Why the Government Shouldn’t Force Bakers—Or Anyone—to Express Support for Same-Sex Marriage

The back-and-forth I described in yesterday’s article has become familiar, even tedious. Christian merchants—bakers, florists, wedding photographers—insist that they are not “anti-gay,” and that they have no objection whatsoever to serving LGBT customers. What they object to, rather, is providing a service that affirms the validity of same-sex marriage. Their critics respond that these merchants are not being asked to affirm or express anything, but merely to supply a cake, some flowers, or a few photographs. In rejoinder, the merchants emphasize the celebratory and expressive nature of a custom-designed cake, a floral arrangement, or an artfully-framed photograph—or of a wedding itself. The critics remain unpersuaded.

Is this the right debate to have? If it is, then the merchants would seem to have a strong case. Jack Phillips, the cake artist in the Masterpiece Cakeshop case currently pending before the Supreme Court, studied art and sculpting in preparation for his current calling. He carefully designs his cakes for particular occasions and clients. Even in ruling against him, the Colorado Court of Appeals noted that the same-sex couple did not walk into Phillips’s shop and ask to buy a cake off the shelf or even request that Phillips make a cake to preexisting specifications: they asked that Phillips himself, as the court put it, “design and create a cake to celebrate their same-sex wedding.”

Suppose we think not about cakes or flowers but more broadly about weddings, or marriage. Wasn’t it precisely because of the expressive dimension of marriage that the California Supreme Court explicitly and later the United States Supreme Court more diffusely ruled that states are required to recognize same-sex marriage and not merely some legally equivalent but expressively distinct form of civil union?

But then again, does any of this really matter? In a thoughtful essay entitled “Who Cares Whether Cake-Baking is Expressive?” NYU Professor Rick Hills argues that the expressive quality of Jack Phillips’s cakes should be constitutionally irrelevant. Appealing to writings of (then) Professor Elena Kagan and Yale Professor Jed Rubenfeld, Hills contends that what should matter for First Amendment purposes is “governmental purpose, not private burdens.” Although Hills sends mixed signals about the proper outcome in Masterpiece Cakeshop, the tone of his essay (and especially of his title) seems to suggest that, however expressive cake artistry or floral arranging may be, the merchants in these cases still have no good claim under the First Amendment’s freedom of expression.

But is this the logical implication of the “governmental purpose” approach advocated by Hills (and many others)? Or is the sound conclusion precisely the opposite—namely, that the complainants’ and the states’ efforts to compel Christian merchants to assist with same-sex weddings violate essential First Amendment commitments?

What Triggers the “Freedom of Expression”?

To appreciate Hills’s perspective, we should quickly review a few basics. The First Amendment provides that Congress (a term that has been extended by interpretation to include both federal and state governments generally) “shall make no law . . . abridging the freedom of speech.” Judges and scholars explain that “speech” means, basically, “expression.” The constitutional protection is not limited to the spoken word; it includes writings (even by people who would not normally be classified as “the press,” which the amendment separately names), and art, and some forms of expressive conduct—dance, for example, or flag burning.

Conversely, there are undeniably expressive utterances or activities that seem not to implicate the freedom of expression at all. Imagine a robber who says, “You can’t prosecute me: when I approached the victim and muttered ‘Your money or your life,’ I was obviously speaking.” The robber’s defense will go nowhere in court: it will not even trigger a balancing of the government’s interest against the robber’s interest in free speech. Or think of a protestor who spray paints “THE MAYOR SUCKS” in large red letters across the face of City Hall. The protestor is undoubtedly expressing himself, but he will get no help from the First Amendment when he is prosecuted for vandalism.

But then how do we decide when the freedom of expression kicks in and when it doesn’t? Judges and advocates have tried a variety of approaches. One approach, evident in a number of Supreme Court decisions but rejected by Professor Hills and others, tries to make headway by distinguishing between “expression” and “conduct.” Justice Hugo Black would likely have said that the robber and the vandal get no constitutional protection because they were mostly acting, not merely speaking. The “expression versus conduct” approach will tend to ask questions like “Was the actor/speaker intending to express something?” and “Would onlookers typically perceive that a message was being expressed?”

Theorists, however, have tended to dismiss the “expression/conduct” distinction. All expression involves conduct, they may say—the moving of lips, the pressing of fingers on a keyboard—and, conversely, all conduct at least implicitly expresses something. In this spirit, Professor Hills dismisses the “expression/conduct” distinction as a piece of “metaphysical silliness.”

But if the “expression/conduct” distinction is empty, and if the speaker’s intention to express or the onlooker’s perception of expression accordingly counts for little, then what should govern constitutional coverage? A common response of theorists—which finds considerable support in the actual caselaw—contends that what matters is the government’s purpose in regulating. Is government regulating something because of what it expresses? If so, the government’s expression-oriented purpose triggers full First Amendment scrutiny. Conversely, if government is regulating to secure interests unrelated to expression, then the First Amendment does not kick in, even though the subject of regulation is undeniably expressive in character.

The examples mentioned above can illustrate the idea. Government goes after the robber because he stole money; it matters not whether the robber accomplished this end through a spoken threat or by simply grabbing the cash. Government prosecutes the protester-vandal because he defaced government property; the fact that the vandal happened to be criticizing the mayor is irrelevant to the government’s purpose in prosecuting.

There is much to be said for (and against) this “governmental purpose” approach to understanding the freedom of expression. In any case, the approach enjoys considerable support among both theorists and jurists. And it is this approach that informs Professor Hills’s criticism of the arguments and amicus briefs that emphasize the expressive dimension of designing and creating wedding cakes. Perhaps wedding cakes are profoundly expressive. Who cares? For the “governmental purpose” approach, that fact is not decisive. What matters, rather, is why the government is regulating cake bakers under antidiscrimination laws.

And so we have to ask: why is government (in collaboration with and on behalf of same-sex couples) going after the bakers and florists and photographers in the marriage cases?

The Centrality of Expression

The question takes us back to the argument of yesterday’s essay. We saw there that in the litigated cases, the states and the same-sex complainants have not primarily relied on the contention that a Christian merchant’s refusal to assist with a same-sex wedding has deprived the couple of any needed product, service, or opportunity. In Masterpiece Cakeshop, another baker supplied complainants with a wedding cake for free; in Arlene’s Flowers, the case of the Washington florist that is currently on appeal to the Supreme Court, the same-sex couple claimed and received $7.91 in damages for the cost of driving to another florist. That was not why the couple and the state brought the lawsuit.

So why were these suits brought? Advocates are often forthright in explaining that these cases are not mainly about material deprivations, which are likely negligible or nonexistent, but rather about the “dignitary harm” or offense suffered when a same-sex couple is in effect told that a merchant regards their marriage as morally wrong or contrary to God’s will.

That claim may be perfectly sincere. But it amounts to a complaint that the couples feel injured by the communication of a message of disapproval. The injuries, in short, are primarily or exclusively expressive in nature. And the remedies sought by the plaintiffs and granted by the courts have likewise been expressive in their content and purpose. Objecting merchants have been ordered to assist with same-sex weddings in the future—not because their services are needed, but because complainants and the states seek to compel them to participate in, to borrow language from the Colorado court, “celebrat[ing] . . . same-sex wedding[s].”

Far from deflecting freedom of expression issues, therefore, the “governmental purpose” approach advocated by Hills and others actually serves to underscore why the marriage-merchant cases do trigger full First Amendment scrutiny.

Put it this way: if you believed that the scope of the freedom of expression should be governed by the “expression/conduct” distinction, there would be room to argue about which side of the line a wedding cake falls on. But if, like Professor Hills, you think that what really matters is the government’s purpose, then the conclusion is hard to avoid. The complainants and the state are not trying to remedy any material deprivation. The whole point, once again, is to compel vendors to acquiesce in performing a service that everyone concerned—the government, the plaintiffs, the vendors—cares about mainly or exclusively because of what it expresses.

Discriminating Discrimination

The “governmental purpose” approach also serves to distinguish the marriage cases from other more familiar applications of antidiscrimination cases. Typically, as I noted yesterday, the primary purpose of antidiscrimination law and litigation has not been to regulate or compel expression, but rather to remedy the denial of needed goods, services, or opportunities.

To be sure, application of antidiscrimination laws in the typical case might also have an impact on expression. In refusing to serve black customers, the owner of Ollie’s Barbecue may well have been expressing some conviction, or at least some prejudice. Who cares? (As Professor Hills might ask.) Under the “governmental purpose” approach, if the purpose of antidiscrimination law and litigation was, as the Supreme Court explained in Katzenbach v. McClung, to remedy a situation in which African-Americans faced serious difficulties in finding food and lodging when traveling in entire regions of the country, the First Amendment would not be implicated. So even if there was an inhibiting effect on Ollie’s expression, relative to the government’s purpose, that effect was incidental.

In the current marriage cases, by contrast, expression is hardly incidental. On the contrary: as we have seen, expression and messages are the main or sole reason why the cases are brought. That purpose fully implicates the First Amendment and, more specifically, the celebrated declaration in West Virginia State Board of Education v. Barnette that “[i]f there is any fixed star in our constitutional constellation, it is that no official, high or petty, can prescribe what shall be orthodox in politics, nationalism, religion, or other matters of opinion, or force citizens to confess by word or act their faith therein.” Even if the prescription and compulsion are presented in the beguiling language of “antidiscrimination.”

Unconstitutional as Applied

Application of the “governmental purpose” analysis to antidiscrimination laws might, to be sure, result in a mildly awkward conclusion. The analysis suggests that the same antidiscrimination law could be constitutionally unproblematic in most cases—namely, where the main purpose is to remedy the denial of goods, services, or opportunities—but not in other cases, where the main purpose is to redress the “dignitary harm” of offense incurred from the expression of another citizen’s constitutionally protected beliefs.

But this sort of awkwardness is scarcely a stranger in the law. Lawyers and judges are perfectly familiar with the idea that a law may be constitutional in some of its applications but not in others. So there is nothing anomalous about the idea that antidiscrimination laws are fully within the government’s authority—but only when the government is not using such laws as part of a campaign to compel people to express “by word or act” their support for a government-prescribed orthodoxy. Unfortunately, that is precisely what is going on in the marriage cases.

Professor Smith assisted in writing and submitting an amicus brief in the Masterpiece Cakeshop case, from which portions of this essay have been adapted.