Garrett Epps has an article in The Atlantic, "Who Pays the Price for the First Amendment?," which makes two interesting points. The first is flagged in the title. Protecting A's First Amendment rights often (always?) imposes some costs. Sometimes -- as in the classic sedition cases -- the costs are broadly distributed (across the entire population, which has to bear the ensuing risk of law-violation). Sometimes, as in Snyder v. Phelps and libel cases, the costs are concentrated (on the targets). Protecting A's rights also provides benefits, not only to A but to society generally, through the availability of information and ideas. In the "broad costs/broad benefits" configuration, there's not so much to worry about. But, should we treat the "concentrated costs/broad benefits" configuration differently? Existing doctrine doesn't, but maybe it should.

The other problem Epps exposes is about the heckler's veto. He accurately describes the consensus solution to the heckler's veto problem, which is that the government must devote extra police resources to ensure that the speaker isn't heckled down/assaulted. (As one of my students pointed out, this can also illustrate the "distributed costs" issue -- his car was broken into when police were diverted from their regular patrol to defending demonstrators against the possibility of a heckler's veto.) But, in the case he describes, the hecklers were part of a group that was itself exercising its First Amendment rights (to hold a civic celebration). And, as a result of the requirement that they put up with the hecklers, the organizers decided not to hold the celebration again. So, the people protected against a heckler's veto used their First Amendment rights to induce others not to exercise their First Amendment right. My guess is that there's a relatively simple solution to this "how to choose between competing hecklers?" problem, residing in the state action doctrine. But, I do think that Epps has identified an interesting puzzle.