ORDER

HAND, Senior District Judge.

The matter before the Court is a districting plan for the election of the Dallas County Commission to replace the five single member plan invalidated by this Court in its Order and Judgment of March 29, 1999, (Docs. 136 and 137). The Court, in its Order of February 10, 2000, (Doc. 199), rejected the plan presented jointly by the Dallas County Commission defendants and the United States Department of Justice, as well as the plan presented by the plaintiffs, because both made greater use of race than is permitted by the constitution. The Court also announced its intentions to produce its own plan and notified the parties that it would be assisted in this effort by Professors Katharine I. Butler and Harold W. Stanley.

The Court announced its intent on March 21, 2000 (Doc. 206) to impose its own Court Plan # 3 and submitted this plan and the process from which it culminated to the parties for their review and comment. A hearing was held on March 30, 2000, to identify and resolve any technical errors or defects in the Court's proposed plan.

In response to concerns expressed by County Commissioner Curtis Williams, the Court's plan was modified slightly to place him a district with more of his constituents. It was then provided to the parties with the possibility that slight modifications might need to be made if some of the district boundaries selected by the experts using census data proved for some reason unworkable. At the recommendation of the Probate Judge, who is responsible for determining into which district voters are to be placed, a few additional changes were made. The Dallas County Commission defendants and the Probate Judge were instructed to cause the adjusted Court Plan # 3 to be posted in a conspicuous place and published in a manner which would notify the public and potential candidates of the districts for which the elections would be conducted.

This opinion will address in greater detail not only the invalidity of the parties' respective proposed plans but the propriety of the Court's remedial plan.

I. FINDINGS OF FACT CONCERNING THE PARTIES' PLANS A. Background

In 1987, the at-large method of electing Dallas County Commissioners was found to have diluted the voting strength of the county's black citizens and thus violated § 2 of the Voting Rights Act. Under state law, the Dallas County Commission consisted of four commissioners elected at large, from residency districts. The Probate Judge was the ex-officio fifth member of the Commission, and by statute voted on matters before the Commission only to break a tie.

To remedy the dilution, this Court ordered that the four commissioners be elected from single member districts, two of which were majority black in voting age population. However, the Court's remedy was overturned on appeal on the grounds that the continued at-large election of the Probate Judge was dilutive. The Eleventh Circuit, on its own motion, sketched out a five single member district plan, replacing the probate judge with a fifth commissioner. The Eleventh Circuit's plan was based on a proposal of the Government's expert, Alan Lichtman. In accordance with the directive of the Eleventh Circuit, this Court filled in the details of the plan and ordered its implementation. Slight modifications were made to accommodate populations shifts after the 1990 census.

In the instant case, plaintiffs, who were not parties to the prior litigation, challenged the Eleventh Circuit's authority to change the number of commissioners from four to five and to remove the Probate Judge as an ex-officio member of the Commission, contrary to state law. This Court agreed that the Eleventh Circuit's remedy exceeded the Court's authority in light of Holder v. Hall, 512 U.S. 874, 114 S.Ct. 2581, 129 L.Ed.2d 687 (1994), and ordered the parties to produce remedial plans which were in accordance with state law, meaning four commissioners were to be elected from single member districts and the Probate Judge was to be restored to his position as an ex-officio member and presiding officer of the commission.

Pursuant to this Court's Order of April 22, 1999 (Doc. 144), the Dallas County Commission published notice to the public that the parties and all other interested persons were directed to devise a new election plan under which four members of the Dallas County Commission would be elected from single member districts and to conduct a public hearing on July 13, 1999, to receive and preserve comments and/or evidence in favor of and in opposition to each of the submitted plans. Prior to the commencement of the required public meeting, four plans had been presented and posted for review: a plan submitted by the plaintiffs; a plan submitted by the United States Department of Justice under seal as merely an "illustrative" plan but ordered to be unsealed and made public by the Court (hereinafter referred to as the "Arrington 1 Plan"); a plan presented by the Dallas County Commission and identified as "Lillie Plan 2"; and a plan identified by the Commission as the "Citizen's Plan." See, Dallas County Commission Defendants' Brief and Redistricting Plan (Doc. 172) filed August 2, 1999, at 4.

Only two of these plans were discussed at the July 13, 1999, public hearing, the plaintiffs proposed plan and the plan submitted by the Department of Justice. The plaintiffs presented their proposed plan and the testimony of Mr. Kin Reynolds from the Reapportionment Office in Montgomery, Alabama, who actually drew the plan. Dr. Theodore Arrington, an expert hired by the Department of Justice, presented the Department's illustrative plan.

The Dallas County Commission defendants ultimately withdrew their proposed plan essentially conceding that it did not comply with the one person one vote requirement and posed serious concerns of continuity as charged by the United States. See e.g., United States' Memorandum (Doc. 171) filed on August 2, 1999 at 4, n. 1. The Citizen's Plan was neither discussed at the public hearing on July 13, 1999, nor presented to the Court.

The parties met on July 15, 1999, in an effort to iron out their differences and reach a consensus but were unable to agree on a plan. At the behest of the Department of Justice, however, Dr. Arrington drafted a second districting plan based on a plan prepared by the County in its negotiations at the July 15th meeting. Dr. Arrington's new plan ("Arrington 2") was subsequently presented to the Commissioners, who voted to adopt it as their own. The Department of Justice provisionally precleared its expert's plan, which was then submitted to the Court as the County's proposed remedial plan. See, Doc. 171 and accompanying Exhibits, including the Supplemental Declaration of Dr. Arrington dated July 18, 1999.

B. The Plaintiff's Plan

It is unnecessary to discuss the plaintiffs' plan (Doc. 164 with attachments) in any depth as it was rejected by both the Dallas County Commission and the United States Department of Justice in part for its impermissible use of race and its failure to accommodate the incumbent County Commissioners to any extent. The four districts in the Plaintiffs' Plan contained the following total population and voting age population by race:

C. The Arrington 2 Plan

The four districts in Dr. Arrington's Plan contained the following total population and voting age population by race:

The most significant of Dr. Arrington's adjustments were those designed to assure what he described as an "equal political opportunity" for "both races." He concluded that the means to do this was to manipulate the racial makeup of the districts.

Dr. Arrington's original goal, reflected in his first "illustrative plan," rejected by the Commissioners for other reasons, was to "draw two districts that would be a clear opportunity for minority citizens . . . to elect candidates of their choice, . . . [o]ne district which would allow white citizens to elect a candidate of their choice and one district which was at sixty-one percent [black]." [Tr. 109]. He characterized the sixty-one percent black voting age majority district in his illustrative plan as a "swing district," by which he meant that it could elect either a black or a white. The only explanation he offered for why a district in which blacks made up 61% of the potential voters was not a clear "opportunity" district was that past discrimination and lower socio-economic status of blacks meant that blacks were not mobilized to vote at the same rate as whites and that black voters were less able to raise money.

As can be seen from the statistics above, Arrington 2, the plan ultimately adopted by the Dallas County Commission, also contained three majority black districts. The least black of the three was 57% black in voting age population, and was the plan's swing district. The Commission's July 15th plan also contained three majority black districts, one that was 62.7% black in voting age population and two that were 54% black in voting age population. Dr. Arrington increased the black percentages in the two 54% districts to "create districts which more equally represented all the races" and further testified that he indeed "did have a racial motive in that progression." [Tr. 154].

Dr. Arrington conceded that the plans he had drawn with three majority black districts out of four might be seen as "tilted slightly in favor of black voters and against white voters." [Tr. 113]. He justified the tilt in the districts by "taking into account the Probate Judge [who] would be the presiding officer . ." [and] ". . . most likely [to be] supported mainly by white voters." [Id.]. This fact was important, he opined, because he considered "the Probate Judge to be a full participant in the Count Commission." [Id.].

Dr. Arrington offered no race-neutral rationale for his decision to manipulate the black percentage of three of the four districts. His over-arching goal was to design "a Commission that has at least two members who are elected with the support of white voters and at least two who are elected with the support of black voters and perhaps one district which can swing either way." [Tr. 134]. According to his reasoning, one district would elect a white candidate, two districts would elect black candidates, and the Probate Judge would be a white candidate. The remaining district, with a 57% black voting age population, would swing. In short, he believed it was necessary to provide "both races" with a chance to elect a controlling number of commissioners, which could only be accomplished, in his opinion, by improving black voters' chances to elect an additional commissioner from a single member district.

His rationale for treating the Probate Judge as just another commissioner for the purpose of "control of the commission" was that the Probate Judge could vote to break a tie, which was obviously possible given that there were an even number of commissioners. Thus, Dr. Arrington reasoned, the Probate Judge had "exactly the same voting power" as the commissioners. [Tr. 158]. His rationale for "counting" the Probate Judge in the "white" column was that he was not likely to be the choice of the black community. Dr. Arrington was not concerned, he said, with whether the Probate Judge would ignore the concerns of black voters, but solely with the fact that he was not likely to receive the support of a majority of black voters. [Tr. 165-166, 168].

II. CONCLUSIONS OF LAW RELEVANT TO THE VALIDITY OF THE ARRINGTON 2 PLAN

Redistricting is a function committed to the State. Chapman v. Meier, 420 U.S. 1, 27, 95 S.Ct. 751, 766, 42 L.Ed.2d 766 (1975). Consequently, even when a jurisdiction's plan has been held to violate federal law, the affected body should be given an opportunity to propose a remedy, which should be adopted unless it too violated federal law. Wise v. Lipscomb, 437 U.S. 535, 542-43, 98 S.Ct. 2493, 2498-99, 57 L.Ed.2d 411 (1978). Moreover, when the Court is forced to assume the unwelcome task of creating election districts, it should make only such changes in the jurisdictions plan as are necessary to correct the constitutional or statutory defect. Upham v. Seamon, 456 U.S. 37, 43, 102 S.Ct. 1518, 1522, 71 L.Ed.2d 725 (1982).

Thus, a legislative body is entitled to considerable deference in the manner it chooses to remedy problems with its districting scheme. That deference, however, does not extend to proposed remedies which violate the constitution or other federal laws.

A jurisdiction that deliberately uses the race of voters to construct election districts violates the equal protection clause unless it demonstrates that such use has been "narrowly tailored to [achieve] a compelling state interest." Miller v. Johnson, 515 U.S. 900, 920, 115 S.Ct. 2475, 2490, 132 L.Ed.2d 762 (1995). The mere fact that race-based districts are regular and compact does not eliminate the constitutional issue. Evidence that a jurisdiction ignored standard districting criteria, combined with bizarrely shaped districts, may point to a racial motive, but the constitutional evil is the dominance of racial concerns in the process, not the appearance of the end result. Miller, 515 U.S. at 912-13, 115 S.Ct. at 2486-87, explaining, Shaw v. Reno, 509 U.S. 630, 644-49, 113 S.Ct. 2816, 2825-28, 125 L.Ed.2d 511 (1993) ("Shaw I"). When the evidence demonstrates that race is the overriding, predominant force in the construction of a districting plan, the Court must reject the plan unless the State satisfies strict scrutiny, the most rigorous and exacting standard of review. Miller, 515 U.S. at 920, 115 S.Ct. at 2490. See also, Shaw v. Hunt, 517 U.S. 899, 907, 116 S.Ct. 1894, 1901, 135 L.Ed.2d 207 (1996) ("Shaw II").

Ordinarily, it is the plaintiffs' burden to demonstrate, through circumstantial evidence of a district's shape and demographics, or through direct evidence of legislative purpose, that race was the primary factor motivating the decision to include a significant number of voters inside or outside a district. Shaw II, 517 U.S. at 905, 116 S.Ct. at 1900, citing, Miller, 515 U.S. at 916, 115 S.Ct. at 2488. When, however the districting plan is offered as a replacement for one invalidated by the court and will be implemented solely by virtue of the court's power, the court has an independent duty to assess its constitutionality, and cannot ignore substantial evidence of improper racial motivation.

A jurisdiction, particularly one covered by the special provisions of the Voting Rights Act, has a compelling interest in compliance with § 5 and § 2 of the Voting Rights Act, which in many cases will require sensitivity to, and even actual consideration of, the racial make-up of election districts. However, the judiciary retains an independent obligation to ensure that the State's use of race has been narrowly tailored to achieve compliance with the Act. Miller, 515 U.S. at 922, 115 S.Ct. at 2491. The State cannot, under the guise of concern for the Voting Rights Act, manipulate the racial make up of its election districts beyond that which is in fact necessary to accomplish compliance.

The basis of Dr. Arrington's very precise manipulation of the racial make-up of the districts in the Arrington 2 plan was the "need" to provide an equal opportunity for blacks to elect candidates of their choice, which the Court understands to mean compliance with Section 2 of the Voting Rights Act. The Court concludes that a judicial finding of racial vote dilution — a race-based violation of federal law —, even one over a decade old, is prima facie evidence that the County has a compelling interest in continuing a race-based remedy.

The more difficult issue under such circumstances is whether the use of race is narrowly tailored to achieve that objective. As the Supreme Court indicated in Miller, "compliance with federal anti-discrimination laws cannot justify race-based districting where the challenged district was not reasonably necessary under a constitutional reading and application of those laws." Miller, 515 at 921, 115 S.Ct. at 2490, citing, Shaw I, 509 U.S. at 653-55, 113 S.Ct. at 2830-31. The state must demonstrate that the means chosen to achieve its purpose is "specifically and narrowly framed to accomplish that purpose." Shaw II, 517 U.S. at 915, 116 S.Ct. at 1905, quoting, Wygant v. Jackson Bd. of Educ., 476 U.S. 267, 280, 106 S.Ct. 1842, 1850, 90 L.Ed.2d 260 (1986) (opinion of Powell, J.)

A race-based remedy is limited to neutralizing the effects of a race-based deprivation of federal rights so that affected minority citizens are able to enjoy the same rights as others similarly situated. In Shaw II, the Supreme Court noted that it had "not always provided precise guidance on how closely the means (the racial classification) must serve the end (the justification or compelling interest) [but had] always expected that the legislative purpose would substantially address, if not achieve, the avowed purpose." Shaw II, 517 U.S. at 915, 116 S.Ct. at 1905, citing, Miller, 515 U.S. at 922, 115 S.Ct. at 2491. It noted "race-based state action must be remedial"; "a reapportionment plan would not be narrowly tailored to the goal . . . if the State went beyond what was reasonably necessary to [meet it]"; "the remedy must . . . be related to `the condition alleged to offend the Constitution . . .'" and must be "remedial in nature, that is, it must be designed as nearly as possible `to restore the victims of discriminatory conduct to the position they would have occupied in the absence of such conduct.'" Shaw II, 517 U.S. at 915, 116 S.Ct. at 1905 (internal citations omitted).

To assess "narrow tailoring" requires an examination of the standards for compliance with § 2 of the Voting Rights Act in the context of single member districts, giving due consideration to the fact that the prior districting scheme was itself adopted as a judicial remedy for racial vote dilution. The essence of a § 2 violation is that the jurisdiction's election structure interacts with a past history of discrimination and present racial politics to deprive black voters of the opportunities enjoyed by other voters. The remedy for dilution is to adopt an alternative election structure that restores black voters to the level of political participation they would have enjoyed, absent dilution.

Thus far, the only remedy to receive judicial sanction has been to substitute standard single member districts for the offending election plan, if so doing will permit blacks to control some number of legislative seats. The remedy is thus inherently limited by the concentration of the black population because only geographically concentrated groups can take advantage of standard single member districts. See, Thornburg v. Gingles, 478 U.S. 30, 50-51, 106 S.Ct. 2752, 2766, 92 L.Ed.2d 25 (1986) (limits § 2's relief to those groups that are "sufficiently large and geographically compact to constitute a majority in a single member district."). The remedy is also limited by the number of seats in the legislative body that are susceptible to election from single member districts. See, Hall, 512 U.S. at 875, 114 S.Ct. at 2588 (the number of seats is not subject to challenge under § 2 of the Voting Rights Act).

In contrast to at-large elections, a single member district election system provides all voters with "district" or "area based" representation, thereby permitting voters to have their area-specific interests represented in the governing body. It is precisely the "area-specific" feature of district elections that provides an advantage to black voters who, in part as a result of past discrimination in housing, often live in racially identifiable neighborhoods. The presence of polarized voting, a core element of racial vote dilution, however, means that black voters cannot be assured of the benefits of single member districts unless they are a majority of some number of districts.

Dallas County's 52.8% black voting age majority, according to the 1990 census, clearly is sufficiently large and sufficiently geographically concentrated to constitute a majority of some number of standard single member districts. Thus the real questions here are in how many districts must the group be an effective voting majority to be assured of equal political participation and how much of a majority must they be to take advantage of a district? The answers are important because they provide the limits which the County must not exceed in its construction of race-based districts in order to satisfy the "narrow tailoring" requirement of the equal protection clause.

The Court has not found directly on point cases to resolve these mixed questions of law and fact. The closest case is Johnson v. De Grandy, 512 U.S. 997, 114 S.Ct. 2647, 129 L.Ed.2d 775 (1994). In De Grandy, a § 2 challenge to a portion of Florida's legislative apportionment plan, the Court concluded that Hispanic citizens who constituted an effective majority in a proportional number of state legislative districts did not suffer vote dilution. The Court reached this conclusion even though polarized voting and past discrimination were present; even though additional Hispanic districts could have been created; and even though on a state-wide basis Hispanics might not control a proportional number of districts. Id., 512 U.S. at 1014-18, 114 S.Ct. at 2660. While the Court refused to accept the State's position that minority control of a proportional share of the relevant districts insulated a districting plan from a Section 2 challenge, it offered no examples relevant to this case of when proportionality might not be a sufficient remedy for dilution. Id., 512 U.S. at 1018-19, 114 S.Ct. at 2660-61.

This Court therefore concludes, based on DeGrandy, that a prima facie remedy for dilution is provided when a districting plan includes minority controllable districts in numbers that roughly match the group's percentage of the electorate. Consequently, in order to narrowly tailor its use of race, the County Commission must not manipulate the racial makeup of more than that number of districts, unless it specifically undertakes to demonstrate that the prima facie remedy is in fact inadequate.

The Commission and Department of Justice would perhaps argue that the inclusion of the Probate Judge as a sometimes voting member of the Commission distinguishes this case from DeGrandy. The Court disagrees. The Probate Judge is a separate and distinct office from a County Commissioner. His duties are different. His term of office is different. He obviously must be elected at large. The mere fact that he is an ex-officio member of the Commission, and can vote in case of a tie, is not a reason to throw his seat into the mix when deciding whether a prima facie remedy has been provided by a proportional number of minority districts. Should minorities be able to claim they have not been assured proportional representation in a state legislative body because the lieutenant governor, the body's presiding officer, can vote to break a tie? The Court thinks not.

Section 2 specifically disavows any group's right to proportional representation. This caveat, in conjunction with the actual right protected by § 2 — the right to an opportunity to political participation equal to that of others in the electorate — is an additional reason not to include the Probate Judge's ex-officio membership on the County Commission when deciding how many minority set-aside districts equal a prima facie remedy for dilution.

If the remedy is merely to put minority voters on "equal footing with others in the electorate," there comes a point when manipulation of electoral outcomes via manipulation of the racial make-up of the districts goes beyond "equal treatment" and infringes upon the rights of others. Minority voters for whom more than a proportional share of district seats are set aside are no longer treated the same "as others in the electorate." They receive an excess share of the area-based representation, plus they retain their influence over the at-large probate judge.

Non-minority voters are unfairly deprived of the benefits of single member districts when fewer of the district seats are actually designed for area specific representation. These voters are not "treated equally" by virtue of their ability to help elect the Probate Judge, an official they must share with the entire county, who has limited voting power on the Commission, and who has many duties not directly related to the functions of the County Commission. The loss of a district-specific commissioner to race-based districting is not compensated for by a fraction of an at-large Probate Judge!

Thus the Court concludes that the fungible commissioner seats, those subject to election from single member districts, are the ones to be counted when deciding the prima facie adequacy of a remedial plan. Proportionality of fungible seats may not always be an adequate remedy, but it is up to the party seeking to make further use of race to establish that it is not. This is particularly true where the finding of dilution for which the race based remedy is now offered was based on circumstances over a decade old.

The Court wants to make clear that it is not concluding that the remaining single member districts must be majority white. Rather it is the Court's conclusion that the racial make-up of the remaining districts should not be manipulated. If more than a proportional number of districts quite naturally turn out to be majority black, there is no equal protection problem. The constitution is not offended if the racial make up of the districts by happenstance favors one group in the electorate over another. It is offended if the districts are deliberately constructed to achieve that outcome. To paraphrase Justice Holmes, even a dog knows the difference between being stumbled over and being kicked. O.W. Holmes, Jr., The Common Law 3 (1881).

As a final note, there is also no merit to the Department of Justice's contention that any redistricting plan for Dallas County that fails to provide three majority black districts violates § 5 of the Voting Rights Act. A redistricting plan violates the substantive preclearance standards for § 5 only if it is adopted for a discriminatory purpose, or is "retrogressive" as to the ability of black voters to participate in the political process when compared to the last legally enforceable plan implemented for the jurisdiction. A four single district plan containing two majority black districts is not retrogressive when compared to the last plan legally implemented, the state prescribed four commissioner at-large plan. The only possible benchmark plan is this at-large system, which was required by state law and utilized in the county until this Court held that it violated Section 2 of the Voting Rights Act. The four single member district plan, adopted by this court, was never implemented. The five member plan adopted by the appellate court was not a legal remedy. The appellate court's very rationale for replacing the probate judge with a fifth seat was to provide a single member "swing district" which in fact was a third majority black district of 61.3% black voting age population. United States v. Dallas County Commission, 850 F.2d 1433, 1440-41 (11th Cir. 1988) ("Dr. Lichtman's plan which creates "two solid white, two solid black and one swing district," consists of five single-member districts, two of which contain black voter majorities of 72.4-percent and 70-percent; two white voter majority districts of 65-percent and 64-percent; and finally a fifth swing district containing a black voter majority of 61.3-percent."). There would have been no fifth district to be a "third majority district," but for the appellate court's error as to the extent of its remedial power. Thus the third majority black district having come into existence via a violation of federal law, the plan containing it, the very plan whose invalidity requires a remedy, cannot logically serve as its replacement's benchmark. See, Abrams v. Johnson, 521 U.S. 74, 97, 117 S.Ct. 1925, 1938, 138 L.Ed.2d 285 (1997) ("Section 5 cannot be used to freeze in place the very aspects of a plan found unconstitutional.").

III. ULTIMATE CONCLUSIONS CONCERNING THE INVALIDITY OF THE ARRINGTON 2 PLAN

As specifically applied to the Arrington plan, Dr. Arrington readily conceded that the primary factor driving the creation of his remedial election plan was to control the black percentage of the voting age population in three districts. His justification for manipulating the third district was to make it a swing district, which he saw as "fair" to both races, and required by § 2 of the Voting Rights Act. While the matter is not free from doubt, the Court assumes that the deliberate creation of two majority black districts, with sufficient black voting age population to permit the group to control electoral outcomes, was a permissible response to the Commission's compelling interest in continuing a remedy for vote dilution. Although no objective evidence was offered to support Dr. Arrington's position that at least a 62% black voting age population was needed to assure blacks' an opportunity to elect their choices in a district, the Court finds that figure to be a reasonable one for the County Commission to use. If it is permissible for the County Commission to use race to construct some number of districts, it is also permissible for the County Commission to determine the degree such use is necessary for those districts to provide a clear opportunity for black electoral success.

Deference to the County Commission's position ends when it uses race beyond the point that appears to be reasonably necessary to effect a § 2 remedy. When the County Commission provided its black citizens, who are slightly more than 50% of the potential electorate, with fifty percent of the district-based commissioners seats, the § 2 remedy was presumptively complete, and thus the further use of race presumptively prohibited. This is not to say that the remedy might not be inadequate in some way but merely that the party urging the further use of race must prove it.

Dr. Arrington's manipulation of the racial makeup of the districts beyond creating two safe black districts, puts the burden on the County Commission and the Department of Justice to demonstrate that this additional race-based action was necessary to continue the remedy for dilution. The Court concludes that they did not satisfy that burden.

Dr. Arrington deliberately created the third majority black district based on his belief that an equal opportunity to participate for a group that is roughly 50% of the electorate must include a chance to elect a controlling number of the Commission's voting members. He reasoned that the Probate Judge's ability to vote in case of a tie made him the fifth vote. The Probate Judge was likely, he reasoned, to be elected by white voters. Consequently, to give officials who were the choice of black voters a chance to control the Commission, it was necessary to manipulate the black percentage of a third district. The Court finds a number of flaws with Dr. Arrington's conclusions.

Manipulating district lines to compensate for an ex-officio, non-fungible, member of the Commission is not within the remedy available under § 2. First, as noted, the Court doubts that "manipulating" such an opportunity is a part of the remedy available under § 2. Once control of a proportional share of the seats subject to district elections has been provided to the group, the remedy has gone as far a feasible without unfairly infringing upon the opportunity of others to enjoy the benefits of single member districts. Equal protection means that if the benefits of district representation are provided to some they should be provided to all. Non-minority citizens do not receive these benefits if "their" district is designed, not for representation of their area of the county, but rather to represent the interests of a group of which they are not a part.

There is no objective evidence that the Probate Judge's position is beyond the potential of the County's black majority. Second, even if the remedy is not limited to the single member districts, the Court finds that the County Commission and Department of Justice have not established that electing the Probate Judge is beyond the potential of the black electorate.

Dr. Arrington's opinion was that blacks, despite being a majority of the potential electorate, lack the opportunity to elect the Probate Judge because black political participation lags behind that of whites. Just to bring blacks up to "even" would, in his opinion, require a district in which blacks were 57% (per his inclusion of a district with this percentage in Arrington 2) or 61% (per his testimony) of the voting age population. Yet, he offered no objective support for the existence of a disparity between black and white political participation rates in Dallas County, or for the proposition that, if a disparity exists, it is beyond the power of blacks themselves to correct. Instead he merely repeated the mantra that past discrimination has depressed blacks' socio-economic status, which in turn depresses their political participation.

That blacks in Dallas County were once the victims of discrimination aimed at denying them access to the ballot is undeniable. That they suffer the effects of that discrimination today to the point that they are unable to take advantage of their majority status in the county is at least questionable. The Voting Rights Act, which eliminated discriminatory registration practices, has been in place for thirty-five years. All black citizens under the age of fifty-six came of voting age after its protections were in place. The so-called federal "Motor Voter" act has been in place for five years, thus providing an opportunity for previously unregistered citizens to register to vote in places where they routinely conduct other business, such as the driver's license bureau. Moreover, blacks have controlled the County Commission since 1989.

While none of these facts demonstrate that past discrimination no longer depresses black political participation, in the aggregate they do suggest that the party asserting the need to "compensate" for past discrimination should provide actual evidence that special electoral aid in the amount claimed is in fact necessary. The evidence cited by Dr. Arrington does not support under mobilization of the black electorate.

First, Dr. Arrington noted that local officials believed that the black percentage of the county's population has actually increased since the 1990 census. (Sup.Af.Jul 28, 1999). If true, this should increase the group's chances to determine the outcome of county-wide elections, even if there is some lag in blacks' participation. Second, the scant election information he reported does not support under mobilization caused by past discrimination. Dr. Arrington relied on the outcome in a 1996 election contest in a 61% commissioner district to support the need for a super-majority black district. Curtis Williams, a black candidate running as an independent, prevailed in the district's general election against the white nominee of the Democratic Party. Williams was elected to the Commission with 58% of the vote. No evidence was offered of where his votes came from but, regardless of the race of his supporters, Williams' 58% of the vote was close to the 61% black voting age percentage of the district. The share of the vote Williams received hardly justifies the fourteen point advantage in voting age population (57% black to 43% white) Dr. Arrington argued was necessary just to give blacks an even chance to elect their choices.

Nor does the election contest support attributing depressed participation, if depressed it is, to the continuing effects of past discrimination. Dr. Arrington noted that elections in this district were much closer in "low turnout situations such as party primaries" than in the general election won by Commissioner Williams. But if blacks turnout in greater numbers in the general election, as Dr. Arrington's figures suggest, surely this demonstrates that variations in turnout are a function of something other than past discrimination. It is hard to conceive of a theory for why past discrimination depresses black turnout in the primary but not in the general election held some months later.

Dr. Arrington discounted Williams seemingly comfortable margin of victory in the general election by observing that most elections were still won in the Democratic Primary. The Court finds Dr. Arrington's position counter-intuitive. It is common knowledge that growth of the Republican Party in the South has taken voters, the overwhelming majority of whom have been white, from the Democratic Party. Logically, the Democratic Primary electorate should be substantially more black than the electorate as a whole, thus giving blacks super-majority status in the primary county-wide.

Third, readily available evidence which could have supported Dr. Arrington's "depressed black voter mobilization" thesis was strikingly absent from the record. The County Commission did not introduce comparative voter registration and turnout statistics, or evidence of blacks' portion of Democratic Party electorate. All of this information, if it is not directly available, can be estimated by the same methods that social scientist use to estimate black and white voter support for candidates. See, B. Grofman, L. Handley R. Niemi, Minority Representation and the Quest for Voting Equality, 84-88 (Cambridge at the University Press, 1992).

Fourth, the fact that the present Probate Judge has not receive majority black support is at best very weak evidence that the black electorate has no opportunity to elect that office. Dr. Arrington surmised that the Probate Judge was likely to be the choice of white, but not black, voters in the future because the present occupant of that office had not receive the support of a majority of black voters. He testified that the present Probate Judge had received more black support than most white candidates but he did not recall that he had received more than a third of the black vote. But if two-thirds of black voters supported a losing candidate for Probate Judge, that candidate's defeat is just as attributable to the one-third of the black electorate who did not support him as it is to the failure of white support. Moreover, if a third of black voters in Dr. Arrington's putative "swing" district "split" from the group, the group's candidate will lose, even if blacks constitute more than 70% of the district's voters.

In the final analysis, this Court is not prepared, based on the scant evidence presented, to find that black voters in the year 2000 are incapable of mobilizing to take advantage of their majority status in the county to elect their choice as Probate Judge. Without more substantial evidence that black political participation is depressed beyond the group's ability to correct on its own, there is no basis to assume that this office is beyond the influence of the black majority if it votes cohesively. An equally plausible explanation for the repeated reelection of the current Probate Judge is that he has retained his position by including black voters in his winning coalition and keeps their support by being sensitive to their concerns. If so, his failure to remain sensitive to the black electorate would likely prompt its mobilization against him. Given that the county's electorate is closely divided between blacks and whites, and that the districts inevitably will reflect that breakdown, the Probate Judge is the one member of the County Commission most motivated to build bridges, instead of fences, between the groups.

Even if black voters cannot control the election of the Probate Judge, there is insufficient evidence to support the proposition that their votes will be diluted unless a third majority black district is deliberately created. The Court is not prepared to conclude that, even if black control of the Probate Judge's election is unlikely, racial manipulation of 75% of the County's single member districts is necessary to provide blacks with fair influence over the election of the Dallas County Commission and over its legislative agenda. As Justice Souter implies in DeGrandy, when minority voters can form coalition with other voters to help elect candidates, this too is a form of political influence. He notes that these "candidates may not represent perfection to every minority voter, but minority voters are not immune from the obligation to pull, haul, and trade to find common political ground, the virtue of which is not to be slighted in applying a statute meant to hasten the waning of racism in American politics." 512 U.S. 997, 1020, 114 S.Ct. 2647.

It was incumbent upon the County Commission and Department of Justice to establish that race-based action beyond providing for proportional control of the district elections was necessary to provide a complete remedy. In short, they offered no evidence that black control of two seats, plus blacks' majority status in the County, with all that means in terms of influence over the Probate Judge, was insufficient to provide blacks with an opportunity equal to that of others to participate in the political process and to elect representatives of their choice. Therefore, the Court resolves the inherent clash in the equal protection clause's mandate of "narrow tailoring of racial remedies" and the Voting Rights Act's mandate of a race-based remedy for dilution, by concluding that the County Commission and the Department of Justice did not establish that the use of race in their proposed districting plan was narrowly tailored.

IV. THE COURT'S PLAN

Although the remedial plan drafted by the Department of Justice and adopted by the County Commission (namely, Arrington 2) must be rejected because it violates the constitution, the Court ordinarily would be compelled to give the County Commission another chance to fix the constitutional infirmity. White v. Weiser, 412 U.S. 783, 796-97, 93 S.Ct. 2348, 2355, 37 L.Ed.2d 335 (1973). Moreover, even if the court must undertake to devise its own plan, it ordinarily should defer to the County Commission's judgment to the extent possible, changing only those portions of the plan that offend federal law. Upham v. Seamon, 456 U.S. 37, 43, 102 S.Ct. 1518, 1522, 71 L.Ed.2d 725 (1982).

In this case, however, the constitutional infirmity with Arrington 2 was that it made a greater use of race than was necessary to provide a remedy for vote dilution. The problem was not that the plan contained three majority black districts, but rather that the district lines had been manipulated to reach that precise result. Given Dr. Arrington's knowledge of the extent and location of the county's black population, and that of the others involved in the plan's production, the Court believed it unlikely that any subsequent plan could be guaranteed not to be based on the same over-emphasis on race. Thus, the Court concluded that a remedy for the over use of race would be more convincingly provided by someone who had no prior knowledge of the location and extent of the County's black population.

The Supreme Court affirmed a similar approach by the Court in Abrams, 521 U.S. at 74, 117 S.Ct. at 1925. There, as here, the plan to be replaced was unconstitutional because it made excessive use of race. The lower court drew its own plan, which the Supreme Court held was justified because to use the invalidated plan as the basis for a remedy "would validate the very maneuvers that were a major cause of the unconstitutional districting." Abrams, 521 U.S. at 86, 117 S.Ct. at 1933.

Furthermore, since the problem with the County Commission/Department of Justice plan was racial manipulation of the districts, not their ultimate racial make-up, the Court could not make the County Commission's plan constitutional simply by lowering the black percentage in the so-called swing district. Any black percentage chosen by the Court would produce just as "racially manipulated" a district as was produced by Dr. Arrington. There were only four districts in the plan. Consequently, all their boundaries were so inter-dependent that it was not feasible to leave two in place, and expect that the remaining two would be free of the effects of the initial racial manipulation, even if they were re-drawn without knowledge of race. See Abrams, 521 U.S. at 86, 117 S.Ct. at 1933 (distinguishing the situation at issue in Upham).

In light of these obstacles, the Court resolved to follow the example of the District Court in Abrams and start over completely. To produce a plan that did not rely excessively on race, the Court employed two experts who had no prior knowledge of the location and relative concentration of the County's population. The Court asked the experts to devise a strategy to produce a single member district plan that would respect the County Commission's expressed non-racial districting goals and take into consideration its interest in continuing a remedy for the earlier § 2 violation, but would make no further use of race. In consultation with the Court, Professor Katharine Butler, developed goals for the remedial plan and a methodology for the other expert, Dr. Harold Stanley, to produce a plan consistent with the goals. The purpose in defining the methodology before confronting the actual demography of the county was to have a process in place to make neutral decisions about issues that might have racial or political implications.

GOALS FOR THE COURT'S PLAN.

The goals decided upon were a combination of the legal requirements and a consideration of the race-neutral districting concerns expressed by the commissioners in various hearing. They are set out below:

1. Comply with one person, one vote.

2. Create sensible districts in light of functions of this body — basically, address commissioners' concerns about each district containing both county and city voters; and about equalizing road mileage. Since the primary function of the Commission is to maintain rural county roads, these considerations increase the commissioner's ability to be aware of and respond to citizen needs within his district.

3. Create districts that follow natural and known boundaries, such as rivers, prominent streams and lakes, highways, and streets.

4. Include two majority black districts in the final plan to continue the remedy for vote dilution found in 1987.

5. To the extent possible, consistent with 1. and 4., create districts so that each incumbent retains some portion of his former district, and so as to avoid to the extent possible, placing two or more commissioners in the same district.

6. To the extent possible, take notice of population growth since the 1990 census and factor that information into the plan so that massive adjustments to the districts lines will not be required following the 2000 census.

7. Make no use of race in the creation of the plan, beyond that necessary to assure a remedy for racial vote dilution.

METHODOLOGY TO ACHIEVE THE COURT'S GOALS

The method to be utilized in an effort to achieve the aforementioned goals was set forth as follows:

1. Do not include race and voting age information in the data base used to create the initial percentage of the individual districts' populations. A plan drawn totally without regard to race thus should provide a measure of the naturally occurring racial consequence of race neutral districting.

2. Create the first tentative district by including all the geographic area between the Alabama River and the eastern border of the county. Add sufficient urban/suburban population to equal a standard district for one person, one vote purposes. Try to include a bridge over the River to connect the urban and rural parts of the district.

3. Using an "eye-ball" approach, divide the remaining area outside Selma and the surrounding suburbs, so as to roughly equalize road mileage.

4. Using recognizable boundaries, divide the population of Selma so as to equalize the population within the districts.

5. Examine available information to determine likely changes in the location and density of the population since 1990. This information should include a comparison of the portion of registrants in each precinct in 1990 and in comparable precincts today. Absent evidence to the contrary, an increase in the portion of all registrants found within a particular precinct should indicate that the underlying population in that precinct has increased relative to the county as a whole. Additional information on population shifts may be available from the sub-county estimates of the population from the Census Bureau.

If this information indicates that the districts as drawn will be significantly malap-portioned when the new census figures come out and it is possible to adjust the 1990 population with the districts to lessen this problem, adjust the district lines accordingly. Make sure that the sum total of the adjustments does not produce a total deviation from population equality in excess of 6% for the plan as a whole. While court ordered plans must comply more strictly with one person, one vote standard, a 6% deviation should be justifiable as a means to cut down on the modifications that will be mandated by the 2000 census.

6. Make a hard copy of this plan, and label it Court-1, Tentative Districting Plan.

7. Now add the race and age information into the data base. Print the population and voting age population by race for each district. Attach this information to Court-1 and preserve both a hard copy and a computer file.

8. If at least two districts are 62% or more black in Voting Age Population (VAP), make no race-based adjustments in the districts lines, regardless of what the racial make up of the districts is otherwise. If there are not two districts at least 62% black in VAP, make adjustments in the following order until that state is achieved.

All movement of population should be consistent with the creation of compact districts, having recognizable boundaries. Whenever possible, population adjustments should be made within the urban/suburban areas of the county. If the two majority black districts objective cannot be accomplished without distorting the district lines using option A, go to option B, and so on.

A. Move black population from any district with more than a 62% black VAP to one with less than a 62% black VAP, if this will create, or will substantially contribute to the creation of, a second 62% black VAP district.

B. Swap adjacent black and adjacent white populations between districts.

C. Make more substantial changes but without distorting district lines.

D. Stop making adjustments, if after steps A-C, two districts have not yet emerged, but there are two districts with black voting age majorities of at least 54% black. Should the process reach this point, we should reevaluate whether two 62% black VAP majority districts can be created without violating standard districting principles.

When two districts have been created, or step C has been completed without producing two districts, print the plan, with its racial breakdown by district. Label it "Court 2 Tentative Districting Plan." Preserve the computer file and a hard copy of this plan.

9. Compare Court-2 with the recently invalidated five single member district plan, and locate the residences of the current commissioners on both Court-2 and the five member plan. Adjust the lines to the extent possible to place one commissioner in each of the districts, matching each with the district that contains the most of his present district. Since one district must contain two commissioners, select the two to be paired on the basis of which old districts most overlap within a single new district. Make any other adjustments necessary to restore population balance, preserve the black majority districts, and further the other goals listed above.

10. Print this plan and label it Court-3 Tentative Districting Plan. Preserve the computer file and a hard copy.

13. Provide all three plans to the parties for comments. Written comments from the parties will be considered, and if data errors, or errors in underlying factual assumptions are brought to light, corrections will be made in the final plan necessary.

ACTUAL IMPLEMENTATION OF THE METHODOLOGY

Dr. Stanley produced a plan, following as closely as possible the above-quoted Methodology. Maps of the final plan, Court # 3, and the two intermediate plans, Court 1 and Court 2, were provided to the parties along with the breakdown of the population by race and the voting age population by race for each district in each plan and the census description of each district.

Dr. Stanley produced Court 1 without knowledge of the racial data, essentially following Steps 1-4 of the above-quoted Methodology. Two majority black districts emerged with no effort to create, or not to create, them. The first district produced in accordance with the prearranged methodology, contained the part of the county east of the Alabama River, augmented by sufficient population west of the river to comply with population equality. This district turned out to be 70.4% black in voting age population. The second majority black district that emerged was District 3, which was slightly under the 62% black VAP which Dr. Arrington had indicated was necessary for a "safe" black district. Court 1 was modified to raise District 3's black voting age majority above 62%. The plan as adjusted was designated Court 2.

Court 2 was then further adjusted to separate the incumbents, and, to the extent feasible place each of them in the district which contained some portion of their constituents from the old five member plan. In Court 1 and Court 2, incumbents Commissioners Roy Moore and Erskine Minor were both in District 1. District 2 had no incumbent. District 1 contained most of Minor's constituents and few of Moore's. Consequently the boundary of District 2 was extended south to pick up Moore, thereby placing him in a district with some of his constituents.

In Court 2, incumbent Commissioners Curtis Williams and Kimbrough Ballard ended up together in District 4. Neither of their residences was close to District 2, the "vacant" district. Adjusting the boundary between Districts 3 and 4 to put Williams into the district with Perry Varner was considered, and ultimately rejected. More voters from Williams' old district were in District 4 than were in District 3. Thus, leaving Williams and Ballard together in the district in which they had ended up in the blindly drawn plan seemed preferably to intentionally moving Williams in with Varner.

As it turned out, population growth, at least as reflected in changes since 1990 in voter registration, did not appear to have been significant. Thus, Goal 6 of the plan was abandoned.

The parties were given an opportunity to examine the three plans and the methodology by which they were produced. At the hearing on the plan, Commissioner Williams testified that he believed that his chances for re-election were better in District 3, a majority black district, than in District 4. Williams requested that the district line be adjusted. Hearing no objections from any other party, the Court instructed Dr. Stanley to make the change.

It was not possible to achieve goal 3, create districts the follow natural and known boundaries, as completely as would have been desired. The amended plan, reflecting the change to include incumbent Commissioner Williams in District 3, was provided to the parties, with the possibility that slight modifications might be needed in case some of the district boundaries prove to be too uncertain to permit the Probate Judge to accurately assign voters to the correct districts. Dr. Stanley worked with the Probate Judge to make these changes, all of which were minor in their impact on the plan.

Shortly after the amended plan was provided to the parties, a request was made for the Court to swap some area between District 1 and 2. It was pointed out that the only portion of District 1 that is north of the Alabama River contains segments of two county roads that are otherwise in District 2. The remaining county roads in District 1 are between the River and the County's eastern boundary. Transferring the area containing these roads from District 1 to District 2, in exchange for an area of equal population, but containing no roads, would allow for more efficient road maintenance. Each commissioner is responsible for maintaining the roads within his district, which would mean that the District 1 Commissioner periodically would have to move his road maintenance machinery across the river to take care of these two road segments.

Dr. Stanley indicated that the requested change would involve moving a total of 924 people but would have virtually no impact on the racial make up of the two districts. A motion (doc. 214) filed by the Dallas County Commission defendants to amend the Court's Plan # 3 to conform to these changes was granted on April 19, 2000. The final statistics for the plan are as follows: Total Population Voting-Age Population Deviation from ideal Percent from ideal

V. CONCLUSION AND FINAL ORDER REGARDING THE IMPOSITION OF A REMEDIAL PLAN

For the reasons stated above, the Court concludes that each of the plans submitted by the parties made impermissible use of race and thus violates § 2 of the Voting Rights Act. In contrast, the Court's Plan # 3 complies with all constitutional and statutory mandates, including the principles emphasized in Abrams, 521 U.S. at 85-90, 117 S.Ct. at 1933-35. It is therefore ORDERED, as previously directed in the Court's Order of April 4, 2000, (Doc. 212), that the election for the Dallas County Commission shall be CONDUCTED pursuant to the Court's Plan # 3 already provided to the parties commencing with the Primary Election on June 6, 2000.