France and New Zealand spearheaded the adoption on May 15 of the Christchurch Call to Eliminate Terrorist & Violent Extremist Content Online, a voluntary pledge endorsed by 18 countries and many tech companies (including Microsoft, Google, Facebook and Twitter). The United States refused to join, citing to free speech concerns. The Christchurch Call was named after the city in New Zealand where a horrific terrorist attack killed 51 people and injured 50 at two mosques in March. That massacre was live-streamed on Facebook, spreading quickly on that platform as well as other social media sites and raising concerns about how such content goes viral. U.S. isolation amidst close allies with respect to this initiative has led to questions about what were the First Amendment hurdles that prevented the U.S. from joining this pledge, especially given it constitutes a political commitment rather than a legally binding document.
Here’s some background about how the U.S. approaches speech issues in international settings. The longstanding position of the U.S. government is that it will not call on itself or even on other countries to take action that is inconsistent with First Amendment free speech principles, even if such calls to action are in non-legally binding documents. There are essentially two reasons for this policy. First, the United States seeks to avoid undertaking political commitments on the world stage to do things that it knows are unconstitutional and therefore cannot be implemented. To do otherwise would place the U.S. in the position of making political promises in bad faith, knowing it will never implement them. Second, it is also U.S. policy to promote the broadest possible protections for speech abroad because it believes the First Amendment’s approach has benefits for individuals outside its borders. If the U.S. were to join international initiatives that are inconsistent with that vision, its diplomatic efforts to promote broad free speech protections abroad would be damaged. In the context of joining legally binding international agreements, the U.S. government has a long track record of making clear (in its reservations, understandings and declarations) that it does not take on any obligations inconsistent with the First Amendment.
This policy has been implemented many times in previous administrations. For example, in an April 2009 statement explaining why the U.S. government would not attend a world conference to combat racism, the State Department noted serious concerns with the (not legally binding) outcome document that was set to be adopted because it contained language that, among other things, ran “counter to the U.S. commitment to unfettered free speech.”
Similarly, in 2016, the U.S. was relatively isolated in voting against a (not legally binding) U.N. General Assembly resolution about the glorification of Nazism and cited free speech concerns. When negotiating the 2011 Human Rights Council Resolution 16/18 to combat religious intolerance, the U.S. made sure any bans on speech that were included in this (not legally binding) resolution aligned with First Amendment principles, an understanding that was even publicly highlighted by the Organization of Islamic Cooperation’s Permanent Observer to the UN.
In 2012, when President Barack Obama requested Senate advice and consent for the Convention on the Rights of Persons with Disabilities, the transmittal package included a proposed understanding in its discussion of Article 8 that nothing in the Convention would require prohibitions on speech inconsistent with the Constitution. Such an understanding was triggered by certain Convention clauses that, for example, called on governments to encourage appropriate media coverage of disabled persons. Similarly, when the United States became a party to the International Covenant on Civil and Political Rights in 1992 and the Convention on the Elimination of Racial Discrimination in 1994, it took reservations to those treaty obligations involving hate speech, making clear it was not taking on any requirements to act inconsistently with the First Amendment.
Given this practice, the question for U.S. policymakers was whether the text of the Christchurch initiative essentially calls on the U.S. or others to act inconsistently with First Amendment principles. The answer is yes for a variety of reasons.
For starters, the Christchurch Call contains a number of pledges geared towards the elimination of “terrorist and violent extremist content” on the Internet, but that phrase is never defined in the pledge (nor does the phrase otherwise have a universally understood definition). While the First Amendment does allow for bans on advocacy of incitement to imminent violence as well as true threats, it does not permit the U.S. government to ban speech that falls in what appears to be the broader and amorphous concept of “terrorist and violent extremist content,” which could potentially encompass content that is being reported by journalists, dissidents, and civil society as well as other protected speech under the Constitution. This lack of a definition of the content to be suppressed (including the lack of any requirement that the illicit speech likely triggers any harm) is a threshold problem that taints the rest of the text.
In addition, the Christchurch Call contains a variety of commitments in which governments and companies pledge to work in unison to eliminate “terrorist and violent extremist content” from the Internet. While U.S. companies are private actors and not bound by the First Amendment, it would pose constitutional problems for the U.S. government to work with private actors to help them remove otherwise lawful speech. As set forth in the Christchurch Call, such collaboration between government and industry is to include prevention of uploading of the illicit content (which triggers concerns about prior restraints on speech) as well as redirecting users from problematic content (which triggers issues in terms of government interference with how individuals impart and receive information).
All in all, given the particular phrasing of the Christchurch Call, the U.S. decision to decline to endorse a text that is not in line with its free speech values is consistent with its past practice. That said, it is understandable that the sincerity of its free speech claims might be questioned because of the White House’s past positions on speech (e.g., attacks on the press and desire to expand defamation laws) as well as its reluctance to condemn hate groups.
While much of the press coverage (e.g., here, here, and here) focused on the U.S. refusal to join this initiative, a more interesting issue sparked by the adoption of the Christchurch Call could be its implications for respecting human rights while countering terror. Though the text repeats numerous times that various commitments to eliminate terrorist and violent extremist content from the Internet are to be undertaken consistent with “human rights” or “human rights law,” there are several red flags that indicate this language may not reflect a meaningful effort to grapple with the thorny freedom of expression issues that are triggered by the initiative’s primary focus on content removal.
First, the Christchurch Call promotes a variety of techniques that have been criticized for their potential adverse human rights impacts. For example, the text emphasizes use of upload filters to stop the spread of violent content. This is problematic because most content moderation decisions require human evaluation of context and not merely the use of automated content filtering systems to avoid infringing on freedom of expression, as highlighted by Access Now, an advocacy organization that fights to protect human rights in the digital age. In addition, the initiative seems to allocate most of the judgment calls to companies to decide what is violent content and therefore to remove it. This outsourcing of governmental functions to tech companies is problematic for a variety of reasons, including the lack of judicial oversight or meaningful appeals processes.
Second, many countries that supported the Christchurch Call have broad bans on terror-related content that don’t pass muster under international human rights law, as highlighted in Amnesty’s Dangerously Disproportionate report about terror laws across Europe. For example, France criminalizes and actively prosecutes public “apology” of terrorism, which covers positive speech about terrorists or acts of terror even if such expression does not rise to the level of incitement to violence. This law has been criticized by various components of the U.N.’s human rights machinery, including the U.N. Special Rapporteur on promoting human rights while countering terrorism. Also problematic are laws in Spain and the United Kingdom that criminalize the “glorification” of terrorism. A 2016 joint statement by the U.N. Special Rapporteur on freedom of expression and regional expression experts noted their alarm at the criminalization of vague concepts such as “apology” or “glorification” of terrorism.
Will the Christchurch Call trigger France and other supporters to evaluate whether their laws are consistent with human rights law standards or will the initiative be used to justify existing laws? Unfortunately, it appears likely that the pact could spur continued implementation and expansion of such laws and related EU initiatives.
Third, a bellwether for how seriously human rights issues are being considered by governments involves how well civil society is included in deliberations on such initiatives. In this case, civil society was largely shut out of the text negotiations until the day before its adoption, which was too late to be impactful. This lack of consultation with civil society was lamented by many, including the Director of the Free Expression Project at the Center for Democracy & Technology, Access Now, the Electronic Frontier Foundation and the Global Network Initiative, a multi-stakeholder initiative the promotes expression and privacy rights online. Civil society groups rallied to make their voices heard anyway by producing a joint statement.
As has been previously highlighted on this blog, regulation after horrific acts of terror often results in governmental measures that overreach and undermine human rights. While it is imperative that governments prioritize combatting terrorism, hate and intolerance in all their manifestations (and laudable that Christchurch Call supporters are seeking to promote this goal), it is also important to achieve these ends while respecting human rights. It will therefore be essential for civil society to continue its outreach to supporters of this pledge to promote implementation that is truly consistent with international human rights protections, particularly given that the U.S. statement only articulated reasons for not joining the pledge in terms of its domestic law and policy concerns without raising concerns about the international human rights aspects of this and similar initiatives.