote 6}

The Terrorist Surveillance Program and the Snowden Leaks

Troubling questions about the FBI’s counterterrorism operations are compounded by revelations about the actions of the NSA. When lobbying Congress to develop and pass the Patriot Act, the Bush administration did not propose major changes to warrant requirements for the capture of private conversations in counterterrorism operations. The law governing such eavesdropping—the Foreign Intelligence Surveillance Act (FISA) of 1978—remained largely intact. Instead, having learned that the NSA could do “nothing more within existing [legal] authorities” to uncover planned terrorist operations, President Bush asked NSA Director Michael Hayden what the agency could do “with more authority.” Hayden responded by outlining what became the Terrorist Surveillance Program, which Bush secretly authorized on October 4, 2001.^{Footnote 7}

Under this secret program, the NSA intercepted and recorded the communications (telephone calls, e-mails, and faxes) “into and out” of the United States of unspecified individuals about whom there would be “a reasonable basis to conclude that one party to the communication is a member of al Qaeda” or somehow connected to the terrorist organization. The president’s order further required NSA officials to “maintain the secrecy of the activities carried out under the program.” And, rather than targeting specific individuals, NSA analysts relied on data mining to identify suspected terrorists and terrorist sympathizers. The selection criteria were, as a consequence, inherently speculative (e.g., communications expressing an interest in flight schools or sent to specific foreign regions such as Pakistan). NSA officials, in addition, did not rely exclusively on the agency’s interception capabilities; private telecommunication companies, including AT&T, Verizon, and BellSouth, provided the NSA with access to their transmissions.^{Footnote 8}

This secret surveillance program was publicly compromised in December 2005 through a leak to the New York Times. In a dramatic front-page article, the Times reported that President Bush had secretly ordered the NSA to intercept the international communications of suspected terrorists without complying with the warrant requirements of FISA. Attempting to downplay the controversy over these revelations, Bush and senior intelligence officials defended this program as having been “a vital link in our war against terrorism” and as having narrowly targeted “people with known links to al Qaeda and related terrorist organizations.” In response to growing criticism, however, the Bush administration eventually reached a compromise with congressional leaders—first with the Protect America Act of 2007 and then with the FISA Amendments Act of 2008, which placed some restrictions on the NSA’s domestic spying. Specifically, in return for granting immunity from prosecution to the private telecommunication companies that had assisted the NSA with the Terrorist Surveillance Program, the administration agreed that a special intelligence court would now need to give prior approval for the NSA to intercept the communications of U.S. persons. The new legislation, nonetheless, allowed the NSA to intercept “inside the United States any communication of non-U.S. persons believed to be located outside the United States, provided a significant purpose of the acquisition pertained to foreign intelligence.”^{Footnote 9}

When Congress amended FISA in 2007 and 2008, it required the inspectors general of the relevant federal intelligence agencies to “conduct a comprehensive review” of the Terrorist Surveillance Program. The resulting unclassified report of 2009, combined with other revelations about the questionable counterterrorism benefits of NSA surveillance, challenged the Bush administration’s positive characterization of the program’s value. Prior to the release of the report and based on interviews with FBI officials who had been the recipients of NSA-supplied information, the New York Times reported in 2006 that the NSA’s “steady release of information” to the FBI had led to “dead ends or innocent Americans” and had often amounted to “pointless intrusions on Americans’ privacy.” The 2009 report by the inspectors general echoed this characterization, disclosing that most of the intelligence agents interviewed by their staff “had difficulty citing specific instances” or “precise contributions” that had led to the uncovering of terrorist plans stemming from the intercepted communications. Indeed, the agents who were interviewed during the compilation of the report variously described the Terrorist Surveillance Program as merely “one tool among many,” as offering “value in some counterterrorism investigations,” or as “generally play[ing] a limited role in the FBI’s overall counterterrorism efforts.”^{Footnote 10}

Denied independent access to NSA records, members of Congress, the media, and the public could neither evaluate how the program had operated nor determine the validity of first the Bush and then the Obama administration’s benign characterization and claims of effectiveness. In June 2013, however, this monopoly over information about NSA operations was shattered when NSA contract employee Edward Snowden leaked to the media thousands of pages of classified NSA records. Snowden’s dramatic disclosures rebutted the claims of both the Bush and Obama administrations that NSA operations had been tightly circumscribed. Snowden’s leaks instead revealed that the NSA had massively monitored individuals with no terrorist plans, had indiscriminately compromised privacy rights, and had collected information about the personal and political conduct of U.S. citizens and foreigners (including prominent foreign leaders, notably German Chancellor Angela Merkel and Brazilian President Dilma Rousseff).

The Snowden revelations offered confirmation that the NSA had mined the communication records of millions of Americans, had intercepted personal content from online communications (including baby pictures, risqué photos from webcam chats, medical records, and conversations about sexual liaisons), and had tapped the main communication links of Internet companies Yahoo! and Google. Furthermore, the pilfered records confirmed that nine out of ten NSA intercepts involved individuals who were not direct targets of NSA surveillance, that NSA analysts at times shared among themselves intercepted sexually explicit material, that the communications of a U.S. law firm representing Indonesia in a trade case had been monitored, that software had been implanted in nearly one hundred thousand computers around the world, and that various online games had been tracked. In one known case, the NSA had acquired information about the pornographic interests of a Muslim critic of U.S. policy, which was then exploited to demonstrate his hypocrisy.^{Footnote 11}

In reaction to the Snowden leaks, President Barack Obama, NSA officials, and the chairs of the House and Senate Intelligence committees initially defended the NSA’s surveillance programs—insisting that they offered essential protection to the country from foreign threats and that they had uncovered numerous terrorist plots. In addition, they condemned Snowden’s actions as having caused significant harm to U.S. counterterrorism operations. This sweeping defense and alarmist claims were soon abandoned, however. In June 2014, the newly appointed NSA director, Michael Rogers, conceded that the “sky was not falling” and that the Snowden revelations had not permanently undermined foreign-intelligence gathering. He also did not dispute one of the conclusions from an independent audit by the Privacy and Civil Liberties Oversight Board—namely, that the Terrorist Surveillance Program had only proved to be of “minimal” benefit to the government’s counterterrorism operations.^{Footnote 12}

The question arises then: Why, given these revelations of FBI and NSA abuses, did Congress not act to curb FBI and NSA surveillance authority? Research into previously secret records, which confirm past abuses of surveillance powers during World War II and the Cold War, challenges this equanimity about the wisdom of deferring to executive branch actions. The historical record provides grounds for even more serious concerns that granting such latitude to executive branch officials, primarily motivated by the need to avert any possible internal security threat, will ensure illegal, obtrusive, and politically biased practices. In what follows, I detail some of the worst government abuses from these earlier periods to demonstrate that, left unmonitored, intelligence officials often conduct investigations driven by dubious political agendas.

The FBI, World War II, and Cold War Surveillance

Between the mid-1930s and the mid-1970s, the FBI conducted highly secretive and mostly undocumented investigations against Americans—many of whom represented no genuine threat to national security. Presidents and other top-level officials would typically give verbal authorizations to these investigations, if they gave approval at all, in order to avoid the development of an incriminating paper trail. Not only did these informal orders prevent Congress and the courts from learning about these investigations, but they also gave FBI agents tremendous leeway to use intrusive, unconstitutional methods and to target a wide array of innocent individuals, including prominent political figures.

In August 1936, President Franklin Roosevelt met with FBI Director J. Edgar Hoover to convey concern about (in Hoover’s phrasing) the “question of subversive activities in the United States, particularly Fascism and Communism.” The FBI director responded to the president’s query by recounting what he characterized as communist plans to “get control” of the International Longshoremen’s and Warehousemen’s Union, the United Mine Workers, and the Newspaper Guild. With union control, communists could then “paralyze the country” by halting shipping, industrial output, and newspaper publications. (Again in Hoover’s phrasing) Roosevelt expressed his interest in “obtaining a broader picture of the general movement and its activities as may affect the economic and political life of the country as a whole.” The president specifically cited the role of Germany and the Soviet Union, claiming that fascism and communism were movements “international in scope and that Communism particularly was directed from Moscow.” Roosevelt thereupon orally authorized the FBI director to investigate this matter but rejected a suggestion that his decision be recorded in writing, insisting instead that it be “handled confidentially.” Indeed, the only records of this presidential decision were two memoranda maintained in the FBI director’s secret office file.^{Footnote 13}

In his oral directive, Roosevelt had not set the parameters that would govern the scope of FBI surveillance operations. In effect, he had simply authorized the FBI to move beyond its law enforcement responsibilities to conduct “intelligence” investigations (i.e., investigations that were not predicated on a “probable cause” violation of a federal statute standard). The president’s discretionary authorization, combined with his insistence on secrecy and his decision to exclude the attorney general, emboldened FBI officials to initiate investigations that went well beyond ascertaining possible German or Soviet influence over the American fascist and communist movements. Indeed, on the basis of Hoover’s briefing about the president’s decision (and before the FBI director even briefed the attorney general about Roosevelt’s order), FBI Assistant Director Edward Tamm outlined to senior FBI officials the “general classifications” that were to govern future FBI “intelligence” investigations. The proposed targets were: “Maritime industry, Governmental affairs, steel industry, general strike activities, Armed Forces, educational institutions, general activities—Communist and affiliated organizations, Fascisti, anti-Fascisti movements, and adherents of organized labor organizations.”^{Footnote 14}

The president might have wanted the FBI to ascertain possible German and Soviet control of the American communist and foreign movements, and thereby forestall planned espionage and sabotage operations. However, the resultant FBI investigations were never confined to monitoring individuals who might engage in espionage and sabotage. Instead, FBI agents began to monitor those whom FBI officials feared could influence the popular culture. For example, those targeted included: Washington Times-Herald gossip columnist Inga Arvad;^{Footnote 15} Washington, D.C.–based detective and influence peddler Henry Grunewald;^{Footnote 16} anti–New Deal activists Lillian Moorehead, John G. O’Brien, and Allen W. Guillon;^{Footnote 17} Hollywood directors, producers, writers, and actors;^{Footnote 18} and German émigré writers and playwrights.^{Footnote 19} Additionally, the FBI went after individuals who were listed in a special index because they were “in a position to influence others against the national interest or are likely to furnish material aid to subversive elements due to their subversive associations and ideology . . . [such as] 1. professors, teachers, and educators, 2. labor union organizers or leaders, 3. writers, lecturers, entertainers and others in the mass media field, 4. lawyers, doctors, and scientists, 5. other potentially influential persons on a local and national level, 6. individuals who could potentially furnish material financial aid.”^{Footnote 20}

Surveillance operations, moreover, were not confined to individuals whom FBI officials deemed suspect because of their “subversive associations and ideology.” Prominent Americans were also kept under watch. For example, among those whose political and personal activities were monitored were First Lady Eleanor Roosevelt,^{Footnote 21} Illinois governor and Democratic presidential nominee Adlai Stevenson,^{Footnote 22} John Kennedy throughout his entire public life,^{Footnote 23} President Dwight Eisenhower,^{Footnote 24} syndicated columnist Joseph Alsop,^{Footnote 25} former White House aide and prominent Washington, D.C., attorney Thomas Corcoran,^{Footnote 26} U.S. Supreme Court Justices and their clerks,^{Footnote 27} prominent writers (Ernest Hemingway, Archibald MacLeish, Lillian Hellman, Dashiell Hammett, Sherwood Anderson, Pearl Buck, Truman Capote, Allen Ginsburg, Arthur Miller, John Steinbeck, and Upton Sinclair),^{Footnote 28} and prominent sociologists (Talcott Parsons, C. Wright Mills, Helen Lynd, Edwin Sutherland, and as well the American Sociological Association).^{Footnote 29}

Early Warrantless Wiretaps

Senior FBI officials did not rely solely on the investigative skills of FBI agents (and their recruited informers) to monitor so-called subversives. They authorized FBI agents to employ intrusive, if illegal, investigative techniques: wiretaps, bugs, break-ins, and mail opening. In some cases, FBI Director Hoover sought and obtained the president’s or the attorney general’s secret authorization for the use of these techniques; in other cases he acted on his own. In either case, Hoover’s awareness of the potentially damaging ramifications should the FBI’s use of such illegal techniques become known led him to institute a series of sophisticated records procedures to preclude the possible discovery of these actions (and his own authorization).

The FBI’s uses of wiretaps posed a particular legal problem. In 1934, Congress passed the Communications Act that banned wiretapping. Then, in 1937 and 1939, the Supreme Court ruled in Nardone v. U.S. that this ban applied to federal agents and that any indictment based on an illegal wiretap would have to be dismissed. The Court’s rulings, however, had little immediate impact. Deeply troubled by German and Soviet involvement in Europe in “the organization of so-called ‘fifth columns’ and in the preparation for sabotage, as well as actual sabotage,” President Roosevelt in May 1940 secretly authorized FBI wiretapping during “national defense” investigations—although on the condition that each proposed wiretap would be subject to the prior review and approval of the attorney general.^{Footnote 30}

Concerned about the political ramifications of secret surveillance being publically exposed, Attorney General Robert Jackson instructed the FBI director to keep minimal files on any investigative approvals. Specifically, his authorization was to be recorded in “a memorandum book” in the FBI director’s office that would list “the times, places, and cases in which this procedure is utilized”—thereby minimizing the discovery of this secret policy.^{Footnote 31}

Despite these precautions, the FBI’s use of wiretaps during an investigation of Justice Department employee Judith Coplon became known during her trial in 1949. A series of disclosures during the trial both embarrassed FBI officials and raised the possibility that Congress might launch an investigation into FBI practices. Indeed, the trial had raised questions about the extent of FBI wiretapping, about the FBI’s wiretapping of Coplon before and after her arrest (including the interception of her conversations with her attorney), about the actions of an FBI agent who gave false court testimony on FBI wiretapping, and about an FBI decision to destroy the Coplon wiretap logs in view of the “imminence” of her trial. To avert the reoccurrence of this public relations problem and to avoid any future demand for an investigation of the FBI’s operations, FBI Director Hoover ordered two fundamental changes.

First, in June 1949, he instituted a new records procedure, code-named June Mail. The heads of FBI field offices (Special Agents in Charge, or SACs) were to caption JUNE any record sent to FBI headquarters that reported information that had been obtained from “highly confidential sources” (wiretaps, bugs, break-ins, and mail opening) or from particularly sensitive sources, such as “governors [and] secretaries to high officials who may be discussing such officials and their attitudes.” On receipt at FBI headquarter, JUNE-captioned records were not to be included in the subject’s main case file but instead were to be routed to the Special File Room where they would be maintained “under lock and key” with access limited to a “need to know” basis.

Second, in December 1949, the FBI director spelled out the conditions that henceforth were to govern the actions of FBI agents who could be called to testify during a court trial. Any agent, supervisor, or SAC, Hoover ordered, should not “engage in any searches, physical surveillances or any types of work which might make him a competent witness in the event . . . it should be decided to prosecute. Thus, the Agent who might in these instances be competent witnesses would have no specific testifiable knowledge of the existence of a technical surveillance [wiretap] in that particular case.”^{Footnote 32}

The Expanding Scope of Warrantless Investigations

President Roosevelt’s May 1940 directive for warrantless surveillance applied only to wiretaps in national security cases. But later administrations decided to give the FBI broad discretion in conducting intrusive investigations. Tacit approval of black bag operations encouraged the FBI not only to use increasingly aggressive methods but also to institute multiple, and sometimes complex, ways to conceal their activities.

Starting in the 1940s, FBI officials on their own authorized agents to employ bugs during “intelligence” investigations and, until 1952, did not inform or seek the attorney general’s approval for such installations. Then, when briefed on the FBI’s uses of bugs in 1952, Attorney General J. Howard McGrath responded by affirming that he “could not authorize the installation of microphones involving a trespass under existing law” (emphasis in the original).

In 1954, however, exploiting the Supreme Court’s critical ruling concerning a bug employed in a California case, Irvine v. California, FBI officials urged Attorney General Herbert Brownell to adopt the same policy for FBI bugging installations as that governing wiretaps, namely, that each installation would have to obtain the prior review and approval of the attorney general. Brownell rejected this recommendation and, instead, issued a vaguely worded secret order approving such uses but not explicitly requiring the attorney general’s advance approval—effectively giving FBI officials a blank check. As a result, most FBI microphone installations (until Attorney General Nicholas Katzenbach’s 1965 order) were neither known nor approved by the attorney general. In effect, for a time FBI officials operated on their own. A May 1966 memorandum by FBI Assistant Director James Gale documents the extent of this independence. In this memorandum, Gale reported that of the 738 bugs installed by FBI agents between 1960 and 1966, the Justice Department had been “notified” about only 158. Gale then explained why FBI officials had bypassed their ostensible overseers: “We know that the [Justice] Department completely lacks security, leaks confidential information to the press, and has demonstrated a propensity for going forward and advising the courts whenever they have knowledge of the existence of microphone coverage regardless of whether or not that coverage had any bearing on the case under consideration.”^{Footnote 33}

In most cases when installing bugs, FBI agents had to break into the target’s office or residence. Moreover, FBI agents had also resorted to break-ins for the additional purpose of photographing the correspondence, financial documents, and other records of a targeted individual or organization. Having concluded that break-ins were “an invaluable technique in combating subversive activities of a clandestine nature aimed at undermining our nation,” FBI officials willingly authorized such actions. They nonetheless privately conceded that break-ins were “clearly illegal” and that they could not secure the attorney general’s authorization.

Recognizing the potential adverse political costs in the event that an FBI agent’s break-in could be discovered, FBI Director Hoover instituted a sophisticated records procedure. Before a break-in could be conducted, the SAC had to seek the approval of senior FBI officials at headquarters in each proposed case and, when seeking such approval, had to “completely justify the need for the use of this technique and at the same time assure that it could be safely used without any danger of embarrassment to the Bureau.” Normally, field office communications to FBI headquarters would have been serialized and indexed in the targeted individual’s or organization’s file in the FBI’s central records system—thereby both documenting the creation of this record and making the record vulnerable to a court-ordered discovery motion or congressional subpoena. To finesse this potential problem, SAC break-in requests were to be captioned Do Not File and, on receipt at FBI headquarters, were to be routed either to Hoover’s office or the office of a designated FBI assistant director. The SAC would also maintain an “informal memorandum” of the break-in request and FBI headquarters’ authorization in his personal office safe “until the next inspection by Bureau Inspectors, at which time [the memorandum was to be] destroyed.” This crafty procedure would allow FBI officials to affirm, truthfully, should Congress or the courts inquire about FBI practices, that a search of the FBI’s “central records system” uncovered no evidence of illegal conduct—insofar as any break-in record would not have been serialized or indexed in that system. Furthermore, because the Do Not File record had been maintained in an FBI assistant director’s office file, it could be safely purged (every six months pursuant to another Hoover order).^{Footnote 34}

The Do Not File system was not a one-time initiative. In fact, it was a spin-off of another record destruction system that Hoover instituted in 1940 and subsequently refined in 1943, 1944, and 1949. Under this procedure, especially sensitive communications among senior FBI officials were to be typed first on blue and then pink paper (thus distinguishing them from other FBI records typed on white letterhead paper) and were to bear the following caption at the bottom: “This Memorandum Is for Administrative Purposes to Be Destroyed after Action Is Taken and Not Sent to the Files.” Then, when a copy of a pink memorandum surfaced during the Coplon trial, Hoover abandoned the use of pink paper and instead required that such communications in the future were to be submitted as “informal” memoranda, typed on white nonletterhead stationery without any “to be destroyed” caption. The FBI director’s eventual concern over the failure of senior FBI officials to destroy these records led him in March 1953 to order the regular destruction of the contents of office files—every six months in the case of FBI assistant directors and every ninety days for FBI supervisors.^{Footnote 35}

FBI officials also devised yet another special record system, in this case to safeguard their policy of having FBI agents collect derogatory and background information about members of Congress. From the 1940s onward, FBI officials had compiled records about the actions of members of Congress in an ad hoc informal manner. They decided to refine and systematize this collection system in the late 1950s. SACs were accordingly instructed to review their office’s files and forward to FBI headquarters “by routing slip not letter” all information that had been obtained on members of Congress. In this way, these communications would not be serialized or indexed in the FBI’s central records system. This submitted information was then supplemented by information about the member of Congress that had been obtained at FBI headquarters’ files to constitute a “summary memorandum” on that member of Congress.

The contents of these summary memoranda were to be “updated on a continuing basis” and included background information on the individual, the member’s attitude toward law enforcement, and any “allegations of criminal or corrupt practices, subversive activities, and immoral conduct.” When in rare instances SACs submitted the requested information “in formal communications,” they were “orally instructed to remove from the files all copies of the material and the communications were treated here [FBI headquarters] as informal ones with no record being created in Bureau files.” The resultant summary memorandum, moreover, was maintained separate from the FBI’s Central Records System in the FBI’s Administrative Review Unit.^{Footnote 36} FBI officials as a result could truthfully deny that the FBI maintained “files or dossiers” on members of Congress in its “central records system.”

Politically Motivated Investigations

We are aware, although in an admittedly limited way, how FBI officials exploited the acquired derogatory personal and political information. Testifying before the so-called Pike Committee in 1975, a former FBI agent recounted a conversation that he and other agents attending a training school at FBI headquarters had with FBI Assistant Director Cartha DeLoach. One of the attending agents, he reported, had asked DeLoach, “What do we use all this memorandum stuff we put in about things we see—what do you do with it?” He then quoted DeLoach as replying: “The other night we picked up a situation where this senator was seen drunk, in a hit-and-run accident, and some good looking broad was with him.” Continuing, the agent contended that DeLoach “and this is an exact quote—he said, ‘By noon of the next day the good senator was aware that we had the information and we never had any trouble with him on appropriations since.’”^{Footnote 37}

This example of how acquired information was used to advance the FBI director’s political and bureaucratic agenda was not an atypical practice. Dating from the 1940s, FBI officials purposefully exploited acquired information (which had no prosecutive value either because it was illegally obtained or because it did not document a violation of a federal statute). They did so through a series of orchestrated leaks to members of Congress and the media who shared Hoover’s political agenda but on the strict condition that the recipient not disclose the FBI as the source. And, through another procedure designed to enable the FBI director to affirm the “confidentiality” of FBI files and to deny that the FBI had been the source of a recipient’s information, the recipient was never given access to or copies of FBI files. Rather, the derogatory material was conveyed through “blind memoranda” (typed on plain white non-letterhead stationary without identifying the sender or recipient). The recipients, who were carefully selected because they shared the FBI director’s political ideology, constituted the major participants in the militantly anticommunist politics of the so-called McCarthy era: notably, the House Committee on Un-American Activities,^{Footnote 38} the Senate internal Security Subcommittee,^{Footnote 39} Senator Pat McCarran,^{Footnote 40} Senator Karl Mundt,^{Footnote 41} Senator Joseph McCarthy,^{Footnote 42} Congressman/Senator Richard Nixon,^{Footnote 43} state governors under a code-named Responsibilities Program,^{Footnote 44} and a host of conservative reporters and columnists whom FBI officials at first assisted informally and then, beginning in the mid-1950s, more formally under a code-named Mass Media Program. The recipients included Don Whitehead, Walter Trohan, Courtney Ryley Cooper, Fulton Oursler, and Frederick Woltman.^{Footnote 45}

The NSA and Early Mass Surveillance Programs

The abusive practices of FBI officials were not unique to that agency. On the contrary, during the Cold War years, NSA officials had instituted at least two known illegal surveillance programs that adversely affected the privacy rights of American citizens.

The first of these, code-named Operation SHAMROCK, originated during World War II. Working closely with three international telegraph companies and relying on the authority of wartime censorship legislation,^{Footnote 46} the NSA’s predecessor, the Army Signals Security Agency (ASSA), captured the telegraph messages of the Soviet Union and Axis Powers that were transmitted through company terminals in the United States.

Despite the end of World War II (and the concurrent expiration of wartime censorship legislation), ASSA officials pressured the executives of these companies to continue their surveillance assistance. This time they targeted the communications of the Soviet Union and Soviet bloc countries. At first willing, company executives soon had second thoughts, having concluded that their actions were legally questionable and that their employees might publicly disclose their cooperation with the government. Secretary of Defense James Forrestal soon secured the continued assistance of the telegraph companies based on two assurances. First, Attorney General Tom Clark pledged, with President Harry Truman’s concurrence, that the government would not prosecute their firms for participating in Operation SHAMROCK. Second, he promised that the Truman administration would introduce legislation shortly legalizing this surveillance practice.

Appropriate legislation was drafted and introduced in Congress. However, while the proposed bill was under consideration by the Senate Judiciary Committee, Truman administration officials asked the committee’s chairman, Alexander Wiley, not to report the bill for floor consideration, fearing that doing so would provoke a divisive debate during the 1948 presidential campaign. The draft bill was set aside and was not reintroduced in succeeding years. Instead, in 1951, Congress approved a seemingly minor section of an omnibus bill that was intended to clarify sections of the U.S. code. Among the proposed changes was a seemingly innocent and noncontroversial section that made it illegal for any “unauthorized” person to disclose information relating to codes and code-breaking. Since one purpose of the SHAMROCK interception program was code-breaking, this provision, at least indirectly, addressed the central concern of company executives over possible disclosures by their employees.^{Footnote 47}

In 1975, when Congress first learned about Operation SHAMROCK, congressional investigators requested access to the program’s records. This request could not be honored, NSA officials responded, because the massive volume of intercepted communications had been regularly destroyed for housekeeping reasons. They nonetheless estimated that millions of telegrams had been intercepted over the duration of the program (1947–75), adding that during the past two to three years NSA analysts had reviewed approximately 150,000 telegrams monthly. They also acknowledged that the program had expanded to intercept “virtually all the international telegrams of Americans carried by RCA Global and ITT World Communications.”^{Footnote 48}

NSA officials instituted another illegal interception program, code-named Project MINARET, in 1967. Under great pressure from the Lyndon Johnson administration to document Soviet, Chinese, and North Vietnamese influence over the militant civil rights and anti–Vietnam War movements, Central Intelligence Agency (CIA) and FBI officials turned to the NSA for assistance. They specifically sought to exploit the NSA’s technological ability to intercept international communications. FBI and CIA officials proposed an arrangement under which they would provide the NSA with the names of individuals whose international communications they sought. NSA officials agreed to honor this request but on the strict condition that, on receipt of NSA-generated reports, FBI and CIA recipients would either destroy them within two weeks or return them to the NSA. To further ensure that their participation would not be discovered, NSA officials required that the resultant reports not be “identified with the National Security Agency” and that all records relating to this program were “not serialize[d]” or filed with other NSA records, were classified “Top Secret,” and were stamped “Background Use Only.” Agency officials made these stipulations because they recognized that this interception program violated the Communications Act of 1934.

The NSA’s assistance ended up being far-reaching. Over the course of this program (1967–73), NSA analysts prepared 1,900 reports (of which 950 were provided to the FBI) based on the interception of the communications of 1,030 U.S. citizens and 2,400 foreign citizens. Not surprisingly, those individuals who were targeted included well-known political activists, including Jane Fonda, Joan Baez, Dr. Benjamin Spock, Rev. Martin Luther King Jr., Rev. David Abernathy, Eldridge Cleaver, Abbie Hoffman, and David Dellinger.^{Footnote 49}

The Costs of Secrecy

The record of American surveillance policy and practices from the World War II era to the present offers clear and compelling lessons for contemporary debates. First, in times of real or perceived threat, elected policymakers are prone to defer to the judgment of intelligence officials about the nature of the threats and the requirements of security. Both secret presidential directives (Roosevelt’s orders of August 1936 and May 1940 and Bush’s order of October 2001) and laws openly enacted by Congress (the USA Patriot Act of 2001 and the FISA Amendments Act of 2008) granted the FBI and NSA broad discretionary authority to monitor suspected spies, saboteurs, and terrorists. As Roosevelt’s case illustrates, the disposition to give security officials the authority they ask for is often shared by left-of-center Democrats as well as conservative Republicans.

Second, under cover of secrecy, intelligence officials not only exploit this broad officially sanctioned deference but stretch and evade statutory and judicial constraints to expand their freedom of action. Their aggressive strategies are highlighted by the FBI’s past use of special records and record-destruction procedures to avoid detection of illegal methods and by the extraordinary scale of the NSA’s more recent unauthorized collection of telephone and e-mail communications. Intelligence officials struggle to break free of legal and constitutional constraints partly because, as an occupational matter, they develop enlarged, if not alarmist, views of the prospective internal security threats. The lack of both specific legislative criteria and critical oversight has facilitated the abuses of expanded FBI and NSA surveillance powers. In a broader sense, they have seriously compromised the American political system’s fundamental values of accountability and the rule of law.

Finally, the abuses of privacy resulting from this combination of broad authorized discretion and aggressive unauthorized freelancing on the part of intelligence agencies have had politically invidious effects. Protected by secrecy, intelligence officials have indulged their own ideological prejudices, or have taken cues from partisan politicians, in deciding which individuals and groups constituted genuine threats. Abusive investigations have targeted political dissidents, critics of administration policy, advocates of progressive reform, and religious minorities, among others. Looking ahead, if democratic political institutions prove unable to define and enforce constraints on the intelligence methods used to protect the public from terrorist attacks, the threat of terrorism will seriously erode American democracy.