The Library: Timeline of Foreign Censorship Laws
Legislation, codes, agreements, and other documents pertinent to the ongoing speech debate.
Since J.D. Vance’s speech in Munich on February 14th, European internet regulations have been at issue. The Racket Library attempts to compile pertinent documents in one place, by theme, with the aid of readers. If you think a document or regulation should be on this list but is not, please write to library@racket.news.
This list is far from complete. It will be updated and is meant to be understood as an introduction to the overall timeline of U.S.-European relations with regard to civil liberties, censorship, and hate speech laws. We’ve elected to speed past early efforts to define Internet law like France’s 2004 LCEN, or Loi pour la confiance dans l’économie numérique, which lacked the muscle and ambition of later counterparts.
Aggressive European speech laws began to be implemented in the mid-2010s. Though American officials were often sympathetic and hoped to implement similar ideas here, these efforts were generally rebuffed, as “experts” observed that the First Amendment made EU-style laws an impossibility. European rights guarantees are often framed as “freedom of expression, but,” as in the case of the French Declaration of the Rights of Man and of the Citizen. It calls the “free communication of ideas” “precious… except what is tantamount to the abuse of this liberty.” Americans consider their rights inherent.
The difference didn’t matter, until recently. Now, it matters a lot.
Again, to submit documents to the timeline, please write to library@racket.news. Some key moments on the road to the current standoff:
2016 EU
Code of Conduct on Countering Illegal Hate Speech Online. This “voluntary” agreement went into effect May 30, 2016, after terror attacks in Paris and Brussels.
Signed by Facebook, Microsoft, Twitter and YouTube, it is a landmark moment in content moderation history, and very strict, requiring companies to “review the majority of valid notifications for removal of illegal hate speech in less than 24 hours and remove or disable access to such content.” It anticipated the future Digital Services Act by creating the concept of “trusted reporters.” The DSA would employ “trusted flaggers.”
Failure to comply with recommendations on “hateful content” threatened signatories with fines of billions of Euros. In a move Silicon Valley had resisted, the signatories were asked to “intensify cooperation between themselves and other platforms and social media companies to enhance best practice sharing,” which ultimately led companies to announce the launching of a joint digital database of “terrorist” content.
The firms did not want to take this step. As one executive told the Washington Post with regard to a similar effort in June of 2016, “As soon as governments heard there was a central database of terrorism images, they would all come knocking.” The code would become a model for other rules around the world:
2016 UK
The Investigatory Powers Act gives the UK power to “obtain communications data” upon request, targeting groups or locations instead of individuals. Current Prime Minister Sir Keir Starmer was critical of Theresa May for supporting the bill, which at the time he called “the most intrusive surveillance”:
The actual bill granted the state extraordinary investigative power, far beyond the USA PATRIOT Act:
2017 Germany
Passage of Network Enforcement Act, or NetzDG. Broad law on hate speech and “misinformation.” Sometimes known as the “Facebook Act.” English version here.
2018 EU
Code of Practice on Disinformation. Another leap forward in “voluntary” content moderation, the Code of Practice on Disinformation went into effect in 2018 after the Cambridge Analytica scandal and alleged Russian interference in the 2016 election.
The original 2018 agreement signed by Facebook, Google, Twitter, and Mozilla:
2019 Spain
The Prosecutor General’s office issued a circular explaining Internet offenses under Article 510 of the criminal code. Updated in 2020, it outlines broad categories of speech offenses, such as “Aporophobia” (hatred or rejection of the poor), “Anti-multiculturalism,” and “Intolerance” (which includes “denying human diversity”).
2019-2020 USA
Virginia Democrat Jennifer Wexton introduces the “COVID–19 Disinformation Research and Reporting Act of 2020,” in what is considered an answer to European legal advances on curbing “misinformation”:
Still, overt efforts to pass speech laws repeatedly failed, and the U.S. began to be seen as lagging behind its Western partners. New York Times reporter Cecilia Kang went on CBS to summarize the frustration of American officials over the “thorniest” problem for implementing hate speech regulation, namely that “we have in our law… the First Amendment.” As a result, Kang said, “the US is sort of in a position where they have to abide by really our constitutional limits,” while the “United Kingdom, Germany, New Zealand, Australia [and] India” are taking action.
Despite the domestic outcry, companies like Facebook, Google, and Twitter were already working to stay in compliance with a series of “voluntary” agreements, like the European Code of Practice on Disinformation. During the first year of the pandemic, Twitter sent monthly compliance reports:
2020 Switzerland
Swiss amend laws to ban anti-gay discrimination. Rules on “Discrimination and Incitement to Hatred” in the criminal code would be updated several times. See Article 261bis:
2021 Australia
Australia issues its own voluntary “Code of Practice on Disinformation,” whose signatories quickly include Facebook, Microsoft, Twitter, TikTok, and RedBubble. Its main objective is “reducing the risk of Harms that may arise from the propagation of Disinformation and Misinformation.” It asks signatories to prioritize “credible” and “trusted” content, provide funding for “fact-checking” services, and remove verifiably “false” content, though what this means in practice is left vague, as was the case in the European law.
The Australian code:
2021 EU
Anticipating tighter cooperation from the U.S. on speech issues after the election of Joe Biden, the EU issues a parliamentary report titled, Harnessing the new momentum in transatlantic relations: Potential areas for common action during the Biden presidency. Written after early negotiations with American counterparts, it imagined “we could be going” to a place where the United States used the authority of the FTC to install its own version of the Code of Practice on Disinformation:
Forging common standards on the rule of law, countering disinformation, and tackling corruption to uphold the fundamental rights of citizens could potentially provide for a stronger transatlantic partnership. The USA could seek inspiration from the EU's Code of Practice on Disinformation and enforce it through the consumer protection authority in the Federal Trade Commission.
This report also talked at length about the U.S. and Europe developing common fact-checking standards and other forms of cooperation. It’s an important read:
2021 Scotland
Hate Crime and Public Order Act. The Scottish law added yet another layer of prohibitions atop existing EU law and Britain’s 1986 Public Order Act, which barred “insulting” behavior likely to cause “harassment, alarm or distress.” This new law specified a number of new protected group categories, including transgender identity.
2022 EU
Europe updated its Code of Practice On Disinformation, which now had 42 signatories. The main change? A fully voluntary agreement was restructured to become a “co-regulation” for “Very Large Online Platforms,” with stiffer penalties and more tools for demonetization. Ostensibly, the Code was “strengthened” in response to Covid “conspiracy theories” and “aggressive pro-Kremlin disinformation.” The strengthened Code was also meant to help prepare signatories for the coming Digital Services Act, which would be even broader in scope and fully remove the “voluntary” aspect of the regulation.
2022 USA
Throughout this period there were repeated efforts to introduce European-style public-private content bureaucracies in the United States. Some were successful, at least temporarily (the Election Integrity Partnership and the Virality Project being examples). However, numerous bills stalled in Congress, including an ambitious Department of Homeland Security project called the Disinformation Governance Board. The DGB would serve as “a central point of contact for interagency partners such as the White House, State’s Global Engagement Center, DOJ, HHS, DOD, and the Intelligence Community,” tackling “misinformation” issues with DHS “stakeholders” in government and the private sector. Amid public outcry, the project shuttered.
The Department of Homeland Security has (partly) complied with FOIA requests, but most of the important documents were redacted, a situation that may be remedied shortly. Some internal memoranda did get out, including docs obtained by Senator Chuck Grassley’s office. These were important in shaping public opinion and are reproduced below:
2022-2024 EU
The Digital Services Act. A massive undertaking, sometimes called Europe’s “Man on the Moon” moment, the DSA is the most comprehensive censorship law ever passed in a Western democracy:
The broad European Commission law required adherence to the recommendations of “trusted flaggers” that went into force on November 20, 2022, but began applying to “all platforms” as of February 17th, 2024. There is a long list of publicly available documents about the DSA, including data from complaints as well as polling numbers and transcripts of debates for and against. For now, we’ve just included a few:
Regulation as passed from October 27, 2022:
Addendum on supervisory fees, March 2, 2023:
Explanatory Memorandum, October 20, 2023:
2022 USA (Twitter Files)
On July 29, 2022, an executive of Twitter’s Legal Policy office in Dublin sent their higher-ups stateside a progress report on Twitter’s compliance with Germany’s Network Enforcement Act, a.k.a. NetzDG. “Hi everyone,” the employee began:
I am delighted to share with you that yet another NetzDG report (9th version) has been successfully completed and is now publicly available. This was truly a great team effort which you can see from the list of contributors below.
He went on to give a shout-out to 31 other full-time Twitter employees, hinting at the substantial resources Twitter was dedicating to complying with foreign laws. Twitter officials seemed compelled to e-congratulate each other whenever they secured favor with NetzDG, which spoke volumes.
Another executive made a list of all Twitter Legal Policy employees fully or partially devoted to keeping the company in compliance abroad. “People on LP spending 50% or more on global regulation,” read one section, followed by a list of names. The first was listed as working on “DSA, LCEN, TBA, FICA, TERREG, Any other EMEA reg for WR, various country-specific transparency reports…”
Some of those laws are already mentioned on this timeline. Others include South Korea’s Telecommunications Business Act (TBA), sometimes called the first Internet censorship law, which was frequently expanded to include criminal penalties even for true statements (“publicly alleging facts”). FICA and POFMA are Singaporean statutes. TERREG is another EU law against terrorist content. These entries were just for the first employee.
Twitter’s efforts to comply with laws like NetzDG are suspiciously similar to the informal efforts involving agencies like the FBI and the Departments of State and Homeland Security in the United States. American efforts were both more convoluted and more camouflaged. Execs handling American content mostly didn’t work in departments with positions like “Legal Takedowns” or “Civil Takedowns.”
2023 UK
Online Safety Act. Britain’s crown jewel censorship law, which has a namesake in Australia, imposes harsh penalties for speech violations, placing internet speech under the oversight of UK media regulator OFCOM:
2025 Germany
J.D. Vance speaks at the Munich Security Conference:
Racket will continually update the list. European residents, lawyers, and others interested in the subject can contribute by pointing us to primary sources.
Write to us at library@racket.news, and we’ll take a look ASAP.
Matt, this is an outstanding effort to document such a critical issue. The timeline lays out, in plain sight, how ‘voluntary’ agreements quickly turn into mandates, and how private companies get roped into enforcing regulations that should be debated in public, not behind closed doors. Appreciate the thorough research—this is the kind of work that keeps the record straight. Keep going!
I understand you are all having a conversation about Europe vs the US. I live in India, and just from a media criticism perspective, this may be helpful.
The trend in the past few years in the mainstream media in the West is to claim democracy is "backsliding" in some countries around the world. In fact, this phrase "democracy backsliding" has its own wikipedia entry and there are "experts" in it who show up on CNN.
Also keep in mind the narrative that around Trump's election in 2016, "right-wing-populists" had made significant advances around the world. This narrative has been revived recently with Trump's second election.
One of the countries that was dragged into this narrative was India. The words "authoritarian", "right-wing-populist", "democratic backsliding" have been found in mainstream media articles in NY Times, WaPo, CNN, BBC, Guardian with reference to India.
The Indian opposition leader Rahul Gandhi, the dynastic leader of the Indian National Congress party has been treated the way exiled princes from monarchies were treated back in the day when colonialists had fun - he has been invited to conferences all around the West to give his opinion of "democratic backsliding" in India since the current "Hindu nationalist" govt took over.
In fact, during the 2024 elections the coverage was so intense that it was treated like the last elections in India.
However, I want to show you how these kind of narratives are misinformation.
Back in 2008, the current opposition party, Rahul Gandhi's party was the ruling party in India.
In the Indian Parliament, his party brought in a set of Amendments to the Information Technology Act. Among these was section 66A.
Section 66A made it possible for the police in various state governments around India to punish social media users for content.
This became law, and in the period between 2008 and 2015, when the Indian Supreme Court struck down this section, hundreds of civilians across the country were arrested and imprisoned for mostly harmless content. In one case, two young women who liked a post were arrested!
This is the party that is now being held up as "Resistance to Hindu fascists" by not only the mainstream media in the US, but also by Congresswomen such as Ilhan Omar.