Concept
Presumption and Burden
Prior restraint
Presumption and Burden
Once a measure is identified as a prior restraint, the question becomes what the government must show to defend it. Near v. Minnesota supplies the answer: prior restraints are generally disfavored, the government bears a heavy burden to justify one, and that burden is ordinarily met only by a special societal harm, such as a serious threat to national security. The Court's own example is the wartime publication of troop movements.
Earlier sections showed how to recognize a prior restraint by the timing of the government's intervention and why the law distrusts advance suppression. This section asks how heavy the burden is and who must carry it. It states the standard, follows it through the Pentagon Papers case, the leading modern application, and ends with the Justices' separate opinions in that case, which show how little agreement exists about what would satisfy the burden.
The Heavy Presumption
The standard joins two ideas that are easier to apply when separated. A presumption is the court's starting point, and here the court starts by assuming that the restraint is invalid. A burden is the task of dislodging that assumption, and here the task belongs to the party defending the restraint. The presumption tells the court where to begin; the burden tells it who must supply the reasons for ending anywhere else. The Court has called prior restraints “the most serious and least tolerable infringement on First Amendment rights,” and it has said that any system of them comes to the Court bearing a heavy presumption against its constitutional validity.1 A government that offers only a general worry about harm therefore loses before the analysis has fairly begun.
The Court applies the same formula to injunctions. In Organization for a Better Austin v. Keefe, an injunction barred peaceful leafleting, and the Court struck it down. It reaffirmed that peaceful pamphleteering is protected communication and that “[a]ny prior restraint on expression comes to this Court with a ‘heavy presumption’ against its constitutional validity.” The plaintiff, as the party seeking the restraint, could not meet the burden that presumption creates.2
The Court also said that courts may not “concern themselves with the truth or validity of the publication” when they decide whether such a restraint can stand.3 That refusal follows from what the presumption attaches to. The presumption responds to the form of the government's action, an order that removes a message before anyone has weighed it, so a showing that the message is false cannot lift it. Falsity does its work later, in a libel action over what has already been published. Near said as much of public officials, whose conduct remains open to debate and free discussion in the press: they find their remedy for false accusations in libel actions and not in proceedings to restrain publication.4
No bright-line test sorts valid restraints from invalid ones. The cases supply a presumption and a burden, and the question that carries the analysis is whether some special societal harm justifies suppressing this speech in advance. A real harm does not end the inquiry. The order must also reach no more speech than necessary, as the Court showed when it upheld some restrictions in a state injunction and struck down others, such as certain buffer zones, because they burdened more speech than necessary to serve a significant government interest.5 The restraint must also come with procedural safeguards, which later sections develop. A court that finds a real harm has therefore completed only the first step.
The burden sits on the government because of how restraints work. An order takes effect before any adjudication of the speech it covers, so a mistaken order inflicts its injury at once, and for news that loses value with delay the injury is often permanent. The decision-maker is also forecasting the effects of words not yet spoken, and forecasts of that kind drift toward restraint because an error in favor of suppression is rarely exposed. If the speaker had to prove the message harmless, the burden would fall on the party least able to carry it, since no one can show what a story will do before it runs. Placing the burden on the government reverses that disadvantage and demands evidence of harm in place of prediction.
The leafleting injunction in Keefe and Near's wartime example mark the two sides of the line. A private plaintiff's grievance with leaflets did not carry the burden.6 Near, by contrast, recognized that a government may prevent publication of “the sailing dates of transports or the number and location of troops” in wartime.7 Publication would itself endanger the ships and the people aboard, and no later damages award or prosecution could undo that harm. The wartime example is the paradigm of a special societal harm, and a private grievance about reputation or commercial standing is not of that kind.
A common error is to ask only whether the speech is harmful, or whether it is false, and so to skip the presumption. Nearly all speech that someone wants restrained is said to cause some harm. The real inquiry is whether the harm is grave and special enough to overcome the presumption, and whether the order is drawn narrowly enough to address it.
The Pentagon Papers Case
The leading modern application of the presumption is New York Times Co. v. United States, the Pentagon Papers case, in which the government asked federal courts to bar two newspapers from publishing a classified study of the Vietnam War.9 The Supreme Court answered with a per curiam opinion, an unsigned opinion issued in the name of the whole Court, whose core is a single sentence of law: “Any system of prior restraints of expression comes to this Court bearing a heavy presumption against its constitutional validity.”10 The opinion draws a consequence from that sentence: the government “thus carries a heavy burden of showing justification for the imposition of such a restraint.”11 The word “thus” shows the burden flowing from the presumption. A restraint presumed invalid must be defended by the party that wants it, and that party must supply the reasons.
The case moved fast. Daniel Ellsberg, a former Defense Department employee, gave the New York Times and then the Washington Post a copy of a 47-volume report on the Vietnam War. Three days after the Times began publishing, the Nixon Administration obtained a temporary restraining order against further installments, and within three weeks the Justices had issued an order forbidding further restraint on publication.12 The government sought injunctions on national security grounds. The district courts refused them, the D.C. Circuit affirmed in the Post case, and the Second Circuit reversed in the Times case and approved an injunction.13 The war was still under way, and the government's position was that publication would jeopardize the war effort.14
Speed mattered because a restraint operates from the moment it issues. Court orders generally must be obeyed until “reversed for error by orderly review,” and in Walker v. Birmingham the Court held, five to four, that marchers could not defend against contempt charges by asserting that the injunction was unconstitutional.15 The four dissenters said the majority had elevated a “rule of judicial administration above the right of free expression.”16 A newspaper under a restraining order therefore cannot ordinarily publish first and contest validity later, and every day the order stays in force keeps the articles out of print. That is the freezing effect described earlier in this chapter, and for news feeding a live public debate it has a real cost. The same concern lies behind the rule that a restraint imposed before a final judicial determination on the merits must be limited to preserving the status quo for the shortest fixed period compatible with sound judicial resolution, a rule developed with the procedural safeguards in a later section.17
The per curiam disposes of the case in three short paragraphs.18 After stating the presumption and the burden, it recites that the lower courts in the Times and Post cases had held that the government failed to meet that burden, and it says simply, “We agree.”19 The vote was six to three, and all nine Justices accepted the presumption and the burden, so the disagreement concerned whether the government had carried it and whether any restraint could ever be justified.20 Nine Justices produced ten opinions, one from each Justice plus the per curiam. The short per curiam is the only one with staying power, so the rule of the case should be taken from its text and not from any single concurrence.21
The opinion's brevity leaves the reasoning to the presumption itself, and the outcome fits the reasons for it. The government's request rested on a prediction that publication would jeopardize the war effort.22 A judge asked to act on a prediction has only an executive's forecast of what words not yet printed will do, and forecasts of that kind drift toward restraint. The Court demanded more, and it did so in a setting close to Near's wartime example of publishing the sailing dates of transports or the number and location of troops. Even claims of national security during an ongoing military action did not satisfy that exception.23
What the opinion left open matters as much as what it decided. Three questions remain open on its face:
- It does not decide whether national security can ever justify a prior restraint on the press, a question on which the separate opinions divide.24
- It does not say what showing would carry the burden, and the standard illustration is a danger comparable to a publication that would endanger a transport at sea during wartime.25
- It does not decide whether the newspapers could be punished after publication, and several Justices suggested that the reluctance to restrain would not necessarily extend to a later criminal prosecution.26
The third point follows the logic of the first section of this chapter. The presumption responds to the timing of the government's intervention, so a ruling that the government cannot stop the presses says little about what it may do once the story has run.
An exam question usually varies the government's evidence and asks whether the presumption yields.
The government in the Pentagon Papers litigation stood on the other side of that line, since its showing, a claim that publication would damage the war effort, did not carry the burden in the Court's judgment.27 Two misreadings recur on exams. Treating the per curiam as a holding that the press may never be restrained adopts the position of only Justices Black and Douglas, who would have barred every prior restraint on the press.28 Treating the label “national security” as enough to carry the burden ignores the opinion's central sentence, which places the burden on the government and requires a justification that survives a heavy presumption of invalidity.
The Separate Opinions
Nine Justices wrote separately, and all nine accepted that any system of prior restraint bears a heavy presumption against its constitutional validity and that the government carries a heavy burden to justify it.30 They parted over whether any restraint on the press can ever be justified and, if so, what the government must show. Their opinions sort into three camps. Because the per curiam decided so little beyond the presumption, the camps are the best guide to how a later court might rule on facts it did not reach.
Justices Black and Douglas, the first camp, would have barred every prior restraint on the press. Black wrote that allowing the publication of news ever to be enjoined would “make a shambles of the First Amendment.”31 His position rests on text and history: the amendment says Congress shall make no law abridging the freedom of the press, so he rejected the Solicitor General's suggestion that the general powers granted in the original Constitution could narrow a later, more specific guarantee.32 Douglas reached the same place by reading the dominant purpose of the amendment as ending the suppression of information that embarrasses those in power, and he stressed that no statute barred this publication.33 Their view requires no weighing of harm, which is also why it has never commanded a majority.34 Because Black and Douglas were the only two Justices who would have barred every restraint, the other seven accepted that a restraint could be justified in some case.35
Justices Brennan, Stewart, White, and Marshall, the second camp, left the door open to some restraints but held it nearly closed on these facts. Their opinions follow two strands. The first demands a strict showing of harm. Brennan observed that the government's whole claim was that publication “could,” “might,” or “may” prejudice the national interest, and he held that the First Amendment tolerates no prior judicial restraint predicated on surmise or conjecture.36 He recognized one narrow class of exceptions, arising when the nation is at war, and even there he required allegation and proof that publication would “inevitably, directly, and immediately” cause an event kindred to imperiling the safety of a transport already at sea.37 That standard reaches Near's wartime examples and little else, so it operates as something close to a prohibition.38 It also explains his refusal to tolerate even the short stays granted to give the courts time to study the claim, since the executive must put the basis for its request before the judiciary before any aid issues.39 Stewart, joined by White, argued that the Executive may justify a restraint to protect the confidentiality its defense and foreign-affairs duties require, but only by showing that disclosure will cause “direct, immediate, and irreparable damage to our Nation or its people,” and he found that the government had not met that test.40
The second strand turns on the missing statute. White said he was convinced that disclosure would do substantial damage to the public interest, yet he concurred because the government had not met the very heavy burden required, and he implied that congressional authorization for a restraint might lessen that burden.41 He added that the government's failure to justify a restraint did not decide whether it could proceed against the newspapers for criminal publication afterward, and he noted that Congress had addressed the protection of defense secrets through criminal laws without authorizing an injunction against threatened publication.42 Marshall grounded his concurrence entirely on the gap. Using the contempt power to restrain conduct that Congress had chosen not to prohibit would offend the separation of powers, and he reasoned that when Congress specifically declines to make conduct unlawful, it is not for the Court to redecide that question and overrule Congress.43
A missing statute matters because of who is making the judgment. When the Executive asks for an injunction with no statute behind it, the court must decide for itself, on the strength of an executive forecast, which disclosures are dangerous enough to silence, and then enforce that decision through contempt. A statute would supply a legislative judgment about which disclosures warrant an injunction, so the court would be applying a rule instead of creating one. White and Marshall differ in degree. For White a statute might lighten the government's burden, while Marshall relied on the fact that Congress had actually refused the power the President was claiming.44
Chief Justice Burger and Justices Harlan and Blackmun, the third camp, would have upheld the restraint or kept it in place longer.45 Burger focused on the speed of the proceedings, complaining that the Justices “literally do not know what we are acting on,” and he would have left the injunction in force until an orderly hearing could be held.46 Harlan invoked the separation of powers in the opposite direction from Marshall. Because the Executive holds primacy in foreign affairs, he reasoned, a court may ask only whether the dispute falls within that power and whether the head of the responsible department made the determination, and it may not redetermine the probable effect of disclosure on national security.47 The two Justices share a principle and part ways on what it commands: Marshall would keep courts from restraining what Congress left lawful, and Harlan would keep courts from second-guessing what the Executive has found dangerous. Blackmun added that the First Amendment is one part of a Constitution that also vests foreign affairs and national safety in the Executive, and he called for weighing the broad right of the press to print against the very narrow right of the government to prevent, under standards that had not yet been developed.48
The votes do not add up to an absolute ban on prior restraints, and they do not add up to a single test either. The per curiam resolved that case and virtually no others, and the separate opinions only hint at the standards that would govern a different case, so how the heavy presumption is to be applied remains vague.49 An answer on this topic should therefore take the presumption and the burden from the per curiam and treat each camp as a competing test to be applied to the facts. Two questions were left open: what circumstances, if any, would justify a court order preventing publication to protect national security, and what difference a statute authorizing a restraint would make.50
The second question reached a court in United States v. Progressive, Inc., where a federal district court enjoined a magazine from publishing an article on how to build a hydrogen bomb. Unlike the Pentagon Papers case, a provision of the Atomic Energy Act appeared to authorize the injunction, and the government argued that preventing nuclear proliferation justified the restraint.51 The district judge reasoned that a mistake against the magazine would curtail First Amendment rights, while a mistake against the United States could bring thermonuclear annihilation, and he relied on that “disparity of risk.”52 The appeal was dismissed after others published the same information elsewhere, so the Supreme Court never reviewed the ruling.53 Set beside the Pentagon Papers litigation, which involved no authorizing statute and only a general claim of harm to the national interest, the pair shows two variables changing at once, the statute and the gravity of the harm. It cannot show which of them carries the weight, and that uncertainty is the open question.
A published illustration of the same problem imagines a monthly, Modern Mercenary, about to print instructions for building a thermonuclear device, with the Atomic Energy Act authorizing injunctions against disclosing restricted data and government experts saying only that publication will “possibly” help a medium-sized nation develop the weapon sooner.54 Brennan's test would likely forbid the injunction, because a possible speedup is speculative and cannot be called inevitable, direct, and immediate. A district judge could still reasonably decide that the statute and the seriousness of the potential harm satisfy Near's troopship exception, and that conclusion might be upheld.55
Step 3 is the one students skip, because it concerns a statute that is absent from the facts. In the Pentagon Papers litigation the missing authorization was a reason for at least two votes, so a fact pattern that includes an authorizing statute changes a variable the Court considered important.