Colleton County Council v. McConnell

201 F.Supp.2d 618 (2002)

COLLETON COUNTY COUNCIL; Steven Murdaugh, individually and in his official capacity as Chairman of Colleton County Council; Dr. Joseph Flowers, individually and in his official capacity as a member of Colleton County Council; Janice H. Alexander, Andrew F. Calcutt, Joseph Hamilton, Hazel Harrelson, Jas. P. Harrelson, William K. Padgett, Daisy F. Rizer, I.N. Rizer, David M. Smalls, A.L. Smoak, Jr., Kathleen V. Steedly, Wendell M. Steedly, Cheryl R. Tilman, and W. Gene Whetsell, Plaintiffs, and
Sel Hemingway, individually, Intervenor-Plaintiff,
v.
Glenn F. McCONNELL, in his official capacity as the President Pro Tempore of the South Carolina Senate; David H. Wilkins, in his official capacity as the Speaker of the South Carolina House of Representatives, and James H. Hodges, in his official capacity as the Governor of South Carolina, Defendants.
Hugh K. Leatherman, individually and as Senator from the 31st District, Scott H. Richardson, individually and as Senator from the 46th District, and Robert W. Hayes, Jr., individually and as Senator from the 15th District, Plaintiffs,
v.
Glenn F. McConnell, in his capacity as President Pro Tempore of the Senate and Chairman of the Senate Judiciary Committee, David H. Wilkins, in his capacity as Speaker of the House of Representatives, and James F. Hendrix, in his capacity as Executive Director of the State Election Commission, Defendants,
James H. Hodges, in his official capacity as the Governor of South Carolina, Intervenor-Defendant.
Kamau Marcharia, James Melvin Holloway, Ann Johnson, and Elder James Johnson, Plaintiffs,
v.
James H. Hodges, in his official capacity as Governor of South Carolina; Glenn F. McConnell, in his official capacity as President Pro Tempore of the South Carolina Senate; David H. Wilkins, in his official capacity as Speaker of the South Carolina House of Representatives; and James F. Hendrix, in his official capacity as Executive Director of the South Carolina
State Election Commission, Defendants.

Nos. CIV.A. 01-3581-10, CIV.A. 3:01-3609-10, CIV.A. 3:01-3892-10.

United States District Court, D. South Carolina, Columbia Division.

March 20, 2002. Order Clarifying Opinion, April 18, 2002.

Carl Frederick Muller, John C. Moylan, III, James C. Parham, Alice Parham, Matthew Richardson, Wyche, Burgess, Freeman, and Parham, P.A., Greenville, SC, for Plaintiffs Colleton County Council, Steven Murdaugh, Joseph Flowers, Janice H. Alexander, Andrew F. Calcutt, Joseph Hamilton, Hazel Harrelson, Jas. P. Harrelson, William K. Padgett, Daisy F. Rizer, I.N. Rizer, David M. Smalls, A.L. Smoak, Jr., Kathleen V. Steedly, Wendell M. Steedly, Cheryl R. Tilman, and W. Gene Whetsell.

William L. Pope, Pope and Rogers, Columbia, SC, for Intervenor-Plaintiff Sel Hemingway.

John F. Hardaway, Columbia, SC, Robert N. Hunter, Jr., Patton, Boggs & Blow, Greensboro, NC, for plaintiffs, Hugh K. Leatherman, Scott H. Richardson, Robert W. Hayes, Jr.

Herbert Edward Buhl, III, Columbia, SC, Laughlin McDonald, American Civ. Liberties Union Foundation, Inc., Atlanta, GA, Maha Zaki, Atlanta, GA, for plaintiffs Kamau Marcharia, James Melvin Holloway, Ann Johnson, and Elder James Johnson.

Charles A. Terreni, Terreni Law Firm, Columbia, SC, Steven W. Hamm, Richardson, Plowden, Carpenter and Robinson, P.A., Columbia, SC, Michael A. Carvin, Louis K. Fischer, Jones, Day, Reavis and Pogue, Washington, SC, for defendant Glenn F. McConnell.

Reginald I. Lloyd, Paige Jones Gossett, Willoughby and Hoefer, P.A., Columbia, SC, Charles Fennell Reid, McNair Law Firm, P.A., Columbia, SC, for defendant David H. Wilkins.

Brent O. E. Clinkscale, Womble, Carlyle, Sandridge, & Rice, Greenville, SC, Richard M. Gergel, David E. Rothstein, Carl Solomon, Gergel, Nickles, & Solomon, Columbia, SC, A. Lee Parks, Parks, Chesin, Walbert & Miller, Atlanta, GA, I.S. Leevy Johnson, Johnson, Toal & Battiste, P.A., Columbia, SC, Brenda Reddix-Smalls, Reddix-Smalls & Carter Law Firm, Columbia, SC, for defendant James H. Hodges.

Charlie Condon, Atty. Gen. of South Carolina, Treva Ashworth, Deputy Atty. Gen., J. Emory Smith, Jr., Asst. Deputy Atty. Gen., Columbia, SC, for James F. Hendrix.

Before WILLIAM B. TRAXLER, Jr., United States Circuit Judge, and JOSEPH F. ANDERSON, Jr., Chief Judge, and PERRY, Senior District Judge.

ORDER

WILLIAM B. TRAXLER, Jr., Circuit Judge.

The United States Constitution requires the governing officials of the State of South Carolina to enact new districting plans for the South Carolina Senate, the South Carolina House of Representatives, and the United States Congressional districts within the state on an equipopulous basis every ten years, in accordance with population changes revealed by the decennial census. Unfortunately, the governing officials of the State of South Carolina, following receipt of the 2000 census data, failed to successfully fulfill this duty and have now reached an impasse. After a lengthy period of mapping, the South Carolina General Assembly, in which Republicans constitute a majority of both houses, prepared redistricting plans for all three bodies, but the plans were vetoed by Governor James H. Hodges, a Democrat. The General Assembly failed in its attempt to override the veto, prompting the filing of these consolidated lawsuits. In each case now before the court, the plaintiffs seek a declaration that the existing districting plans for each elective district are unconstitutional and the implementation of interim court-ordered plans in time for impending 2002 elections. Thus, this court has once again been placed into the center of partisan politics in South Carolina, assigned the "unwelcome obligation" of devising and approving redistricting plans for each legislative body. Connor v. Finch, 431 U.S. 407, 415, 97 S.Ct. 1828, 52 L.Ed.2d 465 (1977).^[1]

I. Background

South Carolina's General Assembly is composed of two bodies: a Senate with forty-six single-member district seats, see S.C. Const. art. III, §§ 1 & 6, and a House of Representatives with 124 single-member district seats, see S.C. Const. art. III, § 3. South Carolina is also entitled to six representatives in the United States House of Representatives. Within quite limited variances, federal law requires that the South Carolina General Assembly redraw each seat in its bicameral state legislature, see Reynolds v. Sims, 377 U.S. 533, 577-80, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964), as well as each of its six congressional seats in the United States House of Representatives, see U.S. Const. art. I, § 2, cl. 3; Karcher v. Daggett, 462 U.S. 725, 730-31, 103 S.Ct. 2653, 77 L.Ed.2d 133 (1983), on an equipopulous basis in accordance with the results of the decennial census.

In anticipation of receiving the results of the 2000 census, and the inevitable malapportionment of existing districts it would reveal, the South Carolina General Assembly set upon a course to adopt new redistricting plans for its two governing bodies, as well as for its six United States Congressional districts. The process began in January 2001, with the introduction of skeleton bills H.3003 (for the South Carolina House of Representatives) and H.4182 (for the United States House of Representatives). The bills were given first reading and referred by the House Speaker to the House Judiciary Committee for use as redistricting vehicles in the General Assembly.

The House Judiciary Committee received the year 2000 census data from the United States Census Bureau on March 15, 2001. During the month of June 2001, the House Election Laws Subcommittee held public hearings in several locations throughout the state, taking testimony from citizens and public officials regarding the proposed House, Senate, and Congressional plans. The full House of Representatives began consideration of House and Congressional redistricting plans passed by the House Judiciary Committee on August 13, 2001. Plans for the House (H.3003) and Congressional seats (H.4182) were subsequently passed by the House and then submitted to the Senate for consideration on August 17, 2001.

The Senate, through a similar Redistricting Subcommittee of its Judiciary Committee, had also been working on a redistricting plan for the South Carolina Senate. Upon receipt of the passed House Plan and the House version of the Congressional Plan, the Senate combined its Senate Plan with the House Plan into H.3003, attached an amended Congressional Plan, and returned H.3003 to the House on August 22, 2001. The House, after concurring in the Senate amendments, ratified H.3003 and sent the bill to Governor Hodges on August 27, 2001.

Three days later, Governor Hodges returned a veto message for H.3003 to the General Assembly. The Governor's stated reason for vetoing the legislatively passed redistricting plan centered on the claim that the House and Senate plans should have created more so-called minority "influence districts," defined by the Governor as districts with a black voting age population ("BVAP") of between 25% and 50%, and a claim that the Congressional Plan unnecessarily split several counties within the state. On September 4, 2001, the House attempted to override the Governor's veto, but failed by a vote of 73 to 46. Consequently, H.3003 was never enacted as law. See S.C. Const. art. IV, § 21 (requiring that all laws be passed by the General Assembly and signed by the Governor to be effective, unless two-thirds of both the House and the Senate vote to override a gubernatorial veto).

Our involvement in this uniquely state matter resulted from the filing of three separate lawsuits, all of which allege that the existing election districts for the South Carolina General Assembly and the United States Congressional seats in South Carolina violate the "one-person, one-vote" requirement of the United States Constitution. See Karcher, 462 U.S. at 731, 103 S.Ct. 2653; Reynolds, 377 U.S. at 568, 84 S.Ct. 1362.

On September 4, 2001, citizens of Colleton County^[2] filed suit against Glenn F. McConnell, in his official capacity as the President Pro Tempore of the South Carolina Senate; David H. Wilkins, in his official capacity as the Speaker of the South Carolina House of Representatives, and James H. Hodges, in his official capacity as the Governor of South Carolina. On September 6, 2001, an action against McConnell, Wilkins, and James F. Hendrix, in his official capacity as Executive Director of the State Election Commission, was brought by Hugh Leatherman, individually and as Senator from the existing 31st Senate District; Scott H. Richardson, individually and as Senator from the existing 46th Senate District; and Robert W. Hayes, Jr., individually and as Senator from the existing 15th Senate District. On October 1, 2001, plaintiffs Kamau Marcharia, James Melvin Holloway, Ann Johnson, and Elder James Johnson, African-American registered voters who reside in South Carolina, brought suit against Hodges, McConnell, Wilkins, and Hendrix. And, on October 30, 2001, the Chairman of the Georgetown County Council, Sel Hemingway, in his individual capacity, was allowed to intervene in the pending cases and to file a complaint against McConnell, Wilkins, and Hodges.

Although each group of plaintiffs seeks to advance its own unique interests, all seek essentially the same broad relief — a declaration that, based on the population changes revealed by the 2000 census, the existing districts for the South Carolina Senate and South Carolina House of Representatives are malapportioned in violation of the one-person, one-vote principle of the Fourteenth Amendment to the United States Constitution, a declaration that the existing districts for South Carolina's Congressional seats are malapportioned in violation of Article I, § 2 of the United States Constitution, and a declaration that the legislative redistricting process in South Carolina has reached an impasse, necessitating judicial intervention. Additionally, the parties seek implementation of interim districting plans by the court for all upcoming elections. All of the pending cases were consolidated and designated to be heard before this three-judge panel, appointed pursuant to 28 U.S.C.A. § 2284(a) (West 1994).

Due to the unusual complexity surrounding this type of case, and the skills and expertise which it requires, the court provided early notice of its intention to appoint a technical advisor to assist the court in understanding and utilizing the relevant technology needed to issue an order in the requisite expedited time frame. See Reilly v. United States, 863 F.2d 149, 154-56 (1st Cir.1988). Specifically, the court notified all counsel of its intent to appoint Mr. Bobby Bowers, Director of the South Carolina Budget and Control Board Office of Research and Statistics, as its technical advisor.^[3] The parties were also asked to advise the court of any objections to the proposed appointment. No such objections were interposed and on November 26, 2001, Mr. Bowers was appointed in this capacity by order of this court.

Following expedited discovery, trial in this matter commenced on January 14, 2002. The trial was conducted in three phases, with the House phase of the litigation commencing on January 14 and concluding on January 18, 2002. Following a week-long recess, trial on the Congressional phase ran from January 29 to January 31, 2002, and trial on the Senate phase ran from February 1 to February 12, 2002. Having considered the voluminous evidence and testimony presented in these cases, we agree that the requested relief is necessary. The existing plans for each elective body are unconstitutional, necessitating the court's implementation of courtordered remedial plans in their place.

II. Remedial Authority

The results of the 2000 census revealed that South Carolina's population has increased from 3,486,703 persons to 4,012,012 persons over the past decade, for an overall population growth of 15.1%. The percentage of the state's black population has remained relatively constant at just under 30% of the total population. The largest population growth occurred in the coastal areas of South Carolina — Horry, Beaufort, Georgetown, and Jasper Counties. With the exception of Greenville County, the urban counties of the state (Charleston, Greenville, Richland, and Spartanburg) did not keep pace with the 15.1% state-wide growth rate, but areas surrounding some existing urban centers — Lexington County, Kershaw County, and Calhoun County surrounding Columbia; York County and Cherokee County south of Charlotte, North Carolina; and Edgefield County and Aiken County near the North Augusta/Aiken metropolitan area — did experience substantial growth. The rural counties were the hardest hit in population losses. Four counties — Allendale, Bamberg, Marlboro, and Union — lost population, and nine others — Chester, Darlington, Dillon, Fairfield, Lee, Marion, Newberry, Orangeburg, and Williamsburg — did not keep pace with the 15.1% state growth rate.

These large population shifts from the rural areas to the urban and coastal areas of the state have resulted in severe malapportionment of both houses of the state legislature and of the United States Congressional districts. The parties have all stipulated that the existing plans for all three bodies are unconstitutionally malapportioned after the 2000 census, see, e.g., Karcher, 462 U.S. at 730-31, 103 S.Ct. 2653; Reynolds, 377 U.S. at 568, 84 S.Ct. 1362, and the court granted summary judgment as to the malapportionment issue in advance of each phase of the trial.

The primary responsibility for drafting and implementing a redistricting plan in South Carolina always rests with the South Carolina General Assembly, subject to the approval of the Governor. See U.S. Const. art. I, § 2; S.C. Const. art. III, § 3. "[J]udicial relief becomes appropriate only when a legislature fails to reapportion according to federal constitutional requisites in a timely fashion after having had an adequate opportunity to do so." White v. Weiser, 412 U.S. 783, 794-95, 93 S.Ct. 2348, 37 L.Ed.2d 335 (1973) (quoting Reynolds, 377 U.S. at 586, 84 S.Ct. 1362).

[L]egislative reapportionment is primarily a matter for legislative consideration and determination, for a state legislature is the institution that is by far the best situated to identify and then reconcile traditional state policies within the constitutionally mandated framework of substantial population equality. The federal courts by contrast possess no distinctive mandate to compromise sometimes conflicting state apportionment policies in the people's name. In the wake of a legislature's failure constitutionally to reconcile these conflicting state and federal goals, however, a federal court is left with the unwelcome obligation of performing in the legislature's stead, while lacking the political authoritativeness that the legislature can bring to the task. In such circumstances, the court's task is inevitably an exposed and sensitive one that must be accomplished circumspectly, and in a manner free from any taint of arbitrariness or discrimination. Connor, 431 U.S. at 414-15, 97 S.Ct. 1828 (citations and internal quotation marks omitted).

Unfortunately, the elected officials of South Carolina have failed to redistrict the General Assembly and South Carolina's Congressional seats in accordance with their constitutional obligations. Primary elections for Congress and the South Carolina House of Representatives are currently scheduled for June 11, 2002, with the general election to follow on November 5, 2002. All of the parties appearing before us have stipulated, and have presented persuasive evidence, that the General Assembly is at an impasse with the Governor and that there is no chance that the governing officials will reach a compromise in time for the impending 2002 elections.

We too are satisfied that the governing officials cannot complete the requisite redistricting in time for the impending elections. Accordingly, the court is required to take steps "to insure that no further elections are conducted under the invalid plan[s]," Reynolds, 377 U.S. at 585, 84 S.Ct. 1362, and to implement remedial districting plans, see Connor, 431 U.S. at 415, 97 S.Ct. 1828.^[4]

III. Remedial Factors

Although the court "perform[s] in the legislature's stead" when the latter has failed to redistrict in accordance with the Constitution, see id., we in fact operate under more stringent requirements than those imposed upon the state legislature. Accordingly, we begin with a summary of the specific redistricting standards under which we operate, each of which is discussed in more detail later.

First, we look to the federal standards which guide us. Because the constitutional wrong we remedy is the malapportionment of the existing districts, the one-person, one-vote requirement of the United States Constitution is always the paramount concern of a court-ordered remedial plan. The plan for redistricting of a state's bicameral legislature "must ordinarily achieve the goal of population equality with little more than de minimis variation," Chapman v. Meier, 420 U.S. 1, 27, 95 S.Ct. 751, 42 L.Ed.2d 766 (1975), whereas in the congressional redistricting process, a "good-faith effort to achieve precise mathematical equality" is required. Karcher, 462 U.S. at 730, 103 S.Ct. 2653 (internal quotation marks omitted).^[5] And, in fashioning these constitutionally mandated equipopulous plans, the court must comply with the racial-fairness mandates of § 2 of the Voting Rights Act, 42 U.S.C.A. § 1973, and the purpose-or-effect standards of § 5 of the Voting Rights Act, 42 U.S.C.A. § 1973c. See Abrams v. Johnson, 521 U.S. 74, 90, 96, 117 S.Ct. 1925, 138 L.Ed.2d 285 (1997).

Second, in satisfying these federally mandated requisites, we look to the historical redistricting policies of the state, but only insofar as "those policies do not lead to violations of the Constitution or the Voting Rights Act." Abrams, 521 U.S. at 79, 117 S.Ct. 1925; see also Upham v. Seamon, 456 U.S. 37, 39, 102 S.Ct. 1518, 71 L.Ed.2d 725 (1982) (noting that "[a]lthough a court must defer to legislative judgments on reapportionment as much as possible, it is forbidden to do so when the legislative plan would not meet the special standards of population equality and racial fairness that are applicable to court-ordered plans"); White, 412 U.S. at 795, 93 S.Ct. 2348 (noting that federal courts, in the context of reapportionment, "should follow the policies and preferences of the State, as expressed in statutory and constitutional provisions or in the reapportionment plans proposed by the state legislature, whenever adherence to state policy does not detract from the requirements of the Federal Constitution").

Finally, we reiterate the Supreme Court's admonition that, in implementing a court-ordered remedial plan for malapportioned legislative and congressional districts, the court must always act "circumspectly, and in a manner free from any taint of arbitrariness or discrimination." Connor, 431 U.S. at 415, 97 S.Ct. 1828 (internal quotation marks omitted). Federal courts, unlike state legislatures, are not in a position to reconcile conflicting state policies on the electorate's behalf, nor at liberty to engage in political policymaking decisions. See id. We are limited to correcting the unconstitutional aspects of the state's plan while complying with the Voting Rights Act requirements. To the extent we can, we will also follow traditional state districting principles, but we do not possess the latitude afforded a state legislature to advance political agendas.

To assist us in our task of implementing a remedial districting plan for each elective body, the parties have presented proposed plans for the two bodies of the South Carolina General Assembly and for the six congressional districts, which they urge us to either adopt in toto or, at a minimum, to use as templates in formulating our own plans. Indeed, in all phases of the trial, the bulk of the evidence presented was devoted to the parties' various redistricting proposals and, in particular, the issue of which of the proposed plans best complies with the one-person, one-vote requirement of the United States Constitution and with the mandates of §§ 2 and 5 of the Voting Rights Act.

We have carefully reviewed the various plans offered by the parties to this litigation, and the evidence presented in support of each, and we are satisfied that none are acceptable for wholesale adoption as a judicially approved redistricting plan. Among a myriad of individual deficiencies, the evidence overwhelmingly demonstrates one fatal problem that permeates them all. The General Assembly and the Governor, being controlled by and a member of, respectively, two different political parties, have proposed plans that are primarily driven by policy choices designed to effect their particular partisan goals. And, in many cases, the choices appear to be reflective of little more than an individual legislator's desire to strengthen his or her ability to be re-elected to the seat in question.

Simply stated, the General Assembly, in which Republicans hold a majority in both bodies, passed plans that the majority of its members believed were favorable to them, and the incumbent Governor, a Democrat, vetoed those plans in order to advocate the implementation of alternative plans that are favorable to the views of his political party and its legislative and congressional members. The Leatherman plaintiffs, all of whom are currently members of the Republican Party, have proposed plans that both advance the Republican Party's agenda in general and strengthen their individual seats in particular. And, the Marcharia plaintiffs, represented by the American Civil Liberties Union ("ACLU"), did not advance plans of their own, but have either endorsed the plans offered by the General Assembly, or suggested modifications of their own, which they feel best serve the interests of black voters in South Carolina.^[6]

Such is the political process. We find it interesting, but are mindful that none of the plans carry the imprimatur of the state and that, whatever merit exists for drawing a plan for political reasons or based upon a particular political ideology,^[7] it does not exist for us. While all parties are entitled to advocate a legislative redistricting plan that furthers their partisan interests, it is inappropriate for the court to engage in political gerrymandering.^[8]

Accordingly, the court begins its remedial task of redistricting the malapportioned districts on a equipopulous basis and in the racially fair manner mandated by the Voting Rights Act. To the extent we can simultaneously observe districting principles demonstrated to have been of traditional concern to South Carolina in past redistricting, without compromising our obligations under federal law, we attempt to do so.^[9] We are, however, confident in the knowledge that those who disagree with the plans may seek alternative ones from those in the most appropriate position to redraw them — the General Assembly and Governor of the State of South Carolina. And, because they are indeed in a better position to address the needs of South Carolina and her citizens, we encourage these elected officials, and those who follow in their footsteps, to renew and continue efforts to fulfill their constitutional duties in this respect.

A. Equality of Population and Acceptable Variances; the "One-Person, One-Vote" Requirement of the United States Constitution.

1. Congressional Reapportionment

The origin of the constitutional guarantee of "one-person, one-vote" for the election of congressional representatives is found in Article I, § 2 of the United States Constitution. In relevant part, it provides that "[t]he House of Representatives shall be composed of Members chosen every second Year by the People of the several States" and that such "Representatives ... shall be apportioned among the several States which may be included within this Union, according to their respective Numbers." U.S. Const. art. I, § 2.

The provision sets "a high standard of justice and common sense for the apportionment of congressional districts: equal representation for equal numbers of people." Karcher, 462 U.S. at 730, 103 S.Ct. 2653 (internal quotation marks omitted). "[C]onstrued in its historical context," the Supreme Court has held, "the command ... that Representatives be chosen `by the People of the several States' means that as nearly as is practicable one man's vote in a congressional election is to be worth as much as another's." Wesberry v. Sanders, 376 U.S. 1, 7-8, 84 S.Ct. 526, 11 L.Ed.2d 481 (1964) (emphasis added). This standard is a demanding one, "requir[ing] that the State make a good-faith effort to achieve precise mathematical equality." Karcher, 462 U.S. at 730, 103 S.Ct. 2653 (internal quotation marks omitted); see also White, 412 U.S. at 790, 93 S.Ct. 2348. "[A]bsolute population equality [is] the paramount objective ... in the case of congressional districts, for which the command of Art. I, § 2, as regards the National Legislature outweighs the local interests that a State may deem relevant in apportioning districts for representatives to state and local legislatures...." Karcher, 462 U.S. at 732-33, 103 S.Ct. 2653. "Unless population variances among congressional districts are shown to have resulted despite [a good-faith] effort, the State must justify each variance, no matter how small." Kirkpatrick v. Preisler, 394 U.S. 526, 531, 89 S.Ct. 1225, 22 L.Ed.2d 519 (1969). "[T]here are no de minimis population variations, which could practicably be avoided, but which nonetheless meet the standard of Art. I, § 2, without justification." Karcher, 462 U.S. at 734, 103 S.Ct. 2653.

Small population deviations may be justified by legitimate state policies such as "making districts compact, respecting municipal boundaries, preserving the cores of prior districts, and avoiding contests between incumbent Representatives," if the state policies are consistent with constitutional norms. Karcher, 462 U.S. at 740, 103 S.Ct. 2653. "As long as the criteria are nondiscriminatory, these are all legitimate objectives that on a proper showing could justify minor population deviations." Id. (emphasis added) (internal citation omitted). However, a state must "show with some specificity that a particular objective required the specific deviations in its plan, rather than simply relying on general assertions." Id. (noting that differences in the numbers of eligible voters and projected population shifts might justify small variations in congressional district populations).

2. Redistricting of the South Carolina General Assembly

The Equal Protection Clause of the United States Constitution requires that the seats in both houses of a bicameral legislature — here the South Carolina Senate and House of Representatives — also be apportioned on an equipopulous basis. See Reynolds, 377 U.S. at 568, 84 S.Ct. 1362. This one-person, one-vote principle, established by the Court in Reynolds and grounded in equal protection, ensures that each person's vote receives equal weight. See id.

In all cases, "the overriding objective [of redistricting] must be substantial equality of population among the various [legislative] districts." Id. at 579, 84 S.Ct. 1362. States, however, are permitted somewhat more flexibility with respect to state legislative apportionment than in congressional districting in order to pursue other legitimate state policies, including "maintain[ing] the integrity of various political subdivisions," id. at 578, 84 S.Ct. 1362, "provid[ing] for compact districts of contiguous territory," id., and recognizing "natural or historical boundary lines," Swann v. Adams, 385 U.S. 440, 444, 87 S.Ct. 569, 17 L.Ed.2d 501 (1967). Generally, a "safe harbor" exists for legislatively implemented plans achieving less than a 10% deviation. See Voinovich v. Quilter, 507 U.S. 146, 161, 113 S.Ct. 1149, 122 L.Ed.2d 500 (1993) (noting that a plan with a maximum deviation under 10% is generally considered to fall within the category of minor deviations); see also Connor, 431 U.S. at 418, 97 S.Ct. 1828; White v. Regester, 412 U.S. 755, 764, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973).

Court-ordered remedial plans for bicameral state legislative bodies, in contrast, are held to a much more stringent standard of population equality. See Chapman, 420 U.S. at 26, 95 S.Ct. 751. Although not required to attain the mathematical preciseness required of courts and state legislatures for congressional redistricting, see id. at 27 n. 19, 95 S.Ct. 751, a court-ordered plan "must ordinarily achieve the goal of population equality with little more than de minimis variation," id. at 27, 95 S.Ct. 751; see also Abrams, 521 U.S. at 98, 117 S.Ct. 1925; Connor, 431 U.S. at 414, 97 S.Ct. 1828. "[A]ny deviation from approximate population equality must be supported by enunciation of historically significant state policy or unique features." Chapman, 420 U.S. at 26, 95 S.Ct. 751.

B. The Voting Rights Act

In exercising our equitable power to redistrict, we must also comply with the requirements imposed upon states by the racial-fairness mandates of § 2 of the Voting Rights Act, 42 U.S.C.A. § 1973, and the purpose-or-effect standards of § 5 of the Voting Rights Act, 42 U.S.C.A. § 1973c. See Abrams, 521 U.S. at 90, 96, 117 S.Ct. 1925, 138 L.Ed.2d 285 (1997). Indeed, the number and composition of majority-minority districts in each proposed plan accounted for the lion's share of the evidence in every phase of this case.^[10]

1. Section 2: The Prohibition Against Minority Vote Dilution

Section 2 of the Voting Rights Act of 1965, 42 U.S.C.A. § 1973, was enacted by Congress "to help effectuate the Fifteenth Amendment's guarantee that no citizen's right to vote shall `be denied or abridged ... on account of race, color, or previous condition of servitude.'" Voinovich, 507 U.S. at 152, 113 S.Ct. 1149 (alteration in original) (quoting U.S. Const. amend. XV). Specifically, § 2 provides that "[n]o voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color." 42 U.S.C.A. § 1973(a) (West 1994). A violation of § 2 "is established if, based 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 protected class of citizens] 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." 42 U.S.C.A. § 1973(b) (West 1994) (emphasis added).^[11]

Simply stated, § 2 prohibits the implementation of an electoral law that "interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by black and white voters to elect their preferred representatives." Thornburg v. Gingles, 478 U.S. 30, 47, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986); see also Voinovich, 507 U.S. at 155, 113 S.Ct. 1149 (explaining that § 2 "focuses exclusively on the consequences of apportionment. Only if the apportionment scheme has the effect of denying a protected class the equal opportunity to elect its candidate of choice does it violate § 2.").

In Gingles, the Supreme Court set forth the specific framework for evaluating a § 2 vote dilution claim.^[12] Three preconditions must first be established: (1) that the minority group "is sufficiently large and geographically compact to constitute a majority in a single-member district"; (2) that the minority group is "politically cohesive"; and (3) that the majority "votes sufficiently as a bloc to enable it ... usually to defeat the minority's preferred candidate." Id. at 50-51, 106 S.Ct. 2752. If these factors are established, "the court considers whether, `on the totality of circumstances,' minorities have been denied an `equal opportunity' to `participate in the political process and to elect representatives of their choice.'" Abrams, 521 U.S. at 91, 117 S.Ct. 1925 (quoting 42 U.S.C.A. § 1973(b)).^[13]

The voting strength of a politically cohesive minority group may be diluted by fragmenting the minority voters among districts:

In the context of single-member districts, the usual device for diluting minority voting power is the manipulation of district lines. A politically cohesive minority group that is large enough to constitute the majority in a single-member district has a good chance of electing its candidate of choice, if the group is placed in a district where it constitutes a majority. Dividing the minority group among various districts so that it is a majority in none may prevent the group from electing its candidate of choice: If the majority in each district votes as a bloc against the minority candidate, the fragmented minority group will be unable to muster sufficient votes in any district to carry its candidate to victory.

Voinovich, 507 U.S. at 153, 113 S.Ct. 1149 (emphasis added). Or, minority voting strength may be diluted by concentrating minority voters within a district:

How such concentration or "packing" may dilute minority voting strength is not difficult to conceptualize. A minority group, for example, might have sufficient numbers to constitute a majority in three districts. So apportioned, the group inevitably will elect three candidates of its choice, assuming the group is sufficiently cohesive. But if the group is packed into two districts in which it constitutes a super-majority, it will be assured only two candidates. As a result, we have recognized that "dilution of racial minority group voting strength may be caused" either "by the dispersal of blacks into districts in which they constitute an ineffective minority of voters or from the concentration of blacks into districts where they constitute an excessive majority."

Id. at 153-54, 113 S.Ct. 1149 (internal alteration omitted) (quoting Gingles, 478 U.S. at 46, 106 S.Ct. 2752).

Although "[o]n its face, § 2 does not apply to a court-ordered remedial redistricting plan ..., courts should comply with the section when exercising their equitable powers to redistrict." Abrams, 521 U.S. at 90, 117 S.Ct. 1925. Thus, to prevail on a § 2 claim, a plaintiff must demonstrate that, under the totality of the circumstances, the election scheme enacted by a legislature dilutes the minority's voting strength by either unnecessarily fragmenting or packing a politically cohesive minority population. See Voinovich, 507 U.S. at 157, 113 S.Ct. 1149; Gingles, 478 U.S. at 46, 106 S.Ct. 2752. In the context of preparing a court-ordered redistricting plan, we interpret the Supreme Court's command as requiring us to consider the Gingles test as we draw, paying particular attention to those areas of the state where a "politically cohesive minority group ... is large enough to constitute the majority in a single-member district." Voinovich, 507 U.S. at 153, 113 S.Ct. 1149. In this way, we operate much like the state legislature should, taking steps to ensure that our plan will not have the effect, albeit unintended, of diluting the voting strength of that majority-minority population. See Abrams, 521 U.S. at 90-91, 117 S.Ct. 1925.

2. Section 5: Avoiding Retrogression

Section 5 of the Voting Rights Act likewise requires us to consider race in our draw. This provision requires certain covered jurisdictions — those with a history of discrimination against minorities — to obtain either administrative preclearance by the Attorney General or approval by the United States District Court for the District of Columbia for any change in a "standard, practice, or procedure with respect to voting." 42 U.S.C.A. § 1973c. To obtain preclearance, the covered jurisdiction must demonstrate that the proposed change has neither the purpose nor effect "of denying or abridging the right to vote on account of race or color." 42 U.S.C.A. § 1973c; see McDaniel v. Sanchez, 452 U.S. 130, 149, 101 S.Ct. 2224, 68 L.Ed.2d 724 (1981).

Enacted as "a response to a common practice in some jurisdictions of staying one step ahead of the federal courts by passing new discriminatory voting laws as soon as the old ones had been struck down," Beer v. United States, 425 U.S. 130, 140, 96 S.Ct. 1357, 47 L.Ed.2d 629 (1976) (internal quotation marks omitted), § 5 "shift[s] the advantage of time and inertia from the perpetrators of the evil to its victim, by freezing election procedures in the covered areas unless the changes can be shown to be nondiscriminatory," id. (emphasis added; internal quotation marks omitted). "[T]he purpose of § 5 has always been to insure that no voting-procedure changes would be made that would lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise." Id. at 141, 96 S.Ct. 1357 (emphasis added).

Although they share the goal of addressing the evils of racial discrimination, § 5 differs in important respects from § 2 of the Voting Rights Act. First, § 2 applies to all states, whereas § 5 applies only to those states, such as South Carolina, which have the worst reputation for historical and ongoing discrimination against blacks. Second, by requiring the creation of majority-minority districts where vote dilution under the Gingles test would occur, § 2 looks beyond the status quo to ensure that a redistricting plan affords blacks an equal opportunity to elect the representatives of their choice as white voters enjoy. See 42 U.S.C.A. § 1973(b); Gingles, 478 U.S. at 47, 106 S.Ct. 2752. Section 5, in contrast, maintains the status quo. It only prevents "backsliding" in those jurisdictions subject to its requirements by prohibiting the implementation of any proposed voting change that has been enacted for a retrogressive purpose, see Reno v. Bossier Parish Sch. Bd., 528 U.S. 320, 335, 120 S.Ct. 866, 145 L.Ed.2d 845 (2000) ("Bossier Parish II"), or that has a retrogressive effect on minority voting strength, see Beer, 425 U.S. at 141, 96 S.Ct. 1357. Section 5 requires no separate inquiry into the Gingles factors to determine whether an opportunity district must be created, but "mandates that the minority's [existing] opportunity to elect representatives of its choice not be diminished, directly or indirectly, by the State's actions." Bush v. Vera, 517 U.S. 952, 983, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996).

The distinction between § 2 and § 5 was expounded upon by the Supreme Court in Reno v. Bossier Parish School Board, 520 U.S. 471, 117 S.Ct. 1491, 137 L.Ed.2d 730 (1997) ("Bossier Parish I"). There, the Court struck down the Department of Justice's denial of preclearance under § 5 on the basis that the plan was, in the opinion of the Department of Justice, violative of § 2. See id. at 477, 117 S.Ct. 1491. This position, the Court held, would "make compliance with § 5 contingent upon compliance with § 2," and would effectively "replace the standards for § 5 with those for § 2." Id. The Court explained that such an interpretation would contradict its "longstanding recognition that the two provisions were designed to combat different evils, and, accordingly, to impose very different duties upon the States." Id.

Thus, in the redistricting context, Section 5 has a quite limited substantive goal — to insure that the state does not adopt a redistricting plan that, when compared to the existing or "benchmark" plan, "would lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise." Beer, 425 U.S. at 141, 96 S.Ct. 1357; see Holder v. Hall, 512 U.S. 874, 883, 114 S.Ct. 2581, 129 L.Ed.2d 687 (plurality opinion) ("Under § 5, then, the proposed voting practice is measured against the existing voting practice to determine whether retrogression would result from the proposed change."); Bossier Parish I, 520 U.S. at 478, 117 S.Ct. 1491 (explaining that "the jurisdiction's existing plan is the benchmark against which the `effect' of voting changes is measured"). Thus,

[i]n § 5 preclearance proceedings — which uniquely deal only and specifically with changes in voting procedures — the baseline is the status quo that is proposed to be changed: If the change "abridges the right to vote" relative to the status quo, preclearance is denied, and the status quo (however discriminatory it may be) remains in effect.

Bossier Parish II, 528 U.S. at 334, 120 S.Ct. 866.

"Section 2, on the other hand, was designed as a means of eradicating voting practices that minimize or cancel out the voting strength and political effectiveness of minority groups." Bossier Parish I, 520 U.S. at 479, 117 S.Ct. 1491 (internal quotation marks omitted). It involves changes to the status quo itself. See Bossier Parish II, 528 U.S. at 334, 120 S.Ct. 866. Because a determination of vote dilution requires "the existence of an `undiluted' practice against which the fact of dilution may be measured, a § 2 plaintiff must also postulate a reasonable alternative voting practice to serve as the benchmark `undiluted' voting practice." Bossier Parish I, 520 U.S. at 480, 117 S.Ct. 1491. Thus, "the comparison must be made with a hypothetical alternative: If the status quo results in an abridgement of the right to vote or abridges the right to vote relative to what the right to vote ought to be, the status quo itself must be changed." Bossier Parish II, 528 U.S. at 334, 120 S.Ct. 866 (alterations and internal quotation marks omitted).

Because South Carolina is a covered jurisdiction under the Voting Rights Act, any redistricting plan that it adopts is subject to the preclearance requirements of § 5. See Allen v. State Bd. of Elections, 393 U.S. 544, 548-49, 89 S.Ct. 817, 22 L.Ed.2d 1 (1969). A legislative plan still reflects "the exercise of legislative judgment" and "the policy choices of the elected representatives of the people"; such plans, therefore, remain subject to preclearance under § 5 even if later reviewed by a court. See McDaniel, 452 U.S. at 153, 101 S.Ct. 2224.

Generally, "a decree of the United States District Court is not within reach of Section 5 of the Voting Rights Act such that it must be precleared," Abrams, 521 U.S. at 95, 117 S.Ct. 1925 (internal quotation marks and alterations omitted), but "[t]he exception applies only to judicial plans devised by the court itself," not to legislative plans which are submitted to the court by the parties for consideration, id.; see Lopez v. Monterey County, 525 U.S. 266, 286, 119 S.Ct. 693, 142 L.Ed.2d 728 (1999). Because we have independently devised plans for each elective body from scratch, however, none of the plans implemented by this court order are subject to the preclearance requirements of § 5 and, accordingly, must be immediately implemented for the upcoming elections.

Even where the court elects to devise its own plan, such that it need not be precleared, the court must still "`follow the appropriate § 5 standards, including the body of administrative and judicial precedents developed in Section 5 cases.'" Abrams, 521 U.S. at 96, 117 S.Ct. 1925 (quoting McDaniel, 452 U.S. at 149, 101 S.Ct. 2224). Thus, like state legislatures, we must ensure that our redistricting plan does not "lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise" in South Carolina. Beer, 425 U.S. at 141, 96 S.Ct. 1357.

3. Equal Protection Constraints Upon The Voting Rights Act Requirements; the Limits of Racial Gerrymandering.

In complying with the Voting Rights Act in the redistricting context, we must also remain cognizant of the limitations the Fourteenth Amendment of the United States Constitution imposes upon the permissible use of race in the process. The Equal Protection Clause of the United States Constitution requires racial neutrality in governmental decision-making. See U.S. Const., amend. XIV, § 1 (providing that no State shall "deny to any person within its jurisdiction the equal protection of the laws"). And, the Supreme Court has repeatedly held that dividing voters according to their race in the redistricting context is subject to the strictures of the Equal Protection Clause. See Shaw v. Hunt, 517 U.S. 899, 904-05, 116 S.Ct. 1894, 135 L.Ed.2d 207 (1996) ("Shaw II"); Miller v. Johnson, 515 U.S. 900, 905, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995); Shaw v. Reno, 509 U.S. 630, 644, 113 S.Ct. 2816, 125 L.Ed.2d 511 (1993) ("Shaw I").

By prohibiting election plans that have the effect of diluting minority voting strength or that lead to retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise, however, the Voting Rights Act necessarily forces states to consider race in the redistricting context, placing the Act in obvious tension with the Equal Protection Clause. The end result is that the Voting Rights Act must always be considered in tandem with the strictures of the Equal Protection Clause, with the latter operating as a constant limit upon the degree to which state legislatures — and this court acting in its remedial capacity — can engage in race-based districting to achieve the goals of the Voting Rights Act.

Race-based classifications, of course, are generally subjected to strict scrutiny and will pass constitutional muster only if the state action is narrowly tailored to achieve a compelling state interest. See, e.g., Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 224, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995). In this case, however, we have no occasion to apply strict scrutiny, at least not in the usual way. Strict scrutiny is a particularly stringent level of judicial review applied to determine the constitutionality of a law or other action already taken by the state. Because the General Assembly failed to implement redistricting plans, we have no pre-existing plans to which strict scrutiny could be applied. Instead, we are drawing the plans ourselves, not reviewing plans drawn by the General Assembly. Nonetheless, because our draw is guided by similar limitations as those imposed upon states in the redistricting context, the purposes of a strict scrutiny review are of importance in this case as well.

a. When does the consideration of race trigger strict scrutiny analysis?

The Equal Protection Clause, as interpreted by Adarand, would seem to compel strict scrutiny of any governmental classification based on race. However, the Court has held that strict scrutiny in the redistricting context only applies where "race was the predominant factor motivating the legislature's decision to place a significant number of voters within or without a particular district." Miller, 515 U.S. at 916, 115 S.Ct. 2475. Because a state must observe the Voting Rights Act requirements when redistricting its legislative and congressional districts, such redistricting decisions necessarily differ from other types of governmental decision-making:

[T]he legislature is always aware of race when it draws district lines, just as it is aware of age, economic status, religious and political persuasion, and a variety of other demographic facts. That sort of race consciousness does not lead inevitably to impermissible race discrimination.... [W]hen members of a racial group live in one community, a reapportionment plan that concentrates members of the group in one district and excludes them from others may reflect wholly legitimate purposes. The district lines may be drawn, for example, to provide for compact districts of contiguous territory, or to maintain the integrity of political subdivisions.

Shaw I, 509 U.S. at 646, 113 S.Ct. 2816. So long as race is only a consideration, and other traditional districting principles such as compactness and contiguity are not subordinated to it, the challenged majority-minority district is not subject to strict scrutiny analysis.

The question of when race predominates over nonracial factors, thereby triggering strict scrutiny, can be difficult to answer. In cases arising outside the redistricting context, "statutes are subject to strict scrutiny under the Equal Protection Clause not just when they contain express racial classifications, but also when, though race neutral on their face, they are motivated by a racial purpose or object." Miller, 515 U.S. at 913, 115 S.Ct. 2475; see also id. at 920, 115 S.Ct. 2475 ("[W]here the State assumes from a group of voters' race that they think alike, share the same political interests, and will prefer the same candidates at the polls, it engages in racial stereotyping at odds with equal protection mandates." (internal quotation marks omitted)). These principles also apply in redistricting cases. See id. at 914, 115 S.Ct. 2475.

Strict scrutiny of a districting plan may also be triggered by an allegation that the state "adopted a reapportionment scheme so irrational on its face that it can be understood only as an effort to segregate voters into separate voting districts because of their race." Shaw I, 509 U.S. at 658, 113 S.Ct. 2816; Miller, 515 U.S. at 916, 115 S.Ct. 2475. Thus, "appearances do matter" in the area of redistricting. Shaw I, 509 U.S. at 647, 113 S.Ct. 2816. "Shape is relevant not because bizarreness is a necessary element of the constitutional wrong or a threshold requirement of proof, but because it may be persuasive circumstantial evidence that race for its own sake, and not other districting principles, was the legislature's dominant and controlling rationale in drawing its district lines." Miller, 515 U.S. at 913, 115 S.Ct. 2475.^[14]

A plaintiff making an equal protection challenge to a redistricting plan must "show, either through circumstantial evidence of a district's shape and demographics or more direct evidence going to legislative purpose, that race was the predominant factor motivating the legislature's decision to place a significant number of voters within or without a particular district." Miller, 515 U.S. at 916, 115 S.Ct. 2475.

To make this showing, a plaintiff must prove that the legislature subordinated traditional race-neutral districting principles, including but not limited to compactness, contiguity, and 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."

Id. (quoting Shaw I, 509 U.S. at 647, 113 S.Ct.2816). In sum, "compliance with traditional districting principles such as compactness, contiguity, and respect for political subdivisions may well suffice to refute a claim of racial gerrymandering," but does not suffice where "those factors [a]re subordinated to racial objectives." Id. at 919, 115 S.Ct. 2475.^[15]

b. When race predominates in the redistricting context.

The fact that race has predominated in the decision-making process, however, does not automatically render the district constitutionally infirm. Rather, "[i]f race is the predominant motive in creating districts, strict scrutiny applies and the districting plan must be narrowly tailored to serve a compelling governmental interest" to pass constitutional muster. Abrams, 521 U.S. at 91, 117 S.Ct. 1925; see also Miller, 515 U.S. at 920, 115 S.Ct. 2475.^[16] And, because compliance with the Voting Rights Act is a compelling state interest, states are afforded a very narrow ability to further that interest through the consideration of race.^[17] If there is "a strong basis in evidence for concluding that creation of a majority-minority district is reasonably necessary to comply" with the Act, and the race-based districting substantially addresses the violation, the plan will not fail under the Equal Protection analysis. See Bush, 517 U.S. at 977, 116 S.Ct. 1941 (citations and internal quotation marks omitted); see also Abrams, 521 U.S. at 91, 117 S.Ct. 1925.

Of course, to be narrowly tailored to achieve the compelling interest of § 2, there must be evidence that the Gingles test is met in the first instance. Section 2 "does not require a state to create, on predominately racial lines, a district that is not reasonably compact. And the § 2 compactness inquiry should take into account traditional districting principles such as maintaining communities of interest and traditional boundaries." Abrams, 521 U.S. at 91, 117 S.Ct. 1925 (emphasis added; internal quotation marks and citations omitted); see also Bush, 517 U.S. at 977, 116 S.Ct. 1941 ("A § 2 district that is reasonably compact and regular, taking into account traditional districting principles such as maintaining communities of interest and traditional boundaries, may pass strict scrutiny without having to defeat rival compact districts designed by plaintiffs' experts in endless `beauty contests.'").

In preparing a redistricting plan in its remedial authority, the court must also comply with §§ 2 and 5 of the Voting Rights Act and, where necessary, narrowly tailor the draw to do no more than is necessary to serve these compelling state interests. The task requires a balancing of all the relevant inquiries. Avoiding minority vote dilution under § 2 may require the creation or maintenance of particular majority-minority districts, even on a predominately racial basis, if this can be done in a reasonably compact manner, taking into account traditional districting principles. Avoiding vote dilution does not, of course, require the creation of a majority-minority district wherever it can be done; indeed, equal protection forbids such a course. See Miller, 515 U.S. at 926-27, 115 S.Ct. 2475 (rejecting the Department of Justice's position that § 2 requires the "maximization" of minority voting strength). Bizarrely shaped districts are not "narrowly tailored to serve the State's interest in avoiding liability under § 2, because § 2 does not require a State to create, on predominately racial lines, a district that is not reasonably compact." Bush, 517 U.S. at 979, 116 S.Ct. 1941 (internal quotation marks omitted).

Nor does § 5 compel the maximization of black voters in the district. See Miller, 515 U.S. at 927, 115 S.Ct. 2475 ("[T]he Justice Department's implicit command that States engage in presumptively race-based districting brings the Act ... into tension with the Fourteenth Amendment.... There is no indication Congress intended such a far reaching application of § 5, so we reject the Justice Department's interpretation of the statute...." (emphasis added)). To achieve nonretrogression in the position of racial minorities, we are permitted to use race as a consideration in the draw, but must narrowly tailor the remedy. "A reapportionment plan would not be narrowly tailored to the goal of avoiding retrogression if the State went beyond what was reasonably necessary to avoid retrogression." Bush, 517 U.S. at 983, 116 S.Ct. 1941 (internal quotation marks omitted).

In sum, where the Voting Rights Act requires that a majority-minority district be drawn or maintained, there are points in the drawing of the district where race must predominate, in the sense that we choose to draw the line in one particular direction over another because of race, though either direction would be consistent with traditional districting principles. In these limited situations, race might be said to predominate in the decision-making process at a particular moment, but never operates to the exclusion of all other districting principles, and even then we are careful to narrowly tailor our draw so that this predominate use of race is limited to the accomplishment of the purposes of the Voting Rights Act and no more.

4. Section 2 Considerations: Racially Polarized Voting in South Carolina and the Point of Equal Opportunity.

In conducting our § 2 inquiry, we must first evaluate the presence of the Gingles factors. We may not assume that racial bloc voting and minority-group political cohesion exists in a particular area of the state; rather, these factors must be proven in each case. See Shaw I, 509 U.S. at 653, 113 S.Ct. 2816.

The history of racially polarized voting in South Carolina is long and well-documented — so much so that in 1992, the parties in Burton stipulated that "since 1984 there is evidence of racially polarized voting in South Carolina." Burton, 793 F.Supp. at 1357-58. The three-judge court in Smith made a similar finding:

In South Carolina, voting has been, and still is, polarized by race. This voting pattern is general throughout the state... There is only one exception according to Defendants' expert, Dr. Ruoff, who has studied the voting history of South Carolina for a number of years. He testified, "Whites almost always vote for whites and blacks almost always vote for blacks unless the candidate is a black Republican and then never."

Smith, 946 F.Supp. at 1202-03. The court also noted the wide socio-economic gap between white and black voters, as evidenced by, among other things, the disparity in unemployment rates, household income, education, and access to private transportation. See id. at 1203.

In this case, the parties have presented substantial evidence that this disturbing fact has seen little change in the last decade. Voting in South Carolina continues to be racially polarized to a very high degree, in all regions of the state and in both primary elections and general elections. Statewide, black citizens generally are a highly politically cohesive group and whites engage in significant white-bloc voting. Indeed, this fact is not seriously in dispute.

Dr. John Ruoff, who was retained by the Southern Poverty Law Center for the purpose of conducting an extensive analysis of racially polarized voting patterns in South Carolina during the last ten years,^[18] has appeared again to testify regarding his findings and conclusions about the continued racially polarized voting in South Carolina. Specifically, he offered undisputed testimony that South Carolinians are still very divided in terms of where they live and that elections throughout South Carolina continue to be marked by very high levels of racial polarization in voting. Black voters are generally politically cohesive and white voters almost always vote in blocs to defeat the minority's candidate of choice. Racial polarization is highest in black-white elections — those involving a black candidate running against a white candidate. And, there is a well-documented hierarchy in the preference of black voters. With few exceptions, black voters demonstrate an overwhelming preference for black Democrats as their representatives, followed by white Democrats, particularly in a general election, but black voters virtually never vote for a Republican candidate. The other experts retained by the parties in this case substantially concurred in this portion of Dr. Ruoff's opinion, which was likewise supported by the statistical and other evidence presented to the court by all of the parties.

By way of summary, the evidence revealed that in black-white, single seat elections, the median level of black voters voting for black candidates was 98% in general elections and 86% in primary contests. Although white voters will cross over to vote for black candidates at a rate of 21% in general elections, they will cross over to vote for a black candidate in primary elections at a rate of only 8%. In addition, voter mobilization among blacks in general elections is lower than among white residents, but greater in black-majority districts.^[19]

In addition, the ACLU provided extensive documentation of the history of voting-related racial discrimination in South Carolina, which was submitted largely as a stipulation among the parties. Evidence of the depressed socio-economic and educational status of blacks in the state which hinders their ability to participate effectively in the political process and to elect representatives of their choice was also presented.^[20]

This evidence overwhelmingly demonstrates that the first two Gingles factors, necessary for the creation of majority-black legislative and congressional districts in areas where minorities are sufficiently large and geographically compact to constitute a majority in a single-member district, are present statewide. Minority voters are generally politically cohesive to a very high degree and, as a rule, the majority usually votes sufficiently as a bloc to defeat the minority's preferred candidate. See Gingles, 478 U.S. at 51, 106 S.Ct. 2752. Thus, we can and should consider race in each of our redistricting plans to ensure that they do not have the unintended effect of diluting the voting strength of a reasonably compact, majority-minority population.

The measure that must be used to ensure that the compact, majority-minority population actually has an "equal opportunity" to elect the candidate of its choice in a particular district, on the other hand, is often in sharp dispute. We measure equal opportunity by the percentage of minority voting age population necessary for the minority voters to elect the candidate of their choice, see Johnson v. Miller, 922 F.Supp. 1556, 1568 nn. 18 & 19 (S.D.Ga. 1995) (noting that minority voting age population is the appropriate measure for analyzing vote dilution and that the alternative use of black registered voter percentages condones voter apathy), aff'd sub nom. Abrams v. Johnson, 521 U.S. 74, 117 S.Ct. 1925, 138 L.Ed.2d 285 (1997), which seems to create little controversy. The actual percentage required in a particular district, and how that percentage is reached, are much different matters.^[21]

Dr. Ruoff testified on behalf of the General Assembly defendants on this subject as well. He testified that state-wide black voters are only able to elect candidates of their choice in majority-black or very near majority-black districts. In addition to the high levels of voter polarization, this is in part because black political mobilization is lower than white political mobilization overall, but higher in districts where black candidates have a strong opportunity to be elected. In other words, blacks in a majority-black district are aware of their opportunity and tend to register and turn out to vote in districts that are majority-black or very near majority-black. Dr. James W. Loewen, who testified on behalf of the Senate in the Senate phase of this litigation, generally concurs in this view as well.

David Epstein, an expert favoring the Governor's proposed House and Senate plans, does not share this view. Using a new technique called "probit analysis" (which he professes to have pioneered), Dr. Epstein testified that the point of equal opportunity for the state as a whole occurs at just over 47.11% BVAP. By region, Dr. Epstein opined that the Upstate's point of equal opportunity is 44.61%, the Midlands' is 47.45%, the Pee Dee's is 48.19%, and the Low Country's is 45.58%. Dr. Epstein also testified that once a district exceeds approximately 40% BVAP, the addition of more black votes does not alter the particular representative's voting patterns; Democrats of both races support the minority-favored position in General Assembly votes. This is because, according to Dr. Epstein, blacks now turn out in numbers roughly equivalent to whites, there is white cross-over voting at about 20%, and a black's tendency to vote against a minority candidate is only about 2%. For these reasons, Epstein testified, redistricting plans drawn to lower BVAP levels, by allowing for the draw of influence districts, will better advance the substantive representation of the black community in South Carolina.^[22]

Having carefully considered all of the expert testimony offered by the parties on these important issues, as well as the supporting documentation behind their views, we find that in order to give minority voters an equal opportunity to elect a minority candidate of choice as well as an equal opportunity to elect a white candidate of choice in a primary election in South Carolina, a majority-minority or very near majority-minority black voting age population in each district remains a minimum requirement.

Section 2 requires the court to consider whether, "`on the totality of circumstances,'" a plan will deny minority voters "an `equal opportunity' to `participate in the political process and to elect representatives of their choice.'" Abrams, 521 U.S. at 91, 117 S.Ct. 1925 (quoting 42 U.S.C.A. § 1973(b)). The evidence presented overwhelmingly demonstrates that, with rare exceptions, blacks currently prefer to be represented first by black Democrats. In the absence of either the choice, such as in a primary, or the opportunity, due to percentages too low to outvote a cohesive majority bloc vote, blacks prefer to be represented by white Democrats. Equal opportunity does not equate to a guarantee of success, but it does require an equal opportunity to elect the minority's candidate of choice which, in South Carolina, is almost always a minority candidate. In sum, the Voting Rights Act protects the minority voters' opportunity to elect their candidate of choice, not just a minority incumbent and not just the minority's opportunity to elect an incumbent of any race. Rather, a minority's equal opportunity to elect a candidate of choice must include the opportunity to elect a minority candidate of choice, and since a minority candidate cannot win the seat without having an equal opportunity to win the party primary, equal opportunity must be measured at every step in the electoral process. The precise point of equal opportunity, however, must always be evaluated on a district by district basis.

5. Section 5 Considerations in South Carolina; the Benchmark Plan and the Measure of Retrogression.

In order to ensure that our plan complies with § 5 of the Voting Rights Act, we must first identify the "benchmark" plan from which to measure any retrogressive effect of our actions. See Bossier Parish I, 520 U.S. at 478, 117 S.Ct. 1491. As a general premise, the benchmark plan for purposes of measuring retrogression is the last "legally enforceable" plan used in the jurisdiction. 28 C.F.R. § 51.54(b)(1) (providing that the last "legally enforceable" plan is the benchmark plan for purposes of measuring retrogression); see Holder v. Hall, 512 U.S. 874, 883-884, 114 S.Ct. 2581, 129 L.Ed.2d 687 (1994). A potential problem arises, however, due to the well-documented use of racial gerrymandering during the 1990s round of redistricting to maximize black representation. See, e.g., Abrams, 521 U.S. at 80, 117 S.Ct. 1925; Smith, 946 F.Supp. at 1185. The Governor asserts that this "max-black" policy has resulted in an unnecessarily high black voting age percentage in most majority-minority districts. In Abrams, however, the Supreme Court reiterated that the last legally enforceable plan used by the jurisdiction is to serve as the baseline for comparison in a § 5 retrogression analysis. See Abrams, 521 U.S. at 96-97, 117 S.Ct. 1925; see also Bossier Parish II, 528 U.S. at 334, 120 S.Ct. 866 (noting that "[i]n § 5 preclearance proceedings — which uniquely deal only and specifically with changes in voting procedures — the baseline is the status quo that is proposed to be changed: If the change abridges the right to vote relative to the status quo, preclearance is denied, and the status quo (however discriminatory it may be) remains in effect" (internal quotation marks omitted)); Holder, 512 U.S. at 883-84, 114 S.Ct. 2581 ("The baseline for comparison is present by definition; it is the existing status. While there may be difficulty in determining whether a proposed change would cause retrogression, there is little difficulty in discerning the two voting practices to compare to determine whether retrogression would occur."); Smith, 946 F.Supp. at 1209 (noting that "any benchmark ... should be the last plan that was legally adopted by the General Assembly that has not been set aside by the court or superseded by action of the General Assembly that has not been altered by the court"). And, in January 2001, the Justice Department released its guidelines for redistricting, taking the position that an actual "finding of unconstitutionality under Shaw" would be a prerequisite to rejecting the last legally enforceable plan as the benchmark. Guidance Concerning Redistricting & Retrogression, 66 Fed.Reg. 5412, 5412 (Jan. 18, 2001).

We hold that, so long as a current district or plan has not been formally declared to be the product of an unconstitutional racial gerrymander under the principles of Shaw and its progeny, it will serve as the benchmark for measuring retrogression. The determination that a particular district is the product of a racial gerrymander is a fact-intensive inquiry and no Shaw challenge to any district has been brought in these proceedings. Furthermore, any such benchmark districts have now been declared unconstitutional under the one-person, one-vote strictures of the Equal Protection Clause, and a Shaw inquiry would unnecessarily embroil this court in extended mini-trials over the moot issue of whether the district is constitutionally infirm for other reasons as well.

This conclusion does not, however, equate to a requirement that we must accept the configuration of the existing districts. The benchmark has no direct effect upon the constraints we now know to be imposed by the Equal Protection clause. The fact that a suspect district has not been formally declared unconstitutional does not affect our duty to draw districts in accordance with the goals of the Voting Rights Act and the principles enunciated by the Supreme Court in Shaw and its progeny. Those principles operate in tandem to require us to "clean up" cores of existing majority-minority districts should the application of federal law and traditional state districting principles now dictate a somewhat different configuration. If, in fact, the prior district was gerrymandered to achieve an unnecessarily — and indeed unconstitutionally — high percentage of black voting age population, the constraints of our current remedial factors will necessarily bring that percentage down to a more acceptable and reasonable level.

The question of how retrogression should be measured and evaluated in the 2000 redistricting process is a more difficult one. In January 2001, the Department of Justice issued its Guidance Concerning Redistricting and Retrogression Under Section 5 of the Voting Rights Act. According to the Department:

A proposed plan is retrogressive under the Section 5 "effect" prong if its net effect would be to reduce minority voters' "effective exercise of the electoral franchise" when compared to the benchmark plan. The effective exercise of the electoral franchise usually is assessed in redistricting submissions in terms of the opportunity for minority voters to elect candidates of their choice. The presence of racially polarized voting is an important factor considered by the Department of Justice in assessing minority voting strength. A proposed redistricting plan ordinarily will occasion an objection by the Department of Justice if the plan reduces minority voting strength relative to the benchmark plan and a fairly-drawn alternative plan could ameliorate or prevent that retrogression.

... [The state] bears the burden of demonstrating that a less-retrogressive plan cannot be reasonably drawn....

... If it is determined that a reasonable alternative plan exists that is nonretrogressive or less retrogressive than the submitted plan, the Department will interpose an objection.

66 Fed.Reg. at 5413 (Jan. 18, 2001) (emphasis added; internal citation omitted). Although § 5 "does not require jurisdictions to violate the one-person, one-vote principle" or "to violate Shaw v. Reno and related cases," the Department concludes that § 5 "may require the jurisdiction to depart from strict adherence to certain of its redistricting criteria." Id. By way of example, the Department advises that state criteria that "require the jurisdiction to make the least change to existing district boundaries, follow county, city, or precinct boundaries, protect incumbents, preserve partisan balance, or in some cases, require a certain level of compactness of district boundaries may need to give way to some degree to avoid retrogression." Id.

In accordance with these precedents, the standard of nonretrogression prohibits our implementation of a plan that will diminish the existing opportunity for minority voters to elect the candidate of their choice in a particular district if a fairly-drawn alternative exists that would not have that effect. However, "[n]onretrogression is not a license for the State to do whatever it deems necessary to ensure continued electoral success; it merely mandates that the minority's opportunity to elect representatives of its choice not be diminished, directly or indirectly, by the State's actions." Bush, 517 U.S. at 983, 116 S.Ct. 1941.

Accordingly, for purposes of § 5, we must consider race in redistricting where necessary to avoid retrogression. To do so, we first examine the court plan to ensure that it does not unfairly retrogress the overall position of minority voters by reducing the total number of majority-minority districts in the benchmark plan and, secondly, to ensure that it does not unnecessarily reduce the opportunity of minority voters to elect their preferred candidate in a particular district, where meaningful comparisons can be made. See Abrams, 521 U.S. at 97, 117 S.Ct. 1925.

Because we must narrowly tailor any race-based changes designed to prevent retrogression in the minority's voting strength, however, we do not arbitrarily strive to achieve any benchmark BVAP, which only represents the current BVAP in a malapportioned district. The history of race-based districting by state legislatures and courts in the early 1990s to maximize the drawing of majority-minority districts without regard to whether there was in fact a geographically compact minority population in the area, believed to be both constitutional and required by the Department of Justice policies, is well-documented. We have not examined individual districts to determine whether we believe they were racially gerrymandered because we do not think that inquiry is appropriate at this stage, nor have the parties proffered evidence which would be sufficient to establish any such finding. But, we do recognize that this may explain, at least in part, why the court, operating under its currently known constraints, cannot constitutionally achieve the existing BVAP in a fairly-drawn district. Additionally, unavoidable reductions in the BVAP percentages have resulted in some areas because of increased residential integration. While the 2000 census reveals that the percentage of minorities remained constant, minorities have moved away from historically black neighborhoods, resulting in substantial population deviations in existing majority-minority districts. Nearly all of the existing majority-minority districts need population and, to remain a majority-minority district, need predominately black population. The residential integration of blacks and whites revealed by the 2000 census results is an encouraging sign in South Carolina. But if it has any downside, it is that it makes it more difficult for courts and state legislatures to add the concentrations of black population necessary to replace the concentrations lost in predominately black areas to avoid retrogression without engaging in bizarre draws.

Cognizant of these constraints on our task, we consult the benchmark BVAP in each of the proposed majority-minority districts, and weigh it along with the other factors that affect the ultimate goal of avoiding a draw that causes an unnecessary diminution of the effectiveness of a majority-minority district. In sum, the court plans avoid retrogression by preserving the number of electable black majority districts in the current plan, and preventing the unnecessary reduction of BVAP in those effective majority-minority districts, within the parameters of the Equal Protection Clause, the Voting Rights Act, and traditional districting principles.

C. Traditional Districting Principles

We turn now to the "traditional districting principles" of the state. We must consider these principles along with the federal mandates imposed by the one-person, one-vote requirement of the Equal Protection Clause and the racial fairness and nonretrogression principles of the Voting Rights Act, but we apply them only to the extent that they do not conflict with the federal principles. See Abrams, 521 U.S. at 79, 117 S.Ct. 1925; Upham, 456 U.S. at 39, 102 S.Ct. 1518.

Generally speaking, traditional redistricting principles in South Carolina have directed courts to maintain, where possible, recognized communities of interest and the cores of existing districts, as well as political and geographical boundaries delineated within the state. See South Carolina State Conference of Branches of the NAACP v. Riley, 533 F.Supp. 1178, 1180 (D.S.C.), aff'd, 459 U.S. 1025, 103 S.Ct. 433, 74 L.Ed.2d 594 (1982). This includes maintaining county and municipal boundaries where possible, and protecting the cores of existing districts by altering old plans only as necessary to achieve the requisite goals of the new plan. See id. at 1180-81.

Maintaining the residences of the incumbents who serve those core constituents within the district is also a districting principle that historically has been observed in South Carolina. See id. Although this is usually referred to as "incumbency-protection," we view the principle as more accurately protecting the core constituency's interest in reelecting, if they choose, an incumbent representative in whom they have placed their trust. Provided it does not conflict with other nonpolitical considerations such as communities of interest and compactness, it is one worthy of consideration by this court. See, e.g., Bush, 517 U.S. at 964, 116 S.Ct. 1941 ("[W]e have recognized incumbency protection, at least in the form of avoiding contests between incumbents, as a legitimate state goal.") (internal quotation marks and alteration omitted); Johnson, 922 F.Supp. at 1565 (noting the protection of incumbents as a legitimate consideration, albeit an inherently more political one).

Metropolitan areas that overflow county boundaries, such as Charleston/Berkeley/Dorchester, Greenville/Spartanburg, and Richland/Lexington, previously have been recognized as sharing a special community of interest. See Riley, 533 F.Supp. at 1181. And, of course, South Carolina has traditionally adhered to the principle that "districts should be as compact as reasonably possible" and "that district boundaries are not gerrymandered to dilute the voting strength of minorities," id.

The amount of deference owed to the historical districting principle of maintaining county boundaries, and the position it holds in the hierarchy of pertinent considerations, was the subject of a fair amount of dispute at trial. In the 1980s redistricting process, the three-judge court in Riley noted the State's "substantial interest in the preservation of county lines," primarily "because the residents of a county have a community of interest." Id. at 1180. As noted then, county residents "are accustomed to voting together for county officials. There is much administrative convenience in drawing district lines along county lines, and it facilitates the process of organizing constituencies and campaigning for the support of constituents." Id. The Riley court, however, did not elevate the respect for county lines to the level of excluding all other factors. Rather, it likewise noted the state's historical concern for preserving the core of existing districts and protecting the incumbents who serve them, the state's recognition of other non-county-based communities of interest, the need for compact districts, and the principle of racial fairness. See id. at 1181.

Ten years later, the three-judge court in Burton seemed to elevate the importance of maintaining county lines. There, the court stated that, "in the quarter century since Reynolds the General Assembly has consistently stated, through its plans and specific statements of policy, that among various state policies, preserving county lines should enjoy a preeminent role in South Carolina's redistricting process. This preeminence is highly rational." Burton, 793 F.Supp. at 1341. Although the Burton court stated that the preservation of county lines was "[t]he only cognizable state policy [it] considered," id. at 1360, the court split 27 of the state's 46 counties in its Senate plan. Ultimately, Burton was summarily vacated by the Supreme Court and remanded for "further consideration in light of the position presented by the Solicitor General in his brief for the United States filed May 7, 1993." Statewide Reapportionment Advisory Comm. v. Theodore, 508 U.S. 968, 968, 113 S.Ct. 2954, 125 L.Ed.2d 656 (1993). The Solicitor General had maintained that the three-judge court failed to apply a proper analysis under § 2 of the Voting Rights Act and, as a part of that failure, placed undue emphasis upon preserving county and precinct lines.

We likewise recognize South Carolina's strong preference for minimizing the splits of counties within her borders. Many governmental services, such as fire and police protection, are organized along political subdivision lines, and counties and cities are often representative of a naturally existing community of interest. See Burton, 793 F.Supp. at 1341 & n. 29; Riley, 533 F.Supp. at 1180. We also understand the desire expressed by county representatives, particularly those in small counties, to remain "whole" in a district, or a substantial part of a district, in order to maximize their voice on a state level.

Nevertheless, we are also cognizant of the evidence that the trend of population movement continues to be from small rural counties to urban/suburban and coastal counties, and that often the cost of keeping small counties "whole" or their splits minimal is to increase the number of splits in more populated counties. Furthermore, as demonstrated by the testimony and other evidence presented to us in this proceeding, the principle of preserving county lines, while accorded much importance, has not been an inviolate policy or even a superior policy in all districts. Different parts of a county may also lack commonality of interest, most notably those counties located in coastal and metropolitan regions of the state which are now divided rather starkly upon rural/resort or rural/urban lines.

Finally, we note that it has long been recognized that "[t]he policy of maintaining the inviolability of county lines ..., if strictly adhered to, must inevitably collide with the basic equal protection standard of one person, one vote." Connor, 431 U.S. at 419, 97 S.Ct. 1828. These population shifts and the stringent deviation to which we are held make maintaining county lines a more difficult goal for the court to achieve, particularly in the context of the smaller House districts, than for the state's elected officials who possess more latitude in this regard.

For these reasons, we agree that we must honor South Carolina's important and longstanding state policy of maintaining county boundaries, but only insofar "as that can be done without violation of the [one-person, one-vote] principle of Wesberry v. Sanders and consistently with other state interests." Riley, 533 F.Supp. at 1180. Like all traditional districting principles adhered to by the state legislature, the principle of preserving county lines occupies a subordinate role to the federal directives embodied in the United States Constitution and the Voting Rights Act when the court is called upon to implement remedial redistricting plans. We do not find that the preservation of county lines continues to enjoy a preeminent role in the court's redistricting task. It is required to be considered as an important, guiding principle in our decision and, where appropriate, accorded great, but not necessarily greater, weight.

Having considered the history of redistricting in South Carolina and the evidence of traditional districting principles presented in this proceeding, we find that the traditional districting principles and existing redistricting policies observed in South Carolina direct us first and foremost to remedy the population deviations in existing districts by maintaining the core of those districts present in the malapportioned plan, while adding or subtracting population in a compact and contiguous manner to achieve the population equality required by the Constitution. In doing so, we consider all of the districting principles historically observed by the state. Generally speaking, however, we find that the cores in existing districts are the clearest expression of the legislature's intent to group persons on a "community of interest" basis, and because the cores are drawn with other traditional districting principles in mind, they will necessarily incorporate the state's other recognized interests in maintaining political boundaries, such as county and municipal lines, as well as other natural and historical communities of interest.

IV. The South Carolina House of Representatives

A. Background

The South Carolina House of Representatives consists of 124 members elected from single-member districts, see S.C. Const., art. III, § 3, to serve two-year terms, see S.C. Const., art. III, § 2. Prior to the Supreme Court's one-person, one-vote mandate enunciated in Reynolds v. Sims, 377 U.S. 533, 84 S.Ct. 1362, 12 L.Ed.2d 506 (1964), the House was apportioned to counties based on population, but each county received at least one representative regardless of population. See Smith, 946 F.Supp. at 1177 (detailing the history of legislative reapportionment since Reynolds); see also O'Shields v. McNair, 254 F.Supp. 708, 709-11 (D.S.C. 1966) (invalidating the South Carolina Constitution's method of apportioning the General Assembly as violative of the one-person, one-vote principle).

In the 1980s, the House successfully redistricted itself without court intervention. However, in 1992, the legislature reached an impasse with the Governor, who believed additional majority-minority districts were required by § 2 of the Voting Rights Act, resulting in a court-ordered plan. See Burton, 793 F.Supp. at 1340. The Burton court determined that the state would follow a predominate policy of maintaining county lines and held that, due to the exigency of the time pressures caused by a pending election schedule, it need not conduct a full § 2 Voting Rights Act inquiry. See id. at 1341, 1353. The Supreme Court summarily vacated and remanded the case to the district court "for further consideration in light of the position presented by the Solicitor General in his brief" to the Court. See Statewide Reapportionment Advisory Comm. v. Theodore, 508 U.S. 968, 113 S.Ct. 2954, 125 L.Ed.2d 656 (1993). As previously noted, the Solicitor General had argued that the district court placed undue emphasis upon the maintenance of county lines and gave inadequate consideration to the requirements of § 2 of the Voting Rights Act. More particularly, the Solicitor General argued that the district court did not adequately address the question of whether additional compact and contiguous districts with black majorities should be created to avoid dilution of black voting strength and that the district court lacked an appropriate basis to conclude that a bare 50% majority-black voting age population should be automatically considered a "black opportunity district" for purposes of § 2 of the Voting Rights Act. See Smith, 946 F.Supp. at 1181. Two weeks later, the Supreme Court issued its decision in Shaw v. Reno, 509 U.S. 630, 113 S.Ct. 2816, 125 L.Ed.2d 511 (1993), which denounced the use of racial gerrymandering to create majority-minority districts under § 2.

In 1994, the legislature successfully passed and precleared a redistricting plan for the House. However, nine of those districts were subsequently challenged as being racially gerrymandered in violation of the Shaw/Miller principles. Following a trial before a three-judge district court panel, six of the nine districts were held to be the product of an unconstitutional gerrymander. See Smith, 946 F.Supp. at 1210. In 1997, the General Assembly passed a House plan which corrected the unconstitutional districts struck down by the court in Smith, which the parties agree is the most recent, legally enforceable House plan.

As noted previously, the parties agree that the current House districts are malapportioned and, therefore, constitutionally infirm under the one-person, one-vote strictures of the Equal Protection Clause. Applying the 2000 census numbers, South Carolina's total population of 4,012,012 persons, split between the existing 124 House districts, compels a target population for each House district of 32,355 persons. The existing House plan has a total population deviation of 95%, with District 118 at -38.21% below the ideal district population and District 79 at 56.79% over the ideal district population.^[23]

The malapportionment is pervasive. By way of example, House District 79, located in the suburban area of Columbia in northeastern Richland County and southern Kershaw County, is the most populated house district in the state, with a total population of 50,729, resulting in a deviation of 56.79%. Without regard to the deviations in adjoining or other nearby districts, District 79 requires the removal of more than 18,000 persons from the district to achieve population equality — over half the population needed for a single district. Similar areas of large growth include the Greenville-Spartanburg area, Lexington County, and the suburbs of Charleston, as well as the beach communities of Horry County and Beaufort County.

House District 118, the former home of the now-closed Charleston Naval Base, demonstrates the opposite problem. Located in the heart of Charleston, it, like many other urban-based house districts, has become extremely underpopulated, and it also experiences the additional problem of being surrounded by other underpopulated urban-based districts. Consequently, District 118 is now the least-populated house district, with a total population of 19,992, for a deviation of — 38.21%. In order to bring it into population compliance, District 118 would require the addition of more than 12,000 persons — nearly half the population needed for a single district — from surrounding districts which themselves are severely underpopulated.

B. The Proposed Plans

During the House phase, four plans were proposed to the court: H.3003, which was passed by the state legislature, but vetoed by Governor Hodges; Governor Hodges' plan, which was submitted as an amendment to the House plan, but overturned on the floor; a plan submitted by Colleton County, which was also submitted to the House but overturned on the floor; and a plan submitted to us by Georgetown County. The bulk of the evidence presented by the House and the Governor consisted of explanations of why this court should not adopt and implement the plans submitted by the other. By way of summary, the House charged the Governor with preparing a plan that racially gerrymandered districts to increase the BVAP percentages in predominately white districts to between 25% and 50%. These so-called "influence districts," the House charges, are designed to promote the election of white Democratic candidates and the concomitant ousting of white Republican incumbents, at the expense of reducing BVAP percentages in existing majority-minority districts to marginal levels.

The Governor charges the House with a converse racial gerrymandering argument, asserting that H.3003's version of the House plan was intentionally drafted to "racially polarize" the state. Specifically, the Governor points to a handful of House representatives and charges that they intentionally gerrymandered their districts on a racial basis, often with the assistance of racial "dot-density" maps, by moving one or two predominately black precincts out of their Republican-controlled district and into a majority-minority or otherwise Democrat-controlled district. The effect, the Governor charges, was the intentional "bleaching" of Republican districts at the expense of existing minority influence districts.^[24] The Governor also asserts that his proposed plan adheres more closely to traditional districting principles in South Carolina and maintains more closely the cores of existing districts than does the legislatively passed House plan.

The ACLU initially proposed three alternative plans, but does not now advocate the plans for adoption by this court. Although they submitted the plans as illustrative of plans that satisfy the Voting Rights Act,^[25] the ACLU asserts that they only seek to ensure that any plan adopted by the court is in compliance with the racial-fairness standards of § 2 and the nonretrogression standards of § 5. They further assert that all of the proposed plans satisfy the nonretrogression standard of § 5 and that the House plan, with minor modifications in a few majority-minority districts (particularly Districts 12, 102 and 116), is acceptable from a § 2 standpoint. Finally, the ACLU specifically denounced the Governor's advocation of "influence districts" both as a policy matter and, given its partisan-based origin, as a valid consideration for this court in the draw.

Colleton County and Georgetown County advance more local concerns. Under the present 1997 plan, Colleton County is divided among five House districts, comprising a majority in none. The House and Governor both propose to now split Colleton County into four House districts, which Colleton County also opposes. Colleton County asks to be split into no more than three House districts. Additionally, Colleton County specifically opposes the House plan's proposal to separate the 742 persons on Edisto Beach from their neighbors on Edisto Island in District 119, and move them to the predominately Charleston-based District 116, which is contiguous only by water. Colleton County asserts that the split is unconstitutionally race-based, whereas the House asserts that the split was motivated by a coastal community of interest.

Georgetown County contests the decision of the House and Governor to split the City of Georgetown between House Districts 103 and 108. Georgetown contends that the City of Georgetown shares a strong community of interest with the coastal portion of Georgetown County and should therefore remain wholly within District 108. The decision to do otherwise, Georgetown charges, represents an unconstitutional, race-based gerrymander, performed with the exclusive intent to maintain House District 103 as a majority-minority district in derogation of all other traditional districting principles.

Having reviewed the evidence and testimony received, the court concludes that none of the proposed plans comply with the criteria required of a judicially drawn plan. First, the House and the Governor have submitted plans which exceed the range of de minimis population deviation and, therefore, could not be adopted by this court even if they were to survive the preclearance process under the Voting Rights Act. Neither disputes this fact. The House plan has a total deviation of 4.86% and the Governor's plan has a total deviation of 3.13%. The other plans submitted are likewise beyond an acceptable range of deviation for a court-ordered plan.

Second, the plans are of limited utility as a template to guide us. For the most part, the House and Governor have opted to attack each other's plan as being race-based and politically motivated. The evidence bears both sides out; both plans plainly represent the culmination of very specific political and, in some cases, individual agendas. It is well-documented in the evidence, and indeed not in dispute by the parties, that black citizens in South Carolina vote almost unanimously for Democratic candidates against Republican candidates and, in a primary, for black Democrats against white Democrats. Given this high level of racial polarization and high correlation between black voters and the Democratic Party which currently exists, it is hardly remarkable that when one places a pen in the hands of incumbents to draw their own districts, a Republican might draw fewer blacks in the district to maximize its Republican base; a white Democrat might draw more blacks in the district to increase its Democratic majority, but not so many as to be threatened with defeat by a black challenger; and a black Democrat might draw enough blacks in the district as necessary to ensure reelection. In view of the mandate that race must be of at least some consideration given the high levels of racial polarization in the state, we find it most difficult to determine whether and when the incumbents' proper motives gave way to improper ones.

Fortunately, we need not do so. In some areas of the state, we were able to consult the submitted plans for common, and therefore noncontroversial ground, although this assistance is limited in those parts of the state that experienced substantial population disparity. In this limited manner, unenacted legislative and gubernatorial plans were considered as evidence of state policy, but always with a healthy regard for each party's highlighting of the opponent's political motivations. In each case, we consulted both, but we ultimately found it necessary to draw the districts largely from scratch, in accordance with our own interpretation of the remedial factors that govern us in our draw.

C. The Court's Plan

1. Population Equality

Because our primary charge is to remedy the malapportionment present in the existing plan, we were guided in our remedy by traditional districting principles, seeking to preserve the core of each existing district where possible, and adding or subtracting population surrounding the core in a compact and contiguous manner. In doing so, we remained mindful of the existing majority-minority districts which §§ 2 and 5 of the Voting Rights Act compel us to respect, as well as any additional areas that the § 2 inquiry would direct us to protect from inadvertent vote dilution.

Because the substantial shifts in population alone would inevitably cause major change to the district cores in some areas, we began our charge with the question of whether we should collapse any existing House districts that have experienced extreme population loss and relocate them as new districts in areas of extreme population growth. In this regard, at least, we found the proposed plans to be of substantial benefit.

Both the House and the Governor agreed that the population shifts in South Carolina require the relocation of two House districts. Both proposed to collapse two existing districts in or near the areas of the state with the greatest population loss. They also agreed that the explosive population growth in the coastal counties of Horry and Beaufort called for the creation of one new House district in each. Horry County grew by 52,576 persons, a 36.5% increase in population, and Beaufort County grew by 34,512 persons, a 39.9% increase in population.^[26]

We agreed that two house districts must be collapsed and relocated elsewhere. Unless we collapsed and moved districts to account for the substantial population losses in the Pee Dee and downtown Charleston area and the substantial population growth in Horry County and Beaufort County, gross disruption of the cores of all existing districts would have "rippled" throughout the Pee Dee and Low Country regions of the state. See, e.g., Johnson, 922 F.Supp. at 1563 (noting that the appropriate place to locate Georgia's new congressional district was the high population growth area near Atlanta). By collapsing districts in those areas of greatest population loss, population is freed up to be relocated to adjoining districts, which were also low in population. Similarly, by relocating these districts to the areas of greatest population gain, we minimized the "ripple" effect of the population gains in the other area districts.

Of course, this determination led us to the difficult question of exactly which two districts in the existing plan should be collapsed, and where those districts should be relocated. Like the House and Governor, we found that the most logical places to locate new districts were in the areas where the most population growth occurred — Horry County and Beaufort County. And, when we moved to these areas to locate the new districts, the cores of each new district naturally landed in the areas of those counties that, in fact, experienced the most growth.

Although the parties agreed that two districts should be collapsed and moved, they disagreed as to which two specific districts should be collapsed. The parties did agree that an existing district must be collapsed in the Charleston area (which grew only 5.0%) and moved to the Beaufort area as described above, but disagreed as to which district in Charleston should be collapsed. The House, which is currently controlled by the Republican Party, proposed the collapse of District 113 in Charleston County, whereas the Governor, a Democrat, proposed the collapse of District 119 in Charleston County. The parties wholly disagreed as to the specific area of the state where the second district should be collapsed. The House proposed the collapse of District 80 in Richland County, an area which grew at 12.0%, whereas the Governor proposed the collapse of District 67 in Sumter County, an area which grew at only 3.3%. Unsurprisingly, the House proposed the collapse of two Democratic districts and the Governor proposed the collapse of two Republican districts.

In order to maintain as much as possible the cores of the remaining districts in the area, the court found that a district must be collapsed in the Charleston area, which experienced the second greatest overall population loss, in large part due to the closure of the Charleston Naval Base in the 1990s. The court rejected, however, both parties' proposed district for collapse. Instead, the court found that the most logical district to collapse in Charleston County was District 118, the district that experienced the greatest negative population deviation in the Charleston area (38.21%) and, indeed, statewide.

With regard to the second district to be collapsed, the court rejected the House proposal to collapse District 80 in Richland County, which was only 7.29% low in population, as opposed to the Governor's proposal to collapse a district in Sumter County. Collapsing a district in Richland County would result in the unnecessary shift of districts in the Midlands area of the state, which has not experienced a population loss as a whole, whereas the Pee Dee area experienced the largest overall negative population deviation in the state. However, the court rejected the Governor's proposal to collapse District 67 in the Pee Dee, which experienced a population deviation of - 13.14%. Instead, we found that District 68 was the most logical district to collapse because it, like the district we collapsed in Charleston, experienced the highest negative population deviation in its area (- 17.94%).

Having made these difficult threshold decisions, we then turned to our task of drawing equipopulous districts. There, we began to draw the districts in a compact and contiguous manner, identifying at the outset the core of each existing district and moving about its borders to add or subtract compact and contiguous areas of population as the individual district dictated. This approach not only served the goal of minimal change and constituency consistency in the individual district, it created the least amount of "ripple" effect in other districts. Where possible, and where the decision was not outweighed by other competing interests, we attempted not only to avoid new county splits, but to also eliminate existing county splits to the extent possible. Our plan does not create any county or city splits that were not already present in the benchmark plan, although it does eliminate several splits that did exist. Beyond this, we were motivated by an attempt to follow natural and other easily identifiable boundaries, such as major roads, to create a smooth district border that election officials could easily pinpoint.

In the end, the court's plan achieves the requisite population equality, with a total de minimis deviation of plus or minus one percent variation, see, e.g., Chapman, 420 U.S. at 26-27, 95 S.Ct. 751, and results in the least change in terms of constituency movement. The plan contains three less total county splits than the 1997 benchmark plan and, as compared to the proposed plans of the House and Governor, the court plan contains seven less total county splits and nine less total county splits, respectively. With the exception of Districts 67 and 113, which adjoined the two collapsed districts, the court plan also does not pit any incumbents against one another. Finally, we note that the split of Colleton County has been reduced from five to three splits, and the constituents of Edisto Beach remain with their Edisto Island neighbors in District 121. And, being in agreement with the position that the City of Georgetown shares a strong community of interest with the coastal portion of Georgetown County, the court does not split the City of Georgetown, which remains wholly within District 108.

2. The Voting Rights Act

As noted previously, in carrying out our remedial task of devising and implementing a redistricting plan, the court is also directed to consider the requirements imposed upon South Carolina by the Voting Rights Act. For purposes of § 5, the House and Governor agree that the "benchmark" plan is the current 1997 plan with the 2000 census numbers applied to it. Although the ACLU urges us to also consult the 1997 plan with the 1990 census numbers applied to it for comparison, they agree that both parties have drawn reasonable districts for purposes of the Voting Rights Act. The 1997 plan with the 1990 census numbers contains 32 majority-minority districts. As a result of the population losses and shifts revealed by the 2000 census results, the benchmark plan now contains only 25 majority-minority districts.

Prior to trial, the parties agreed that no § 5 issues were presented by the plans proposed by the Governor or House because each proposed majority-minority districts in excess of the 25 districts in the current benchmark plan. The Governor's plan proposed 28 majority-minority districts, whereas the House plan proposed 27 majority-minority districts. The ACLU's illustrative plan contained 32 majority-minority districts, but was not drawn in accordance with traditional districting principles, and the ACLU agrees that the numbers presented by the Governor and House are sufficient to avoid retrogression. Although we did not elect to adopt a submitted plan, the court-drawn plan, which also exceeds the 25-district benchmark, raises no retrogression issues under § 5.

The court plan also complies with § 2 of the Voting Rights Act. In preparing the new districting plan, we considered every existing majority-minority district present in the benchmark plan, as well as other areas pointed out to us by the parties as ones that § 2 might require. First, in keeping with our paramount goal for every existing district — save District 118 in which population deviation compelled a collapse — we attempted to maintain the core of each existing majority-minority district. With few exceptions, these districts required the addition of population to achieve the one-person, one-vote requirement. And, in every case, we examined the geographic territory surrounding the existing minority district to determine whether additional concentrations of minority voters were present in a compact and contiguous area such that we could achieve the concurrent goals of equalizing population, avoiding retrogression, and preserving traditional districting principles. Where we believed it necessary to honor the principles established in the Shaw/Miller line of cases, we eliminated lines that, although never challenged, appeared to us to possibly be the result of the 1990s racial gerrymandering or that otherwise did not comport with the identified traditional districting principles.

Second, we considered those districts that were not majority-minority under the benchmark plan, but that one or more parties urged us to consider under § 2. If the minority population existed in a reasonably compact area, otherwise met the Gingles test,^[27] and could be drawn utilizing traditional districting principles, it was drawn as a majority-minority district. House Districts 12 and 82, for example, fell easily into this category. The minority population in each area is sufficiently compact and has a sufficiently strong community of interest to require its draw as a § 2 district. District 103, in contrast, could have been drawn as a majority-minority district if we had split the City of Georgetown. The Governor and the House did so, placing its BVAP above the 50% mark. We declined to do so. The proposed split violated the strong community of interest between the residents of the City of Georgetown and its coastal neighbors. The court did not split municipal boundaries elsewhere in the plan, unless they were already split or were split on a county line or where municipal population size mandated a split. And, the loss of the district as a majority-black one did not cause a net retrogressive effect.

Although in drawing a majority-minority district applying traditional districting principles, the court might have produced a district where the BVAP exceeds the percentage necessary to be considered an "equal opportunity" district under § 2, the court did not allow race to predominate so as to remove the excess black voters to create an "influence" district elsewhere. We protected the status of an equal opportunity district and prevent retrogression where we must, as we are compelled to do by the mandates of the Voting Rights Act. Beyond that mandate, however, the lines were drawn wherever traditional districting principles led us without the consideration of race. Although demographic changes did not always allow for a constitutionally appropriate draw that reached the benchmark BVAP in every district, there is no net loss of majority-minority districts in the state-wide plan and the black voting age population in every district closely approximates that percentage which the parties agreed was sufficient to be both a minority opportunity district under § 2 and nonretrogressive under § 5.

3. The Specific Majority-Minority House Districts

We turn now to the specific majority-minority districts present in the court plan. As noted previously, the Governor's plan proposes 27 majority-minority districts, whereas the House plan proposes 28 majority-minority districts. The court-drawn plan contains 29 majority-minority districts. As to each, we make the following additional findings.

First, the parties agreed that twenty-one districts — Districts 23, 25, 31, 49, 50, 51, 57, 59, 62, 66, 70, 73, 74, 76, 77, 82, 91, 95, 101, 109, and 121 — all satisfied the Gingles test and that the proposed plans had all been drawn to a level sufficient to provide black voters an equal opportunity to elect the candidates of their choice. This was confirmed by Dr. Ruoff in his testimony. In addition, Dr. Ruoff testified that Districts 111 and 122, while in danger of losing their opportunity status due to anticipated development, also met the Gingles test and were drawn to provide the requisite equal opportunity. We agree that the cores of these districts are located in areas that satisfy the Gingles test and, while not adopting either plan's iteration in full, the court's draw of these districts accomplishes similar levels of opportunity.

The draws of two districts — Districts 41 and 64 — in the proposed plans are, according to Dr. Ruoff, at questionable levels to be called "equal opportunity" districts, but cannot be drawn at a higher BVAP level without racially gerrymandering. Having determined that the areas do meet the Gingles factors, the court's plan maintains these districts as majority-minority and at a level similar to that proposed by the parties and which the court believes will be sufficient to provide the black voters an equal opportunity.

According to Dr. Ruoff, three districts — Districts 102, 103, and 116 — are drawn to a marginal level of equal opportunity by the proposed plans, but should be maintained as majority-minority districts. Having considered the areas in conjunction with other districting principles, the court agrees that Districts 102 and 116 meet the Gingles factors. Under the court plan, they contain BVAP levels that approximate those proposed by the parties and that the court believes is sufficient to provide an equal opportunity. As noted previously, District 103 could not be drawn to the proposed level of equal opportunity without violating the existing community of interest between the residents of the City of Georgetown and the Georgetown County coastal community; therefore, it falls just shy of a majority-minority district.

Finally, Dr. Ruoff testified that District 118 was in serious jeopardy and that District 12 was a questionable equal opportunity district. As previously noted, the court found that District 12 could and should be maintained as an equal opportunity district, and the court has drawn that district accordingly. However, for the reasons previously discussed, the court weighed all of the relevant considerations and concluded that the severe population variance in District 118 and its surrounding districts dictated that it be collapsed and relocated to an area of substantial population growth in Beaufort County. As a result of that collapse, however, District 113 recouped much of its population shortfall from the constituents of the collapsed District 118, causing its BVAP to increase from 32.15% to 50.11%, which we also believe provides the minority constituents in that area an equal opportunity to elect the candidate of their choice. Consequently, the number of majority-minority districts in this area suffered no loss.

V. The South Carolina Senate

A. Background

The South Carolina Senate consists of 46 members elected from single-member districts. See S.C. Const., art. III, § 1; State ex rel. McLeod v. West, 249 S.C. 243, 153 S.E.2d 892, 894 (1967). Prior to Reynolds' one-person, one-vote mandate, each of South Carolina's 46 counties was entitled to elect one senator to serve a four-year term. See S.C. Const., art. III, § 6; see also Smith, 946 F.Supp. at 1177; O'Shields, 254 F.Supp. at 711. In Burton, the three-judge panel implemented a court-drawn plan for the Senate, see Burton, 793 F.Supp. at 1358-63, a plan that was subsequently vacated by the Supreme Court. See Statewide Reapportionment Advisory Comm., 508 U.S. at 968, 113 S.Ct. 2954. The Burton court plan had created eleven black-majority districts, ten of which had a majority-black voting age population. See Burton, 793 F.Supp. at 1362. On remand, the General Assembly passed a new redistricting plan for the Senate, which received approval from the Department of Justice and became law. See Smith, 946 F.Supp. at 1181, 1201-02. This legislatively drawn plan created 12 black-majority districts. See id. at 1203.

However, in Smith, three of those districts — Districts 29, 34 and 37 — were subsequently declared to be the product of an unconstitutional gerrymander. See Smith, 946 F.Supp. at 1210. In 1997, the General Assembly passed a Senate plan to remedy the unconstitutional Senate districts. It is the plan currently in use and, therefore, is the last legally enforceable Senate plan. The current plan, applying either 1990 or 2000 census numbers, contains 10 majority-minority districts.

As with the House districts, it is undisputed that the existing Senate districting plan is now malapportioned. Based on the state's total population of 4,012,012, the ideal Senate district would have a population of 87,218 people. The total deviation of the existing Senate districts is 48.16%, with District 46 overpopulated by 21.15% and District 43 underpopulated by 27.01%. Nine of the ten majority-minority districts in the Senate benchmark plan, measured by total black population, are substantially low in population.^[28]

B. The Proposed Plans

During the Senate phase, six plans were proposed to the court: H.3003's version of the Senate plan, which was passed by the legislature, but vetoed by Governor Hodges; two plans prepared by Governor Hodges, designated Plan A and Plan B; one plan proposed by Leatherman; one plan proposed by Colleton County; and one plan proposed by Georgetown County. The ACLU proposed an illustrative plan as well, but now place their full support behind the Senate version of the Senate plan, asserting that it fully complies with the one-person, one-vote requirements of the Constitution and the Voting Rights Act. As in the House case, the bulk of the evidence presented was designed to encourage this court to adopt the plan of the advocate and to point out the improprieties of the other plans.

The Senate asserts that its plan better serves the goals of least change, compactness, preserving communities of interest and county lines, and accommodating incumbents where practical. The Senate also charges that the Governor's plan cannot be adopted because it facially violates § 5 of the Voting Rights Act by reducing the overall number of effective majority-minority districts under the existing plan and by reducing the effectiveness of individual majority-minority districts by deliberately removing black population from majority-black districts and moving them into adjacent districts. The effect, as alleged by the Senate, is the deliberate reduction of the BVAP in majority-minority districts below the point of equal opportunity in order to create so-called "influence districts" — majority white districts where blacks cannot elect their candidate of choice, but will substantially aid in the election of a white Democratic candidate over a Republican candidate. While the Senate asserts that such influence districts may have merit from a purely political standpoint, i.e., to increase the probability that the Democratic party will be able to oust incumbent Republicans and regain control of the South Carolina Senate, it asserts that political goal cannot be achieved at the expense of retrogressing other districts and that this court cannot engage in such a political design. Simply stated, the court cannot draw a plan that intentionally aids one political party at the expense of another.

The Governor, for his part, charges that the Senate plan is racially and politically gerrymandered for the purpose of "bleaching" Republican-held districts and rendering them "safe" Republican districts. As before, this generally amounts to the converse of the Senate's argument that the Governor has racially gerrymandered districts to achieve the opposite result. By increasing the BVAP percentage in Republican-held districts, the Governor increases the probability of a white Democrat ousting the Republican incumbent in the general election, but at the expense of lowering the BVAP in an adjoining majority-minority district and the concomitant ability for blacks to elect a black Democrat in that district over a white Democrat. By increasing the BVAP in current majority-minority districts, the Senate Republicans avoid that result and make the adjoining "superwhite" districts Republican strongholds. Leatherman has also proposed a Senate plan that generally advances the Republican point of view, basing communities of interest upon votes cast for a particular candidate. Again, these are political arguments that we cannot and do not endorse, and involve issues that are not appropriate for our consideration in the draw.

Colleton County asserts, as before, the importance of county lines to our inquiry. The Senate Plan splits Colleton County into three districts, District 38, 39 and 45. Noting that the approximately 38,000 people in Colleton County could fit within one senate district (and even then not hold the majority), they ask us to consider their plan for splitting Colleton County into only two districts, Districts 38 and 45.

Georgetown County is divided between Senate District 32, located west of the City of Georgetown in a more rural part of the County, and District 34, located on the coastal side of the County. Essentially, Georgetown's plan advocates a proposal that maintains the core of District 34 along the coast of Georgetown, with Georgetown as the hub or core. As part of its proposal, Georgetown also advocates the retention of the City of Georgetown in District 34, based upon the assertion that it shares the coastal-based community of interest (fishing, beach and environmental concerns in particular) with the other parts of that district.

Again, we have carefully considered the plans proposed by all of the parties, but conclude that none comply fully with the criteria that govern a court-drawn plan. As was the case in the House phase of this litigation, the evidence presented demonstrated that political policies and goals were the driving force in determining the draws of the plans submitted by the Senate, Leatherman, and the Governor. In particular, we note that while we may consider race to the extent authorized by the Voting Rights Act — to avoid minority vote dilution and retrogression — we are aware of no authority, and have been pointed to none, that would permit us to draw district lines based upon what amounts to political policy decisions as to who best can represent the wishes of minority voters. Accordingly, this court has drawn its own plan in accordance with the criteria that guide us, consulting the proposed plans for common ground where possible.

C. The Court's Plan

1. Population Equality

First, the court plan satisfies the one-person, one-vote standard required by the Equal Protection clause. As we did in the House case, the court generally sought to maintain the cores of the existing districts, adding or subtracting compact and contiguous population as individual district requirements dictated to correct the population deviations. The district-wide plan has a total de minimis variation of plus 0.96% and minus 0.85%. And, the plan does not pit any incumbents against each other in the same districts.

2. The Voting Rights Act

The parties agree that none of the proposed plans run afoul of the requirements of § 2 of the Voting Rights Act. Neither does the court plan. Rather, the application of § 5 of the Voting Rights Act to the Senate districts is the source of the dispute.

All proposed plans, save those of the Governor, maintain the current benchmark of nine majority-minority districts, plus District 7, at levels sufficient to provide the minority voters an equal opportunity to elect their preferred candidate. The Governor's proposed plan maintains only eight majority-minority districts, plus District 7, at a level of equal opportunity, which the others contend is retrogressive.

More specifically, all of the proposed plans maintain Districts 19, 21, 30, 32, 36, 39, 42, and 45 as majority-minority districts — that is, each district has a BVAP that exceeds the 50% mark — and at levels that each plan's advocate believes to be sufficient to provide the minority voters an equal opportunity to elect the candidate of their choice.

Two other districts — District 7 in Greenville and District 17 in the Chester/Fairfield area — have historically been majority-minority opportunity districts under the existing plan, but cannot be maintained as such. Under the benchmark plan, just over 50% of the total population of District 7 is black, but the current BVAP falls just below the 50% mark. Additionally, the district has lost so much population (- 15.06% deviation) that the parties agree it can no longer be drawn with at least 50% total black population or black voting age population, without improper racial gerrymandering. However, all parties agree that the BVAP in District 7 is still high enough for it to be an equal opportunity district and the district is currently represented by a black incumbent. District 17 has a total population and BVAP in excess of 50% under the benchmark plan, but also suffers from a substantial population shortage (- 12.77%). However, it is obvious that it can no longer be drawn with a BVAP over 50% or as one of equal opportunity without unconstitutionally gerrymandering the district.

The Senate plan, however, would add black voting age population to District 40, located in the Orangeburg area, to make it a majority-minority district where blacks have an equal opportunity to elect the candidates of their choice. Although the Governor's plan produces a district with only 46.96% BVAP, Senator Brad Hutto, the Democratic Senator who currently represents District 40, encouraged the court to increase District 40's BVAP to over 50%.

Minority Districts 30, 32 and 36 generated the greatest amount of discussion. All located in the Pee Dee area and currently low in population, they surround the Republican-held District 31 in Florence. While most of the submitted plans keep these districts at a BVAP level of 58% to 59%, the Governor's plan reduces the BVAP in District 30 (currently represented by a minority candidate) from 58.69% to 52.00%; reduces the BVAP in District 32 (which has never elected a minority candidate) from 58.27% to 50.93; and reduces the BVAP in District 36 (which also has never elected a minority candidate) from 58.37% to 52.46%. As a result, it appears, the Governor's plan can maintain, and even slightly increase, the BVAP percentage in District 31 from 29.58% to 31.91%, compared to the Senate's BVAP percentage of 22.34%.

3. The Specific Senate Majority-Minority Districts

The court-drawn plan contains nine majority-minority districts developed from the cores of the existing districts, plus District 7, at sufficient levels to afford the minority voters an equal opportunity to elect their preferred candidate. All of these districts were drawn using the state's traditional districting principles along with consideration of the compact and contiguous black population in and adjoining the cores of those districts.

Eight of the nine majority-minority districts in the court plan are drawn as majority-minority districts in both the Senate and Governor's plans; they are Districts 19, 21, 30, 32, 36, 39, 42, and 45. Also, like the Senate and Governor, the court plan does not attempt to maintain District 17 as a majority-minority district given the overwhelming evidence that it cannot be done in a manner to ensure equal opportunity without racially gerrymandering the district. Thus, the only substantial difference in this respect lies in the court's draw of District 40 as a majority-minority district, as the Senate plan did, in order to both prevent retrogression in the state-wide voting strength of blacks and to protect the naturally compact voters within the area from the dilution of their voting strength.

The BVAP of each preexisting majority-minority district in the court plan was reduced somewhat from the benchmark plan, this being caused primarily by the existence of abnormally high levels of black population established by the previous gerrymandering of the district lines and by the subsequent loss in population within the districts. However, we are satisfied that the BVAP for each district in the court's plan is the best that could be reached consistent with our responsibilities under the Voting Rights Act and other districting principles. And, we are confident that the BVAP levels are sufficient in each district to provide the minority voters of each an equal opportunity to elect the candidate of their choice. In reaching these conclusions, we have carefully considered the voluminous expert testimony and statistical evidence presented to us, and note that, with one exception, the BVAPs in the court plan closely approximate or exceed the points of equal opportunity arrived at by each expert.

The one exception is District 36, located in the Pee Dee area of the state. Both the Senate and the House draw the district as a majority-minority district, but differ substantially on the BVAP level necessary to render it an equal opportunity district. The Governor's expert, Dr. Epstein, opined that a district in the Pee Dee region generally requires a BVAP of only 48.19% to give black voters an equal opportunity, and the Governor draws the district at 52.46% BVAP. Dr. Ruoff and Dr. Loewen, on the other hand, both testified that the district needed a BVAP in excess of 57% to give minority voters an equal opportunity to elect a minority candidate of their choice, but base this conclusion on a demonstrated lack of cohesion among the black voters in that district in the last several elections. Everyone acknowledges, however, that District 36 is currently represented by a popular and long-serving white Democratic senator.

We agree with the parties that District 36 should be maintained as a majority-minority district. However, the experts' analyses of the election data clearly reveals that District 36 is atypical for South Carolina and, in particular, for the Pee Dee region of South Carolina in that blacks are crossing over in sufficient numbers to vote with a highly polarized white community to elect a white incumbent over a black challenger. Nevertheless, the district is currently a majority-minority district in the benchmark plan, and the racial composition of the population residing there and traditional districting principles naturally led to District 36 remaining a majority-minority district. Additionally, we view the lack of cohesion in the black community as induced by the years of faithful constituent service by the long-serving white Democratic senator, and we question the testimony that an extraordinarily high BVAP is necessary to make District 36 one in which minorities have an equal opportunity to elect the candidate of their choice for purposes of the Voting Rights Act. We must remember that the question is not what BVAP would be necessary to defeat a popular incumbent; the question is what BVAP is required to insure that the minority population has an equal opportunity to elect a minority candidate of choice in an open election. We are convinced by the testimony that the bloc voting by black citizens that has been demonstrated to exist in South Carolina generally and in other parts of the Pee Dee area in particular would also occur in District 36 during an open election and, therefore, that a BVAP consistent with that required in other parts of the Pee Dee is proper. Accordingly, we believe the BVAP that exists under the court's plan, 55.15%, also will be sufficient to make this an opportunity district for minorities in an open election.

The court plan also draws District 40 as a majority-minority district. District 40 previously had a BVAP of 48.73%, almost enough to make it a majority-minority district. The black population in this highly polarized area is sufficiently large and geographically compact to satisfy the Gingles test, and this population coupled with traditional districting principles quite naturally led to its being drawn as an opportunity district. We are also satisfied that the BVAP we establish (51.02%) will, based upon the expert testimony of Dr. Loewen, make this district one of equal opportunity. And, by drawing District 40 as an opportunity district, we avoid the retrogression in the total number of such districts state-wide that the loss of District 17 would otherwise cause.

In sum, there is no impermissible retrogression in the court's Senate plan. The number of majority-minority seats, measured by black voting age population, remains the same as in the benchmark plan. And, the changes in voting strength reflected in the BVAP for each district, caused by our correction of the likely racial gerrymandering of the 1990s and the subsequent drops in population, are minor and do not represent a diminution of the effective voting strength of the minority population in those areas. Consequently, in our judgment, § 5 of the Voting Rights Act has been satisfied. Section 2 of the Voting Rights Act has also been satisfied. The only new district required by § 2 has been created. Additionally, we have weighed the testimony and verified that the BVAPs in all of the majority-minority districts and in District 7 are sufficient for minorities to have an equal opportunity to elect the candidate of their choice.

VI. The United States House of Representatives

A. Background

Based upon the results of the 2000 census, South Carolina is entitled to keep its six seats in the United States Congress. With a total population of 4,012,012, the ideal congressional district should contain 668,669 persons. Four of South Carolina's six existing districts are overpopulated: the First District is overpopulated by 16,096; the Second District is overpopulated by 62,353; the Third District is overpopulated by 1,470 persons; and the Fourth District is overpopulated by 1,666 persons. The Fifth District is underpopulated by 13,144 persons. The Sixth District, which is South Carolina's only existing majority-minority district, is severely underpopulated by 68,443 persons. The parties concede that all of the existing congressional districts are malapportioned based upon the 2000 census, and therefore in violation of Article I, § 2 of the Constitution, and that the General Assembly and Governor are at an impasse in the redistricting process.

B. The Proposed Plans

A number of plans have been submitted for consideration by the court. First, the House and Senate have a joint legislatively passed plan (the "General Assembly plan"), which was vetoed by the Governor; the Governor has offered two proposed plans; Leatherman has offered two proposed plans; and Colleton County has offered a proposed plan. The General Assembly plan and the plans proposed by Leatherman all have a total deviation of plus or minus one person. The Governor's plan deviates from plus nine to minus eleven persons.

With regard to the Voting Rights Act requirements, the parties agree that there remains a single possibility for a majority-minority congressional district in South Carolina — the existing majority-minority Sixth Congressional District, which has as its core the substantial and predominately black populations in the Pee Dee area and in portions of adjoining Richland and Charleston Counties. The Sixth District, which is currently represented by the Honorable James Clyburn (the only African-American congressional incumbent in South Carolina), has a current BVAP, based upon the 2000 census population, of 57.78%. All parties agree that the Voting Rights Act requires the maintenance of this single district as a majority-minority district in South Carolina, and all proposed plans incorporate the core areas of the current district. The dispute among the parties arises over the specific lines that should encompass constituents of the Sixth Congressional District.

The General Assembly plan achieves a BVAP in the Sixth District of 54.12%. To resolve the severe population loss in the district, the General Assembly plan proposes to eliminate the current split in Colleton County between the Second and Sixth Districts and to place the county entirely in the Sixth District. The splits of Orangeburg and Calhoun County between the Second and Sixth Districts are maintained. The splits of Lee County and Darlington County between the Fifth and Sixth Districts are removed, but Sumter County, in the Pee Dee, and Richland County remain split, as do Dorchester, Berkeley, and Charleston Counties. Georgetown County is also split in the General Assembly plan.

The Governor has proposed two congressional redistricting plans: Governor's Plan A and Plan B. The two plans are nearly identical, achieving BVAP percentages of 53.05% and 53.01%, respectively. Both plans reflect the Governor's desire to eliminate the existing splits of Orangeburg County and Calhoun County between the Second and Sixth Districts and to place those counties entirely within the Sixth District, as well as the Governor's desire to avoid a split of Georgetown County. To accomplish these goals, the Governor eliminates the split of Colleton County, but places it whole within the Second District. Like the General Assembly plan, the Governor eliminates the splits of Lee County and Darlington County, but maintains the splits in Sumter County, Richland County, Dorchester County, Berkeley County, and Charleston County. However, the Governor's plans also move the Savannah River Site in Aiken County from its home in the Third District to the Second District. And, Governor's Plan B, at the apparent request of Congressman Clyburn, proposes the relocation of Fort Jackson, the largest military installation in South Carolina, from its present home in the Second District, which is currently represented by Congressman Joe Wilson, to the Sixth District, which is currently represented by Congressman Clyburn.

Both of Leatherman's plans make dramatic changes to the Second, Fifth, and Sixth Congressional Districts. Designated the "Court E Plan" and "No Retrogression Plan," the plans achieve a BVAP percentage of 56.57% and 58.37%, respectively. They are admittedly partisan-based draws, designed to maximize the election of Republican candidates. For the same reasons, they are inconsistent with this court's goal of achieving population equality with the minimal effect upon the status quo. While we have considered these plans, we find them to be of quite limited utility to us in our draw.

Finally, Colleton County has appeared for the sole purpose of advocating that the County, which is currently split between the Second District and the Sixth District, be kept wholly within one congressional district in the court-imposed plan. Prior to the court-drawn plan of Burton, Colleton County was placed wholly within the First District. However, Colleton County takes no position as to which district it would prefer to be placed within, so long as it is placed within it whole.

C. The Court's Plan

1. Population Equality

In keeping with our overriding concern, the court plan complies with the "`as nearly as practicable'" population equality requirement of Article 1, § 2 of the Constitution, Karcher, 462 U.S. at 730, 103 S.Ct. 2653, with a deviation of plus or minus one person. In drawing the court plan, the court consulted both the existing 1994 plan, originally drawn by the court in the Burton litigation and later enacted by the General Assembly, as well as the predecessor Congressional plan from the 1980s. As we did in the House and Senate cases, we generally sought to maintain the cores of the existing congressional districts, adding or subtracting compact and contiguous population as individual district requirements dictated to correct the population deviations.

2. The Voting Rights Act

The parties agree that the 1994 plan enacted by the General Assembly is the benchmark plan for purposes of the Voting Rights Act. This existing congressional plan was substantially drawn by the three-judge court in Burton in 1992, and later codified with slight modifications by the General Assembly. See S.C.Code Ann. § 7-19-40 (Law.Co-op.Supp.2001). In 1996, the Sixth District was challenged under Shaw/Miller principles as an unconstitutional racial gerrymander, see Leonard, et al. v. Beasley, et. al., 3:96 CV 3640 (D.S.C. filed Dec. 6, 1996), but the parties settled the matter without resolving the constitutionality of the Sixth District. Instead, the state defendants agreed to concede that racial considerations predominated over traditional districting principles in the draw of the Sixth District should the Leonard plaintiffs bring another such challenge after the 2000 census and redistricting process was completed.

In this case, no § 2 concerns have been raised by the parties. Indeed, the parties agree that § 2 legally requires maintenance of the Sixth District as a majority-minority district and that its core population is the substantial and predominately black population located in the Pee Dee and adjoining counties of Richland and Charleston. Furthermore, the creation of the district in 1994 by the General Assembly, and the proposed plans of the current General Assembly and Governor, are persuasive evidence that the preservation of the district is now a nonpartisan districting policy of the state.

However, the district is presently over 68,000 persons low in population and, therefore, could only maintain its status as a § 2 district if there is sufficiently compact majority or near majority-black population in areas contiguous to the core of the existing district which shares a community of interest with the existing constituents and in a number sufficient to make up the existing deficit without losing majority-minority status overall. We agree that there is and, therefore, that § 2 and § 5 of the Voting Rights Act require the maintenance of the Sixth District as a majority-minority district. We believe the minority population in the core areas of the Sixth District, as drawn by the court, is sufficiently compact and shares a sufficiently strong community of interest to warrant being a majority-minority district.

Application of the 2000 census figures to the 1994 benchmark plan produces the current BVAP of the Sixth District at 57.78%. Apparently acknowledging the impossibility of drawing a congressional district that maintains this BVAP level, the General Assembly and Governor do not challenge each other's proposed plans on the basis of § 5 of the Voting Rights Act. As noted previously, both plans contain a BVAP of between 53% and 54%. Rather, only Leatherman raises § 5 as an issue in the Congressional case, asserting that we should strive to achieve the current benchmark BVAP in the Sixth District, and proposing two alternative draws that achieve a BVAP of 56.57% and 58.87%, respectively.

In drawing the borders of the Sixth District, we generally followed the existing lines of the 1994 plan. In doing so, we also remedied those aspects of the 1994 plan that appeared to reflect the unnecessary subordination of traditional districting principles to race. The result was our elimination of some of the rougher lines and "fingers" that plague the existing plan and that were apparently the subject of prior challenge in the Leonard proceeding. Having done so, we found that there was still a sufficiently numerous, geographically compact, and politically cohesive minority population in the area to require the preservation of a minority district there.

In drawing the district, however, we discovered that the magnitude of the population shortage in the Sixth District revealed by the 2000 census, coupled with our correction of some of the questionable aspects of the existing plan, only allowed for a constitutionally proper draw that has a 53.75% BVAP in the district. We are satisfied that we have narrowly drawn the district and achieved a BVAP that does not result in a dilution of the effective voting rights of the district's minorities. Indeed, Congressman James Clyburn testified that a BVAP of 53% or above would be sufficient to allow the minority constituency a fair opportunity to elect a non-incumbent black candidate of choice in the district. In sum, the court's draw of the Sixth District does not run afoul of the standards imposed by § 5 of the Voting Rights Act. Although the overall voting strength of the minority voters in the Sixth District is lowered, the result cannot be avoided without running afoul of the strictures imposed upon us by the Equal Protection Clause and is not one which causes "a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise." Beer, 425 U.S. at 141, 96 S.Ct. 1357 (emphasis added); see also Abrams, 521 U.S. at 95, 117 S.Ct. 1925.^[29]

3. Other Considerations in the Draw of District Six

We also considered the areas of dispute among the parties as to the precise lines of the Sixth District. Although our draw of the Sixth District incorporates the same cores of those in the existing plan and the proposed plans, our ultimate plan varies somewhat from all of the others.

Under the current plan, the Second District contains all or portions of 11 counties and the Sixth District contains all or portions of 16 counties. The House plan places the Second District in all or part of 10 counties and places the Sixth District in all or portions of 14 counties. The Governor's plan places the Second District in all or part of 9 counties and places the Sixth District in all or portions of 12 counties. The court's plan places the Second District in all or part of 10 counties and places the Sixth District in all or part of 14 counties. Therefore, the court's plan, like those of the General Assembly and the Governor, improves upon the current plan in at least this limited respect.

The major differences between the General Assembly and the Governor stem from the proffered splits of Orangeburg, Calhoun, and Georgetown Counties. The General Assembly proposed to maintain the existing splits of Orangeburg and Calhoun Counties between the Second and Sixth Districts, eliminate the split of Colleton County and place it wholly within the Sixth District, and split Georgetown County between the Sixth and First Districts. The Governor, on the other hand, proposed to eliminate the existing splits of Orangeburg and Calhoun Counties and place those counties wholly within the Sixth District; eliminate the existing split of Colleton County, but place it wholly within the Second District; and keep Georgetown County wholly within the First District.

In the court's plan, Darlington County is removed from the Sixth District and placed wholly within the Fifth District. Beaufort County is placed wholly within the Second District, eliminating that split as well. The split of Colleton County is eliminated by placing Colleton County wholly within the Sixth District, but a split is created in Georgetown County between the Sixth District and the First District.

With regard to Orangeburg County and Calhoun County, the court considered eliminating the existing splits, but elected to maintain the status quo in those areas by maintaining the splits (albeit in a slightly cleaner fashion), and to provide needed population to the Sixth District instead from the remainder of Colleton County residents. According to the evidence, the western portions of Orangeburg County and Calhoun County are an important part of the existing core of the Second District. Indeed, in the 1980s, and for most of the twentieth century, Calhoun County and Orangeburg County were located wholly within the Midlands-based Second District. It was the creation of the majority-minority district compelled by § 2, and first drawn by the Burton court in 1992, that removed the eastern portions of those counties from the Second District, adding the predominately black population in that area to the compact and contiguous majority-black population of the Pee Dee to create the Sixth District. As a result, the Second District was forced to partially migrate away from its Midlands base, picking up Allendale, Hampton, Jasper, and Beaufort counties.

The legislature's draw after Burton also necessitated removal of a portion of Colleton County from its home in the First District (a predominately coastal district) and placement of it within the Second District. Thus, unlike the western portions of Orangeburg County and Calhoun County, which have always been in the Second District, the Colleton County constituents in the Second District are relative newcomers, placed there as a result of compliance with the Voting Rights Act in the 1990s.

This court, therefore, was faced with a number of competing considerations in this area. Viewing the Sixth District in isolation from its surrounding districts, we considered adding the needed population from the contiguous majority-black area of Colleton County, as well as the contiguous majority-black counties of Allendale, Hampton, and Jasper, currently in the Second District. But, this decision would have isolated Beaufort County from the Second District, necessitating its inclusion in the First District and resultant large changes to the cores of all three districts. Accordingly, we elected to maintain the split of Orangeburg and Calhoun Counties, which preserved these core areas of the Second District and caused the fewest changes elsewhere.

Our decisions in these areas did, however, impact Georgetown County. Under the current plan, Georgetown County is located wholly within the First District. Under the General Assembly plan, Georgetown County is split between the First and Sixth districts. The Governor proposed that Georgetown County be contained wholly within the Sixth District.^[30] The court plan splits Georgetown County (albeit in a different fashion than that proposed by the General Assembly plan), dividing its residents between the Sixth District and First District.

Weighing all the competing considerations, we believe the county splits in our plan, however unfortunate they may seem in isolation, best serve the concurrent goals of achieving population equality, preserving the cores of the existing congressional districts, and maintaining the Sixth District as a majority-minority district under the Voting Rights Act while recognizing state redistricting principles.

4. Fort Jackson and the Savannah River Site

We also considered, but rejected, Governor Hodges' proposed relocation of Fort Jackson from the Second District to the Sixth District in his Plan B. Congressman Clyburn testified that he requested the relocation of Fort Jackson from the Second District to the Sixth District upon the death in the fall of 2000 of Congressman Floyd Spence, who represented the Second District. Prior to his death, Congressman Spence served as Chairman of the House Armed Services Committee. The Second District is now represented by Congressman Joe Wilson, who was elected in a special election after Congressman Spence's death.

According to Congressman Clyburn, who testified at trial, as a senior member of Congress and a member of the House Appropriations Committee, he is now in a better position than the recently elected Congressman Wilson to serve the interests of Fort Jackson. Not surprisingly, Congressman Wilson expressed a contrary opinion. He testified that Fort Jackson should remain in its historical place in the Second District and that, in any event, he has now been elected a member of the House Armed Services Committee and can serve the Fort's interests as well or better than Congressman Clyburn.

We express no opinion on the issue of whether Congressman Clyburn or Congressman Wilson is best suited to serve the interests of Fort Jackson. We are confident that both men will attempt to serve this important South Carolina interest regardless of its district home. However, even were we to agree that the move had some political benefit, such an important change to the core of an existing district in a redistricting plan, based upon nothing more than our determination that one elected official will do a better job than another, is clearly beyond the scope of our remedial authority. The Governor's attorney has implicitly conceded the same, admitting that the requested relocation of Fort Jackson has no relevance to the task of remedying the malapportionment of the existing congressional districts in a manner consistent with the mandates of the Voting Rights Act. We think that obvious truth, and admirable concession, ends the matter. For largely the same reasons, we also rejected the Governor's proposal to move the Savannah River Site entirely from the Third District to the Second District.

VII. Conclusion

In drawing the plans to remedy South Carolina's unconstitutional districting plans, we have approached our task with great concern, attempting to apply federal redistricting principles in such a way as to preserve, where possible, the status quo in South Carolina and to minimize the damage to existing districts where large population changes have occurred. We have also attempted to adhere to the requirements that the Voting Rights Act would impose upon the South Carolina legislature in the redistricting context, in order to avoid the imposition of a plan that might have the unintended effect of diluting the voting strength of South Carolina's minority population or a plan that unnecessarily retrogresses the fragile gains minority voters have achieved over the past two decades in this state. Like our predecessors faced with the task in the 1990s, we encourage the General Assembly and the Governor to work together to adopt any plan that could improve upon what we have done. They are, of course, in the best position to do so. We have done our best in the interim.

ACCORDINGLY, IT IS HEREBY ORDERED that the State of South Carolina is hereby enjoined from conducting any further elections under the existing electoral districts for the South Carolina House of Representatives, the South Carolina Senate, and the United States Congress.

IT IS FURTHER ORDERED that the redistricting plans for the South Carolina House of Representatives (set forth at Exhibit A), for the South Carolina Senate (set forth at Exhibit B), and for the South Carolina Congressional Delegation (set forth at Exhibit C)^[31] shall be the lawful election districts for each of those bodies for the elections scheduled in 2002 and for all subsequent elections until the South Carolina General Assembly, with the approval of the Governor and in accordance with § 5 of the Voting Rights Act, ends its current impasse and enacts a redistricting plan for any or all of them.^[*]

IT IS SO ORDERED.

ORDER OF CLARIFICATION

By order dated March 20, 2002, this three-judge panel issued an order declaring the existing electoral districts for the South Carolina Congressional Delegation to the United States Congress, the South Carolina House of Representatives, and the South Carolina Senate unconstitutionally malapportioned and enjoining the State of South Carolina from conducting any further elections under the existing electoral schemes. Such action, the court held, was necessary because the elected officials of South Carolina had failed to redistrict the state's General Assembly and its six Congressional Seats and were at an impasse on the issue, and there was no chance that the governing officials would be able to reach a compromise in time for the regularly-scheduled primary and general elections scheduled for June 11, 2002, and November 5, 2002, respectively.

On April 4, 2002, the South Carolina Attorney General advised the court that two special elections were currently scheduled to fill two existing vacancies in the South Carolina General Assembly — one in the South Carolina House of Representatives and one in the South Carolina Senate. While advocating no position on the issue, the Attorney General expressed some question as to whether the two special elections could proceed under the existing electoral districts as planned without running afoul of this court's March 20, 2002 redistricting plan and order.

Although the State of South Carolina is not technically a party to this litigation, the court is sensitive to the unusual circumstances of this case and to the State's request that we clarify our intentions under the March 20th ruling in order to ensure that the upcoming special elections and any future special elections not be unnecessarily subjected to challenge under our decision. Accordingly, we agree that a clarification of our ruling is appropriate.

First, although not raised as an issue by any of the parties before it, the court specifically asked all parties during the trial whether the special election for the South Carolina Senate seat — which had been vacated by Congressman Joe Wilson and of which the court had become aware — would be impacted by this litigation. The Governor's counsel affirmatively represented that it was not so impacted, and no other parties raised a contrary position.^[1]

Second, to the extent our March 20, 2002 order can be read to require the pending special elections to be conducted under the 2002 court-ordered redistricting plans, or that any other special elections held prior to the next round of regularly-scheduled elections be conducted under the 2002 court-ordered redistricting plans, we disavow that intent. Such an interpretation would be neither prudent nor practical. In every case, the new district lines omit some persons and/or add others and, in the case of two state House districts, the districts have been completely collapsed and moved to high-growth areas of the state. If special elections held prior to the next round of regularly-scheduled elections were to be held under the new district lines, the effect would be to leave some persons with no representation and others with double representation.^[2] Such a situation is obviously not one that the United States Constitution would allow, much less require. Cf. Gaona v. Anderson, 989 F.2d 299 (9th Cir.1993) (per curiam) (rejecting Voting Rights Act challenge to use of old version of state senate district for an interim special election); Political Action Conference of Illinois v. Daley, 976 F.2d 335 (7th Cir.1992) (holding that city need not alter its redistricting scheme even though four-year terms of aldermen elected in 1991 resulted in a four-year delay in using new 1990 census data); French v. Boner, 963 F.2d 890 (6th Cir.1992) (holding that city had no constitutional duty to re-conduct elections held after the new decennial census data became available in 1991, but before a new apportionment plan could be put into effect).

Under our March 20, 2002 order, the State of South Carolina is only enjoined from conducting the regularly-scheduled 2002 primary and general elections under the existing electoral districts for the United States Congress and for the South Carolina House of Representatives, and after the November 2002 General Election, from conducting any further elections under the existing electoral districts for the United States Congress and the South Carolina House of Representatives. All such regularly-scheduled elections, and all special elections held after the November 2002 General Election, must be conducted in accordance with the redistricting plans for the South Carolina Congressional Delegation and South Carolina House of Representatives adopted by this court in its March 20, 2002 order, unless and until the South Carolina General Assembly, with the approval of the Governor and in accordance with § 5 of the Voting Rights Act, ends its current impasse and enacts an alternative redistricting plan for the legislative body at issue.

The State of South Carolina is likewise enjoined from conducting the regularly-scheduled 2004 primary and general elections under the existing electoral districts for the South Carolina Senate and, after the November 2004 General Election, from conducting any further elections under the existing electoral districts for the South Carolina Senate. All such regularly-scheduled elections, and all special elections held after the November 2004 General Election, must be conducted in accordance with the redistricting plans for the South Carolina Senate adopted by this court in its March 20, 2002 order, unless and until the South Carolina General Assembly, with the approval of the Governor and in accordance with § 5 of the Voting Rights Act, ends its current impasse and enacts a redistricting plan for the South Carolina Senate.

IT IS SO ORDERED.

NOTES

[1] The redistricting process in South Carolina has historically been a troubled one. Much of that history is discussed in detail in Burton v. Sheheen, 793 F.Supp. 1329, 1337-40 (D.S.C. 1992), vacated sub nom. Statewide Reapportionment Advisory Comm. v. Theodore, 508 U.S. 968, 113 S.Ct. 2954, 125 L.Ed.2d 656 (1993), in which a three-judge court was called upon to address a near-identical impasse for all three elective bodies following the release of the 1990 decennial census, and in Smith v. Beasley, 946 F.Supp. 1174, 1177-79 (D.S.C.1996), in which a three-judge court was called upon to address the constitutionality of nine House of Representative election districts and three Senate election districts drawn by the South Carolina General Assembly after the Burton court plans were vacated by the Supreme Court.

[2] The original complaint was brought by the Colleton County Council, Steven Murdaugh, individually and in his official capacity as Chairman of Colleton County Council, Dr. Joseph Flowers, individually and in his official capacity as a member of Colleton County Council, and by fourteen additional residents of Colleton County. Colleton County, as well as Steven Murdaugh and Dr. Joseph Flowers in their official capacities, were subsequently dismissed by order of this court.

[3] Mr. Bowers served in the capacity of technical advisor to the South Carolina District Court in the previous redistricting litigation in this state and has extensive experience in the area. See Burton, 793 F.Supp. at 1339.

[4] Because regular elections for the South Carolina Senate seats will not be held until 2004, we considered sua sponte the propriety of implementing a remedial plan for the Senate at this time, due to the possibility that the impasse could be broken after the Gubernatorial and House elections in November 2002. However, we concluded that the matter is ripe. The existing Senate plan was challenged as unconstitutional under the one-person, one-vote mandate, the parties agreed that the existing plan is unconstitutionally malapportioned (and offered no opposition to the entry of summary judgment as to this issue), and the parties stipulated that the Governor and General Assembly are indeed at an impasse. Principles of judicial economy also counseled us to act upon the matter. The parties and the court have expended time and other resources preparing to address the matter without objection and, in any event, the court-imposed plan is only an interim remedy; the plan is only enforceable unless and until the South Carolina General Assembly, with the approval of the Governor and in accordance with § 5 of the Voting Rights Act, ends its current impasse and adopts its own redistricting plan.

[5] The Supreme Court has also held that courts must utilize single-member districts "[u]nless the ... [c]ourt can articulate ... a singular combination of unique factors" that justifies a different result. Chapman v. Meier, 420 U.S. 1, 21, 95 S.Ct. 751, 42 L.Ed.2d 766 (1975) (internal quotation marks omitted). The South Carolina General Assembly is already organized exclusively into single-member districts.

[6] For simplicity's sake, we refer to the Leatherman plaintiffs simply as "Leatherman" and the Marcharia plaintiffs as the "ACLU."

[7] See, e.g., Hunt v. Cromartie, 526 U.S. 541, 551, 119 S.Ct. 1545, 143 L.Ed.2d 731 (1999) (recognizing that a redistricting plan can be driven, at least to some extent, by partisan interests); Smith, 946 F.Supp. at 1206 (noting that "[t]he drawing of district lines for political purposes has often been criticized, but it is not illegal").

[8] The citizens of Colleton County and of Georgetown County have offered plans that substantially incorporate the state-wide deficiencies of the other proffered plans, making unique modifications in their areas which address their particular localized concerns. On the whole, these proffered modifications do not appear to be politically motivated, but rather driven by county-based desires. Accordingly, we address their unique requests in isolation and, where appropriate, incorporate their requests into our plans.

[9] Because the parties have raised the issue of constructing their plans by Voting Tabulation Districts (VTDs) and precincts, we note that in many cases there is a difference between the two. VTDs are geographical boundaries established by the United States Census Bureau prior to collecting data for the decennial census. Ideally, the VTDs would supply data from the census which coincide with precinct boundaries. However, precinct boundaries in many instances do not follow either natural or manmade features. While VTD boundaries are established prior to the census and remain in place for the decade, precincts are units created by the General Assembly for use in elections and for the collection of election data on a county-by-county basis. The latter may be changed at the will of the legislature and, in fact, there have already been a number of precinct changes since the creation of the VTD boundaries that were used in the 2000 decennial census. The end result, of course, is that VTDs and precincts are not synonymous in all instances, but any precinct splits that exist in the court plans can be remedied in any legislative session.

[10] A majority-minority district is one in which the minority population comprises greater than 50% of the total population in the district. It is undisputed that the black population in South Carolina is the only minority present in significant concentrations in any area to warrant protection under the Voting Rights Act. Accordingly, we use the terms "minority" and "black" interchangeably.

[11] In City of Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980), the Supreme Court held that a vote dilution claim under § 2 was only actionable if the challenged practice was adopted or maintained for the purpose of discriminating against minorities. In response, Congress added subsection (b) of § 1973 in 1982 to forbid practices having either the purpose or the effect of preventing minorities from electing representatives of their choice.

[12] Although Gingles actually set forth the requirements for establishing a § 2 challenge to a multi-member districting plan, see Gingles, 478 U.S. at 50-51, 106 S.Ct. 2752, the Supreme Court has subsequently held that the same three prerequisites are necessary to establish a § 2 claim with respect to singlemember districts. See Growe v. Emison, 507 U.S. 25, 39-42, 113 S.Ct. 1075, 122 L.Ed.2d 388 (1993); Voinovich, 507 U.S. at 158, 113 S.Ct. 1149.

[13] The legislative history of § 2, particularly the Senate Report, sets forth a variety of factors considered relevant in determining if a legislatively enacted plan violates § 2 because it "results" in discrimination. These factors include:

1. the extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process;

2. the extent to which voting in the elections of the state or political subdivision is racially polarized;

3. the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;

4. whether the members of the minority group have been denied access to any candidate slating process;

5. the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process;

6. whether political campaigns have been characterized by overt or subtle racial appeals; and

7. the extent to which members of the minority group have been elected to public office in the jurisdiction.

See Thornburg v. Gingles, 478 U.S. 30, 43-45, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986) (citing S.Rep. No. 417, 97th Cong.2d Sess. 28-29 (1982), 1982 U.S.Code Cong. & Admin.News, p. 177).

[14] The three-judge panel in Smith provided a helpful summary of the factors that have been identified by the Supreme Court as evidencing race-based line drawing:

They are: using land bridges in a deliberate attempt to bring black population into a district (Miller); use of sophisticated computers with block-by-block racial data (Bush); evidence of demographer's ... purpose in drawing the challenged lines (Shaw and Miller); creation of new, noncompact and oddly shaped districts beyond those necessary to avoid retrogression (Miller and Shaw); creation of districts that exhibit disregard for city limits, local election precincts, and voting tabulation districts (Bush); districts that wind "in snakelike fashion" until enough black neighborhoods are included to create a black-majority district (Shaw I); establishment of a fixed percentage of minority population with[in] a district ... to establish a safe black district (Bush); evidence that race or percentage of race could not be compromised (Shaw II); statements by legislators indicating that race was the predominant factor in creating the challenged districts (Miller); and statements made in submission for preclearance to the Department of Justice that its purpose was to comply with the dictates of the Department's rejection letter (Shaw II).

Smith, 946 F.Supp. at 1207.

[15] Justice O'Connor reconciled the competing considerations involved in the review of equal protection challenges made in the context of a Voting Rights case as follows:

Today's decisions, in conjunction with the recognition of the compelling state interest in compliance with the reasonably perceived requirements of § 2, present a workable framework for the achievement of the[] twin goals [of "combating the symptoms of racial polarization in politics and eliminating unnecessary race-based state action"]. I would summarize that framework, and the rules governing the States' consideration of race in the districting process as follows.... [S]o long as they do not subordinate traditional districting criteria to the use of race for its own sake or as a proxy, States may intentionally create majority-minority districts and may otherwise take race into consideration, without coming under strict scrutiny. Only if traditional districting criteria are neglected and that neglect is predominately due to the misuse of race does strict scrutiny apply.

Bush, 517 U.S. at 993, 116 S.Ct. 1941 (O'Connor, J., concurring) (emphasis added; citations omitted). Under Justice O'Connor's view, states may take race into consideration and intentionally create majority-minority districts without triggering strict scrutiny, so long as the state does not allow the use of race to predominate over the use of traditional districting considerations. However, "districts that are bizarrely shaped and noncompact, and that otherwise neglect traditional districting principles ... for predominately racial reasons are unconstitutional." Id. at 994, 116 S.Ct. 1941.

[16] Districts drawn by a state legislature which deviate from traditional districting principles for the purpose of serving partisan politics do not trigger the strict scrutiny analysis. See Easley v. Cromartie, 532 U.S. 234, 257, 121 S.Ct. 1452, 149 L.Ed.2d 430 (2001). But as noted previously, this court may not draw its plans to serve partisan purposes.

[17] In Bush, five members of the Court "assumed without deciding" that compliance with § 2 of the Voting Rights Act is a compelling state interest. 517 U.S. at 977, 116 S.Ct. 1941 (plurality opinion); id. at 1003, 116 S.Ct. 1941 (concurring opinion of Thomas, J., joined by Scalia, J.). Justice O'Connor, however, who authored the plurality opinion, also wrote a separate concurring opinion in which she expressed her opinion that compliance with the Act is a compelling state interest, see id. at 992, 116 S.Ct. 1941 (concurring opinion of O'Connor, J.), a view that seems to be shared by the four dissenting justices as well, see id. at 1004, 116 S.Ct. 1941 (dissenting opinion of Stevens, J., joined by Ginsberg and Breyer, JJ.); id. at 1065, 116 S.Ct. 1941 (dissenting opinion of Souter, J., joined by Ginsberg and Breyer, JJ.).

[18] Dr. Ruoff examined 401 elections in the period from 1992 to 2000, including 234 black-white elections, 142 white-white elections, 24 black-black elections and 1 Hispanic-white election, using correlation analysis, double-equation ecological regression analysis, and extreme or homogenous precinct analysis, which are the generally accepted techniques for analyzing racial polarization. See, e.g., Gingles, 478 U.S. at 52-53, 106 S.Ct. 2752.

[19] Dr. Loewen testified that, based upon his study of black-white contests, 85% of the whites voted white and 84% of the blacks voted black (compared to 90% and 85%, respectively, in the 1970s and 1980s).

[20] In analyzing this same data, which remains the most recent, the Smith court noted that "there is a socio-economic gap between the average white citizen and the average black citizen. There is a larger percentage of blacks than whites below the poverty level; the household income of blacks is generally less than that of whites; unemployment is greater among blacks; and the level of formal education among blacks is less.... More blacks than whites are without private means of transportation, more whites than blacks own their own homes. Infant mortality is greater among blacks." Smith, 946 F.Supp. at 1203.

[21] The 2000 census data, unlike that of prior decades, reports the black population in several racial categories, including "black only," which would report the least numerous count, "all or any part black" (AP-Blk), which would report the most numerous count, and "non-Hispanic black," which would fall somewhere between the two. The parties differed in their choice of category for purposes of analyzing the black population in this case; however, the use of the differing black population categories has an insignificant impact on the analysis before this court. Like the House and the Governor, we use the "black only" category in our order.

[22] Because of the high level of racial polarization in the voting process in South Carolina, "influence districts" allow the Democratic Party the opportunity to gain control of the General Assembly. It is, therefore, an inherently politically based policy. With the aid of a substantial (but not majority) black population that votes nearly exclusively for a Democratic candidate, a white Democrat can usually defeat a black Democrat in the primary election and then use the black vote to defeat any Republican challenger in the general election. See Smith, 946 F.Supp. at 1183. Although the Governor asserts that a draw based on "traditional districting principles" will naturally lead to the creation of such districts, he does not advance a claim that this court can or should consider race to intentionally draw such "influence districts."

[23] An individual district's deviation is the difference in total population from the ideal district size. A plan's total population deviation is derived by adding the largest positive deviation in the plan and the largest negative deviation in the plan.

[24] As previously noted, the Governor specifically denied that he was claiming a right to influence districts under § 2 of the Voting Rights Act, but rather argued that the House plan could not be adopted by us because it was the product of "white gerrymandering" in violation of the Equal Protection Clause.

[25] The ACLU admitted that their plans did not consider any community of interest factors or any political factors other than incumbency protection.

[26] Other areas of substantial growth included Lexington County, which grew 28.9%, and York County, which grew 25.2%. Conversely, many areas in the Pee Dee and Low Country lost population: Allendale County lost 4.4%, Bamberg County lost 1.4%, Marlboro County lost 3.0%, and Union County lost 1.5%.

[27] As noted previously, there is no real dispute that the remaining Gingles prerequisites — racially polarized and white bloc voting — exist statewide.

[28] As noted in more detail infra, District 7 is a majority-minority district in terms of total black population only. Thus, the benchmark plan contains ten majority-minority districts, if defined as districts with a total black population that exceeds the 50% mark, but only nine majority-minority districts that exceed the 50% mark in terms of BVAP. Unless otherwise noted, our reference to majority-minority districts hereafter refers to those districts that contain a BVAP greater than 50%. An opportunity district, in contrast, is one in which it is demonstrated (usually by election data and expert testimony interpreting it) that the BVAP percentage in a district is sufficient to actually afford the minority population an equal opportunity to elect the candidate of its choice.

[29] We note that the court attempted several alternative draws of the Sixth District that would maintain it as a Pee Dee and Low Country district and eliminate the splits of either Richland County or Charleston County, or both. However, we found that it was not possible to create a majority-minority district that achieved a greater than 53% BVAP without splitting nearly every adjoining county along racial lines and causing major changes in every other congressional district in the state.

[30] Interestingly, although Georgetown participated in the other phases of this litigation to advocate its particular interests, the citizens of Georgetown did not propose a plan nor otherwise participate in the Congressional phase.

[31] A compact disc of the court-proposed districts with population summary statistics, including block equivalency files based on the Census 2000 PL94-171 data file, are attached as Exhibit D.

[*] Due to publication problems, Exhibits A, B, C & D have not been reproduced for publication.

[1] Pursuant to the court's inquiry, the following colloquy took place:

Judge Traxler: Is the special election in Lexington County to fill Senator's Wilson's seat impacted by this litigation? Does everybody agree there is no impact on that particular election in this litigation?

Mr. Parks: That's right.

Mr. Gergel: (Nods head in the affirmative). (Transcript of Trial, February 1, 2002, p. 7).

In response to the Attorney General's request, all parties have remained consistent with their position at trial on this issue — except the ACLU, which has now taken the position that the court-ordered plans should apply to all special elections which occur prior to the next applicable round of regularly-scheduled elections. We believe the ACLU is barred from raising a contrary position at this late date. However, our decision to clarify the order pursuant to our authority under Fed.R.Civ.P. 60(a) should suffice to address their position.

[2] For example, under the March 20, 2002, redistricting plan for the South Carolina House of Representatives, House District 68, located in the severely under-populated Pee Dee area, was collapsed and moved to a substantially over-populated coastal area in Horry County. If the current representative of House District 68 were to leave office and a special election conducted, the new representative would be elected from Horry County, leaving Horry County residents in the "new" District 68 with two representatives in the House and "old" District 68 residents in the Pee Dee with none.