Facebook, free expression and the power of a leak

SOCIAL media platforms aren't bound by the First Amendment, so it's important to understand their user-speech policies.

The First Amendment protects our right to use social networks like Facebook and Twitter, the Supreme Court said last week; the decision, which overturned a North Carolina law barring sex offenders from social networks, called social media "one of the most important places" for the exchange of views. It is a reminder of the role such networks play in our speech, access to information and, consequently, our democracy.

The government cannot block people from social media, but these private platforms can. So although Facebook recently updated its mission statement to pledge that it will give "people the power to build community and bring the world closer together", it also says in its mission statement that it is committed to removing "bad actors and their content quickly to keep a positive and safe environment".

Until recently, details of the types of posts it barred were a mystery - and Facebook was under no legal obligation to share.

That changed on May 21, when The Guardian released more than 100 pages of documents revealing Facebook's internal rules, a development that could make Facebook accountable to the public in its decisions about user speech.

Facebook has often been pressured to explain or alter its approach to moderating users' speech, in cases involving, for example, pictures of breastfeeding and Donald Trump's posts about barring Muslims from the US.

The leak thus gives some answers. Facebook's content policies resemble United States law - but they also have important differences.

For example, Facebook generally allows the sharing of animal abuse. But diverging from First Amendment law, Facebook will remove that same imagery if a user shows sadism.

Similarly, Facebook's manual on credible threats of violence echoes First Amendment law on incitement and true threats by focusing on the likelihood that violence will actually occur.

But there are also crucial distinctions. Where First Amendment law protects speech about public figures more than speech about private ones, Facebook does the opposite: If a user calls for violence, however generic, against a head of state, Facebook deems it a credible threat against a "vulnerable person".

It's fine to say: "I hope someone kills you"; it's not fine to say: "Somebody, shoot Trump." The government can't arrest you for saying it, but Facebook will remove the post.

These differences are to be expected. Courts protect speech about public officials because the Constitution gives them the job of protecting fundamental individual rights in the name of social values like autonomy or democratic self-governance. Facebook probably constrains speech about public officials because as a large corporate actor with meaningful assets, it can be pressured to cooperate with governments.

Unlike in the American court system, there's no due process on these sites; Facebook users don't have a way to easily appeal if their speech gets taken down. And unlike a government, Facebook doesn't respond to elections; instead, it acts in response to bad press, powerful users, government requests and civil society groups.

That's why the transparency provided by the Guardian leak is important. If individual users are to influence Facebook's speech governance, they'll have to know how this system works, just as citizens understand what the Constitution protects - and leverage that knowledge.

For example, before the Guardian leak, a private Facebook group, Marines United, circulated nude photos of female Marines and other women. This prompted a group called Not in My Marine Corps to pressure Facebook to remove related pages, groups and users. Facebook announced in April that it would increase its attempts to remove nonconsensual nude pictures.

But the Guardian leaks showed that the Marines United pictures were largely not covered by Facebook's substantive "revenge porn" policy. Advocates, using information from the leaks, are now pressuring Facebook to do more to prevent the nonconsensual distribution of private photos.

Civil liberties groups and user rights groups should do just this: Take advantage of the increased transparency to pressure these sites to create policies advocates think are best for the users they represent.

Today, as social media sites are accused of spreading false news, influencing elections and allowing horrific speech, they could respond by stepping up their policing of content. Clarity about their internal speech regulation is more important now than ever. NYTIMES