Gerrymandering—the manipulation of political districting processes and boundaries for partisan political advantage—has proven a troubling and difficult area of constitutional concern. This is partly due to the exceptionally divergent standards of judicial review applicable depending upon the basis for the gerrymander claim. The Supreme Court has consistently held that racial gerrymanders are subject to strict scrutiny review and presumptively violate the Equal Protection Clause of the Fourteenth Amendment. The Court has recently declared that partisan gerrymanders, on the other hand, are a political question and non-justiciable.
This Article argues that current guidance from the Supreme Court on standards for evaluating gerrymandering claims is inadequate to guard against constitutional violations because of the problem of entanglement: the race and partisan preferences of voters are so deeply intertwined in many contexts that it is practically impossible to discern whether race or partisanship was the basis for political districting decisions. The entanglement of race and politics in political districting processes means that there is a dangerous risk that unconstitutional racial gerrymanders will escape judicial review under the cover of partisanship.
This Article explicates the problem of entanglement in gerrymandering cases and evaluates several possible solutions. Presenting original research drawn from the 2020 decennial census and voter data from the 2020 presidential election, this Article establishes an empirical basis for the problem of entanglement. Although prior legal scholarship has emphasized the problem of “conjoined polarization”—the overlap in partisan and racial preferences—as an enabling factor in partisan redistricting processes, this Article claims that racial residential segregation plays a more central and dynamic role than has generally been acknowledged in undergirding the entanglement of race and politics in political redistricting processes.
The manipulation of political districting processes for political advantage—popularly known as gerrymandering—has long bedeviled the United States, but concerns about the abuse of this practice have intensified in recent decades due to a confluence of factors: intensifying partisan political polarization, widening racial political polarization, the use of detailed voter files to predict voting behavior, the emergence of sophisticated computer technology to generate ever-more precise political maps, and a sharp divergence in the Supreme Court’s jurisprudence governing different forms of this practice.
In reviewing suits brought to challenge gerrymandering practices, the Supreme Court has held that state legislative efforts to draw political districts based on race violate the Equal Protection Clause, a natural extension of the Court’s general prohibition on the use of racial classifications in policymaking. On the other hand, the Supreme Court has held that legislative efforts to draw political districts based upon partisanship or for partisan political advantage are “political questions” and non-justiciable.
In this regard, these two forms of gerrymandering are treated in the utmost extreme: racial gerrymandering is subject to the highest level of judicial scrutiny while partisan political gerrymandering is treated as non-justiciable, meaning not that it is subject to the lowest level of judicial review, rational basis review, but that the practice is deemed unsuitable for judicial review at all. Racial gerrymanders are subject to strict scrutiny judicial review whereas partisan political gerrymanders are not subject to judicial review whatsoever.
The Supreme Court’s broader equal protection clause jurisprudence supplies a basis for treating these two types of claims differently. Prevailing equal protection jurisprudence treats race as a “suspect” class in government policymaking subject to strict scrutiny review, while most other classifications are reviewed under a rational basis test. But strict adherence to this approach would compel a very different result than the determination that partisan gerrymanders are non-justiciable. Lower courts would still be able to entertain such cases, just under a much lower level of review, rational basis.
If there were no relationship between race and partisanship in voting patterns, then political gerrymanders and racial gerrymanders could be regarded as separate and distinct categories and there would be no logical inconsistency in a jurisprudence that regulated one but not the other. Partisan gerrymanders would have no observable racial effect, or vice versa. In practice, however, race has long been highly correlated with partisan political affiliation. Although racial political polarization waxes and wanes over time, it is strong enough that a jurisprudence of gerrymandering cannot neatly divide the two types.
The Court’s racial gerrymandering jurisprudence makes clear that sorting voters into separate political districts on the basis of race is unconstitutional, just as it is presumptively unconstitutional to sort pupils into different schools on the basis of race. In racially diverse states with racially polarized voting patterns and merely modest levels of racial residential segregation, however, it is likely that partisan gerrymandering will effectively sort people into different districts on a racial basis. In much of the country, race and partisanship are entangled, such that redistricting efforts on one basis are largely indistinguishable from the other. As a consequence, unregulated partisan gerrymanders have a dangerous potential to subvert the constitutional rule against racial gerrymandering.
Although political scientists have long recognized the correlation of race and partisan affiliation (what political scientists term “conjoined polarization”), prior analysis of gerrymandering jurisprudence has underexamined the specific role of racial residential segregation in facilitating the entanglement of race and politics in redistricting processes. In recent legal scholarship analyzing this problem, segregation is either completely absent from the discussion, mentioned in passing, or is treated as an assumed operative background condition. The role of segregation in relation to gerrymandering processes is both more central and more dynamic than is generally appreciated.
This Article argues that it is the interaction of racial residential segregation and racial political polarization that creates the entanglement problem in redistricting processes, not merely “conjoined polarization” by itself. Where the level of racial residential segregation is higher, the entanglement of race and politics in districting processes is likely to be greater, not only because of the geographic concentrations of people that facilitate political district line-drawing, but also because regions with higher levels of racial residential segregation have both greater racial political polarization and partisan political polarization.
This Article presents original analysis of the 2020 presidential election results and 2020 census data to demonstrate that racial segregation and partisan segregation are strongly correlated. Moreover, regions with higher levels of racial residential segregation appear to have higher levels of partisan polarization. As a result, partisan gerrymanders in those regions are likely to result in the segregation of voters into different political districts on the basis of race and vice versa.
Part I provides a brief history of gerrymandering, including the types and forms of political districts that were historically practiced. Political districts were far more varied in the early years of the republic than is generally appreciated or understood today. More importantly, Part II notes that although gerrymandering practice can be traced to the early decades of the republic, efforts to curb it also extend back into the nineteenth century. Standards and norms for democratic practice have improved and evolved since the framing of the Constitution, laying the groundwork for particularized claims brought to challenge this practice.
Part II compares racial gerrymandering and partisan political gerrymandering cases, rulings, and reasoning. It analyzes points of divergence and convergence between the two lines of cases. The partisan and racial gerrymandering cases germinate from the same seed and the same soil but have produced extremely divergent results in the body of the Supreme Court’s precedent governing these cases. This creates a problem in cases brought that challenge redistricting where race and partisan affiliation are largely co-extensive. In such cases, racial gerrymandering could escape judicial scrutiny under the cover of partisanship.
Part III explicates the entanglement problem, that purely partisan redistricting maps are in many cases objectively indistinguishable from redistricting maps that explicitly use race. The key components of this problem are racial political polarization and racial residential segregation. When these factors coincide, partisan gerrymandering is likely to sort people into different districts on a racial basis. Part III also shows that racial residential segregation plays a larger role than is generally appreciated in both racial and partisan gerrymandering processes. It presents original and other recent empirical research suggesting that regions with higher levels of racial residential segregation have both more racial political polarization and political segregation.
Part IV reviews three possible ways to address the entanglement problem in terms of current constitutional law and text, weighing the merits of each. First, any hybrid gerrymandering case in which race appears to play a significant role but is co-extensive with partisanship could be categorically exempted from judicial review if the state raises such a defense. This approach is not a functional solution because it would formalize a loophole for subverting the Constitution as long as racial gerrymanders are clothed in the guise of partisanship.
Second, any case where race and partisanship are co-extensive could instead be drawn within the racial gerrymandering line and held to strict scrutiny review, even though race cannot be said to “predominate.” This approach would better align with the Court’s broader anti-classification jurisprudence but would require adjustments to the standards applicable to racial gerrymandering cases.
Finally, the Court could reverse its judgment that partisan gerrymanders are non-justiciable. The Court only recently gathered a majority of Justices in support of that view. It could reverse course and direct lower courts to review such claims under a lower standard of review within the equal protection jurisprudence or some other constitutional provision or basis altogether. In this regard, Part IV makes the case for revisiting the Court’s Guarantee Clause jurisprudence based upon principles and concerns articulated by the framers of the Constitution.
I. A BRIEF HISTORY OF GERRYMANDERING
Although the United States was still a young nation at the time of the ratification of the Constitution in 1787, the framers already enjoyed decades of cumulative experience with democratic political processes, including political districting, based upon the collective experiments already underway in the various states since the Revolution. The Federalist Papers, for example, note political districts of varying size and composition both within and between states as a matter of fact.
In Federalist No. 57, James Madison observes that different sized political districts contribute to both the federal and state legislatures: “The city of Philadelphia is supposed to contain between fifty and sixty thousand souls. It will therefore form nearly two districts for the choice of federal representatives. It forms, however, but one county, in which every elector votes for each of its representatives in the State legislature.” Thus, Pennsylvania’s county-districting system for electing state legislators necessarily resulted in large population disparities between political districts in that state at the time of the adoption of the Constitution. But in Federalist No. 61, Alexander Hamilton notes that although the New York State Assembly is drawn from counties, the New York State Senate is drawn from districts composed of two to six counties apiece.
While acknowledging the existence of population disparities between political districts (and implicitly, the existence of inequities in political representation), both Madison and Hamilton unequivocally maintain throughout The Federalist Papers that the principle of majoritarianism—that the majority should prevail—is the fundamental basis of free government and republican government. In Federalist No. 58, for instance, Madison asserts that “the fundamental principle of free government” is that the “majority would rule.” In that context, he was writing against the suggestion made by critics of the proposed Constitution that supermajorities should be required for either a quorum or a decision (such as passing a law) in the House of Representatives. As he explains: “In all cases where justice or the general good might require new laws to be passed, or active measures to be pursued, the fundamental principle of free government would be reversed. It would be no longer the majority that would rule: the power would be transferred to the minority.”
Indeed, this is one of the chief objections the framers had with the Articles of Confederation, which gave equal suffrage to each state in the federal legislature (unlike the Constitution, which does so only in one legislative chamber, the Senate). Prior to the constitutional convention in Philadelphia in the summer of 1787 where the Constitution was hammered out, Madison privately wrote to Thomas Jefferson expressing his hopes for systemic changes to the federal government. Chief among these concerns was converting from a system in which each state receives equal voting power in Congress to a system of representation based upon population.
Hamilton firmly agreed. In Federalist No. 22, Hamilton maintains that “the fundamental maxim of republican government . . . requires that the sense of the majority should prevail.” Therefore, in his view, Every idea of proportion and every rule of fair representation conspire to condemn a principle, which gives to Rhode Island an equal weight in the scale of power with Massachusetts, or Connecticut, or New York; and to Deleware [sic] an equal voice in the national deliberations with Pennsylvania or Virginia, or North Carolina.
In a powerful and eloquent denunciation of the principle of equal suffrage between states, Hamilton goes on to develop the argument on the “impropriety of an equal vote between States of the most unequal dimensions and populousness” in various ways.
This argument, however, and all of the reasoning developed in support of it, would appear to have equal force against political districts within states of “most unequal dimensions and populousness.” Indeed, in Federalist No. 46, Madison asserts that “[e]very one knows that a great proportion of the errors committed by the State legislatures proceeds from the disposition of the members to sacrifice the comprehensive and permanent interest of the state, to the particular and separate views of the counties or districts in which they reside.”
How can the principles, reasoning, and keen insights developed by Hamilton in pushing for more proportional representation between states in the federal government be reconciled with the apparent lack of concern with unevenly populated political districts within states or the absence of an explicit mechanism regulating it? The answer is not clear. It may have been an oversight. There were many weighty matters that preoccupied the Constitutional Convention, and the issue of unequal political districts within states may not have been a topline concern. Or, if it were a serious concern, perhaps any concerned framers were either outnumbered by those who were not or sensed efforts to regulate it were either impracticable or not a winnable issue. Despite their reputation, especially Madison’s, as “author” of the Constitution, historians have noted that most of the proposals Madison or Hamilton introduced or supported at the convention were defeated.
Or perhaps they assumed that unequal populations across political districts within states, to the extent that they were found, would exist within tolerable limits, or would not result in the vast and strikingly unequal representation of political interests to the same extent found when small states enjoyed the same voting power as the larger states. After all, the proponents of the Constitution were largely concentrating their reform efforts on addressing defects in the experience of government under the Articles of Confederation and the paralysis that resulted from allowing small states to block legislation necessary to advancing the interests of the nation.
Yet another possibility is that they believed that a separate and sufficient mechanism existed for addressing the districting problem. As an example of the latter, perhaps they believed that Article I, Section 4, allowing Congress to alter state electoral rules in federal elections, would suffice to remedy any particularly egregious or extreme case that might arise within a state. This possibility is purely speculative given that none of the three lengthy Federalist Papers (59–61) dedicated to defending the inclusion of this provision mention the composition, size, or population of political districts. In any case, this provision quickly fell into disuse, because it would be decades before Congress passed a law under this authority.
Regardless of the framers’ concern—or lack thereof—for the issue of inequities between political districts within states, state legislatures took full advantage of the maneuvering room granted them by the Constitution’s silence on this matter. In one of the most notorious instances of abuse of this power, the Massachusetts legislature passed a redistricting law in 1812 designed to minimize the political power of the Federalist Party in the next election by concentrating Federalist voters into a small number of districts while spreading Republican voters into a wider range of districts. The plan worked because the Republican party won twenty-nine seats compared to eleven for the Federalist Party, despite winning only 49% of the vote. The map, however, so conspicuously divided up the Boston region in an unnatural manner that critics likened the shape to a salamander. Governor Elbridge Gerry, a leading proponent of the plan, lent his name to political history when a political cartoonist dubbed the plan a “Gerry-mander: A new species of Monster.”
The type and form of political district found in the early republic was more diverse and less uniform than those that exist today. Not only were political districts of unequal population and dimension regularly employed, and the manipulation of those districts common to the extent of political tolerance, but the form and type of district were not nearly as uniform as is the case today. All districts organized for electing members to the House of Representatives today are what political scientists call “single-member plurality” districts. This means that each district elects a single member, and that member is elected by a plurality of the vote (winning more votes than any other candidate—also called “first past the post”). This was not, however, the case in the early years of the republic. A variety of district types co-existed, from multi-member to at-large districts. Many of these district types were designed to maximize or entrench partisan political power.
The Constitution neither prescribes nor prohibits particular types of districts or methods of electing representatives, aside from the requirement that voting qualifications be the same as those employed for the most “numerous branch of the state legislature.” It only requires a certain number of representatives for each state based upon relative population. Consequently, most of the original thirteen states used multi-member districts in the first congressional elections.
Between 20% and 44% of House members were elected from multi-member districts until the Twenty-Eighth Congress. This means that each district elected more than one representative. They did not use proportional representation systems, as is common to most modern parliamentary democracies, as such systems had not yet been developed. Instead, they gave some districts greater political representation relative to others. Later, some states used “at-large” voting, meaning that House members were elected in some states by a vote of the entire state, as United States senators are elected today.
As is true of many aspects of our political system and institutions, there was a gradual trend toward greater uniformity. In 1842, Congress passed the first of a series of laws, generally known as “Apportionment Acts,” which outlawed at-large, statewide House districts under its Article I, Section 4 authority. Although ostensibly aimed at giving political minorities within states more opportunity to elect members to Congress, it also had the effect of outlawing multi-member districts, not just at-large systems. There were serious doubts about the constitutionality of such laws, and at least a few states continued to use at-large systems in violation of the law.
It was not until an apportionment act in 1872 that Congress added that districts should not only be geographically contiguous and single-member, but also that they should contain “as nearly as practicable an equal number of inhabitants.” This requirement was reiterated in similar subsequent enactments, although it was not yet a constitutional principle.
In 1967, Congress passed another law prohibiting multi-member and at-large districts (in states with more than one representative) based on concerns that southern states might resort to at-large, statewide systems in response to the Voting Rights Act of 1965. Both multi-member and at-large districts could be used to dilute Black voting strength in southern states.
The key developments, however, were in the courts of the 1960s. In 1962, the Supreme Court ruled that political districting processes were “justiciable” and could be reviewed by courts in the case of Baker v. Carr. Two years later, the Supreme Court ruled that political districts should approximate equal population and announced the principle of “one person, one vote.” Unequal political districts undermined this principle. If districts could be devised of unequal size, then some people enjoy greater electoral influence, and their votes might count more than others. This legal principle has been serially re-affirmed and strengthened such that the Court has struck down districting laws drawing districts with deviations of less than 1% in population between them. Permitting suits challenging inequities and disparities in the design of political districts opened the door for the challenges to partisan political and racial gerrymandering processes.
II. THE GERRYMANDERING CASES
This Part of the Article will briefly review the major challenges to racial gerrymandering and partisan political gerrymandering reviewed by the Supreme Court and conclude with some comparative observations and analysis.
A. RACIAL GERRYMANDERING
The first notable racial gerrymandering case actually precedes Baker v. Carr. In Gomillion v. Lightfoot, the Supreme Court considered a challenge to a redistricting plan in Alabama that would have rendered the city of Tuskegee a twenty-eight-sided political district for no perceptible reason other than to disenfranchise the town’s Black population, which lived virtually exclusively in the districts outside the newly drawn city boundaries. The Court held that complaints alleging racial gerrymandering of municipal boundaries were cognizable under the Equal Protection Clause of the Fourteenth Amendment.
A decade later, in White v. Regester, the Supreme Court affirmed a lower court ruling that a redistricting plan adopted in Texas had elements designed to exclude Mexican-Americans from electing representatives in the state legislature through the employment of multi-member districts. The Court, however, rejected a similar claim involving multi-member districts in Indiana that had a disparate effect on Black voters in racially segregated urban neighborhoods. These cases, however, did not involve the drawing of districts so much as the type of district.
Racial gerrymandering claims received a significant boost in a series of cases considered by the Supreme Court after the 1990 census, beginning with the landmark case of Shaw v. Reno, in which the Court first recognized this claim as such in the drawing of district lines. After the 1990 census, North Carolina was awarded an additional congressional seat. The state legislature’s initial apportionment plan was rejected by the Department of Justice under the preclearance provision of the Voting Rights Act (“VRA”). A revised VRA-compliant plan created a second majority-Black district. Five white North Carolina residents sued, arguing that the redistricting plan violated the Equal Protection Clause of the Fourteenth Amendment. Specifically, the residents claimed that the state engaged in an “unconstitutional racial gerrymander.”
In an opinion authored by Justice Sandra Day O’Connor, the Court articulated several principles that helped lay the foundation for a clear rule against racial gerrymanders. First, the Court situated the case firmly within its racial classification jurisprudence, affirming that “laws that explicitly distinguish between individuals on racial grounds fall within the core of [the Equal Protection Clause’s] prohibition,” and that “[e]xpress racial classifications are immediately suspect.” Furthermore, the Court asserted that the harms of racial classification are as present in the electoral context as they are in other contexts that the Court had reviewed:
Racial classifications of any sort pose the risk of lasting harm to our society. They reinforce the belief, held by too many for too much of our history, that individuals should be judged by the color of their skin. Racial classifications with respect to voting carry particular dangers. Racial gerrymandering, even for remedial purposes, may balkanize us into competing racial factions; it threatens to carry us further from the goal of a political system in which race no longer matters—a goal that the Fourteenth and Fifteenth Amendments embody, and to which the Nation continues to aspire.
The Court acknowledged, however, that all redistricting is necessarily race-conscious, drawn with an awareness of racial demographics, just as the legislature is aware of many other demographics features when drawing legislative districts. Therefore, the Court held that not all race-conscious redistricting is unconstitutional. The Court did not, however, identify the line between permissible race-conscious political redistricting and impermissible racial gerrymandering. It concluded that the plan in question was so “bizarre on its face that it was ‘unexplainable on grounds other than race.’ ” Therefore, the Court held that the appellants stated a claim strong enough to survive a motion to dismiss and remanded the case for further determinations.
A similar set of facts led to another suit that the Supreme Court considered involving Georgia in the case of Miller v. Johnson. In announcing its decision in an opinion authored by Justice Anthony Kennedy, however, the Court both affirmed critical parts of Shaw and helped indicate where to draw the line between race-conscious political districting and impermissible racial gerrymandering.
The Court first specifically rejected the state of Georgia’s claim that “evidence of a legislature’s deliberate classification of voters on the basis of race cannot alone suffice to state a claim under Shaw.” Kennedy also observed that “the essence of the equal protection claim recognized in Shaw is that the state has used race as a basis for separating voters into districts.” Critically, however, the Court promulgated a “predominant factor” test to guide it’s application of the facts to the law in this context.
To establish a racial gerrymandering claim, a plaintiff must prove that “race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.” To make this showing, a plaintiff must establish that the legislature subordinated traditional race-neutral districting principles, including but not limited to compactness, contiguity, respect for political subdivisions or communities defined by actual shared interests, to racial considerations. Where these or other race-neutral considerations are the basis for redistricting legislation, and are not subordinated to race, a state can “defeat a claim that a district has been gerrymandered on racial lines.”
In several subsequent cases involving similar fact patterns and southern states, the Court further clarified that to bring racial gerrymandering claims, individuals must reside in the district that they claim is gerrymandered and that such claims should be brought on a district-by-district basis, not against an entire plan. This is consistent with the view that these cases are concerned with the dangers of racial classification. An individual residing outside of a district that it claims has been gerrymandered has not, in some sense, been “classified” on the basis of race by government as the Court has construed this concept. It should be noted that most of the contemporaneous cases heard by the Court at this time involved drawing districts to increase or preserve minority representation in Congress under the VRA.
There remain a number of ambiguities in how to apply the predominance test, some of which the Supreme Court has grappled with, but has not necessarily fully resolved. In particular, the Court has tried to clarify how the consideration of race is to be viewed in relation to other considerations before triggering strict scrutiny. In a case heard in the 2015 term, the Court rejected a lower court’s ruling that race did not “predominate” as a consideration in the redistricting plan because of “non-racial factors,” including the goal of creating districts of equal population. The Court clarified that the equal population factor is not a factor to be considered in the ordinary course of redistricting, but a constitutional mandate. While that may seem like an easy case, the Court considered a harder question in the 2016 term.
The state of Virginia defended a Republican-led redistricting plan against a racial gerrymandering claim on the grounds that the “predominance test” should only be applied if the use of race is in conflict with “traditional districting principles,” as it had presumed in cases such as Shaw. The Court rejected this position in a seven-to-one decision presented in an opinion by Justice Kennedy. He concluded that racial gerrymanders can exist or arise even under plans that otherwise conform to traditional factors such as compactness and contiguity. As he explained, “The Equal Protection Clause does not prohibit misshapen districts. It prohibits unjustified racial classifications.”
Critically, the Court emphasized that an unconstitutional racial gerrymander can arise or exist even if an identical plan could have been adopted without consideration or use of race. Justice Kennedy explained that “[t]he racial predominance inquiry concerns the actual considerations that provided the essential basis for the lines drawn, not post hoc justifications the legislature in theory could have used but in reality did not.” Thus the predominance test is a factual inquiry into considerations used as part of the actual districting process. Holding otherwise would provide a state with constitutional cover for unconstitutional behavior. As the Court observed, “By deploying [non-racial] factors in various combinations and permutations, a State could construct a plethora of potential maps that look consistent with traditional, race-neutral principles. But if race for its own sake is the overriding reason for choosing one map over others, race still may predominate.”
B. PARTISAN GERRYMANDERING
Since the early 1960s, the Supreme Court has become more solicitous of racial gerrymandering claims even has it has become more hostile to partisan political gerrymandering claims. This Section will now review the latter line of cases.
The Supreme Court squarely confronted a partisan gerrymandering claim in the 1986 term in the case of Davis v. Bandemer. In that case, the Court reviewed a redistricting plan proposed by the Republican-controlled Indiana state legislature. This plan yielded an immediate partisan advantage in which 43 out of the 100 seats in the state House of Representatives were filled by Democratic candidates even though 51.9% of the statewide votes went to Democratic candidates. Upon review, a deeply divided Court produced a fragmented set of opinions in which most justices agreed on the result—a holding that the plan was not unconstitutional—but disagreed on the rationale and basis for that judgment.
Six of the Justices, and therefore the Court, endorsed the legal principle that partisan or political gerrymanders are justiciable, while three Justices, Justice O’Connor, Justice Burger, and Chief Justice Rehnquist, preferred to rule that such claims are not. Even among the Justices in the majority and plurality, however, there was a lack of consensus on the grounds for doing so and on the standards that should be adopted to evaluate such claims. The debate between the Justices prefigures most of the issues that have been subsequently debated in this context.
The plurality of Justices White, Marshall, Brennan and Blackmun begin with the principle that in order to establish a partisan gerrymandering claim, plaintiffs must “prove both intentional discrimination against an identifiable political group and an actual discriminatory effect on that group.” They caution, however, that the mere fact that “a particular apportionment scheme makes it more difficult for a particular group in a particular district to elect the representatives of its choice does not render that scheme constitutionally infirm.” In addition, they reject the notion that mere disproportionality in representation is a sufficiently adverse effect to establish a constitutional violation. The plurality advises that reviewing courts examine both the district individually as well as the state’s overall districting plan holistically. The plurality also asserts that a constitutionally infirm redistricting plan can occur either when a minority manipulates boundaries to consistently thwart the will of the majority, or when a majority uses its power to shut a minority out of the political process or a meaningful chance to influence it.
Ultimately, the plurality held that “unconstitutional discrimination occurs only when the electoral system is arranged in a manner that will consistently degrade a voter’s or a group of voters’ influence on the political process as a whole.” In applying this standard, the plurality emphasized that claimants would need to examine more than a single election to conclude that an impermissible partisan gerrymander exists, and that they should be able to distinguish between a meaningful structural disadvantage established by districting compared to mere lack of success in persuading voters. The plurality felt that the district court’s conclusions were less persuasive than the state’s defense on this point and voted to reverse the lower court.
Justice Powell and Justice Stevens found the plurality’s approach too restrictive and would have affirmed the lower court. In particular, Justice Powell wrote in favor of a multi-factor analysis:
The most important [factor to consider is] the shapes of voting districts and adherence to established political subdivision boundaries. Other relevant considerations include the nature of the legislative procedures by which the apportionment law was adopted and legislative history reflecting contemporaneous legislative goals. To make out a case of unconstitutional partisan gerrymandering, the plaintiff should be required to offer proof concerning these factors, which bear directly on the fairness of a redistricting plan, as well as evidence concerning population disparities and statistics tending to show vote dilution. No one factor should be dispositive.
Justice O’Connor and Justice Burger argued strenuously that the majority’s holding would prove unworkable in practice, as illustrated by the inability of the six Justices in favor of holding partisan gerrymanders as justiciable to get behind a single standard and therefore was a “political question” as set out in the framework for evaluating such questions in Baker v. Carr. In addition to the lack of a clearly defined and easily applicable standard for adjudicating such claims, they had prudential concerns. Justice O’Connor argued that without a clear standard, opening federal courts to claims of partisan gerrymandering would lead to “pervasive and unwarranted judicial superintendence of the legislative task of apportionment.” She wondered if there was any logical stopping point short of “roughly proportional representation for every cohesive political group.”
Despite its inviting holding, Bandemer established a heavier burden for plaintiffs to overcome so as to make out a discriminatory political gerrymandering claim than its authors probably imagined. Over the next eighteen years, no federal courts, at any level, ruled that a redistricting plan was an unconstitutional political gerrymander. This was the situation when the Court heard the case of Vieth v. Jubelirer in 2004.
In Vieth, the Court considered a suit brought by members of the Democratic Party, claiming that the state of Pennsylvania’s partisan redistricting plan violated the Constitution. Five Justices agreed that the plan did not violate the Constitution, but four of those Justices took the position of Justice O’Connor, Justice Burger, and Chief Justice Rehnquist in Bandemer, arguing that political gerrymanders should be non-justiciable and voting to overturn Bandemer. The fifth Justice, Justice Kennedy, however, agreed that current standards were unworkable, but he preferred to leave the door open to discovering one. The four dissenting Justices agreed with Bandemer’s holding but disagreed on the specifics of how to operationalize that principle.
The Court subsequently heard a case that had been held over until Vieth was decided, League of United Latin American Citizens v. Perry. In Perry, the petitioners challenged the 2003 redistricting plan enacted by the Republican-controlled Texas legislature. Drawn merely a year after the 2002 midterm elections, the 2003 plan led to an election result where twenty-one Republicans and eleven Democrats were elected in congressional elections the following year, a far more lopsided result than the aggregate vote count would suggest. The petitioners argued that the mid-decennial nature of the redistricting plan revealed the legislature’s sole motivation to gain partisan advantage, which should be sufficient to trigger heightened scrutiny. In the majority opinion, Justice Kennedy entertained the petitioners’ proposed “sole-intent” test, but ultimately found it “not convincing” because some contested district lines were drawn based on more local interests and a number of line-drawing requests by Democratic legislators were honored.
Interestingly, Justice Stevens and Justice Breyer, in their dissent, proposed a different test wherein a plaintiff must prove partisan aims by showing that: (1) the legislature “subordinated neutral districting principles to political considerations” and (2) their predominant motive was to “maximize one party’s power,” along with a showing of discriminatory intent by establishing that (1) the plaintiff’s candidate of choice won election under the old plan; (2) the plaintiff’s residence is now in a district that is a safe seat for the opposite party; and (3) the plaintiff’s new district is less compact than the old district. However, this complex alternative test was clearly not entertained by the majority.
The next major partisan gerrymandering case was heralded when the Court agreed to review a redistricting plan arising from Wisconsin in Gill v. Whitford, based on the fact that a new technical standard had been developed and proposed: the efficiency gap. This was a formula devised by social scientists to provide metrics that could gauge the specific disadvantage created by certain redistricting schemes. The authors of this formula claimed it provided clear and workable standards for operationalizing them. The result was a letdown when the Court dismissed the case for lack of standing. The Court also breezily dismissed another case the same year in a per curiam opinion regarding Maryland’s Democratic-favored political gerrymandering. This case was reconsidered, and then dismissed, in the 2019 term.
Justice Kennedy’s departure from the Court led to a more decisive result in this line of cases. In Rucho v. Common Cause, a majority of the Court held for the first time that partisan gerrymanders are a political question and non-justiciable largely for the reasons developed in the dissent in Bandemer and the concurrences in Vieth. This would seem to be, thus far, the end of the line for claims of partisan gerrymandering.
C. POINTS OF CONVERGENCE AND DIVERGENCE
The juxtaposition of the racial gerrymandering and partisan political gerrymandering cases is itself revealing. The partisan and racial gerrymandering cases germinate from the same seed and the same soil. They both arise out of a political context in which democratic representation is being distorted or manipulated for specific advantage, and in a legal context in which the Constitution provides scant direct guidance (despite the provision of a political mechanism by which Congress, via Article 1, Section 4, can alter state electoral rules in federal elections). The germ for both claims is the Court’s greater solicitude toward challenges to various electoral schemes in the early 1960s. And yet the Court has evolved vastly different frameworks and conclusions for regulating these forms of political districting activity for reasons that are not entirely convincing or coherent.
In theory, the different levels of scrutiny and accord given to race versus other classifications under the Equal Protection Clause could explain the differences in treatment, but the proponents of treating political gerrymanders as non-justiciable decline to ground their reasoning on this basis. After all, if this were the critical explanatory factor, then partisan gerrymandering claims would be justiciable, just at a much lower level of judicial review.
Instead, the main contention of the non-justiciable position emphasizes the lack of workable standards and the constitutional structure and history of gerrymandering as a political practice. Recall, for example, that Justice White and the plurality in Bandemer emphasized whether a districting scheme created a severe structural disadvantage in access to the political process, whereas Justice Powell (joined by Justice Stevens) would have applied a multi-factor, holistic approach. Justice O’Connor and the Justices who joined her opinion argued that the Baker factors for evaluating whether an issue is a political question squarely fit, and that there was no logical stopping point short of proportional representation once such claims were entertained.
The most obvious rejoinder to the claim that political gerrymandering suits cannot be grounded onto a workable standard is the fact that workable standards have already been developed and adopted in the racial gerrymandering context. Logically, if the standard is workable in one context, it should be workable in another, absent some factor that would render it otherwise. Indeed, this is a prominent theme of the Justices who support judicial review of extreme partisan gerrymanders.
In Bandemer, all six Justices in the majority signed onto a sharp critique of Justice O’Connor’s opinion (joined by Burger and Rehnquist) arguing that she failed to “point out how the standards that we set forth here for adjudicating this political gerrymandering claim are less manageable than the standards that have been developed for racial gerrymandering claims.” This was also a central point of contention in Vieth. The plaintiffs in that case modeled their claim on the standard adopted in Miller, that the partisan objective was a “predominant factor” in the redistricting process. As the plurality in Vieth forthrightly noted, “Appellants contend that their intent test must be discernible and manageable because it has been borrowed from our racial gerrymandering cases.” Yet, it disagreed, for reasons that Justice Stevens upbraided in his dissent:
Especially perplexing is the plurality’s ipse dixit distinction of our racial gerrymandering cases. Notably, the plurality does not argue that the judicially manageable standards that have been used to adjudicate racial gerrymandering claims would not be equally manageable in political gerrymandering cases. Instead, its distinction of those cases rests on its view that race as a districting criterion is “much more rarely encountered” than partisanship, and that determining whether race—“a rare and constitutionally suspect motive”—dominated a districting decision “is quite different from determining whether [such a decision] is so substantially affected by the excess of an ordinary and lawful motive as to [be] invali[d].” But those considerations are wholly irrelevant to the issue of justiciability.
Moreover, the stronger the argument advanced by the plurality in Vieth and the majority in Rucho against porting the standards used in the racial gerrymandering cases to the political gerrymandering context, the more they reveal the weaknesses of the doctrine they have developed in the racial gerrymandering context. In Vieth, for instance, Justice Scalia writes on behalf of the plurality that the “predominant motivation” test is “vague . . . when used to evaluate single districts, [but] it all but evaporates when applied statewide.” Any vagueness, however, is built into the notion that courts can discern the predominance of any particular factor in a complex legislative process. Qualitative considerations in legislative processes are not like mechanical inputs into the production of widgets. They inherently resist the quantification necessary to determine “predominance.”
The argument Justice Scalia develops in the guise of a rhetorical question is especially revealing:
And how is the statewide “outweighing” to be determined? If three-fifths of the map’s districts forgo the pursuit of partisan ends in favor of strictly observing political-subdivision lines, and only two-fifths ignore those lines to disadvantage the plaintiffs, is the observance of political subdivisions the “predominant” goal between those two? We are sure appellants do not think so.
The exact same argument could be developed at the scale of a single political district in the racial gerrymandering context, using neighborhoods, census tracts, or other census designated geographies as the smaller scale analogue subdivision. The “predominant factor” test seeks to quantify the inherently qualitative. Any deficiency in the predominant factor test in the partisan gerrymandering context is likely equally applicable to the racial gerrymandering context.
Finally, Justice Scalia tries to draw another distinction to justify this difference in treatment:
Determining whether the shape of a particular district is so substantially affected by the presence of a rare and constitutionally suspect motive as to invalidate it is quite different from determining whether it is so substantially affected by the excess of an ordinary and lawful motive as to invalidate it.
But this argument utterly ignores the fact that Shaw and its progeny never characterized the pursuit of partisan political advantage as an ordinary (“traditional”) districting principle and also readily acknowledged that virtually all political districting is race-conscious, drawn with the awareness of racial demographics. To that extent, it is impossible to declare that consideration of race is any less “ordinary” than consideration of partisan advantage in the redistricting process. Moreover, in both cases, the argument is that it is only the predominance of this consideration that renders it illegal, not its mere presence. To that extent, Justice Scalia’s comment assumes what it concludes when it declares one legal and the other unlawful.
Chief Justice Roberts makes a similar leap in logic in Rucho: “A permissible intent—securing partisan advantage—does not become constitutionally impermissible, like racial discrimination, when that permissible intent predominates.” If true, it is only because he and his colleagues make it so. If they decided that securing partisan advantage was unconstitutional when it predominates the redistricting process, then it would be so. Both Justice Scalia and Chief Justice Roberts assume that pursuit of partisan advantage is permissible to justify their conclusion that the case lines are different. Further, if the main reason that partisan gerrymandering is non-justiciable is the lack of workable standards, then the constitutionality of such processes is based upon their judgment in that regard rather than an evaluation of valiant, but ineffective, efforts.
As unconvincing as these arguments are for why the predominance test cannot be reasonably applied to the partisan context, there is an even stronger argument for why partisan and racial gerrymanders cannot be treated in such categorically distinct ways, and it is a point that the dissenters have not developed: the entanglement of partisanship and race in redistricting processes.
Against a backdrop of even moderate racial residential segregation, racial political polarization in voting patterns combined with the race-conscious districting processes mean that partisan gerrymandering cannot be clearly distinguished from racial gerrymandering. Even if mapmakers completely ignore racial demographics in generating districting options, a redistricting process motivated by the pursuit of partisan advantage is likely to produce maps strikingly similar to those that might occur if the state had “used race as a basis for separating voters into districts.” The next Part of this Article confronts and unpacks this problem.
III. THE ENTANGLEMENT PROBLEM
The entanglement of race and partisanship in political districting processes is rooted in a few critical phenomena. To understand the entanglement problem, we must first examine the concept of polarization, especially as it is applied to politics.
Polarization is itself an elusive notion, and it is not always clear what it means. Sometimes polarization is invoked to describe the salience of political conflict. This simple characterization is misleading. Although there may be an underlying relationship, partisanship and political polarization are not coterminous.
The existence of closely contested elections versus landslide elections cannot tell you whether the electorate is polarized or not. Landslides might occur in deeply polarized contexts but in which one group dominates another, as in the Jim Crow South, or low-polarization contexts in which a broadly popular, mainstream candidate triumphs. Similarly, closely contested elections can occur when both candidates are popular or both are deeply unpopular. Aggregate returns are not indicative.
More often, polarization is used to refer to intensity of disagreement or political conflict. This conceptualization, however, also fails to capture the essence of polarization. Extreme (corrosive or toxic) partisanship can exist in a state of either high or low political polarization. It is the distance or scope of disagreement, not necessarily the intensity of feeling or contestation, that is key to understanding polarization. This is perhaps easier to understand in an economic context.
Suppose that most people in a certain society are middle-income, with far fewer extremely rich and extremely poor individuals. Such an income distribution would be depicted graphically in the form of what statisticians call a “normal distribution” (or a “Gaussian” distribution) with a bulge in the middle and smaller downward sloping tails toward the edges of the income distribution. In contrast, suppose that most people in another society are either extremely poor or very wealthy, with fewer people as middle-income. This is what statisticians call a bimodal distribution, with bulges towards the ends of the distribution. The mean or median results could be identical in both situations. See Figure 1.
FIGURE 1. A Normal vs. a Bimodal Distribution
Polarization refers to a process by which a distribution resembling a normal distribution evolves into something more closely resembling a bimodal distribution. Thus, if the income distribution hollows out this way, economists may describe this as “economic polarization.” The same is true of politics.
As political scientists and surveyors have demonstrated, American political attitudes have indeed become more polarized in recent decades. The Pew Research Center, which conducts regular surveys on political attitudes, found that in both 1994 and 2004, more Americans expressed mixed or centrist political opinions than either conservative or liberal views. By 2014, however, this had reversed, a trend that has continued to the present.
One particularly important dynamic contributing to political polarization is that Republicans have become more conservative and Democrats have become more liberal. In 1994, only 64% of Republicans were more conservative than the median Democrat, and only 70% of Democrats were more liberal than the median Republican. By 2014, those figures had risen to 92 and 94%, respectively.
Another form of polarization is racial political polarization. This tends to refer to distinctive partisan preferences between racial groups. Table 1 below indicates the significant racial gaps in support for partisan candidates in presidential elections since 2000.
TABLE 1. Republican Support by race in Presidential Elections, 2000–2020
Although there is non-trivial variation between electoral cycles, this table shows consistently high political polarization between racial groups in the popular vote for President (and by implication for political party) over two decades. In presidential elections since 2000, Black support for the Republican candidate narrowly ranges from 4% to 12%. White support ranges from 55% to 59%. Latino support ranges from 26% to 44% (in 2004). And Asian support ranges from 27% to 43%. The white-Black voting gap ranges from 46 to 53 points. Although political polarization by race seems to reach a peak in 2012, the figures as of 2020 are just as large, if not larger, as they were in 2000. In short, racial political polarization has worsened over time, but it has been consistently high throughout the twenty-first century.
Similar tables could be generated for congressional races, midterm elections, governor’s races, and the like. Political scientists have observed that racial political polarization in the United States has been pronounced at least since 1965. This connection between race and partisan affiliation is described as the “conjoined polarization.” Although this phenomenon might be troubling in its own right, it is not automatically a problem in the context of political redistricting processes.
If racial groups were evenly distributed across space, then political polarization by race would not factor into political redistricting processes. Gender political polarization offers a contrasting illustration. Gender political polarization has increased significantly in recent years too, and a table similar to Table 1 could be presented illustrating it. But the absence of gender political segregation means that gender is not a meaningful basis upon which to conduct political redistricting processes.
Residential segregation is the critical element that enables gerrymandering. This is true generally, but in the context of racial gerrymandering, racial residential segregation is the critical factor.
When racial groups are highly concentrated into certain neighborhoods or communities, and there is a moderate to high degree of racial political polarization, then racial gerrymandering is relatively easy. On the other hand, even if the political community is racially polarized, if racial groups are highly integrated across space, then racial gerrymandering would prove extremely difficult to impossible. If either condition is absent, it is not possible to either use race to draw political districts for partisan advantage nor to draw districts for partisan advantage that segregate people into different political districts on the basis of race. Only when both conditions hold does the entanglement problem arise. Table 2, below, indicates the relationship.
TABLE 2. Conditions for Entanglement
The entanglement of race and politics in political redistricting processes, then, is not simply a function of the conjoined polarization of racial political polarization and partisan political polarization. It depends on the existence of racial residential segregation as well.
Unfortunately, racial residential segregation is persistently moderate to high across much of the United States. Using traditional measures of segregation such as the dissimilarity index, national Black-white dissimilarity in 2020 was 55.2, meaning that 55% of Black Americans would have to move into a predominantly white neighborhood to have no racial residential segregation between Black and white Americans. This is in a range social scientists regard as “moderately high.” Asian-white dissimilarity scores are 40.0, and Hispanic-white dissimilarity scores are 45.3, in the moderate range. Even though most of these numbers have improved in recent decades, they demonstrate the persistence of racial residential segregation. In many cities, these figures are much higher.
Other popular measures of segregation are the Isolation/Exposure Indices, which describe neighborhood composition of the typical (median) or average person by race. The “typical” Black and white Americans reside in vastly different neighborhood milieus. As of 2020, the average white resident of a metropolitan area resides in a neighborhood that is 69% white, 9% Black, 12% Hispanic, and 6% Asian. In contrast, a typical Black resident lives in a neighborhood that is 41% Black, 34% white, 17% Hispanic, and 6% Asian. Not only are these demographically different worlds, these figures mean that Black “exposure” to white people is 34, roughly the same level it was in 1940.
Using a different measure of segregation, the Divergence Index, which can account for multiple racial groups simultaneously and provide a single holistic score for every city or metropolitan region, racial residential segregation appears stubbornly persistent. The Divergence Index compares demographic proportions of smaller geographies to larger geographies and then sums those population-weighted differences to yield a holistic score. Over 50% of cities and metro areas have a higher (more segregated) Divergence score as of 2020 than in 1990.
Given the central role of residential segregation in facilitating or impeding the manipulation of district boundaries for political advantage, it is strange that prior legal scholarship has under-emphasized or ignored this factor in analyzing gerrymandering cases. Perhaps that is because it is assumed that segregation is a neutral or binary background factor that either exists or does not, and does not actually affect the level of partisan political polarization in a region. If so, there are several findings that challenge this assumption.
A study conducted by the political scientist Jessica Trounstine identified a direct relationship between racial residential segregation and partisan political polarization. She finds that the relationship between segregation and racial political polarization is statistically powerful: a city in the 10th percentile of segregation has a 35% point divide in racial support for a political candidate, compared to a 63% point divide at the 90th percentile. In other words, the more racial residential segregation, the more racial political polarization.
Could this difference be explained by the fact that white people vote more conservatively in certain states or regions than others? She tests for this and finds that the relationship between segregation and polarization is unaffected by the conservatism of the local white population. In fact, she found that “cities with more conservative white populations have smaller racial divides.”
In short, racial residential segregation is not a background condition that exists in a binary state, either existing or not, to undergird the entanglement problem. Rather, it is a condition that interacts dynamically with racial political polarization. A higher level of racial residential segregation seems to coincide with higher racial political polarization. Similarly, racial residential segregation interacts dynamically with partisan political segregation and (by inference) partisan political polarization, as will be shown below.
To appreciate the full relationship between race and politics, it is important first to emphasize that the critical role of residential segregation in relationship to gerrymandering is general, not specific to racial segregation and racial gerrymandering. Partisan political polarization, by itself, is neither a necessary nor a sufficient condition to enable partisan gerrymandering. Political segregation, however, is necessary. If people of different political preferences or partisan affiliation are evenly distributed across space, regardless of how polarized they may be, it is extremely difficult to draw political districts for partisan advantage. On the other hand, high levels of partisan residential segregation (say, Republicans living in one community and Democrats in another) make it much easier to draw districts for partisan political advantage. For simplicity of illustration, compare hypothetical voting precincts of equal size with voting totals depicted in Tables 3 and 4 below.
TABLE 3. Hypothetical #1 Voting Precinct Electoral Results
Although the vote total in Table 3 is relatively close (only a two-point margin between the winner and loser), the precincts are extremely divergent from the aggregate vote total, suggesting a high degree of political segregation between precincts. In contrast, consider this a different election with the same aggregate result in Table 4 below.
TABLE 4. Hypothetical #2 Voting Precinct Electoral Results
In this case, the winner’s vote margin is the same as in the first case, but the precincts only marginally diverge from the aggregate result. The largest gap for support for Candidate A and the aggregate result is just 4 points, in Precinct 1, compared to a 37–40-point gap between each precinct and the ultimate results in the first case. The first case suggests a high level of political segregation across precincts. If these results were stable over time, and not specific to a particular candidate but consistent across candidates and issues, then we could reasonably describe that region as having a high degree of political segregation. It should be much easier for mapmakers to draw political districts for partisan advantage in cases like the first rather than the second (although a larger number of precincts or sub-precinct level voting data would be needed).
Unfortunately, the United States has widespread political segregation as well, which maps vividly illustrate. Democrats and Republicans are not just more divergent in their views, they are increasingly residing in different communities. Using this type of data or other geographically disaggregated vote tabulations, it is possible to calculate the degree of political segregation that exists in different regions. Using the formula for the Divergence Index (described above), I have calculated the relative degree of political segregation for different major metropolitan regions of the United States.
Using 2020 presidential election precinct tabulation results, out of the 314 largest metropolitan areas in the United States (those with a population of 200,000 or more), Table 5 below lists the top 20 most politically segregated regions of the United States along with their divergence score.
TABLE 5. The 20 Most Politically Segregated Metropolitan Areas in the United States (2020 Presidential Election)
The dynamic of political segregation can also be depicted visually, providing a more intuitive and easily comprehensible approach. Compare maps of Jackson, Mississippi (the most politically segregated metro), with Carson City, Nevada (ranked 273 out of 314), which has one of the lower political divergence scores), in Figure 2, below. The contrast is vivid.
FIGURE 2. Political Segregation in Jackson, Mississippi, and Carson City, Nevada
As the maps displayed in Figure 2 illustrate, the metropolitan area of Jackson ranks first in political divergence, indicating the presence of ideological extremes where precincts overwhelmingly voted in favor of one party while neighboring precincts voted in favor of the other party. In contrast, all of the precincts in Carson City approximate the regional average, suggesting a much lower level of political segregation. The high level of political segregation in Jackson makes it especially vulnerable to partisan gerrymandering. This is partly what makes districting such a dilemma in the United States: the high degree of political geographic or residential segregation.
One of the striking features of the list presented in Table 5, above, is that the vast majority of the most politically segregated regions are in the South (17 of the top 20). Given that region’s racial history, it raises the question of whether there is a relationship between political residential segregation and racial residential segregation in the United States. Figure 3, below, confirms the relationship between the two, as shown in a scatterplot.
FIGURE 3. Racial Segregation and Political Segregation in the United States, 2020 (314 Metros)
Figure 3 depicts the 314 largest metropolitan areas in the United States. The horizontal axis indicates the divergence index score (calculated as described above) for political segregation and the vertical axis indicates the relative percentile rank for racial residential segregation using the same formula but with data from the 2020 decennial census. The scatterplot indicates a clear positive relationship, with a 0.50 correlation between the two variables.
Not only does this analysis suggest a relationship between the two phenomena, but it also means that where racial segregation is higher, partisan gerrymandering should be easier. Conversely, where partisan segregation is higher, racial gerrymandering should be easier. In short, they go hand in hand, but in a dynamic relation. The problem lies in their coincidence. This is the crux of the entanglement problem, not the simple fact of conjoined polarization between race and politics.
Whether the context is partisan gerrymandering or racial gerrymandering, the active ingredient is segregation, not polarization. Partisan gerrymandering only requires some degree of political segregation. Racial gerrymandering, however, requires both racial political polarization and racial residential segregation. Prior legal scholarship treats segregation as the backdrop condition and conjoined polarization as the central or proximate problem when the truth is the opposite.
In practical terms, the persistence of racial political polarization means that the strong correlation between political segregation and racial residential segregation easily facilitates both racial and partisan gerrymandering in ways that are essentially indistinguishable. State officials can draw maps at a hyper-granular level that may relocate a small number of people from one district to another, with full awareness of their race and the fact of conjoined racial identity with partisan preference, and nonetheless claim that the decision was based on partisan motivations. In other words, racial residential segregation enables partisan gerrymandering that will result in the political segregation of people between districts on the basis of race. Even if the map-makers were to scrub all data regarding race from their software, a map drawn on partisan or other non-racial characteristics could appear objectively indistinguishable from maps drawn in cases like Shaw.
This problem is not speculative or theoretical. The Supreme Court has already heard cases touching on this problem. In oral argument in Wittman v. Personhuballah, Chief Justice Roberts asked “if race and partisanship are co-extensive, which one predominates?” In that case, several Republican members of Congress appealed a lower court’s decision to strike down a redistricting plan it found to be based on race. This question led to a brief dialogue among the Justices and the lawyer for the original plaintiffs regarding this issue, but the case was ultimately dismissed for lack of standing among the members of Congress to bring their appeal.
In another case decided that same term, Cooper v. Harris, the Court acknowledged that many of the Shaw considerations (compactness, for example) used to assess whether race predominated become less probative when the state raises the defense of partisanship. As it explained, “political and racial reasons are capable of yielding similar oddities in a district’s boundaries. That is because, of course, ‘racial identification is highly correlated with political affiliation.’ ” In that case, the Court rejected the claim that partisan goals are a complete defense to racial gerrymandering claims. As it explained, the predominance “inquiry is satisfied when legislators have ‘place[d] a significant number of voters within or without’ a district predominantly because of their race, regardless of their ultimate objective in taking that step.”
But this guidance merely sidesteps the more difficult question of how to determine whether a legislature’s districting decisions were “because of race” in such cases, and assumes that racial motives can be disentangled from partisan ones (either as a means or an end). Indeed, the entire concept of “predominance” assumes that the factors considered by entities charged with redistricting, and which are being reviewed by courts, are separate and independent elements. Because of the difficulties introduced by entanglement, lower courts may be hesitant to rule that race “predominates” when race and partisanship are highly entangled or when the state supplies reasons to believe that any apparent use of race was merely partisanship.
Again, this is not a speculative concern. The NAACP Legal Defense Fund and the Lawyers Committee for Civil Rights brought a suit against Georgia in 2017 alleging that a mid-decade redrawing of political districts was both a racial and political gerrymander. The district court acknowledged that ascertaining the existence of a racial gerrymander was “particularly hard to do when the State offers a defense rooted in partisan gerrymandering, as it did here. We did not move these voters because they are black, the State tells us. We moved them because they were Democrats.” The court ultimately sided with the state for that reason.
Experience demonstrates that this epistemological problem created by the entanglement of racial and partisan gerrymanders already exists and may be intensifying. The provision of block level census data following the 1990 census meant that state legislatures could draw more fine grain political districts based on race than was ever possible before using computer programs. Indeed, the Court confronted this fact in Bush v. Vera, in which the Court noted that the computer program “REDAPPL enabled districters to make more intricate refinements on the basis of race than on the basis of other demographic information.”
This technology has only improved in the intervening decades. It is now possible to generate thousands of potential maps at a keystroke with computer processing and programs that draw from large voter or census files. Large data files can be cross-referenced not only to generate demographic profiles, but also psychographic information, such as predicting propensity to vote, donate money, or even respond to certain campaign communications.
Upon the death of a North Carolina Republican strategist involved in redistricting efforts, his daughter made public his personal computer files against the wishes of the party and company he worked for. The trove contained thousands of documents detailing the various ways that he sought to generate political advantages for his clients, describing gerrymandering as legal vote stealing.
In Vera, however, Justice O’Connor stated that “[i]f district lines merely correlate with race because they are drawn on the basis of political affiliation, which correlates with race, there is no racial classification to justify.” Although likely dicta, this approach nonetheless does not resolve the issue because, if there is a perfect identity or correspondence of race and partisanship, how is a court supposed to judge whether the lines were drawn “because of political affiliation” or “because of race”? If a state legislature is trying to conceal its racial intentions, it would simply develop a record of partisan purposes and other traditional districting considerations. The entanglement of race and partisanship would then allow state legislatures to subvert the Constitution. The vastly divergent standards for both forms of gerrymandering makes it even more difficult to regulate this problem. There are, however, solutions.
This Article argues that the wildly divergent standards established by the Supreme Court governing partisan political gerrymandering and racial gerrymandering claims are untenable. Justices across the ideological spectrum agree that the use of race in drawing political districts may run afoul of the Constitution, but the Court’s extremely divergent rules regulating gerrymandering cases make it extremely difficult, if not impossible, to know whether race has been used or not. In too many contexts, partisan political gerrymanders will entail conduct that violates the principles and standards established by the foundational racial gerrymandering cases, or run so closely up to them that any attempt to fully disentangle partisan ends from racial ones is likely to be hopelessly futile or end up subverting the Court’s racial gerrymandering jurisprudence by allowing racial gerrymanders to escape under the cover of partisanship.
Beyond the categorical differences grounded in the Court’s precedent regulating the use of race at a higher level of judicial review than most other classifications, the Court has tried to manage the entanglement problem in two steps: first, by recognizing that all political districting processes are inherently race-conscious, as they are conscious of other demographic and community characteristics. Thus, the Court has distinguished between awareness of race and actions or policy decisions that use race in the sorting of people into different districts. The Court has held that it is the latter that violates the Constitution, not the former.
Second, by requiring that racial considerations actually “predominate” other factors, the Court has drawn a line between impermissible consideration of race and other ordinary or “traditional” considerations such as compactness, contiguity, community boundaries, and so forth. It is notable that in the listing of such ordinary considerations that partisan political advantage is never mentioned or listed. And this is presumably not simply because a bare-faced partisan consideration is unseemly, but because it was not considered by the Court (at least in those decisions) as a regular or ordinary consideration in the districting processes. Nonetheless, the Court’s jurisprudence prompts an objective, factual inquiry into whether race was in fact used or not.
Unfortunately, several demographic factors are converging in a way that makes it much more difficult—if not impossible—to delineate between race and partisanship as a consideration. First, political polarization appears to have increased in recent years. Second, political segregation is highly visible and becoming more pronounced as well. Third, racial political polarization has increased markedly in recent decades. The interaction of these three factors, on top of a fourth—the persistence of racial residential segregation—means that partisanship and race are highly correlated in a way that makes partisan districting largely and increasingly coterminous with racial districting. In simplified terms, “conjoined polarization” and racial residential segregation interact to create the conditions that entangle race and partisanship in political redistricting processes.
That this is a practical problem is evidenced by the fact that many cases heard by the Supreme Court in recent years feature both claims, whereas that would have been anomalous even a few decades ago. With the Court shutting the door on partisan gerrymandering claims, it seems increasingly likely that suits designed to curb the excesses of partisan gerrymandering will be brought under the color of racial gerrymandering. Thus, the Court will eventually need to squarely confront and address this problem.
One possible solution is to carve out an exemption for racial gerrymanders that appear to be largely or entirely based on partisan motives, as Justice O’Connor intimated in Vera, but to extend the exemption to cases in which the objective use of race clearly “predominates.” Under this approach, mapmakers would be permitted to use race in districting processes as long as their purpose was purely partisan. Under this approach, partisanship would be a complete defense to racial gerrymandering claims.
This approach would solve the smaller problem of the difficulties lower courts confront disentangling race and partisan motivates, but it would leave intact the larger problem of allowing racial gerrymanders to persist under the cover of partisanship. To that extent, this is only a partial solution or non-solution, because it would potentially obliterate the racial gerrymandering claim and undermine the constitutional prohibition against the use of race in policymaking in many ordinary cases. Under a rule such as this, any egregiously racially segregated political district could be justified on the basis of mere partisanship. The obvious exception would be cases in which state legislatures are seeking to create “majority-minority” districts under the VRA, because those could not plausibly be defended on partisan-only grounds. This approach would clearly violate the Court’s prevailing anti-classification jurisprudence and the principles and spirit of Shaw and its progeny’s rules against permitting state legislatures to sort people into one district or another on the basis of race.
A second possible solution would be to create an exemption in the opposite direction: a supplement could be drawn to the rule that partisan gerrymanders are non-justiciable if partisan purposes overlay or are essentially indistinguishable from racial considerations. In such a case, partisan gerrymanders could be swept into the racial gerrymandering line, even though it would be difficult or implausible to assert that race “predominates.” This is essentially the tack the Court appeared to take in Cooper v. Harris in its 2016 term, prior to the Court’s more recent decision that partisan gerrymanders are non-justiciable in Rucho.
The NAACP Legal Defense Fund’s president and director-counsel, Janai Nelson, has suggested an approach along these lines. The approach would shield against rearguard incursions into racial gerrymandering from the partisan direction. This option is most consistent with the Court’s anti-classification jurisprudence: any sorting or segregation of people into different political districts based on race violates the Constitution, even if it cannot be said that race “predominates.” This approach drives the presumptive rule against the use of racial classifications to its logical endpoint.
Although requiring a tweak to the Court’s racially gerrymandering jurisprudence, the Court can easily justify the use of a threshold test less than “predominance” when two factors are so tightly entangled that “predominance” becomes nonsensical. In such cases it is either impossible for race to “predominate” because partisan considerations are co-extensive with race or, even if they are greater, it is because their correspondence renders the possibility of calculating predominance by disaggregating and weighing the relative influence of each factor or consideration impossible.
A third possible solution to the entanglement problem is to reverse course on partisan gerrymanders and declare that they are justiciable. Aside from the unlikely chance that the Court will revisit, let alone reverse, it’s recent decision in Rucho, even if it were to do so, there remains the challenge of defining the standard upon which they can or should be reviewed.
The most obvious and straightforward option is the predominance test—to inquire whether partisan considerations “predominate” over other ordinary districting considerations. In cases where race and partisanship are entangled, this could help solve the larger problem of allowing racial gerrymanders to escape under the cover of partisanship, but it does not actually solve the epistemological problem of how courts may distinguish between entangled factors or inputs. Thus, it would have the inverse effect of the first possibility, which is to help address the larger problem, but leave the smaller one intact.
Moreover, this possibility remains only a partial solution to the larger problem. Even if the Supreme Court were to allow courts to review partisan gerrymandering claims, it is unlikely that such partisan gerrymanders would be reviewed at the exactingly high level of scrutiny as racial classifications are. Thus, there is still some risk present that racial gerrymanders will escape regulation in the guise of partisanship due to the gap in the standards of review.
This third option, however, although requiring a reversal of recent precedent, is at least more logically consistent with the idea that these are categorically distinct claims arising from different case law and constitutional concerns. It renders the entanglement problem less urgently in need of resolution since the more extreme partisan gerrymandering cases would be regulated through a parallel structure under (presumably) rational basis review.
This approach has several other meritorious considerations in its favor, especially its potential grounding in various aspects of constitutional jurisprudence. Some versions of this approach, for example, naturally conform to the paradigm famously known as “Carolene Products footnote four.” This famous footnote in Constitutional Law maps neatly to the entanglement problem at issue in this Article. That is because at the heart of this footnote are issues of political process and racial equality and their interrelation, the same issues here. As the Court said in that famous footnote:
It is unnecessary to consider now whether legislation which restricts those political processes which can ordinarily be expected to bring about repeal of undesirable legislation, is to be subjected to more exacting judicial scrutiny under the general prohibitions of the Fourteenth Amendment than are most other types of legislation. . . . Nor need we enquire whether similar considerations enter into the review of statutes directed at particular religious, or national, or racial minorities; whether prejudice against discrete and insular minorities may be a special condition, which tends seriously to curtail the operation of those political processes ordinarily to be relied upon to protect minorities, and which may call for a correspondingly more searching judicial inquiry.
Carolene Products footnote four has been called a “paradigm” within equal protection jurisprudence on account of the fact that it provides a coherent and comprehensive roadmap for judicial review. It is not just that the footnote addresses one possible way of dealing with laws that affect access to the political process, like political redistricting does, but it also specifically deals with the intersection of race and political access: suggesting that laws which impede racial minorities access to the political process should be reviewed more closely.
The Equal Protection Clause arguably provides a sufficient basis by itself for establishing a rule against partisan gerrymanders: they violate fundamental democratic principles such as those that motivated the Court to rule, in the early 1960s, that political districts should be of equal size. They violate majority rule—what Madison called the “fundamental principle of free government” in Federalist No. 58 and Hamilton called the “the fundamental maxim of republican government” in Federalist No. 22. Whether permitting a minority to entrench itself through the manipulation of district boundaries or by manipulating the number of voters in each district, the result can be the same and does violence to the principle of “one person, one vote” either way.
A claim rooted in equal protection could narrowly assert that permitting extreme partisan gerrymanders would violate a person’s right to be treated equally by law or more broadly assert that it would also hinder access to both the political process and the ability to use that process to remedy unfair or unjust legislation. But even if the Equal Protection Clause itself, or some broader more synthetic reading of it, Carolene Products, or even related associational claims grounded in the First Amendment are part of the foundation for rendering partisan gerrymandering claims justiciable, there is another constitutional provision which has lain dormant but could be potentially enlisted to this cause: the Guarantee Clause.
The Guarantee Clause requires the United States guarantee to the states a republican form of government. Although rarely invoked, the prevailing consensus is that this Clause requires majority rule and that representatives serving in state governments be selected by elections. In other words, it is a guarantee to the citizens of those states (and of the nation) that each state government must be republican in form. This clause might be the basis for challenges to features of various state governments that are anti- or un-democratic. In addition to a formal recognition of the problem of entanglement in the gerrymandering cases, a revival of the Guarantee Clause could provide an easily understandable basis for reversing or excepting Rucho.
The problem here is that Supreme Court precedent does not allowed federal courts to entertain claims brought under the Guarantee Clause, thus far. In 1849, and again in 1946, the Supreme Court ruled that claims under this clause are non-justiciable. Prominent and notable jurists, however, would have held otherwise. In his courageous and lonely dissent in Plessy v. Ferguson, the first Justice Harlan would not only have held that the segregative railway statute adopted by the state of Louisiana and reviewed in that case violated the Thirteenth and Fourteenth Amendments to the Constitution, but also the Guarantee Clause. As he explained:
Such a system is inconsistent with the guaranty given by the constitution to each state of a republican form of government, and may be stricken down by congressional action, or by the courts in the discharge of their solemn duty to maintain the supreme law of the land, anything in the constitution or laws of any state to the contrary notwithstanding.
The aforementioned Supreme Court decisions, moreover, occurred prior to the Court’s ruling that districting cases are justiciable in Baker v. Carr. And in any event, these decisions cannot be based on the meaning of that clause as understood by its framers.
Federalist No. 9 explains that the “principles” of republican governance are “now well understood,” and in addition to the fundamental majoritarian principle, they include: (1) “[t]he regular distribution of power into distinct departments”; (2) “the introduction of legislative balances and checks”; (3) “the institution of courts composed of judges holding their offices during good behavior”; and (4) “the representation of the people in the legislature by deputies of their own election.”
Moreover, Federalist No. 39 dealt specifically with the meaning of republican government. As Madison explained there, “we may define a republic to be . . . a government which derives all of its powers directly or indirectly from the great body of the people; and is administered by persons holding their offices during pleasure for a limited period, or during good behaviour.” In other words, it is a government, in the words of Lincoln, “of, by, and for the people.” Madison goes onto explain that
[i]t is essential to such a government, that it be derived from the great body of the society, not from an inconsiderable proportion, or a favored class of it; otherwise a handful of tyrannical nobles, exercising their oppressions by a delegation of their powers, might aspire to the rank of republicans, and claim for their government the honorable title of republic.
Critics might note that, despite Madison and Hamilton’s declarations on the centrality of the principle of “majority rules,” they and their colleagues seemed unconcerned with the use of districting for partisan advantage in terms of either unequal size of districts or manipulation of lines, or else they would have argued against it (as they did against equal suffrage for states) or proposed or included provisions in the Constitution or in their respective roles in the federal government against it. In Part II, a number of possibilities were presented as to why Madison and Hamilton may not have introduced measures relating to this potential problem. A few more may now be considered.
First, any interpretative methodology assessing the Guarantee Clause would have to account for the fact that many categorical exclusions for voting existed at the time of the framing of the Constitution that are now prohibited, including those on the basis of sex, race, and class (through the Fifteenth, Nineteenth and Twenty-Fourth Amendments, respectively). Thus, there are textual reasons to “update” any originalist understanding of the Guarantee Clause based upon the text of the Constitution itself, as amended. After all, it has already been observed that the modern Supreme Court has repeatedly struck down districting plans with population disparities in percentage terms far less than those observed by Madison and Hamilton in The Federalist Papers.
Second, the framers failed to anticipate the extent to which partisanship would manifest in the federal councils and the harmful effects thereof. There is no reason to believe that they should devise measures to address problems they lacked the foresight to see. And, by their own accounts, the precautions and safeguards that the framers believed would curtail the harmful effects of partisanship were based on assumptions and premises that proved fallacious or were quickly refuted by experience as political parties organized themselves in the federal councils.
To be clear, the framers were well-aware of the problem of partisanship. Having observed it within the states and other republics, Alexander Hamilton referred to this problem as “the diseases of faction” and the “demon of faction,” and James Madison called it “mischiefs of faction” and the “rage of party.” What they underestimated was the degree to which political parties would become the primary organizing forces to frame political discourse and focus policy debate in the federal government.
As Madison and Hamilton explained throughout The Federalist Papers, the framers believed that the size and diversity of peoples and interests represented in the national government would ameliorate the effects of faction as observed in the state governments. As Madison concluded in Federalist No. 10, the “variety of sects dispersed over the entire face of [the confederacy of states] must secure the national councils against any danger from that source.” Hamilton arrived at similar conclusions in Federalist Nos. 60 and 61, where he wrote that “a diversity of local circumstances, prejudices, and interests” would make it unlikely that a “predominant faction” would prefer a particular class of electors over another. Not only that, Hamilton felt that the diverse manner in which the various federal branches would be populated would safeguard against this problem, such that he concluded there is “little probability of a common interest to cement these different branches in a predilection for any particular class of electors.”
If at any moment the quantity of highly competitive national political parties had been greater and more numerous or if political parties had not acquired so much significance in organizing political interests at a national level, then this conclusion might have proven correct. But by the end of the first decade of the government’s operation under the Constitution, the problems of partisanship were already manifest to such an extent that it was one of the principal subjects of concern in George Washington’s 1796 Farewell Address. Far from leaving office assured with his good works and sanguine on the young nation’s prospects, he sounded an alarm.
In this regard, any originalist argument would be incomplete without considering George Washington’s remarks in his Farewell Address, which were drafted with input from Hamilton and Madison, his top advisors. The degree to which they underestimated the corrosive effects of toxic partisanship—what they called the “baneful effects of the spirit of party”—is clear from the substance of the address.
Having observed the emergent dynamics of partisanship firsthand as the nation’s first chief executive, Washington expressed a deep-seated fear that political parties presented a danger to the stability of the young republic, and potentially an existential threat. The Senate Historical Office characterizes his remarks concerning the “dangers of parties in state” as reflecting the view that political parties “carried the seeds of the nation’s destruction through petty factionalism.” The remarks specify, in serial form, the harmful effects that flow from extreme partisanship. Among them:
• “It serves always to distract the public councils and enfeeble the public administration.”
• “It agitates the community with ill founded jealousies and false alarms,”
• “kindles the animosity of one part against another,”
• “foments occasionally riot and insurrection.”
• “It opens the door to foreign influence and corruption, which find a facilitated access to the government itself through the channels of party passions. Thus the policy and the will of one country are subjected to the policy and will of another.”
If anything, the experience of the last few years amply illustrates these dangers, especially in the administration of President Donald Trump, which experienced arguably each of these effects, most obviously in the Ukraine scandal that led to the first impeachment, the interactions with Russian officials that led to the Mueller investigation, and the riot and insurrection at the Capitol on January 6, 2021.
Extreme partisanship has fostered a visceral antipathy against the other party (what political scientists call “negative partisanship”) often for no other reason than the “animosity of one part against another,” such that even bipartisanship on broadly popular legislation (such as when Republicans in Congress voted for the 2021 infrastructure bill) is viewed within the faction as a violation of partisan solidarity.
And, in the most extreme case, above all the previously listed concerns, Washington asserted that partisanship could lead to the “destruction of public liberty” in this way:
The alternate domination of one faction over another, sharpened by the spirit of revenge natural to party dissension, which in different ages and countries has perpetrated the most horrid enormities, is itself a frightful despotism. But this leads at length to a more formal and permanent despotism. The disorders and miseries which result gradually incline the minds of men to seek security and repose in the absolute power of an individual; and sooner or later the chief of some prevailing faction, more able or more fortunate than his competitors, turns this disposition to the purposes of his own elevation on the ruins of public liberty.
Some political prognosticators and military leaders have warned that some version of this frightful vision might be realized in the aftermath of the 2024 election.
Critically, however, Washington also characterized the dangers of extreme partisanship in geographic terms, which he called “the danger of parties in the state, with particular reference to the founding of them on geographical discriminations.” Although he may well have been referring to the sectional divide between the North and the South, the specific formulation is a perfect fit for the excesses of gerrymandering. It accurately characterizes the manifestation of partisanship through gerrymandering, which is a process of making geographic discriminations.
Whether a consequence of the framers’ underestimation of the ultimate role of political parties, the extent to which partisanship would infect the federal government, or some other reason, the Constitution itself is silent in regards to the problems posed by extreme forms of political districting. But that does not mean that the framers left the political community helpless, even in the absence of an explicit constitutional provision adopted to solve this problem. Provisions like the Guarantee Clause are open-textured specifically to empower that political community to address problems such as this, empowering both Congress and the courts to regulate such practices as partisan districting as circumstances may necessitate.
Unlike the canonical and venerated document that it is regarded as today, the United States Constitution was recognized by its framers as a political experiment of uncertain prospects. To give the political community governed by it the flexibility to make it work in practice and tailor it to exigencies without violating its text or spirit, the framers provided an amendment process to correct for unforeseen flaws or circumstances and used terse, open-textured language amenable to varying shades of interpretations in enumerating certain powers, rights, or prohibitions.
Nonetheless, the Constitution was chiefly designed to overcome the deficiencies of the Articles of Confederation and other problems that had plagued the young republic. Consequently, it is silent on many issues that subsequently bedeviled the nation governed by it in intervening centuries. Two notable examples that frustrated subsequent generations in the first half of the nineteenth century were the definition of national citizenship and its relationship to state citizenship, and in the latter half of the twentieth century and early twenty-first century, the issue of abortion.
Because the original Constitution did not explicitly indicate how federal citizenship was defined or acquired, the Supreme Court ultimately decided the question of whether persons of African descent were or could become United States citizens, which it did in the most notorious case of Dred Scott v. Sandford. The Court’s decision was reversed by opening line of the Fourteenth Amendment. Similarly, the issue of abortion has proven to be highly divisive and one of the most deeply contested legal matters of the last fifty years. Advocates and jurists often look beyond the explicit text to the structure of the Constitution and the history and traditions of the republic at certain points in time to try to resolve these matters.
Unfortunately, extreme manipulation of political districting processes is another issue upon which the Constitution remains explicitly silent. This does not mean federal officials are powerless to address it. Constitutional text provides indirect solutions, such as the affirmative powers afforded Congress under the Elections Clause to “make or alter” the laws for elections to the federal legislature and implicit protections made by inferences drawn from other provisions, such as the Equal Protection Clause. Nonetheless, as a result of the lack of constitutional specificity regarding this issue, certain problems generated by this underlying phenomenon are treated differently, depending on the circumstances, the class of persons most affected, or the form of the districting problem.
This Article focuses on the problem of the entanglement of race and partisanship in the judicial review of gerrymandering claims. It conducts a brief history of gerrymandering, examines the divergent lines of cases, reveals the factors that contribute to the growing problem of gerrymandering, including the relationship between political segregation and racial residential segregation, and closes with a survey of possible solutions grounded in the constitutional text and structure.
Although Congress could potentially pass laws curbing gerrymandering in the states, and would have the authority to do so under Article I, Section 4, the core of the problem of gerrymandering is that it violates what Madison called the “fundamental principle of free government”—that of majority rule, and therefore should be within the cognizance of the Constitution, not ordinary legislation, to resolve. This is true even though the Constitution is silent on it, an omission that is adequately compensated for by the applicability of indirect provisions that the framers included such as the Guarantee Clause and subsequent Amendments, most notably the Fourteenth, requiring equal protection of the law.
Blame for underestimating the rise of political parties and the harmful effects of extreme partisanship cannot be entirely placed on insufficient foresight of the framers. A number of developments have contributed to the intensity of political partisanship in the federal government, including the enlargement of the sphere of national politics, the evolution of the information and media environment, and technological developments.
There is no way that the framers could have fully anticipated the extremities toward which political districting processes designed for partisan purposes might distort many of the principles of representative government that they sought to institutionalize. In particular, they could not have anticipated the development of modern technological tools such as computer databases and programs such as Geographic Information System (“GIS”) technology that would easily permit state legislators to essentially select their voters rather than the other way around. But the framers’ insufficient foresight does not leave us helpless.
Whether the remedy lies in an act of Congress or the courts applying a synthetic reading of the Constitution as a whole, Carolene Products footnote four, the Guarantee Clause, or a novel reading of the Equal Protection Clause, partisan gerrymandering is a problem for our health of democracy that requires resolution. It is a problem in its own right because it undermines the values and foundation of the republic, and because it causes and results in the racial segregation of voters in clear violation of the Constitution without necessarily running afoul of the standards established by the Supreme Court to secure those protections.
* Stephen Menendian is the Assistant Director at the Othering and Belonging Institute at University of California, Berkeley. The author would like to thank john powell and Dan Tokaji for their insights on this critical issue, Chris Elmendorf and Joshua Clark for their invaluable expert feedback on drafts of this Article, Samir Gambhir and Peter Mattingly for their contributions to the underlying research regarding segregation and assistance in developing the maps and scatterplots, Wenqi (Michael) Xu, Sara Osman, and Yemaj Sheik for their general research and citation assistance.