The Structural Weakness of Internet Speech

It is hard to look at the international protests surrounding the Innocence of Muslims video and the contemporaneous (though seemingly unrelated) fatal attack on the U.S. Consulate in Libya and not feel as though we are witnessing an important moment in the Internet's development.

Of course, posting material online has lead to drastic, even fatal, consequences in the past. But it is hard to think of another time where a single piece of online content has brought about such an overwhelmingly serious and negative reaction. And given that the creator's initial anonymity led news reports to declare the video as coming "from the Internet," it's quite possible that this will remain a video whose origin is attributed primarily by where it was published rather than who made it. In the minds of many, this will remain an action of the Internet – and an action with very serious consequences.

This leads to a fear. We now live in a world where one person is able to cause mass chaos by posting one video. Of course, that is not what happened; the video took nearly 100 people and thousands of dollars, and the environment is especially volatile in the post-revolution countries of Libya and Egypt. (The truth is always much more nuanced.) But I suspect that the question will be asked: Can we tolerate this? Should we censor Internet speech in order to avoid the harm to others, both religious and physical? When should free speech yield to other concerns?

This is a familiar fight to those who defend free speech, even under our uniquely American calibration of the concept. In America the answer is we almost always rule in favor of speech. Strong protection for free speech is a definitional character of our country, and what our Constitution commands. Faced with concerns of national security, personal safety, religious dignity, and even the right to a fair trial, we have consistently held that speech takes precedence.

But an under-appreciated fact about this truism is that it is equal parts substantive and structural. Yes, strong free speech is a fair reflection of many of our values. But also protecting free speech is the fact that it is really, really hard to amend our Constitution to say otherwise. "We the people" are empowered to except certain types of speech from constitutional protection, but we don't because this is very hard to do; it requires popular support in percentages unheard of in modern political fights. If the U.S. Constitution was as easy to amend as the California Constitution I strongly suspect that our free speech rights would be considerably weaker.

This is important to keep in mind when we find ourselves confronting controversial speech on the Internet. As Christian Science Monitor editor (and CMLP alum) Arthur Bright, EFF's (and Berkman alum) Jillian York, and the Washington Post's Craig Timbergeach noted last week, what happens online is subject to extensive regulation by a set of private parties across several different points of control. Many different private parties (e.g., the website, an ISP, and those maintaining the DNS system) all have the ability to censor speech online. And it is often forgotten – despite reminders from folks like Dawn Nunziato, Jillian York (again), and, through a different lens, Judge Kozinski – that these parties are not answerable to the First Amendment. Indeed, American law currently provides little oversight to their actions.

This is not to say that corporations don't care about free speech – we have seen that they do. And I suspect several free speech advocates will agree with Google's decision to take down the YouTube video in certain countries