The Supreme Court long held the drawing of legislative districts within the discretionary purview of the states.1 In a series of cases in the early 1960s, however, the Court began to recognize malapportionment claims under the Equal Protection Clause of the Fourteenth Amendment.2 Establishing the principle of “one person, one vote,”3 the Court stated that “the Constitution visualizes no preferred class of voters but equality among those who meet the basic qualifications.”4 But in these cases, the Court “carefully left open the question what population” states must equalize to achieve that ideal.5 Last Term, in Evenwel v. Abbott,6 a unanimous Court again declined to provide an answer, stating only that “a State may draw its legislative districts based on total population,” without reaching the question whether it must.7 Given this ambiguity, if a state moves to equalize both total population and voter population, then the Court will likely have to weigh nondilution of votes against other values, such as geographic regularity and continuity of communities of interest. Further, if the residential demography of noncitizen immigrants renders such a compromise unworkable, then states may increasingly face a choice between the two measures of equality — and the Court’s precedents indicate important reasons for deference to states as they assess the relevant political tradeoffs.
In 1962, responding to state legislators’ refusal to redistrict despite substantial rural migration, the Supreme Court opened the door to claims of vote dilution by holding them justiciable under the Equal Protection Clause in Baker v. Carr.8 Subsequently, several state apportionment plans — including a state legislative plan imitating the Federal Senate, a dilutive congressional map, and a primary system modeled after the Electoral College — failed the new constitutional test.9 But while the Court grandly declared that “[t]he conception of political equality . . . can mean only one thing — one person, one vote,”10 it consistently avoided identifying whether districts must have equal numbers of voters, citizens, or people. That decision “involves choices about the nature of representation with which we have been shown no constitutionally founded reason to interfere,” as the Court later explained in Burns v. Richardson.11
Following the 2010 Census, Texas adopted a permanent State Senate map (S172) that equalized total population within presumptively acceptable parameters.12 However, the plan created significant inequality among districts under an eligible- or registered-voter standard.13 Plaintiffs Sue Evenwel and Edward Pfenninger — who live in rural districts with relatively large numbers of eligible and registered voters14 — argued that S172 therefore unconstitutionally diluted their votes, asserting that “the ‘population’ that must be equalized for purposes of the one-person, one-vote rule” is the voting population.15
The District Court for the Western District of Texas rejected this argument. Citing Fourth and Fifth Circuit cases, a three-judge panel held that courts could not “‘interfere’ with a choice that the Supreme Court has unambiguously left to the states absent the unconstitutional . . . exclusion of specific protected groups of individuals.”16
The Supreme Court affirmed. Writing for the Court, Justice Ginsburg17 held, “based on constitutional history, this Court’s decisions, and longstanding practice,” that states could use total population as the apportionment base.18 When considering constitutional history, she stressed the Framers’ decision to use total population in dividing representation in the House among the states.19 Further, she argued, opposition to voter-population apportionment motivated the drafters of section 2 of the Fourteenth Amendment to retain the congressional apportionment base. On this point, she quoted at length from a tribute to representation delivered by Senator Jacob Howard while supporting that amendment in the Senate: “Numbers, not voters; . . . this is the theory of the Constitution.”20
Rejecting the use of “selectively chosen language” from Court precedent, Justice Ginsburg observed that “[f]or every sentence appellants quote from the Court’s opinions, one could respond with a line casting the one-person, one-vote guarantee in terms of equality of representation, not voter equality.”21 She also noted that the Court had used total-population statistics to evaluate districting plans and expressly permitted uneven distributions of voters.22
Finally, the “settled practice” of all fifty states and numerous local jurisdictions weighed against intervening.23 Following these three rationales against forcing states to choose voter population, Justice Ginsburg added two policy concerns to affirmatively support that choice: access of nonvoters to constituent services and their “important stake in many policy debates.”24 But since the former reasons proved sufficient to dismiss the appeal, Justice Ginsburg declined to decide whether “States may draw districts to equalize voter-eligible population.”25
Justice Thomas concurred in the judgment. He agreed that states may choose to equalize total population, but wrote separately to note that the “Court has never provided a sound basis for the one-person, one-vote principle.”26 He characterized the conflicting language referenced by the majority and appellants as only natural because the Constitution does not require any particular measure.27 In seeking to promote the common good, which they understood as “objective and not inherently coextensive with majoritarian preferences,” the Framers perceived a tension between allowing the people to govern themselves and guarding against tyranny of the majority.28 Thus, they charged states in Article IV, section 4 with striking a balance to maintain a “Republican Form of Government.”29 Further, he wrote, the Reconstruction Amendments left this original understanding unchanged.30
Justice Thomas then criticized the majority’s “attempt to impose its political theory upon the States.”31 Warning that the judicial branch had “arrogated . . . important value judgments” by mandating its views and providing inconsistent guidance, he rejected the “faulty premise that ‘our system of representative democracy’ requires specific groups to have representation in a specific manner.”32
Justice Alito concurred in the judgment.33 He agreed that the one-person, one-vote rule permitted equalization of total population.34 Further, regarding the dispute between Texas and the United States over the meaning of Burns and “very difficult theoretical and empirical questions about the nature of representation,” he joined Justice Ginsburg in seeing “no need to wade into these waters in this case.”35
Nonetheless, Justice Alito wrote separately to dispute the majority’s “suggest[ion],” urged by the United States, that the Constitution’s plan for congressional representation somehow directs that state legislative districts must equalize total population.36 Noting that that plan itself contradicts one person, one vote, he described how it emerged directly from interstate struggles over political power.37 Characterizing the majority’s inference of a general theory of representation as “profoundly ahistorical,” he then criticized its reliance on words “pluck[ed] out of context” from both the Constitutional Convention and the drafting of the Reconstruction Amendments.38 This history, he argued, supported the Court’s earlier view that “the Founding Fathers clearly had no intention of establishing a pattern or model for the apportionment of seats in state legislatures” when they drafted the Constitution.39
After Evenwel, it is clear that states may continue to equalize total population. And while Burns held that states may instead equalize voter population in at least some circumstances, the extent of its reach remains contested.40 In the face of this ambiguity, as well as a circuit split on the issue, some states could move to equalize both population metrics, as several Justices considered41 — thus assuaging the concerns of both Justice Ginsburg and Sue Evenwel. In such a case, cartographic realities would likely force a choice between nondilution of votes and other important interests, such as geographical regularity of boundaries and contiguity of communities of interest.42 Further, if the residential demographic patterns of noncitizen immigrants render such a panacea practically impossible,43 then states will face a hard choice between the two measures. A state decision to equalize voter population instead of total population would force the Court to confront the tension between the majority’s favorable view of total-population apportionment, which rested in significant part on nonvoters’ role in policy debates, and the Court’s history of deferential review of state laws related to self-government and the democratic process. This tension, which reveals the political nature of the underlying issues, suggests the prudence of judicial deference to state legislatures in such a case.
Three circuit courts have ruled on this question. The Ninth Circuit dismissed an argument similar to that made by the plaintiffs in Evenwel in Garza v. County of Los Angeles.44 But it went further than the Evenwel majority by selectively quoting from the Court’s conflicting language, inferring a right of noncitizens “to influence how their tax dollars are spent,” and holding that “districting on the basis of voting capability . . . would constitute a denial of equal protection” to noncitizens.45 In Evenwel, both the majority and the United States in its amicus brief endorsed several of Garza’s arguments, albeit without following them to Garza’s conclusion.46 The Fourth and Fifth Circuits, however, held that the choice of apportionment method belongs to states; these holdings received some support in Evenwel’s concurrences.47
Given the important considerations supporting both voter equality and equal representation, equalizing both measures would seem a pragmatic solution for at least some states.48 However, like squeezing a balloon, equalizing both metrics at once would likely create distortions elsewhere. When pressed on this point, counsel for Texas and the United States each warned of potential disruptions to “traditional redistricting factors” such as compactness, contiguity, and “keeping communities together.”49 But while the Court has recognized that these traditional redistricting factors protect important values,50 it has also clearly stated that they are not constitutionally required.51 Thus, if states do equalize both populations, courts may defer to that decision — though the level of scrutiny applied could be determinative.52
Even so, some states may find it impossible to equalize both measures,53 and the level of scrutiny may also be pivotal if states choose to equalize voter population instead of total population. This prospect is not merely hypothetical; some believe that citizens of states affected by recent demographic trends may soon join the Evenwel plaintiffs in objecting to the dilution of their votes.54 The number of noncitizen immigrants in the United States has grown twelvefold relative to the citizen population since Baker, and they “tend to cluster in certain . . . areas, often in the same communities” as citizens of similar backgrounds.55 While disparate populations of minors and felons may also play a role in unbalancing voter populations,56 the impact of concentrated populations of noncitizen immigrants has driven most recent high-profile one person, one vote cases.57 If a state does choose to equalize voter population by bending on total population, the Court will be forced to answer the question it evaded yet again in Evenwel — or at least to decide whether to extend Burns to include noncitizen immigrants. Considering this question would require the Court to further develop the two rationales employed by the majority to support its theory of representation: access to constituent services and policy influence, or in other words, “requests and suggestions.”58
The concern about “requests” for constituent services is practical. Representatives provide services to their constituents; a plan that does not equalize total population could dilute access to these services.59 Citizenship classifications with regard to government services have received heightened scrutiny from the Court in the past,60 and some have argued that “judges should take constituent service seriously.”61 But “diluted access” seems difficult to prove; after all, the density of requests surely varies over time, both within and across districts. And as Justice Ginsburg admitted, it has never been determined that a constituent has a right to equal access to, much less a response from, her representative.62 Further, Evenwel’s pages of debate over the Framers’ intended theory of representation would seem peculiar if that theory were based upon provision of “help navigating public-benefits bureaucracies” or distribution of pork spending, since neither concept existed in anything like its modern form when the Constitution was drafted.63 For the constituent-service theory to motivate strict scrutiny, it would need grounding in a more fundamental theory of representation.
The “suggestions” prong of the majority’s reasoning strikes just such a chord — one that resounds with appropriate depth for a potential “theory of the Constitution.”64 It gets to the heart of a fundamental question: What, exactly, does representation mean? While nonvoters, tautologically, cannot vote, Justice Ginsburg wrote that they “have an important stake in many policy debates” and an implied entitlement to influence those debates through their representatives.65 This argument supports total-population apportionment because it gives nonvoters power over the democratic process of policymaking.
For this very reason, however, use of the “suggestions” argument to overturn a state choice of voter-population base would conflict with a line of cases showing rational basis deference to state laws related to democratic sovereignty. Classifications based on alienage usually receive strict scrutiny under Graham v. Richardson,66 which held that noncitizen immigrants were entitled to “heightened judicial solicitude” as a “‘discrete and insular’ minority.”67 And since Burns carefully qualified its deference to state choice with the caveat “[u]nless a choice is one the Constitution forbids,” the application of strict scrutiny to a plan that equalized voter population could prove determinative.68 However, the Supreme Court has made a consistent exception to Graham’s general rule, using rational basis review to scrutinize citizenship classifications related to self-government and the democratic process.
Because “a democratic society is ruled by its people,” these cases upheld state laws entrusting roles involving “important policy responsibilities” exclusively to citizens.69 This exception was limited to those who “‘participate directly in the formulation, execution, or review of broad public policy’ and hence ‘perform functions that go to the heart of representative government’” in Bernal v. Fainter.70 But the Court has made clear that these functions include voting: “[W]e have recognized ‘a State’s historical power to exclude aliens from participation in its democratic political institutions’ as part of the sovereign’s obligation ‘to preserve the basic conception of a political community.’”71
Justice Ginsburg’s vision of representation, to be sure, would not directly transform noncitizens into voters — but it does reveal a basic tension. Supporters of a theory of equal representation will argue that the policy influence envisaged by the Evenwel majority does not meet Bernal’s standard and that a state choice of voter population therefore deserves strict scrutiny.72 Indeed, the Ninth Circuit cited Bernal to assert that the “equal protection right” afforded to aliens under the Fourteenth Amendment “allow[s] political participation short of voting or holding a sensitive public office.”73 But to the extent that the Court may grant certain voters extra weight for the express purpose of enabling their noncitizen relatives and neighbors to influence the democratic process, it seems to contradict itself. Having declared that “it is clear that a State may deny aliens the right to vote,” because this “lie[s] at the heart of our political institutions,”74 can the Court now coherently command states to count noncitizens’ votes by proxy?
Confronting this tension, of course, leads to a number of deeper issues, including the definition of a political community,75 the fundamental meaning of representation, and the prudent balancing of humanitarian values like inclusion and hospitality with cultural and political imperatives like assimilation and national identity.76 These thorny questions have no easy answers. But perhaps this is why Justice Frankfurter warned in Baker that “[t]o charge courts with the task of accommodating the incommensurable factors of policy that underlie these mathematical puzzles is to attribute, however flatteringly, omnicompetence to judges.”77 If forced to address the question that Evenwel — and, as some argue, the Constitution78 — left open, the Court should heed that warning by deferring to states, rather than selecting a political theory to foist upon them.79