Edward B. Foley
Originalism and Election Law (or, The Difference between Reynolds and Benisek)
Election Law as a distinct field of study was founded on decidedly “non-originalist” premises. The seminal case, after Baker v. Carr set aside the “political question doctrine” as an obstacle to the judicial entry to this field, was Reynolds v. Sims. Decided in 1964, during the heyday of Warren Court activism, Reynolds declared that the Equal Protection Clause of the Fourteenth Amendment required equally populated districts for a state’s legislative chamber. The Warren Court made no effort to derive this “one-person, one-vote” requirement from the original meaning of the Equal Protection Clause.
Indeed, the Court obviously could not do so, since the Fourteenth Amendment (in its second section) explicitly acknowledged that states were entitled to deny equal voting rights to their citizens, the only consequence being that states would lose strength in their share of congressional seats nationwide to the extent that they denied voting rights to adult males 21 or older (except for reason of a felony or participation in the Civil War on the side of Confederacy). Moreover, it was necessary to add the Fifteenth Amendment in order to protect the right to vote from race-based discrimination. The Fifteenth Amendment, however, was mere surplusage if the Fourteenth already guaranteed each adult citizen an equal right to vote. In his Reynolds dissent, Justice John Marshall Harlan (one of the jurisprudential giants in the Court’s history) excoriated the majority opinion for abandoning all pretense of fidelity to the Constitution’s actual language and original meaning.
Nor has academic defense of Reynolds endeavored to square its “one-person, one-vote” doctrine with the original meaning of the Equal Protection Clause. On the contrary, the leading scholarly treatment of Reynolds is the one offered by the late (and great) John Hart Ely, who candidly justified Reynolds as a “non-interpretivist” (not merely “non-originalist”) exercise of judicial review warranted by the overarching objective of making American government more democratic in nature. Dubbed a “representation-reinforcing” theory of judicial review, and invoking the famous Footnote Four of the old (and otherwise inconsequential) Carolene Products case, Ely argued that the Supreme Court could retain legitimacy for rulings untethered to specific constitutional clauses as long as those rulings enhanced democratic procedures (while protecting “discrete and insular minorities” from the pathologies of majoritarian tyranny) and scrupulously refrained from imposing substantive policy choices not expressly necessitated by constitutional text.
But Reynolds is over a half-century old, and the progressive activism of the Warren Court is long since over. Nor has Ely’s theory, for all its elegance, fared well as an explanation of the Court’s animating philosophy during the Chief Justiceships of Warren Burger, William Rehnquist, and now John Roberts. During the last several decades, under the energetic influence of Justice Antonin Scalia, the Court has become increasingly originalist in its overall interpretative approach to previously unresolved constitutional questions. And Ely’s idea that the Court could ever be justified in being altogether “non-interpretivist” in the exercise of its power of judicial review under Marbury v. Madison has become entirely untenable. Thus, Reynolds sits uneasily in the canon of contemporary constitutional jurisprudence, a fact made evident when the Court was forced to reexamine the basic premises of the “one-person, one-vote” doctrine in the recent Evenwel case. Was it equal number of voters, or residents, that was constitutionally required by Reynolds, or may a state choose whichever theory of electoral equality it prefers? The answer remains unsettled, except that the Court rejected the claim that an equal number of voters is required if a state prefers to equalize population. (It also doesn’t help that Reynolds was the jurisprudential predicate for the Court’s controversial recount-terminating ruling in Bush v. Gore—a decision more easily justified on Due Process grounds, insofar as Florida improperly changed the rules for counting ballots after they had been cast.)
Initially, the constitutionality of partisan gerrymandering was thought to involve an extension of Reynolds: yes, the districts may be equally populated, but there is still a denial of voter equality caused by the manipulation of district lines. That’s how the claim was argued in Davis v. Bandamer, when in 1986 the Court first confronted a constitutional challenge to partisan gerrymandering. And that’s how the claim was argued again in Vieth v. Jubilerer, when the issue came back before Court in 2004. Moreover, there is even a hefty dose of Reynolds as the constitutional predicate for the argument in the pending Gill v. Whitford case that a districting map with an excessive “efficiency gap” violates Equal Protection.
But insofar as any constitutional attack on partisan gerrymandering is premised on Reynolds, the claim is presumptively dubious for any Justice inclined towards originalism as the proper method of constitutional interpretation. Even if Reynolds need not be overruled as an egregious non-originalist mistake, surely there is no obligation to extend that mistake by sustaining a Reynolds-based challenge to gerrymandering. Moreover, given that so much energy for more than thirty years has been devoted to deriving a successful anti-gerrymandering claim from the flagrantly non-originalist Reynolds, it is understandable that one might assume that it is impossible to mount a genuinely originalist challenge to partisan gerrymandering. That assumption, however, is incorrect. If we bypass Reynolds altogether, and examine the issue from a fresh perspective, we can see that congressional gerrymandering in particular—the type at issue in the pending Benisek case from Maryland—is constitutionally infirm according to a properly originalist interpretation of Article I and its role in the overall design of the original Constitution.
As explained in the previous post in this series, the original understanding of Article I was that elections to the federal House of Representatives are to occur “every second Year,” rather than the longer electoral terms for Senators and the President, so that the House of Representatives serves as the singular part of the federal government directly responsive to the frequently adjusting will of the “People” of the United States. The Framers of the Constitution, moreover, designed these congressional elections as a component of the overall structure of the federal government with the overriding objective that the government genuinely act in accord with the public interest, rather than on behalf of any particular “faction” or self-serving private interest of a mere segment of society. If there is one thing that the Framers wanted their new Constitution to avoid, above all, it was for Congress to become captured by the narrow interests of a “faction” or what we today would term a political party. Thus, insofar as partisan gerrymandering causes Congress to remain beholden to the interests of a political party long after that party has lost public support—and thus Congress pursues the interests of a faction rather than the public interest, and yet gerrymandering has distorted the electoral system so that “the People” are unable to make Congress responsive to the public will rather than the self-serving faction—then gerrymandering contravenes the most basic original understanding of how biennial elections to the federal House of Representatives were supposed to operate. Fidelity to this original understanding requires invalidation of the contravening gerrymander.
Focusing on the originalist antipathy towards faction properly distinguishes the constitutional condemnation of partisan gerrymandering from the anti-originalist imposition of “one-person, one-vote” in Reynolds. There is nothing in the original Constitution, or the Fourteenth Amendment, that required states to district their own legislatures according to a principle of equal population. The Framers knew that there could be valid principles of legislative apportionment, based on legitimate geographical considerations, for why a state’s own legislative districts might deviate from strict population equality. (A desire for balance between downstate and upstate in New York, for example, might be one such legitimate consideration, or between the tidewater and piedmont regions of Virginia.) But if the federal Congress became captured by the mere desire of a faction to perpetuate itself in power, without regard to valid geographical factors, that circumstance would be directly antithetical to what the Framers were aiming to accomplish with their constitutional design, including specifically the biennial nature of elections to the federal House of Representatives as a key component of that design.
The fact that gerrymanders occurred almost immediately after adoption of the Constitution, before the proverbial ink was dry, in no way undercuts this originalist conclusion. In his Vieth plurality, Justice Scalia seemed to think that partisan gerrymanders could not be unconstitutional given their existence going all the way back to the time of the Founding (and, indeed, in the colonial period before). But as Justice Scalia himself recognized in many other writings, the proper methodology of originalism—what he and others have technically called the “original public meaning” of the Constitution—does not work this way. Often provisions are included in a Constitution precisely because its authors are acutely aware that ordinary politicians, succumbing to the inevitable pressures of ordinary politics, will become agents of a self-serving faction rather than of the general public interest, and thus there needs to be a constitutional constraint upon such ordinary political behavior. Consequently, the mere fact that this kind of improper political conduct begins to occur almost immediately after adoption of the Constitution is not an indication that the conduct is constitutionally permissible. On the contrary, it simply underscores the accurate assessment of the need for this constitutional constraint.
The Sedition Act of 1798 can serve as useful example of this basic, but important, point. The fact that within a decade after adoption of the Constitution one faction (the Federalists) criminalized the expression of dissent by the faction’s political opponents (the Democratic Republicans) does not make criminalization of political dissent consistent with the First Amendment. On the contrary, the Sedition Act confirmed the necessity of the First Amendment, to help thwart the efforts of a faction to undermine political opposition—and even worse, with the goal of perpetuating the faction in power regardless of shifting public sentiment, thereby prevent elections from being responsive to the will of the “People” and the popular desire for a change in the political composition of Congress. In New York Times v. Sullivan, the Supreme Court recognized that the Sedition Act was contrary to the original meaning of the First Amendment even though the Sedition Act was adopted shortly after the First Amendment itself. In the same way, the Supreme Court now should recognize that extreme partisan gerrymandering is antithetical to responsiveness of congressional elections that Article I is structured to achieve, and this proper originalist point remains true even though partisan gerrymanders started to appear shortly after the adoption of the Constitution itself.
Thus, it might prove surprising to originalists who have not given the issue much thought, but an honest and principled commitment to originalism requires recognition that faction-motivated gerrymandering is contrary to the original understanding of how elections to the federal House of Representatives are supposed to work. Recognizing this truth does not entail acceptance of the anti-originalist Reynolds. Just the opposite: a principled originalist would sharply distinguish congressional gerrymanders from the apportionment of a state’s own legislature. The original Constitution was structured to give states latitude in the particular ways they chose, within the overall federalist system, to pursue their conception of how best to achieve popular sovereignty within the state, or what the Constitution itself calls a “republican form of government.” Reynolds undercuts this latitude the states were supposed to have.
But with respect to the federal House of Representatives, the originalist understanding was altogether different. Above all, factions were not supposed to subvert the direct accountability to the “People” that biennial elections to the House were designed to achieve. Insofar as faction-driven gerrymanders do exactly that, faithful adherence to originalism must condemn them as unconstitutional.