The Senate Proposes Five Amendments to FISA Reform
This week, the Senate will vote on five amendments to H.R. 6172, which would reauthorize certain intelligence-related authorities that expired on March 15 and would also make substantive changes to the Foreign Intelligence Surveillance Act (FISA) and related laws.
Published by The Lawfare Institute
in Cooperation With
This week, the Senate will vote on five amendments to H.R. 6172, which would reauthorize certain intelligence-related authorities that expired on March 15 and would also make substantive changes to the Foreign Intelligence Surveillance Act (FISA) and related laws. Charlotte Butash and I wrote about the substance of the House bill on March 11, when it passed the House on a 278-136 vote. Senate Majority Leader Mitch McConnell had scheduled a procedural vote on the House-passed bill for the evening of March 16 but scrapped that vote at the last minute after reaching a deal with Sens. Mike Lee and Rand Paul to take up the bill at a later date together with a package of amendments. To try to cover the lapse in authorization, the Senate on March 16 passed by voice vote a Senate bill that would have extended key existing surveillance-related provisions for 77 days. The House never took up the Senate bill—so the authorities in question have been inoperative for nearly two months.
Senate consideration of the House-passed bill and attendant agreements will proceed on May 13 and 14. Some outside groups have urged the Senate to reject the House bill entirely, preferring to let the relevant authorities remain lapsed. For his part, McConnell has urged passage of the House bill as-is, stating, “I hope the Senate will pass it next week, free of amendments that would jeopardize important tools that keep America safe.” In pursuit of this goal, McConnell has offered side-by-sides, or alternatives, to two of the bipartisan amendments offered by others.
In an unusual coronavirus-related twist, the agreement for consideration of these amendments set the threshold for passage of each amendment at 60 votes, rather than the usual simple majority, in order to short-circuit the need for an additional vote on cloture while preserving the cloture threshold of 60 votes. It seems that, in the absence of remote voting, there is consensus on reducing the number of votes senators must take in the Senate chamber.
What follows is an account of each of the amendments in question.
The Paul Amendment
The first amendment, by Republican Sen. Rand Paul of Kentucky, would radically change how the FISA regime applies to U.S. persons—a term that includes U.S. citizens, legal permanent residents, associations of U.S. citizens and U.S. corporations. Specifically, it would add to the Foreign Intelligence Surveillance Act a new Title IX outlining broad limitations on the act’s application to U.S. persons.
First, it would prohibit an officer of the United States from requesting, and the Foreign Intelligence Surveillance Court (FISC) from ordering, five different things: electronic surveillance of a U.S. person, physical search of premises or property of a U.S. person, installation and use of a pen register or trap-and-trace device to obtain information on a U.S. person, production of tangible things concerning a U.S. person, and any targeting of a U.S. person for the acquisition of information. In addition, it would limit the use of information concerning U.S. persons acquired under FISA from being used as evidence in court against that person. U.S. persons whose information is collected could use the information in court—presumably to sue the U.S. government or its officers.
The amendment would require that electronic surveillance, use of a pen register or trap-and-trace device, production of tangible things, or targeting of U.S. persons for information can be done only pursuant to a warrant issued by a non-FISA federal court and only under the Federal Rules of Criminal Procedure. (Pen registers are surveillance devices that capture the phone numbers dialed on outgoing telephone calls; trap-and-trace devices capture the numbers identifying incoming calls.) Finally, Paul’s amendment would make any information concerning a U.S. person acquired under Executive Order 12333—which, among other things, authorizes the intelligence community to collect, retain, and disseminate information concerning United States persons in accordance with established procedures—unavailable in any investigation or court proceeding, while any information collected on a person pursuant to Executive Order 12333 could be used by that person in a court proceeding.
This amendment represents a radical departure from how FISA authorities are used with respect to U.S. persons currently. Currently, FISA authorities are available with respect to U.S. persons who are suspected of being agents of a foreign power knowingly engaged in clandestine intelligence-gathering activities that constitute a likely violation of U.S. criminal statutes; or who knowingly engage in sabotage or international terrorism, or related activities, on behalf of a foreign power. Under Paul’s rubric, federal law enforcement authorities could not utilize FISA authorities to collect information on a State Department or FBI employee suspected of selling American secrets to China. Only the more limited criminal collection authorities outlined in the Federal Wiretap Act, 18 U.S.C. §§ 2510‐2522, known as “Title III,” would be available in order to collect information on such a person. These authorities are not designed for investigations that are classified or involve classified information for a number of reasons. They do not enjoy as much secrecy and protection of classified information—notably, grand jurors are not subject to secrecy requirements.
Finally, Paul’s rubric limits what can be done with information collected on a U.S. person. Under Title III authorities, the target of a wiretap is required to be informed of the wiretap. In a FISA investigation, the target is not notified.
Paul’s proposal, in essence, repeals FISA’s authority to engage in collection domestically against U.S. persons—the authorization of which was the core purpose of the statute in the first place.
The Wyden-Daines Amendment
A bipartisan amendment by Democratic Sen. Ron Wyden and Republican Sen. Steve Daines would prohibit Section 215 of the USA Patriot Act from being used for warrantless collection of Americans’ internet website browsing and internet search history information. Section 215 currently allows the government to obtain a FISA court order requiring third parties, such as telephone companies, to hand over any records or other “tangible thing” if deemed “relevant” to certain types of investigations including those related to international terrorism, counterespionage or foreign intelligence. Section 215 orders may have been combined with requests under other provisions, such as Section 216, which governs access to online activity like email contact information and browsing histories. The Wyden-Daines view is that collection of web browsing and internet search information constitutes digital tracking of Americans as they access the internet and should require a warrant. The senators’ summary of the amendment states:
Several Federal courts, including the FISA Court, have found that certain internet search history constitutes the “content” of communications. The narrowness of the findings, inconsistent court decisions, and the absence of U.S. Supreme Court jurisprudence requires that Congress, as a matter of policy, protect Americans from warrantless collection of their web browsing and internet search history.
Courts have struggled with the line between what constitutes the “content” of communications and what does not. For example, most courts have ruled that a first-level web domain, like www.amazon.com, is considered to be dialing-routing-addressing-signaling, not “content,” and as such could be collected without a warrant. But a domain name that results from a search of “anarchist cookbook”—a 1971 book that contains instructions for manufacturing explosives—would be considered “content.” The Wyden-Daines amendment is designed to clarify that all Americans’ internet website browsing and internet search history information would be off-limits without a warrant.
McConnell has authored a side-by-side to this amendment that would seem, on its face, to similarly limit an application under Section 215 to exclude an order authorizing or requiring the production of internet website browsing information or internet search history information. But the side-by-side would limit such an application only “to the extent such information constitutes contents of a communication, as defined in section 2510 of title 18, United States Code.” This qualification is key to understanding the McConnell alternative amendment: Rather than requiring a warrant for collection of Americans’ internet website browsing and internet search history information, it essentially restates and reifies the existing, somewhat confusing, state of affairs. In essence, the McConnell amendment maintains the current status quo regarding treatment of internet website browsing and internet search history information.
The Leahy-Lee Amendment
Finally, a bipartisan amendment by Democratic Sen. Patrick Leahy and Republican Sen. Mike Lee seeks to incorporate additional civil liberties and privacy protections into the FISA system by outlining reforms related to the role of the FISC’s amicus curiae and treatment of exculpatory evidence. A side-by-side amendment authored by McConnell, meant to draw votes away from the Leahy-Lee amendment to defeat its passage, is different in some key respects. Interestingly, an earlier version of the Leahy-Lee amendment shows that the updated version adopts some of the ideas from McConnell’s amendment, including annual reporting to Congress on accuracy and completeness of FISA applications. The updated version also shows a softening of some of the robust language that created a much more aggressive and empowered role for the role of the amicus curiae.
The USA Freedom Act of 2015 authorized the appointment of an amicus curiae in cases that raise “novel or significant” issues, and the authority has been used in only a limited number of cases. The amicus curiae provisions of the Leahy-Lee amendment would require the appointment of an individual with expertise in privacy and civil liberties and expand amicus appointments to include cases that relate to First Amendment protected activities; “sensitive investigative matters” (defined as those involving a public official, candidate, religious or political organization or staff thereof; news media within the United States; or a matter the court finds similar in its sensitive nature to the foregoing); the approval of a new program, technology, or use of technology; and the reauthorization of programmatic surveillance. It would require the amicus to raise “legal arguments regarding any privacy or civil liberties interest of any United States person that would be significantly impacted by the application or motion.” It would allow the amicus to “seek leave to raise any novel or significant privacy or civil liberties issue relevant to the application or motion or other issue directly impacting the legality of the proposed electronic surveillance with the court, regardless of whether the court has requested assistance on that issue.” Finally, it would provide amici with authority to petition for appellate review to the Foreign Intelligence Surveillance Court of Review and the U.S. Supreme Court.
While the expansion of the role of the amicus as outlined in this amendment is significant, parts of the amendment reflect existing practice. For example, the definition of “sensitive investigative matter” in the amendment essentially mirrors the FBI’s revised Domestic Investigations and Operations Guide (DIOG), which says:
A sensitive investigative matter (SIM) is defined as an investigative matter involving the activities of a domestic public official or domestic political candidate (involving corruption or a threat to the national security), a religious or domestic political organization or individual prominent in such an organization, or the news media; an investigative matter having an academic nexus; or any other matter which in the judgment of the official authorizing the investigation, should be brought to the attention of FBI Headquarters ... and other DOJ officials.
In this sense, the Leahy-Lee amendment applies the FBI’s existing definition of a “sensitive investigative matter” to bring it within the ambit of matters with which the amicus can assist the FISC.
More importantly, this part of the Leahy-Lee proposal brings to the fore the question of the exact nature of the role of the FISC’s amicus curiae. The amendment outlines a more aggressive role for the amicus in advocating for and protecting civil liberties, whereas in the existing model the amicus is styled as an assistant to the court should the court request it.
Interestingly, the earlier version of the amendment would have allowed the amicus to “raise any issue with the court at any time, regardless of whether the court has requested assistance on that issue.” The updated version removes this language.
The amendment provides that “an amicus curiae … may petition the Foreign Intelligence Surveillance Court to certify for review to the Foreign Intelligence Surveillance Court of Review of a question of law,” and if the court denies such review, the amicus must record a written statement of the reasons for such denial. If the review is granted, the FISC “shall appoint the amicus curiae to assist the Court of Review in its consideration of the certified question, unless the Court of Review issues a finding that such appointment is not appropriate.” Most significantly, an amicus can “petition the Foreign Intelligence Surveillance Court of Review to certify for review to the Supreme Court of the United States any question of law[.]” It also guarantees the amicus access to applications, certifications, motions and other supporting materials including classified information.
The amendment affords the FISA judge the ability to retain control of the process at every turn, but the proposed changes nonetheless mark a significant shift in the role of the amicus—bringing the role of amicus in the FISC closer to that of an advocate or even a defense attorney for an absent defendant. It puts the amicus in a more adversarial role with respect to the government and the FISA judge—and the consequences of that kind of change have, in the past, given pause to some experts familiar with the functioning of the FISC. In 2014, for example, Judge John D. Bates, the director of the Administrative Office of the United States Courts, wrote to key senators with his view of the Senate version of the USA Freedom Act that “we are concerned that inserting into FISA court proceedings an advocate with a statutory mandate to make specific arguments would raise substantial legal questions and impede the courts’ work without furthering the interests of privacy or civil liberties.” Specifically, Bates noted that “the participation of the special advocate could actually hinder the FISC’s ability to obtain complete and accurate information.”
The amendment also codifies expanded requirements for the attorney general or other federal officer to disclose exculpatory material in the FISA application including “all information in the possession of the Government that might reasonably … call into question the accuracy of the application or the reasonableness of any assessment in the application conducted by the department or agency on whose behalf the application is made” or “otherwise raise doubts with respect to the findings that are required to be made … in order for the court order to be issued.” In addition, the amendment requires a certification by the federal officer making the application for a court order for electronic surveillance or a physical search under FISA to include “a certification that the officer has collected and reviewed for accuracy and completeness supporting documentation for each factual assertion contained in the application.”
This part of the amendment seems like a fairly modest answer to the debate about exculpatory material that has raged as a result of the December 2019 findings by the inspector general of the Department of Justice that FBI personnel had “provided false information … and withheld material information” in connection with four applications to the FISC for authority to conduct electronic surveillance on Carter Page. The debate also involves the March 2020 findings by the inspector general that each of the 29 applications subsequently reviewed by the office as part of its more general review of FISA processes contained inaccuracies including missing files in four FISA applications and errors or inadequately supported facts in the other 25 applications. In December 2019, the presiding judge of the FISC issued a public order giving the FBI a deadline of Jan. 10 to come up with a proposal to fix the problems, which some experts found inadequate.
Some proposals on the issue of production of exculpatory material have been more specific and potentially more expansive than the general language proposed in this amendment. This isn’t to say the language is flawed—the more general statutory language is consistent with a range of practices the executive branch could develop. It does mean, however, that rigorous oversight of the executive branch’s practices—by the FISC, Congress and the Department of Justice inspector general—must continue over time to ensure that both the spirit and the letter of the law are respected.
The McConnell side-by-side amendment, which is designed as an alternative to the Leahy-Lee amendment, is different in some key respects. It narrows the circumstances under which an amicus curiae would be appointed to include only those that involve novel or significant interpretations of the law, present exceptional concerns about the protection of First Amendment rights, or involve a sensitive investigative matter. It would narrow the role of the amicus in a number of ways compared to the Leahy-Lee amendment and also limit the types of information the amicus curiae is entitled to receive.
The McConnell amendment also limits the requirement to provide the court with exculpatory information and certify compliance only to Title I FISA applications and physical searches, rather than to all FISA authorities. It is not clear why this requirement should apply only to a more limited set of FISA applications.
The definition of “sensitive investigative matter” for purposes of appointment of an amicus curiae is different from the Leahy-Lee definition. It excludes investigation of a domestic religious or political organization and the domestic news media—which is puzzling, because it seems that the FBI has itself decided to include these categories. The new definition also includes two new elements: “an application that targets a United States person when the application relies for its criminal predicate on only the provisions of the Foreign Agents Registration Act of 1938” (FARA) and “an act prohibited under section 371 of title 18, United States Code, consisting of a conspiracy to commit an offense under this Act.”
Based on the first new element, it seems the McConnell amendment seeks to bring greater scrutiny to the use of FARA as the criminal predicate in counterintelligence cases involving U.S. persons secretly acting on behalf of foreign governments. But it is not obvious why FISA actions rooted in FARA should necessarily warrant higher scrutiny. The purpose of FARA is to “insure that the U.S. Government and the people of the United States are informed of the source of information (propaganda) and the identity of persons attempting to influence U.S. public opinion, policy, and laws.” FARA does not prohibit any conduct but, rather, requires registration, and it creates civil and criminal penalties for an individual who willfully fails to register or makes a false statement or omission of a material fact in connection with registration. FARA has been criticized as difficult to administer, but that alone would not seem to be a justification for its special treatment here.
What is clear, though, is that FARA has been a stumbling block for a number of the president’s closest friends and advisers. In 2018, Trump’s former campaign manager Paul Manafort pleaded guilty to conspiracy to defraud the United States in connection with his failure to register under FARA as an agent of the government of Ukraine, and to conspiracy in connection with FARA-related false statements and misrepresentations to the Department of Justice in violation of both FARA and the general false-statements statute. At least two other people pleaded guilty to FARA-related charges, or lying about lobbying that should have been reported under FARA, in the context of the Russia investigation: Rick Gates, President Trump’s former deputy campaign chairman and Paul Manafort’s right-hand man; and Sam Patten, a Republican lobbyist who worked with Manafort. Most notably, former national security adviser Michael Flynn pleaded guilty to lying to the FBI about conversations with the Russian ambassador in December 2016 and in FARA filings regarding his work on behalf of Turkey—though the Justice Department, in a controversial action, recently moved to dismiss Flynn’s case. In addition, former Special Counsel Robert Mueller and his prosecutors also relied heavily on FARA in charging 13 Russian individuals and three Russian companies with illegally using social media to influence the 2016 elections.
The meaning and purpose of the latter element—regarding “a conspiracy to commit an offense under this Act”—is unclear. The phrase “this Act” refers to the FISA, and it is unclear whether the language is meant to apply to a conspiracy to commit an offense under FISA—like one of the offenses outlined in 50 U.S.C. § 1809 related to unauthorized electronic surveillance or disclosure of information obtained through unauthorized electronic surveillance—or something else, like a conspiracy under FARA.
A final section of the McConnell amendment amends FISA provisions related to electronic surveillance and physical searches, to include the following provision:
(e) An application that targets a campaign for Federal office may not contain information obtained from another candidate for that Federal elected office or an individual working on behalf of such candidate’s campaign.
Here again, it is not entirely clear that there is a clear policy rationale for such an amendment. Why exclude any source of information from consideration in a FISA application if that information is found to be credible? The provision seems aimed at reifying skepticism of the Steele dossier—a private report compiled in 2016 by an attorney for the Democratic National Committee containing allegations of cooperation between the Trump campaign and the government of Russia that was later used in support of an application for a FISA warrant for Carter Page—rather than achieving any clear policy goal. One can easily imagine President Trump crowing about members voting in favor of an amendment that essentially criticizes the use of the Steele dossier in the context of the Russia investigation.
***
Whether any of these amendments can obtain the necessary 60 votes to pass is not clear at the moment. What is clear is that if any one of them is approved, the Senate will have meaningfully altered the House-passed FISA reform bill—and the resulting bill will have to return to the House for final passage before being sent on to Trump for his signature.