Beyond the Purcell Principle

In Allen v. Milligan III, decided earlier this week, the Supreme Court stayed a lower federal court injunction that prohibited Alabama from using an electoral map (the 2023 map). The stay is technically interim—until the filling of an appeal or a petition for certiorari, depending on which of the consolidated cases is in view. But given the timetable for primaries for the 2026 election, the stay decides the question: the Alabama legislature gets to use its 2023 map, and the three-judge court that found it to be racially discriminatory does not get to block it.

I will leave to election law scholars the tangle of whether Callais requires this rapid reversal, and whether the three-judge court took sufficient notice of the new direction in Callais. But I want to call attention to the application of the Purcell doctrine in the per curiam opinion. The application comes in a single paragraph:

The State has also made a strong showing of irreparable harm and that the equities and public interest favor it. We have repeatedly cautioned that lower federal courts should not “alter the election rules on the eve of an election.” Republican National Committee v. Democratic National Committee, 589 U. S. 423, 424 (2020) (per curiam). Here, the District Court interposed itself into Alabama’s ongoing efforts to conduct its imminent 2026 congressional elections under maps that its elected representatives selected. Its view that conducting the elections under court-imposed maps would be more convenient for the State was not a valid justification for that intervention. While federal courts should not impose changes close to an election, ibid., States are free to decide for themselves whether last-minute changes to an election are in their best interests.

Three brief observations:

First, the Court is treating the Purcell doctrine not as a separate doctrine outside of the equitable analysis, but rather as part of the ordinary equitable analysis for an injunction. That is, the Court is locating Purcell in “irreparable harm and . . . the equities and public interest,” which are conventionally the last three factors of the preliminary injunction and stay tests. This “normalizing” or “naturalizing” or “assimilation” of Purcell is nowhere more obvious than in the fact that the per curiam does not cite Purcell; all the citations are in the dissent. This observation aligns with Derek Muller’s post about a month ago on Abbott v. LULAC, where he concluded: “Abbott v. LULAC shows that Purcell is a refinement of equitable principles in the context of elections.”

Second, the Court is expressly treating Purcell as a one-sided doctrine not a two-sided doctrine. That is, it is a restriction on late-breaking actions by federal courts, but without taking into account late-breaking actions by state legislatures. Now previous decisions from the Supreme Court already had that pattern of differentiation, but Allen III puts a sharp point on it in the last quoted sentence above.

The effect of the Court’s formulation in future cases could be to shut down opportunistic timing of suits by litigants who challenge state election changes, but to green-light the opportunistic timing of election-law changes by states. Yet perhaps that line is just dicta. After all, here the state legislature may have moved as quickly as it could have, given the injunction. Indeed, in this case, deciding whose actions are “late-breaking” is actually complicated, because of Callais and the fallback quality to the latest Alabama legislative action. And it also depends somewhat on Alabama law: the dissent notes various ways Alabama’s hands might be unclean (and unclean hands is relevant), and an additional way might be the possibility that the legislature was violating Alabama law (cf. McClintock’s Handbook of the Principles of Equity: “Almost always the violation of any statute will make a party’s hands unclean.”). Yet all of this is pretty complicated.

Whether there was bad faith here by the legislature, or unclean hands, is not something on which I have formed an opinion—I would need to dig more into the record. But note that all this factual complication simultaneously (1) makes the Court’s action more permissible while also (2) undermining the force going forward of the line that “States are free to decide for themselves whether last-minute changes to an election are in their best interests.” If the state legislature was not opportunistically delaying in order to avoid judicial review, for example, then Allen III can’t control cases where a state legislature is opportunistically delaying in order to avoid judicial review.

Third, these two developments are in tension. I think it makes a lot of sense to assimilate the Purcell principle to equity, treating it as a specific instantiation for election law of general equitable concerns. Within the preliminary injunction and stay tests, this probably fits best with the public interest as a concern for the orderliness of state elections. Late-breaking changes by federal courts, all else being equal, disserve the public interest. And where there is bad faith by one of the parties, that is certainly relevant for irreparable injury and the balance of the equities.

But the question of bad faith, ever-relevant for equitable relief, is a factual one, and it cannot be determined by appeal to presumptions. Any presumption of good faith for the state, in other words, must be defeasible, and whether the presumption is rebutted in any particular case is itself a factual question. (An astute reader of this post noted to me the parallel with Younger abstention: the federal courts will not usually enjoin criminal proceedings in state court, but that presumption of abstention is defeasible, and bad faith is one of the grounds for a federal injunction.)

Moreover, Purcell fits with equity’s concern, manifested in a number of doctrines and maxims, about the court not being manipulated or made an instrument of injustice. That is, Purcell reflects a legitimate concern that litigants challenging election laws can sandbag, hanging back and disrupting elections with last-minute challenges. Some cases in which the Purcell principle is now invoked are far distant from that concern, but the concern itself is legitimate.

But if the Purcell principle is going to be located within the ordinary equitable analysis, it will not fit how equity works to have it be one-sided and selective in its concern for opportunism. Equity is pervasively concerned with opportunism, as Henry Smith has shown in his fantastic article Equity as Meta-Law, which is the place to begin for understanding what equity means by opportunism. The myriad equitable doctrines and maxims that respond to equity are concerned with opportunism in the round.

Take laches, for example: laches will block or limit the equitable relief given to a plaintiff who unreasonably delays in suing, where that unreasonable delay causes harm to the other side. There is no laches doctrine especially restricting property owners, or especially restricting non-property owners. Laches reflects the sensitivity of a court of equity to how its relief is vulnerable to abuse (a theme I develop in The System of Equitable Remedies). But no one gets a free pass.

“No one” needs an asterisk: laches doesn’t usually apply to the government. But equity’s concern with opportunistic action still does apply to the government. Consider the unclean hands doctrine. It is true that it applies differently to the executive branch than it does to private litigants. But again there is no free pass: as Will Baude and I have shown, unclean hands doctrine does apply to the executive branch when it seeks equitable relief. And again consider the example of Younger: state judicial proceedings are treated differently in federal court, but bad faith prosecution is a way for a state to lose that special treatment.

In short, if the Purcell principle is going to be assimilated to equitable analysis, it will not cohere—to put it more sharply, the body of equitable doctrine will reject the transplant—if it is a doctrine that constrains opportunism by challengers and simultaneously green-lights opportunism by state legislatures. The concern for opportunism might play out differently for state legislatures than for challengers, but equity cannot give anyone a “Get Out of Jail Free” card.

Here’s how I put the point last year in Preliminary Injunction Realism, a festschrift piece for Doug Laycock:

The Purcell Principle could be seen as a timing doctrine specific to election law, or as an aspect of the public interest. See Richard L. Hasen, The Stagnation, Retrogression, and Potential Pro-Voter Transformation of U.S. Election Law, 134 YALE L.J. 1673 (2025); see also Brooks M. Chupp, Equitable Balancing in the Purcell Framework, 16 DREXEL L. REV. 507, 521–22 (2024) (listing other ways Purcell could intersect with equitable analysis). There is nothing wrong with equitable considerations outside the four factors, and it is appropriate for a judge’s preliminary injunction analysis to be sensitive to the policies of the substantive law. After all, “equity follows the law,” including election law. But timing-related doctrines in equity, preeminently laches, are alert to abuse by both parties; they appraise each party’s good faith. For the Purcell Principle to cohere with equity, therefore, it needs to be similarly alert to opportunism not only by the plaintiffs who challenge new legal norms, but also by the legislatures and executive officers who enact and enforce them.

Now there is a way to resolve this, even if Purcell, or the doctrine formerly known as Purcell, still remains narrowly about the timing of actions by federal courts. First, the Court will need to say what the connection is between the public interest and the timing of actions by federal courts (e.g., is it, as I suggested above, the public interest in the orderly conduct of elections?). Second, the Court will need to be explicit that the scope of the public interest considerations are not limited to the Purcell principle. This is a statement of the obvious, but there is no reason there can’t be a Purcell principle about equity avoiding opportunistic timing by challengers and a Schmurcell principle about equity avoiding opportunistic timing by those they challenge.

If there were such a Schmurcell principle, there is no certainty it would apply in this case—again, perhaps the legislature moved as quickly as it could and is not acting opportunistically and trying, for example, to avoid judicial review of its actions. And there is no reason they have to be formulated the same way and apply with the same severity.

But however these principles are formulated, and however they might apply in this case, the point is that courts of equity need to be attentive to how their powers can be abused by litigants, and the threats never come just from one direction.