The Right to Vote

by Alexander Keyssar

  The upshot of this assembly of cases was a broadening of the legally operative definition of the right to vote. By 1965, the Constitution was interpreted to mean that individuals not only had the right to register, cast their ballots, and have their ballots counted, but also that they had the right to have their votes count as much as the votes of other citizens. Votes could not be weighted more heavily in some locales than in others; nor could voting districts be significantly unequal in population. The federal government, moreover, had assumed responsibility for judging the legality and legitimacy of federal, state, and local electoral arrangements—and it would have to make those judgments and apply the one person, one vote standard each time that the nation’s population was counted.

  Preclearance and the Totality of Circumstances

  By the late 1950s, the city of Tuskegee, Alabama, contained a sizable number of African Americans who had succeeded in becoming registered voters. Indeed, black registrants were sufficiently numerous that they posed a threat to white political control of the city. To meet this threat, the state legislature completely redrew the city’s boundaries, creating a bizarre twenty-eight-sided municipality with an almost entirely white population: blacks found themselves consigned to the surrounding counties, where they lacked the numbers to wield much influence. The legislature’s step was not unprecedented: racial districting had been a common form of political warfare throughout the South, particularly in the years before the wholesale disfranchisement of the black population. What happened in Tuskegee was unusually flagrant, however, and the city’s black citizens, emboldened by the civil rights movement, mobilized collectively to protest the new city boundaries. In response to these protests, and acting under the Civil Rights Act of 1957, the Justice Department promoted a lawsuit by local black citizens. For the first time in the twentieth century, the Supreme Court in 1960 heard a case involving racial gerrymandering (Gomillion v. Lightfoot), and led by Justice Frankfurter, the Court ruled that the legislature’s action was a clear violation of the Fifteenth Amendment.66

  The events in Tuskegee (and similar developments elsewhere) served notice to civil rights activists and Justice Department lawyers that many white Southerners were not prepared to share political power with blacks. If the federal government insisted on black enfranchisement, conservative Southerners would attempt to vitiate its consequences by altering the structures of representation. Deannexation, as had occurred in Tuskegee, was one way of doing so; annexation (of white populations) was another. Racial gerrymandering could take the form of cracking (i.e., dividing minority voters into several districts) or stacking (i.e., loading voters into a single district). In the numerous cities with white majorities but sizable black populations, whites could maintain a monopoly on political power by having all city council members elected “at large” rather than from single-member districts. A similar result could be achieved by insisting on majority runoffs rather than plurality victories. Given the long history of racial discrimination and the recent history of strenuous resistance to civil rights, it would have been surprising had southern whites in the 1960s not turned to such mechanisms to remain in power.

  No doubt it was with this in mind, as historian J. Morgan Kousser has argued, that the Voting Rights Act insisted on the federal “preclearance” of any new “voting qualification or prerequisite to voting or standard, practice, or procedure with respect to voting” in all locales that had a history of discrimination. As Attorney General Nicholas Katzenbach testified to Congress during hearings on the Voting Rights Act, “there are an awful lot of things that could be started for purposes of evading the Fifteenth Amendment if there is the desire to do so.” The preclearance procedure—requiring targeted states and counties to get federal approval for any new or changed electoral procedures—was a mechanism designed to prevent recalcitrant white Southerners from undermining the effectiveness of black enfranchisement.67

  The scope of the preclearance provision (section 5) was unclear during the first years after passage of the Voting Rights Act, and the mechanism was little used. This changed dramatically in 1969, as a result of a cluster of lawsuits that came before the Supreme Court. Originating in Mississippi and Virginia, these cases involved an array of different alterations of electoral structures: a shift from district to at-large voting for county supervisors, the appointment rather than election of superintendents of education, new rules governing the ability of independent candidates to secure a place on the ballot, and modifications in the procedures for write-in ballots. Bringing the cases together under one decision (Allen v. State Board of Elections), the Court ruled that all of these changes were subject to the preclearance provision of the Voting Rights Act. Rejecting the claim that section 5 applied only to laws governing the registration of voters, the Court concluded that the Voting Rights Act “was aimed at the subtle, as well as the obvious” and thus that all modifications of electoral laws were to be screened by the Justice Department for their potentially discriminatory effects. Two years later, the Court affirmed this broad interpretation, finding that the preclearance provision applied also to municipal annexations and the relocation of polling places.68

  The Court’s actions meant that any state or county covered by the trigger mechanism of the Voting Rights Act (i.e., those with histories of discrimination) had to clear all changes in its electoral rules with the Justice Department. This requirement generated heavy traffic in the wake of the 1970 Census, as states redistricted to meet the one person, one vote requirement of Reynolds v. Sims and tried to seize the opportunity to create racially biased structures and procedures. Despite the initial hesitance of the Nixon administration, the Justice Department enforced the Voting Rights Act with some vigor; it consequently found itself screening thousands of changes in electoral law, more than two hundred of which were rejected. In subsequent years, in response to the ingenuity of southern conservatives, the federal courts enlarged the reach of the preclearance provision, as did Congress with its 1975 extension of the Voting Rights Act to cover language minorities.69

  In addition, the courts wrestled with constitutional challenges to potentially discriminatory schemes that were not subject to the preclearance clause: those adopted before 1965 (and thus were not changes in the law per se) and those in states and counties not under direct federal supervision. In 1971, in an Indiana case, the Court made clear that multimember districts (i.e., those in which more than one representative was elected by a particular geographic unit) were not necessarily unconstitutional, but were suspect—because they created an opportunity for elections to be structurally stacked against black voters. In 1973, the Court went a step further, unanimously striking down multimember districts in two Texas counties because they diluted the votes of African Americans and Mexican Americans. Its decision was based on a close inspection of the “totality of circumstances” enveloping the districting system, including the state’s lengthy history of racial discrimination. Implicit in the ruling was the notion that blacks and Hispanics would have greater political representation if districts were smaller and each elected only one official. The totality of circumstances standard was refined and given specificity by a circuit court in Zimmer v. McKeithen (1973), and the Zimmer test was widely applied for the remainder of the 1970s. In practice, the test rendered at-large and multimember district elections suspect throughout the South.70

  The totality of circumstances standard emerged from the Court’s dawning recognition of the complexity of the problem that it faced: How was a court to tell whether a districting system—or any other electoral arrangement—discriminated against blacks or other minorities? The easy cases were those where discriminatory intent could be established, but by the 1970s, southern lawmakers—having learned the lesson of Gomillion—were sufficiently sophisticated to make intent difficult to prove. New laws invariably were accompanied by racially neutral justifications, such as administrative efficiency. Alternatively, a finding of discrimination could be based on the effects of ne
w electoral arrangements, but this too had hazards: the fact that an electoral structure left minorities unrepresented, for example, did not prove unlawful discrimination. Interest groups that could not muster a majority always were unrepresented as a result of elections; that was how the system worked. The totality of circumstances standard—which embraced both intentions and effects—was a strategy for coping with this dilemma, but it did not necessarily provide clear answers. The standard implied that courts ought to consider the effects of electoral arrangements, but how were the effects of an annexation or districting system to be gauged? What was it that had to be measured, and to what should such measures be compared? What was the yardstick that told whether the effects of at-large voting or deannexation were illegally discriminatory?71

  The most obvious answer to such questions would have been the adoption of some form of demographic proportionality for racial minorities: the percentage of minority elected officials ought to be roughly comparable to the minority percentage of the population. If African Americans constituted one-third of the inhabitants of a city or state, they should have one-third of the representatives, and a fair electoral structure would make that happen. Without such a standard, the voices of a long-oppressed minority would continue to be muffled. Yet this obvious answer was not easy to accept. A system of racially based proportionality was repugnant to the traditional American emphasis on individual rather than group rights; it violated the integrationist norm of race blindness; and it conveyed the balkanizing implication that blacks could only be represented adequately by other blacks. In addition, racially based proportionality privileged the collective political rights of some groups over others. If blacks and Hispanics were entitled to a certain percentage of a city’s representatives, what about Asians, Jews, Catholics, or union members?72

  The Supreme Court struggled, with much internal debate, to find some middle ground on which to stand, a way to be sensitive to the history of racial discrimination without imposing demographic proportionality or opening a new Pandora’s box of claims for representation by nonracially defined social or political groups.73 The result was something of a muddle, shaped in part by the growing conservatism of the post-Warren Court. In 1973, the Court allowed Petersburg, Virginia, to annex white neighborhoods from an adjacent county (which would give the city a white rather than black majority); at the same time, however, it advised the city to switch to district rather than at-large elections to give blacks greater representation. Two years later, the Court permitted nearby Richmond to annex a white neighborhood and adopt a new districting system, because the overall plan “fairly” recognized “the minority’s political potential” and offered African Americans representation “reasonably equivalent to their political strength in the enlarged community.” The next year, a majority of the Court veered in a different direction, sanctioning a New Orleans proposal that would increase black representation on the city council from what it had been, but not to a level proportional to the black population of the city. The Court did not consider the plan to be in violation of the Voting Rights Act because there was no “retrogression” in the “effective exercise of the electoral franchise” by blacks. Liberal dissenters on the Court objected loudly, stressing that the absence of “retrogression” was a woefully inadequate standard and that there had to be “legislative representation roughly proportional to the Negro population.”74

  In 1977, in United Jewish Organizations of Williamsburgh, Inc. v. Carey, the Court dealt head-on, if clumsily, with the balkanizing issue: it rejected the Fourteenth and Fifteenth Amendment claims of a group of thirty thousand Hasidic Jews in Brooklyn who found themselves split up between two districts because of an apportionment plan that sought to produce “nonwhite majorities” in those districts. The Jewish community claimed that the plan greatly diminished its chance of electing one of its own members to office. The Court ruled, however, that for the purposes of districting Hasidic Jews were simply “white” (whatever their own view of their cultural identity), that whites on the whole were adequately represented, and that it was permissible for race to be used as a factor in drawing district boundaries.75

  This less-than-coherent attempt to find middle ground came to an abrupt halt in 1980, when the Supreme Court jettisoned the Zimmer test and the totality of circumstances standard in City of Mobile, Alabama v. Bolden. Mobile had been governed since 1911 by a three-person commission, chosen in at-large elections. Although Mobile’s population was 35 percent black, no African American had ever been elected to the commission. Applying the totality of circumstances approach, a federal district court (sustained by the appeals court) found sufficient evidence of dubious voting practices and historical discrimination to rule that black voting strength had been diluted in violation of the Fourteenth and Fifteenth Amendments and the Voting Rights Act. A very fragmented Supreme Court (issuing six separate opinions) overturned the verdict, concluding that there was insufficient evidence of discriminatory intent and that only actions undertaken with “racially discriminatory motivation” were unconstitutional or illegal under the Voting Rights Act. The Court ruled that the absence of black commissioners was not by itself evidence of discrimination, that there was no Fourteenth Amendment right to proportional representation, and that at-large voting always disadvantaged minorities of different types. Four of the justices further maintained that the Fifteenth Amendment applied only to the actual exercise of the franchise, not to vote dilution or districting.76

  Mobile v. Bolden simplified the law while making it far more difficult for claims of vote dilution to be sustained. Proving an overt discriminatory intent, rather than deducing intent from the consequences of a law, was extremely difficult: in the immediate aftermath of the decision, numerous suits were dismissed or withdrawn, judgments reversed, and challenges to at-large voting systems turned down in the lower courts because of lack of evidence of deliberate discrimination. Civil rights advocates were enraged by what appeared to be the Court’s reversal of precedent and its becoming—in the words of Justice Marshall’s vehement dissent—“an accessory to the perpetuation of racial discrimination.”77

  Thanks to the happenstance of timing, the reign of Bolden was brief. In 1982, when the Voting Rights Act was being renewed by Congress, liberal Democrats, joined by moderate Republicans, rewrote the law to put greater emphasis on effects than on intent. Recognizing the near impossibility of proving deliberate and purposeful discrimination, section 2 of the amended law, as drafted by Senator Robert Dole, prohibited any voting standard or procedure that “results in a denial or abridgement of the right of any citizen . . . to vote on account of race or color.” This replaced the original language, which implied that a state law violated the Voting Rights Act only if discriminatory intent could be established. The amended act underscored this shift by declaring that a standard or procedure was illegal ifbased on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered.

  In a disclaimer important to conservatives and some smaller minority groups, the bill also contained an explicit rejection of demographic proportionality: “Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.”

  The totality of circumstances standard had been restored. Two days after President Reagan signed the Voting Rights Act extension, the Supreme Court signaled the change in direction by easing the requirements for establishing proof of intent in Fourteenth Amendment cases, ruling that circumstantial evidence could be sufficient to establis
h discrimination.78

  Defining Dilution

  Restoring the totality of circumstances approach also meant restoring its attendant problems and confusions. As a result, the lower courts were uncertain about how to proceed and inconsistent in their rulings. In 1986, the Supreme Court attempted to clarify matters in a North Carolina redistricting case, Thornburg v. Gingles. In yet another fragmented decision, the Court articulated a set of conditions that had to be met to establish that a multimember or at-large districting plan illegally diluted the votes of an African-American minority. First, the minority had to be “sufficiently large and geographically compact to constitute a majority of a single district.” Second, it had to be “politically cohesive.” Third, there had to be evidence, over time, that whites voted as a bloc in such a way as to “usually” defeat black candidates. Other factors (such as those from the Zimmer test) could certainly be weighed to help determine whether a plan was illegally discriminatory, but the presence of racially polarized voting—measured through standard statistical methods—was the heart of what became known as the Gingles test. If whites tended strongly to vote for white candidates and blacks for black candidates, then a reasonable districting system ought to produce both white and black representatives.79

  The Gingles decision was hailed as a victory for civil rights advocates, largely because it confirmed the Court’s retreat from the imposition of an “intent” standard. But Gingles did not end either the confusion or juridical zigzagging. For one thing, its guidelines were not readily applicable to all types of vote dilution cases, particularly those where multimember districts were not at stake. In addition, many of the key terms of the Gingles test were subject to interpretation, including geographic compactness, cohesive, usually, and even majority. (Could blacks and Hispanics, for example, be combined to form the majority of a new single-member district?) It also was unclear to the lower courts whether the three prongs of the Gingles standard had replaced the totality of circumstances approach or whether they should serve instead as an efficient way to do a first screening of potential claims. Nonetheless, the decision facilitated the adjudication of many cases, and it encouraged both the lower courts and the Justice Department to promote single-member and “minority opportunity” districts in the numerous locales where the Gingles criteria were met. Indeed, the Court’s ruling served as a mandate, for state lawmakers as well as Justice Department officials engaged in the preclearance process, to create “majority-minority” voting districts in cities and states that contained sizable minority populations and had a record of racially polarized voting.80