Ideally, litigants come to appellate courts with a problem the courts can solve.  Sometimes, though, they bring solutions in search of a problem the courts can create. The plaintiffs in Evenwel v. Abbott have gone even further: Their case brings the U.S. Supreme Court a problem and asks the Court to create more problems, with no solution in sight. They want the Court to completely upend the current system of drawing legislative districts—in a way that would give more power to conservative voters and candidates. Beyond that, they are asking the Court to adopt a new constitutional rule with no constitutional provision attached.

Evenwel, which the Court will hear next Tuesday, is a challenge by a group of registered Texas voters to the state’s plan of districts for the state senate. The Texas legislature drew its new districting plan on the assumption that it should try to make each district roughly equal in population to every other. The plaintiffs in Evenwel challenged that plan, however, on the grounds that the legislature should use eligible voters, rather than total population, as the relevant measure. Each district, in other words, should have roughly the same number of eligible voters, not the same number of people.

The change would produce a political earthquake. Eligible voters as a group are older (no children under 18, to begin with), wealthier, and more Republican—and, even more important in Texas, whiter and more Anglo—than the population at large. Many people in the Southwest—both legal residents and undocumented immigrants—are not citizens. Under the proposed Evenwel rule, only those eligible to vote count.

The plaintiffs cite two seminal cases, Baker v. Carr and Reynolds v. Sims, which together are considered (in shorthand) to have established a rule that districting must be done on a “one person one vote” rule. “The Court,” the plaintiffs argue, “need not look beyond these seminal decisions to resolve the question presented in Appellants’ favor.”

Indeed, the appellants hope the Court won’t look beyond them—because the proposed rule is anchored in scattered language from those opinions, not in the constitutional principle they drew from. In Baker, the Court held for the first time that a state’s legislative districts, if drawn unequally, could be challenged under the Equal Protection Clause. In Reynolds, the Court for the first time struck down a state legislative-districting plan because it drew districts unequal in population. Chief Justice Earl Warren wrote for the six justices that “the weight of a citizen's vote cannot be made to depend on where he lives.” For this reason, he continued, “Population is, of necessity, the starting point for consideration and the controlling criterion for judgment in legislative apportionment controversies ... We hold that, as a basic constitutional standard, the Equal Protection Clause requires that the seats in both houses of a bicameral state legislature must be apportioned on a population basis.”

In a later case, the Court held that Hawaii could use registered voters rather than raw population as the basis for its districts. The majority reached that conclusion because the state had an unusually high number of transient military personnel and tourists—and “only because on this record it was found to have produced a distribution of legislators not substantially different from that which would have resulted from the use of a permissible population basis.” The Hawaii case, Burns v. Richardson, has been read to mean that states have a choice of basis, as long as its plan remains close to overall population numbers. The plaintiffs want to replace that loose rule with a no-choice “eligible voter” requirement. They want the Court to read some of its previous language aggressively (“the weight of a citizen’s vote”) and to ignore other language (”Population is ... the starting point ... and the controlling criterion”).

One would hardly know it from the appellants’ brief, but the basis of all the Court’s districting cases is the Equal Protection Clause of the Fourteenth Amendment. That provision protects the rights of “persons” within a state’s jurisdiction. The framers of the clause knew the difference between “persons,” “citizens,” and voters: Elsewhere in Section One they guarantee the “privileges or immunities” of “citizens of the United States,” rather than of “persons”; in Section Two they discuss the right to vote of some, but not all, “citizens of the United States.”

As for representation, Article I requires representatives to be divided among the states “according to their respective numbers,” meaning population. There was no exception for non-citizens, or for children; slaves were also counted, though only as three-fifths of “free persons.”

The Fourteenth Amendment did away with the three-fifths rule, basing House seats on “the whole number of persons in each State”—population again. The amendment makes an exception—when eligible male citizens (women were not guaranteed the right to vote until 1919) are denied the vote on any basis, states were to lose House seats for all members of the denied group. This provision was never applied, but I read it to mean that if, say, adult male voters were denied the ballot because they were African American, the state would lose representation for its entire African American population—men, women, and children.

I find no hint anywhere in the text that the boundaries of political power could or should ignore children, or women (who were then ineligible to vote), or non-citizens. And the debate over the amendment shows a good deal of awareness of the needs of immigrants, as well as citizens, for “the equal protection of the laws.”

The Evenwel “rule,” in other words, has been patched together more or less out of thin air, supplemented by a highly partial reading of a few Warren Court cases. Its result would be a political system even less tethered to the concerns of ordinary people than the present one. That prospect undoubtedly appeals to some Americans at this moment of demographic panic—of a nostalgia for an alternate universe past when America was “greater,” whatever that means. But it makes little sense within the genuine political universe created by the Constitution we actually have. It makes no sense as law.

Victory for the plaintiffs seems unlikely, however. The Court is hearing this case not so much out of choice but because, as a direct appeal from a three-judge court, it can’t just refuse. Whether it heard arguments or not, its decision would set a precedent. And the plaintiffs are on shaky ground. They are in essence asking the Court to open itself up to a decade or more of hell adjudicating its new rule. Consider this soothing phrase from their brief: “This appeal need not resolve every implementation issue.” Indeed. A rule basing districts on “eligible voters” would be a nightmare to administer, with district courts around the country required to find information about eligible voters—information that’s not available anywhere.  

As an amicus brief for a group of former directors of the Census explains, “there is no actual count of the number of voting age citizens.” The Census counts the number of people in the country. The only systematic information about citizen population is a sample, like a voter poll in other words, taken by the Census. And that sample is too small to produce a reliable estimate of citizens in a given district, the former officials warn: “Adequate data to support Appellants’ positions simply do not exist.”