Last week a divided three-judge panel of the U.S. Court of Appeals for the Fifth Circuit decided a case involving access to mail-in ballots. The case may not have earth-shattering effects, but neither are its consequences trivial this year. Above all that, it was clearly wrongly decided, and illustrates how some judges have bad interpretive instincts when it comes to navigating the tricky but ultra-important voting rights realm.
The case, Texas Democratic Party v. Abbott (one of many, unrelated cases bearing that name), involved a challenge to a relatively simple Texas statute that, as the Fifth Circuit put it: “permits early voting by mail [but only] for voters who: (1) anticipate being absent from their county of residence; (2) are sick or disabled; (3) are 65 years of age or older; or (4) are confined to jail.” The third category of the statute thus prefers people who are 65 or older, giving them an entitlement to early voting by mail that younger persons do not enjoy unless they satisfy additional criteria.
Plaintiffs challenged this law as violative of the Twenty-Sixth Amendment of the Constitution, which provides—also quite straightforwardly—that “[t]he right of citizens of the United States, who are eighteen years of age or older, to vote shall not be denied or abridged by the United States or by any State on account of age.”
On the face of things, the plaintiffs’ challenge would seem strong. Texas is definitely treating people under 65 differently with respect to voting “on account of [their] age.” (Indeed, no one even tried to deny that Texas was differentiating between would-be voters due to, or on the basis of, their age.)
Nonetheless, two judges of the Fifth Circuit decided the plaintiffs’ Twenty-Sixth Amendment claim fails. To its credit, the Fifth Circuit did not embrace the position, advanced by the defendants, that the claim loses because the Twenty-Sixth Amendment does no more than lower the voting age to 18. This reading would be impossible to square with the Amendment’s clear text, which does more than alter the minimum voting age; it prohibits discrimination “on account of age” against anyone 18 or over in the right to vote. Why, then, did two judges rule against the claim? Because, they said, the statute did not “deny or abridge” the right of anyone under 65 to vote. The court interpreted “abridgment,” by reference to some dictionaries and a few cases not on point, as the taking away or reduction of meaningful voting liberties that someone enjoyed before the enactment in question was adopted:
Rejecting the plaintiffs’ arguments, we hold that an election law abridges a person’s right to vote for the purposes of the Twenty-Sixth Amendment only if it makes voting more difficult for that person than it was before the law was enacted or enforced. . . . [A] law that makes it easier for others to vote does not abridge any person’s right to vote for the purposes of the Twenty-Sixth Amendment.
Here, because the Texas statutory backdrop was that no early voting by mail was originally allowed—exceptions to that starting point were made beginning in 1917, culminating in the entitlement at issue today of persons over 65 in 1975, after the Twenty-Sixth Amendment had been ratified—persons under 65 had never enjoyed early voting privileges, so they lost nothing by the conferral of such privileges to older voters but not to them. Again, the idea is that giving something to older folks takes nothing away from younger folks.
This reasoning is flawed for many reasons. First, the Fifth Circuit’s approach is open to seemingly arbitrary outcomes. The panel’s reasoning suggests the result would have been different—even though the effect and words of the statute would have been the same—had early voting been the rule rather than the exception. In other words, if everybody had enjoyed early voting prior to 1975, and in that year Texas revised the statute to limit it to only persons over 65, now all of a sudden the rights of younger folks have been abridged?
A related problem—one that the panel mentioned but did not address other than to say courts can manage such difficulties—is that of manipulation. Suppose Texas wants to make it harder for young people to vote but has in place a law that allows everyone to vote early. Exempting only young people would, as just suggested, constitute an abridgment in the eyes of the panel. So what should Texas do? Repeal early voting for all, wait a while, and then reinstate it only for older persons—voila! How is a court to know when such a sequence reflects a good-faith effort at legislative experimentation, or a cynical effort to circumvent a manipulable test? This is an especially dicey task in an era when the message the current Supreme Court generally sends is that inquiry into legislative motivations is to be avoided, and that statutes, rules, and executive orders should be analyzed on their face and by their operation.
But beyond (and much more important than) these practical concerns, the Fifth Circuit’s reasoning is completely wide of the mark because it ignores the equality dimension—which is the essence—of the Twenty-Sixth Amendment. The words of and history behind the Twenty-Sixth Amendment make clear that its proponents thought not only that young people were responsible enough to vote, but also that they were—as a class—equally valuable and entitled as older folks are to vote. Putting aside what the “right to vote” the Court has discussed in the context of the Fourteenth Amendment may mean, the voting rights covered by the Twenty-Sixth Amendment (and by the earlier specific voting rights amendments—the Fifteenth, Nineteenth, and Twenty-Fourth—which served as intellectual and textual templates for the Twenty-Sixth) involve not an absolute right to vote, but a right to be treated equally with respect to the vote.
That is why Representative Richard Poff, 20-year Virginia Republican member of the House and one of the leaders behind the Amendment, described the Amendment as conferring “a plenary right on citizens 18 years of age or older to participate in the political process, free from discrimination on account of age.” (emphasis added).
And that is why the language of the Amendment is written the way it is. As noted earlier, it does not say merely that every state shall reduce its voting age to 18. Rather, it provides that the right of persons 18 or older to vote cannot be abridged on account of age: textually, then, age cannot be used as a criterion for regulating the core political right of voting. In this sense, the Amendment self-consciously and precisely tracks the language of the Fifteenth and Nineteenth Amendments, with the same intended consequences. For the purposes of all these specific voting rights amendments, the meaning of “abridge” does not and cannot be limited to “take away or reduce what was enjoyed before,” but also has to include “deprive” (another dictionary definition of abridge) or “withhold” on unequal terms. In other words, although the Fifth Circuit may have been correct in suggesting that the word “abridge” implies a baseline, the baseline need not be a moment in the past; instead it can—indeed must—be what other people are currently receiving on account of their race or sex (or age).
As longtime House member Claude Pepper, Democrat from Florida, elaborated in this anti-discriminatory vein: “What we propose to do . . . is exactly what we did in . . . the 15th amendment and. . . the 19th amendment. Therefore, it seems to me that this proposed amendment is perfectly in consonance with those precedents.” And to the same effect, hear the powerful words of Poff:
What does the proposed constitutional amendment accomplish? . . . [I]t guarantees that citizens who are 18 years of age or older shall not be discriminated against on account of age. Just as the 15th amendment prohibits racial discrimination in voting and just as the 19th amendment prohibits sex discrimination in voting, the proposed amendment would prohibit age discrimination in voting.
Even the Fifth Circuit acknowledged, as it had to, that “[t]he language and structure of the Twenty-Sixth Amendment mirror the Fifteenth, Nineteenth, and Twenty-Fourth Amendments” and that it is proper, in interpreting a part of the Constitution, to “focus [on] how the same or at least similar terms that also appeared elsewhere in the Constitution ha[ve] been interpreted.” Indeed, comparison to the Fifteenth, and Nineteenth led the Fifth Circuit to conclude that plaintiffs had individual rights under the Twenty-Sixth Amendment such that they can sue (a preliminary issue the court addressed).
Given the obviously (and admittedly) intentional patterning of the Twenty-Sixth Amendment on the Fifteenth and Nineteenth, the case was a very easy one. The Fifth Circuit judges need merely have asked—and this really should have been enough to decide the case—what would happen if Texas had never permitted early voting by mail but then extended that option to Whites but not Blacks, or to men but not women. Would there be a question in anyone’s mind that the State had in these events violated the Fifteenth or Nineteenth Amendments, respectively? It would be absurd to say these laws did not “abridge” the freedom to vote within the meaning of the Fifteenth and Nineteenth Amendments simply because the laws weren’t taking anything away that Blacks or women had previously enjoyed, but instead were simply giving something new to other groups. Indeed, the clear unconstitutionality of these laws would be evident even if early voting by mail weren’t a particularly useful option to have (which it is). A law that changed the preexisting (uniform) closing time for the voting polls such that polls close two minutes later for White voters than for Black voters would undeniably abridge voting rights of Blacks and blatantly violate the Fifteenth Amendment.
The short of it is that when a state uses a facial classification based on race, sex, or age to condition access to voting in general or to any method of voting in particular, the government abridges the voting equality rights explicitly written into the Constitution. And unlike cases under the Fourteenth Amendment’s voting rights jurisprudence, we needn’t even ask what the “standard of review” is, or what interests the state might have to justify its differential treatment. As the Supreme Court observed in the Fifteenth Amendment context, “race . . . [c]olor and previous condition of servitude too are [simply] forbidden criteria or classifications.”
The Fifth Circuit intimated that plaintiffs may, on remand, be able to argue they have a claim under the Fourteenth Amendment’s Equal Protection Clause, in which case the strength of the government’s interest might be relevant (although it’s hard to imagine the government ever having a good reason to withhold any voting access on account of race, sex, or age). But the Fourteenth Amendment (under which age, I might point out, is ordinarily not a suspect classification) is simply beside the point. The voting equality amendments—including the Twenty-Sixth—cover voting much more explicitly than does the Fourteenth Amendment, the history of which suggests it was not designed to apply to political rights at all. Plaintiffs clearly should have won under the Twenty-Sixth Amendment, regardless of what claim they might have been able to make in its absence. (And it is frankly bizarre that the Fifth Circuit even technically leaves open on remand the possibility that plaintiffs can pursue their Fourteenth Amendment claim—how on earth could they win on the Fourteenth and lose on the Twenty-Sixth?)
None of this is to say difficult questions might not arise concerning what “abridge” means in the context of laws that do not overtly make use of age (or race or gender) classifications but that have a disparate impact against groups along any of these lines. For example, if a state were to close its polls at 6 pm (which would be earlier than other states) and there were evidence that such a decision adversely impacted racial minorities or women or young adults in particular (because of the jobs they tended to have), there would be complicated questions of what level of improper intent a plaintiff challenging such a law would have to prove. As Professor Daniel Ortiz pointed out a few decades ago, in the voting rights context the intent requirement the Court normally insists on in the equality-rights realm has been watered down such that the Court has been much more willing to accept a disparate impact theory than in other areas of equal protection law.
But in the Abbott case we needn’t even worry about such nuances. The law overtly discriminates against people based on their age. And that should have been the end of it.
The Fifth Circuit’s failure to understand any of this, and also to appreciate the group-equality nature of voting and voting rights, is extremely troubling, and somewhat surprising. The right to vote is an individual entitlement, to be sure. But voting is a hybrid right (as Alan Brownstein and I elaborated in a law review article in the late 1990s); the reason the Constitution singles out certain criteria, and the groups that are defined by those criteria, is that voting is more than an individual act—it involves a collective effort to exert political power to elect groups’ preferred candidates and enact groups’ preferred policies. There are obviously known (or knowable) partisan implications—both in the 1970s when the Twenty-Sixth Amendment and the Texas law were enacted and today as well—that ensue from giving any particular groups (including groups defined by age) greater or fewer voting options; that is one reason why, as noted above, even certain disparate impacts alone are sometimes problematic in the voting rights arena. And overt, facial discrimination against persons—on the grounds explicitly identified by the Constitution’s clear words as impermissible—is impossible to countenance. And yet a Fifth Circuit panel did so. We’ll see if this clear error gets rectified before the election.