Free Speech, Censorship, Moderation And Community: The Copia Discussion

from the not-an-easy-issue dept

As I noted earlier this week, at the launch of the Copia Institute a couple of weeks ago, we had a bunch of really fascinating discussions. I've already posted the opening video and explained some of the philosophy behind this effort, and today I wanted to share with you the discussion that we had about free expression and the internet, led by three of the best people to talk about this issue: Michelle Paulson from Wikimedia; Sarah Jeong, a well-known lawyer and writer; and Dave Willner who heads up "Safety, Privacy & Support" at Secret after holding a similar role at Facebook. I strongly recommend watching the full discussion before just jumping into the comments with your assumptions about what was said, because for the most part it's probably not what you think: Internet platforms and free expression have a strongly symbiotic relationship -- many platforms have helped expand and enable free expression around the globe in many ways. And, at the same time, that expression has fed back into those online platforms making them more valuable and contributing to the innovation that those platforms have enabled. And while it's easy to talk about government attacks on freedom of expression and why that's problematic, things get really tricky and really nuanced when it comes to technology platforms and how they should handle things. At one point in the conversation, Dave Willner made a point that I think is really important to acknowledge:

I think we would be better served as a tech community in acknowledging that we do moderate and control. Everyone moderates and controls user behavior. And even the platforms that are famously held up as examples... Twitter: "the free speech wing of the free speech party." Twitter moderates spam. And it's very easy to say "oh, some spam is malware and that's obviously harmful" but two things: One, you've allowed that "harm" is a legitimate reason to moderate speech and two, there's plenty of spam that's actually just advertising that people find irritating. And once we're in that place, it is the sort of reflexive "no restrictions based on the content of speech" sort of defense that people go to? It fails. And while still believing in free speech ideals, I think we need to acknowledge that that Rubicon has been crossed and that it was crossed in the 90s, if not earlier. And the defense of not overly moderating content for political reasons needs to be articulated in a more sophisticated way that takes into account the fact that these technologies need good moderation to be functional. But that doesn't mean that all moderation is good.

This is an extremely important, but nuanced point that you don't often hear in these discussions. Just today, over at Index on Censorship, there's an interesting article by Padraig Reidy that makes a somewhat similar point, noting that there are many free speech issues where it is silly to deny that they're free speech issues, but plenty of people do. The argument then, is that we'd be able to have a much more useful conversation if people admit:

Don't say "this isn't a free speech issue", rather "this is a free speech issue, and I’m OK with this amount of censorship, for this reason.” Then we can talk."

Soon after this, Sarah Jeong makes another, equally important, if equally nuanced, point about the reflexive response by some to behavior that they don't like to automatically call for blocking of speech, when they are often confusing speech with behavior. She discusses how harassment, for example, is an obvious and very real problem with serious and damaging real-world consequences (for everyone, beyond just those being harassed), but that it's wrong to think that we should just immediately look to find ways to shut people up:

Harassment actually exists and is actually a problem -- and actually skews heavily along gender lines and race lines. People are targeted for their sexuality. And it's not just words online. It ends up being a seemingly innocuous, or rather "non-real" manifestation, when in fact it's linked to real world stalking or other kinds of abuse, even amounting to physical assault, death threats, so and so forth. And there's a real cost. You get less participation from people of marginalized communities -- and when you get less participation from marginalized communities, you lead to a serious loss in culture and value for society. For instance, Wikipedia just has fewer articles about women -- and also its editors just happen to skew overwhelmingly male. When you have great equality on online platforms, you have better social value for the entire world.

That said, there's a huge problem... and it's entering the same policy stage that was prepped and primed by the DMCA, essentially. We're thinking about harassment as content when harassment is behavior. And we're jumping from "there's a problem, we have to solve it" and the only solution we can think of is the one that we've been doling out for copyright infringement since the aughties, and that's just take it down, take it down, take it down. And that means people on the other end take a look at it and take it down. Some people are proposing ContentID, which is not a good solution. And I hope I don't have to spell out why to this room in particular, but essentially people have looked at the regime of copyright enforcement online and said "why can't we do that for harassment" without looking at all the problems that copyright enforcement has run into.

And I think what's really troubling is that copyright is a specific exception to CDA 230 and in order to expand a regime of copyright enforcement for harassment you're going to have to attack CDA 230 and blow a hole in it.

She then noted that this was a major concern because there's a big push among many people who aren't arguing for better free speech protections:

That's a huge viewpoint out right now: it's not that "free speech is great and we need to protect against repressive governments" but that "we need better content removal mechanisms in order to protect women and minorities."

From there the discussion went in a number of different important directions, looking at other alternatives and ways to deal with bad behavior online that get beyond just "take it down, take it down," and also discussed the importance of platforms being able to make decisions about how to handle these issues without facing legal liability. CDA 230, not surprisingly, was a big topic -- and one that people admitted was unlikely to spread to other countries, and the concepts behind which are actually under attack in many places.

That's why I also think this is a good time to point to a new project from the EFF and others, known as the Manila Principles -- highlighting the importance of protecting intermediaries from liability for the speech of their users. As that project explains:

All communication over the Internet is facilitated by intermediaries such as Internet access providers, social networks, and search engines. The policies governing the legal liability of intermediaries for the content of these communications have an impact on users’ rights, including freedom of expression, freedom of association and the right to privacy.

With the aim of protecting freedom of expression and creating an enabling environment for innovation, which balances the needs of governments and other stakeholders, civil society groups from around the world have come together to propose this framework of baseline safeguards and best practices. These are based on international human rights instruments and other international legal frameworks.

In short, it's important to recognize that these are difficult issues -- but that freedom of expression is extremely important. And we should recognize that while pretty much all platforms contain some form of moderation (even in how they are designed), we need to be wary of reflexive responses to just "take it down, take it down, take it down" in dealing with real problems. Instead, we should be looking for more reasonable approaches to many of these issues -- not in denying that there are issues to be dealt with. And not just saying "anything goes and shut up if you don't like it," but that there are real tradeoffs to the decisions that tech companies (and governments) make concerning how these platforms are run.

Filed Under: censorship, community, copia, free expression, moderation, section 230
Companies: copia, copia institution