Columnists

OPINION | Xenophobia unleashed: Facebook a megaphone for Operation Dudula’s hate speech

Listen to this article
-00:00
Subscribers can listen to this article
Operation Dudula and community members during a protest over spaza shops in Diepkloof, Soweto, last year.
Operation Dudula Protest Over Spaza Shops In Diepkloof Operation Dudula and community members during a protest over spaza shops in Diepkloof, Soweto, last year. (Gallo Images/Luba Lesolle)

Heidi Swart considers how Meta’s rollback of professional fact-checking and ineffective moderation has emboldened the spread of hate speech and division in South Africa.

Tech giant Meta announced earlier this year it would terminate professional third-party fact-checking on its social media platforms in the United States, eliciting intense criticism from its own ethics oversight board, civil society, and the African Union.

Yet indications are that Meta aims to eliminate professional fact-checking globally. With social media’s role in stoking public violence well documented, the elimination of paid fact-checking could reduce the already limited effects of Meta’s content moderation in South Africa.

READ | Phillip de Wet: Facebook's changes call for a Great (Political) Internet Wall for South Africa

Facebook and Instagram allow crucial connections between friends, families, and businesses. But people also post misleading and heinous content, and it’s in Meta’s interest to ensure its platforms aren’t overrun by falsehoods and banalities; it needs to operate in the mainstream to continue to grow.

Yet, vitriolic, divisive content is what gets the clicks, also a growth requirement. Add in a gargantuan user base (Facebook alone has more than 3 billion users, posting in more than 100 languages) and Meta’s premium on free speech, and content moderation becomes Sisyphean.

Meta has Community Standards – hundreds of rules defining what can’t be posted. It relies on two tiers of “personnel” to implement these standards.

The first tier – artificial intelligence and users’ own reporting – detects content potentially violating standards. (The efficacy of AI is questionable. In 2021, leaked internal company communications indicated that Facebook’s AI detected only 3% to 5% of hate speech.)

The second tier is the review phase. A post can be removed, have a warning or disclaimer label added, or show up less frequently in feeds. Content moderators look at whether posts encourage violence, call for segregation or exclusion of certain groups of people or whether posts depict some people as sub-human. Fact-checkers examine posts’ veracity, labelling them as false, altered, partly false, missing context, satire and true.

Community notes

But the third-party professional fact-checking programme in the US has now been scrapped. In January, the company announced that it wanted to “undo the mission creep that has made our rules too restrictive ...” Meta vowed to eliminate restrictions on subjects like immigration and gender identity.

As an alternative, it introduced community notes; platform users themselves will deal with posts that potentially violate community standards. The model will be tested in the US before international expansion.

The move has been widely criticised.

In May, the African Union’s African Commission on Human and Peoples’ Rights said that community notes were not a viable substitute for professional fact-checking.

Meta’s own ethics oversight board accused it of failing to consider the impact on human rights.

Online posts can have dire real-life consequences. Facebook has been used to organise events that have led to deadly violence, including the 6 January uprising in the United States and conflict and genocide in Myanmar and Ethiopia.

So, can we rely on Meta to keep South African users’ posts truthful and on the right side of human rights? A brief glance at Facebook users supporting ideological extremes suggests not.

Anti-migrant organisation, Operation Dudula, uses Facebook to mobilise protesters against foreigners in South Africa. A search for Operation Dudula on Facebook yields just over 250 groups and pages with the words “Operation Dudula” in the title. (To date, xenophobic actions have led to 128 778 persons being displaced, 5 647 spaza shops being looted, and 693 people dying. That’s according to Xenowatch, an initiative tracking xenophobic incidents in SA.)

As of February, Operation Dudula was facing legal action, brought by some 500 foreigners accusing the movement of illegally and violently evicting them from their homes in a derelict building in 2022.

In June, rights group Kopanang Africa Against Xenophobia took Operation Dudula, the Department of Home Affairs and the South African Police Service to court in a bid to stop Operation Dudula’s xenophobic and racist speech and conduct against foreigners.

READ | Mbhazima Shilowa: Resentment over healthcare for immigrants: Is SA repeating history?

A look at Facebook user accounts affiliating themselves with Operation Dudula reveals ample posts promoting violence and negative sentiment towards foreigners.

A video filmed at night “Cable theft...Suspects caught in the act” shows a man with a bludgeoned face sitting on the ground with his knees drawn to his chest.

Caught in a torchbearer’s circle of light, his interrogator wants to know where he is from and where he is taking the power cables he’s allegedly stolen.

The man mumbles through swollen lips, “I was born in Tanzania.”

He apparently brushes his hand against the interrogator, who then responds: “Touch me again and I’ll break your fingers.”

Below the video, a comment: “Foreigners man, they are not here to help us grow economy, they are here to destroy this country” (sic). Another comment reads: “Cut off his strong hand . He’l never steal again.” (sic)

The Facebook group in which this video was posted has more than 2 100 followers, with several posts propagating a negative worldview of foreigners living in South Africa. One post below a video of a man in a nurse’s uniform is captioned: “WATCH: An illegal Ethiopian foreigner is currently working at Baragwanath Hospital as a nurse with no formal medical qualifications.” Another video shows two men ripping a large section of piping from the side of a building, the comment below stating, “Foreigners are here to destroy”.

Then there’s another extreme on Facebook: two posts both displaying the apartheid-era South African flag. One post shows a white soldier holding the old flag, with a caption encouraging those who “served” under the flag to repost the image.

Reported for hate speech

The second post included the apartheid-era flag, a whites-only beach and white children posed next to a black man with a bicycle selling ice cream. The caption reads that these images represent the “good old days”.

Notably, 184 Facebook users reported this last post, primarily for hate speech. Both of the above posts were viewed by millions.

Following its content moderation process, Meta decided to leave the posts up. Users then escalated the matter to Meta’s oversight board – an entity to which users can turn their complaints when they have exhausted all avenues offered by Meta.

In April, the board, via majority vote, decided to leave the posts up, stating that they do not clearly advocate for exclusion or segregation, nor can they be understood as a call for people to engage in violence or discrimination”.

A remarkable finding, given that in 2023 the South African Supreme Court of Appeal upheld an Equality Court ruling that the public display of the old SA flag constitutes hate speech, unfair racial discrimination and harassment.

Ultimately, it is clear that when it comes to racial hatred and xenophobia, Facebook leaves South Africans with ample freedom of speech.

Problem with algorithms

Even when challenged, Facebook still fails to remove content considered hate speech by the country’s own courts. But there may be a silver lining: the community notes system might just give SA users impetus to call out xenophobia and racial hatred.

However, Facebook’s algorithms are geared to give users more of what they want to see. Thus, the very people who would potentially call out xenophobia and racism probably won’t see harmful posts, not unless they actively look for them.

It will be a time-consuming and thankless endeavour. It’s a thin silver lining indeed.

- Heidi Swart is an independent analyst for RiskCollective.

*Want to respond to the columnist? Send your letter or article to opinions@news24.com with your name and town or province. You are welcome to also send a profile picture. We encourage a diversity of voices and views in our readers’ submissions and reserve the right not to publish any and all submissions received.

Disclaimer: News24 encourages freedom of speech and the expression of diverse views. The views of columnists published on News24 are therefore their own and do not necessarily represent the views of News24.

5482

Heidi Swart