The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Free Speech

Ban on "Political" Clothing in Polling Places Is Unconstitutionally "Unreasonable"

But a more precise ban on clothing in polling places (which the Court treats as "nonpublic fora") that mentions candidates, parties, or ballot measures may be constitutional.


In Minnesota Voters Alliance v. Mansky, the Supreme Court today struck down a ban on people, including ordinary voters, wearing any "political badge, political button, or other political insignia" in a polling place. Here's the Court's reasoning:

[1.] The inside of a polling place is a "nonpublic forum"—properly owned (or, for some polling places, temporarily borrowed for the day) and controlled by the government. In most places, such as on private property or on "traditional public fora" such as streets and sidewalks and parks, the government may not impose content-based restrictions (unless the restrictions fall within a few narrow exceptions, such as for "fighting words" or "true threats" or libel). But in nonpublic fora, content-based restrictions are fine if they are (a) viewpoint-neutral and (b) reasonable.

[2.] Content-based restrictions on speech in polling places may well be reasonable, because they are aimed at making voting "an island of calm in which voters can peacefully contemplate their choices," which helps keep voting as "a weighty civic act," "a time for choosing, not campaigning." The Court spoke favorably of clear restrictions on speech that mentions parties or candidates (or ballot measures).

[3.] But this ban, which isn't limited to speech that mentions, say, political parties or candidates, is not reasonable:

[T]he statute prohibits wearing a "political badge, political button, or other political insignia." It does not define the term "political." And the word can be expansive. It can encompass anything "of or relating to government, a government, or the conduct of governmental affairs," Webster's Third New International Dictionary 1755 (2002), or anything "[o]f, relating to, or dealing with the structure or affairs of government, politics, or the state," American Heritage Dictionary 1401 (3d ed. 1996). Under a literal reading of those definitions, a button or T-shirt merely imploring others to "Vote!" could qualify.

The State argues that the apparel ban should not be read so broadly. According to the State, the statute does not prohibit "any conceivably 'political' message" or cover "all 'political' speech, broadly construed." Instead, the State interprets the ban to proscribe "only words and symbols that an objectively reasonable observer would perceive as conveying a message about the electoral choices at issue in [the] polling place."

At the same time, the State argues that the category of "political" apparel is not limited to campaign apparel. After all, the reference to "campaign material" in the first sentence of the statute—describing what one may not "display" in the buffer zone as well as inside the polling place—implies that the distinct term "political" should be understood to cover a broader class of items. As the State's counsel explained to the Court, Minnesota's law "expand[s] the scope of what is prohibited from campaign speech to additional political speech." …

[F]ar from clarifying the indeterminate scope of the political apparel provision, the State's "electoral choices" construction introduces confusing line-drawing problems. For specific examples of what is banned under its standard, the State points to the 2010 Election Day Policy—which it continues to hold out as authoritative guidance regarding implementation of the statute. The first three examples in the Policy are clear enough: items displaying the name of a political party, items displaying the name of a candidate, and items demonstrating "support of or opposition to a ballot question."

But the next example—"[i]ssue oriented material designed to influence or impact voting"—raises more questions than it answers. What qualifies as an "issue"? The answer, as far as we can tell from the State's briefing and argument, is any subject on which a political candidate or party has taken a stance. For instance, the Election Day Policy specifically notes that the "Please I. D. Me" buttons are prohibited. But a voter identification requirement was not on the ballot in 2010, so a Minnesotan would have had no explicit "electoral choice" to make in that respect. The buttons were nonetheless covered, the State tells us, because the Republican candidates for Governor and Secretary of State had staked out positions on whether photo identification should be required.

A rule whose fair enforcement requires an election judge to maintain a mental index of the platforms and positions of every candidate and party on the ballot is not reasonable. Candidates for statewide and federal office and major political parties can be expected to take positions on a wide array of subjects of local and national import. Would a "Support Our Troops" shirt be banned, if one of the candidates or parties had expressed a view on military funding or aid for veterans? What about a "#MeToo" shirt, referencing the movement to increase awareness of sexual harassment and assault? At oral argument, the State indicated that the ban would cover such an item if a candidate had "brought up" the topic.

The next broad category in the Election Day Policy—any item "promoting a group with recognizable political views"—makes matters worse. The State construes the category as limited to groups with "views" about "the issues confronting voters in a given election." The State does not, however, confine that category to groups that have endorsed a candidate or taken a position on a ballot question.

Any number of associations, educational institutions, businesses, and religious organizations could have an opinion on an "issue[] confronting voters in a given election." For instance, the American Civil Liberties Union, the AARP, the World Wildlife Fund, and Ben & Jerry's all have stated positions on matters of public concern. If the views of those groups align or conflict with the position of a candidate or party on the ballot, does that mean that their insignia are banned?

Take another example: In the run-up to the 2012 election, Presidential candidates of both major parties issued public statements regarding the then-existing policy of the Boy Scouts of America to exclude members on the basis of sexual orientation.6 Should a Scout leader in 2012 stopping to vote on his way to a troop meeting have been asked to cover up his uniform? The State emphasizes that the ban covers only apparel promoting groups whose political positions are sufficiently "well-known." But that requirement, if anything, only increases the potential for erratic application. Well known by whom? The State tells us the lodestar is the "typical observer" of the item. But that measure may turn in significant part on the background knowledge and media consumption of the particular election judge applying it.

The State's "electoral choices" standard, considered together with the nonexclusive examples in the Election Day Policy, poses riddles that even the State's top lawyers struggle to solve. A shirt declaring "All Lives Matter," we are told, could be "perceived" as political. How about a shirt bearing the name of the National Rifle Association? Definitely out. That said, a shirt displaying a rainbow flag could be worn "unless there was an issue on the ballot" that "related somehow … to gay rights."

A shirt simply displaying the text of the Second Amendment? Prohibited. But a shirt with the text of the First Amendment? "It would be allowed."

"[P]erfect clarity and precise guidance have never been required even of regulations that restrict expressive activity." But the State's difficulties with its restriction go beyond close calls on borderline or fanciful cases. And that is a serious matter when the whole point of the exercise is to prohibit the expression of political views.

It is "self-evident" that an indeterminate prohibition carries with it "[t]he opportunity for abuse, especially where [it] has received a virtually open-ended interpretation." Election judges "have the authority to decide what is political" when screening individuals at the entrance to the polls. We do not doubt that the vast majority of election judges strive to enforce the statute in an evenhanded manner, nor that some degree of discretion in this setting is necessary. But that discretion must be guided by objective, workable standards. Without them, an election judge's own politics may shape his views on what counts as "political." And if voters experience or witness episodes of unfair or inconsistent enforcement of the ban, the State's interest in maintaining a polling place free of distraction and disruption would be undermined by the very measure intended to further it.

[4.] Justice Sotomayor, joined by Justice Breyer, dissented, arguing that the Court should have certified the case to the Minnesota Supreme Court, to give it an opportunity to issue a clarifying interpretation of the statute "which likely would obviate the hypothetical line-drawing problems that form the basis of the Court's decision today." (For the Court's response to this point, and the dissent's rebuttal, see the opinion; I omit here because it's too tangential to the free speech question.)

[5.] A broader takeaway for future cases: The Court has in recent years somewhat cut back on the void-for-vagueness doctrine, by insisting that courts (even in free speech cases, where the doctrine is stronger than in others) should generally "consider whether a statute is vague as applied to the particular facts at issue, for '[a] plaintiff who engages in some conduct that is clearly proscribed cannot complain of the vagueness of the law as applied to the conduct of others.'" If the law clearly applies to your speech, you can't challenge it as vague even if it's vague as applied to a large zone of other speech. And this means that laws generally can't be challenged as vague on their face (the way they can be challenged as overbroad), unless they are so vague as to be generally indeterminate.

But vagueness arguments can still be made within other First Amendment doctrines, and can thus prevail in facial challenges (such as this one). We saw that in Reno v. ACLU (1997), where the vagueness of terms in the Communications Decency Act led the Court to conclude that the Act was unconstitutionally overbroad. We see that here, where the vagueness led the Court to conclude that the law was unreasonable. The same logic could lead to a conclusion that the law is not narrowly tailored to a government interest, and thus fails intermediate or strict scrutiny in the many cases where such scrutiny is applied. Expect more such challenges to vague statutory terms in First Amendment cases.