(Editor’s Note: This is part of a series on the FISA Section 702 reauthorization and reform debate.)
On November 16, members of the House Permanent Select Committee on Intelligence (HPSCI), identified as “the Majority FISA Working Group” of that committee, released a report on Section 702 of the Foreign Intelligence Surveillance Act (FISA). As the report explains, the House judiciary and intelligence committees formed a joint working group to examine Section 702, a controversial surveillance authority scheduled to expire on Dec. 31. However, none of the judiciary committee members of that working group signed on to the HPSCI report, and the characterization of the report as a “majority” product indicates that the Democratic HPSCI members of the working group also dissented.
The report acknowledges that intelligence agencies have repeatedly abused Section 702 to conduct warrantless surveillance of Americans, and it declares that the law “cannot [be reauthorized] without significant reforms.” But the report includes false and misleading statements that minimize the serious privacy violations perpetrated under Section 702. And the changes it suggests not only would fail to address the most serious abuses of Section 702 — they would, in some cases, actually expand surveillance.
Background
Congress enacted Section 702 to make it easier for the government to address foreign terrorist threats. The law grants the government broad authority to collect the communications of non-Americans located abroad without a warrant. But the surveillance inevitably sweeps in Americans’ communications, as well. Concerned about the impact on Americans’ privacy, Congress directed the government to “minimize” the retention and use of Americans’ private phone calls, emails, and text messages.
Instead, intelligence agencies have turned Section 702 into a go-to domestic spying tool. Last year alone, they performed more than 200,000 warrantless searches of Section 702-acquired data for the express purpose of finding and reviewing Americans’ communications. By itself, these “backdoor searches” (or “U.S. person queries,” in the government’s terminology) are an egregious circumvention of privacy rights, but the problem is compounded by what the FISA Court has described as the FBI’s “persistent and widespread” violations of the minimal rules that apply to such searches. These abuses include baseless searches for the communications of members of Congress, 19,000 donors to a congressional campaign, and 141 racial justice protesters, among many others.
Lawmakers from both parties are thus vowing not to reauthorize Section 702 without significant reforms. In response, those who are more hawkish on surveillance have grown increasingly strident in opposing meaningful reforms, often claiming or implying that changes to the law could give free rein to terrorists, narcotics traffickers, and other threats to national security — claims that are not supported by the available evidence, as we have discussed elsewhere.
False and Misleading Statements in the HPSCI Report
In defending Section 702, the HPSCI report makes several false or misleading statements. Three of them merit particular attention here.
First, the report claims that “‘[w]arrantless backdoor searches’ are not permitted under Section 702” because incidental collection “does not authorize the U.S. government to target” Americans. This claim conflates the collection of Americans’ communications with subsequent searches for them. Section 702 prohibits the government from “targeting” Americans at the collection stage ( i.e., from collecting all of their communications), but the law does allow the government to “incidentally” collect any communications an American has with a foreign target. Once collected, intelligence officials can search through these communications without a warrant, including when they are looking for Americans. It is at this second stage that a backdoor search occurs. The government freely admits to running searches of communications collected under Section 702 for the express purpose of retrieving and reviewing Americans’ communications; that is why the government calls them “U.S. person queries.”
Second, the report falsely claims that “all courts find Section 702 constitutional.” In fact, while four district court judges and the FISA Court have found backdoor searches to be constitutionally reasonable, four circuit court judges — including a unanimous Second Circuit panel in United States v. Hasbajrami — have indicated that backdoor searches raise serious Fourth Amendment concerns. Far from ruling backdoor searches constitutional, the Hasbajrami court rejected the government’s argument in support of the constitutionality of backdoor searches and remanded the matter to the district court for further proceedings.
Finally, the report claims that Section 702 is irreplaceable because “Section 702 is the only legal authority within FISA that authorizes the U.S. government to collect foreign intelligence information on non-U.S. persons in this manner” (emphasis added). The statement is technically accurate, insofar as there is no other FISA authority that precisely replicates Section 702. It is nonetheless highly misleading, given that there are multiple other ways in which the government may obtain the communications and other information of non-U.S. persons. For instance, when the government is operating overseas, it may collect foreign intelligence on non-U.S. persons under Executive Order 12333 with even fewer legal restrictions than under Section 702. The government also purchases information about foreigners from data brokers.
The primary relevance of these alternative means of collection is that they, too, enable the government to obtain private information about Americans without a warrant or other compulsory legal process. Under Executive Order 12333, the government may obtain communications between foreign targets and Americans; it may also collect bulk data in which foreigners’ and Americans’ data is commingled. For example, the CIA, acting pursuant to Executive Order 12333, has collected bulk data that includes Americans’ financial records and what appears to be information related to communications, which the CIA then retrieves through backdoor searches.
Even when acting inside the United States, the federal government claims the right to buy Fourth Amendment-protected data about Americans from brokers. Agencies such as the FBI, Department of Homeland Security, and Department of Defense have routinely exploited this “data broker loophole” to purchase sensitive information about Americans, such as location information and email metadata showing the time, source, and destination of emails.
In short, if Congress fixes only the problems with Section 702, it will find itself playing surveillance whack-a-mole as intelligence agencies rely on other legal loopholes to obtain similar results — which is why it is so critical that Congress not reauthorize Section 702 without comprehensive surveillance reforms.
Ineffective “Reforms”
Unfortunately, the changes recommended in the HPSCI report are far from comprehensive — and are wholly inadequate even as they apply to Section 702. For example, the report recommends codifying the basic tweaks that the FBI already has implemented to its internal Section 702 policies. But even with those changes, the FBI itself assesses that it conducts noncompliant backdoor searches 2 to 4 percent of the time—amounting to 4,000–8,000 unlawful searches for Americans’ private communications every year. And those noncompliant searches continue to include alarming abuses, including recent searches for the communications of a U.S. Senator, a state senator, and a state court judge who reported civil rights violations to the FBI.
A number of the other “changes” the report recommends are also mere restatements or codifications of existing law and policy. For example, the proposed prohibition on queries conducted for the purpose of burdening free expression or disadvantaging someone based on their race, ethnicity, gender, sexual orientation, or religion would mirror restrictions already imposed by the Constitution and reflected in agency policy. The proposed prohibition on the FBI receiving Section 702 information not connected to an open national security investigation, as well as the proposed requirement that the FBI report its annual backdoor search metrics, would also just codify existing policy and practice.
Another proposed change is to reduce the number of FBI personnel who may “authorize” backdoor searches by 90%. Currently, however, most queries do not require any prior authorization, so it is unclear whether or how this change would apply in those cases. Presumably, the majority of abuses will take place in situations where prior authorization is not required. If the intent is to limit the number of FBI agents who actually conduct the queries, that begs the question — not answered in the report — of what criteria might be used to identify that smaller pool. Absent a reliable process to weed out those who are most likely to violate querying restrictions, there is little reason to think this solution would be helpful. Indeed, assuming the number of U.S. person queries will remain constant under HPSCI’s proposal, charging a small number of agents with conducting 200,000 queries each year could be a recipe for haste and carelessness.
Perhaps the most significant change proposed in the report is a requirement that FBI personnel obtain a warrant when performing a search solely for the purpose of finding evidence of a crime. This is a pale imitation of the warrant requirement for all backdoor searches called for by privacy advocates, because it would apply to vanishingly few queries. The vast majority of backdoor searches performed by the FBI — and all of them performed by the other agencies with access to raw Section 702 information — are at least partially for the purpose of finding foreign intelligence information, and so would fall outside the scope of the proposed warrant requirement. Indeed, there were only 16 cases last year in which the FBI ran “evidence-of-a-crime-only” queries that resulted in Section 702 data being returned and accessed. Moreover, the most egregious abuses of Section 702, including those aimed at racial justice protesters, campaign donors, and members of Congress, were justified as efforts to find foreign intelligence.
Even when it comes to improving the workings of the FISA Court — a set of reforms that should be low-hanging fruit, given the widespread acknowledgment of the court’s failings — the report falls short. It proposes only a skeletal version of the reforms that were included in the so-called “Lee-Leahy” amendment, which passed the Senate in 2020 by a resounding vote of 77-19. The HPSCI proposal includes transcribing the FISA Court’s proceedings and appointing amici curiae (independent experts who serve as advisors to the court) to weigh in on annual Section 702 re-certifications, but it excludes other key aspects of Lee-Leahy, such as appointing amici in a wide range of sensitive investigative matters (for instance, those involving religious groups or the media); ensuring that amici have adequate access to underlying materials; and permitting amici to request that the FISA Court certify its decisions for appellate review. It is extremely telling that even the highly popular reforms embraced by a supermajority of the Senate in 2020 go too far for the HPSCI working group majority.
An Expansion of Warrantless Surveillance
Not only does the report fail to recommend sufficient reforms, some of the changes it proposes would actually expand warrantless searches and other forms of surveillance. Most significantly, the report proposes amending Section 702 to allow the government to search Section 702 data for the communications of any non-U.S. person — including people lawfully inside the United States — who applies for immigration benefits, such as a visa, asylum, or a green card. These searches would not have to meet even the minimal existing standard of being “reasonably likely” to produce foreign intelligence or evidence of a crime. This would be a dramatic expansion of the current vetting practice, which, while described only in heavily redacted court documents, appears to involve limited and largely automated queries of a subset of Section 702 information.
In contrast to the “evidence-of-a-crime-only” warrant requirement, which would apply in a tiny fraction of cases, government statistics indicate that this expansion of backdoor searches would affect millions of people. Many of these individuals are inside the United States at the time they apply for benefits, including people with work permits or student visas who are applying for a green card and people who have entered this country seeking asylum.
The report also recommends restoring two provisions of the U.S.A. Patriot Act that Congress chose not to reauthorize in 2020. The report’s justification for these recommendations is one brief paragraph asserting, without elaboration, that the authorities “would help law enforcement and intelligence officials more easily track bad actors” and “would make it easier to identify and prevent acts of terror.” Yet over the past three years, the government has presented no evidence that the lapse of these provisions — one of which was literally never used — has affected national security or created any difficulties in the conduct of investigations.
* * *
What is needed is not the “mere window dressing” provided by the HPSCI working group majority report, but comprehensive surveillance reforms to fix the problems with Section 702 and rein in other easily abused spying authorities. A bipartisan, bicameral bill has already been introduced in Congress that would do just that. The Government Surveillance Reform Act of 2023, led by Senator Ron Wyden, Senator Mike Lee, Representative Warren Davidson, and Representative Zoe Lofgren, would close the backdoor search loophole, check surveillance under Executive Order 12333, and prohibit the government from buying its way around the Fourth Amendment, among many other crucial changes. As we wrote previously, the Government Surveillance Reform Act is both the right approach and potentially the only viable path to reauthorizing Section 702.