Valve's content policy and the First Amendment

Back in May, Valve changed its content policy to crack down on visual novels containing adult content. Just a few weeks later, Valve again changed its policy to allow anything that isn't "illegal" or "straight-up trolling."

This led many developers and consumers down a rabbit hole of questions surrounding what Valve's role should be in curating or censoring content available through the Steam Store. It also led to criticisms arguing that Valve should do more to defend groups that might be offended by specific titles, and also that Valve should do more to defend content that might offend people (whether something is "trolling" is completely up to Valve).

Let's talk about the law stuff:

The First Amendment

When the May policy was announced, many declared it a violation of developers' First Amendment rights. These declarations were, of course, totally off base and incorrect.

The First Amendment is great. It just doesn't apply here. It says, in part: "Congress shall make no law... abridging the freedom of speech, or of the press."

One operative piece of the First Amendment is that opening: "Congress shall..." Congress has been interpreted by the courts to extend beyond just laws passed by Congress, but generally not so far as to apply to private companies. Valve is not the government. The Steam Store is not City Hall. There are some circumstances in which a private space becomes a sort of limited public sphere, but that doesn't really apply here.

So. developers have no First Amendment right to distribute games on Steam. The same is true of any other private platform.

The CDA & DMCA

There are two laws that create limited safe harbors for service providers, like Steam, hosting third-party content.

The first, Section 230 of the Communications Decency Act, gives a limited safe harbor from certain kinds of claims (like libel, for example) arising from content posted by users. The second, Section 512 of the Digital Millennium Copyright Act, gives a limited safe harbor from claims of copyright infringement arising from content posted by users.

Sites like YouTube and Facebook are classic examples. If YouTube and Facebook could be sued every time someone posted something that infringed a copyright to the platform, or every time someone wrote something libelous about their friends and neighbors on the platform, both would be buried in legal fees and judgments. Basically, the CDA and the DMCA allow the internet to function as it does.

Under both the DMCA and the CDA, a service provider can lose its safe harbor if it exercises too much control over the content. If it converts the speech or infringement from that of its users to that of its self. For example, if an online review site allowed anyone to submit reviews, but only posted those it approved after a content review, it would be in jeopardy of losing the CDA defense if one of those reviews later proved to be libelous.

From the most recent announcement regarding content policy changes:

It is possible that Valve swung the pendulum back in June in order to ensure it stayed within the safe harbors. I can't say that for sure, of course, and there's a really good chance their decision had nothing to do with this, but it also might have.

International Law

This post really just talks about the U.S. Valve accurately points out that "laws vary around the world," and says that they are "going to need to handle this on a case-by-case basis. As a result, we will almost certainly continue to struggle with this one for a while.

Just recently, for example, Valve disabled loot box item trading in CS:GO and Dota 2 because of Dutch regulators.

I suspect this will be an ongoing struggle. Finding the balance between embracing free expression and providing a quality, mostly inoffensive product is a difficult one. The complication of a multijurisdictional compliance need only muddies the waters further.

Enforcement

Valve has always had the right under their distribution agreement to boot a game off the platform for basically any reason. Nothing in the recent announcements has changed that.

However, it's worth paying attention to this specific sentence in their most recent announcement:

What this process will look like has yet to be seen. I would speculate, though, that it sounds like they are going to take an ESRB-like self-reporting tactic sometime in the future.