Just before Thanksgiving, the California Supreme Court rejected the opportunity to overturn a state “pronouns provision.” One section of a state law governing long-term care facilities for LGBT residents prohibits staff from “willfully and repeatedly fail[ing] to use a resident’s preferred name or pronouns after being clearly informed of the preferred name or pronouns,” if their refusal to do so was motivated “wholly or partially on the basis of a person’s actual or perceived sexual orientation, gender identity, gender expression, or . . . [HIV] status.”

On its face, the provision appeared to violate the First Amendment. So argued Taking Offense — no, not me or this blog but an advocacy group funded, in its words, to challenge efforts “to coerce society to accept [the] transgender fiction that a person can be whatever sex/gender s/he thinks s/he is, or chooses to be.” The organization sought a court order order to block enforcement. The trial court refused, but the state Court of Appeal reversed, agreeing that the pronoun provision was not narrowly enough tailored to withstand First Amendment concerns. In Taking Offense v. California, the state supreme court reversed again.

The California Supreme Court viewed the provision in light of its placement in a general law prohibiting discrimination in a state care facility. California has many such anti-discrimination provisions in a variety of settings, including general business establishments, housing, and residential care facilities for the elderly. The state legislature offered several findings to explain its purpose in enacting the LGBT facilities law, noting in particular the marginalization of elderly residents, many lacking family assistance, who require various services that by necessity must be performed by others, and pointing to data showing that “nearly one-half of the participants . . . reported experiencing discrimination in the prior 12 months because of the sexual orientation or gender identity.” More than 40 percent of residents responding to a recent survey reported witnessing or experiencing various types of mistreatment, including verbal or physical harassment by staff, refusal by staff to honor medical power of attorney given to a resident’s spouse or partner, discriminatory restrictions on visitation, and ”staff refusal to refer to a transgender resident by his or her preferred name or pronoun.” The legislative provisions governing health and safety in such residences covered each of the problems the legislature identified, including refusal to use preferred name or pronouns.

The trial court rejected Taking Offense’s claim that the pronoun provision “deprives long-term care facility staff of the ‘right to express offensive speech.’“ The provision applies strictly to speech and conduct within the facility and has no application outside. The trial court also rejected the assertion that the law “unconstitutionally compels and censors speech content, and imposes viewpoint discrimination.”

The appellate court disagreed, holding the pronoun provision to be content based. The court declared the provision fails the strict scrutiny test, requiring not only a compelling state interest (which the appellate court agreed had been demonstrated) but also a showing that it does not restrict more speech than necessary to accomplish the compelling interest — in this case, the elimination of discrimination in long-term care facilities. In the appellate court’s view, the pronoun provision “criminaliz[ed] occasional, off-hand, or isolated instances of misgendering, that need not occur in the resident’s presence and need not have a harassing or discriminatory effect on the resident’s treatment or access to care.”

In rejecting the appellate court’s holding in Taking Offense v. California, the California supreme court looked to the “strong presumption that an act of the Legislature is constitutional.” The case law has long been understood to require statutes to “be upheld as constitutional unless their invalidity, clearly, positively, and unmistakably appears.” These standards make it difficult for a facial challenge to succeed, because to overturn a law on its face, without pointing to a particular use in a factual setting requires showing that no matter the circumstances it will always (or at least in the “vast majority” of cases) run afoul of constitutional commands, in this case, of the First Amendment.

The supreme court concluded that it need not parse the pronoun provision under strict scrutiny, intermediate scrutiny, or overbreadth analysis — the usual tests for protected speech. Instead, the court said, the provision is best seen, in words offered in an amicus brief filed by the Lambda Legal Defense and Education Fund and other organizations, “not as a content-based regulation of protected speech, but instead as a regulation of discriminatory ‘conduct, which triggers no special constitutional scrutiny.’” The state’s highest court saw the pronoun provision as

address[ing] and operat[ing] in a narrow setting. It properly regulates discriminatory conduct aimed at vulnerable seniors who typically constitute a captive audience, residing in long-term care facilities that have become, in effect, their homes. The provision regulates the professional conduct of long-term care staff whose job is to provide and support medical treatment and intimate personal care — and seeks to promote an environment conducive to such care. It is carefully calibrated to achieve those ends, and does not preclude facility staff from expressing their views about gender to anyone (including a resident) in any otherwise lawful manner other than by misgendering a resident — and even then, the prohibition is limited to willful, repeated, knowing acts done because of a protected characteristic.

The court held that “the First Amendment presents no obstacle” to “anti-discrimination laws that prohibit acts that create hostile environments while only incidentally affecting speech.” In other words, when spoken words are used to carry out non-speech objectives that harm someone else, they may be viewed as conduct that can be regulated despite the First Amendment. As the U.S. Supreme Court said in a 1992 case (R.A.V. v. St. Paul), “‘it has never been deemed an abridgment of freedom of speech . . . to make a course of conduct illegal merely because the conduct was in part initiated, evidenced, or carried out by means of language, either spoken, written, or printed.” Otherwise, a “law against treason could not restrict “telling the enemy the Nation’s defense secrets.”

Taking Offense v. California may seem a narrow holding embracing a small portion of the population, but its holding points to a much larger application if other courts follow its logic. During the past two years, there has been an onslaught of commentary and criticism of American universities for tolerating or even perhaps encouraging antisemitic rhetoric on campus. At its broadest, the argument against seeking to deter or eliminate it is that, no matter how offensive, even the nastiest rhetoric is just that, an opinion of the speaker, and is protected by the First Amendment. This is true on a street corner, in a public park, in the opinion columns of newspapers and magazines, and in any other venue where people congregate.

Except, as we have come to understand, pervasive and offensive speech that serves to create a hostile environment can lead to discrimination against people who share protected characteristics, like race, religion, ethnicity, sex, and so on when they are subjected to harassment as a “captive audience.” This understanding is not new, as the U.S. Supreme Court made clear back in 1986 in Meritor Savings Bank v. Vinson, in holding that creating a hostile environment by harassing female employees in the workplace violates Title VII of the federal Civil Rights Act of 1964.

You can hear the loud voices already, protesting that however the captive-audience thesis plays out in certain spaces, like the workplace, it surely does not apply to American universities, which are the quintessential venues for free speech — expressing one’s mind on any topic that occurs, exchanging views and opinions, arguing over matters of truth and falsity and right and wrong — and without limit. You need not go far to find this position proclaimed and elaborated with force and conviction. Start here. Go wherever your search engine wants to take you. And realize that the discussion is not new.

For those who doubt the premise, it’s time to rethink. What student, being hectored in class by virtue of “membership” in a protected class, feels free to leave? When the yelling begins? When it reaches a crescendo? When it is explicit and unmistakable rather than implicit and mushy? To say that a student need not enroll in a particular course is no answer, because it is the very essence of discrimination that one is separated from the mainstream, forced to forgo the company of the majority. It may be wise policy to create designated campus spaces in which students and others can hurl insults at each other. It may be that the First Amendment protects such spaces against the argument that those who are insulted are subject to a hostile environment within the university community of which they are members. The courts have dealt and no doubt will continue to deal with these and related arguments. But in the near future, the debate should be centered on the hostility that individuals and mobs display towards particular people — students in classrooms, employees at work — who are entitled to the protection of the communities in which they seek a life and a living. Taking Offense v. California gives us the beginning of an answer.