The Purcell Principle

As far-right candidates in Hungary, Brazil, and Serbia met their electoral demise in 2026, the world is seemingly holding its breath as the United States barrels towards its own trajectory-defining nationwide elections. The 2026 U.S. midterm elections see all 435 members of the House of Representatives and 33 of the nation’s 100 Senators on the ballot, meaning the composition and majority control of both chambers of U.S. Congress hangs in the balance of November 3rd’s results. Races across the country have been both contentious and contested, with Trump taking a heavy-handed approach in attempting to disrupt election processes and polls reporting swings in the likelihood of party-control over districts and states alike. Simply put, Trump and the Republicans are projected to lose their majority in the House – and perhaps even the Senate – and have initiated legal battles across the country with the potential to disenfranchise thousands if not millions of voters. The U.S. Supreme Court has responded in mystifying manners.

There exists a doctrine in twenty-first-century Supreme Court decision-making called ‘the Purcell principle.’ Coined in 2016 by UCLA Law Professor Richard Hasen and named after the case it emerged from, Purcell v. Gonzalez (2006), the term refers to the concept that generally guides federal courts to leave in place the status quo of election processes when election day is near. Purcell’s intention is to prevent confusion among voters, but the principle has become an encompassing excuse for the justices to modify election conduct. Its core flaws are thus twofold: the Purcell principle itself is derived from a non-argued and non-binding case, and there is no actual outline as to how close an election must be for the principle to apply. The Court therefore has no procedural reason to abide by Purcell or provide explanation for its decisions beyond a simple invocation of Purcell, leaving lower courts guessing and usually at odds with the Court as to when it should guide their decision-making. With Trump’s constant and calculated attacks on long-standing elections processes on the eve of the midterms, Purcell has returned as a dubious factor in many of the federal judiciary’s recent (and in some cases, surprising) decisions, many of which are still being handed down with days to go until the election.

Purcell’s Place in American Jurisprudence

 Purcell v. Gonzalez (2006) seems prima facie a landmark merit decision – that is, a case the Supreme Court scheduled oral arguments for, solicited amici curiae (“friends of the court”) briefs for, and handed down with clearly articulated and precedent-setting majority, concurring, and/or dissenting opinions based on the case’s merits. After all, one might assume its significance may be equated to a merit decision such as Dobbs v. Jackson (2022) or Citizens United v. FEC (2010) for its sharp increase in citations by federal courts since 2016 to base precedent-setting judicial review upon.

Reality could not be further from this conclusion. Purcell is instead a case that existed on the Court’s emergency docket – also termed the ‘interim docket’ or colloquially known as the ‘shadow docket.’ The core difference between the Court’s merit and emergency dockets is that unlike merit appeals, emergency appeals are typically brought to a singular justice and usually referred to the other justices (rather than voted upon for acceptance), not scheduled for argument, decided rapidly, not usually given any opinion, and do not set Supreme Court precedent. Emergency decisions do hold the same weight of law in that they settle disputes on a national level until further briefing may be held in federal court, but they are not intended to guide future decision-making in the manner merit decisions are. Purcell is thus an anomaly.

The controversy at the center of the 2006 decision surrounded a voter-approved 2004 proposition in Arizona prescribing voter IDs as necessary for a ballot to be counted. In early 2006, a midterm election year, a group of Arizona voters sued to block its implementation; after the federal district court kept in place the proposition, they appealed, to which the circuit court overturned the lower court’s decision and blocked Arizona from implementing the voter ID proposition. When on emergency appeal to the U.S. Supreme Court (the contestation was the circuit court’s order itself, not the merits of the proposition, as by this point there were only days until the November elections), the Court ruled against the circuit court to put back in place the proposition, basing its reasoning on the timeliness of the election and how a revocation of the (then two-year-old) proposition would confuse voters.

All of this occurred on the Court’s emergency docket, but in recent years all levels of federal courts have increasingly referred to that decision when considering whether to stay state election conduct disputes on their own emergency dockets. In essence, every application of Purcell has varied – whether it is in regard to the time or the substantive issue – making it a deeply flawed notion and something seemingly decided on a whim based on jurisdiction or, controversially, ideology. It is meant to constrain federal judges in deciding state electoral matters, but even state courts have taken a separate approach in applying Purcell (if at all). This is key to why it is a principle and not a clause. As UC Law SF Professor Rory Little has argued on principles, “principles are not groundless or invalid. Instead, their application simply needs more clear explanation and definition.” So, it should unsettle the reader who expresses concern for the state of democracy that such an inconsistent idea has made its way into the forefront of deciding whether or not consequential election modifications may take effect.

2026 Election Disputes and Purcell’s Place Now (If At All)

At the time of writing this, there are currently 371 active federal cases litigating against the Trump administration’s efforts made in the name of national security. At least 24 have directly concerned election law, with half having been filed since August 20th. The issue at hand is not that plaintiffs have been slow to act against executive orders, rather that the Trump administration accelerated its concerted crusade within these past few months before the general election. The other half of these efforts, however, have already impacted voters for their distance from November. This temporality has seemingly split how the Supreme Court has utilized Purcell for these election disputes.

Take Louisiana v. Callais (2026). Decided April 29, Callais all but gutted Section 2 of the Voting Rights Act, in essence finding that the requirement of a second majority-minority district constituted a violation of the Equal Protection Clause. This triggered an immediate effort among many Republican-controlled states to redraw their congressional districts and dilute the minority vote to flip a seat from Democrat to Republican. Based on previous applications of Purcell, such as in Republican National Committee v. Democratic National Committee (2020) or Democratic National Committee v. Wisconsin State Legislature (2020), this ruling should have been considered too close to the election to modify existing congressional districts. In fact, Louisiana even delayed its primary to gerrymander its maps in line with Callais’s logic. If changing the date of an election is not indicative of a court ruling coming too closely to said election, I am not sure what more could be.

Mail-in voting has also come under attack, as Trump had attempted to modify how mail-in ballots are postmarked and collected with a March executive order – this was finally processed with an August 21 administrative rule from the United States Postal Service. On appeal for allowing the rule to take effect after a lower court blocked the rule, the Supreme Court ruled against the Trump administration to block the rule, citing the belief that the government is unlikely to succeed on the merits. But the more curious aspect underpinning this decision is the implicit diversion from Purcell. Justice Kavanaugh offered the only detailed (albeit three sentence) opinion, framing the issue as related to the Administrative Procedure Act. Where Purcell is inconsistent in this case rests with the lower court preventing the August 21 rule from taking effect. By Purcell’s logic (or, at least what we are able to reasonably infer), this federal court injunction is too close to the November general election and should have been vacated by the Supreme Court. But that’s exactly what the Supreme Court did not do.

On the other hand, there has indeed been some seeming congruence on Purcell. A “procedurally bananas” dispute, as Georgetown Law Professor Stephen Vladeck put it, has emerged in Missouri, in which the state has appealed to the U.S. Supreme Court three times as the justices ruled twice against Missouri’s attempt to implement a gerrymandered 2025 map (Vladeck’s blog gets into the specifics about the procedural ineptitudes evident here). While not explicitly stated, all of the justices (for lack of any public dissent) seem united in applying Purcell as voters in Missouri have already begun voting using the existing 2022 map.

Another late dispute in line with Purcell emerged with the Supreme Court’s October 23 ruling allowing New Hampshire to proceed with disallowing student IDs as a valid form of voter ID. The law amending New Hampshire’s valid voter ID methods passed in April, to which the district court ruled that, while the law may proceed, with supporting evidence student IDs may be considered. The Supreme Court, citing Purcell, struck down the district court ruling again on the basis of potential voter confusion too close to the election as a result of federal court action. But, compared to the mail-in ballot case, the timeline makes little sense, as its question has been germane far longer, and in both instances federal courts blocked a rule from taking effect.

This is all to say that there is a mystifying legal principle circulating among the federal judiciary’s election disputes, and we have seen a sharp increase in how subjective its application is in recent months, weeks, days. Perhaps in the fallout of the midterms we will see the Supreme Court finally decide a case on the merits with regard to Purcell and finally obtain a clear, multi-pronged test for deciding whether an action occurs too closely to an election and its potential for confusing voters. Until then, we Court observers will remain the ones confused.