In spite of (or perhaps because of) the fact that the Supreme Court’s per curiam opinion two weeks ago in the Wisconsin election case, Republican National Committee (RNC) v. Democratic National Committee (DNC), ran fewer than four pages, the 5-4 ruling has engendered a great deal of commentary and criticism about what the justices did. Namely, on the literal eve of Wisconsin’s April 7 contest, the Court effectively removed and rewrote some parts of the injunctive order that a district court judge had imposed on state election officials just a handful of days earlier. A column by Fellow Verdict columnist Michael Dorf last week highlights a main point of attack (and one shared by other critics): that the Court distorted and misapplied the considerations that govern the kinds of remedies that are appropriate for federal courts to provide, and along the way may also have misinterpreted (if not made up out of whole cloth) Wisconsin state law requirements. All this in order to adhere to what Professor Dorf calls “petty sticklerism”—a “commitment to wooden and heartless interpretation of rules.”
While we appreciate some of the points critics raise (and certainly agree with them that the Supreme Court’s reasoning was not well explained), we have a different take on the matter. Rather than concentrate on whether the district court’s injunctive remedy, and the Supreme Court’s trimming back on it, complied with the principles that ought to inform permissible federal court relief, we focus our attention on whether there was any plausible violation of federal law for the federal courts to redress in the first place. All federal judicial remedies are inappropriate if there is no legal violation to prevent or undo. Our concern is that critics of the RNC case, and (for that matter) the Justices themselves, have put the remedial cart before the substantive-law horse. And after more careful examination of the merits of the plaintiffs’ federal claims in the federal district court in Wisconsin, the U.S. Supreme Court’s decision to reverse the lower court looks correct as to result, even if poorly executed as a matter of judicial craft.
What was the federal claim on which the district court granted the relief in question? (We note here that in this setting only claims based on federal law provide a basis for suing in federal court; misconstruction or misapplication of state election law, unless that misapplication itself constitutes a federal law violation, is generally none of the business of federal courts, even if federal judges technically may have “supplemental” jurisdiction over some state law claims; in the Wisconsin case, plaintiffs pressed only federal causes of action.) In brief, the plaintiffs asserted their federal constitutional rights to vote were being violated because they didn’t have time to obtain and return their absentee ballots early enough for those ballots to be tallied and counted under Wisconsin law (which required that all ballots, whether cast in person or returned by mail, be received by the end of the day on Election Day), and because the plaintiffs couldn’t easily vote in person on Election Day itself on account of the health risk presented by the coronavirus pandemic.
As the district court put it: “[E]ven the most diligent voter may be unable to return his or her ballot in time to be counted. Wisconsin clerks are facing a record number of absentee ballot requests [because of the COVID-19 pandemic], and despite diligent efforts, as of [five days before the election], they [were] still working on sending out a backlog of [several thousand] absentee ballot applications. [As a result] ‘[t]here is no practical way that a person [timely] submitting a request for an absentee ballot . . . [would] have the time to receive, vote and return their ballot by Election Day.’” And as to the possibility of simply voting in person on Election Day, it goes without saying that being out in public presents some individual and public health risk even if election officials work hard—as they did in Wisconsin—to respect and enforce social-distance guidelines.
Applying a very open-ended balancing test that the Supreme Court has used in some but not all cases alleging violations of the federal constitutional right to vote, the district court found the “burden placed on absentee voters” in their efforts to have their preferences taken into account to be “severe,” and the state’s interest in enforcing its requirement that all ballots be received by the end of the day on Election Day insufficient to justify this extreme burden voters faced: “The state’s general interest in the absentee receipt deadline is not so compelling as to overcome the burden faced by voters who, through no fault of their own, will be disenfranchised by the enforcement of the law.” Accordingly, the district court found the plaintiffs would likely succeed on their claim of violation of the fundamental constitutional right to vote, and the judge ordered (among other things) that Wisconsin allow people to submit (and have counted) absentee ballots after Election Day. To prevent people from casting ballots after they already knew the provisional results on Election Day itself, the district court enjoined also state officials from releasing election-tally results for several extra days after Election Day night, during which period absentee ballots could be cast. Yet the plaintiffs did not get everything they sought. In particular, even though the district court also agreed that the plaintiffs had demonstrated a likelihood of success on their related claim that (in light of the impact of COVID-19) holding the in-person election as scheduled would also violate the rights of voters, the judge, citing the state’s interests in an orderly electoral process, the risk of voter confusion, and principles of federalism, declined to issue a remedy requiring the state to postpone the election entirely.
The Supreme Court overturned part of the district court order, and in effect freed the state from having to count absentee ballots that were submitted (either by postmark or by in-person dropoff) after Election Day. Before the justices, much of the fight seemed to be over whether the district court had violated the so-called Purcell principle (from a case of the same name), that “lower federal courts should ordinarily not alter the election rules on the eve of an election.” Professor Dorf (along with the dissenting justices at the Court) argues that the principle doesn’t (and can’t) apply just to lower federal courts, and that the Supreme Court majority was itself violating the principle by overturning (parts of) the district court’s order precisely on the day before Election Day.
Our view is that Purcell may overstate the appropriate limitations on federal courts. Where there is a strong and imminent likelihood of a federal violation (constitutional or statutory), a federal court should frequently be empowered to prevent it. (Indeed, fashioning after-the-fact remedies is often more problematic than enjoining violations beforehand.) And, arguably, the Purcell considerations should apply, albeit in a narrower manner, to the Supreme Court in addition to lower federal courts. Nonetheless, if a federal district judge wrongly enjoins an imminent state election, the Supreme Court cannot be powerless to undo that mistake merely because the Supreme Court’s correction (necessarily) occurs even closer in time to the contest. Surely it can’t be the rule that the Court cannot fix a Purcell error because doing so violates Purcell; if so, then lower courts could ignore Purcell with impunity.
But we want to set aside Purcell, and its emphasis on remedial latitude, to examine a much more fundamental question: was there, to use the language we do above, “a strong and imminent likelihood of a federal violation (constitutional or statutory)” for the district judge in Wisconsin to provide any remedy concerning absentee voters? If the answer to that question is “no,” then the Supreme Court’s ruling was correct. This is true regardless of whether (as critics of the Court’s ruling point out) not all the defendants appealed the district court’s ruling and not all aspects of the district court’s order were challenged on appeal. Parties need not challenge on appeal everything a lower court has done but are free to focus on what they consider the most objectionable part(s) of a ruling. Here, then, the RNC’s choice to focus its case in the Supreme Court on the district court’s extension of time to cast absentee ballots did not prevent the RNC from arguing, as its more general basis for reversal by the Court, that the district court lacked power to provide any judicial remedy at all because there was, on the merits, no imminent federal rights violation.
It is remarkable how little attention the district court devoted to the merits of the claim concerning the asserted difficulties of the absentee voters. The court spent only a little more (or less, depending on how you count) than one page of its opinion explaining why the plaintiffs were likely to succeed in their constitutional challenge in this respect. The district court’s legal discussion consisted basically of its citation to two Supreme Court cases laying out a general test of burden-on-voter vs. justification-by-state, and then one district court case (which is obviously not binding) from another state affording relief in arguably similar circumstances. However, the two cases from the Supreme Court involved not election administration but ballot access by candidates, and so are not particularly relevant to evaluating the challenge brought against the Wisconsin election rules. Indeed, more generally, the cases in which the Court has over the decades rigorously demanded substantial state justifications and careful tailoring in the voting arena involve primarily (perhaps almost exclusively) rules that: (1) concern requirements for voter or candidate eligibility; (2) weigh ballots that are cast unequally; (3) reflect intentional or practical discrimination on account of race; or (4) implicate congressional statutes prescribing particular requirements for elections of federal officials.
The Wisconsin dispute involved none of these issues but instead centered on the mechanics of how a state administers its elections on the ground. (Notably, the Wisconsin election this month was devoted almost entirely to issues of state government, including, importantly, who should serve on the state’s Supreme Court and its appellate courts. To be sure, there was a presidential primary, but under Article II of the Constitution states don’t even have to hold elections to pick presidential electors, just as they don’t have to hold elections for state judges. And while there was in parts of the state one congressional primary – and states do have an obligation to hold elections for members of Congress – there was no claim that it ran afoul of any federal election laws regulating congressional contests.)
The U.S. Supreme Court case in recent times that focuses most squarely on the logistics and administration of state elections is Crawford v. Marion County Election Board. That case involved an unsuccessful challenge in 2008 to Indiana’s requirement that voters demonstrate their identity to vote. In rejecting the challengers’ claims, six justices (including Justice Stevens) agreed that, at the very least, “even-handed restrictions” promoting the “integrity and reliability of the electoral process itself” satisfy constitutional standards.
Wisconsin’s own requirement that ballots be received (and its corollary implicit requirement that they be cast) by Election Day is certainly “even-handed.” And no one could easily deny that the goals of securing election results as soon as possible and avoiding unnecessary risk that some people might vote after knowing provisional election results go to the “integrity and reliability of the electoral process itself.”
The district court spent no time analyzing Marion County, or indeed even trying to figure out which category of voting cases from within the Supreme Court’s voting rights jurisprudence was the most appropriate from which to draw guidance. Instead the court simply laid out an unguided balancing test from cases involving voter qualifications and ballot access and determined that Wisconsin could ignore its state-law requirements without losing too much. To repeat the court’s language: “The state’s general interest in the absentee receipt deadline is not so compelling as to overcome the burden faced by voters who, through no fault of their own, will be disenfranchised by the enforcement of the law.”
The district court also failed to analyze carefully what its reasoning would mean for state election law generally. For example, what if a state simply chose not to provide for absentee ballots? As far as we know, the Supreme Court has never suggested that voters must have the option of casting their ballots by mail. If the Constitution imposed such a requirement then wouldn’t the procedures of “caucus” states, where people must attend and participate in person to have their input counted, be problematic? And putting aside the caucus device, if a state decided not to permit absentee (or any other mail-in) ballots (except perhaps as required by federal statute) and instead required in-person voting for all, wouldn’t enforcement of its law also “disenfranchise” many people who for whatever reason can’t vote in person?
Or consider a “normal” (non-COVID-19) election year in Wisconsin. There are always some people who, “through no fault of their own” (to use the district court’s words) submit a timely request for an absentee ballot but who (because of poor mail delivery or other reasons) don’t get their ballots in time to return them by mail early enough to have them counted but who can’t – say, for family or work reasons –vote in person on Election Day. Have those folks also been unconstitutionally “disenfranchised” by Wisconsin’s rule that ballots be received by Election Day? If the answer is “no” (as it has to be), then why is there an unconstitutional disenfranchisement this year?
The distinction surely can’t be that in a normal year people could have planned better and sought absentee ballots earlier. If anything, the need this year for advance planning should have been even more obvious: by early (and certainly mid-) March, the health benefits of social distancing were plain. The President announced national social-distancing guidelines/recommendations against unnecessary group gatherings (of more than 10 persons) on March 16, and the Wisconsin Governor’s stay-at-home directive (which exempted voting from its restrictions) was itself issued on March 24th. And other states were well ahead of that. But even if a voter requested an absentee ballot on March 25th, there very likely would have been time to receive and return the ballot by mail and have it counted. The Wisconsin election website also advises voters more generally that given possible delays in mail delivery they should request and complete absentee ballots well in advance of the state deadline.
If the distinction this year is simply that more people were likely to be involved, it is well to ask why a few hundred folks being unable to vote is permissible but a few thousand is not—and, more generally, to wonder: where is the line? One individual’s constitutional rights don’t become more or less strong depending on the number of other people asserting a violation. The unusual circumstances this year did (as we suggest below) argue in favor of a legislative rescheduling of the election itself, but whether this year has implicated constitutional rights in a way that other years have not is a very different matter.
Indeed, the district court decision itself would seem to rest on an expectation that, even at the end of its own contemplated process this year, some significant number of voters would still be unfairly prevented from casting a ballot. Recall that the district court’s overarching concern was that some Wisconsin voters who had requested absentee ballots would not receive the ballots in time to return them, in person or by putting them back in the mail, by Election Day. But play this out. On Election Day itself, some voters (not having yet even received their absentee ballots) would have to decide whether to show up at the polls (possibly risking their health) or to wait for the requested ballots to arrive in the ensuing days—and hope that they would come quickly enough so that they could be sent back for counting before the expiration of the extended timeline the district court had prescribed. If the mail were slow (plaintiffs argued that in-state delivery could take several days), no vote could be cast and counted for these unlucky folks. Perhaps the thought is that the district court’s order allowed for more rather than fewer absentee ballots to be cast and tallied than would otherwise be the case, and that this is a better outcome. But again, if the state procedures violated the Constitution and required a remedy, how can some individuals’ rights be ignored because the mail, for them, was particularly slow, even if others were lucky enough to have better postal service?
More generally, consider the essential nature of the plaintiffs’ claims. Wisconsin law’s receipt deadline may be a but-for cause of many voters’ problems, but certainly the proximate cause is the COVID-19 pandemic. Should Wisconsin officials themselves have postponed the election? Of course. It is disgraceful that they didn’t. And plaintiffs are understandably angered by the state’s inaction. But an awful failure to act does not mean Wisconsin’s extant regime is unlawful (at least not under the federal Constitution). In essence, what the plaintiffs objected to was not Wisconsin’s decision to have a ballot-receipt deadline, but Wisconsin’s failure to accommodate voters who had trouble with that deadline in 2020. But there are all kinds of real-world constraints on people’s ability to cast their ballots every year that would benefit from accommodation. Moving Election Day from Tuesday to Saturday would facilitate voter turnout. So would making Election Day a holiday. Or providing day care for would-be voters. Or transportation to the polls. Or moving more generally to a vote-by-mail system. But none of these accommodations (some of which we think might be very good ideas) is, under current doctrine, constitutionally required. What that means is that failing to adopt them isn’t actionable. So why was Wisconsin’s failure to accommodate this month of a different kind, constitutionally speaking?
Election Day deadlines for absentee voting are common. Wisconsin’s rules applied evenly to everyone. Large numbers of Wisconsinites were able to request, receive, and timely return absentee ballots (and Wisconsin election officials had agreed to count ballots up to six days after the election provided the ballots were postmarked by Election Day). That others may have been unable to vote does not mean the state was in violation of the Constitution such that a federal remedy was warranted. In a variety of contexts, the Supreme Court has cautioned against displacing state power to administer orderly elections. In so doing the Court has emphasized the relationship between adherence to settled rules and democracy itself. Representative of the Court’s treatment of election mechanics is the formulation the Court used in 1974 in Storer v. Brown: “[A]s a practical matter,” the Court there said, “there must be a substantial regulation of elections if they are to be fair and honest and if some sort of order, rather than chaos, is to accompany the democratic processes.”
We are not suggesting that the district court (and critics of the Supreme Court’s ruling) might not have comebacks to some of the challenging questions we flag. We are doubtful there are compelling responses, but our primary point today is that we shouldn’t be talking about remedies until we have a better sense of what, precisely, violations of federal law need to be remedied. Absent substantial (and not just minimally non-trivial) federal questions, federal courts ought not to be in this game at all. And there has been thus far very little discussion (by the justices or the commentators) of the merits of the plaintiffs’ federal constitutional assertions in the Wisconsin matter.
That brings us to the observation that federal law is not the only game in town. Perhaps there might have been a remedy of accommodation mandated under Wisconsin state law. But we’ll never know when challengers jump to federal court even when the better recourse is under state law. (The quick resort to federal court is also an answer to the argument that the Supreme Court misinterpreted Wisconsin law’s apparent, categorical preference that no votes be cast after Election Day; if that be a misconstruction of the values embodied in Wisconsin law, it is because plaintiffs forced the Supreme Court to make some, perhaps underinformed, judgments about state law by pressing their claims in federal rather than state court.)
To be sure, state courts are not always fair brokers. But where there is no firm and independent federal constitutional or statutory claim, state court manipulation of its own law would have to be severe to implicate the federal Constitution; only if state law is being distorted by state actors in a way that is patently designed to discriminate invidiously should state law (mis)interpretations give rise to federal constitutional violations. Otherwise, states are free to misread their own law, and are subject to state political remedies. (It bears noting that Democrats won big in Wisconsin’s election two weeks ago). This is the reason both of us objected strenuously to the Supreme Court’s intervention in Bush v. Gore: the Court’s decision in that case represented a contrived conversion of state-law issues into newfangled federal constitutional claims. Call us old-fashioned, but just as we deeply lament Bush v. Gore, we are glad the Court didn’t repeat the same mistake this month; having federal courts stick to substantial federal claims doesn’t strike us as “petty sticklerism.” And that’s true even if it lets stand the shameful decision by the Wisconsin legislature requiring people to needlessly risk their lives to effect change in Wisconsin on Election Day.