Concept
When Conduct Speaks
Regulation of expressive conduct
Symbolic Speech and Expressive Conduct
When Conduct Speaks
The First Amendment bars Congress from “abridging the freedom of speech, or of the press,” and a literal reader might stop at spoken and printed words. The Supreme Court has not. In Texas v. Johnson, Johnson burned an American flag at a protest during the 1984 Republican National Convention and was convicted under a Texas law punishing flag desecration. The Court reversed, holding that government may not prohibit the expression of an idea simply because society finds the idea itself offensive or disagreeable.1 The burning was itself the message; no speech or pamphlet was needed to carry it.
That protection has an edge, because almost any act can be performed with an idea in mind. Puerto Rican Nationalists who tried to assassinate President Truman doubtless meant to call attention to the cause of Puerto Rican independence. The attempt was not First Amendment speech, and the government's interest in punishing attempted murder would override any First Amendment claim.2
Every expressive-conduct problem therefore raises two questions: when conduct should be analyzed under the First Amendment at all, and, once it should, what test decides whether the conduct is protected.3 The first question comes first for a practical reason. If every defendant could invoke the First Amendment by claiming that his or her conduct was meant to communicate, virtually every criminal law would have to meet strict scrutiny.4 Conduct that does qualify as expression receives the full range of First Amendment protections, and the more conventional the method, the more likely a court is to see the expressive component.5 The three subsections below answer the first question in turn: which nonverbal acts the Court has protected, how it sorts expressive conduct from other conduct, and where conduct fails that sorting.
Nonverbal Expression Protected
The Court has long recognized that First Amendment protection does not end at the spoken or written word.6 By the time of Texas v. Johnson it had treated a range of nonverbal acts as expressive: wearing black armbands to protest the Vietnam War, staging a sit-in against racial segregation, wearing military uniforms in a play critical of the war, picketing for many causes, and several acts involving flags.7 Those decisions arrived one at a time, and each added a reason why a nonverbal act can be speech. Taken in order, they show how the category was built and why it needs a limit.
The earliest, Stromberg v. California (1931), struck down a California statute that made it a crime to display a red flag “as a sign, symbol or emblem of opposition to organized government.” The statute was so vague that it allowed punishment for the fair use of “the opportunity for free political discussion.”8 Had protection stopped at words, a state could have silenced dissent simply by outlawing its emblem.
In West Virginia State Board of Education v. Barnette (1943), West Virginia required schoolchildren to salute the flag, and children who were Jehovah's Witnesses refused for religious reasons. The Court decided the case on free speech grounds. It explained that using a flag or emblem to stand for a system, an idea, or an institution is “a short cut from mind to mind,” so a flag salute is a form of utterance, and a compulsory salute forces the individual to communicate acceptance of the political ideas the flag represents by word and sign.9 No official, the Court said, “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.”10 Because the reasoning did not turn on the children's religious views, it protects anyone who declines to join in a state-required expression, whatever the reason.11 Stromberg thus protected a person who displays a symbol, and Barnette protects a person who refuses to display one.
Later decisions carried the same reasoning beyond flags. In Brown v. Louisiana (1966), which arose from a peaceful sit-in at a segregated public library, Justice Fortas wrote an opinion that did not command a majority. He said that First Amendment rights “are not confined to verbal expression” and include “the right in a peaceable and orderly manner to protest … unconstitutional segregation of public facilities.”12 Tinker v. Des Moines Independent Community School District (1969) offers the cleaner example of a symbol worn rather than a place occupied. The Court held that wearing a black armband to protest the Vietnam War was “the type of symbolic act that is within the First Amendment,” “closely akin to ‘pure speech.’”13 The case arose in a public school, one of the government settings where the Court generally defers to regulation of expression, and the Court still protected the armbands.14 They qualified because anyone who saw them, in the context of the times, would understand them as a symbol of protest against the war.15
Two further arguments explain why such acts count as speech and not merely as conduct. Words are themselves symbols, and the Court observed that the flag, full of expressive content, signifies the nation as readily as the letters of the word “America” do, so a rule that protected the word but not the flag would draw its line arbitrarily.16 History points the same way. One scholarly argument notes that Framing-era English and American political culture was rich with symbolic expression, used interchangeably with words, including burning effigies, wearing cockaded hats, and raising “liberty poles.”17
The same decisions that protect symbols also explain why the category needs a boundary. Stromberg and Barnette created no test for which expressive conduct, in which circumstances, the First Amendment protects, and the Court began to set boundaries in a draft card burning case.18 In United States v. O'Brien (1968), O'Brien burned his draft registration certificate and argued that freedom of expression includes all modes of “communication of ideas by conduct,” so that his act, done in demonstration against the war and the draft, was protected symbolic speech.19 The Court answered: “We cannot accept the view that an apparently limitless variety of conduct can be labeled ‘speech’ whenever the person engaging in the conduct intends thereby to express an idea.” Even assuming a communicative element in the burning, the Court added, protection did not necessarily follow, because when speech and nonspeech elements are combined in the same course of conduct, a sufficiently important governmental interest in regulating the nonspeech element can justify incidental limitations on First Amendment freedoms.20 That second step belongs to the O'Brien standard, taken up later in this chapter.
The limit is practical. Almost any act can carry some private message in the actor's mind, so a test that turned on intent alone would let every criminal defendant demand First Amendment scrutiny of an ordinary law by announcing what the act was meant to say.21 Johnson later put both halves of the rule together. Conduct may be “sufficiently imbued with elements of communication” to fall within the First Amendment, but the Court has not automatically treated every action involving a flag as expressive, and it looks instead to the context in which the act occurred.22
An exam fact pattern will often describe a defendant who broke an ordinary law and call the act a protest. The first move is to ask whether the act itself, in its setting, conveys a message to those who see it. The next section turns that question into a two-part inquiry.
The motorist's wish to be understood matches the wish of the Tinker students. What differs is the act: an armband worn during the Vietnam War carried its message to everyone who saw it, and the speeding carried none.
The Spence Inquiry
To decide whether Johnson's burning was expressive conduct, the Court asked two questions drawn from Spence v. Washington: whether an intent to convey a particularized message was present, and whether the likelihood was great that those who viewed the conduct would understand the message.24 The two halves do different work. The first looks to the actor and ties protection to a genuine communicative purpose, which answers the worry in O'Brien that nearly any act would qualify on the actor's say-so. The second looks to the audience and screens out private gestures that no onlooker could read as a message, however sincerely meant. Both must be met. A sincere message that no viewer could decode fails the second half, and an act that viewers might read as pointed, but that the actor never meant as a message, fails the first.
Spence shows the halves working together. He hung a flag he owned upside down in his apartment window, with a peace symbol made of removable tape on both sides, to protest the invasion of Cambodia and the killings at Kent State University.25 The Court stressed that he taped the flag at roughly the same time as, and concededly because of, those events, and it noted that Washington had conceded the display was a form of communication.26 The timing supported both halves. It made Spence's claimed message believable, and it made the message easy to grasp: it would have been difficult, the Court said, “for the great majority of citizens to miss the drift of appellant's point at the time that he made it.”27 Flags have also traditionally been used and accepted as a form of communication, and Spence's use fit that tradition.28
Johnson shows the same inquiry as the first step of a larger analysis. Before asking anything about the State's interest, the Court had to decide whether the burning was expressive conduct that permitted Johnson to invoke the First Amendment. Only if it was would the Court ask whether Texas's regulation was related to the suppression of expression.29 The burning ended a demonstration called the Republican War Chest Tour, held during the convention in Dallas. Marchers chanted political slogans and staged die-ins at corporate locations to dramatize the consequences of nuclear war. In front of City Hall, Johnson unfurled an American flag, doused it with kerosene, and set it on fire while the others chanted, “America, the red, white, and blue, we spit on you.”30
Run on Johnson's facts, the first two moves come from his own testimony. He said that he burned the flag as Ronald Reagan was being renominated, that no more powerful statement of symbolic speech could have been made at that time, and that the act juxtaposed new patriotism with no patriotism.31 The setting answers the rest. The burning culminated a political demonstration that coincided with the convening of the Republican Party, and the Court found that “[t]he expressive, overtly political nature of this conduct was both intentional and overwhelmingly apparent.”32 The Texas Court of Criminal Appeals had reached the same view from the observer's side: given an organized demonstration, speeches, slogans, and distributed literature, anyone who saw the act would have understood its message.33 Texas also conceded that the burning was expressive, as Washington had conceded in Spence.34 An exam stem will not hand you a concession, so write the four moves out rather than assuming them.
The same national symbol can fall on the other side of the line when the circumstances show no message. Johnson acknowledged that a tired person might drag a flag through the mud, knowing that others would be offended but with no thought of expressing any idea, and it set aside the prosecution of such a person as a different case from the one before it.35 Contact with a flag appears in Spence, in Johnson, and in the muddy-flag example alike, but only in the first two do the circumstances show a particularized message the actor meant to send. The first half can therefore fail on its own, and so can the second.
Timing and accompanying speech filled the second half in Spence and Johnson, and without them the blue square is a private gesture. The second half also asks about viewers, so the actor's own belief that the conduct carries a message does not end the analysis. In Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission, the Court did not embrace the baker's compelled-speech argument. It called the free speech aspect of the case “difficult” and noted that “few persons who have seen a beautiful wedding cake might have thought of its creation as an exercise of protected speech.”36 The remark echoes the audience half of the inquiry.
Two cautions apply when you use the inquiry. First, “particularized” does not mean a slogan. In Hurley v. Irish-American Gay, Lesbian & Bisexual Group of Boston, the Court said that a narrow, succinctly articulable message is not a condition of constitutional protection, since confining protection to a “particularized message” would never reach the paintings of Jackson Pollock, the music of Arnold Schoenberg, or the verse of Lewis Carroll.37 Second, the wording raises measurement problems. If the actor's intent is judged subjectively, any defendant can claim one after the fact, but if it is judged from a reasonable viewer's perspective, the first half collapses into the second. And because the second half turns on audience understanding, a work whose message a puzzled public fails to grasp seems to lose protection.38 The Court has not used Spence in every case, and it did not in the law-school case discussed next. It may still invoke the two-part test as the first step in an expressive-conduct case, which is how it used the test in Johnson.39
When Conduct Isn't Expressive
A regulated party with more than a private intention can state, publicly and in plain words, what its conduct is supposed to mean. Can that statement carry the conduct into the First Amendment? Rumsfeld v. Forum for Academic and Institutional Rights says no. The Court held there that the expressive nature of the conduct a federal funding condition regulated, a school's choice whether to give military recruiters equal access, was insufficient to bring the conduct within the First Amendment's protection.41 The answer marks the outer edge of the Spence inquiry from the actor's side: the message must come from the conduct, not from commentary about it.
The Solomon Amendment withholds federal education funds from higher education institutions that deny military recruiters access equal to that provided other recruiters.42 Writing for a unanimous Court, Chief Justice Roberts reasoned that the Amendment “neither limits what law schools may say nor requires them to say anything.” It “regulates conduct, not speech”: it affects what law schools must do, not what they may or may not say.43 That left a second question, whether the regulated conduct was itself expressive, and the Court answered no. Unlike flag burning, it said, the conduct the Amendment regulates is not inherently expressive. Before the Amendment, law schools “expressed” their disagreement with the military by treating recruiters differently, but those actions were expressive only because the schools accompanied them with speech explaining them.44
An observer's view shows why the explanation could not rescue the claim. Someone who sees military recruiters interviewing away from the law school cannot tell whether the school is expressing disapproval, whether its interview rooms are full, or whether the recruiters simply preferred another location.45 The act has several readings, and the school's explanation picks one. The Court treated the need for that explanation as strong evidence that the conduct is not inherently expressive, and it named the danger: if combining speech and conduct were enough, a regulated party “could always transform conduct into ‘speech’ simply by talking about it.”46 The schools lost nothing they wished to say, since the statute left their explanations untouched. What they sought was to lend the protection of that speech to an act that said nothing by itself. Johnson is the contrast. There the setting, the timing, and the chants around the burning let viewers read the act itself, while the law schools' interview location stayed open to several readings until the schools' own words supplied one.
A rule that allowed such borrowing would leave expressive conduct with no edge. The Court has observed that some kernel of expression can be found in almost every activity a person undertakes, such as walking down the street or meeting friends at a shopping mall, and that such a kernel does not bring the activity within the First Amendment.47 If a kernel sufficed, virtually every criminal law would draw the same challenge from any defendant who claimed the act was meant to communicate.48
The law-school case is not the only one in which regulated conduct carries no expression of its own. In Nevada Commission on Ethics v. Carrigan, the Court treated a legislator's vote as a tool of governing and not of communication, even though the vote may be exercised out of belief.49 In Arcara v. Cloud Books, Inc., a public-health law led to the closure of a bookstore in which sexual activity was taking place. The Court held the First Amendment was not implicated, because that activity manifested no element of protected expression and the law applied to the premises in which the owners happened to sell books.50 Expression is missing for a different reason in each case. The law schools had a message but no communicative act, a legislator's vote is a tool of governing, and the activity that triggered the bookstore's closure had no communicative element at all.
For content-neutral regulation, the classification matters less than its prominence in the cases suggests. The O'Brien test, taken up in full below, is applied so favorably to the government that it makes very little difference in most cases whether conduct is classified as speech, so long as the regulation is content neutral.51 In Clark v. Community for Creative Non-Violence, the Court assumed that sleeping in symbolic tents in Lafayette Park might be speech and still upheld the camping ban.52 The law-school case added an alternative holding of the same kind: even if the Solomon Amendment regulated expressive conduct, any incidental burden on speech was no greater than essential and promoted a substantial government interest that would be served less effectively absent the regulation, so the Amendment satisfied O'Brien.53 A regulation built around the protester's message is different, because the government can hardly deny that the conduct communicates when the law targets what it communicates.54
The general ordinance easily meets the first, second, and fourth parts of the O'Brien test, which is why the court need not settle whether the burning is speech, while the selective version fails the third part.55 The crucial question therefore shifts from whether the defendant's conduct is speech to whether the government's regulation is based on content.56 Courts can often skip the classification when the law is neutral, but not when it targets a message, and the next sections take up that question and the O'Brien test itself.