What are the limits to the freedom of speech? A classic answer is Mill's harm principle: one has the freedom of speech to the extent that one is not harming another. Depending on how we think about harm in this context, the principle may be too restrictive. If merely offending someone suffices to harm them, than merely offensive speech will not be permitted, regardless of whether the offense is reasonable. This seems obviously too broad.
There's a way of understanding the harm principle that would take the paradigmatic case of harm to be specifically traumatic harm. Speech that merely offends, even if the offense is reasonable and indeed severe, would be protected, whereas speech that characteristically tends to result in traumatic harm would not be protected. Traumatic harm is a sufficient condition for rendering speech outside the protection of political liberties.
This version of Mill's idea-- call it the traumatic harm principle-- can be put to good use in jurisprudence. Consider Snyder v. Phelps, a U.S. Supreme Court case decided this week in an 8-1 ruling that picketing by the Westboro Baptist Church near the funeral service of a U.S. marine was protected by the First Amendment. (Read the New York Times article if you want to know what the signs said, I won't repeat the content here.)
I find it telling that both the opinion of the majority, and Justice Alito's dissenting opinion, reference the severe psychological toll the protests would have taken on the father of the deceased: "it is not necessary to allow the brutalization of innocent victims," writes Alito, while the majority emphasizes that the protests must have deepened the father's "incalculable grief." Though the justices do not appeal to the notion of trauma specifically, these remarks show an awareness of the traumatic effects of the picketer's actions. If we follow the traumatic harm principle I've proposed, we would have sufficient reason to deny this 'speech' constitutional protection.
Beyond this case in particular, there are other fruitful applications of our revised harm principle. Consider speech that we would classify as harassment, sexual or otherwise. This is a clear example of speech that is characteristically traumatic to its target. It is also typically offensive, but that doesn't suffice to explain why it should not be protected. The traumatic harm principle would license an exclusion of sexual harassment and other forms of traumatic speech as beyond the reach of political liberties.
While it would admittedly mark some degree of departure from contemporary First Amendment interpretation to rule that the picketers' speech is not protected in virtue of being traumatically harmful, it would be a move in the direction of a morally sound free speech jurisprudence. There are, of course, difficulties with my suggestion. What exactly constitutes traumatic speech? Must the speech be actually traumatic, or only the kind of speech that tends to be traumatic? These are important questions, though they are very much the same questions that present any account of the limits of free speech.