University of Colorado Law Review > Printed > Volume 91 > Issue 3 > CONCEPTS, NOT NOMENCLATURE: UNIVERSAL INJUNCTIONS, DECLARATORY JUDGMENTS, OPINIONS, AND PRECEDENT

CONCEPTS, NOT NOMENCLATURE: UNIVERSAL INJUNCTIONS, DECLARATORY JUDGMENTS, OPINIONS, AND PRECEDENT

Introduction

The battle lines are drawn on the permissibility and validity of so-called “nationwide” injunctions—injunctions in federal constitutional litigation purporting to halt government enforcement of a challenged law[2] against all possible targets of that law and to protect all rights holders against enforcement, not only the parties to the action. Courts are divided—some granting,[3] with attempts at justification,[4] others rejecting, in practice if not in concept.[5] Justices Thomas and Gorsuch have weighed in against them.[6] Scholars supporting their validity[7] and scholars rejecting them as impermissible[8] have made their positions known.

I have staked my position in the impermissible camp:[9] A court order should protect rights-holders who are parties to a particular case against enforcement efforts by government officials who are also parties to that case. A court order should protect no further.

The requirement that remedies be particularized and limited to the parties arises from the judicial process in which constitutional review and adjudication occur. That process offers rights-holders two ways to challenge enforcement of a constitutionally invalid law.

The first is defensively. The rights-holder is a defendant in a proceeding to enforce some law and raises the constitutional defect in that law as a defense against enforcement. This could be government-initiated proceedings against X, whether criminal,[10] civil,[11] or administrative,[12] or it could be private civil litigation against X.[13] In either context, X challenges the constitutional validity of the law being enforced; a court agreeing with X’s constitutional defense will dismiss the enforcement action or otherwise enter judgment for X.

The second is offensively. The rights-holder initiates a pre-enforcement federal constitutional challenge to the law. X sues the government or, more commonly, the executive official responsible for enforcing the law in what often is labeled an Ex parte Young action.[14] X must seek prospective relief—a declaratory judgment[15] that a law is constitutionally invalid and cannot be enforced by the defendant against the plaintiff, an injunction prohibiting enforcement of that law against the plaintiff,[16] or both.[17] Many cases produce anti-suit injunctions, with the court issuing a prohibitory (or negative) injunction preventing the defendant official from initiating a judicial proceeding to enforce the law against the rights-holder.[18] But Ex parte Young pre-enforcement litigation extends to efforts to stop enforcement of any law, regardless of how that law would be enforced.[19]

Whether the constitutional defect is raised offensively or defensively, the court must interpret the Constitution and decide whether the underlying law being enforced is consistent with the Constitution. A law might be invalid because it exceeds internal limits on the government’s power to enact laws (such as from the Commerce Clause or federalism principles),[20] or because it violates external limits on the government’s power arising from provisions creating individual rights (such as the First or Fourteenth Amendments).[21]

Non-particularized remedies—the subject of this judicial and academic debate—arise in the offensive context. But if a court agrees that a law is constitutionally defective, either context should produce the same result: an opinion declaring the law constitutionally invalid and a judgment premised on that conclusion, prohibiting continued and future enforcement as to the parties to the action. Regardless of remedy—dismissal of ongoing enforcement or injunction prohibiting future enforcement—the judgments should have the same scope. Samuel Bray and Douglas Laycock separately explain the appropriate scope of judicial remedies. Bray argues that a “federal court should give an injunction that protects the plaintiff vis-à-vis the defendant.”[22] Laycock explains that “the court in an individual action should not globally prohibit a government agency from enforcing an invalid regulation; the court should order only that the invalid regulation not be enforced against the individual plaintiff.”[23]

From that starting point, this Article explores four subsidiary issues on the scope of injunctions in constitutional cases. All relate to the fact that judicial review occurs in particular litigation and procedural contexts.

Part I considers what to call these beyond-the-plaintiffs injunctions. The term that has carried the day in the courts is “nationwide,”[24] with many scholars following suit.[25] Bray offers “national” injunction[26] as a close alternative. Justice Thomas adopted “universal” in his concurring opinion in Trump v. Hawaii,[27] Justice Gorsuch called them “cosmic”[28] (although it appears he was joking), and they agreed these injunctions suffered from the “same basic flaw” regardless of framing as universal, cosmic, or nationwide.[29] Michael Morley focuses on whether the injunction is plaintiff-oriented or defendant-oriented.[30] Another conception emphasizes the injunction’s particularity, whether it is particularized to the parties to the case or not particularized to any parties.[31] Nomenclature matters because the correct terminology exposes the real issues, avoids judicial confusion over those issues, and may, in turn, eliminate the need for any terminology.

Part II explores the unique role and status of a different remedy that a court might issue in a pre-enforcement challenge to the enforceability of a constitutionally invalid law: the declaratory judgment—a declaration of the rights and legal relations of the parties[32]—which has developed into a tool to obtain a judicial declaration on the constitutional validity of a law.[33]

Part III emphasizes the distinction between two products of a judicial decision: the judgment resolving the dispute in a particular case and imposing a party-specific remedy and the opinion explaining that judgment. Each carries distinct meanings, characteristics, and consequences. Conflation of these products produces some confusion among courts and commentators, including assigning to the judgment the legal consequences and effects of the opinion.

Part IV shows why the debate over universal or nationwide scope is an unnecessary and unfortunate distraction, diverting courts and litigants from the real issues and producing layers of confusion in constitutional adjudication.

“Universal” or “Non-Particularized,” Not “Nationwide,” Perhaps Merely “Injunctions”

The confusion over nomenclature—over the competing terms for orders that extend beyond the plaintiffs—relates to a separate confusion over distinct elements of an injunction: “who” and “where.”

“Who” refers to the persons bound and protected by a judgment—who is controlled and subject to the limitations of the order and bound to act or refrain from acting in some respect,[34] as well as who enjoys the blanket of the court’s protection and its power to enforce the judgment if it is disobeyed.[35] “Where” refers to the “territorial breadth” or geographic scope of the court order, where the judgment and court’s enforcement power can find and act with respect to those bound or protected by an injunction. In pre-enforcement constitutional litigation, “who” means against what persons government is barred from enforcing the challenged law and against what persons government remains free from judicial decree to enforce the challenged law; “where” means the place in the world in which the government is barred from enforcing the law against that “who” the injunction protects. The former element is not about geographic applicability, but about the “particular people or entities whose rights they are tailored to enforce.”[36]

Where

“Nationwide” best describes an injunction’s “where”—the “geographical bounds” of the plaintiff’s injury and of the remedy for the plaintiff’s injury. So understood, all injunctions are and should be nationwide. All injunctions protect the plaintiff against the defendants’ unconstitutional or unlawful conduct everywhere the plaintiff may be or may go. That is, government officials are and should be prohibited from enforcing the constitutionally defective law against the protected party wherever the protected party may be or may go. A rights-holder’s rights are violated by threatened enforcement of a constitutionally defective law wherever she goes, and the injunction protecting her against those violations protects her wherever she goes. All injunctions thus extend beyond the geographical bounds of the issuing judicial district or affirming judicial circuit.

A simple case illustrates the point. The Northwest Immigrant Rights Project (NWIRP) filed suit in federal court, seeking to halt enforcement of certain regulations in immigration proceedings; those regulations required attorneys to file a notice of appearance (and provide full legal representation), even when seeking to offer incidental advice to parties otherwise appearing pro se. NWIRP could not afford to provide full representation in every case, so the regulations’ effect was to stop NWIRP from providing incidental assistance. NWIRP alleged the regulations violated the First Amendment—by interfering with communications between it and parties to immigration proceedings—and the Tenth Amendment—by infringing on the state power to regulate attorneys.[37]

The district court enjoined enforcement of the regulations against NWIRP, making it properly and appropriately nationwide. In prohibiting the federal government from enforcing the attorney regulations against NWIRP, the district court necessarily prohibited enforcement of the regulations against NWIRP anywhere in the United States where NWIRP might attempt to provide legal services in immigration proceedings. The organization operated in Washington state,[38] so the injunction obviously prohibited enforcement in proceedings held there. But if NWIRP were to begin providing legal services in immigration proceedings in Oregon or Texas or Florida or Maine, the injunction would bar enforcement of the regulations in those proceedings, and attempted enforcement in those places would violate the Washington-based injunction. The protection that the injunction affords NWIRP against enforcement follows NWIRP wherever it might otherwise be subject to enforcement of the challenged law. This is neither exceptional nor controversial. A court would not call this a “nationwide” injunction since the nationwide scope of the protection for NWIRP would be understood—the court would call it an injunction.

An injunction prohibiting enforcement of a state law should be as nationwide as an injunction prohibiting enforcement of federal law—it protects the plaintiff against enforcement of the constitutionally defective state law everywhere she is or might go. In practice, nationwide scope presents less of a practical problem because of constitutional and prudential limits on extraterritorial application of state law. Those limits render it unnecessary for the injunction to protect nationwide; it does not render this conception of nationwide inapplicable. If a court enjoins Florida officials from enforcing a Florida law prohibiting flag burning against the plaintiff, the injunction prohibits Florida officials from enforcing that law anywhere the plaintiff might burn a flag, including outside Florida. Limits on extraterritorial application of Florida’s flag-burning law—Florida cannot prosecute someone who burns a flag in Texas for violating Florida law—add an additional prohibition on that enforcement.

Who

The significant feature at the heart of this legal and scholarly controversy is the injunction’s “who”—when an injunction prohibits, or purports to prohibit, enforcement of the challenged law against the universe of people who might be subject to enforcement of that challenged law, whether parties to the constitutional litigation or otherwise. During argument in Trump v. Hawaii, Justice Gorsuch recognized the problem as injunctions that are “not limited to relief for the parties at issue” stopping enforcement of “a federal statute with regard to anybody anywhere in the world.”[39] Because “nationwide” describes “where” and the real problem is the injunction’s broad “who,” different terminology becomes necessary.

“Universal” works well because the injunctions prohibit enforcement of the challenged law against the universe of people—parties or otherwise—against whom the challenged law might be enforced. I have urged that term for several years,[40] Justice Thomas adopted it as the appropriate term,[41] and Justice Gorsuch appeared to have been joking when he labeled these injunctions “cosmic.”[42]

Further consideration reveals that “universal” is incomplete. An injunction might prohibit enforcement of a law against rights holders other than the plaintiff without being universal—the injunction might protect some rights-holders other than the parties without protecting all right-holders. An injunction protecting the plaintiff and ten nonparties is not universal, although it would be impermissibly overbroad in scope.

The appropriate terminology must capture all injunctions that protect beyond the plaintiff—those that prohibit enforcement against the universe and those that prohibit enforcement against more than the plaintiff but less than the universe. We thus might distinguish “particularized” injunctions from “non-particularized” injunctions. A particularized injunction’s protections are particularized to the parties to the action, but not beyond, consistent with a judicial remedy benefitting the plaintiff “in particular.”[43] A non-particularized injunction’s protections extend beyond the plaintiffs—whether to the universe or to some group larger than the plaintiffs but smaller than the universe.[44] Stephen Vladeck offers “categorical” to describe the injunction prohibiting all enforcement of the challenged law, regardless of the government’s target.[45]

Morley’s focus on the injunction’s “orientation” captures all overbroad injunctions, whether universal or non-universal-but-beyond-the-plaintiff. A plaintiff-oriented injunction “vindicates the plaintiffs’ rights, but otherwise leaves the underlying statute or regulation undisturbed,”[46] the equivalent of a particularized injunction. A defendant-oriented injunction “completely prohibit[s] the defendant agency or official from enforcing a challenged provision against anyone throughout the state or nation,”[47] the equivalent of a non-particularized injunction.

A court might impose an injunction that is both permissibly nationwide and impermissibly universal/non-particularized. Injunctions are nationwide when they protect the named plaintiffs against enforcement of the constitutionally invalid law throughout the nation, wherever the plaintiffs are or might go; this is proper and unremarkable. Injunctions are non-particularized when they protect nonparties—people other than the named plaintiffs—against enforcement of the constitutionally defective law throughout the nation. This is the point of controversy.

Any term—universal, non-particularized, categorical, defendant-oriented—also works regardless of the source of the challenged law. An injunction prohibiting enforcement of a state law can be universal/non-particularized, just as an injunction prohibiting enforcement of a federal law can be universal/non-particularized. The difference is the size of the universes against whom enforcement is proscribed. Because federal law has broader reach than state law, the universe of people subject to enforcement of federal regulations governing immigration proceedings[48] or federal regulations repealing the mandate that employer-provided insurance cover contraception[49] is broader than the universe of people subject to enforcement of any state law. And the scope of an overbroad injunction prohibiting enforcement of federal law is broader than the scope of an overbroad injunction prohibiting enforcement of a state law.

For example, in Koontz v. Watson, the District of Kansas declared constitutionally invalid a Kansas law requiring all persons who contract with the state to certify that they were not involved in boycotts of the State of Israel.[50] The plaintiff, a teacher hired to conduct teacher-training programs, alleged that the law violated the First Amendment, and the court agreed.[51] The injunction prohibited the state from enforcing any statute, law, policy, or practice requiring independent contractors to declare that they are not participating in a boycott of Israel and prohibited the state “from requiring any independent contractor” to certify that they are not participating in a boycott of Israel as a condition of contracting with the state.[52] That injunction was universal/non-particularized—prohibiting enforcement of all state laws against the universe of potential state contractors, regardless of who those contractors are, where they are, what they are contracting for, and what laws they are subject to. But it protected a smaller universe.

Morley argues that the nomenclature discussion errs in looking for a single term, when the debate is over five distinct types of injunctions: (1) nationwide plaintiff-oriented injunctions, prohibiting enforcement of the challenged law against the plaintiff wherever she is; (2) nationwide plaintiff-class injunctions, prohibiting enforcement of the challenged law against all members of a civil-rights injunctive class certified under Federal Rule of Civil Procedure 23(b)(2);[53] (3) nationwide associational injunctions, prohibiting enforcement of the challenged law against members of a plaintiff association;[54] (4) nationwide defendant-oriented injunctions, prohibiting the defendant government from enforcing the challenged law against any person, any place; and (5) nationwide defendant-class injunctions, prohibiting a class of defendants from enforcing the challenged law against the plaintiff.[55] He argues that the first three are proper if the plaintiff can satisfy certain requirements, while the fourth is not appropriate, and the fifth is controversial.[56] Morley’s framework provides helpful specificity, while capturing the essential point that all injunctions should be nationwide in their “where,” but particularized in their “who.”

Recent litigation within the Ninth Circuit demonstrates the mischief when courts fail to use the proper terminology. Four asylum-advocacy organizations sued to stop enforcement of Trump Administration regulations requiring those seeking asylum in the United States to first seek asylum in a third country.[57] The district court enjoined enforcement of the regulations, making the injunction “nationwide” because the Ninth Circuit had “consistently recognized the authority of district courts to enjoin unlawful policies on a universal basis.”[58] In that sentence, the district court demonstrated the nomenclature confusion at the heart of this legal debate—using nationwide and universal as synonyms when they cover distinct aspects of the court order.

The Ninth Circuit granted in part a motion to stay the injunction pending appeal. The injunction remained in effect within the Ninth Circuit, but the court stayed the injunction to the extent it applied outside the Ninth Circuit, “because the nationwide scope of the injunction is not supported by the record as it stands.”[59]

But the Ninth Circuit left the district court with jurisdiction to “further develop the record in support of a preliminary injunction extending beyond the Ninth Circuit.”[60] The district court accepted the invitation, offering factual support for reinstating the injunction’s original scope. That factual support focused on the extent to which some of the California-based plaintiff organizations provided representation and educational programs to asylum-seekers in states outside the Ninth Circuit.[61]

The Ninth Circuit’s stay produced relief that was over- and under-protective. It was over-protective by leaving in place an unstayed injunction protecting not only the four plaintiffs but also the non-particularized universe of enforcement targets within the Ninth Circuit. It was under-protective by failing to protect the named plaintiffs, who were affected by the challenged regulations outside the Ninth Circuit, wherever they operated. In other words, it produced injunctive relief that was insufficiently nationwide to accord the plaintiffs complete relief while being too universal/non-particularized in protecting nonplaintiffs.

By contrast, the district court performed the appropriate analysis in supplementing the record. It focused on the location of the four named plaintiffs, who operated and suffered injury outside the Ninth Circuit and who needed nationwide protection in other states.[62] In other words, it focused on the geographic scope of the plaintiffs’ activities and thus the nationwide scope of an injunction that could protect them in those activities. And the district court never mentioned protecting nonparties, who should not have been within the scope of any court order.

Eliminating Qualifiers

While Morley adds more nomenclature, perhaps the search for correct nomenclature is misguided. Rather than label injunctions as “nationwide” or “universal” or “plaintiff-class-oriented” or “non-particularized,” we should, as Portia Pedro argues, call them “injunctions,” without misleading, inaccurate, undefined, and undefinable qualifiers.[63] That is possible, however, only with agreement on and understanding of the definition and appropriate scope of an injunction.[64] The “universal” or “non-particularized” label attaches when courts attempt to stretch their orders beyond that appropriate scope. The label is necessary to describe and criticize an erroneously broad order.

If Justices Thomas and Gorsuch[65] and scholars such as Bray, Cass, Morley, and myself[66] are correct about the limited remedial power of federal courts in constitutional litigation, the scope-of-injunction debate reduces to two points. First, a judicial remedy is particularized to the litigation at issue and therefore to the parties to that litigation;[67] an injunction should protect the plaintiffs against enforcement, but no further. Second, that injunction should protect the plaintiffs (however defined) wherever in the nation they find themselves.[68] In other words, the nationwide piece of that injunction is necessary and appropriate in every case; the universal/non-particularized (or defendant-oriented) piece of the injunction is unnecessary and inappropriate in every case.[69]

To render labels unnecessary, courts and commentators must agree that every injunction has a nationwide “where” and a non-universal/particularized “who.” An “injunction,” unqualified, in a case brought by X would prohibit the government from enforcing the law against X anywhere X is or might go. That injunction, unqualified, would be silent and inapplicable as to Y. The law remains on the books, enforceable against everyone other than X.[70] As the Supreme Court explained in Doran v. Salem Inn,[71] “neither declaratory nor injunctive relief can directly interfere with enforcement of contested statutes or ordinances except with respect to the particular federal plaintiffs, and the State is free to prosecute others who may violate the statute.”[72]

Universal Declaratory Judgments?

The scope-of-judgment controversy has focused on universal/non-particularized injunctions. But an injunction is not the only available remedy in pre-enforcement Ex parte Young actions. A court also may grant a declaratory judgment—an order “declar[ing] the rights and other legal relations of any interested party seeking such declaration”—alone or with an injunction.[73]

In Steffel v. Thompson, the Supreme Court explained that Congress created the declaratory judgment in 1934 as a delayed reaction to the “storm of controversy” that followed Ex parte Young in 1908 and to legislative hostility toward the power of federal district courts to enjoin enforcement of constitutionally invalid state or federal laws.[74] A declaratory judgment represents a “milder alternative” to the “strong medicine” of an injunction—a federal court engages in judicial review and pronounces that a law is constitutionally invalid and should not be enforced, without imposing the coercive hammer of an injunction. A declaratory judgment is less intrusive on states and political branches because the court does not prohibit enforcement of the law, it only opines on the law’s constitutional validity and enforceability. A declaratory judgment is less coercive because it is not immediately enforceable through contempt if the government disregards the declaration or continues enforcing the challenged law.[75] Declaratory judgments function through persuasion, convincing government defendants—through the force of the court’s reasoning—to refrain from enforcing the challenged law.[76] If persuasion does not work and coercion becomes necessary, the declaratory judgment can form the basis for a subsequent injunction,[77] although it requires an additional round of litigation and a second court order.

Declaratory judgments and injunctions operate as antisuit remedies. An individual threatened with governmental enforcement of a law against her can seek a federal court order that the law is constitutionally invalid and enforcement would violate her constitutional rights. That order could be an injunction prohibiting enforcement of the constitutionally invalid law on pain of contempt, or it could be a declaration that the constitutionally invalid law should not be enforced—these are alternatives serving the same purpose.[78] In fact, many plaintiffs request a pre-enforcement injunction where the primary objective is the declaration of rights, not the coercive force of the injunction and contempt.[79]

Bray rejects “mildness” as the fundamental distinction between injunctions and declaratory judgments.[80] A declaratory judgment is a court order, having the force and effect of a final judgment.[81] A declaration that a law should not be enforced against an individual frees that individual to engage in constitutionally protected conduct without fear of enforcement; it is of no matter that the judgment does not command nonenforcement of the law or carry the pain of contempt.[82] The real distinction between the remedies is the greater detail the court can (indeed, must[83]) include in an injunction, allowing it to manage and oversee the parties and their conduct going forward. Declaratory judgments require less detail and less party management, allowing the court to pronounce the rights without more.[84]

Choosing between an injunction and a declaratory judgment thus depends on the needs of the case—whether ongoing court supervision and management is necessary. That depends, in turn, on the scope of the constitutional action. An injunction is essential in structural-reform litigation, where the purpose of the suit is judicially supervised reform of government institutions, such as schools or prisons.[85] A declaratory judgment may be sufficient in the one-off case in which an individual seeks to stop enforcement of a law against him but does not require broader judicial oversight.[86] Courts may believe the declaration of rights sufficient because the government “will do [its] duty when disputed questions have been finally adjudicated and the rights and liabilities of the parties have been finally determined.”[87]

But declaratory and injunctive relief present similar problems of remedial scope. When a court declares that a law should not be enforced because it is constitutionally invalid—even when the declaration is not accompanied by an injunction—it must determine against whom the law should not be enforced and by whom it should not be enforced. In other words, the question remains whether a court can issue a universal/non-particularized declaratory judgment or whether the declaration must be particularized to the parties. This affects how courts understand whether the government defendant is “doing its duty” and “obeying” the declaration of rights.

A declaratory judgment thus should be as party-particularized as an injunction, limited to declaring the rights of the plaintiff as against the government defendant, but not extending to declare the rights of nonplaintiffs or to bind nondefendants. The declaration establishes that a constitutionally invalid law cannot be enforced against the plaintiff by the defendant but says nothing about the enforceability of that law by or against nonparties. As John Harrison argues:

When a court enjoins an officer from enforcing a statutory rule, the effect is similar to the repeal of the rule as far as the plaintiff is concerned. When a court declares that a statutory rule is not applicable to a party because the rule is unconstitutional, the declaratory judgment again resembles a judicial act of invalidation with respect to the parties involved.[88]

This understanding is consistent with the text of the Declaratory Judgment Act. Section 2201(a) empowers the court to declare the rights or legal relations “of any interested party,”[89] meaning the determination of rights is specific to the parties, but it cannot speak to the law or its enforceability in the abstract.[90] Kevin Walsh’s argument about the nature of constitutional actions holds for declaratory relief as it does for injunctions—it is an in personam claim to stop government officials from enforcing the law against the plaintiff, not an in rem claim to stop the law itself.[91] The requirement of particularity is not unique to injunctions because any remedy resolves a discrete dispute between discrete parties to a discrete action and not beyond.[92]

“Further necessary or proper relief,” namely an injunction against the adverse party, can follow if the declaration proves insufficient to protect the rights declared against enforcement of the challenged law.[93] But only the plaintiff can seek that further relief to protect her declared rights. For example, X having obtained a declaratory judgment, it would be incoherent to allow Y to use X’s declaratory judgment to obtain an injunction protecting Y; if X must pursue the less-coercive remedy against enforcement in a separate step, so must Y. It also would be incoherent to allow X to convert her declaratory judgment into an injunction if the government obeys the judgment as to X but attempts to enforce the challenged law against Y; nonenforcement as to X has given X what she wants, so an injunction prohibiting enforcement of the law as to Y is not “necessary” to protect X’s rights. If X cannot protect Y’s constitutional rights by bringing a lawsuit to enforce those rights,[94] X cannot protect Y’s constitutional rights by converting her declaratory judgment into a non-particularized injunction protecting Y.

In endorsing particularity of federal remedies in Doran, the Supreme Court treated declaratory and injunctive relief as having the same scope and purpose—either remedy halts enforcement of the challenged law against the federal plaintiffs, while either remedy leaves the government free to enforce that law against others who violate it.[95] Moreover, the milder, less-intrusive, weaker declaratory-judgment medicine should not have broader nonparty effects than the stronger, more coercive injunctive medicine. If the injunction only prohibits government officials from enforcing the challenged law against the parties, the less-coercive declaratory judgment should only declare that government officials cannot enforce the challenged law against the parties.

In Martin v. Gross, in two consolidated individual actions, the district court declared invalid a Massachusetts law prohibiting secret recording of government officials but declined to enjoin enforcement.[96] According to the court, the declaratory judgment meant government officials could not enforce the law against the plaintiffs. Enforcement would constitute failure to “do their duty” and would provide a basis for the court to convert the declaratory judgment into an injunction, the earlier remedy having failed to persuade the government to change its conduct.[97]

But the declaratory judgment did not speak, and should not have spoken, to the validity of Massachusetts officials enforcing the law against nonparties. Such enforcement would not have constituted failure of officials to comply with the judgment or with their official duties. Had Massachusetts officials continued to enforce the law against persons who were not party to Martin, those nonparties would have to join or initiate their own actions and obtain their own judgments—declaratory, injunctive, or both—protecting them against enforcement of the law.

Judgments and Opinions, Preclusion and Precedent

A court issues two papers when it decides a case: a judgment and an opinion. Failure to distinguish these papers, their meaning, and their effects explains some of the confusion over the scope of remedies.

  1. Judgments

The binding judgment resolves constitutional litigation involving one plaintiff, one defendant, one law, and one constitutional right.[98] Will Baude argues that the root of the judicial power under Article III is the authority to “issue binding judgments and to settle legal disputes within the court’s jurisdiction. But judgments settle only those legal disputes, not others.”[99]

A judgment must be obeyed by the parties and enforced by the executive, even if erroneous.[100] A party cannot avoid its obligation to obey a judgment and cannot avoid contempt on the grounds that the judgment is wrong. Under the “collateral bar rule,” a party cannot disobey an injunction and challenge the contempt finding on the ground that the underlying injunction is erroneous or invalid.[101] An erroneous judgment can be challenged and corrected only through established judicial processes, such as appellate review of the judgment, subject to the procedural rules and limits that Congress and the courts put in place.[102] But the enjoined party must obey that judgment while appellate review proceeds unless the injunction is stayed pending review.[103] The completion of appellate review produces an Article III final judgment, which cannot be questioned or undone by the other branches[104] and is subject to limited judicial reconsideration.[105] A court may enforce that final judgment on its own or on request of a party, including by holding noncompliant parties in contempt of court and ordering them jailed.[106]

But any judgment is particularized to the litigation at issue and, therefore, to the parties to that litigation.[107] Only parties are bound to abide by the judgment, and the court’s enforcement powers—such as contempt—are limited to the parties.

This is obvious with the judgment in an enforcement action. Imagine the government initiates a proceeding to enforce a law against X, X defends on the grounds that the law is inconsistent with the Constitution, and the court agrees with X that the law is constitutionally invalid and cannot be enforced as the rule of decision in the case. The court dismisses the enforcement action and enters judgment in favor of X and against the government. But this judgment goes no further, speaking to no person other than X. And no one believes or argues otherwise.

It should follow that a judgment in a pre-enforcement Ex parte Young action for declaratory or injunctive relief to halt future enforcement of the challenged law should be as particularized as the action enforcing that law. That judgment should protect the plaintiff against enforcement but should not protect nonparties against separate enforcement. Bray argues that a “federal court should give an injunction that protects the plaintiff vis-à-vis the defendant, wherever the plaintiff and the defendant may both happen to be. The injunction should not constrain the defendant’s conduct vis-à-vis nonparties.”[108] Laycock offers a similar framing—the “court in an individual action should not globally prohibit a government agency from enforcing an invalid regulation; the court should order only that the invalid regulation not be enforced against the individual plaintiff.”[109]

A pre-enforcement action anticipates government enforcement, and the pre-enforcement remedy prevents that anticipated enforcement. The rights-holder’s offensive effort to stop enforcement before it begins is symmetrical to her defensive effort to defeat enforcement once undertaken. The rights-holder’s goal in both is to halt enforcement of the challenged law against her. If the judgment in the enforcement action would be limited to the rights-holder, the judgment in the pre-enforcement action should go no further.

The effect of either judgment is controlled by the law of judgments and the law of preclusion.[110] The final judgment resolves the dispute between parties and is enforceable to ensure that those parties comply. Preclusion then limits the right to relitigate, in a new action, the legal and factual issues considered and resolved by that judgment.[111] Like the judgment, preclusion is limited to the parties to the first action and judgment or to those with a close or privity connection with them; preclusion does not affect those unconnected to the original litigation and the judgment resolving that litigation.[112]

Courts have relaxed this rule somewhat by allowing nonmutual preclusion—a nonparty to Court I avails herself of the preclusive effect of the judgment against a party to Court I, who is denied another bite at the apple in Court II.[113] But under United States v. Mendoza,[114] nonparties cannot use preclusion against the federal government or federal officials,[115] a principle that some courts have extended to state governments and officials.[116] That is, a nonparty to the judgment in Court I cannot use nonmutual preclusion against the federal or state governments (or officials) to resolve new litigation before Court II. If Court II is considering actual or threatened enforcement against Y, Y cannot argue that the constitutional question has been resolved against the government by Court I’s judgment as to X and that preclusion binds Court II to reach the same conclusion on the constitutional question.

This point has been the target of recent scholarly criticism. Zachary Clopton and Alan Trammell independently argue that Mendoza was wrongly decided and that Congress or courts should overrule or narrow it.[117] This would allow nonparties in Court I to obtain the preclusive benefits of the constitutional ruling to bar or halt the government from future enforcement against them in Court II. Y could argue that Court I’s judgment as to X resolved the question of the law’s constitutional validity against the government after the government had a full and fair opportunity to litigate the constitutional issue. That judgment would bind Court II, requiring it to find the law constitutionally invalid and unenforceable in an action involving Y, without Y having to relitigate the constitutional issue and without the government having a second opportunity to litigate the constitutional issue on which it lost.

Clopton and Trammell both tie limiting or overruling Mendoza to the scope-of-injunction debate. Both authors argue that if, in a non-Mendoza world, a nonparty can benefit from Court I’s judgment via nonmutual preclusion, then Court I should—in an appropriate case—be able to skip the middle step and directly protect nonparties via a universal/non-particularized injunction.[118] Nonmutual preclusion and non-particularized injunctions give nonparties the benefits of Court I’s injunction; the latter protects nonparties more directly and without the need for additional litigation in Court II.

The problem with this argument is that the scope of a judgment and the scope of preclusion need not be coextensive. Expanding the preclusive effect of a judgment does not require expanding the permissible scope of that judgment. The symmetry between a judgment in an enforcement action and a judgment in the corresponding pre-enforcement action demonstrates why.

Suppose Mendoza were overruled. The government initiates an enforcement action against X, who defends on the grounds that the law at issue is constitutionally invalid; Court I agrees and dismisses the action against X. Without Mendoza and with nonmutual preclusion available against the government, Y could assert preclusion based on that judgment in a subsequent enforcement action. Y could argue that Court II is bound by the judgment of Court I on the constitutional issue, without Y having to litigate the issue and without the government having an opportunity to relitigate the issue on which it lost. But the judgment of Court I would not protect anyone other than X, regardless of how anyone might wield its preclusive effect in subsequent litigation. The government would violate that judgment, and be subject to contempt, only by attempting to enforce against X, not against any nonparty.

The same should hold if the judgment from Court I came in a pre-enforcement Ex parte Young action by X. Broadening the preclusive effect of that judgment need not broaden the judgment and injunction itself. Regardless of the posture of the litigation that produced Court I’s judgment, the preclusive effect of that judgment matters for subsequent litigation, in which Court II decides what preclusive effect to accord Court I’s prior judgment.[119]

  1. Opinions

The opinion, the second paper the court issues, is a reasoned explanation justifying the judgment. Opinions are “essays written by judges explaining why they rendered the judgment they did. The primary significance of these essays for nonjudicial actors is the guidance they provide in predicting future judicial behavior.”[120] Opinions “explain the grounds for judgments, helping other people to plan and order their affairs.”[121] This giving of reasons for an outcome represents a hallmark of judicial decision-making.[122]

Court I’s opinion—explaining why a law is constitutionally valid or invalid and justifying the judgment—serves as precedent for Court II in considering the constitutional validity of that law or a similar law in a separate action involving a different rights-holder. Precedential force varies by court.[123] A district court opinion as to the validity of a law has persuasive force for the next court, including for judges within that district, but no binding force.[124] A regional court of appeals’ opinion has binding force on other panels of that circuit (and can be reversed only by that circuit sitting en banc) and on district courts within its circuit, but persuasive force on courts of appeals and trial courts elsewhere. A Supreme Court decision has binding force on all courts in all circuits and districts and in all state courts.[125]

There are debates and confusion about when a judicial decision establishes precedent, what that precedent is, and how courts can tell.[126] While important questions, they are beyond the current point that precedent governs a judicial decision’s prospective nonparty effects—on government officials and rights-holders forming their primary conduct in the real world and on courts and parties to future litigation.[127] The effect of binding precedent (whatever its scope) continues until a decision is overruled by the issuing court or a higher court.[128] Persuasive precedent allows different courts to decide issues in their own ways, depending on how convincing they find prior opinions.

The other significant feature of precedent is that it can change.[129] While the opinion allows government officials and individuals to arrange their primary conduct going forward,[130] all must account for the possibility of change in organizing their enforcement activities.[131] This leaves everyone with freedom and with uncertainty. Laws that government officials believe unenforceable under current precedent remain on the books and may become enforceable with a change in precedent. Laws they believe enforceable may cease to be so with a change in precedent.[132] Conduct that individuals believe they may constitutionally engage in free from government restriction may lose its constitutional protection and become subject to restriction with a change in precedent.[133] Conduct that individuals believe to be prohibited may gain constitutional protection and become permissible with a change in precedent.

  1. Judgments, Opinions, and Non-Particularity

Universal/non-particularized injunctions grant courts broader authority to establish the parameters of constitutional law for other persons, beyond resolving the case at hand.[134] But the judgment need not perform that function, either directly via an injunction protecting the universe of targets or indirectly via nonmutual preclusion. Instead, the opinion performs that function. The opinion provides the decision’s wider prospective nonparty authority through the law of precedent and stare decisis. The opinion protects other rights-holders by establishing the parameters of constitutional law and constitutional rights for future litigation.[135]

Judgment and precedent operate differently within the judicial hierarchy. A district court opinion is not binding precedent, even on other judges within the district.[136] But an unstayed district court judgment[137] is and remains binding on the parties, carrying the same force and effect on those parties as an injunction that has been reviewed and affirmed by a higher court. While in effect, the district court injunction places enjoined government officials in the same position as though the injunction was affirmed on review or as though officials declined to seek review. That force remains unless and until the judgment is reversed by a higher court. Allowing universal/non-particularized injunctions thus expands the power and force of one district judge, giving her judgment force that her opinion lacks as precedent.

Similarly, Supreme Court affirmance of the district court’s judgment does not expand the injunction. If the district court entered a non-universal/particularized injunction, the Supreme Court affirms a non-universal/particularized injunction; the injunction does not gain broader scope or force to protect beyond the parties.

Supreme Court affirmance does mean all future enforcement efforts necessarily fail and all pre-enforcement actions to enjoin enforcement necessarily succeed, because all courts are bound by the Supreme Court’s opinion pronouncing that the challenged law is constitutionally defective and not enforceable.[138] But the affirmance resolves the question as a matter of the law of precedent—the binding precedential effect of the Supreme Court’s opinion on any subsequent court deciding a legal issue arising from a new government threat or attempt to enforce the law against nonparties to the first case (who are not protected by the judgment). The affirmance is not a function of the law of judgments or of a non-particularized injunction prohibiting enforcement against those nonparties.

Preclusion and precedent empower the later court. Court II decides the scope and meaning of the precedent set by Court I’s opinion and whether and how to apply it in resolving the new action. Similarly, the preclusive effect of Court I’s judgment “is usually the bailiwick of the second court.”[139] Following a judgment from Court I, the parties in Court II raise the preclusive effect of that judgment before Court II; Court II decides whether preclusion applies and the scope of that preclusion.

The trend toward universal/non-particularized injunctions reflects judicial impatience with this adjudicative process. Universality/non-particularity seeks to empower Court I to control the adjudicative process at the expense of Court II. And it extends the judgment and the law of judgments to swallow the opinion and the law of precedent.

Court I, having declared the challenged law constitutionally invalid in Case I and having extended its judgment beyond the parties, strips Court II of the opportunity (or at least the need)[140] to adjudicate the same issue involving different parties. By issuing a universal/non-particularized injunction, Court I can prevent Court II from deciding either the scope of Court I’s judgment and injunction or the meaning of its opinion as precedent. Court I issues the lone controlling judgment and opinion on the law’s constitutional validity and prohibits all enforcement of that law against anyone, subject only to reversal by its regional circuit or by the Supreme Court. By issuing the lone, universally binding judgment and opinion, Court I can guard both through its enforcement and contempt powers, cutting off any opportunity for disagreement by the parties or by another court.

Supporters of universal/non-particularized injunctions reject individual, atomized litigation of constitutional rights, fearing a flood of duplicative litigation in which each affected individual or entity must file its own action and obtain its own injunction.[141] For example, in imposing a universal/non-particularized injunction in an action challenging regulations stripping federal funds from “sanctuary cities,” the Northern District of Illinois emphasized that thirty-seven counties and cities had filed an amicus brief in that action.[142] Because all had been heard in this case, judicial economy counseled against compelling each to file a separate lawsuit to have a separate court resolve legal issues already addressed.[143]

But multiple or successive litigation is not duplicative litigation. Rather, multiple or successive litigation is necessary to create precedent—persuasive and binding—that subsequent courts can use to guide resolution of subsequent cases. Multiple precedents from multiple litigation in multiple courts allow “percolation of legal questions” through different district courts and courts of appeals, allowing each court to reach its own conclusion, pending final resolution by the Supreme Court.[144] Allowing universal/non-particularized injunctions to preempt further litigation preempts the creation of new precedent.

Although not a constitutional case, Nevada v. United States Department of Labor offers a bizarre example of the problems created by this sort of judicial reach. Several states and business organizations sued in the Eastern District of Texas, challenging—under the Fair Labor Standards Act (FLSA)—the validity of Department of Labor (DOL) regulations raising the salary line at which employees become exempt from overtime requirements (that is, broadening the class of employees entitled to overtime pay). The district court issued a universal/non-particularized preliminary injunction prohibiting enforcement of the regulations,[145] then granted summary judgment for the plaintiffs.[146]

A Chipotle employee named Carmen Alvarez, represented by counsel, filed a separate action in the District of New Jersey, alleging that the company had denied her overtime payments in violation of the DOL regulations.[147] The Eastern District of Texas found Alvarez and her lawyers in contempt of its original injunction; because the injunction was universal/non-particularized, their attempts to enforce a regulation that the court had determined was unenforceable violated a court order to which they were subject.[148]

More than a year later, the Fifth Circuit reversed the contempt finding, rejecting the argument that Alvarez or her attorneys were in privity with DOL, given the absence of evidence of an express or implied legal relationship under which DOL could be said to represent Alvarez’s interests.[149] The court added, “[m]ore generally, Chipotle’s theory that the DOL represents every worker’s legal interests through its enforcement of the FLSA so as to bind every worker in the United States to an injunction where the DOL is the only bound party lacks authoritative support.”[150] Federal labor law gave individuals unique legal rights and the opportunity to enforce those rights in private litigation when violated by a particular actor, distinct from the power of the federal government to enforce federal labor law.[151]

Had the district court’s original injunction been properly particularized, precedent could have done the work here, rather than judgment and contempt. Alvarez’s action against Chipotle should have gone forward with Chipotle urging the District of New Jersey to agree with the Eastern District of Texas’s opinion that the overtime regulations were invalid and unenforceable as to Alvarez and to resolve the lawsuit against her and in favor of Chipotle.[152] Alternatively, because the Eastern District of Texas would not have been binding authority on the District of New Jersey, the latter court could have reached a different legal conclusion about the regulations’ validity and ruled in favor of Alvarez.[153] This would have created a division of authority on the legal question. Either way, this shows the law of precedent and percolation in action.

Mila Sohoni shows that the Supreme Court has long affirmed injunctions that, by their terms, prohibit government conduct as a universal and categorical matter and are not limited to the plaintiffs.[154] This includes some of the Court’s most significant constitutional cases,[155] in which the Court intended and the public understood the Court to have stopped all enforcement of the constitutionally infirm laws, not only enforcement against the plaintiffs.

As one example, Sohoni offers West Virginia Board of Education v. Barnette,[156] in which the Court held that the First Amendment prohibited states from compelling school children to stand and recite the Pledge of Allegiance. Sohoni argues that if, the day after Barnette, the federal government compelled students in D.C. schools to salute the flag, “there is no basis in Article III for thinking that such a hypothetical case would have or should have come out any differently than Barnette did.”[157]

Sohoni is correct that the subsequent court in this hypothetical case would have reached the same result as Barnette. But the reason would have been the law of precedent and the binding nature of Supreme Court opinions, not the scope of the injunction. The Court, having declared in Barnette that compulsory flag salutes violate the First Amendment, established binding precedent; that precedent required lower courts to declare invalid similar government attempts to compel the salute. The new dispute would have been litigated in the District of the District of Columbia, with that court resolving new litigation involving new parties based on binding precedent. The court would apply the binding precedent of the Barnette opinion and conclude that D.C.’s attempted compulsion violated the First Amendment.

In fact, however, the hypothetical D.C. case would have reached the same result had the Barnette injunction been non-universal/particularized. But this would have had nothing to do with the injunction and judgment in Barnette itself. The district court in West Virginia—which issued and must oversee and manage the injunction affirmed in Barnette—would play no role in the new D.C. dispute. The judgment and injunction from Barnette would be irrelevant to any subsequent litigation. The opinion and precedent do the work.

This distinction addresses Pedro’s discussion of a real-world example, in which a district court in 1963 attempted to “redecide” Brown v. Board of Education.[158] The Fifth Circuit quickly and forcefully reversed. The court of appeals stated that “the District Court was bound by the decision of the Supreme Court in Brown. We reiterate that no inferior federal court may refrain from acting as required by that decision even if such a court should conclude that the Supreme Court erred either as to its facts or as to the law.”[159] The court continued:

Thus was the Savannah case ended then, and there it must end now. We do not read the major premise of the decision of the Supreme Court in the first Brown case as being limited to the facts of the cases there presented. We read it as proscribing segregation in the public education process on the stated ground that separate but equal schools for the races were inherently unequal. This being our interpretation of the teaching of that decision, it follows that it would be entirely inappropriate for it to be rejected or obviated by this court.[160]

But, as with Sohoni’s hypothetical reaction to Barnette, the Fifth Circuit’s analysis sounds in precedent and the binding force of the Brown opinion on lower courts, not in judgment and the force of the injunction affirmed in Brown. The district court in Stell erred in allowing the parties to offer evidence and arguments showing the wrongness of Brown.[161] The Fifth Circuit’s point was that it and the district court were bound to follow and apply Brown to new litigation involving new parties and new facts, such that a new injunction against the Savannah-Chatham Board’s actions was legally required. That court did not say that the Board was subject to, and thus in violation of, an existing injunction. Again, that would have been the bailiwick of the trial court in the District of Kansas overseeing the injunction against the Topeka Board that the Supreme Court affirmed in Brown.

An Unnecessary Distraction

Universality/non-particularity ultimately represents an unnecessary and unfortunate distraction. Courts employ language weighted with rhetorical baggage and uncertain meaning[162] when they label an injunction nationwide or universal but fail to recognize the consequences of that rhetoric. Accepting that the Supreme Court has been approving, on paper, universal injunctions for years,[163] neither courts nor litigants have taken that universality seriously, then or now. Nor do they follow universality to its logical conclusion. The purportedly universal/non-particularized scope of the injunction thus diverts attention from significant legal and constitutional issues while having no practical effect.

Litigation over Trump Administration regulations establishing a religious exemption from the Affordable Care Act’s contraception mandate illustrates the problem. Two district courts agreed the regulations violated the Administrative Procedure Act and enjoined enforcement. The Eastern District of Pennsylvania made the injunction universal (while incorrectly labeling it “nationwide”) and offered detailed justifications for that universality.[164] The Third Circuit affirmed, including as to scope.[165] The Northern District of California particularized its injunction to the plaintiff states.[166]

On review of the California case, the Ninth Circuit ordered the parties to brief whether the Pennsylvania universal/non-particularized injunction mooted the California case and whether the mootness analysis changed because the universal injunction came from a district court in another regional circuit.[167] The court of appeals divided on the answer; the majority held that the case was not moot.[168] But the path to that result exposed the problems that arise when courts make their injunctions universal/non-particularized, fail to take non-particularity seriously, and misuse and confuse the nomenclature.

Dissenting in the Ninth Circuit, Judge Andrew Kleinfeld captured the real problem:

That nationwide injunction means that the preliminary injunction before us is entirely without effect. If we affirm, as the majority does, nothing is stopped that the Pennsylvania injunction has not already stopped. Were we to reverse, and direct that the district court injunction be vacated, the rule would still not go into effect, because of the Pennsylvania injunction. Nothing the district court in our case did, or that we do, matters. We are talking to the air, without practical consequence. Whatever differences there may be in the reasoning for our decision and the Third Circuit’s have no material significance, because they do not change the outcome at all; the new regulation cannot come into effect.[169]

The majority missed the point in several respects.

First, it insisted that the California and Pennsylvania injunctions complemented rather than conflicted.[170] The real problem, however, was that the injunctions duplicated: The universal Pennsylvania injunction, purporting to protect the California plaintiffs, rendered a California injunction superfluous. The California plaintiffs faced no real or imminent harm from the challenged regulations while the Pennsylvania injunction remained in effect. The United States could not and would not enforce the revised mandate against the California plaintiffs; any attempt to do so would violate that universal/non-particularized injunction and could be halted with a motion asking the Eastern District of Pennsylvania to enforce its order.

Critics of universal/non-particularized injunctions invoke the risk of conflicting decisions—either plaintiffs jump from court to court until they find a judge who agrees the challenged law is constitutionally invalid and will universally enjoin enforcement[171] (what Bray refers to as “shopping until the statute drops”)[172] or one court orders the government to take some action and another court orders the government to refrain from that action, placing the government in a position where it will violate a court order no matter what it does.[173]

The contraception-mandate cases reveal a third problem. Because of the universal/non-particularized injunction, the rights of the California plaintiffs no longer were violated or at risk of being violated. And if they were, the Eastern District—the court that issued the universal/non-particularized injunction—could protect those rights by enforcing its judgment. The litigation in the Ninth Circuit and the Northern District of California was superfluous because the relief in that case did not provide California with new legal protections that the existing overbroad Eastern District injunction did not provide.[174]

Second, the Ninth Circuit majority declined to conclude that it lacked jurisdiction to issue an injunction affecting conduct covered by a prior injunction, because that would be akin to declaring that the Eastern District (and the affirming Third Circuit) “plainly acted beyond its jurisdiction” in issuing its universal injunction.[175] The court preferred not to intrude on a sister federal court. The Ninth Circuit was correct in one respect: the Eastern District overstepped in issuing that broad injunction. But while the Eastern District and Third Circuit’s error lies beyond the Ninth Circuit’s corrective power, it should not provide a basis for that court to commit the distinct error of issuing an injunction lacking practical effect.

Third, the Ninth Circuit majority expressed concern that the Eastern District injunction might go away, either because it was reversed on appeal or because it was preliminary and might be superseded by denial of a permanent injunction.[176] That would leave the California plaintiffs unprotected and in need of a new injunction. But the possibility of future need does not change present reality—at the time the Ninth Circuit was asked to rule, the California plaintiffs could not be the targets of enforcement, were not being injured or suffering violations of their rights, and did not need the new injunction. A properly scoped Pennsylvania injunction would not have affected the California plaintiffs (who were not party to that case), making it clear to the Ninth Circuit that they needed the California injunction to protect their rights.

Finally, although not discussed in the opinion, the Ninth Circuit suffered from nomenclature confusion when it questioned the effect of the competing injunction coming from a district court outside the circuit. All injunctions are (and should be) nationwide, in that they protect a protected party everywhere he is or he may go.[177] A plaintiff protected against enforcement of a law is protected against enforcement wherever she is, and the bound government is prohibited from enforcing wherever the target is. It follows that if a court has the power to protect nonparties (through universal/non-particularized injunctions), then it protects those nonparties everywhere they go. If the Eastern District had the power to issue this universal/non-particularized injunction prohibiting enforcement against all potential targets of the regulations, that injunction protected those targets everywhere they might have gone, including within California and the Ninth Circuit.

The Ninth Circuit’s consideration and resolution of this question exposes the emptiness of the scope-of-injunction debate. The practice of issuing universal/non-particularized injunctions has not produced a world in which one injunction ends all litigation by prohibiting enforcement of the challenged law as to all persons. There has been no loss of percolation, contrary to critics’ fears—multiple plaintiffs continue to bring multiple actions, producing multiple court orders judging the constitutional validity and enforceability of the challenged laws.[178]

That parties continue to pursue additional litigation and that courts continue to consider constitutional questions and to issue orders, however, reveals the real problem. No one is serious about purportedly universal/non-particularized injunctions and no one follows universality/non-particularity to its true conclusion. Not the courts issuing purportedly universal injunctions who choose not to wield their power to enforce their orders and hold government officials in contempt. Not the plaintiffs who continue pursuing separate litigation despite an existing universal/non-particularized injunction that by its terms protects them. And not the later courts who hear subsequent cases and issue subsequent injunctions that are, in practice, superfluous.

Conclusion

The real question becomes why courts should or do bother with universality or non-particularity. Courts and parties could avoid uncertainty and confusion—and end an academic and jurisprudential debate—by keeping their injunctions to themselves. Orders and judgments, whether injunctions or declaratory judgments, should be particularized to the parties before the court and protect those parties everywhere they are or might go. But those orders and judgments should not protect or purport to protect anyone else. They should be accompanied by an opinion having some precedential effect (depending on the court) on future litigation involving different persons in different courts.

By adopting the approach proposed here, courts need not worry about qualifiers such as “universal” or “particularized” or “non-particularized” or “nationwide” or “cosmic.” They can simply issue “injunctions.” And everyone will know what an injunction is and what should be the proper scope of that injunction.

  1. *Professor of Law, FIU College of Law. Thanks to Suzette Malveaux, the University of Colorado Law School, and the University of Colorado Law Review for inviting me to participate in the 2019 Rothgerber Constitutional Law Conference.
  2. . “Law” throughout the paper includes all enforceable legal rules regardless of source: statutes, administrative regulations and policies, executive policies and orders, and judge-made common law.
  3. . E.g., Regents of the Univ. of Cal. v. U.S. Dept of Homeland Sec., 908 F.3d 476, 511–12 (9th Cir. 2018); Pennsylvania v. Trump, 351 F. Supp. 3d 791, 830–35 (E.D. Pa. 2019); Batalla Vidal v. Nielsen, 279 F. Supp. 3d 401, 437–38 (E.D.N.Y. 2018); City of Chicago v. Sessions, 264 F. Supp. 3d 933, 951–52 (N.D. Ill. 2017); City of Santa Clara v. Trump, 250 F. Supp. 3d 497, 539–40 (N.D. Cal. 2017).
  4. . E.g., Pennsylvania, 351 F. Supp. 3d at 830–35; City of Chicago v. Sessions, 2017 WL 4572208, *4 (N.D. Ill. 2017).
  5. . E. Bay Sanctuary Covenant v. Barr, 934 F.3d 1026, 1028–30 (9th Cir. 2019); California v. Health & Human Servs., 351 F. Supp. 3d 1267, 1300 (N.D. Cal. 2019); see also City of Chicago v. Sessions, 321 F. Supp. 3d 855, 881 (N.D. Ill. 2018) (staying nationwide scope of injunction).
  6. . Dep’t of Homeland Security v. New York, 140 S. Ct. 599, 600 (2020) (Gorsuch, J., joined by Thomas, J., concurring in the grant of stay); Trump v. Hawaii, 138 S. Ct. 2392, 2424–25 (2018) (Thomas, J., concurring); but see id. at 2446 n.13 (Sotomayor, J., dissenting).
  7. . E.g., Zachary D. Clopton, National Injunctions and Preclusion, 118 Mich. L. Rev. 1, 5–7 (2019); Amanda Frost, In Defense of Nationwide Injunctions, 93 N.Y.U. L. Rev. 1065, 1069 (2018); Suzette M. Malveaux, Class Actions, Civil Rights, and the National Injunction, 131 Harv. L. Rev. F. 56 (2017); Portia Pedro, Toward Establishing a Pre-Extinction Definition of “Nationwide Injunctions, 91 U. Colo. L. Rev. 847 (2020); James E. Pfander & Jacob Wentzel, The Common Law Origins of Ex Parte Young, Stan. L. Rev. (forthcoming 2020) (manuscript at 56–57); Mila Sohoni, The Lost History of the “Universal” Injunction, 133 Harv. L. Rev. 920, 924 (2020); Alan M. Trammell, Demystifying Nationwide Injunctions, 98 Tex. L. Rev. 67, 73–74 (2019).
  8. . E.g., Douglas Laycock, Modern American Remedies 436 (4th ed. 2010); Samuel L. Bray, Multiple Chancellors: Reforming the National Injunction, 131 Harv. L. Rev. 417, 469 (2017); Ronald A. Cass, Nationwide Injunctions’ Governance Problems: Forum-Shopping, Politicizing Courts, and Eroding Constitutional Structure, 27 Geo. Mason L. Rev. (forthcoming 2020); Michael T. Morley, Disaggregating Nationwide Injunctions, 71 Ala. L. Rev. 1, 7–8 (2019) [hereinafter Morley, Disaggregating]; Michael T. Morley, Nationwide Injunctions, Rule 23(b)(2), and the Remedial Powers of the Lower Courts, 97 B.U. L. Rev. 615, 620 (2017) [hereinafter Morley, Nationwide]; Howard M. Wasserman, Precedent, Non-Universal Injunctions, and Judicial Departmentalism: A Model of Constitutional Adjudication, 23 Lewis & Clark L. Rev. 1077, 1080–81, 1093–94 (2020) [hereinafter Wasserman, Departmentalism]; Howard M. Wasserman, “Nationwide” Injunctions Are Really “Universal” Injunctions and They Are Never Appropriate, 22 Lewis & Clark L. Rev. 335, 353 (2018) [hereinafter Wasserman, “Nationwide”].
  9. . Wasserman, Departmentalism, supra note 7, at 1080–81, 1093–94; Wasserman, “Nationwide”, supra note 7, at 353.
  10. . E.g., Bond v. United States, 564 U.S. 211 (2014); United States v. Stevens, 559 U.S. 460 (2010); United States v. Morrison, 529 U.S. 598 (2000); United States v. Lopez, 514 U.S. 549 (1995); Texas v. Johnson, 491 U.S. 397 (1989).
  11. . E.g., United States v. Windsor, 570 U.S. 744 (2013); cf. Huffman v. Pursue Ltd., 420 U.S. 592, 604–05 (1975).
  12. . E.g., Middlesex Cty. Ethics Comm’n v. Garden State Bar Ass’n, 457 U.S. 423 (1982).
  13. . E.g., Snyder v. Phelps, 562 U.S. 443 (2011); Miami Herald Publ’g Co. v. Tornillo, 418 U.S. 241 (1974); N.Y. Times v. Sullivan, 376 U.S. 254 (1964).
  14. . Ex Parte Young, 209 U.S. 123 (1908).
  15. . 28 U.S.C. §§ 2201–2202 (2018).
  16. . E.g., Obergefell v. Hodges, 135 S. Ct. 2584 (2015); Reno v. ACLU, 521 U.S. 844 (1997); Doran v. Salem Inn, Inc., 422 U.S. 922 (1975); Steffel v. Thompson, 415 U.S. 452 (1974).
  17. . 28 U.S.C. § 2201(a).
  18. . Bray, supra note 7, at 449–50; John Harrison, Ex Parte Young, 60 Stan. L. Rev. 989, 990, 1014–15 (2008); Va. Office for Prot. & Advocacy v. Stewart (“VOPA”), 563 U.S. 247, 262 (2011) (Kennedy, J., concurring).
  19. . VOPA, 563 U.S. at 256–57; Pfander & Wentzel, supra note 6 (manuscript at 49); see, e.g., United States v. Sanchez-Gomez, 138 S. Ct. 1532 (2018); City of Chicago v. Sessions, 264 F. Supp. 3d 933 (N.D. Ill. 2017); Miller v. Davis, 123 F. Supp. 3d 924 (E.D. Ky. 2015).
  20. . Bond v. United States, 564 U.S. 211, 214 (2014); United States v. Morrison, 529 U.S. 598, 601–02 (2000).
  21. . Obergefell, 135 S. Ct. at 2608; United States v. Stevens, 559 U.S. 460, 481–82 (2010); ACLU, 521 U.S. at 849; Brown v. Bd. of Educ., 347 U.S. 483, 495 (1954).
  22. . Bray, supra note 7, at 469.
  23. . Laycock, supra note 7, at 276.
  24. . Trump v. Hawaii, 138 S. Ct. 2392, 2424–25 n.1 (2018) (Thomas, J., concurring).
  25. . See Frost, supra note 6, at 1071; Trammell, supra note 6, at 67–71.
  26. . Bray, supra note 7, at 419, 419 n.5.
  27. . Hawaii, 138 S. Ct. at 2424–25, n.1; see also Wasserman, “Nationwide”, supra note 7, at 350.
  28. . Transcript of Oral Argument at 72–73, Hawaii, 138 S. Ct. 2392 (No. 17-965).
  29. . Dep’t of Homeland Security v. New York, 140 S. Ct. 599, 600 (2020) (Gorsuch, J., joined by Thomas, J., concurring in the grant of stay).
  30. . Morley, Disaggregating, supra note 7, at 9–10; Morley, Nationwide, supra note 7, at 621.
  31. . Wasserman, Departmentalism, supra note 7, at 1094–95.
  32. . 28 U.S.C. § 2201(a) (2018).
  33. . John Harrison, Severability, Remedies, and Constitutional Adjudication, 83 Geo. Wash. L. Rev. 56, 82–83, 83 n.130 (2014).
  34. . See Fed. R. Civ. P. 65(d)(2).
  35. . Bray, supra note 7, at 419 n.5.
  36. . Morley, Disaggregating, supra note 7, at 9.
  37. . N.w. Immigrant Rights Project v. Sessions, 2017 WL 3189032, *2–6 (W.D. Wash. 2017).
  38. . Id. at *1.
  39. . Transcript of Oral Argument, supra note 27, at 72–73.
  40. . Wasserman, “Nationwide”, supra note 7, at 350.
  41. . Trump v. Hawaii, 138 S. Ct. 2392, 2424–25 n.1 (2018) (Thomas, J., concurring).
  42. . Transcript of Oral Argument, supra note 27, at 72–73; see also Dep’t of Homeland Security v. New York, 140 S. Ct. 599, 600 (2020) (Gorsuch, J., concurring in the grant of stay).
  43. . Kanuszewski v. Mich. Dep’t of Health & Human Servs., 927 F.3d 396, 426 (6th Cir. 2019) (Rogers, J., concurring).
  44. . Wasserman, Departmentalism, supra note 7, at 1093–94.
  45. . Stephen I. Vladeck, The Solicitor General and the Shadow Docket, 133 Harv. L. Rev. 123, 139 (2019).
  46. . Michael T. Morley, DeFacto Class Actions? Plaintiff- and Defendant-Oriented Injunctions in Voting Rights, Election Law, and Other Constitutional Cases, 39 Harv. J.L. & Pub. Pol’y 487, 500 (2016).
  47. . Id.; Morley, Nationwide, supra note 7, at 616.
  48. . Northwest Immigrant Rights Project v. Sessions, No. C17-716, 2017 WL 3189032, at *6–7 (W.D. Wash. July 27, 2017).
  49. . Compare California v. Health & Human Servs., 351 F. Supp. 3d 1267, 1300 (N.D. Cal. 2019), with Pennsylvania v. Trump, 351 F. Supp. 3d 791, 830–31 (E.D. Pa. 2019).
  50. . 283 F. Supp. 3d 1007, 1012–13 (D. Kan. 2018).
  51. . Id. at 1012.
  52. . Id. at 1027.
  53. . Cf. Wasserman, “Nationwide”, supra note 7, at 366–69.
  54. . Id. at 368–69.
  55. . Morley, Disaggregating, supra note 7, at 9–10.
  56. . Id.
  57. . E. Bay Sanctuary Covenant v. Barr, 391 F. Supp. 3d 974, 976 (N.D. Cal. 2019); E. Bay Sanctuary Covenant v. Barr, 385 F. Supp. 3d 922, 929–30, 960 (N.D. Cal. 2019).
  58. . E. Bay, 385 F. Supp. 3d at 960.
  59. . E. Bay Sanctuary Covenant v. Barr, 934 F.3d 1026, 1028–29 (9th Cir. 2019).
  60. . Id. at 1030–31.
  61. . E. Bay, 391 F. Supp. 3d at 982–84.
  62. . This point was rendered moot when the Supreme Court stayed the injunction in full. Barr v. E. Bay Sanctuary Covenant 140 S. Ct. 3 (2019).
  63. . Pedro, supra note 6, 863–64, 868–69.
  64. . Id. at 69.
  65. . Dep’t of Homeland Security v. New York, 140 S. Ct. 599, 600 (2020) (Gorsuch, J., joined by Thomas, J., concurring in the grant of stay); Trump v. Hawaii, 138 S. Ct. 2392, 2424–25 (2018) (Thomas, J., concurring).
  66. . Bray, supra note 7, at 469; Cass, supra note 7, at 5; Morley, Nationwide, supra note 7, at 616; Morley, Disaggregating, supra note 7, at 7–8; Wasserman, “Nationwide”, supra note 7, at 353; Wasserman, Departmentalism, supra note 7, at 1093–94.
  67. . William Baude, The Judgment Power, 98 Geo. L.J. 1807, 1811 (2008); Bray, supra note 7, at 469; Robert. J. Pushaw, Jr., The Inherent Powers of Federal Courts and the Structural Constitution, 86 Iowa L. Rev. 735, 860 (2001).
  68. . Bray, supra note 7, at 469.
  69. . This is subject to litigation and procedural mechanisms, expanding who qualifies as the plaintiffs so as to gain the protections of the injunction. Wasserman, Departmentalism, supra note 7, at 1098–1104; Wasserman, “Nationwide”, supra note 7, at 366–75.
  70. . Harrison, supra note 32, at 87–88.
  71. . 422 U.S. 922 (1975).
  72. . Id. at 931.
  73. . 28 U.S.C. § 2201(a) (2018).
  74. . Steffel v. Thompson, 415 U.S. 452, 466–67 (1974). Congress responded one year after Ex parte Young by creating three-judge district courts for actions seeking injunctions against enforcement of state laws, 28 U.S.C. § 2281, later extending it to actions seeking to enjoin enforcement of federal laws. 28 U.S.C. § 2282. See Michael E. Solimine, Congress, Ex Parte Young, and the Fate of the Three-Judge District Court, 70 Pitt. L. Rev. 101, 113 (2008).
  75. . Perez v. Ledesma, 401 U.S. 82, 111–12 (1971) (Brennan, J., concurring in part and dissenting in part); Cass, supra note 7, at 49.
  76. . Steffel, 415 U.S. at 470–71.
  77. . 28 U.S.C. § 2202; Samuels v. Mackell, 401 U.S. 66, 72–73 (1971).
  78. . Steffel, 415 U.S. at 466; Mackell, 401 U.S. at 73; Harrison, supra note 32, at 87–88.
  79. . Emily Sherwin & Samuel L. Bray, Ames, Chafee, and Re on Remedies 1071 (3d ed. 2020).
  80. . Samuel L. Bray, The Myth of the Mild Declaratory Judgment, 63 Duke L.J. 1091, 1120–21 (2014).
  81. . 28 U.S.C. § 2201(a).
  82. . Bray, supra note 79, at 1120–21.
  83. . Fed. R. Civ. P. 65(d)(1)(B), (C).
  84. . Bray, supra note 79, at 1124–25.
  85. . Id. at 1128; see, e.g., Brown v. Plata, 563 U.S. 493 (2011); Malcolm M. Feely & Edward L. Rubin, Judicial Policy Making and the Modern State: How the Courts Reformed America’s Prisons (1998).
  86. . Bray, supra note 79, at 1124–25.
  87. . Martin v. Gross, 380 F. Supp. 3d 169, 172 (D. Mass. 2019).
  88. . Harrison, supra note 32, at 87–88.
  89. . 28 U.S.C. § 2201(a) (2018).
  90. . Harrison, supra note 32, at 82–83, 82 n.130.
  91. . Kevin C. Walsh, Judicial Departmentalism: An Introduction, 58 Wm. & Mary L. Rev. 1713, 1725 (2017).
  92. . Bray, supra note 7, at 469; Cass, supra note 7, at 7; Morley, Nationwide, supra note 7, at 616; Morley, Disaggregating, supra note 7, at 7–8; Wasserman, “Nationwide”, supra note 7, at 353; Wasserman, Departmentalism, supra note 7, at 1094–96.
  93. . 28 U.S.C. § 2202; Samuels v. Mackell, 401 U.S. 66, 72 (1971).
  94. . City of L.A. v. Lyons, 461 U.S. 95, 105 (1983).
  95. . Doran v. Salem Inn, Inc., 422 U.S. 922, 931 (1975).
  96. . Martin v. Gross, 380 F. Supp. 3d 169, 172–73, (D. Mass. 2019).
  97. . Id.
  98. . Baude, supra note 66, at 1811; Gary Lawson & Christopher D. Moore, The Executive Power of Constitutional Interpretation, 81 Iowa L. Rev. 1267, 1327 (1996); Thomas W. Merrill, Judicial Opinions as Binding Law and as Explanations for Judgments, 15 Cardozo L. Rev. 43, 44–45 (1993); Pushaw, supra note 66, at 860; see also Fed. R. Civ. P. 58.
  99. . Baude, supra note 66, at 1811.
  100. . Id. at 1826; Harrison, supra note 32, at 87; Pushaw, supra note 66, at 860; Wasserman, Departmentalism, supra note 7, at 1105, 1110–11; but see Lawson & Moore, supra note 97, at 1325.
  101. . Walker v. City of Birmingham, 388 U.S. 307, 314–16 (1967).
  102. . Id. at 314.
  103. . Nken v. Holder, 556 U.S. 418, 434 (2009); Josh Blackman & Howard M. Wasserman, The Process of Marriage Equality, 43 Hastings Const. L.Q. 243, 283 (2017).
  104. . Plaut v. Spendthrift Farm, Inc., 514 U.S. 211, 218–19 (1995); Lawson & Moore, supra note 97, at 1319.
  105. . See Fed. R. Civ. P. 60(b); Rufo v. Inmates of Suffolk Cty. Jail, 502 U.S. 367, 379–80 (1992).
  106. . Fed. R. Civ. P. 70; 28 U.S.C. §§ 401–402 (2018); see Juidice v. Vail, 430 U.S. 327, 335–36 (1977); Miller v. Caudill, 936 F.3d 442, 446 (6th Cir. 2019); Reynolds v. Roberts, 207 F.3d 1288, 1297 (11th Cir. 2000).
  107. . Baude, supra note 66, at 1826; Pushaw, supra note 66, at 860.
  108. . Bray, supra note 7, at 469.
  109. . Id. at 469 n.10.
  110. . Richard H. Fallon, Jr., Fact and Fiction About Facial Challenges, 99 Calif. L. Rev. 915, 924 n.31 (2011) [hereinafter Fallon, Fact]; Richard H. Fallon, Jr., As- Applied and Facial Challenges and Third-Party Standing, 113 Harv. L. Rev. 1321, 1339–40 (2000) [hereinafter Fallon, As Applied]; Harrison, supra note 32, at 88; Adam N. Steinman, Case Law, 97 B.U. L. Rev. 1947, 1957 (2017).
  111. . Clopton, supra note 6, at 10–13; Trammell, supra note 6, at 92–95; see United States v. Mendoza, 464 U.S. 154, 158 (1984).
  112. . Mendoza, 464 U.S. at 159–60; Clopton, supra note 6, at 10–13; Harrison, supra note 32, at 88.
  113. . Clopton, supra note 6, at 12–13; Trammell, supra note 6, at 95.
  114. . 464 U.S. 154 (1984).
  115. . Id. at 162.
  116. . Idaho Potato Comm’n v. G&T Terminal Packaging, Inc., 425 F.3d 708, 713–14 (9th Cir. 2005); Morley, Nationwide, supra note 7, at 623–24.
  117. . Clopton, supra note 6, at 37; Trammell, supra note 6, at 99–102.
  118. . Clopton, supra note 6, at 6, 19, 36–38; Trammell, supra note 6, at 102, 120–22.
  119. . Smith v. Bayer Corp., 564 U.S. 299, 307 (2011).
  120. . Merrill, supra note 97, at 62. Oliver Wendell Holmes famously defined all law as “the prediction of the incidence of the public force through the instrumentality of the courts.” Oliver Wendell Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 457 (1897).
  121. . Baude, supra note 66, at 1844; Lawson & Moore, supra note 97, at 1327; Merrill, supra note 97, at 44–45, 62.
  122. . Lon L. Fuller, The Forms and Limits of Adjudication, 92 Harv. L. Rev. 353, 387–88 (1978).
  123. . Randy J. Kozel, The Scope of Precedent, 113 Mich. L. Rev. 179, 185–86 (2014).
  124. . Camreta v. Greene, 563 U.S. 692, 709 n.7 (2011); ASARCO Inc. v. Kadish,

    490 U.S. 605, 617 (1989); Fallon, Fact, supra note 109, at 924 n.31; Fallon, As- Applied, supra note 109, at 1340; but see Morley, Disaggregating, supra note 7, at 53 (proposing that district court opinions be given intradistrict or intracircuit stare decisis effect).

  125. . Cass, supra note 7, at 47–49; Fallon, Fact, supra note 109, at 923 n.31; Fallon, As-Applied, supra note 109, at 1339; Harrison, supra note 32, at 88; Steinman, supra note 109, at 1957.
  126. . See generally Randy J. Kozel, Settled Versus Right: A Theory of Precedent (2017); Steinman, supra note 109, at 1950; Richard M. Re, Narrowing Supreme Court Precedent From Below, 104 Geo. L.J. 921 (2016); Richard M. Re, Narrowing Precedent in the Supreme Court, 114 Colum. L. Rev. 1861 (2014); Frederick Schauer, Precedent, 39 Stan. L. Rev. 571 (1987).
  127. . Baude, supra note 66, at 1844; Lawson & Moore, supra note 97, at 1327; Merrill, supra note 97, at 44–45, 62.
  128. . Jonathan F. Mitchell, The Writ-of-Erasure Fallacy, 104 Va. L. Rev. 933, 946–47, 1017 (2018).
  129. . Walsh, supra note 90, at 1715.
  130. . Baude, supra note 66, at 1844; Lawson & Moore, supra note 97, at 1327; Merrill, supra note 97, at 44–45, 62.
  131. . Mitchell, supra note 127, at 1008.
  132. . Id. at 987.
  133. . Id. at 948, 987–88.
  134. . Frost, supra note 6, at 1087–89, 1092–95; Malveaux, supra note 6, at 62–63.
  135. . Baude, supra note 66, at 1844; Fallon, Fact, supra note 109, at 923 n.31; Fallon, As Applied, supra note 109, at 1339; Merill, supra note 97, at 44–45; Morley, Disaggregating, supra note 7, at 24–25, 39–40.
  136. . See supra notes 125–130.
  137. . Nken v. Holder, 556 U.S. 418, 434 (2009); Blackman & Wasserman, supra note 102, at 283.
  138. . Blackman & Wasserman, supra note 102, at 252–53; Steinman, supra note 109, at 1957; Walsh, supra note 90, at 1715, 1727–28.
  139. . Smith v. Bayer Corp., 564 U.S. 299, 307 (2011) (emphasis in original); see supra notes 117–119 and accompanying text.
  140. . See infra Part IV.
  141. . City of Chicago v. Sessions, 2017 WL 4572208, *3 (N.D. Ill. Oct. 13, 2017); Frost, supra note 6, at 1101; Malveaux, supra note 6, at 61–62; Pedro, supra note 6, at 851–52, 865; Trammell, supra note 6, at 112.
  142. . City of Chicago, 2017 WL 4572208, at *3.
  143. . Id.
  144. . Bray, supra note 79, at 420; Wasserman, “Nationwide”, supra note 7, at 383.
  145. . Nevada v. U.S. Dep’t of Labor, 218 F. Supp. 3d 520 (E.D. Tex. 2016).
  146. . Nevada v. U.S. Dep’t of Labor, 275 F. Supp. 3d 795 (E.D. Tex. 2017).
  147. . Nevada v. U.S. Dep’t of Labor, 321 F. Supp. 3d 709, 720 (E.D. Tex. 2018).
  148. . Id. at 720, 726.
  149. . Texas v. U.S. Dep’t of Labor, 929 F.3d 205, 208 (5th Cir. 2019).
  150. . Id. at 213.
  151. . Id.
  152. . To the extent there was privity among DOL, Alvarez, and her attorneys, that should have been left for a preclusion analysis, not judgment and contempt, in the second court. Chipotle could have urged the District of New Jersey to apply nonmutual defensive preclusion based on that privity but left that court to determine the first injunction’s preclusive effect. But the Eastern District of Texas did not want to surrender that control. And the overbroad universal/non-particularized injunction it issued in Nevada’s lawsuit allowed it to maintain that control over the legal issues in a subsequent case, even as to nonparties.
  153. . That would have remained true had the Fifth Circuit affirmed on appeal in Nevada. A court of appeals decision is persuasive to, but not binding on, a district court in a different circuit; the District of New Jersey is not in the Fifth Circuit. See supra notes 122–125.
  154. . Sohoni, supra note 6, at 926–28; Pfander & Wentzel, supra note 6 (manuscript at 57).
  155. . Sohoni, supra note 6, at 989–90; see, e.g., W. Va. Bd. of Educ. v. Barnette, 319 U.S. 624 (1943); Hague v. CIO, 307 U.S. 496 (1939).
  156. . 319 U.S. 624 (1943).
  157. . Sohoni, supra note 6, at 991.
  158. . Pedro, supra note 6, at 854; see Stell v. Savannah-Chatham Cty. Bd. of Educ., 220 F. Supp. 667 (S.D. Ga. 1963), rev’d, 333 F.2d 55 (5th Cir. 1964).
  159. . Stell, 333 F.2d at 61.
  160. . Id.
  161. . Id.
  162. . Pedro, supra note 6, at 870–71.
  163. . Sohoni, supra note 6, at 924–26.
  164. . Pennsylvania v. Trump, 351 F. Supp. 3d 791, 830–35 (E.D. Pa. 2019).
  165. . Pennsylvania v. President of the United States, 930 F.3d 543, 575–76 (3d Cir. 2019).
  166. . California v. Health & Human Servs., 351 F. Supp. 3d 1267, 1300–01 (N.D. Cal. 2019).
  167. . California v. U.S. Dep’t of Health & Human Servs., Order of Apr. 29, 2019.
  168. . California v. U.S. Dep’t of Health & Human Servs., 941 F.3d 410, 422 (9th Cir. 2019).
  169. . Id. at 434 (Kleinfeld, J., dissenting).
  170. . Id. at 422.
  171. . Bray, supra note 79, at 460; Wasserman, “Nationwide”, supra note 7, at 363–64.
  172. . Bray, supra note 79, at 460.
  173. . Id. at 362; Wasserman, “Nationwide”, supra note 7, at 383–84.
  174. . The majority rejected mootness because the Eastern District injunction was preliminary and the subject of a Supreme Court petition for writ of certiorari. California v. U.S. Dep’t of Health & Human Servs., 941 F.3d 410, 422–23 (9th Cir. 2019). Bray argues that the proper focus should have been not Article III mootness but equitable mootness and the lack of an equitable need for the court to address or resolve the issue. Samuel Bray, National Injunctions and Equitable Mootness, Volokh Conspiracy (Apr. 30, 2019), https://reason.com/2019/04/30/national-injunctions-and-equitable-mootness/ [https://perma.cc/V9AV-CLYN].
  175. . California, 941 F.3d at 422.
  176. . Id. at 423.
  177. . See supra Part I.
  178. . Frost, supra note 6, at 1108–09; Spencer E. Amdur & David Hausman, Nationwide Injunctions and Nationwide Harm, 131 Harv. L. Rev. F. 49, 53 (2017).