Ugly: an inquiry into the problem of racial gerrymandering under the Voting Rights Act.

• Author: Polsby, Daniel D.
• Position: Symposium: The Future of Voting Rights After Shaw v. Reno

INTRODUCTION

If you have to ask, you can't afford it. -- Attributed to J.P. Morgan(1)

How ugly is too ugly? Shaw v. Reno(2) examines that important question in the course of rejecting North Carolina's 1990 effort to comply with the preclearance provisions in section 5 of the Voting Rights Act (VRA).(3) But the problem presented is a general one that is apt to arise perennially in any territory-based system of representative democracy. What has "shape" to do with representation? Quite a lot, as we hope to demonstrate.

In Shaw, North Carolina's congressional redistricting map -- ugly enough as first proposed(4) -- was made uglier to accommodate the creation of a second "safe" district for an African-American representative. 5 In the jargon of electoral districting, an "ugly" map is one filled with irregular, uncompact shapes that do not evidently correspond to established political or natural boundaries. An ugly map implies that a human ambition of some kind, with politically strategic ulterior motives, has been hard at work.

Pared to its simplest form, the question is this: Why does an effort to comply with conditions set by the Attorney General for preclearance under the VRA(6) by consciously creating a district with certain racial demographics -- through gerrymandering or otherwise -- violate the Constitution? In North Carolina's case, the government created the very ugly Twelfth District to satisfy administrative preclearance procedures established by the VRA. Although the Supreme Court has read the VRA expansively in several cases,(7) the Shaw Court -- shocked to find racial gerrymandering going on in the back room -- remanded for a finding whether this sort of thing is absolutely necessary.(8)

Of course, whether a racial gerrymander is necessary will depend upon a number of variables. Is it right to allow any form of race-based districting? If it is, may the districtmaking authorities attend to other political business first, and only take race into account afterward? The construction of nonugly districts might have been easier if the districtmakers were not trying to do so many things at once. The protection of minority voting interests was a concern to which North Carolina mapmakers turned only after they had other, more important fish fried, including ensuring reelection of incumbent members of Congress(9) -- such as Steve Neal of the ugly Fifth District, chairman of the subcommittee that supervises the regulation of financial institutions; and Charlie Rose of the very ugly Seventh District, who chairs the House subcommittee with jurisdiction over tobacco and peanut subsidies and who aspires to be Speaker of the House.(10)

In addition to incumbent protection, mapmakers also have the overall partisan makeup of the delegation to consider.(11) Every decennial census indicates that the Democratic Party's traditional hold on North Carolina's voters is eroding. As recently as 1960, Democratic candidates for Congress in North Carolina outpolled Republicans by a wide margin and held eleven of the state's twelve seats in the House of Representatives; in 1970, the Democrats still led by fifty-three to forty-seven percent and held seven of eleven seats.(12) By 1992, Republican vote totals had almost matched those of the Democrats: out of 2.48 million votes cast for congressional candidates, Democratic votes surpassed Republican by only about three percentage points.(13) Yet, in the end, the state's delegation of twelve was Democratic by an eight-to-four margin.(14) This result partly reflects one of the properties of the single-member district with plurality voting rules (SMDPV), as will be discussed below.(15) But one can infer that a certain amount of artifice has entered into the outcome; North Carolina seems to have gerrymandered its districts a little.

In the discussion that follows, we focus on the case of congressional districting rather than on districting in general. Although we proceed in this manner for the sake of clarity, it is also true that no single, all-purpose normative theory of electoral mechanics will cover every case of democratic representation, from county commissions to mosquito control districts to sovereign legislatures. We do not claim that one can generalize our argument to every sort of election to which the VRA might apply. Yet we think our argument does approximate a theory of general application.

The Shaw decision is nebulous and fits uncomfortably into the Supreme Court's VRA jurisprudence. It raises new problems rather than working with the ones already in play. Most puzzling is how Shaw treats the current constitutional status of "compactness" as a principle of districtmaking. On the one hand, the Court says that compactness is not constitutionally mandated;(16) on the other hand, it says that North Carolina's Twelfth District was simply too ugly to be legal.(17) Why? What did the North Carolina defendants do wrong? If racial gerrymandering is unconstitutional, where does that leave statutes like the VRA, which courts and commentators have always assumed to entail race-conscious districtmaking?

Shaw addresses a fundamental problem of representative democracy -- namely, how far a legislature may go in controlling who is elected to it. In some respects, this question is the most fundamental that we can ask in our constitutional system. The compactness of districts is pertinent to this inquiry. We argue that compactness in some sense is ordinarily a property of single-member territorial districts. We offer a proof that gerrymandering -- whether it is conducted under the auspices of the VRA or is freelance legislative tinkering -- can spoil the game for representation by single-member district. In order to avoid that destiny, an antigerrymandering principle must be defined and administered outside normal political channels.

1. THE GENESIS OF SHAW

   In the beginning, the Voting Rights Act was simple enough. Acts that were prohibited -- that is, impediments to the exercise of the franchise such as poll taxes, chicanery by vote registrars, obstructing polling places, and the like -- and the reasons Congress prohibited these obstacles were well understood by everyone connected with the election process.(18)

   After the Supreme Court decided Mobile v. Bolden(19) in 1980, however, matters took an esoteric turn. Bolden involved the at-large election of city commissioners. The African-American plaintiffs complained that this procedure deprived them of Fourteenth and Fifteenth Amendment rights because it led to a situation in which a simple majority (fifty percent plus one) could win one hundred percent of the seats. The justiciability of this sort of vote-dilution claim had been recognized in prior Supreme Court cases.(20) Following the rule of Washington v. Davis,(21) Arlington Heights v. Metropolitan Housing Development Corp.,(22) and Personnel Administrator v. Feeney,(23) the Supreme Court held that plaintiffs must show purposeful discrimination to state a cause of action. However, proving the state of mind of wary public officials, even ones who were acting in bad faith, would not be easy. Shortly after Bolden, therefore, Congress amended the VRA to give Bolden-type plaintiffs relief that the Constitution, as interpreted by the Supreme Court, would not give them. The VRA amendments protected minority voters' "opportunity . . . to participate in the political process and to elect representatives of their choice."(24) A violation exists if the participation-election opportunities of voters in the protected class are less than those of other voters. Courts relieve plaintiffs hoping to make such a showing of the obligation to prove intentionally discriminatory acts. Amended section 2 charted a new course for voting-rights law not only because it established a new cause of action for plaintiffs, but because section 2 constitutes one of the section 5 preclearance criteria that the Attorney General employs in cases like Shaw.(25) In order to understand section 5, therefore, it is first necessary to understand section 2, a difficult task. The legislative history of the 1982 amendments is filled, as Professor Issacharoff has noticed,(26) with obviously strategic talk and rejoinder designed to influence the spin that courts would put on the legislation. But a problem remains at the heart of the statute that partisan infusions in legislative history can do little to fix. The statute is paradoxical in that it seems to mean two conflicting things at once. In amending section 2, it seems that Congress proposed to abate Bolden and to resurrect the standard in White v. Regester,(27) which propounded a "totality of circumstances" methodology for determining when the denial of a section 2 "opportunity to participate and to elect" had occurred.(28) This standard later appeared in the Senate Judiciary Committee's Majority Report.(29) What, then, does section 2 mean by "opportunity to participate and to elect," over and above freely voting and having one's votes honestly counted?

   The answer to this question, such as it is, appears in the Supreme Court's decision in Thornburg v. Gingles.(30) The "opportunity to participate and to elect"(31) Means not to have one's vote "diluted."(32) But dilution is neither a statutory nor intuitive term. When does dilution take place? Dilution occurs when an electoral structure "submerges" a protected minority.(33) What does the term submergence mean?(34) Gingles sets forth a three-prong test for determining when a statutory violation has occurred:

   First, the minority group must be able to demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single-member district. . . . Second, the minority group must be able to show that it is politically cohesive. . . . Third, the minority must be able to demonstrate that the white majority votes sufficiently as a bloc to enable it . . . usually to defeat the minority's preferred...