[AnimalMuppet]

IANAL, and I'm not sure precisely where the lines are drawn, but I think at least this much is probably correct: If I use a racial epithet, or a derogatory word for someone's sexual preference or gender, that might be hate speech, but it's not a threat.

Moving on to shakier ground, I think the (US) rules are that if I even say "X should all be killed", that's definitely hate speech, but it's still not a threat.

If I'm talking to others, and I say "Go kill all the X" (or "Let's go kill all the X"), that's incitement to violence, but I'm still not sure that's a threat.

If I say "I'm going to kill you, you X", that's a threat.

As I said, IANAL. I welcome corrections from those who actually know this stuff...

[jfengel]

I believe you are correct, though I wanted to clarify something. I didn't mean that there was necessarily an overlap between threats and hate speech, but only that threats were another category of speech that was forbidden but was also vague.

There also happens to be some overlap, confusing the issue. But I don't know precisely on what grounds threats can be forbidden. I believe they should be, but the constitutional justification for that is unclear to me.

[schoen]

I'm not sure if this will help much, but the current U.S. constitutional law around this is based on this concept:

https://en.wikipedia.org/wiki/True_threat

I imagine one thing that people don't like about this doctrine is that a speaker might have the intention or effect (or both) of causing someone to feel afraid, but a court might conclude that this feeling was nonetheless unreasonable, and therefore the speech was protected.

That's a thorny problem because feeling frightened is a very serious kind of suffering, but courts don't want that possibility to swallow all of free expression doctrine (and generally don't want to give hearers a way to put a stop to someone's speech through their reactions alone https://en.wikipedia.org/wiki/Heckler%27s_veto).

Strangely, writing about this has made me think about https://repository.upenn.edu/cgi/viewcontent.cgi?article=123... and an idea that having one sender and two receivers (here, a person to whom the speech is directed, and a judge, assessing whether the speech contains a threat or not) allows for "attacks" based on the receivers' different interpretations of the same message. Maybe I'll have to think about that more and write something about it...

Edit:

Oh, also, by "such a category of unprotected speech", I didn't mean to say that there are no categories of unprotected speech in U.S. constitutional law, just that "hate speech" isn't one of them! So I think you might have been confused by an ambiguity in how I phrased that.

While there are several categories of unprotected speech in U.S. constitutional law, hate speech isn't one of them. A sort of analogy is that when various people tried to legally restrict minors' access to violent video games, opponents of this legislation successfully argued that there was no tradition in U.S. law of limiting the protection of speech because it contained explicit violence, as opposed to because it was sexually explicit. In that case the opponents agreed that there are prior concepts of "obscene as to minors" and "harmful to minors", but that these have only ever been held to relate to sexuality, not violence.

While this goes against a lot of people's intuitions, U.S. civil libertarians pretty uniformly celebrated it because the unprotected speech categories were successfully contained and kept from growing further.

[curryst]

IANAL either, but my understanding is that only speech that presents "imminent lawless action" is illegal [1]. The unlawful action being advocated must be both imminent and likely in order for the speech to be unlawful.

> If I'm talking to others, and I say "Go kill all the X" (or "Let's go kill all the X"), that's incitement to violence, but I'm still not sure that's a threat.

So in this case, there would need to be evidence that you are advocating to take that action on some kind of timeline (i.e. tomorrow, or next week, or next month) as well as evidence that it's likely to actually happen as a result of your speech. This probably wouldn't be prohibited because it's unlikely to actually happen. Lots of people spew some kind of discriminatory comments like this, but they're protected because they can reasonably claim that they thought it was unlikely anyone would actually do anything.

> If I say "I'm going to kill you, you X", that's a threat.

This is actually grey-er than you might think. There would be a big legal debate about whether this threat is "imminent" or not. I would guess it was deemed a threat, though. Now if you had said "I'm going to kill you one of these days, X", that would be protected because the threat isn't imminent. It's remarkably similar to the case that set the precedent, Hess v. Indiana [2]. TLDR; Vietnam War protesters were told to get off of a street, Hess uttered "We'll take the fucking street later" and was arrested. The court held that the speech was protected because although it advocated unlawful action, the threat of unlawful action was not imminent. If Hess had said "We'll take the street back tomorrow", it would not have been protected, because the threat was imminent.

I too would love input from someone more knowledgeable than me, though.

As an aside, I find it maddening that we need lawyers to figure out whether a particular action is actually illegal. Well, technically, it's worse than that, you actually need the Supreme Court to weigh in before you have a definitive answer. Lawyers can be wrong, and lower court judges can be overruled. I just can't fathom how we seem to be okay with not knowing whether a particular action is illegal unless someone has already done it, or you can somehow convince the Supreme Court to answer your hypothetical.

[1] https://en.wikipedia.org/wiki/Imminent_lawless_action [2] https://en.wikipedia.org/wiki/Hess_v._Indiana

[schoen]

> I would guess it was deemed a threat, though. Now if you had said "I'm going to kill you one of these days, X", that would be protected because the threat isn't imminent.

I think there's a difference between the "true threats" and "imminent lawless action" issues: that is, I think true threats are unprotected even if they aren't imminent. I think the imminent lawless action issue relates more to incitement issues, while the true threats issue relates more to, well, threats to harm specific people. But

> I too would love input from someone more knowledgeable than me, though.

... same here!