Government speech and compelled speech

Yesterday I discussed specialty plate programs in lower courts and the parties’ arguments in Walker v. Texas Division, Sons of Confederate Veterans. Today I begin to analyze the questions in the case.

But first, a disclaimer.


The following is my understanding of First Amendment law, gleaned from years of reading numerous free speech opinions, summaries, and analyses. I’m generally confident in this explanation, but I may well have made mistakes, or simply missed nuance present in the cases but not in the summaries I’ve read. Please point out mistakes in the comments.

Of course, I really have no business trying to explain First Amendment jurisprudence, if I want it explained correctly. First Amendment law is incredibly complex. My haphazard reading will miss things.

But I’m barging ahead anyway, for a few reasons. First, I want to talk about this. Second, it’s fun to talk about it! Third, you don’t learn unless you’re willing to look like a fool from time to time. Fourth, the law is not this recondite, bizarre arcana that only lawyers and judges can understand. It may require some work to correctly understand laws, terms of art, rules of statutory construction, and relevant past decisions in the common law. But any intelligent person can do it if they make the effort.

And fifth, nobody with any sense will unconditionally rely on this as authoritative, not when there are far better places to look for the finest in free Internet legal advice.

Government speech

The “recently minted” government speech doctrine occupies an uneasy place in the realm of speech. For when government speech occurs, non-governmental speech open to First Amendment challenge is reduced. There must be some government speech: otherwise we’d absurdly conclude that the government’s World War II war-bond propaganda must be accompanied by anti-bond propaganda. Government programs often have viewpoints suppressible only in the voting booth. But this mechanism is sluggish and imperfectly responsive, and government speech’s discretion can be abused. So it’s best to be careful anointing government speech.

This is your government. This is your government on beef. Any questions?

Certainly some license plates — the state’s default designs and designs ordered by the legislature — are government speech, even if they’re also individual speech under Wooley v. Maynard. In each case the government wholly chooses what it wishes to say, and that message is government speech. The individual’s choice to assist in conveying it, under Wooley, isn’t government speech.


But Texas’s government-speech argument, applied beyond plates it designs itself, is laughable. The linchpin of Texas’s argument is that because they control the program, that makes it government speech they can control. This argument is completely circular! By starting from their control over the program’s speech, they’ve assumed their conclusion.

This doesn’t mean Texas is wrong. But their circular central government-speech argument can prove nothing. This logical flaw is blindingly obvious. Texas’s lawyers can’t have missed this. If they made this their lead argument, they’re scrambling.

Compelling Texas to speak?

Texas’s better argument is that vehicle licenses and plates are its program, implicating its right to speak or not speak under Wooley. But the First Amendment restrains government power, not individual power. And many courts (although so far not the Supreme Court) have held that government can be compelled to “speak” in accepting advertising in government-controlled places (public transit systems, for a common example). The problem is Texas voluntarily created a specialty plate program open to all for speech. No “compulsion” derives from a voluntary act.

Texas didn’t fully control the specialty plate program, but rather opened it to anyone with money. (As Chief Justice Roberts noted in oral argument: “They’re only doing this to get the money.”) It’s possible there’s government speech in Texas SCV‘s plate, perhaps the occasionally-proposed “hybrid” speech. But once Texas opens the program to all, it loses full control over what’s said.

How then do we consider specialty plate programs? What controls may Texas exercise? Now we must decide how to classify the specialty-plate program with respect to First Amendment-protected speech. What sort of forum for speech is Texas’s specialty-plate program?

Tomorrow, First Amendment forum doctrine.


Specialty plates in circuit courts, and the parties’ arguments at the Supreme Court

Yesterday I discussed the background to Walker v. Texas Division, Sons of Confederate Veterans. Stated briefly, Texas denied Texas SCV‘s application for a specialty license plate with a Confederate flag on it, because the design might be “offensive”. The question is whether Texas is required by the First Amendment to grant the application.

Today I discuss how specialty plate programs have fared in lower courts, and the arguments Texas and Texas SCV bring to the case.

In the courts

Almost every circuit court has required that specialty plate programs be viewpoint-neutral, not restricting designs because of their views. (And the one exception judged a program without an open invitation for designs.) So it’s unsurprising that Texas SCV won its Fifth Circuit case.

Texas appealed to the Supreme Court, which agreed to answer two questions. Are specialty plate programs “government speech” that need not be viewpoint-neutral, such that the design can be rejected as “offensive” (or, indeed, for almost any reason)? And did Texas discriminate by viewpoint in rejecting Texas SCV’s design?

Texas’s argument

Texas says license plates are entirely the government speaking, and it can say or not say whatever it wants. Texas relies on two cases: Pleasant Grove City v. Summum, in which a city’s approval of a limited set of monuments in its city park (and denial of a particular monument) was deemed government speech; and Johanns v. Livestock Marketing Association, in which a government beef-promotion plan that exacted a fee from beef producers to support speech (including the Beef. It’s What’s For Dinner tagline) was deemed government speech that program participants couldn’t challenge on the grounds that it compelled them to speak.

According to Texas, its specialty plates are government speech because Texas “effectively control[s]” the whole program. What matters is whether Texas “exercises final approval authority over every word used” — and it does. Texas allows private citizens to participate, but it has “final approval authority” over every design. Texas also argues that it can’t be compelled to speak by displaying the Confederate flag. By making a license plate, the state’s authority backs (or doesn’t back) every design approved or rejected. Plate purchasers shouldn’t be able to force Texas to espouse the views of an unwanted specialty plate, which drivers would then ascribe to Texas.

The Texas-approved Mighty Fine Burger specialty plate

And of course, Texas says ruling against them would lead to “untenable consequences”. For every “Stop Child Abuse” plate there’d have to be an opposing plate supporting child abuse, and so on for the whole parade of horribles. Texas particularly notes that the Eighth Circuit forced Missouri to let the Ku Klux Klan join the state’s Adopt-a-Highway program under this logic. (The person behind me in the oral argument line related that one of the highways entering Arkansas was adopted by the KKK under that rule, giving Arkansas visitors that delightful first impression of the state.)

Texas also asserted that assessing how members of the public view a specialty plate is “an objective inquiry”, so that deciding a specialty plate “might be offensive” doesn’t discriminate on the basis of the specialty plate’s viewpoint. As to the Fifth Circuit’s criticism of the “unbridled discretion” provided by the “might be offensive” bar, Texas instead describes it as “discriminating among levels of offensiveness”, such latitude permitted because the state is “assisting speech”.

Texas SCV’s argument

Texas SCV says Texas is being hypocritical. The Capitol gift shop sells Confederate flags. Texas recognizes a state Confederate Heroes Day. It maintains monuments to Confederate soldiers. Either Texas doesn’t really think the Confederate flag is offensive to the public, or its other “government speech” is flatly inconsistent with its specialty-plate stance.

Texas SCV also distinguishes the plates designed by the state legislature from plates designed by private entities. The former are the product of the government, but the only government involvement in the latter is in approval or disapproval. The driver has ultimate control, because only when he designs a plate and ultimately drives a vehicle with it does speech occur. And under Wooley v. Maynard — a case where a Jehovah’s Witness protested New Hampshire’s fining of people who covered up “Live Free or Die” on their license plates, and the Court said New Hampshire couldn’t force a person to espouse the state motto — it’s the individual’s speech (at least for non-legislatively-designed plates).

Texas SCV brushes off Summum and Johanns. Permanent monuments in parks have always been associated with the government, because parks physically can’t accommodate all monuments. Not so for license plates. (And Texas’s $8000 deposit covers startup costs that might justify treating rare plates differently.) And while the beef-promotion messages were part of a “coordinated program” by government to “advance the image and desirability of beef and beef products”, privately-designed specialty plates are not — especially as their fullycontradictory messages are “consistent” only as a fundraiser.

Finally, given that privately-designed specialty plates are private speech, the First Amendment requires that restrictions be viewpoint-neutral. By restricting Texas SCV’s message based on its potential for offensiveness, Texas endorsed viewpoints that deem the Confederate flag racist and discriminated against viewpoints that do not.

Tomorrow, analysis of Texas’s government speech and compelled speech arguments.


On the topic of Afghan “censorship” and the rights and duties of a free press

Merriam-Webster defines the word censor thus:

: to examine in order to suppress or delete anything considered objectionable <censor the news>; also : to suppress or delete as objectionable <censor out indecent passages>

For censorship of expression to occur, therefore, two things must be present. The expression must be examined for objectionability. The expression then must be deleted or suppressed, which is to say that it must be prevented from being made. Merriam-Webster gives this as the relevant definition of suppress:

: to keep from public knowledge: as b: to stop or prohibit the publication or revelation of <suppress the test results>

It is not enough to simply discourage an expression, but rather it must be prevented entirely.

Recent headlines have been making hay over a request of the government of Afghanistan. By way of background, it is well known that Afghanistan is going through a period of relative instability as it comes under attack from various factions and terrorist elements. These disruptive elements within (and to some extent without) Afghan society attempt to use force to bring about political, social, and cultural changes, through suicide bombings, military attacks, and other violent tactics. These actions compel Afghanistan to consider actions other nations may never even need to consider, in the interests of furthering what stability is possible now and in the future.

On Thursday Afghanistan will elect its chief executive, the president, in one of the finest traditions of democratic society. This occasion will naturally draw violence in an attempt to influence voting: recall the 2004 Madrid train bombings which occurred three days before Spain’s general elections, which undoubtedly strongly affected voting and which very possibly may have changed the outcome entirely. Yet note: election-time violence will directly affect only a few dozens or hundreds, but reports of that violence will indirectly affect several orders of magnitude more people.

Given this, and given that Afghanistan at the moment cannot control election violence nearly as well as most other nations, it is clear that when Afghanistan sees actions can be taken that will reduce the impact of violence, it is reasonable to consider taking them. In particular, Afghanistan has called on domestic and foreign media not to cover election-day violence to avoid such reports potentially driving away voters. Some denounce this request as “an attempt to censor the reporting of violence”. But is it really censorship as is claimed?

Recall the definition of “censor” given above. First, the expression must be examined for objectionability. Assuming that we admit as “examination” consideration of future expression based on past expression upon similar topics in similar situations, we indeed have this: Afghanistan has evaluated coverage surrounding election day and determined that election-day coverage of violence, during voting hours, may do more harm than good. Now, however, consider the second arm of the censorship test: the expression must be prevented from being made. On this point Afghanistan’s actions do not conform to the requirements of censorship. Afghanistan’s government has made a request that media not cover election-day violence; it has not used its democratically-permissible monopoly on force to prohibit such coverage. While a request may under certain circumstances be to some extent a prohibition, as when the request constitutes an implicit threat of future sanctions, nothing of the sort is suggested in the article to which I link. Therefore, Afghanistan’s actions, despite the inveighings of such groups as Human Rights Watch, do not constitute censorship.

A further three points are worth noting. First, the request extends only from 06:00 to 20:00, the period when polls are open in Afghanistan. The request in no way asks that coverage be diminished outside (and most particularly, after) this period, neither of then-present violence nor of violence which occurred during polling hours on election day. Reports of violence would not be “[kept] from public knowledge”, except temporarily, and only in case of voluntary restraint by individual media organizations — not by power of the government of Afghanistan. Second, the request extends only to the particular topic of violence and to no other topic. While this may certainly exclude much newsworthy information, it is limited in scope to that which gives clear cause for concern. Third, note that the request is of both “domestic and foreign media”. Afghanistan cannot censor foreign media because the responsibility for the actions of such organizations lies outside its national borders and sphere of influence (meager as that perhaps may be). The government of Afghanistan might (I have no knowledge of the limits or extent of Afghan government authority, and I only suggest a plausible possibility) have the power to censor domestic media if it chose to do so in light of current circumstances (in the same way that, in case of rebellion, the writ of habeas corpus may be suspended in the United States by the government), but it has not exercised that power, if even such a power exists. A government may have claim to extensive power in theory, but it is the extent to which excessive power is not exercised when it might have been that makes that government worthy of the consent of the governed.

Accepting, then, that media organizations may choose to honor or not honor the request, should they honor the request? Here the issue is not nearly so clear, and strong arguments may be made for either choice. I am inclined to believe that voluntary restraint with respect to a specific topic, for fourteen hours during a singularly important political event with clear cause to do so, ending completely upon its termination and in no way restricting publication of coverage from those fourteen hours after their conclusion, is likely the best choice for most organizations. Media organizations should not allow insurgents and terrorists to wield their coverage as part of an Afghan suicide pact.

One fact, however, remains clear: that organizations have a choice as to whether they will allow themselves to be used in this manner demonstrates that the government of Afghanistan is not engaging in censorship.

« Newer