SZ: Mr. Schwarz, your study sees a connection between the use of social media and hate crime. Is it really that easy to imagine: You read hateful posts on Facebook and become violent as a result?
Carlo Schwarz: Just because you go to the right and read comments against refugees does not of course make you radicalized and commit attacks. We suspect that these are people who were previously critical of refugees and are therefore in these right-wing Facebook groups. Through the permanent exchange with like-minded people, these people are pushed over a line at some point. Then they are actually ready to attack refugee homes or refugees.
Specifically, you examined how attacks on refugees in Germany are related to the regional use of the AfD’s Facebook group. What did you want to find out?
The idea behind it is relatively simple: If the mood on Facebook plays a role, more attacks should take place in communities in which there are many users of these right-wing groups.
And according to your study, that is the case. Are there concrete examples where it can be safely said that hate on the Internet led to the assault?
I don’t know of any study that could establish this connection beyond any doubt. To do this, you would have to run an experiment: some people start reading these pages. Other people don’t look at them. And then you wait to see what happens. This is of course not possible for ethical reasons.
But how can you be sure that the attacks on refugees are really related to the use of Facebook?
In the study, we take advantage of the fact that there are repeated Internet outages, so that some users no longer have access to Facebook. There were 313 such failures during the life of our study. And if there is no internet in one place, you can see that the number of attacks is significantly lower in the week of the failure. If the attacks were driven by the general mood in society, i.e. independent of social media, an internet failure would have no effect.
Wouldn’t platforms like Facebook then have to do much more to counter the hatred on their channels?
It’s too easy for yourself to say: Social media should just delete all hate comments. Identifying these comments is not easy, even with the latest methods. And what “hate speech” is is also not clearly defined. This is a debate to be had as a society. Mark Zuckerberg shouldn’t decide what to say on social media.
In Germany there has been a law for some time that is supposed to combat hate crime on the Internet: The Network Enforcement Act, or NetzDG for short, obliges large platform operators to delete obviously criminal content within 24 hours of receiving a complaint. Is that a solution?
Well, the NetzDG was a first attempt to create some kind of legal basis for regulation. Am I a big friend of NetzDG? Neither. Because the incentives are set in such a way that Facebook tends to delete too many comments. If Facebook decided tomorrow to simply remove all comments that users have marked as “possibly questionable”, the law would be completely fulfilled. I don’t think there will be a perfect solution. Instead, there will be a compromise in society: these are things that we want to remove – and these are things that should remain. Nevertheless, there will always be wrong decisions.
Isn’t it a little late for such a compromise?
Even with cars, it took several decades before someone said: A seat belt might be a good idea.