We are writing from the Global Project Against Hate and Extremism (GPAHE), a nonprofit, civil society organization based in the US, to request that the Facebook Oversight Board recommend that Meta take additional and meaningful action to stop the spread of hate speech in non-English languages, specifically the Polish word “murzynów,” strengthen its enforcement of its policy to protect migrants, especially in non-English languages, and enforce its community standards and policies when it comes to political speech. By Meta’s own admission, political figures “often have broader influence” and “therefore, they may pose a greater risk of harm when they violate our policies.” For example, GPAHE has documented many elements of the spread of the dangerous, anti-immigrant “Great Replacement” conspiracy theory and the political and public figures who spread it. Given the volatile and polarized global narrative around migration and its place in political discourse, it is especially imperative that this Board also recommend to Meta that it enforce its public figures and civil unrest policy.
Violence associated with online anti-immigrant rhetoric
There are too many anti-immigrant violent incidents in Europe to list here. And there have been many studies documenting online activity with offline consequences. Examples include studies in Spain, Sweden, Germany, and the UK. News coverage includes this New York Times article titled “How Online Hatred Toward Migrants Spurs Real-World Violence” which focused on Ireland, Portugal and the UK.
Additionally, the racist and anti-immigrant “Great Replacement” conspiracy theory which is found around the world inspires hate, violence, and mass murders and is especially prevalent as refugees from North Africa and the Middle East make their way to Europe. And nowhere does it thrive and spread more than in online spaces like Facebook and Instagram.
As a technology company that impacts every corner of the globe, it is incumbent upon Meta to accept the responsibility that comes with great power and protect its users and our communities from blatantly false conspiracy theories, hate speech, and violence-inspiring disinformation before more damage is done.
The conspiracy theory is most often associated with the idea that immigration is meant to overwhelm the vote of those of European descent (white vote), thereby destroying their political power, erasing “traditional” and cultural “values,” which ultimately will “destroy” the country. This language has been widely adopted by far-right media and political figures and parties such as the Polish Konfederacja Wolność i Niepodległość political party.
A May 2022 AP poll in the US found that one-third of those polled believed “an effort is afoot to replace native-born Americans with new immigrants for electoral purposes,” a 2023 UK poll found that one-third of Brits also believe in the conspiracy theory, and a 2021 poll found that 67 percent of the French believed the great replacement would happen. This is a dangerous global phenomenon. Dozens have been murdered by killers who believed the conspiracy theory, most of whom had a significant connection to online spheres, with at least one using Meta platforms to livestream the horrors.
Mass attacks related to the “Great Replacement” conspiracy theory
- 2023 Jacksonville, Florida – three Black people murdered.
- 2023 Allen, Texas – eight people murdered.
- 2022 Buffalo, New York – ten Black people murdered.
- 2021 Toronto, Canada – four members of a Muslim family were run down and killed.
- 2019 El Paso, Texas – gunman hoped to kill Latinos, killing 23 people.
- 2019 Poway, California – a synagogue was targeted because Jews were “planning a genocide” of Europeans. One person murdered. Killer had earlier set a mosque on fire.
- 2019 Christchurch, New Zealand – 51 people were murdered at mosques. Livestreamed on Facebook.
- 2019 Halle, Germany – gunman targeted a synagogue but was unable to enter. One passerby was murdered.
- 2018 Pittsburgh, Pennsylvania attack – targeted a synagogue, murdering 11 people.
- 2011 Oslo, Norway–77 killed, many of them teenagers. Gunman targeted individuals he thought would bring mass Muslim immigration into the country.
The Polish word “murzynów”
It is frankly appalling that Meta believes itself to be able to determine whether a word is offensive to the Black community in Poland. If Polish people say that it’s a derogatory term, then
it simply is, even if the whole country is not offended. A word is not deemed offensive by consensus, but by the feelings of hurt and fear it inspires in the offended person. And the image in the post under consideration leaves no doubt as to the intent of the poster.
Public and political figures’ content moderation
In addition to Meta not being the entity to determine how an entire country views the offensiveness of a word, it is very likely that Meta declined to remove the post because the Konfederacja Wolność i Niepodległość political party posted it. Meta has wrongfully decided that content from political parties is subject to its “newsworthiness” exemption from the rules and policies that all other users must follow, allowing a massive amount of harmful content to proliferate on its platforms.
There is no question but that Facebook community standards should be applied to political speech and rigorously enforced. All Meta platforms state that politicians and public figures are subject to the policies, but they also have “public interest” or “newsworthiness” exemptions, effectively rendering the rules, including those on hate speech and misinformation, useless, if indeed political speech is adequately reviewed for violations.
Hate speech has a measurable impact on people’s willingness and ability to participate in the democratic process. It can exert psychological constraints on members of a targeted group causing them to withdraw from public discourse, the so-called “silencing effect.” Moreover, hate speech causes desensitization, a loss of the ability to understand others’ pain, destroying a common basis for political communication. As history continues to show, hate speech coupled with disinformation can lead to stigmatization, discrimination and large-scale violence. Hate speech has been identified as a precursor to atrocity crimes, including genocide, such as the Rohingya genocide of 2017. And violence against LGBTQ+ people reached its highest point in the past decade in Europe and Central Asia in 2022 and the US against the backdrop of “rising and widespread hate speech from politicians, religious leaders, right-wing organizations and media pundits.” Hate speech from politicians and state officials was reported in 23 countries across Europe, as well as Azerbaijan.
As UN Secretary-General António Guterres has said, “Addressing hate speech does not mean limiting or prohibiting freedom of speech. It means keeping hate speech from escalating into something more dangerous, particularly incitement to discrimination, hostility, and violence.”
It is specifically important that politicians not use social media to spread hate speech. A body of research suggests the incendiary rhetoric of political leaders can make political violence more likely, gives violence direction, complicates the law enforcement response, and increases fear in vulnerable communities. Political leaders’ remarks do not disappear on social media, especially as the social media platforms’ algorithms tend to amplify more incendiary remarks, quickly magnifying rhetoric against their political opponents, minority groups, and other targets. Leaders with large social media followings will see their remarks shared with millions of followers. This then drives coverage in more traditional news outlets and serves as a cue to local politicians whose similar content is in turn amplified by their communities and the company algorithms.
Politicians use social media for their campaigns, but they are also able to bypass rules and norms of traditional media. And for those who wish to engage in hate and demonizing speech, the results can be damaging.
The post in question for this appeal should have been taken down under the existing hate speech and misinformation policies, without regard to the identity of the poster. The post clearly should have been reviewed based on Facebook protections of “race, ethnicity, national origin and religious affiliation” and the fact that “refugees, migrants, immigrants and asylum seekers are protected against “the most severe attacks.”
