Concept
Speech Raises the Bar
Vagueness
Speech Raises the Bar
The due process baseline asks two things of a penal law: can a person of ordinary intelligence tell what it forbids, and does it guard against arbitrary and discriminatory enforcement? When a law reaches expression, a third question joins them: does it leave enough breathing space for First Amendment rights?1 The Court answers by demanding more clarity from the legislature, and it grants broader relief when the legislature falls short.2
The three subsections below move from rule to application. The first states the heightened standard and explains it. The second shows laws failing because their operative word means only what the person applying it feels. The third shows rules failing because the line they draw between covered and uncovered speech dissolves when a speaker tries to find it. Vagueness is not overbreadth, though challengers often assert both. A statute can be quite specific and still reach too much, as a law forbidding “the display of a nude human body on a motion picture screen” would, and the next major section separates the two doctrines.3
Greater Specificity Required
Justice Powell stated the rule in Smith v. Goguen: “Where a statute’s literal scope, unaided by a narrowing state court interpretation, is capable of reaching expression sheltered by the First Amendment, the [vagueness] doctrine demands a greater degree of specificity than in other contexts.”4 The trigger is the statute’s literal scope. A court reads the words, asks what protected expression they could sweep in, and, if the answer is some, holds the drafters to the higher standard. The phrase “unaided by a narrowing state court interpretation” signals that an authoritative narrowing can change the answer, a point the later section on narrowing constructions develops.
Buckley v. Valeo restates the rule twice in reviewing the federal campaign-finance statute. Due process requires adequate notice to a person of ordinary intelligence, and where First Amendment rights are involved, “an even ‘greater degree of specificity’ is required,” a proposition the Court supported with Goguen, Grayned v. City of Rockford, and Kunz v. New York.5 Elsewhere the Court added that close examination of a statute’s specificity is required “where, as here, the legislation imposes criminal penalties in an area permeated by First Amendment interests,” citing Goguen, Cramp v. Board of Public Instruction, and Smith v. California. It framed the test as whether the language affords the “[p]recision of regulation [that] must be the touchstone in an area so closely touching our most precious freedoms.”6 Criminal penalties raise the stakes for a speaker who guesses wrong, and the First Amendment setting raises the cost of the guess.
A speaker who cannot see where a prohibition ends will stay well back from its edge. Baggett v. Bullitt, which invalidated a loyalty oath for teachers, made the point: vague statutes cause citizens to “steer far wider of the unlawful zone” than they would if the boundaries of the forbidden areas were clearly marked, “restrict[ing] their conduct to that which is unquestionably safe,” and “[f]ree speech may not be so inhibited.”7 Buckley, quoting an earlier decision, describes the same speaker as one who must “hedge and trim.”8 The result is a chilling effect. A vague statute, like an overbroad one, creates the risk that protected speech will be deterred, and courts respond with rulings of facial invalidity.9
The blur also serves whoever enforces the rule. A boundary the speaker cannot find is a boundary the enforcer can place after the fact, so the speech setting sharpens the enforcement ground along with the notice ground. The remedy follows the same logic. First Amendment vagueness challenges, like overbreadth challenges, typically produce facial invalidations, while a statute found vague as a matter of due process typically is invalidated only as applied to a particular defendant.10 The chilling effect likely explains the difference: a ruling that frees one defendant and leaves the text in force does nothing for the speakers who remain uncertain, and silent. A successful facial challenge usually invalidates the law entirely rather than as to certain applications, and the doctrine is available even to a person whose own speech is unprotected.11
Courts make that call by reading the text, as a question of law. Facial invalidation is the more sweeping use of judicial review, a power now firmly established even though the proper scope of judicial action under the Constitution remains a subject of continuing debate.12 Justice Douglas, dissenting in Dennis v. United States, urged the contrary on a different speech question: it would be a matter for the jury, he wrote, and the Court errs when it treats the question as one of law.13
The heightened standard follows the law’s effect on speech, not the label on the lawsuit. When there is no danger that a law will deter lawful speech, the Court is less strict in enforcing the vagueness requirement.14 In Village of Hoffman Estates v. Flipside, the Court upheld a statute requiring a business to obtain a special license if it sold items “designed or marketed for use with illegal drugs,” because it did not believe the statute deterred constitutionally protected speech.15 A retailer unsure whether a product is covered is deciding what to stock. A teacher unsure what a loyalty oath demands, as in Baggett, can stay safe only by saying less. The usual exam error is to treat the label “First Amendment case” as the trigger, when the real question is whether the law’s uncertainty gives people reason to give up expression.
Subjective Speech Standards
Some speech laws fail because their operative word has no content apart from the reaction of whoever applies it. Coates v. Cincinnati shows the pattern. The ordinance there made annoyance the test, and the Court explained the defect: “Conduct that annoys some people does not annoy others. Thus, the ordinance is vague, not in the sense that it requires a person to conform his conduct to an imprecise but comprehensible normative standard, but rather in the sense that no standard of conduct is specified at all. As a result, ‘men of common intelligence must necessarily guess at its meaning.’”18 That distinction organizes the whole pattern. An imprecise but comprehensible standard gives citizens and officers something to argue about and apply. A standard that means whatever the listener happens to feel gives them nothing.
A word like “annoying” fails both vagueness grounds at once. On the notice side, a speaker cannot predict whether some passerby will be bothered, so the speaker must guess. On the enforcement side, an ordinance that specifies no standard leaves the standard to whoever applies it. Speech sharpens the danger, because the speakers most likely to annoy are those who voice unpopular views, and a subjective standard reaches them first. Add the chilling effect from the previous subsection, and people with something to say stay home.
Smith v. Goguen applies the same reasoning to a single word. A state law made it a crime when any individual “publicly mutilates, tramples upon, defaces or treats contemptuously the flag of the United States,” and the defendant was convicted for sewing a small cloth replica of the flag to the seat of his pants. The Court held the law void for vagueness because of the inherent ambiguity in deciding what is “contemptuous” treatment of the flag.19 Contempt is an attitude, so the statute gave the citizen no fact to check in advance, and the enforcer had to infer the attitude from conduct.
A scienter requirement does not obviously repair a word like “contemptuous,” because the trouble lies in what the word means, not in whether the defendant knew what he was doing. The Court has sometimes reached the opposite view: it upheld an indictment under a statute punishing only knowing violation of an I.C.C. regulation, reasoning that the regulation applied only where the defendant knew of a safer route and deliberately took the more dangerous one. Justice Jackson, dissenting, argued that scienter, at least as traditionally defined, cannot cure vagueness in a statute or regulation.20 A later section on scienter requirements returns to the point.
Content-based regulation shows the same pattern in modern form. A law that regulates speech by the substance of its message is content specific, presumptively unconstitutional, and subject to strict scrutiny, while a content-neutral law receives intermediate scrutiny.21 In Reno v. ACLU, the Court struck down part of the Communications Decency Act, a statute built on the undefined terms “indecent” and “patently offensive.” A speaker deciding whether a web page is indecent has no definition to consult and can stay safe only by saying less than the law may permit. Content-based review asks a different question from vagueness: not whether the line can be found, but whether the government may draw it at all. New York’s “Son of Sam” law, challenged in Simon & Schuster, shows a line drawn by subject matter, since it required payment to the Crime Victims Board of any proceeds due to a person accused of, convicted of, or admitting to a crime for a book or other work describing the crime.22
Fighting-words laws show how much a drafting choice matters. A statute punishing verbal acts, carefully drawn so as not unduly to impair liberty of expression, is not too vague for a criminal law, as the Court said in Chaplinsky v. New Hampshire.23 Fighting words may be prohibited and offensive speech may not, and the line between them is ill-defined. One reason the Court has not clarified it is that the Court has increasingly used vagueness and overbreadth to invalidate disorderly conduct and breach of the peace statutes on their face, as in Gooding v. Wilson.24 A fighting-words law therefore survives only if it is specific and narrowly tailored to speech outside First Amendment protection; otherwise it is void for vagueness or invalid as overbroad.25
A subjective-sounding word does not doom a law on its own. The question is whether the law offers a standard that is imprecise but comprehensible or no standard at all, and whether anything beyond the enforcer’s reaction, such as a carefully drawn text or an authoritative narrowing construction, gives the word content.26
Blurry Lines in Speech Rules
Other speech rules fail because the line they draw between covered and uncovered speech dissolves in use. Buckley v. Valeo supplies the leading example. The expenditure ceiling in section 608(e)(1) limited “any expenditure . . . relative to a clearly identified candidate.” The Act defined “expenditure,” “clearly identified,” and “candidate,” but nothing defined which expenditures are “relative to” a candidate. The Court held that so indefinite a phrase fails to mark the boundary between permissible and impermissible speech unless other portions of the provision make the covered range sufficiently explicit.29 The provision carried criminal penalties in an area permeated by First Amendment interests, so the Court examined its specificity closely.
A looser reading would have left the deficiencies described in Thomas v. Collins, where the supposedly clear-cut distinction between discussion, laudation, general advocacy, and solicitation puts the speaker “wholly at the mercy of the varied understanding of his hearers.” Such a distinction, the Court said, “offers no security for free discussion” and “compels the speaker to hedge and trim.”30 That is a vagueness defect, not merely a burden. The speaker’s notice depends on a fact the speaker cannot control, namely how others will understand the words, and an official deciding afterward can draw the line wherever he prefers. Fair notice and guarded enforcement fail together, and the cautious speaker responds by saying less.
The Court found the cure in the communication’s own text. The deficiencies could be avoided only by reading section 608(e)(1) as limited to communications that include explicit words of advocacy of election or defeat of a candidate, much as the definition of “clearly identified” requires an explicit and unambiguous reference to the candidate.31 To preserve the provision against invalidation on vagueness grounds, the Court construed it to apply only to expenditures for communications that in express terms advocate the election or defeat of a clearly identified candidate for federal office. Express words work because they are a feature of the message the speaker writes. A speaker can read a draft, see whether it urges the election or defeat of a named candidate, and know in advance which side of the line it falls on, and an official reviewing the draft sees the same marker. The express-advocacy reading answered the vagueness objection only. The Court went on to hold that the ceiling unconstitutionally infringes First Amendment rights.32
A second provision, section 434(e), drew a similar objection. In its effort to be all-inclusive, the Court said, the provision “raises serious problems of vagueness, particularly treacherous where, as here, the violation of its terms carries criminal penalties and fear of incurring these sanctions may deter those who seek to exercise protected First Amendment rights.”33 Where definiteness is at stake, the Court also recognized an obligation to construe the statute, if that is consistent with the legislature’s purpose, to avoid the shoals of vagueness.34 Buckley’s per curiam opinion applied a separate test to compelled disclosure: the subordinating interests of the State must survive exacting scrutiny, with a substantial relation between the governmental interest and the information required to be disclosed.35 Vagueness is thus one inquiry among several, and a clear law can still fail another.
Gentile v. State Bar of Nevada shows the same blur inside a safe harbor. After a defense lawyer spoke to the press, the state bar reprimanded him for violating the extrajudicial publicity rule.36 Silencing orders directed to lawyers, which arise in contexts such as standing rules developed for the governance of the bar, are one alternative trial judges have turned to since Nebraska Press virtually invalidated gag orders against media personnel.37 The rule in Gentile contained a safe harbor allowing attorneys to describe the “general nature” of the defense “without elaboration.” The Court held the rule void for vagueness as applied to Gentile’s conduct, because the safe harbor does not provide sufficient guidance for lawyers trying to fit within it. It may mislead them into believing they cannot be prosecuted for publicly discussing possible defenses even if they should reasonably know the discussion might materially prejudice an adjudicative proceeding.38
A blurred safe harbor does special damage because a safe harbor invites reliance. A lawyer who plans a press statement around a promise to protect the “general nature” of the defense “without elaboration” cannot tell where a general description ends and elaboration begins, so following the rule offers no protection. Because a wrong guess brings professional discipline, the careful lawyer’s response is silence, which is the chilling effect again. An indeterminate edge also lets the disciplinary authority fix the line after the statement is made.
The standards are the same whether a rule is attacked on its face or as applied: it must give adequate notice of what conduct is forbidden and provide adequate enforcement standards. A rule vague on its face fails that test for any conduct, while a rule vague as applied fails only for specific conduct, and there may be conduct to which it can readily be applied.39 Gentile shows that relief in a speech case can be as-applied, so facial relief is typical but not automatic.
Rule 3.6(c) shows what a safe harbor with checkable limits looks like. It permits a lawyer to make “a statement that a reasonable lawyer would believe is required to protect a client from the substantial undue prejudicial effect of recent publicity not initiated by the lawyer or the lawyer’s client,” limited to “such information as is necessary to mitigate the recent adverse publicity.”40 The lawyer has two things to check before speaking: a trigger, which is adverse publicity that someone else started, and a purpose-based limit, which is what is necessary to mitigate it. The old safe harbor offered only a pair of elastic words. The new rule still uses reasonableness language, but that language operates on a defined trigger, and an imprecise but comprehensible standard of that kind is ordinarily tolerable, as a later section explains.
Express advocacy and the Rule 3.6(c) trigger work the same way. Each replaces a line that depends on how hearers react, or on an elastic term of degree, with a marker the speaker can find in the communication or in a defined event before speaking. An essay stem built on this pattern will describe a speech rule resting on a distinction, such as issue discussion versus electoral advocacy or a stated defense versus an elaborated one, and ask whether the rule survives. The usual exam error is to ask whether the government’s interest is strong enough, which is a different question from whether the line can be found.