Civil rights protesters have long been the target of state surveillance. Recent efforts to keep tabs on Black Lives Matter protests and demonstrations against family separation, for instance, echo the FBI’s counter-intelligence program used to spy on civil rights leaders from Martin Luther King, Jr. to Huey Newton.
Yet today’s surveillance analysts have a new source of information: social media. Consider the recent protests over the police killings of George Floyd and Breonna Taylor. As demonstrations spread across the country, the FBI and local police monitored social media and made arrests based on what people have posted online. In one instance, police in Wichita, Kansas even arrested a teenager on suspicion of incitement to rioting based on a threatening Snapchat screenshot he shared. The teen’s post added a note cautioning readers to “stay tf away from” his hometown—rhetoric intended to denounce the call to violence, not to foment it.
Unfortunately, the surveillance of social media is a growing trend. In recent years, social media posts have landed individuals of color in overbroad and unreliable gang databases, and even been used to justify keeping them imprisoned. One New York teen spent more than a year on Rikers Island, based in large part on the district attorney’s incorrect assessment that he was a member of a criminal gang. The D.A. relied on Facebook photos of the teen with members of a local crew—a group of kids loosely affiliated by block or housing development—and several posts from crew members that he had “liked.” In reality, the teen was simply connected to crew members because they were his neighbors and family members. The extent to which communities of color are viewed with suspicion can create a self-fulfilling prophecy where basic social media etiquette is mistaken for membership in a criminal enterprise. Social media is highly contextual and prone to misinterpretation, magnifying the risk that one person’s innocuous post will be taken as something more sinister.
From protests to public housing, social media monitoring raises civil liberties and civil rights concerns that are currently going unaddressed. Establishing a framework that balances public safety and the right to privacy, free expression, and equal protection under the law requires updates to our existing regulatory controls.
How do police monitor social media?
To begin with, how do the police watch social media? Most commonly, an officer views publicly available posts by searching for an individual, group, hashtag, or another search vector. Depending on the platform and the search, it may yield all of the content responsive to the query or only a portion. When seeking access to more than is publicly available, police may use an informant (such as a friend of the target) or create an undercover account by posing as a fellow activist or alluring stranger. This allows officers to communicate directly with the target and see content posted by both the target and their contacts that might otherwise be inaccessible to the public.
Police have also used software to monitor people, groups, associations, or locations in a more automated manner. This software included tools that mapped clusters of activity and a platform for linking undercover accounts. This tactic is less common now after the major platforms prohibited app developers from receiving automated access to public content for surveillance.
Dataminr, the prominent social media analytics firm, appears to have found a partial workaround to this prohibition by providing police with “public sector alerts.” Dataminr’s automated systems analyze public data feeds and deliver automated alerts to law enforcement clients, including the FBI, about shootings and natural disasters. It is unclear whether law enforcement can customize the service to go beyond the types of alerts included in Dataminr’s marketing material.
Law enforcement may also request social media data as part of a criminal investigation. By deploying subpoenas and warrants, law enforcement can collect an array of data directly from social media companies. For example, a template warrant drawn up by the Department of Justice to serve on Facebook contemplates collecting an array of data, including contact information, photos, status updates, private messages, friends lists, group affiliations, “friend” requests, future and past event postings, privacy settings, and more.
Social media monitoring is pervasive. According to a 2017 survey by the International Association of Chiefs of Police, 70% of responding police departments use social media for intelligence gathering and to monitor public sentiment. Similarly, Facebook’s latest transparency report states that the company received over 50,000 government requests for data between July and December 2019.
What does the law say?
Despite widespread use of social media by police, there are few laws that specifically constrain law enforcement’s ability to engage in social media monitoring. In the absence of legislation, the strongest controls over this surveillance tactic are often police departments’ individual social media policies and platform restrictions, such as Facebook’s real name policy and Twitter’s prohibition against using its API for surveillance. While the constitutional landscape is unsettled, constitutional protections for privacy, freedom of speech and association, and equal protection may provide mechanisms for individuals to challenge the government’s collection and use of their social media data.
Fourth Amendment protections
The Fourth Amendment guarantees the right of the people to be free from unreasonable searches and seizures; the inquiry for whether a search was unreasonable, outside of the core protections of the Fourth Amendment for “persons, houses, papers, and effects,” generally comes down to whether a person has a “reasonable expectation of privacy” and whether society recognizes that expectation as reasonable. While courts have begun to recognize that privacy doesn’t require absolute secrecy, they have nevertheless typically held that individuals do not have a recognized expectation of privacy in data publicly shared online. As one appeals court put it, “If you post a tweet, just like you scream it out the window, there is no reasonable expectation of privacy.”
When police want access to information that is not readily available to the public, however, the Supreme Court generally requires law enforcement to meet a higher standard. The Court has held, for instance, that when law enforcement uses digital tools to view information that would otherwise require a warrant to obtain, the use of that tool requires a warrant as well. Nevertheless, courts have generally allowed police to engage in undercover operations (both online and in the real world) without obtaining a warrant—though individual law enforcement agencies may put additional restrictions in place.
This permissive approach largely grows out of a separate legal principle called the “third-party doctrine.” Under this doctrine, when people share information with a third party, whether another person or a business, they should expect that the data could be disclosed to the government.
Courts have extended this reasoning to social media. In United States v. Meregildo (2012), for instance, a New York district court found that while a person’s privacy settings might have signaled an intent to maintain secrecy, any expectation of privacy was “extinguished” when they shared content with Facebook friends because those individuals were free to use that information however they saw fit—including by sharing it with law enforcement.
But the boundaries of the third-party doctrine are becoming murkier, and the Supreme Court is increasingly signaling a need to reconsider the doctrine in light of technological advances. Most recently, in the 2018 case Carpenter v. United States, the court ruled that despite the third-party doctrine, the police had to obtain a warrant before it could access historical location information held by cell phone providers. And in the Supreme Court’s 2012 decision in United States v. Jones, which held that the government had to get a warrant to install a GPS device that enabled constant location tracking, Justice Sonia Sotomayor observed in her concurrence that secrecy might need to stop being treated as a “prerequisite for privacy” in order to account for the volumes of sensitive data shared with third parties on a daily basis.
The application of the Fourth Amendment has been rooted in practical limitations, but the force multiplier effect of social media surveillance calls those limitations into question. If police want to physically trail a car, there are limits to the number of vehicles that can be followed. Having to make determinations about how best to allocate an officer’s valuable time provides some rough limits, even if those resource decisions are not always made judiciously. By contrast, a single officer can monitor the social media accounts of dozens of people all at once, without having to leave her workstation. Undercover social media accounts offer greatly expanded power as well. An officer who would have to take care to create and maintain a single undercover persona in real life could have twenty different personas online. And sophisticated software may be able to assist police in monitoring thousands of accounts at the push of a button.
In 1983, the Supreme Court predicted that this kind of wide-scale surveillance could trigger a different legal analysis. “If such dragnet-type law enforcement practices…should eventually occur,” it may be time “to determine whether different constitutional principles may be applicable,” Justice William Rehnquist wrote in United States v. Knotts, in which the court decided that tracking a suspect using a surreptitiously planted beeper device did not constitute a Fourth Amendment search.
Finally, law enforcement can also seek account information directly from social media companies. Under the Stored Communications Act (SCA), law enforcement can serve a warrant or subpoena on a social media company to get access to information about a person’s social media profile. The SCA also permits service providers to voluntarily share user data without any legal process if delays in providing the information may lead to death or serious injury.
While numerous defendants have challenged the constitutional validity of warrants to search their social media data, courts have generally upheld the warrants. For example, courts have upheld warrants looking for IP logs to establish a suspect’s location, for evidence of communications between suspects, and to establish a connection between co-conspirators.
First and Fourteenth Amendment protections
The First and Fourteenth Amendments offer protections where surveillance is based on political or religious beliefs, associations, racial and ethnic identities, and other protected categories or activities. Surveillance can have a chilling effect on First Amendment freedoms, and surveillance that disproportionately targets a protected class can give rise to equal protection harms under the Fourteenth Amendment.
Social media monitoring can have serious chilling effects on protected speech and association. When police target individuals for surveillance because of their political viewpoints, people may choose to censor their online activity and associations to reduce the risk of governmental monitoring. Likewise, law enforcement may use social media to compile dossiers on people on the basis of First Amendment-protected activities and may share them among local, state, and federal agencies—this can increase the risk that protesters later face retaliatory targeting in the form of deportation proceedings or unrelated criminal prosecutions.
The case law on whether these impacts give rise to concrete First Amendment claims continues to evolve. In Laird v. Tatum (1972), the Supreme Court adopted a restrictive approach to the First Amendment, ruling that plaintiffs in the case did not have standing to challenge an Army intelligence program stood up in response to 1960s civil rights protests. That program collected information from newspapers and police departments and sent Army intelligence agents to attend public meetings. But, the Court ruled, the mere existence of a surveillance program that may have chilled speech did not cause sufficient harm to give the plaintiffs the right to sue under the First Amendment.
Where the targeting of First Amendment-protected activity leads to a concrete harm, affected individuals may have standing to raise a claim. For example, in Baird v. State Bar of Arizona, a lawyer was prevented from joining the state bar because she refused to answer if she had ever been a member of the Communist Party. There, the Supreme Court ruled that the government “may not inquire about a man’s views or associations solely for the purpose of withholding a right or benefit because of what he believes.” Applying this ruling to the digital era, if a person’s social media is surveilled on the basis of her political beliefs or associations, and she is later denied a civil benefit or prosecuted for an unrelated crime in retaliation for the beliefs or associations revealed, she should have standing to bring a First Amendment claim. In a time when social media monitoring allows the government to discover a person’s beliefs and associations easily and covertly, there is a particular need to ensure that this information is not misused to target disfavored individuals for adverse government action. Unchecked, discriminatory surveillance can have chilling effects on unpopular but lawful associations.
Fortunately, in Hassan v. City of New York (2015), the Third Circuit ruled that where discriminatory government monitoring dissuades individuals from exercising their constitutional rights, they may have standing to challenge the surveillance. The Hassan court evaluated the NYPD’s post-9/11 surveillance of Muslim communities in New York and New Jersey. This surveillance program targeted mosques, student groups, businesses, cafes, and more, leading many individuals to limit or withdraw from being active participants in their community. The Third Circuit ruled that where surveillance is racially or religiously biased, or undertaken in retaliation for exercising First Amendment rights, impacted individuals have standing to challenge the practice in court—even if the discrimination is not “motivated by ill will, enmity, or hostility.”
Extending Hassan to social media monitoring, surveillance that targets protected speech or disproportionately targets a racial or religious group, and leads to a concrete harm, can give rise to a viable First or Fourteenth Amendment challenge, even if the surveillance was not animated by “overt hostility or prejudice.” By intentionally targeting Black Lives Matter activists, police may be engaging in this kind of discriminatory surveillance. The same analysis could also apply to the surveillance of teens and pre-teens of color suspected of gang activity.
These discriminatory surveillance programs, which target their online lives for disparate treatment based on their ethnicity, may stamp them with a badge of inferiority that violates their rights to equal protection under the law. In Boston, a defense attorney recently won a discovery order requiring the police department to produce records related to its surveillance on Snapchat, which will offer an opportunity to determine whether the program disproportionately targets Black and Latino men.
Room for reform
The availability of social media has dramatically expanded the scope of law enforcement surveillance. At the same time, few departments have publicly available policies governing their use of social media for intelligence, data collection, and criminal investigations. There are a number of practical steps that can be taken to begin to address this gap.
- Every jurisdiction should be required to hold public hearings and obtain local government approval before police engage in social media monitoring. Where departments are already engaged in this practice, they should pause the bulk of these operations pending public hearings and evaluate whether existing surveillance programs disproportionately target constitutionally protected groups or associations.
- Every law enforcement agency that uses social media for data gathering purposes should have a publicly available policy that describes their use of social media. These policies should detail the restrictions and procedures for social media monitoring and should specify the standards governing collection, use, retention, and sharing of personal information. Social media policies should contain clear prohibitions against surveillance based on race, religion, gender, sexual orientation, immigration status, or a person’s exercise of First Amendment freedoms. The policy should also specify the legal processes that must be followed before law enforcement can seek social media data from companies. The restrictions imposed by a department’s social media policy should be legally enforceable, such as by state attorneys general or the Department of Justice.
- Not every investigation warrants the use of invasive covert accounts. There should be strict controls on the use of this technique, including ongoing monitoring, supervisory approval and oversight, and time limitations. Law enforcement should be banned from impersonating an actual person without that person’s permission. Where law enforcement wants to use a covert account, they should be required to document that no less-invasive means are available and to submit the documentation to an external body for oversight and approval. Judicial oversight of online undercover activity would offer the most robust protection.
- Police are prevented from interviewing minors without notifying their parent or guardian. This protection should be extended to the online space with a flat prohibition against police connecting with minors via social media.
- Social media monitoring should be subject to ongoing reporting and audit requirements. For example, police should have to regularly disclose information such as the number of social media investigations that are open and closed and those that are extended past their original closure date. The reports should also indicate where investigations may impact protected classes of people. Each police department’s social media monitoring practices should be audited by an independent entity on an ongoing basis to ensure compliance with constitutional protections and safeguards.
These recommendations are intended as starting points for enacting needed reforms. They represent a regulatory floor, and jurisdictions should be empowered to enact stronger prohibitions against certain types of monitoring. This may prove necessary in jurisdictions where police departments exhibit repeated and consistent practices that violate constitutional rights and freedoms. Establishing the right regulatory controls will require input and action from a number of stakeholders, from communities to civil society to police departments and local government to Congress and the Department of Justice. But this important regulatory action is overdue and should not be put off any longer.
This piece was originally published by the Brookings Institution.