A couple months ago I wrote a post about a U.S. judge who ruled that online hate speech is physical harassment and I optimistically thought this would have an impression on people.

More recently a computer science student told me he refused to engineer social media content controls because he thought John Stuart Mill argued for absolutely unlimited speech.

I’m not sure where this crazy interpretation of Mill comes from or how, but it is very wrong (as I wrote back in 2011). That student was not the first to grab onto it.

A bad Mill reference deflated my recent optimism. The famous philosopher very clearly had a harm principle, widely discussed and documented; it really shouldn’t be hard to understand what Mill considered harmful and why he developed a principle to restrict speech.

It is these unfortunate misunderstandings of famous philosophers, let alone ignoring rulings by judges in actual courts, that surely leads to people saying their freedom of speech is some weird absolute.

Then again, maybe yet another U.S. judge can help here. Here’s a new ruling in the “hate state” of North Carolina that using speech for violent threats on Twitter can and will put people in jail.

“Social media allows you to share your views with the world in seconds, but it does not give you the right to threaten violence against others. The FBI stands ready to investigate whenever threatening language crosses the line to a crime,” said Special Agent in Charge Strong. According to filed court documents and evidence presented at trial, on March 13, 2018, Twitter user @DaDUTCHMAN5, later identified as Vandevere, used his social media account to send a message that contained a threat to injure an individual identified in court records as Q.R.

The defendant literally argued that sending a death threat wasn’t a threat because it was a political exchange of views while also not being a political view:

The tweet included a picture of a lynching and read, “VIEW YOUR DESTINY.” Vandevere argued his indictment must be dismissed on First Amendment free speech grounds, claiming the communication in question was not a “true threat.” […] “In 2019, the political arena necessarily includes the public exchange of political views that occurs daily on Twitter and other social media sites,” wrote his attorney, Andrew Banzhoff.

The judge wasn’t impressed by these weak arguments. Obviously telling a stranger their destiny is a lynching is a clear threat, and an exchange of political views is harmed by threats of violence.

The 4th U.S. Circuit Court of Appeals has held that “mere political argument, idle talk or jest” are not true threats, Cogburn noted. “However,” the judge added, “a true threat dressed up in political rhetoric or artistic expression alone does not render it a non-threat.”

What is especially interesting about this case is the victim claimed it was one of many similar threats on Twitter.

“It spikes any time there is (anti-Muslim) rhetoric from the political leadership of this country,” he said. “It’s almost predictable.”

This case therefore also provides credible evidence that federal anti-Muslim rhetoric has been increasing hate crimes, as already documented by researchers.

But what we think is interesting is that Trump’s tweets and hate crime only appear to be correlated after the start of his presidential run. It is also interesting that this correlation seems to be driven by areas with many Twitter users.

The above case is exactly what hate crime researchers have been warning about, given areas with a history of domestic terrorism (lynchings) and certain types of Twitter users:

…40 hate groups active in North Carolina. And many have argued that the 2015 murder of three Muslim students in Chapel Hill was fueled by Islamophobia, though the killer was not charged with a hate crime.