“Virality on the Internet” and OHPI

In a recent article, Reddit moderators talk about the challenges of moderating the online discussion platform.

A few things the article highlights are:

  • The company is bound by US laws, which are clearly on the side of unfettered free speech regardless of context. Other countries take on a more nuanced view of things. We have explained the Australian legal position on the issue here. We take a more stringent view on hate speech under the law, and Australians social media users are bound by it.
  • While governments have to follow the free speech legislations, private companies and enterprises can set their own rules of engagement for their platforms. They are well within their rights to ban certain kinds of speech and behaviours.
  • The way social media is built encourages the creation of echo chambers, and lets people think that their views have legitimacy. When they hit critical mass, their views overflow across communities and swamp the debate. (To put it simply, the content “goes viral”.)
  • Reddit’s way of tackling the problem is to allow these communities to exist but “only in a small corner with no resources and no permission to aggressively promote an agenda across the site”. In short, with reduced ability to let their views and posts go viral. It is still in the process of building tools to achieve this goal.

OHPI supports Reddit’s attempts to tackle the “virality” of content on social media. Unfortunately, until the tools are built and rolled out, we can’t comment on whether they go far enough to tackle the problem.

However, the discussion gives us the opportunity to clarify our stand on hate speech.

Many people think OHPI stands for outright banning of all hate speech. We don’t. We stand for regulating hate speech from going viral, that is, finding a large audience, who were not seeking their viewpoint, and dominating the dialogue with the help of social media technologies. Our mission is to prevent the harm caused by the distribution of hate speech within society to individuals and communities.

For example, we don’t fight to remove individual websites and blogs from publishing hate speech. We don’t call for the close down of closed online communities, even if we think that they stand for hateful and discriminatory views.

What fight against is the amplification of their voices through social media, because content on social media can easily be distributed and circulated in a way that gives such views and speech legitimacy. If racist speech keeps appearing on your newfeeds unchecked, it starts appearing acceptable simply through its wide presence. We start thinking is it ok to say such things.

The difference is between allowing people to debate and discuss hateful ideas in their drawing rooms vs allowing them to spray paint those ideas onto every house in the neighbourhood.

Currently, social media platforms have no way to tackle speech from going viral. All it takes is enough likes, shares and retweets for discriminatory ideas to pollute the entire pool. No boundaries or walls are built into the infrastructure of such platforms to identify and contain hateful speech. Only getting it removed can ensure that it doesn’t find a wider audience. So OHPI fights to get it removed off the platform.

To see all our publications on social media policies, go here. For our most recent article on Reddit, go here.

Please share this article, if you found it helpful.


Your support via our Facebook page will help in our work to combat all forms of online hate. You can also stay informed about our work by joining our mailing list for a monthly summary of our activities. For further publications by OHPI on hate against Indigenous Australians, see here.