Note from the Editor: The Federalist Society takes no positions on particular legal and public policy matters. Any expressions of opinion are those of the author. We welcome responses to the views presented here. To join the debate, please email us at [email protected].

Narrow tailoring is a legal concept that requires a logical cause-and-effect relationship between the burden a law imposes on First Amendment rights and the goal the law seeks to achieve. Whether this relationship is present is determined through a means-ends test. Narrow tailoring is important because it holds government accountable to show that it made a serious effort to minimize any impact its regulation has on our rights.

For example, if a government wanted to ensure that drivers have adequate visibility, it might limit the application of stickers on the front windshield of a car to just the edges, leaving the center of the windshield clear. For some drivers, those stickers may be expressive, so limiting them would burden speech rights. While this approach might not be the only way of achieving good visibility—i.e. a driver may still be able to drive safely with a sticker in the middle of the windshield—there is a clear relationship between the means (no sticker in the middle of the windshield) and the ends (good visibility).

Narrow tailoring is a required element of the strict scrutiny and exacting scrutiny standards that courts apply to evaluate whether certain burdens on speech and association satisfy the First Amendment.

But narrow tailoring is sometimes misconstrued to mean narrow applicability. Not only does this approach get the law wrong, but it raises other constitutional concerns.

If rather than forbid stickers in the middle of the windshield, the law instead forbade any stickers on windshields of silver cars, that law would be narrowly applicable—affecting fewer stickers and drivers than the original rule—but it would not be narrowly tailored. It would be narrowly applicable because it only applies to a subset of cars: the silver ones. But unless there is some identifiable relationship between the color of the car and whether stickers on the windshield obstruct the driver’s view, the means-ends test has not been satisfied.

Clarifying this distinction is important because many courts have not been properly applying narrow tailoring; some judges have been convinced that narrow applicability is a lawful substitute for narrow tailoring. But narrowing the pool of potential targets for a law based on irrelevant characteristics raises its own constitutional issues.

For example, one test used to evaluate whether a limitation on speech or assembly is constitutional is to examine whether the law is overbroad or underinclusive. Narrow application fails both tests. Using the windshield example: if the sticker ban applied to all stickers on a silver car, then it is overbroad in prohibiting stickers that create no problem for visibility. It may sweep up parking permits in the lower corners, or oil change reminders in the upper corners, or college logos across the top, which have no meaningful effect on driver visibility. But it would also be underinclusive because it would leave the problem of visibility unresolved for cars of every other color even if it solved the issue for silver cars.

The problems with narrow applicability are even worse if the law allows for discretionary enforcement. If, for example, the sticker-enforcer has discretion to determine whether a car is subject to the law because he deems the car to be silver—rather than grey, white, ice blue, or platinum—the enforcer can sweep drivers in or out of coverage at-will. Thus the driver of a grey car could risk enforcement by displaying the logo of a college that prevailed in last week’s game against the enforcer’s favorite, while a platinum car with the enforcer’s preferred logo would be safe. These discretionary enforcements cannot hide behind a claim of narrow application when the relevant point is that they fail narrow tailoring.

Furthermore, where a fine for non-compliance with the sticker mandate can only be challenged by appearing in court or a similarly burdensome process, the logical way to avoid the burden would be to self-censor. If the bias of the enforcer is known, then self-censorship would be viewpoint-specific, applying only to supporters of the wrong college. But if the bias is not known, or is changeable depending on the mood, viewpoint, or employment status of the enforcer, the chill would be broader, applying to any car that could plausibly be deemed silver or any sticker that could refer to a college or college-adjacent entity.

The chill could go even further by dissuading people from buying or driving silver (or near-silver) cars at all. Choice of car color may or may not generally be considered expressive behavior—although it could be—but the unintended consequences of applying a law that depends on criteria that are decoupled from the law’s purpose and inflicted at the discretion of the enforcer could extend well beyond the safety concerns that sparked the law in the first place.

Here's a real world example of narrow applicability masquerading as narrow tailoring. Under the New Jersey Attorney General’s “power of inquisition,” the AG can “investigate merely on the suspicion that the law is being violated, or even just because [he] wants assurance that it is not.” In other words, the NJ AG has almost limitless discretion over whom to investigate—making viewpoint-based misuse of the investigatory power a risk. Ordinarily, such a risk can be managed through run-of-the-mill legal procedures. But when it comes to burdens on First Amendment rights, the chill imposed from threatened disclosure attaches immediately, and the procedures that usually shield against overreach come too late to protect against unconstitutional burdens on association.

An example of this First Amendment problem can be seen in First Choice Women’s Resource Centers, Inc. v. Platkin, which may be headed to the Supreme Court. The attorney general’s viewpoint on pregnancy centers is well-known; among other things, he’s published an open letter in which he pledged to take legal action against pregnancy centers that do not provide or refer for abortion.

In keeping with this pledge, he issued an investigatory subpoena demanding disclosure of donor information for thousands of donations to a charity that provides pregnancy support but not abortion. This type of demand, which would expose the identities of charitable donors without any allegation that the donors had violated the law in any way, falls squarely within AFPF v. Bonta and thus requires narrow tailoring before donors’ rights to anonymous association and speech can be overcome. In other words, there must be a means-ends justification for burdening the First Amendment rights of these donors at this time and in this way. And because narrow tailoring flows from constitutional protection, it must prevail over the policy preferences of the attorney general. But because narrow tailoring is not a factor in issuing a subpoena, there is no requirement that the investigatory subpoena rest on anything but the attorney general’s policy preference. Thus, donors to pregnancy centers that do not refer for abortion are at risk of exposure while donors to abortion providers are not, with no meaningful justification for why that is so.

Narrow tailoring, with stringent adherence to a means-end test, is critical to limiting burdens on speech or assembly to only those cases necessary to effectuate a compelling governmental purpose. Protecting First Amendment rights requires vigilance in rejecting laws that purport to be narrowly tailored, but that really only narrows their application by irrelevant characteristics while failing to satisfy the means-ends test that narrow tailoring requires.