Published by The Lawfare Institute
in Cooperation With
The U.S. Department of Justice is using a once-obscure law, most commonly used today to require the registration of those who lobby in the United States on behalf of foreign governments, to target foreign-based anonymous disinformation actors on social media. What is the significance of this unprecedented move, and what can be done to develop a sustainable enforcement regime for these anonymous online actors?
The Foreign Agents Registration Act, or FARA, is the central law governing agents of foreign entities’ political activities in the United States. The law was enacted in 1938, in the midst of the first House Un-American Activities Committee’s investigation into the effects of Nazi propaganda efforts on U.S. soil. The law was designed to insulate American institutions of government from covert foreign influence by requiring that foreign propagandists disclose their political activities in the United States. In the first decades after its enactment, the Justice Department enforced FARA against propagandists from Germany and the Soviet Union through criminal prosecutions. Later changes, beginning with amendments to FARA in the 1960s and culminating in the passage of the Lobbying Disclosure Act in 1995, precipitated a more lax enforcement model centered around voluntary compliance. This enforcement model has led many observers to criticize the law as ineffectual and innocuous.
That all changed in the past few years, when FARA once again began making headlines. In November 2017, then-Special Counsel Robert Mueller charged President Trump’s former campaign chairman, Paul Manafort, and his deputy, Richard Gates, with violating FARA for acting as agents of the Ukrainian government without filing the appropriate paperwork with the Justice Department. Also in 2017, after receiving media criticism for his work on behalf of the Turkish government, President Trump’s former national security adviser, Michael Flynn, retroactively filed paperwork under the act as a foreign agent of Turkey. In late 2019, former Obama White House Counsel Gregory Craig was charged with making false statements under the act (charges that were later dismissed by court order). The Justice Department’s new “zero-tolerance” crackdown on FARA nondisclosures has led to a 30 percent increase in registrations since 2016.
But the Justice Department has also begun to employ FARA in an entirely new context, a move that has so far failed to attract significant attention or analysis. In February 2018, Mueller indicted 13 Russian individuals and three organizations allegedly involved in the 2016 Russian coordinated disinformation campaign. Mueller’s team claimed that these actors violated a number of laws—including failing to register as foreign agents under FARA. But why is the government charging these foreign-based actors with violating FARA, a statute that is currently being used to regulate U.S. lobbyists?
As the case against these actors has proceeded in the U.S. District Court for the District of Columbia, the government indicated in October that it intends to argue at trial that the defendants conspired to cause a number of individuals or organizations to act as agents of a foreign principal, for which the individuals and organizations or the conspirators (or both) would have had a legal duty to register under FARA with the Justice Department. In other words, the government is arguing that foreign online disinformation actors such as the Russia-based Internet Research Agency, and perhaps those Americans acting at the behest of foreign disinformation organizations, have a legal duty to register their activities with the department.
It is unprecedented for the Justice Department to use FARA in the context of social media disinformation campaigns that originate abroad. As stated above, the department has used the act in recent decades to target lobbyists working on behalf of foreign governments in Washington, D.C. As the attorney for one of the indicted Russian organizations put it: “[N]ever before has a foreign corporation … with no presence in the United States, been charged criminally for … the political speech of individuals on social media, at rallies, or in advertisements during a U.S. presidential election campaign.” Similarly, a former head of the FARA Unit at the Justice Department stated: “[T]his case may represent the first time the [Justice Department] has charged foreign nationals, operating predominantly from a foreign country, with criminal violations of FARA.”
As I have explored in a recently published paper, using FARA in this novel context poses a number of interesting statutory construction and enforcement questions. FARA was enacted at a time when Nazi propaganda took place out in the open on American soil. The propagandists charged in Mueller’s indictment, unlike those Nazi propagandists who operated at the time of FARA’s enactment, are anonymous and work covertly to influence U.S. political opinion from abroad. While FARA’s purpose is to disclose sources of foreign political influence in the United States, using FARA in this new context is not an easy fit.
This post identifies and discusses the challenges that exist in using FARA in the online disinformation context. It also explores potential enforcement mechanisms that could be introduced if the Justice Department intends to begin requiring FARA registration from foreign social media actors.
Applying FARA to Online Disinformation Actors: Statutory Complications
FARA generally requires that “agents of foreign principals” who undertake certain activities on behalf of foreign interests—including engaging within the United States in political activities for, or in the interests of, foreign principals—register with the Justice Department, file copies of any “informational materials” that they distribute with the department, and maintain records of their political activities in the United States. The act is not designed to censor political propaganda that originates abroad. Rather, it aims to promote transparency about the nature and motivations of the speaker by shining “a spotlight of pitiless publicity,” as the House of Representatives described the law at its enactment, on foreign-based organizations seeking to influence U.S. politics.
From the act’s passage until 1966, FARA registration efforts were aimed primarily at traditional sign-holding political propagandists and the Justice Department used criminal prosecutions as the primary method of enforcement. Beginning in the 1960s, as the Smith Act (otherwise known as the Alien Registration Act, which set criminal penalties for advocating the overthrow of the U.S. government by force or violence) took FARA’s place as the central law regulating subversive activity, FARA enforcement shifted from criminal prosecutions to civil fines and encouraging voluntary compliance. This trend continued into the 1990s and early 2000s, during which time FARA was used primarily to coax lobbyists to voluntarily register themselves as agents of foreign powers. Recently, the Justice Department has begun to pursue the registration of lobbyists more vigorously, as evidenced by the recent high-profile prosecutions. Given FARA’s poor recent record of criminal enforcement, and a history of lax voluntary compliance, a number of legislative proposals have been introduced to consider ways to strengthen FARA’s effectiveness in combating traditional, non-anonymous sources of political influence. These include a proposal to provide the Justice Department with civil investigative authorities.
But these legislative proposals do not address some of the statutory and enforcement challenges that arise when applying FARA to social media actors.
The first issue that needs to be addressed by either Congress or a court concerns the location of the actor, or “agent.” FARA applies to an agent who “engages … within the United States in political activities for or in the interests of [a] foreign principal.” The organizers of coordinated disinformation campaigns such as the Internet Research Agency, however, are often operating from abroad, not from “within” the United States. This would appear to make it difficult to apply FARA’s statutory language to foreign-based disinformation propagandists.
This issue is not unique to FARA. Determining the proper venue for computer network crimes has confounded federal courts because “in today’s wired world of telecommunication and technology, it is often difficult to determine exactly where a crime was committed, since different elements may be widely scattered in both time and space, and those elements may not coincide with the accused’s actual presence.” But some courts have begun to address this problem in the realm of computer network crime. The U.S. Court of Appeals for the 4th Circuit, for instance, has found that venue may lie “where the effects of the defendant’s conduct are felt, but only when Congress has defined the essential conduct elements in terms of those effects.”
Borrowing the 4th Circuit’s logic, a court may therefore deem evidence that the effects of a foreign-based social media disinformation actor’s activity are felt within the United States to be sufficient to determine that the foreign actor is acting “within the United States.” Congress has helpfully defined FARA’s essential conduct elements in terms of those effects, not in terms of the origin of the actor. The statutory language of acting “within” the United States serves to narrow FARA’s application to foreign propaganda affecting American political opinion. Neither the statute nor its legislative history indicates that the language of the statute was intended to establish the importance of the propagandist’s physical location. Even so, Congress could make the statute yet more specific: It could amend the statutory language to “engages in political activities aimed at the United States” (emphasis added) to more explicitly define the statutory conduct elements in terms of the geography of the intended effect.
Another issue that a court or Congress would have to address is how FARA’s terms apply when the “agent” and the “foreign principal” occupy interchangeable positions because they are both foreign persons located abroad. The traditional FARA paradigm is that a “person outside of the United States,” a foreign principal, sends an agent inside the United States to influence the American political process. This paradigm no longer holds if a foreign organization employs foreign-based social media trolls, in which case the agent is also a “person outside of the United States.”
Where both the agent and the foreign principal operate from abroad, there is no meaningful legal distinction in identifying them as an “agent” or as a “foreign principal.” In this scenario, a court would likely find that those foreign actors working to influence U.S. political opinion are required to register under FARA, regardless of what label they would receive.
But in situations where foreign disinformation actors co-opt U.S. persons as their agents, the mechanics of the agent-principal relationship become more important. FARA statutorily mandates that violators of the act must run afoul of its terms “willfully,” so an unwitting American agent would not be required to register. If the U.S. person was willfully acting on behalf of a foreign power, however, the application of FARA would depend on whether the actor was acting as an “agent” of the foreign principal as the act defines it.
Under the act’s current language, FARA covers a broad range of possible agency relationships. FARA’s disclosure and record-keeping requirements are imposed on “persons” who act as “agents of a foreign principal.” A “person” is an “agent of a foreign principal” when the person “acts as an agent, representative, employee, or servant, or any person who acts in any other capacity at the order, request, or under the direction or control, of a foreign principal.” Justice Department regulations have not further clarified the necessary elements of an agent’s relationship with a foreign principal, but they have defined “control” to include “the possession or the exercise of power, directly or indirectly, to determine the policies or the activities of a person, whether through… contract, or otherwise.” This definition does not provide much guidance as to what kinds of activities, in practice, suffice to establish a principal-agent relationship.
The case law interpreting FARA suggests that the principal-agent relationship does not require that the parties expressly enter into a contract establishing the relationship. While financial support from a foreign principal alone is insufficient to establish agency, the U.S. Court of Appeals for the D.C. Circuit has held that financial support is not required to establish agency. FARA also does not include any threshold requirements for a certain degree of activity on the agent’s part (i.e., there is no set minimum amount of political activity that the actor has to engage in to properly qualify as an agent).
That said, in 1981, the U.S. Court of Appeals for the 2nd Circuit introduced a FARA-specific definition of agency for the purposes of determining “whether the relationship warrants registration by the agent to carry out the informative purposes of the Act.” The court cautioned that a broader reading, such as requiring that any person who acts at a foreign principal’s “request” register under FARA, would “sweep within the statute’s scope many forms of conduct that Congress did not intend to regulate.” Such a broad reading of agency might mean that a person who saw public information on Facebook about a protest, and attended the protest, would be considered an “agent” of the individual who organized the protest.
The court went on to state that the “surrounding circumstances” will usually provide sufficient indication as to whether a “request” by a foreign principal requires the recipient to register. The court suggested that it is important to ascertain whether “those requested to act were identified with specificity by the principal”—specifically, whether “a particular individual, or a sufficiently limited group of identifiable individuals, is asked to act.” The court also suggested that the specificity of the action requested is relevant: “Once a foreign principal establishes a particular course of conduct to be followed, those who respond to its ‘request’ for complying action may properly be found to be agents under the Act.”
In applying the 2nd Circuit’s test, a fact-specific investigation would be necessary to determine whether the surrounding circumstances suggest that a certain relationship qualifies as agency under FARA. The government would likely have to prove that the “agent” took specific direction from a foreign party, which would undoubtedly be difficult in light of the challenges in making high-confidence attribution determinations of social media accounts.
Establishing an Enforceable FARA Regime for Online Disinformation Actors
The issues addressed above reflect the challenges that already exist if the Justice Department intends to argue that the charged defendants in Mueller’s indictment had a duty to register under FARA. But if the department takes the next step and attempts to establish a FARA registration regime for all prospective foreign social media disinformation actors, a number of other issues arise.
The most glaring problem when it comes to the registration of foreign social media disinformation actors is that these actors’ raison d’etre is to remain covert. The Internet Research Agency would shut down an identified disinformation account and open a new one rather than register as a foreign agent with the Justice Department. Additionally, the department does not have jurisdiction to compel a foreign entity to submit its paperwork under FARA. Because FARA is a transparency statute, not a censorship statute, the Justice Department has no power to compel the takedown of disinformation accounts.
The situation is slightly less dire, however, when it comes to FARA’s labeling requirement. FARA has a labeling requirement for “any informational materials” transmitted “for or in the interests of [the] foreign principal,” requiring that these materials include a conspicuous statement that “the materials are distributed by the agent on behalf of the foreign principal.” The current definition of “informational materials” in the act states that they can be “in any form which is reasonably adapted to being … disseminated or circulated among two or more persons.” The act does not give further meaning to the phrase “disseminated or circulated,” suggesting that it is possible to apply these statutory terms to materials that are disseminated online.
Over the past few years, social media companies such as Facebook and Twitter have taken steps to remove coordinated disinformation campaign content from their sites. But posts that do not violate these companies’ terms of service may not be taken down. Twitter’s recent political ad ban, for example, does not apply to unpaid posts, which tend to display disinformation more frequently than paid posts. There may, therefore, be room for the institution of a FARA labeling regime for a certain segment of social media content.
Because FARA has no jurisdiction over foreign actors, a labeling requirement placed on the disinformation actors themselves, such as the Internet Research Agency, would be toothless. Indeed, the Justice Department may not even be able to attribute the precise identity of the coordinators of a disinformation campaign.
The Justice Department does have jurisdiction, however, over the social media companies that provide a platform for this content. But compelling social media companies to label posts on their sites would come with challenges under the First Amendment. The Supreme Court has held that an individual or group is not required to foster, display or help promote those ideas with which the individual or group disagrees. Requiring social media companies to place labels on certain accounts may, therefore, be problematic in cases where the companies disagree with the labeling decision. A labeling requirement would also potentially run into Section 230 of the Communications Decency Act, which provides companies like Facebook and Twitter with civil immunity for any action undertaken to restrict access or availability of material on their platforms (though courts have held that 230 does not provide immunity from criminal laws).
A labeling regime might be more palatable if it involved a voluntary partnership between the Justice Department and the social media companies to identify and label suspected accounts. But this would require the government to identify with some degree of confidence the social media accounts that warrant FARA labeling. The government would also have to engender trust in its labeling decisions. This suggests that an independent third-party body, outside the confines of the Justice Department, may be best suited to make the determinations. Congress would also have to enact a statutory means of redress to ensure that adversely labeled account owners were provided with due process.
Taken together, the difficulties in instituting a registration and labeling regime under FARA may suggest that the Justice Department is best suited leaving FARA out of the disinformation campaign conversation. But FARA was designed to be the primary law addressing foreign propaganda in the United States, and the current foreign-based social media propaganda threat will only worsen. If appropriately amended, FARA could play an increasingly important role in the coming years in helping to tackle a threat that is in many ways more dangerous to our democracy than the actions of “foreign agents” like Paul Manafort, Michael Flynn and Gregory Craig.