A couple of observations:
1. Justice Stevens’ dissent, joined by Justices Souter and Ginsburg, seems to endorse some restriction on pro-illegal-drug-use speech (and not just speech that falls in the narrow generally unprotected category of speech intended to and likely to incite imminent illegal conduct). It is “willing to assume that … the pressing need to deter drug use supports JDHS’s rule prohibiting willful conduct that expressly ‘advocates the use of substances that are illegal to minors,'” and treats it as an implication of the “unremarkable proposition” “that deterring drug use by schoolchildren is a valid and terribly important interest.” “[T]he First Amendment protects student speech if the message itself neither violates a permissible rule nor expressly advocates conduct that is illegal and harmful to students.” And while Justice Stevens says that he is only “assum[ing] for the sake of argument” that “Given that the relationship between schools and students ‘is custodial and tutelary, permitting a degree of supervision and control that could not be exercised over free adults,’ it might well be appropriate to tolerate some targeted viewpoint discrimination in this unique setting,” the tenor of all these statements strikes me as supporting, if grudgingly, restrictions on express advocacy of illegal drug use.
So it seems that everyone on the Court, with the possible exception of Justice Breyer, is willing to endorse some viewpoint-based restriction on some pro-drug speech (even outside the narrow and long-established incitement exception). Moreover, everyone with the exception of Justice Thomas and the possible exception of Justice Breyer, is willing to endorse what seems like a viewpoint-based First Amendment exception to do it. (Justice Thomas would say that pretty much all speech by K-12 students is unprotected from the government acting as educator, a viewpoint-neutral exception that allows a wide range of viewpoint-based speech restrictions.) The question is chiefly about the difference in the scope of the exception.
2. On the facts, Justice Stevens’ argument that Frederick’s speech wasn’t advocacy of drug use strikes me as quite weak. Here is what he argues (one paragraph break added):
To the extent the Court independently finds that “BONG HiTS 4 JESUS” [can most reasonably be interpreted as] the advocacy of illegal drug use … that conclusion practically refutes itself. This is a nonsense message, not advocacy. The Court’s feeble effort to divine its hidden meaning is strong evidence of that. Ante, at 7 (positing that the banner might mean, alternatively, “‘[Take] bong hits,'” “‘bong hits [are a good thing],'” or “‘[we take] bong hits'”). Frederick’s credible and uncontradicted explanation for the message — he just wanted to get on television — is also relevant because a speaker who does not intend to persuade his audience can hardly be said to be advocating anything. FN7 But most importantly, it takes real imagination to read a “cryptic” message (the Court’s characterization, not mine) with a slanting drug reference as an incitement to drug use.
Admittedly, some high school students (including those who use drugs) are dumb. Most students, however, do not shed their brains at the schoolhouse gate, and most students know dumb advocacy when they see it. The notion that the message on this banner would actually persuade either the average student or even the dumbest one to change his or her behavior is most implausible. That the Court believes such a silly message can be proscribed as advocacy underscores the novelty of its position, and suggests that the principle it articulates has no stopping point.
But it seems to me clear that the most plausible interpretation of Frederick’s speech is precisely that marijuana use is good, and should be engaged in. The alternative meanings mentioned by the majority, which Justice Stevens points to, are quite consistent with that; even just “[we take] bong hits” has as its “most probable” interpretation that “[we take] bong hits [and that’s good].” Ask high school students or recent high school graduates whom you know how they would likely interpret the banner. I asked three this morning, and all of them treated the statement as being generally pro-drug.
Doubtless the statement is ambiguous, as many short slogans are. For instance, it’s not clear whether the “4 Jesus” really tries to bring in a religious meaning. But if you’re looking for one “most plausible interpretation,” “drug use is good” is surely part of it.
As to Justice Stevens’ claim that the message is too “dumb” to be effective advocacy, I’m not sure: The theory even with most express advocacy is that the messages persuade when aggregated together, not that they will move many minds standing alone, and even jocular and “dumb” support of marijuana use may in the aggregate lead students to think of marijuana use as cool and fun. But more importantly, even dumb advocacy is still advocacy.
3. Perhaps, though, Justice Stevens should be arguing not that the speech isn’t advocacy, but that it’s not express enough to be advocacy — that only messages that expressly say “take bong hits,” “use marijuana,” “people should use marijuana,” and the like should be proscribable. “BONG HiTs 4 JESUS” definitely doesn’t seem express about anything much; and limiting the exception to express advocacy would indeed limit the exception’s scope.
At the same time, First Amendment law almost never draws distinction between express messages and pretty clear implicit messages, partly because any such distinction would either be very easy for speakers to evade, or attempts to prevent such evasion would eat away at whatever limitations the “express[ness]” requirement imposes. (Consider the related debates about express advocacy of election or defeat of a candidate in campaign finance law — a debate that I unfortunately don’t have the time to get into this weak, despite the fascinating-looking FEC v. Wisconsin Right to Life, Inc. decision.) And Justice Stevens doesn’t at all confront these issues, as it seems to me an effective argument for limiting the exception to truly express advocacy should.
4. Finally, note that the debate about whether First Amendment tests should focus on (1) whether speech could reasonably be interpreted as advocating illegal drug use, making false statements, making threats, and the like (the majority’s view, at least in this case) or on (2) whether speech “can most reasonably be interpreted” as advocating illegal drug use, making false statements, making threats, and the like is a very interesting one, and should get more attention than it has historically gotten. Courts must somehow determine what a statement “means,” but there are plenty of statements that have multiple plausible meanings, each of which is reasonable and none of which is the one “most reasonable” one.
I’ve tried to find good discussions of this in the libel context, where the factual problem of ambiguity often comes up. As I understand it, the dominant view is that the jury must find the reasonable meaning. A small minority of states (perhaps only Illinois, I think) take the view that so long as a reasonable meaning is innocent rather than defamatory, the statement must be found nondefamatory. But in practice, unless I’m mistaken, the question is often the opposite, which is whether a reasonable person could interpret the statement as defamatory (even if other reasonable people would take the opposite view). I’d love to hear from people who have researched the subject more thoroughly than I have, and have some good cases — likely libel cases — that discuss this.
Nonetheless, it seems to me that Justice Stevens erred in deciding that “BONG HiTs 4 JESUS” would “most reasonably be interpreted” as simply being a “nonsense message”; instead, the most reasonable interpretation is a message of vague support for marijuana use, the very interpretation that the majority adopted using its focus on what constitutes a reasonable meaning. Thus, this interesting debate (whether we look to the reasonable meaning, which is to say the most reasonable meaning, or a reasonable meaning, which is to say some meaning that at least a substantial minority would accept) isn’t fully implicated here.