Belgian Court Penalizes Meta For Failing To Boost & Promote Far-Right Politician
EU internet regulations and courts never fail to stupefy.
The entire concept of shadowbanning" has gotten distorted and changed over time. Originally, shadowbanning was a tool for dealing with trolls in certain forums. The shadowbanned trolls would see their own posts in the forums, but no one else could see them. The trolls would think that they had posted (because they could see it) but were just being ignored (the best way to get trolls to give up).
But, around 2018, the Trumpist crowd changed the meaning of the word to instead be any sort of downranking or limitation within a recommendation algorithm or search result. This is nonsensical because it's got nothing to do with the original concept of shadowbanning. But, nevertheless, that definition has caught on and is now standard.
Ever since the redefinition, though, angry people online (especially among the far right) seem to act as if shadowbanning" is the worst crime man could conceive of. It's not. The concept of shadowbanning" as now conceived (being downranked in algorithmic results) is no different than giving an opinion. Any algorithm ranks some things up and some things down, and the system is trained to do that with various variables, and some of them may be we don't think this account is worth promoting."
The freakout (and misunderstandings) over shadowbanning continue, though, and now a Belgian court has fined Meta for allegedly shadowbanning" a controversial far-right politician.
I will warn you ahead of time that my thoughts here are based on a series of English articles, automated translations of articles in other languages, and an automated translation of the actual ruling.
The basics are pretty straightforward. Tom Vandendriessche, a Belgian member of the EU Parliament, representing the far-right Vlaams Belang Party claimed that he was shadowbanned by Meta. According to Meta, Vandendriessche had violated the company's terms of service by using hateful language. And rather than banning him outright, the company had chosen to limit the visibility of his posts.
The ruling is strange and problematic for many reasons, but I'm still perplexed at how this result makes any sense at all:
According to the court, Meta was unable to provide sufficient evidence that the Vlaams Belang lead candidate actually engaged in the activities they accused him of.
It also found that the company had profiled the politician based on his right-wing political beliefs, a process that is forbidden under the European Union's GDPR regime.
This latter violation prompted the court to award Vandendriessche 27,279 in compensation, with the sum designed to cover the additional advertising costs the MEP incurred due to the shadowbanning.
A further 500 was also awarded to the politician to compensate for any damage Meta had done to his reputation.
Since this ruling is under the GDPR and not the newly in-place DSA, I've heard some saying that the result doesn't much matter, since future such disputes will be under the DSA.
But, really, the EU's approach to all of this is completely mixed up. The DSA was put in place because EU officials claim that websites aren't doing enough to stop things like hate speech (which is why I keep pointing out that the DSA itself is a censorship bill and will have problematic consequences for free speech). Yet, here, we're being told that the GDPR somehow creates a form of must carry" law that says that you have to host nonsense peddler speech and recommend it in your algorithms.
How can that possibly make sense?
It goes beyond must carry" to must promote." And that seems like a form of compelled speech, which is very problematic for free speech.
Of course, Vandendriesche is falsely claiming this forced promotion and forced recommendation is a victory for free speech. But that's nonsense because it involves forcing others to speak on your behalf, which is the opposite of free speech.
If you're wondering why Vandendriesche might have faced some limitations on the reach of his speech, well... you don't have to look far:
The European Parliament is currently conducting an investigation into racist language used by Vlaams Belang MEP Tom Vandendriessche during a plenary session in Strasbourg in January.
Vandendriessche was also blocked on Facebook in early 2021 after a post in June 2020 after likened Black Lives Matter movement as akin to book-burning in Nazi Germany. At the time, he wrote: After street names, TV series and statues, it will be books' turn. And finally ours. Until our civilisation is completely wiped out. If fascism returns one day, it will be under the name of anti-fascism."
Should he be allowed to say such ridiculous and hateful things? Sure. Should Meta be required to promote them? That seems utterly crazy.
Meanwhile, the day after this ruling, it was announced that he's also under a separate, new investigation as well for some sort of potential fraud, though the details are scant. Of course, he's still expected to be returned to the EU Parliament following the elections this weekend.
Either way, everything about this case makes no sense. If a platform judges that someone violates their rules (for example by posting what they and their users consider hate speech), how could it possibly make sense for a court to say you have to promote this person's speech to make sure it reaches as far and wide an audience as possible?