The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent


A Reply to Bray's Response to The Lost History of the "Universal" Injunction


Dear readers, you may recall Professor Mila Sohoni's forthcoming article in the Harvard Law Review, The Lost History of the "Universal" Injunction. A few days ago I wrote a response. Mila has now written a reply, which I post below. Both the response and the reply are cross-posted on the Notice & Comment blog. To vigorous scholarly debate!

In Multiple Chancellors, Professor Bray argued that federal courts should give only a "plaintiff-protective injunction, enjoining the defendant's conduct only with respect to the plaintiff," "[n]o matter how important the question and how important the value of uniformity," with respect to federal defendants. (MC, p. 420; p. 424 (noting that this rule would "logically apply" to state defendants).) He cited Article III in support of this rule: "Article III gives the judiciary authority to remedy the wrongs done to those litigants, not the wrongs done to others." (MC, p. 471.) It will therefore not come as a surprise that Bray has some critical thoughts about my draft article, The Lost History of the "Universal" Injunction, 133 Harvard Law Review (forthcoming 2020). My article documents around a dozen non-plaintiff-protective injunctions issued by federal courts in the first half of the twentieth century against various local, state, or federal laws. (See Lost History, at pp. 23-73; all citations are to the 9/25/2019 draft.) In one of these cases—Pierce v. Society of Sisters—the three-judge federal court granted, and the Supreme Court affirmed, a universal injunction that barred the enforcement of the Oregon compulsory public-schooling law in a suit brought by two schools suing for themselves alone. Two other decrees—an injunction issued by the Supreme Court in 1913 and an injunction issued by the D.C. Circuit in 1939—each disprove Bray's claim in MC that the first national injunction issued in 1963. After setting out and assessing these decisions and others, Lost History argues that injunctions that reach beyond the plaintiffs are consistent with the limits of traditional equity and the power conferred Article III to decide "cases … in equity": "There is only one 'judicial power,' and that power includes the power to issue injunctions that protect those who are not plaintiffs." (Lost History, p. 7.)

Rather than engage with my article, Bray's response mischaracterizes its arguments and evidence. To read Bray's response, one would imagine that I had said that the Supreme Court's 1940 decision in Perkins v. Lukens was a ringing endorsement of the universal injunction against federal agency action and that I had omitted the Court's language concerning its breadth. He states that "[t]he article says that the Supreme Court … was silent on the scope of the injunction." In fact, my article directly quotes (in above-the-line text) all of the following language from Perkins: the Court's description of the D.C. Circuit's decision as "sweeping"; its observation that "[i]n this vital industry, by action of the [D.C. Circuit], the Act has been suspended and inoperative for more than a year"; its framing of the question before it as whether "a Federal court, upon complaint of individual iron and steel manufacturers, may restrain the Secretary and officials who do the Government's purchasing from carrying out an administrative wage determination by the Secretary, not merely as applied to parties before the Court, but as to all other manufacturers in this entire nation-wide industry," and its statement that "In our judgment the action of the [D.C. Circuit] goes beyond any controversy that might have existed between the complaining companies and the Government officials." (Lost History, pp. 65-66.) I also note that "the Perkins Court was deeply skeptical of the D.C. Circuit's universal injunction" (Lost History, p. 66 (emphasis in original). "This is all in plain sight in the opinion of the Court," says Bray. It is also all in plain sight in my article.

As I then go on to explain, however, Perkins was clear that no injunction of any kind could issue in that case—not even a purely plaintiff-protective injunction—because the steel companies lacked standing to challenge the Secretary's wage determination: "[A] wage determination by the Secretary contemplates no controversy between parties and no fixing of private rights; the process of arriving at a wage determination contains no semblance of these elements which go to make up a litigable controversy as our law knows the concept." The limitations in the Walsh-Healey Act were not "a bestowal of litigable rights upon those desirous of selling to the Government"; rather, the act was "a self-imposed restraint for violation of which the Government—but not private litigants—can complain." (Lost History, p. 66 (quoting Perkins).) Did Perkins did say anything about what the proper remedy would be in suits brought by parties that hadstanding or in suits that did implicate private rights? No: Perkins instead took care to distinguish cases that involved "regulatory power over private business or employment" and cases in which official action "invade[d] private rights in a manner amounting to a tortious violation." Notably, one of the cases Perkins distinguished was Pierce v. Society of Sisters, which had earlier affirmed a universal injunction against a state law (Lost History, pp. 38-41).

To Bray, none of this discussion of standing bears significance, even though it is the basis of the actual holding of the case. Bray says "the article's separation of 'standing' from 'remedies' is artificial"; rather than put the concepts of "standing," "judicial power," "equity" and "remedies" into "separate boxes," he says, we must treat them as "reinforcing and interrelated concepts" that all speak to "the fundamental relationship of judicial power to the political branches." This leads him to read Perkins as "undermin[ing]" the contention that federal courts may issue injunctions that reach beyond the plaintiffs and to treat it as a case that somehow "confirms the previously recognized timeline for the development of the national injunction" (emphasis added).*

But Perkins cannot be pressed that far. Let us bring Perkins down to brass tacks. In 1939, the D.C. Circuit gave a universal injunction against federal agency action. (That alone is worth knowing, because it is 24 years before the 1963 date that Bray identified—which, I suppose, Bray would now have us know as the year of the "first non-reversed national injunction." (But see infra, regarding Lewis.)) In 1940, the Court reversed the D.C. Circuit's injunction on standing grounds, while both showing that it was "deeply skeptical" (Lost History, p. 66) of that injunction and also taking care to stress that its decision rested on the line between public rights and private rights. Beyond that, Perkinsmeans very little. If Perkins had held that the plaintiffs had standing and were entitled only to an injunction shielding them, then Perkins would be very helpful to Professor Bray's account. If Perkins had held that the plaintiffs had standing and were entitled to the "sweeping" injunction they received from the D.C. Circuit, Perkins would be very helpful to my account. But as matters unfolded, Perkins did neither of those things. Thus, just about all that can be drawn from Perkins is to point out what the case did not disturb, which is all that I do: Perkins "left intact the propriety of the injunction that reached beyond the plaintiffs as a remedy in a case brought by plaintiffs with standing" (Lost History, p. 5); Perkins "did not hold—or even say in dictum—that a universal injunction was categoricallyinappropriate or that it would be improper in suits brought by parties that had standing or in suits that did implicate private rights" (Lost History, p. 67 (emphasis in original)); and Perkins "did not implicitly limit or reject the injunction that reached beyond the plaintiffs" (Lost History, p. 70). I offer further support for that reading of the impact of Perkinsby showing that both just before and just after Perkins, the Court continued to affirm relief that went beyond the plaintiffs in cases brought by plaintiffs with standing and implicating private rights. (Lost History, p. 67-71.)

Bray offers an equally off-base treatment of the 1913 universal preliminary injunction against enforcement of a federal law issued by the Supreme Court in Lewis. (To set the record straight, the "entirety of the opinion" in Lewis does not just say "Granted." What it says is, "Per Curiam. On consideration of the motion for a restraining order of the appellees herein, It is now here ordered by the court that the motion be, and the same is hereby, granted.".) After repeating the same facts set forth in my article about the litigation in Lewis, Bray says that Lewis is "more plausibly" read "as a matter of estoppel: the government didn't need to make the representation that it wouldn't enforce the provision against newspapers during the pendency of the suit, but once it did, it should be held to it." Bray does not explain why a representation as to non-enforcement of a federal law made by the Taft Administration's Postmaster General (Hitchcock) could rightly estop the Wilson Administration's Postmaster General (Burleson) from enforcing that federal law. Nor does he address why, if parties in this era were understood to be able to seek relief only for themselves, the Lewis plaintiff had any business seeking and receiving assurances of non-enforcement against all newspapers. More importantly, dismissing Lewis as "a matter of estoppel" does not explain why the Court had the power to bind the federal government to its word. As Bray has elsewhere noted, "[a] court's equity powers are not determined by the concessions of the parties; many equitable doctrines protect the public and the court itself." (MC, p. 441.) A national injunction is a national injunction, whether the federal government concedes it should issue or not. Indeed, Harlem Valley v. Stafford likewise was a case in which the federal government did not contest the injunction's scope and indeed insisted that a national injunction issue—yet Bray (correctly) counted Harlem Valley as an example of a national injunction. (MC, p. 440-41.) The Lewis decision may be short. But it represents the Court's per curiam disposition of a single plaintiff's petition for an order that would enjoin a then-Cabinet-level federal officer from enforcing a federal law against either the plaintiff or against thousands of "other newspaper publishers" pending the Court's resolution of the case. My claim concerning Lewis therefore stands: Lewis issued a universal preliminary injunction against enforcement of a federal law.

Bray also cites my treatment of West Virginia Board of Education v. Barnette as an instance of a "pattern" of "inadequate contextualiz[ation]." He notes that the Barnette complaint sought an injunction "protecting the children of Jehovah's Witnesses" and that the Barnette decree protected "the children of the plaintiffs, or any other children having religious scruples against [saluting the American flag]." To those who have read my article, all of these points will have been familiar: as I say, the "three plaintiffs sued on behalf of themselves and a class of Jehovah's Witnesses," and that the decree "enjoined the state officers from requiring the flag salute to be taken by 'the children of the plaintiffs, or any other children having religious scruples against such action.'" (Lost History, p. 70 (emphasis in original).) Adding nothing substantive beyond what is already set out in the article, Bray claims it is unclear whether "the three-judge district court granting the injunction was thinking of 'children having religious scruples against the flag salute' as a broader set than 'children of Jehovah's Witnesses'"; says that the Jehovah's Witnesses "would have a claim analogous to the bill of peace"; and concludes "this looks like what we would now call a class action for all Jehovah's Witness school children in West Virginia." He then says that all this "is more plausible than saying the court rendered a 'universal injunction,'" which, incidentally, is not a label that I anywhere claim for Barnette. (See Lost History, pp. 5, 70-71, 86.)

The nub of Barnette is this: the suit was brought on behalf of a class of Jehovah's Witnesses, and yet the three-judge court in Barnette issued, and the Supreme Court affirmed, a decree that protected "any other children with religious scruples," Jehovah's Witnesses or not, from having to salute the flag. Bray cannot mean that this decree would have been read as allowing the state to coerce, say, a Mennonite child to salute the flag. The three-judge court's opinionconcluded by enjoining the flag salute requirement "in so far as it applies to children having conscientious scruples against giving such salute," 47 F. Supp. 251, 256 (S.D. W. Va. 1942), and contemporaneous press coverage duly reported that the decree extended beyond just the children of Jehovah's Witnesses. (See Court Throws Out Flag Salute, Charleston Gazette, Oct. 7, 1942 (reporting that the decree enjoined the board "from requiring the flag salute of children of the Jehovah's Witnesses sect and of all others having conscientious religious scruples"); Jehova[h]'s Witnesses Do Not Have To Salute Flag To Attend School, Raleigh Register, Oct. 7, 1942, at 5 ("The state board of education was enjoined today from requiring the flag salute of children of the Jehovah's Witnesses sect and all others having conscientious religious scruples….") (emphases added).) If Barnette was "like what we would now call a class action for all Jehovah's Witness school children," then the relief it gave went beyond that class. And calling this a "claim analogous to the bill of peace" simply concedes that the bill of peace and its analogues have had far greater power than Bray has described them as possessing.

This provides a natural segue to the next point: what exactly are the bill of peace and its analogues? Is the modern-day Rule 23 injunctive class action the sole such analogue, or is the universal injunction as well?** Bray and others contend that the modern-day Rule 23 injunctive class action, and not the universal or non-plaintiff-protective injunction, is the only way that a federal court may legitimately offer relief beyond the actual plaintiff. When class certification makes the class into "the plaintiff," say these critics, then a decree may shield that class, for the class will be bound; until that point is crossed, however, only the named plaintiff may be shielded. (See Lost History, p.56 n.321 (collecting quotes).)

But to vest the event of class certification with such significance is to miss the import of what Article III empowers federal courts to do. The Equity Rule 38 suits I describe demonstrate in practice what should already have been obvious in principle: class certification (or some equivalent device that creates preclusive effect on absentees) is not a necessary precondition for a federal court to afford broad-scale injunctive relief to non-plaintiffs. As I show, through the Equity Rule 38 device, plaintiffs in the period between 1913 and 1938 were able to gain preliminary and even final injunctive relief from three-judge federal courts against the enforcement of state laws on behalf of very large numbers of non-parties that they alleged to be similarly situated to them. (Lost History, pp. 42-52.) To pluck out an example, one such suit (Jackson) was brought by a single store-owner and resulted in a decree from a three-judge federal court that one newspaper estimated as shielding 44,000 retail stores in Indiana from its anti-chain-store tax. (Lost History, 50-52, 58 n.328.) That injunctive relief inured to the benefit of these 44,000 non-parties without imposing any burdens on them whatsoever; no class certification device existed at that period, and the absent non-parties would not have been bound by an adverse judgment had the state prevailed. See James W. Moore & Marcus Cohn, Federal Class Actions, 32 Ill. L. Rev. 307, 319-20 n.97 (1937-38) (listing, inter alia, Jackson as a "spurious" class suit). The Jackson decree (and the other Equity Rule 38 decrees I describe) therefore functioned like a modern-day universal injunction and unlike a modern-day injunctive class action, in that these decrees allowed those who did not participate in a lawsuit to receive the benefits of an action brought by someone else, while not bearing the burden of an adverse judgment.

Bray appears to doubt the utility of knowing such actual facts about how the Equity Rule 38 mechanism was used; after all, he seems to say, we have injunctive class actions now. But anyone interested in the constitutionality or the pedigree of today's universal injunctions should find it useful to know that a century ago federal courts regarded the principles of equity as allowing them to offer injunctive relief to non-plaintiffs, even when those non-plaintiffs included all sorts of parties with nothing in common except their shared vulnerability to enforcement of the challenged law. Can we doubt that injunctions shielding those "similarly situated" to the plaintiffs may be consequential today? Consider Trump v. IRAP, in which the Court retained a nationwide preliminary injunction against an iteration of the Trump travel ban "only with respect to parties similarly situated to Doe, Dr. Elshikh, and Hawaii," Trump v. Int'l Refugee Assistance Project, 137 S. Ct. 2080, 2088 (2017). The IRAP Court allowed this nationwide preliminary injunction to remain in effect as to all those similarly situated to a handful of plaintiffs, though no class had been certified and those similarly situated non-parties would therefore not be bound by any subsequent judgment. Yet the IRAP Court was not inventing the wheel when it issued this decision; it was simply doing what many Article III courts before it have done.

Bray casts my article as "conced[ing], as it must," that the Equity Rule 38 representative suit was a progenitor of the modern-day class, as "try[ing] to escape th[e] close connection" between the bill of peace, the Equity Rule 38 suit, and the modern-day class action, and as indulging in pointless "differentiations" between these devices. But what Bray casts as concession and evasion could better be called "accuracy" and "precision." If we collapse together the modern-day class action with the representative suit and the bill of peace, we miss the significance of how federal courts utilized their equitable power to shield non-plaintiffs in this important intermediate phase of American law—as well as the functional equivalence between those earlier decrees and the decrees we are seeing today. (See Lost History, p. 56-57 & nn. 321-22 (comparing Ramos and Pierce); p. 74 n.430.) In short, it will not do to run together the-bill-of-peace-and-the-representative-suit-and-the-modern-day-injunctive-class-action as if they were a single word—as Bray would do. These mechanisms are related, but they are also different. Moreover, their relation poses no problem for me, whereas their differences pose a significant problem for critics of the modern-day universal injunction.

This brings me to Hill v. Wallace, a case that forces me to return the charge of "inadequate contextualiz[ation]" to Bray. Multiple Chancellors described Hill as a "prototypical example[]" of a "stockholder suit" and as a case that "allow[ed] eight members of the Chicago Board of Trade to sue the Secretary of Agriculture on behalf of all 1610 members, seeking an injunction that would restrain the enforcement against the Board of an allegedly unconstitutional statute." (MC, at p. 431 & n.69 (emphasis added)). In his response, Bray largely repeats this characterization, calling Hill "a thoroughly unremarkable instance of an American bill of peace brought by some shareholders of a corporation on behalf of all of them." But on both occasions Bray omits that the eight shareholders of the CBoT were doing more than just trying to stop the enforcement of a law against the corporation in which they were shareholders. (Compare, e.g., Pollock v. Farmers' Loan and Trust.) They were also trying to stop the enforcement of a federal law against themselvesand all the other 1,600 shareholders. The Future Trading Act imposed record-keeping obligations on individual traders and it imposed punitive taxes and criminal penalties on them individually if they traded in futures on markets not certified as "contract markets." (Lost History, p. 26.) A decree that shielded just the CBoT from enforcement of the act would not have shielded all of these individual traders from these taxes and criminal penalties—which is why the eight plaintiffs sought and obtained an order from the Court that temporarily protected all 1,600 non-party members of the CBoT from the tax, criminal, and record-keeping provisions of the act. (Lost History, p. 28.)

Bray appears to think I misunderstood the fact that Hill was a representative suit. On the contrary, as I explain, Hill is noteworthy both because it was a representative suit and because of who that representative suit was against—the federal government: "Hill demonstrates that … the representative suit in equity was allowable against the federalgovernment with respect to enforcement of federal law."  (Lost History, p. 30 (emphasis in original).) The reason this is noteworthy is because Multiple Chancellor's discussion of the "American bill of peace" does not acknowledge that the bill of peace or the representative suit could be used in this fashion. In MC, Bray explains that at the end of the nineteenth century, American courts issued bills of peace against municipal ordinances or taxes or county taxes, and that some state courts did so against state taxes, but stresses that the bill of peace was not used against federal laws or state laws: "Again what was challenged were not federal or state laws but municipal ordinances." (MC, p. 427.) Multiple Chancellors does not thereafter discuss how the representative suit or the bill of peace was used in the early twentieth century as to either state or federal law, with one important exception: Bray casts Frothingham as tacitly rejecting such representative suits against the federal government. I quote Bray here in full, as I do in my article: "Equity allowed certain kinds of representative suits, and in nineteenth-century American law the prototypical examples were suits against municipal corporations and public corporations by one or more individual plaintiffs (taxpayers and stockholders, respectively). But the scale and relationship of the individual to the national government were 'very different.'" In a case like this, Justice Sutherland wrote, 'no basis is afforded for an appeal to the preventive powers of a court of equity.'" (MC, p. 431 (emphasis added); quoted at Lost History, p.30 n.81). As my account of Hilldemonstrates, however, by 1922 the Supreme Court had accepted that the representative suit could be used against "the national government" to stop the enforcement of a federal law against individuals. (Lost History, p. 30.) Hill might have been more "unremarkable" if Bray's article had not made that point so worthy of remark.

Finally, I learn from Bray that I have failed "an important test of any legal intellectual history: take the proposed conclusions, assume they're true, and ask how the relevant actors would therefore have behaved." If it were true, or thought to be true, that Article III or traditional principles of equity forbade the injunction that protects non-plaintiffs, then the Supreme Court would not have granted a universal preliminary injunction against a federal law in Lewis in 1913, or affirmed the universal injunction against the Oregon law in Pierce v. Society of Sisters in 1925, or affirmed the injunction that protected "any other children having religious scruples" in Barnette in 1943. The D.C. Circuit would not have issued a universal injunction against federal agency action in Lukens. And multiple other lower federal courts in decisions spanning decades would not have issued the many non-plaintiff-protective decrees against state or local laws discussed at length in my article. (Lost History, pp. 37-61, 67-71.) The reader may judge whether my account has "fail[ed]" Bray's test—or whether Bray's has.

* In a footnote to his response, Bray toys with the idea that the Perkins decree was a purely plaintiff-protective injunction, not a national injunction; if that is so, then perhaps Wirtz v. Baldor Electric would not be a national injunction either, but a purely plaintiff-protective one, for it involved the same statutory scheme for minimum-wage determinations and the same competitive-bidding considerations for government contracts as Perkins did.

** For a rich discussion of this question, see Brief of Amici Curiae Legal Historians in Support of Plaintiff and Appellee the City of Chicago, 2018 WL 6173238, *9-*12.