Seamon v. Upham, 536 F. Supp. 931 (E.D. Tex. 1982)

U.S. District Court for the Eastern District of Texas - 536 F. Supp. 931 (E.D. Tex. 1982)
February 27, 1982

536 F. Supp. 931 (1982)

A. M. SEAMON
v.
Chet UPHAM, et al.

Civ. A. No. P-81-49-CA.

United States District Court, E. D. Texas, Paris Division.

February 27, 1982.

*932 *933 *934 *935 Leighton Cornett, Paris, Tex., for plaintiff.

David R. Richards, Austin, Tex., for Juanita Craft, et al.

Luis M. Segura, Jesse Roy Botello, San Antonio, Tex., for M. Garcia, A. Luna, A. Gonzales, B. Eureste, R. Padilla, R. Palomo, M. Tamez, R. Cristan and H. DeHoyos.

George J. Korbel, Jose Camacho, Austin, Tex., Richard Gonzales, Alpha Hernandez, Rolando Menchaca, Texas Rural Legal Aid, Inc., Del Rio, Tex., Raul Noriega, Texas Rural Legal Aid, Inc., San Antonio, Tex., for B. Gonzales, A. Dolorosa and J. Adame.

Joaquin G. Avila, Mexican American Legal Defense and Educational Fund, San Antonio, Tex., for A. Garcia and J. Rodriguez.

Rafael Quintanilla, Jr., Austin, Tex., for Democratic Party of Texas.

Mark White, Atty. Gen., Austin, Tex., for plaintiff.

John M. Harmon & R. James George, Jr., Graves, Dougherty, Hearon & Moody, Austin, Tex., for Eric Clifford and Chester Upham.

Robert C. Slagle, Jr., Sherman, Tex., pro se.

R. C. Slagle, III, Sherman, Tex., for Brady Fisher.

Steve Bickerstaff, Richard E. Gray, III, Asst. Atty. Gen., Austin, Tex., for William Clement, Geo. Strake, State of Texas.

*936 James Allison, Asst. Atty. Gen., Austin, Tex., for Roger Peterson and Margaret Coplin.

Before SAM D. JOHNSON, Circuit Judge, JUSTICE, Chief District Judge, and PARKER, District Judge.

 
OPINION

SAM D. JOHNSON, Circuit Judge:

Since the Supreme Court's 1962 decision in Baker v. Carr, 369 U.S. 186, 82 S. Ct. 691, 7 L. Ed. 2d 663 (1962), federal courts have found themselves undesirably in the center of the "political thicket" of congressional and legislative apportionment. This Court finds itself in just such a position.

The first called session of the 67th Legislature of Texas enacted Senate Bill No. 1 (S.B. 1), which apportioned Texas into twenty-seven single-member congressional districts, on August 10, 1981. The enactment was signed by the Governor of Texas on August 14, 1981. Tex.Rev.Civ.Stat.Ann. art. 197f (Vernon Supp. 1981). Since Texas is a jurisdiction covered by the submission and preclearance provisions of section 5 of the Voting Rights Act, 42 U.S.C. § 1973c (1976); Briscoe v. Bell, 432 U.S. 404, 97 S. Ct. 2428, 53 L. Ed. 2d 439, S.B. 1 was submitted to the Attorney General of the United States for preclearance on or about September 11, 1981.[1]

Prior to any objection or affirmative indication that an objection would not be interposed to S.B. 1 by the Attorney General of the United States, suit was filed in the federal district court for the Eastern District of Texas against the State of Texas and several state officials. Numerous plaintiffs were allowed to intervene in the suit, which presented the question of whether S.B. 1 constituted an impermissible gerrymander that resulted in the dilution of minority voting strength. One group of minorities argued basically it was "packed" into a single congressional district in order to minimize the political influence it had enjoyed previously. A group of minorities from another area of the State argued S.B. 1 "fragmented" the minority population in order to decrease its voting strength. Plaintiffs challenged the constitutionality of S.B. 1 under the fourteenth and fifteenth amendments of the United States Constitution. In addition, plaintiffs claimed the legislation violated section 2 of the Voting Rights Act of 1965. 42 U.S.C. § 1973 (1976).

This three-judge court was empaneled pursuant to 28 U.S.C. § 2284 (1976). On November 30, 1981, this Court held an evidentiary hearing. At the three-day hearing, the parties presented evidence relevant to the issues involved in the constitutional and statutory challenge to S.B. 1. At the conclusion of the hearing, the Court took the case under advisement, recognizing that congressional reapportionment plans enacted by the State of Texas are not effective as law unless and until they receive clearance pursuant to section 5 of the Voting Rights Act. Accordingly, this Court withheld deciding and addressing the constitutionality of S.B. 1 until appropriate action was taken by the Attorney General of the United States. See Wise v. Lipscomb, 437 U.S. 535, 98 S. Ct. 2493, 2497-98, 57 L. Ed. 2d 411 (1978) citing Connor v. Finch, 431 U.S. 407, 97 S. Ct. 1828, 1832, 52 L. Ed. 2d 465 (1977) and Connor v. Waller, 421 U.S. 656, 95 S. Ct. 2003, 44 L. Ed. 2d 486 (1975).

On January 25, 1982, this Court entered an order postponing the filing deadline for prospective candidates for the office of representative to United States Congress for sixteen of twenty-seven congressional districts, as those districts were defined by S.B. 1, until February 22, 1982.[2] The Texas *937 Election Code, Tex.Elec.Code Ann. art. 13.12(c) (Vernon Supp.1981), provides that an application to have one's name placed on the ballot as a candidate for nomination by a political party shall be filed no later than 6:00 p. m. on the first Monday in February preceding the date of the primary election. See also Texas Educ.Code Ann. art. 11.22(e) (Vernon Supp.1981). Consequently, the filing deadline in the year 1982 was Monday, February 1. This Court's action was taken as a result of its opinion that such action was necessary because of the delay by the Attorney General of the United States in meeting his obligations under section 5 of the Voting Rights Act. This delay placed the people of the State of Texas in the awkward position of facing filing deadlines for United States congressional offices that were to be filled by elections that, because the United States Attorney General had failed to preclear the relevant legislative enactment, would be unenforceable. 42 U.S.C. § 1973c.

The Attorney General of the United States did interpose an objection to S.B. 1 as provided for in 42 U.S.C. § 1973c by letter dated January 29, 1982 to the Secretary of State of the State of Texas. January 29, 1982, a Friday, was only three days prior to the February 1, 1982, filing deadline. It was, however, some 140 days after the initial information was submitted to the office of the United States Attorney General by the State of Texas.[3]

The effect of the objection to S.B. 1 by the Attorney General of the United States was to render implementation of S.B. 1's provisions legally unenforceable. 42 U.S.C. § 1973c. Accordingly, this Court, on February 2, 1982, ordered that a hearing be set to determine the extent of this Court's jurisdiction in the case sub judice. In addition, assuming the Court's jurisdiction and the need for implementing a court-ordered congressional apportionment plan, the parties were ordered to provide written submissions along with maps, plats, or other relevant data.

Such a hearing was held on February 9, 1982. At that time, the Court granted plaintiffs' motions to amend their complaints to include challenges to the congressional apportionment plan that S.B. 1 was meant to supplant. This plan, based upon 1970 census figures, was instituted after substantial litigation. See Graves v. Barnes, 343 F. Supp. 704 (W.D.Tex.1972); Graves v. Barnes, 378 F. Supp. 640 (W.D. Tex.1973); Graves v. Barnes, 408 F. Supp. 1050 (W.D.Tex.1976); Graves v. Barnes, 446 F. Supp. 560 (W.D.Tex.1977); White v. Weiser, 412 U.S. 783, 93 S. Ct. 2348, 37 L. Ed. 2d 335 (1973). This existing congressional apportionment plan provides for only twenty-four districts. In addition, utilizing 1980 census data, its districts have an average deviation of ± 17 percent from the ideal population for a congressional district and a deviation of ± 84.3 percent at its extremes. Because of the obvious constitutional inadequacies of this plan, this Court was asked by the parties to this litigation, including the State of Texas, to exercise equitable jurisdiction and devise and implement a congressional apportionment plan that would include twenty-seven congressional districts and satisfy constitutional requirements.[4]

*938 Subsequent to the February 9, 1982 hearing, this Court entered a second order postponing the filing deadline for prospective candidates for representative to the United States Congress. The deadlines were postponed until 6:00 p. m. on March 19, 1982. This order, dated February 15, 1982, affected only the sixteen congressional districts, as they were defined by either S.B. 1 or otherwise, that were affected by the January 25, 1982 order.[5] It also enjoined the State of Texas from holding any election for the purpose of choosing any congressional representative from the State of Texas. The injunction was to be in force and effect until this Court implemented a court-ordered congressional apportionment plan or took other action it deemed appropriate.

This Court recognizes the clear directives of the Supreme Court that the task of reapportioning congressional districts belongs to the state legislature in the first instance and the federal courts should make every effort not to preempt the state legislature's primary jurisdiction and responsibility. See Wise v. Lipscomb, 98 S. Ct. at 2497; Connor v. Williams, 404 U.S. 549, 92 S. Ct. 656, 658 n.4, 30 L. Ed. 2d 704 (1972); White v. Weiser, 93 S. Ct. at 2354; Ely v. Klahr, 403 U.S. 108, 91 S. Ct. 1803, 1807, 29 L. Ed. 2d 352 (1971); Reynolds v. Sims, 377 U.S. 533, 84 S. Ct. 1362, 12 L. Ed. 2d 506 (1964). The Texas State Legislature, however, is the only state body empowered to enact a new congressional apportionment plan.[6] There has been no action taken by the Governor of Texas to call a special session of the Legislature, which meets biennially and is not scheduled to convene again until 1983. To the contrary, the Governor, through his attorneys, has requested this Court to implement a congressional apportionment plan of its own.

In the absence of a congressional apportionment plan that has been precleared pursuant to section 5, the Voting Rights Act "essentially freezes the election laws of the covered state." Georgia v. United States, 411 U.S. 526, 93 S. Ct. 1702, 1709, 36 L. Ed. 2d 472 (1973). As a result, the State of Texas faces the possibility that the date for party primary elections which, under Texas law, is May 1, 1982, Tex.Elec.Code Ann. art. 13.03 may pass with no election. There are critical immediate concerns, in addition. In order to conduct an election on May 1, 1982, there is substantial mechanical preparation necessary. This preparation includes conducting proper voter registration, printing ballots, mailing absentee ballots, advising voters regarding the districts of their residence, and, of course, providing as nearly as possible for the time frame established through the legislative process of this state *939 for candidates to appraise, evaluate, organize, and take other action necessary to submit themselves for election.

A present Texas legislative effort to apportion Texas' twenty-seven congressional districts appears to be precluded. Accordingly, if Texas' elective process is not to be completely frustrated and reduced to utter chaos and confusion, this Court is obligated, indeed it is required, to initiate a remedial decree that will allow the State of Texas to conduct congressional elections under a constitutional apportionment plan.[7]Wise v. Lipscomb, 98 S.Ct. at 2497-98; Connor v. Finch, 97 S. Ct. at 1834; Mahan v. Howell, 410 U.S. 315, 93 S. Ct. 979, 989, 35 L. Ed. 2d 320 (1973).

 
I. Appropriate Standards

In meeting the "unwelcome obligation" of devising and implementing a congressional apportionment plan, this Court turns to the guidance provided by the experiences of other federal courts placed in the political thicket of apportionment. This assistance is not generally in the form of affirmative prescriptions and directions regarding proper methodology and procedure to be utilized in fashioning apportionment plans. Instead, it is in the form of warnings and admonitions against taking certain actions or setting certain goals.

These admonitions, however, can be cast into a basic conceptual framework that consists of two coterminous constitutional duties or requirements that must be fulfilled in order for a court to properly implement a congressional apportionment plan. One function or goal of the Court must be to provide voter equality. Providing voter equality is a majoritarian principle. It implicates the requirement many times characterized as the necessity of satisfying the constitutional command of one person, one vote. McDaniel v. Sanchez, 452 U.S. 130, 137, 101 S. Ct. 2224, 2230, 68 L. Ed. 2d 124 (1981); Connor v. Finch, 97 S. Ct. at 1833; id. at 1840 (Blackmun, J., concurring); Wyche v. Madison Parish Police Jury, 635 F.2d 1151, 1159 (5th Cir. 1981); Marshall v. Edwards, 582 F.2d 927, 938 (5th Cir. 1978), cert. denied, 442 U.S. 909, 99 S. Ct. 2820, 61 L. Ed. 2d 274 (1979).

A companion constitutional requirement is that the court-ordered apportionment plan be racially fair.[8] This requirement involves the rights of people who are members of minority groups. A racially fair plan avoids racial discrimination by not diluting the value of the votes to be cast by members of minority groups. McDaniel v. Sanchez, 101 S. Ct. at 2230; Connor v. Finch, 97 S. Ct. at 1828; id. at 1840 (Blackmun, J., concurring); Wyche v. Madison Parish Police Jury, 635 F.2d at 1159; Marshall v. Edwards, 582 F.2d at 938.

At the outset, this Court acknowledges that its attempt to accomplish its constitutional objectives is generally evaluated under stricter standards than a legislature's *940 attempt at properly apportioning a state.[9]Connor v. Finch, 97 S. Ct. at 1833. A federal court simply does not operate with the same latitude as a state legislature. This is a function of the Supreme Court's perception that "[t]he least representative branch of the government must take care when it reforms the most representative branch." Marshall v. Edwards, 582 F.2d at 934. See also Wyche v. Madison Parish Police Jury, 635 F.2d at 1159.[10]

Consequently, many choices for apportionment that might pass muster if enacted by the state's representative body will fail if ordered by a federal court. This is because "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." Connor v. Finch, 97 S. Ct. at 1833-34 quoting Roman v. Sincock, 377 U.S. 695, 84 S. Ct. 1449, 1458, 12 L. Ed. 2d 620 (1964). As a result, this Court may not consider the purely political considerations that might be appropriate for legislative bodies. Wyche, 635 F.2d at 1160.

Since reapportionment "inevitably has sharp political impact," a district court, in fashioning a reapportionment plan, "should not pre-empt the legislative task nor `intrude upon state policy anymore than necessary.'" White v. Weiser, 93 S. Ct. at 2355 quoting Whitcomb v. Chavis, 403 U.S. 124, 91 S. Ct. 1858 at 1878, 29 L. Ed. 2d 363. See also Connor v. Finch, 97 S. Ct. at 1833; Reynolds v. Sims, 377 U.S. 533, 584, 84 S. Ct. 1362, 1393, 12 L. Ed. 2d 506 (1964); Graves v. Barnes, 446 F. Supp. 560, 563-64. In other words, the federal courts should avoid encroaching upon the legislature's lawmaking prerogative by following "the policies and preference of the state, as expressed in statutory and constitutional provisions or in the reapportionment plans proposed by the state legislature." White v. Weiser, 93 S. Ct. at 2355.

As long as the state's proposals do not run afoul of the strict requirements encountered by the district court, they may be utilized. The Court is allowed to pay deference to certain policy choices, such as providing compact and contiguous districts, utilizing natural and historical boundaries, and preserving the integrity of county lines. See Connor v. Finch, 97 S. Ct. at 1833; Kilgarlin v. Hill, 386 U.S. 120, 87 S. Ct. 820, 822, 17 L. Ed. 2d 771 (1967); Mahan v. Howell, 93 S. Ct. 979; Wyche v. Madison Parish Police Jury, 635 F.2d 1151; Taylor v. McKeithen, 5th Cir., 499 F.2d 893. Indeed, if the Court follows the state's proposals of this type, it may reduce the incidence of its being found acting in a partisan manner. Connor v. Finch, 97 S. Ct. at 1828-29 (Blackmun, J., concurring).

This deference to the state legislature's lawmaking prerogative is not without limitations, however. If adherence to the state's proposals would result in the district court's being unable to satisfy either of its *941 coterminous constitutional goals providing voter equality or racial fairness the court should not defer. This limitation necessarily may require some variance from the state's proposed plans in order for the Court to satisfy the stricter standards it faces. Connor v. Finch, 97 S. Ct. at 1834 (1977); White v. Weiser, 93 S. Ct. at 2355. See also Graves v. Barnes, 446 F. Supp. 560, 564 (1977). It should never be forgotten that the plan ultimately implemented is to be evaluated as a court-imposed plan.

If the state's proposals are not immediately capable of incorporation into the Court's plan because they would not help serve the goals required of federal courts, the district court is not allowed to go behind the proposal and inquire into the other, perhaps numerous, policy choices made by the state in arriving at its proposal. The overriding objective of the district court's plan must be to provide substantial equality of population, while not diluting the political strength of members of minority groups. Wyche v. Madison Parish Police Jury, 635 F.2d 1151. The Court is bound to apply equitable standards of voting equality and racial fairness. Mahan v. Howell, 93 S. Ct. at 989; Connor v. Finch, 97 S. Ct. at 1833; Wyche v. Madison Parish Police Jury, 635 F.2d at 1160-61. See also Beer v. United States, 425 U.S. 130, 96 S. Ct. 1357 at 1363 n.11, 47 L. Ed. 2d 629 (1976) quoting City of Richmond v. United States, 422 U.S. 358, 95 S. Ct. 2296, 2301, 45 L. Ed. 2d 245 (1975). The watchword for a court-ordered apportionment plan is "fairness."

The federal court's task of providing and implementing a congressional apportionment plan 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 v. Finch, 97 S. Ct. at 1834 quoting from Roman v. Sincock, 84 S. Ct. at 1458. Accordingly, if a state proposal submerges the voting potential of racial minorities, the district court is not required to defer to the state's choice of apportionment. Indeed, absent "any insurmountable barrier to devising [an alternative] plan," the district court has a duty to refrain from the state's proposal. Connor v. Johnson, 402 U.S. 690, 692, 91 S. Ct. 1760, 1762, 29 L. Ed. 2d 268 (1971).

 
A. Providing Voter Equality

The population of the State of Texas expanded substantially during the decade of the Seventies, growing at a rate of 27.1 percent. As a result of this growth, Texas was apportioned three additional representatives to the United States Congress. Instead of the twenty-four representatives it was apportioned previously, Texas is now entitled to twenty-seven Congressmen.

Accordingly, the State of Texas must be divided into twenty-seven parts. These twenty-seven parts or districts must contain an approximately equal number of voters. This requirement is a function of "the command of article 1 § 2 [of the United States Constitution] that Representatives be chosen `by the People of the Several states,' [which] means that as nearly as is practicable one man's vote in a congressional election is to be worth as much as another's." Mahan v. Howell, 93 S. Ct. at 983 (1973) citing Kirkpatrick v. Preisler, 394 U.S. 526, 89 S. Ct. 1225, 22 L. Ed. 2d 519 (1973), Wells v. Rockefeller, 394 U.S. 542, 89 S. Ct. 1234, 22 L. Ed. 2d 535 (1969), and Wesberry v. Sanders, 376 U.S. 1, 84 S. Ct. 526, 527, 11 L. Ed. 2d 481 (1964). See also White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973); Gaffney v. Cummings, 412 U.S. 735, 93 S. Ct. 2321, 37 L. Ed. 2d 298 (1973); Reynolds v. Sims, 377 U.S. 533, 84 S. Ct. 1362, 12 L. Ed. 2d 506; Wyche v. Madison Parish Police Jury, 635 F.2d 1151.[11]

*942 The analysis serving as the foundation for the one-person, one-vote or population equality principle is easily understandable. It suggests that a person living in a congressional district with a larger population than another district is "underrepresented." People living in a congressional district with a relatively smaller population are "overrepresented."[12]

In order for this Court to achieve its constitutional goal of population equality in the apportionment of the State of Texas, it must devise and implement a plan "with little more than de minimis variation" of population[13] among the twenty-seven congressional districts. McDaniel v. Sanchez, 101 S. Ct. at 2230; Chapman v. Meier, 420 U.S. 1, 95 S. Ct. 751, 766 and n.19, 42 L. Ed. 2d 766 (1975); Connor v. Finch, 97 S. Ct. at 1833; Connor v. Coleman, 440 U.S. 612, 99 S. Ct. 1523, 1527 n.4, 59 L. Ed. 2d 619 (1979) (Marshall, J., dissenting) (court-ordered plan given less deference than legislative apportionments as to variances from population equality). Based upon the 1980 census data, the ideal population for one of Texas' twenty-seven congressional districts is 526,977. In other words, this Court achieves its goal of population equality if it implements a plan that apportions the State of Texas into twenty-seven congressional districts, each of which has a population of 526,977 or a de minimis variation from this number.

While the principle of "no more than de minimis variation" is easy to state, application of the rule is more difficult both theoretically and practically. It is theoretically problematic because there is no set maxim of what variances from the ideal population are de minimis and what are constitutionally impermissible. Indeed, the Supreme Court has rejected arguments by states that "there is a fixed numerical or percentage population variance small enough to be considered de minimis and to satisfy without question" the requirement that "as nearly as practicable one man's vote in a congressional election is to be worth as much as another's." Kirkpatrick v. Preisler, 394 U.S. 526, 89 S. Ct. 1225, 22 L. Ed. 2d 519 quoting Wesberry v. Sanders, 84 S. Ct. at 530 (1964).

Beyond the theoretical problems in achieving de minimis population variance, there are practical difficulties that have been recognized by the Supreme Court as justifications for variations. Initially, the Supreme Court has recognized that the basic statistical materials utilized by courts census data taken at ten-year intervals are only "as accurate as such immense undertakings can be, but they are inherently less than absolutely accurate." Gaffney v. Cummings, 93 S. Ct. at 2327. "Some deviation is permitted for purposes of administrative convenience, adherence to historical or geographical boundaries, and recognition of separate political units." Wyche v. Madison Parish Police Jury, 635 F.2d at 1158, citing Mahan v. Howell, 410 U.S. 315, 93 S. Ct. 979, 35 L. Ed. 2d 320 (16.4% deviation approved) and Abate v. Mundt, 403 U.S. 182, 91 S. Ct. 1904, 29 L. Ed. 2d 399 (1971) *943 (11.9% deviation upheld). The geographic location of people also may make mathematical exactness impossible if there is an attempt to provide compact and contiguous districts.

In any event, the Supreme Court has recognized that, as a practical matter, mathematical exactness or precision is hardly a workable constitutional requirement. Gaffney, 93 S. Ct. at 2326 quoting Reynolds v. Sims, 84 S. Ct. at 1390. Such an acknowledgment explains the Court's refusal to establish a theoretical norm for what a constitutionally de minimis variation would be. This Court, however, in its effort to achieve the constitutional objective of voter equality, has attempted to satisfy the most severe requirements possible for mathematical precision.[14]

 
B. A Racially Fair Plan

In order to achieve this Court's second constitutional objective implement a plan that provides racial equality this Court is guided by the Supreme Court's recent opinion in the case of McDaniel v. Sanchez. In that opinion, the Supreme Court referred to the legislative history of section 5 of the Voting Rights Act, which stated that "in fashioning [a] plan, the court should follow the appropriate Section 5 standards, including the body of administrative and judicial precedents developed in Section 5 cases." McDaniel v. Sanchez, 101 S. Ct. at 2236 quoting S.Rep.No.94-295, 94th Cong., 1st Sess., 18-19 (1975).

Section 5 of the Voting Right Act is applicable whenever a covered jurisdiction "shall enact or seek to administer any voting qualification or prerequisite to voting or standard, practice, or procedure with respect to voting different from that in force or effect on November 1, 1972." 42 U.S.C. § 1973c. The Supreme Court in Georgia v. United States, 93 S. Ct. at 1708 expressly held that an apportionment plan is a standard, practice, or procedure requiring section 5 clearance either by a declaratory judgment of the United States District Court for the District of Columbia or by appropriate action of the Attorney General of the United States. Clearance of the apportionment plan is achieved by a determination of the United States District Court for the District of Columbia or the United States Attorney General that the plan "does not have the purpose and will not have the effect of denying or abridging *944 the right to vote on account of race or color, or in contravention of the guarantees set forth in section 1973b(f) (2) of this title." 42 U.S.C. § 1973c. Put another way, if the appropriate parties are satisfied that the apportionment plan does not have a racially discriminatory purpose or effect, the plan will be enforceable as law.[15]See 28 C.F.R. § 51.19; Beer; Panior v. Iberville Parish School, 5th Cir., 536 F.2d 101 at 104-05.

Consequently, this Court, in order to implement a racially fair plan, must devise a proposal that has neither a racially discriminatory purpose nor effect.[16] Regarding the purpose requirement, a district court's plan, by definition, cannot have a racially discriminatory purpose if one of its two express goals is to implement a racially fair scheme. While this is arguably a bootstrap proposition, it is, nevertheless the appropriate analysis. The federal courts are at times summoned to implement remedial plans as a result of a state's purposeful use of an apportionment scheme to dilute minority voting strength. Accordingly, a federal court is called upon to apply the appropriate analysis and implement a plan that is racially fair and avoid any possible malevolent purpose of the states.[17]

As noted, the second requirement of section 5 is a congressional apportionment plan that does not have the "effect" of denying or abridging the right to vote on account of race or color. The Supreme Court case of Beer v. United States is the definitive statement on what the section 5 term "effect" means. See also Georgia v. United States, 411 U.S. 526, 93 S. Ct. 1702, 36 L. Ed. 2d 472 (1973). In Beer, after analysis and review of the legislative history of section 5, the Court stated that "the purpose of section 5 has always been to ensure that no voting procedure changes would be made that would lead to a retrogression in the position of racial minorities with respect to their *945 effective exercise of the electoral franchise." 96 S. Ct. at 1364.

In order for this Court to facilitate implementation of its own plan, it is necessary to analyze the Supreme Court's definition of the section 5 effect standard. Initially, it is helpful to review the Supreme Court's position regarding an "effective exercise of the electoral franchise."

There are certain truisms regarding the effects of apportionment plans. Initially, no group is entitled to be included in a single legislative district, see United Jewish Organizations v. Carey, 430 U.S. 144, 97 S. Ct. 996, 51 L. Ed. 2d 229 (1977), or to have its political clout maximized. Beer v. United States, 96 S. Ct. at 1361 n.8; City of Richmond v. United States, 422 U.S. 358, 95 S. Ct. 2296, 45 L. Ed. 2d 245; Whitcomb v. Chavis, 403 U.S. 124, 91 S. Ct. 1858, 1875-78, 29 L. Ed. 2d 363 (1971); Taylor v. McKeithen, 499 F.2d at 909 quoting Turner v. McKeithen, 490 F.2d 191, 197 (5th Cir. 1973). It is equally true that no group is entitled to proportional representation. City of Mobile v. Bolden, 446 U.S. 55 at 67, 100 S. Ct. 1490, 1504, 64 L. Ed. 2d 47 (1980); White v. Regester, 93 S. Ct. at 2339; Whitcomb v. Chavis, 91 S. Ct. at 1872. "The Constitution ... does not demand that each cognizable element of a constituency elect representatives in proportion to its voting strength. White v. Regester [412 U.S. at 765-66, 93 S. Ct. at 2339, 37 L. Ed. 2d at 234]." Wyche v. Madison Parish Police Jury, 635 F.2d at 1159-60 quoting Nevett v. Sides, 571 F.2d 209, 216 (5th Cir. 1978), cert. denied, 446 U.S. 951, 100 S. Ct. 2916, 64 L. Ed. 2d 807 (1980). See also Whitcomb v. Chavis, 403 U.S. 124, 91 S. Ct. 1858, 29 L. Ed. 2d 363 (1971); Marshall v. Edwards, 582 F.2d at 934-36; Dove v. Moore, 539 F.2d 1152, 1155 (8th Cir. 1976).

Further, a minority group is not entitled to be represented by a member of the group. This Court recognizes there is language indicating the basis of some constitutional claims to be that a particular election system prevents a minority group from electing "members of its race" to the governing bodies. However, the race of the person ultimately elected is not the factor determinative of whether an apportionment plan is racially fair.[18]

Indeed, the implementation of an apportionment plan that attempts to provide proportional representation or "safe" districts districts in which a racial minority group "probably" could elect a member of its group has no guarantee of being efficacious.[19] Initially, such a plan must necessarily presume that racial and ethnic groups will vote as a single unit. However, "racial and ethnic groups do not always vote in solid phalanxes and ... the realities of partisan politics may enable a minority in some circumstances to exact more political concessions by swinging its vote to one of two candidates whose majority-race following is approximately equal than it could by selecting a candidate of its own identity." Wyche v. Madison Parish Police Jury, 635 F.2d 1160-61 citing Marshall v. Edwards, *946 582 F.2d at 934-36.[20] Additionally, "[a] court-ordered racial gerrymander which would assure that blacks form a sizable electoral majority in a single district may not be nearly as effective a guarantee of access as the creation of two or more districts with substantial black voter populations such that all candidates in those districts must be responsive to the needs and aspirations of the black electorate." United States v. Board of Sup'rs of Forrest Cty., 5th Cir., 571 F.2d 951 at 956. See also Panior, 536 F.2d at 105; Taylor v. McKeithen, 499 F.2d at 902; Marshall v. Edwards, 582 F.2d 927, 936-37. Finally, there can be little doubt that effective representation is not a function of ethnicity alone.

Accordingly, the efficacy of an apportionment plan that attempts to provide proportional representation or safe districts is dubious at best. It is at least in part because of this factor that the courts have been admonished against fashioning an apportionment scheme with an eye toward affirmatively creating racially balanced representation, even as a remedial measure.[21]Wyche v. Madison Parish Police Jury, 635 F.2d at 1161; Marshall v. Edwards, 582 F.2d 927.

The decisive factor in every case is whether a minority group is allowed "effective participation in the political process." Whitcomb v. Chavis, 91 S. Ct. at 1968 citing Reynolds v. Sims, 84 S. Ct. at 1383; Lodge v. Buxton, 639 F.2d 1358 at 1374.[22] While not entitled to be placed in legislative district, to maximize their voting potential, to proportional representation, or to have members of their group elected to office, members of minority groups have a right not to be placed outside the political process in such a manner that they may be unable, or less able, to influence elected officials whether the official is black, Hispanic, or white to be sensitive and responsive to their needs. A "system of government that serves the interest of the people must serve the interest of all the people." Lodge v. Buxton, 639 F.2d at 1374. This means, by definition, members of minority groups *947 have a right to vote without practical hindrance. Beyond this, however, it means the "political process leading to nomination and election [must be] equally open to participation by the group in question." The members of the group must have no less opportunity than do other residents to "participate in the political process and to elect legislators of their choice."[23]White v. Regester, 93 S. Ct. at 2339 citing Whitcomb v. Chavis, 91 S. Ct. at 1872. See also Allen v. State Board of Elections, 393 U.S. 544, 89 S. Ct. 817 at 833, 22 L. Ed. 2d 1 citing Reynolds v. Sims, 377 U.S. 533, 84 S. Ct. 1362, 1370, 12 L. Ed. 2d 506; Georgia v. United States, 93 S. Ct. at 1707 (1973); Wyche v. Madison Parish Police Jury, 635 S. Ct. at 1160; Taylor v. McKeithen, 499 F.2d 893, 902.

"[T]he standard [under Section 5] can only be fully satisfied by determining ... whether the ability of minority groups to participate in the political process and to elect their choices to office is augmented, diminished, or not affected by the change affecting the voting." Beer, 96 S. Ct. at 136 quoting H.R.Rep.No.94-196, p. 60 (emphasis in original). Accordingly, a court-ordered congressional apportionment plan must not cause retrogression in the ability of minority groups to elect responsive members to Congress regardless of the color of the skin of the person elected.

The ability of this Court to define "access to the political process," while at the same time avoiding the appearance of arbitrariness, is a difficult task. Neither the Congress nor the courts have articulated a precise prescription. There is no mechanical formula. Indeed, stating that a certain percentage of a minority population in a single district will always indicate access to the political process, and that less than that fixed percentage will always indicate no access or reduced access to the political process, would be, in many instances, an error by the Court. This is because the size of the minority population necessary for political impact depends on the district itself. This Court would be amiss to attempt to establish a mechanical or fixed percentage that could be universally applied without question.[24]

This Court faces a similar and related problem with defining and applying a principle of retrogression. Again, neither Congress nor the courts have articulated explicit directives. Determining whether a new apportionment plan increases or decreases the voting power of minority group members relative to the prior plan is necessarily imprecise. It is, above all things, problematic. See Beer, 96 S. Ct. at 1370 n.12 (Marshall, J., dissenting).

This Court's standard is fairness and equity. This standard, together with its inherent limitations, pretermits uniform explanations in every instance. This Court is directed to exercise its equitable jurisdiction and discretion and avoid the taint of arbitrariness by providing reasonable and rational explanations for its actions. It is not required to view every diverse facet with a single myopic eye. Therefore, no fixed percentage or formula will be seen as applicable to every circumstance without variance. This is particularly true of a state with the size and diversity of the State of Texas.

This Court begins its approach by recognizing that, in examining the effect of its plan and analyzing whether that effect causes a retrogression in minority voting strength, the Court cannot limit its evaluation *948 to those districts that contain such a high population of minorities that it "probably will be represented" by a member of the minority group. As noted previously, minority groups are capable of significant access to the political process in districts in which there is little chance that they probably will be able to elect a member of a minority group. This form of access is a function of the group's ability to "swing" the election's outcome in favor of one candidate or another.[25]

Another related factor in determining retrogression is the size of the minority population that the district had under the previous plan. If a district has a significant minority population, such as a percentage over 65 percent, minor changes occasioned by an apportionment plan, either up or down, probably will have little influence on the minority group's access to the political process. It will depend, of course, upon whether the district is truly safe or, for whatever reason, only marginally safe.[26] On the other hand, minority group members residing in districts that have historically had small minority populations are not likely to realize a retrogression in voting strength as a result of a decline in their numbers. They simply had little electoral impact from the outset. Conversely, their electoral clout can be increased only through an apportionment plan providing substantial increases in their percentages.

Districts with moderate minority populations neither high nor low percentages are the most difficult to analyze. This is a function of the difficulty in determining the "magical" percentage that might place a minority group in the position of being capable of influencing an election. When the district's minority percentage is moderate, political factors such as incumbency, economics, education, and the minority group's ability to build political coalitions become increasingly critical to the determination of whether the minority group has the ability to elect government officials of its choice.

Significantly, this Court may not consider the innumerable political factors that may affect a minority group's access to the political process. It is left to consider only the minority populations and from those percentages determine whether its plan is retrogressive. It is also clear that figures for total population are not the determinative criteria for analyzing retrogression. "[T]he Supreme Court has recognized that `if it is the weight of a person's vote that matters, total population even if stable and accurately taken may not actually reflect that body of voters whose vote must be counted and weighed for purposes of reapportionment, because `census persons' are not voters.'" Wyche, 635 F.2d at 1162 quoting Gaffney v. Cummings, 412 U.S. 735, 93 S. Ct. 2321, 2328, 37 L. Ed. 2d 298 (1973). As a result, "[a]n important factor, perhaps the single most significant one in determining vote dilution," is the voting age population of the congressional districts. Id. at 1162 and cases cited therein.

This Court also acknowledges the point of view that an apportionment plan must be evaluated on a statewide basis and focus should not be isolated on individual areas of the state that may be less favorable to a minority. Piecemeal review of the plan may be misleading. Connor v. Finch, 97 S. Ct. at 1840 through 41 (Blackmun, J., concurring). The Court, nevertheless, must face the realities of the state and recognize that minority group members are not spread evenly throughout its borders; in many instances, if not most, minority populations are concentrated in particularized *949 geographic areas. Consequently, much of the Court's analysis will necessarily focus on individual areas of the State. Specifically, in Texas, three major minority concentrations appear in the Dallas, Houston, and South Texas areas.

This particularized analysis focuses on a possible result of the Court's definition of districts in areas having concentrated minority populations. The Court must take care to avoid "packing" a group of minorities into a single congressional district and, by doing so, reduce the political strength previously enjoyed by a part of the minority group. This Court briefly addressed this issue in its foregoing discussion regarding the problem of attempting to guarantee that the creation of a "safe" district would be an effective means of enhancing a minority group's access to the political process. As noted, there can be no guarantee that such an election system will be efficacious. To the contrary, "[t]here is no agreement on whether the political interests of a minority group are best maximized by an overwhelming majority in a single district, bare majorities in more than one district or a substantial proportion of the voters in a number of districts." Panior, 536 F.2d at 105 quoting Turner v. McKeithen, 490 F.2d 191, 197 n.24 (5th Cir. 1973).

Attempts to create a single strong district normally may be accomplished only by diluting minority voting power in surrounding districts to the point that elected officials "could ignore with impunity the special needs of [minority group members] in those districts." See Taylor v. McKeithen, 499 F.2d at 902. A congressional scheme such as this could reasonably be termed retrogressive. This is particularly true if the so-called single strong district fails even to provide a "safe" seat but only enhances the minority group's ability to swing an election one direction or another, since the minority group conceivably was capable of achieving this result before the reapportionment.[27]

An additional and final factor faced by this Court is the fact that Texas was apportioned three new congressional districts. This precludes a one-on-one comparison between the existing apportionment plan and this Court's plan. Accordingly, this Court must provide some reasonable methodology for analyzing the increased number of districts in order to determine whether the Court's method of handling these three districts has a retrogressive effect. This Court initially bases its analysis of the new districts and their contribution to the Court's plan by a two-step analysis. First, the Court determines what percentage of the state is minority population. This is found to be 33 percent. Second, the Court compares the average minority population of the three new congressional districts to the statewide minority percentage of 33 percent. If the average minority population percentage of the three new congressional districts is greater than the statewide minority percentage, which it is,[28] a reasonable, although somewhat superficial, conclusion may be made that the method of incorporating the three new congressional districts did not cause a retrogression in minority voting strength.

*950 An additional method for analyzing the effect of the three new congressional districts is to determine the net difference between the three districts' minority populations and the statewide minority population. This makes mathematical adjustments for any fragmentation of the minority population among the three districts. It is achieved by first finding the numerical variances in the minority percentages for the individual districts and the statewide minority population percentage of 33.0%. District 25 has a minority population of 38.7 percent. This makes for a + 5.7 percent difference from the statewide total. There is a + 31.25 percent difference between the statewide total and District 27, which has a minority population of 64.25 percent. District 26 has a minority population of only 7.01 percent. As a result, it differs from the statewide minority population by-25.99 percent.

The net difference between the minority population in the three new districts and the total statewide minority population is + 10.96. This positive difference leads to the reasonable, although again somewhat superficial conclusion that the Court's method of incorporating the three new congressional districts into a congressional apportionment scheme, standing alone, does not cause a retrogression in minority voting strength.[29]

 
II. The Court-Ordered Plan

As noted, this Court acknowledges it has two coterminous constitutional duties or requirements that must be fulfilled in order for the Court to properly implement a congressional apportionment plan. The Court must devise and implement a plan that provides both voter equality and racial fairness. As will be demonstrated in greater detail, this Court, in fashioning its remedy, has paid great deference to the proposals of the State of Texas as they were expressed by the Legislature in passing the ill-fated S.B. 1. Specifically, the great majority of S.B. 1's provisions are incorporated in full. All but six of the districts created by S.B. 1 District 3, 4, 15, 24, 26, and 27 are utilized in the Court's plan precisely as they are defined in the legislative enactment.

 
A. Voter Equality

The State of Texas is entitled to twenty-seven representatives to the United States Congress. The 1980 census data reveals that the State of Texas has a total population of 14,228,338 people. Accordingly, each of the twenty-seven districts from which Texas will elect its representatives should have a total population of 526,977, or a de minimis variation from this number. The Court's plan satisfies the constitutional objective of voter equality under the most stringent requirements of mathematical exactitude.

The average deviation in district population from the ideal district under the Court's plan is ± 0.05%. The average variance in population is ± 257 people. The population and deviation from the ideal population for each district is set out for convenience.

 
  POPULATION EQUALITY ACROSS DISTRICTS
         IN COURT-ORDERED PLAN
    Ideal District Population = 526,977
District       Population           Deviation
   1            527,016               +.01
   2            526,772               -.04
 
*951
  POPULATION EQUALITY ACROSS DISTRICTS
          IN COURT-ORDERED PLAN
District    Population     Deviation
   3          526,853        +.02
   4          526,991         -0-
   5          526,792        -.04
   6          527,393        +.08
   7          527,083        +.02
   8          527,528        +.10
   9          526,443        -.10
  10          526,943        -.01
  11          526,871        -.02
  12          527,074        +.02
  13          526,840        -.03
  14          526,920        -.01
  15          527,203        +.04
  16          527,401        +.08
  17          526,831        -.03
  18          527,393        +.08
  19          527,805        +.16
  20          526,333        -.12
  21          527,044        +.01
  22          526,602        -.07
  23          526,976        -.08
  24          526,677        -.06
  25          526,801        -.03
  26          527,285        +.06
  27          526,941        -.01

While only one district has a 0% deviation in population from that of the ideal district, the other districts have only very minor deviations. The congressional district with the largest total population is District 19. It has a population of 527,805 people or 828 people more than ideal. Therefore, it deviates from the population norm by + 0.16%. On the other end of the scale, Congressional District 20 has the smallest total population with 526,333 people. This is 644 people less than ideal. As a result, it deviates from the population norm by -0.12%. The difference in population between the largest and smallest districts is 1,472 people. The top to bottom deviation is ± 0.28%.

Even the de minimis deviations from absolute population equality can be explained. As nearly as practicable, the plan attempts to comply with the State of Texas' goals of preserving certain traditional and historical boundaries. Additionally, there has been an attempt to provide compact and contiguous districts. Finally, the Court recognizes there is no precise numerical definition of a de minimis deviation from the ideal population. However, the Court's remedial directive comes as close to mathematical exactness as may be practically possible.

 
B. Racial Fairness

In fashioning a remedial apportionment plan, this Court follows the appropriate standards of section 5 of the Voting Rights Act, which requires a proposal that has neither a racially discriminatory purpose nor effect. The purpose of the Court's plan is articulated in its two express objectives providing voter equality and racial fairness. Consequently, one prong of section 5 is met.

Regarding the effect prong of section 5, the plan must not lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise. Accordingly, this analysis begins with an evaluation of the position of racial minorities under the apportionment plan S.B. 1 was to replace.[30]

 
*952 (1) The Pre-Existing Congressional Apportionment Plan

The previous election system, with twenty-four congressional districts, provided minorities with three districts in which they "probably" could elect a member of a minority group to represent the district. This is a result of the high minority profile of the districts. Those districts were Districts 15, 18, and 20. District 15 had a minority population of 77.8 percent. District 18 was 74.0 percent minority, and District 20 had a minority population of 77.6 percent.[31]

District 16 under the previous plan also should be mentioned. It reasonably could not have been classified as a "safe" district, since it had a total minority population of only 60.5 percent. However, it did provide substantial access to the minority population because of its high minority profile. The same is true of District 23, which had a total minority population of 57.8 percent. Under the previous plan, District 14 had a minority profile of 45.4 percent. This percentage reveals minorities had a significant potential for influencing an election.[32]

The voting age population of Districts 15, 18, and 20 echoes the fact they were probably safe minority districts. Of the voting age population in District 15, 72.23 percent was minority. The minority breakdown of the voting age population in Districts 18 and 20 was 67.93 percent and 72.27 percent, respectively.

Regarding the previous District 16, minorities comprised 55.20 percent of the district's overall voting age population. 52.15 percent of the voting age population in District 23 was minority. In District 14, minorities constituted 40.39 percent of the voting age population.

This analysis addresses a statewide view of districts in which minorities reasonably could be characterized as having substantial access to the political process. This access is a product of the group's sheer numbers in relation to the individual districts' total population and voting age population. This Court's analysis must proceed to an evaluation of minority voting strength in particular areas of the State. This is necessary because of the noteworthy minority population that is generally concentrated in three areas of the State. As a result of this concentration, a retrogression in access may result from fairly simple and minor movements of district boundaries.

The Dallas County area is one of the three areas containing a significant minority population. This population is heavily concentrated very near the exact geographic center of the county.

Under the previously existing congressional districting scheme, Dallas County was divided among four districts Districts 3, 5, 6, and 25. District 3 drew its population from the northwest corner of Dallas County, the southeast corner of Denton County, and the southern part of Collin County. Congressional District 3 had a very small total minority population. Its total minority percentage was only 8.7 percent. In addition, minorities constituted only 7.88 percent of the district's voting age population. Consequently, minorities had little electoral impact in District 3.

District 6, which ran along the south side of Dallas County and then southeast through central Texas, had a total minority population of 21.5 percent. Minorities constituted 18.08 percent of the district's voting population. Standing alone, these figures demonstrate limited minority impact. In addition, it is critical to note that only a *953 small proportion of District 6 actually impacted on Dallas County. The limited total minority population of District 6 was spread throughout the district's boundaries, which ran from Parker County at its far northern point to Brazos County at its most southerly point.

Districts 5 and 24 had significant minority populations. The line serving as a boundary for these two districts was the Trinity River, which runs through the center of the minority community in Dallas. Under the pre-existing electoral system, Congressional District 5 had a minority population of 29.1 percent. Minorities constituted 26.18 percent of the district's voting age population. In District 24, minorities comprised 37.4 percent of the total population and 32.68 percent of the voting age population.

These percentages manifested themselves in rather substantial political influence in both District 5 and 24.[33] Minority group members had access to the political process as a result of their ability to "swing" the outcome of a congressional election to one candidate or another. In the 1978 general election, the white congressional representative of District 24 actually lost the overall white vote to his opponent by approximately 5,000 votes. He carried black precincts, however, by approximately 11,000 votes or 75 percent of those cast by black voters. As a result, he won an election by 5,887 votes in which approximately 72,500 votes were cast. In the 1980 general election, the representative of District 24 carried the white vote against a black opponent only by about 10,000 votes. He received, however, 94 percent of the black votes cast. As a result, he defeated his opponent by 34,518 out of 152,862 total votes cast.[34]

The white congressional representative of District 5 defeated his closest opponent in the 1978 general election by 852 votes. In the 1980 general election, he won by 3,044 votes. The Congressman lost the white vote in both elections by a significant margin. However, he carried over 90 percent of the black vote in those elections. His ability to capture such a high percentage of the vote from a group that actually constituted the minority of voters resulted in his success in the two elections.[35]

In summary, a particularized evaluation of Dallas County as it existed under the previous plan reveals a concentrated minority with significant access to the political process in two congressional districts.

A second area of the State containing a concentrated minority population is the Harris County or Houston area. Under the pre-existing congressional apportionment plan, Harris County was divided among five congressional districts Districts 7, 8, 9, 18, and 22.

A segment of the northeast corner of Harris County provided part of the population for Congressional District 9. The remainder *954 of the district's population was located in Galveston, Chambers, and Jefferson Counties. Minorities in District 9 comprised 29.6 percent of the total population and 26.77 percent of the voting age population. These percentages indicate the minority population had a degree of influence in District 9.

Congressional District 22 picked up part of its population from the southern tip of Harris County. The remainder of the district's population was spread throughout all of Brazoria and Fort Bend Counties, and part of Waller County. District 22 had a total minority population of 31.2 percent. In addition, minorities constituted 27.99 percent of the district's voting age population. Considering raw percentages, it is reasonable to conclude minorities possessed influence in District 22.

District 7, 8, and 18 were totally contained within the boundaries of Harris County. District 18, as already noted, possessed such a very high percentage of minorities that it reasonably could be labeled a "safe" minority district. While not a safe district, District 8, due to its high minority profile, provided minority group members with potentially significant impact. Minorities made up 40.1 percent of the total population and 36.07 percent of the voting age population. District 7, on the other hand, provided minorities with little impact. Only 14.0 percent of the total population was comprised of minorities, and minorities constituted a mere 12.86 percent of the voting age population.

An examination of the Harris County area reveals that, under the pre-existing congressional apportionment plan, minorities had a high level of political influence. One district, District 18, could be characterized as a safe minority district. Three others Districts 9, 22, and 8 provided minority group members with moderate to significant political influence.

The third and final specific area of concentrated minority population exists in South Texas. This area, which borders on the Gulf of Mexico and the Country of Mexico, has a significant Hispanic population. Under the pre-existing congressional apportionment plan, there were basically six districts apportioned to the South Texas area Districts 14, 15, 16, 20, 21, and 23.

Districts 14, 15, 16, 20, and 23 have already been discussed. Districts 15 and 20, with total minority populations of greater than 70 percent, reasonably could be labeled safe minority districts. While not safe districts, Districts 14, 16, and 23, with total minority populations of greater than 45.4 percent, provided minorities with potentially substantial political influence.

The remaining district District 21 had a minority population that comprised 28.5 percent of its total population. In addition, minorities made up 24.04 percent of the district's voting age population. These percentages, while not dominant, indicate minority group members had the potential for being able to exercise an effective political voice.

In summary, minority group members had the potential for influence in the electoral process under the previous apportionment plan. Three districts had minority percentages greater than 70 percent. Two others had percentages greater than 50 percent. In areas of particularized minority concentrations, the pre-existing plan provided several districts in which, based upon percentages, minorities enjoyed the opportunity for electoral influence.

 
II. The Court's Plan

The apportionment plan devised and implemented by this Court provides racial fairness. In other words, it does not lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise.[36]

*955 At the outset, the same three safe minority districts are maintained. These districts are Districts 15, 18, and 20. Under the Court's plan, minorities constitute 72.21 percent of District 15's total population. They constitute 66.61 percent of the voting age population. Under the Court's plan, minorities in District 18 comprise 72.0 percent of the district's total population and 66.17 percent of the voting age population. Finally, in District 20, minorities constitute 70.5 percent of the total population and 64.8 percent of the voting age population.

The Court recognizes that in each of these three districts there is some decline in the total minority population. In no instance, however, does the population fall below 70.0 percent. Districts 15, 18, and 20 each experience a decline in the percentages. Once again, however, the decline is to a percentage that is so substantial it cannot reasonably be said to have resulted in a retrogression.

Minorities also maintain the three districts in which they had substantial access as a result of their percentages of the total population. The minority composition of District 16 actually increased on a percentage basis. Under the Court's plan, minorities now make up 63.9 percent of the total population. They also enjoyed an increase in their percentages of voting age population. They now comprise 58.76 percent of the total voting age population. In District 23, the difference between the court-ordered plan and the pre-existing plan is miniscule. In District 14, there is a decline in the minority population in terms of both total population and voting age population. The decline, however, is to a percentage that still allows impact in the electoral process.

In addition to maintaining three safe districts and the two districts with minority population percentages greater than 50 percent, *956 minorities gain an additional district with a minority percentage above 50 percent. Indeed, District 27, which is one of the three new districts Texas is entitled to receive, has a total minority population of 64.25 percent. Minorities constitute 58.23 percent of the voting age population. Due to its high minority profile, District 27 adjusts for the rise in Texas' minority population relative to the rise in the State's total population. Hispanic population in Texas increased from 18 percent of the total population to 21 percent between 1970 and 1980. While not required to guarantee proportional representation for minorities, an adjustment for the increase in minority population should be made in order to avoid implementing a plan that occasions a retrogression in the effect of the minority's vote. District 27 serves that function. While this district may fall short of being a safe district, it does provide a potentially significant voice for minority group members.

At this point, the Court's analysis is directed at determining whether, because of some shift in boundary lines, there is a retrogression in minority voting strengths. Accordingly, it is appropriate to address the particular areas of the State that have concentrated minority populations.

Dallas County is impacted by the same four districts that previously divided it. Additionally, the court-ordered plan adheres to county lines to the degree practicable in an effort to satisfy an express goal of the State Legislature. To the extent county lines are not preserved, the resulting district configurations are in accordance with traditional and historical configurations of congressional districts in the Dallas County area.[37]

Minorities enjoyed very limited electoral influence in District 3 under the pre-existing apportionment plan. Because of this, it cannot be said the Court's plan leads to a retrogression in minority voting strength, although the minority population is decreased 1.29 percentage points under the Court's plan, from 8.7 percent to 7.4 percent. District 6 also has a lower minority population than it did previously. Once again, however, the small decline in minority population occurs in a district that previously provided most limited access. It cannot be labeled retrogression.

In the two dominant Dallas County districts Districts 5 and 24 the Court's plan avoids a retrogression in minority voting strength. The plan utilizes the historical and natural boundary the Trinity River to divide the two districts. The result is that minorities comprise 31.87 percent of District 5's total population and 28.12 percent of its voting age population. In District 24, minorities make up 45.7 percent of the total population and 40 percent of the voting age population.[38] In both instances, there is a slight increase in minority population of these two districts that serve as the foundation for minorities being capable of effectively exercising the elective franchise in Dallas County.[39]

*957 In the Harris County area, the Court's plan does not cause a retrogression in minority voting strength. As previously noted, the Court's plan preserves District 18 as a safe minority district. In addition, District 7 under the Court's plan maintains its basic characteristics. It contained a small percentage of minorities originally, and continues to do so with a minority percentage of 10.3 percent. The impact on District 9 was to remove any part of Harris County from the congressional district's boundaries. Minorities, nevertheless, maintained virtually the same impact previously enjoyed in District 9.

Districts 8 and 22 have different characteristics under the Court's plan. This is a result of Harris County's growth and the necessity of creating a new congressional district in the area. District 8 reflects a decline in minority population. The decrease, however, is not so dramatic that it rules out potential minority influence. Minorities continue to comprise 29.2 percent of the district's total population and 25.8 percent of the voting age population. These percentages indicate minorities will continue to have the ability to impact elections significantly.

District 22 realized an arguably significant decrease in minority population. It now has a total minority population of 23.2 percent, as opposed to the 31.2 percent population under the previous plan. The drop in voting age population was from 27.99 percent to 20.93 percent.

This decline, together with the decline in District 8, does not amount to a retrogression in Harris County minority voting strength, however. This is because of the gain in a completely new district in which minorities will have potentially substantial electoral impact. The new district is District 25, which contains a minority population of 38.7 percent. Minorities make up 34.19 percent of the district's voting age population.

Accordingly, considering the districts together, there is no retrogression in minorities' voting strength. Generally, minorities in the Harris County area have one safe district District 18. Additionally, there are three districts District 8, 9, and 25 in which minorities have a potential for influencing the outcome of elections.

Another means of examining the Harris County area is to exclude Districts 9 and 22 from the evaluation. This is done for a practical reason. The reason is District 9 is now entirely outside the boundaries of Harris County and District 22's population is now comprised of a very small percentage of people living in Harris County. Focus on *958 the four districts that, as a practical matter, are Harris County districts Districts 7, 8, 18, and 25 reveal substantial minority influence. Minorities have population percentages ranging from 29.2 percent to 72.0 percent in three out of the four districts. Consequently, the Court's plan cannot be said to dilute minority voting strength in Harris County.[40]

Finally, the effect of the court-ordered plan on the South Texas area must be examined. At the outset, it should be recognized that the South Texas area, due to its population growth, has been apportioned an additional district. There are now seven congressional districts bordering the Gulf of Mexico and the Country of Mexico.

As noted, the two safe minority districts that existed previously are preserved. In addition, Districts 14, 16, and 23 maintain significant minority populations. The total minority populations range from a low of 31.6 percent in District 14 to a high of 63.9 percent in District 16. In all three districts, minorities constitute greater than 27 percent of the voting age population. It should be noted that District 21 had a slight drop in minority population. However, this cannot be said to have caused a retrogression in minority voting strength for two reasons. First, the decline was small. Second, minorities had limited access in District 21 under the previous plan.

The major factor in South Texas was the creation of a new congressional district. This district, District 27, was to compensate for the large population growth in South Texas particularly among Hispanics. This district serves two functions. It provides an additional district to help compensate for overall population growth in the State. It also, due to its minority profile, adjusts for the rise in minority population. As discussed, the district with a minority population of 64.3 percent provides minorities with substantial access to the political process.[41] Consequently, the Court's District 27, together with the remaining districts in South Texas, do not cause a retrogression in minority voting strength.

 
III. Conclusion

This Court, in fashioning a remedial decree, has sought two constitutional objectives. First, it has attempted to provide a plan that guarantees voter equality. Second, the Court has sought racial fairness. In recognition of the fact that apportionment is a legislative task in the first instance, this Court has deferred to the wisdom of the State Legislature in every instance it found acceptable under the standards applicable to a court-ordered plan.

The exercise of this unwelcome judicial responsibility is taken for the sole purpose *959 of insuring and facilitating, to the degree practicable, a timely and orderly elective process provided for by the laws of this State. The Court's plan is a temporary interim plan. It is to be in effect, however, for the 1982 primary and general elections, and for any other intervening or further congressional elections as may hereafter occur until such time as the Texas Legislature enacts a plan of its own for the apportionment of Texas' congressional districts and such plan is found to be legally enforceable by this Court. It is to be emphasized that this order is without prejudice to the legislative and executive branches of the State of Texas to proceed with the consideration and adoption of a constitutionally permissible plan of congressional districting at a called or regular session of the legislature of the State of Texas for any congressional elections other than the 1982 primary or general elections. This Court expressly retains jurisdiction over this proceeding, including, but not limited to, applications for implementation, modification, enforcement, or other relief relating to the apportionment of Texas' congressional districts.

The Court's plan is attached hereto as an appendix.

 
APPENDIX 
District 1

Same as Congressional District 1 as defined by Senate Bill No. 1, which was enacted by the First Called Session of the 67th Legislature of Texas on August 10, 1981, and signed by the Governor of Texas on August 14, 1981 (hereinafter referred to as S.B. 1).

 
District 2

Same as Congressional District 2 as defined by S.B. 1.

 
District 3

District 3 is composed of that part of Collin County included in census tracts 313.02, 317, 318.01, and 320.01; and that part of Dallas County included in census tracts 2.01, 71.01, 73.01, 73.02, 74, 75.01, 75.02, 76.01, 76.02, 76.03, 76.04, 77, 78.01, 78.04, 78.05, 78.06, 78.07, 78.08, 78.09, 79.02, 79.03, 79.04, 79.05, 80, 81, 82, 94, 95, 96.03, 96.04, 96.05, 96.06, 96.07, 96.08, 96.09, 97.01, 97.02, 99, 128, 129, 130.02, 130.03, 130.04, 131.01, 131.02, 131.03, 132, 133, 134.01, 134.02, 135, 136.01, 136.04, 136.05, 136.06, 136.07, 136.08, 136.09, 136.10, 137.01, 137.02, 137.04, 137.05, 137.06, 137.07, 137.08, 138.01, 138.02, 139, 140.01, 181.07, 181.08, 181.09, 181.10, 181.11, 181.12, 181.13, 181.14, 185.02, 190.03, 190.04, 190.06, 190.07, 190.08, 190.09, 190.10, 190.11, 190.12, 191, 192.01, 192.02, 192.03, 192.04, 192.05, 192.06, 192.07, 193.01, 193.02, 194, 195.01, 195.02, 196, 197, and 198, and block groups 1 and 4 of census tract 1.

 
District 4

Same as Congressional District 4 as defined by S.B. 1.

 
District 5

District 5 is composed of that part of Dallas County included in census tracts 2.02, 3, 4.01, 4.02, 4.03, 5, 6.01, 6.03, 6.04, 7.01, 7.02, 8, 9, 10, 11.01, 11.02, 12, 13.01, 13.02, 14, 15.01, 15.02, 16, 17.01, 17.02, 18, 19, 21, 22.01, 22.02, 23, 24, 25, 26, 27.01, 27.02, 28, 29, 30, 31.01, 31.02, 32.01, 32.02, 33, 34, 35, 36, 37, 38, 39.01, 39.02, 40, 71.02, 72, 83, 84, 85, 90.01, 90.02, 91.01, 91.02, 92.01, 92.02, 93.01, 93.03, 93.04, 98.01, 98.02, 100, 114.02, 115, 116.01, 116.02, 117, 118, 119, 120, 121, 122.02, 122.03, 122.04, 122.05, 123, 124, 125, 126, 127, 148.01, 148.02, 150, 167.02, 168, 169.01, 169.02, 169.03, 169.04, 170, 171, 172, 173.01, 173.02, 174, 175, 176.01, 176.02, 177, 178.01, 178.03, 178.04, 178.05, 179, 180, 181.04, 181.05, 181.06, 181.15, 182.01, 182.02, 183, 184.01, 184.02, 184.03, 185.01, 186, 187, 188.01, 188.02, 189, 190.13, 190.14, and 190.15, and block groups 2 and 3 of census tract 1.

 
District 6

Same as Congressional District 6 as defined in S.B. 1.

 
District 7

Same as Congressional District 7 as defined in S.B. 1.

 
District 8

Same as Congressional District 8 as defined in S.B. 1.

 
District 9

Same as Congressional District 9 as defined in S.B. 1.

 
*960 District 10

Same as Congressional District 10 as defined in S.B. 1.

 
District 11

Same as Congressional District 11 as defined in S.B. 1.

 
District 12

Same as Congressional District 12 as defined in S.B. 1.

 
District 13

Same as Congressional District 13 as defined in S.B. 1.

 
District 14

Same as Congressional District 14 as defined in S.B. 1.

 
District 15

District 15 is composed of Atascosa, Brooks, Duval, Frio, Hidalgo, Jim Hogg, Jim Wells, Karnes, LaSalle, Live Oak, McMullen, San Patricio, Starr, Wilson, and Zapata counties; that part of Gonzales County included in enumeration districts 229, 230, and 231B; and that part of Nueces County included in census tract 37.

 
District 16

Same as Congressional District 16 as defined in S.B. 1.

 
District 17

Same as Congressional District 17 as defined in S.B. 1.

 
District 18

Same as Congressional District 18 as defined in S.B. 1.

 
District 19

Same as Congressional District 19 as defined by S.B. 1.

 
District 20

Same as Congressional District 20 as defined by S.B. 1.

 
District 21

Same as Congressional District 21 as defined by S.B. 1.

 
District 22

Same as Congressional District 22 as defined by S.B. 1.

 
District 23

Same as Congressional District 23 as defined by S.B. 1.

 
District 24

District 24 is composed of that part of Dallas County included in census tracts 20, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 59.01, 59.02, 60.01, 60.02, 61, 62, 63.01, 63.02, 64, 65, 67, 68, 69, 86.01, 86.02, 87.01, 87.03, 87.04, 87.05, 88.01, 88.02, 89, 101, 102, 103, 104, 105, 106, 107, 108.01, 108.02, 108.03, 109, 110.01, 110.02, 111.01, 111.02, 112, 113, 114.01, 140.02, 141.01, 141.02, 141.03, 141.04, 142, 143.01, 143.02, 143.03, 143.04, 144.01, 144.02, 145, 146, 147, 149, 151, 152.01, 152.02, 153.01, 153.02, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165.01, 165.02, 165.06, 166.01, 167.01, and 199; and that part of Tarrant County included in census tracts 1217.02, 1218, 1219.01, 1219.02, 1220, 1221, 1222, 1223, 1224, and 1229, and block groups 4, 5, 6, 7, 8, and 9 of census tract 1130.

 
District 25

Same as Congressional District 25 as defined by S.B. 1.

 
District 26

District 26 is composed of Denton County; that part of Collin County included in census tracts 303, 304, 305, 306, 307, 314, 315, 316.01, 316.02, 316.03, 316.04, 316.05, 316.06, 316.07, 318.02, 318.03, 319, 320.02, and block groups 4, 5, and 6 of census tract 308; and that part of Cooke County included in enumeration districts 325, 328, 336, 337, 338A, 338B, 339T, 339U, 340T, and 340U; and that part of Tarrant County included in census tracts 1013.01, 1023.02, 1042.01, 1042.02, 1054.01, 1054.03, 1054.04, 1055.01, 1055.02, 1055.03, 1055.04, 1056, 1057.01, 1057.02, 1065.02, 1065.03, 1065.04, 1065.05, 1108.03, 1109.01, 1109.02, 1110.01, 1110.03, 1110.04, 1112.01, 1113.01, 1113.02, 1115.03, 1115.04, 1115.05, 1115.06, 1115.07, 1115.08, 1115.09, 1115.10, 1131, 1135.03, 1135.04, 1135.05, 1135.06, 1136.03, 1136.04, 1136.05, 1137.01, 1137.02, 1216.01, 1216.04, 1216.05, 1216.06, 1216.07, 1217.01, 1225, 1226, 1227, and 1228, and block groups 1, 2, and 3 of census tract 1130.

 
*961 District 27

District 27 is composed of Cameron, Kenedy, Kleberg, and Willacy Counties; and that part of Nueces County not included in District 15.

JUSTICE, Chief Judge, concurring in part and dissenting in part.

For reasons which I shall set forth, the decision of the court to implement a court-ordered congressional apportionment plan, in the wake of the failure of the Texas legislature to adopt a suitable plan, requires that the plan ultimately adopted be free of any taint of arbitrariness or discrimination. Alternatively, even if a court-ordered plan which mirrors the boundaries of a legislative plan is to be evaluated under the more relaxed standards applicable to legislative plans, the court must conduct an intensely local appraisal of the design and impact of the congressional apportionment plan as passed by the legislature. On the basis of either mode of analytical review, I believe the conclusion is inescapable that a portion of the plan implemented by the court offends the Fourteenth and Fifteenth Amendments of the United States Constitution. Accordingly, I dissent from those portions, and concur separately in the remainder of the court's plan.

 
I. 
The Court's Unwelcome Obligation

On January 29, 1982, the Attorney General of the United States interposed an objection, pursuant to Section 5 of the Voting Rights Act of 1965, as amended, 42 U.S.C. § 1973c, to Senate Bill No. 1 (1981), the congressional reapportionment plan passed by the Texas State Legislature. As the Attorney General stated in his objection letter, "the effect of this objection is to render the implementation of the provisions of Senate Bill No. 1 legally unenforceable." In the absence of a reapportionment plan which has been pre-cleared pursuant to Section 5, the Voting Rights Act "essentially freezes the election laws of the covered state." Georgia v. United States, 411 U.S. 526, 538, 93 S. Ct. 1702, 1709, 36 L. Ed. 2d 472 (1973). In the case of congressional reapportionment in the State of Texas, this result is impermissible. As a constitutional matter, under the 1980 census figures, Texas is entitled to 27 Congressional seats. The existing apportionment plan provides for only 24 Congressional seats. Accordingly, the existing Congressional apportionment plan for Texas must be modified before the 1982 elections.

It is axiomatic that reapportionment is a matter properly delegated to legislative bodies. However, recent experience teaches that these entities frequently fail to perform their duties in a manner that satisfies the requirements of the Constitution and the Voting Rights Act. In such instances, "it becomes the `unwelcome obligation' of the federal court to devise and impose a reapportionment plan pending later legislative action." Wise v. Lipscomb, 437 U.S. 535, 540, 98 S. Ct. 2493, 2497, 57 L. Ed. 2d 411 (1978), quoting Connor v. Finch, 431 U.S. 407, 415, 97 S. Ct. 1828, 1834, 52 L. Ed. 2d 465 (1977). This doctrine of equitable jurisdiction was developed in the context of litigation in which the legislative plan had been struck down by a federal court on constitutional grounds. However, the court in Wise went on to note the "distinctive impact of § 5 [of the Voting Rights Act] upon the power of States to reapportion themselves." 437 U.S. at 541-42, 98 S. Ct. at 2497-98. After detailing the procedural requirements of the Voting Rights Act, and the inherent potential for delay, the Court noted:

 
Pending such submission and clearance, if a State's electoral processes are not to be completely frustrated, federal courts will at times necessarily be drawn further into the reapportionment process and required to devise and implement their own plans.

437 U.S. at 542, 98 S. Ct. at 2498.

Thus, the court today assumes the "unwelcome obligation" of adopting a congressional apportionment plan to fill the void left by the dereliction of the state legislature. In assuming this duty, the court participates in one of the fundamental acts of *962 democratic self-governance. It is precisely the crucial nature of the reapportionment process which renders it typically a legislative task. The legislature's presumed sensitivity to the political and social cross-currents of the state cloaks it with a "political authoritativeness" that other institutions may lack. Connor v. Finch, 431 U.S. at 415, 97 S. Ct. at 1834. Lacking the innate credibility of the legislature, federal courts must act "circumspectly, and in a manner `free from any taint of arbitrariness or discrimination.'" Id., quoting Roman v. Sincock, 377 U.S. 695, 710, 84 S. Ct. 1449, 1458, 12 L. Ed. 2d 620 (1964). In other words, "[a] more stringent standard is applied to judicial apportionments" than is applied by federal courts when they are reviewing the validity of legislatively drafted apportionment schemes. Wise v. Lipscomb, 437 U.S. at 541, 98 S. Ct. at 2497.

The signification of this stringent criterion is not clear, for several reasons. Initially, the analytical framework to be applied to legislative apportionments is imprecise, since the tidewaters of constitutional litigation of voting rights challenges have left a confusing and uncertain legacy. The Supreme Court's decision in Bolden v. City of Mobile, 446 U.S. 55, 100 S. Ct. 1490, 64 L. Ed. 2d 47 (1980), held that the intent requirement applies to constitutional challenges in the area of voting. However, the lack of a majority opinion in Bolden, and also the conflicting discussions of the evidentiary requirements concerning the intent standard, diminish the clarity of that holding. Moreover, the heightened nature of the scrutiny directed to judicial plans has never been clarified, in terms of its effect on the proper framework for analysis. Even if the standard of review mandated by Bolden were clear, the effect on that standard is wholly uncertain, as it relates to the increased sensitivity demanded of courts in Wise v. Lipscomb. Close analysis of the language quoted from Roman v. Sincock, supra, 377 U.S. at 710, 84 S. Ct. 1458, seemingly indicates that one effect of the closer examination is elimination of the intent requirement applied to legislative actions by Bolden. As noted, the Court in Roman v. Sincock commanded that courts act in a manner free from "any taint of arbitrariness or discrimination." Id. This language would appear to be focused solely on the effects of an apportionment plan adopted by the court. The effects test may be the product of the indisputable premise that federal courts are always presumed to act without invidious intent. This assumption then serves to shift the focus of the inquiry away from the secret corridors of intent to the more accessible realm of effect. Hence, courts must take special care that the plans they choose do not, unwittingly, have the impermissible effect of diluting minority voting strength.

The relationship between intent and effect is extremely problematic, as this dissent will seek to demonstrate. Indeed, careful analysis of the various holdings of the Supreme Court indicates that an exhaustive inquiry into effect provides crucial and revealing circumstantial evidence of intent. See infra, at pp. 976-979.

In this action, the court has chosen to implement a plan which substantially tracks the one adopted by the Texas legislature. Two portions of the plan have been altered. Districts 15 and 27 were objected to by the Attorney General as dilutive of Hispanic voting strength in South Texas. Those two districts have been redrawn, in a manner that remedies the defects of S.B. 1 with respect to those two districts. Additionally, districts 5 and 24, both in Dallas County, have been altered. These districts survived the scrutiny of the Attorney General. However, on the basis of analysis under the standards applicable to court-ordered apportionment plans, a majority of this court has found that districts 5 and 24, as drawn by S.B. 1, impermissibly dilute the ability of minority residents of Dallas County to participate in the electoral process. I wholly concur in that judgment.

The Attorney General's letter of objection notwithstanding, this court has the duty to conduct its own review of S.B. 1, to insure that its terms do not transgress the constitutional standards of the Fourteenth and Fifteenth Amendments. The majority *963 has decided to defer to the state legislature in devising its plan, incorporating the district boundaries as drawn by S.B. 1, as to all areas in dispute except Districts 15, 27, 5, and 24.

The source of this deference is puzzling. It is plainly held in Wise v. Lipscomb, supra, that "[a] more stringent standard is applied to judicial apportionments." 437 U.S. at 541, 98 S. Ct. at 2497. The plan imposed by the court today is, without question, a judicial apportionment. The decision to incorporate large components of S.B. 1 does not contravene that fact. Nor does the limited nature of the Attorney General's letter of objection vitiate the nature of the court's equitable remedy. Senate Bill No. 1 is a nullity as a result of the objection letter, and, in light of this fact, the court has been presented with the task of sculpting its own plan, at least pending later legislative action. There is no legal requirement that, in fashioning a court-ordered plan, the court must mirror the choices of the legislature. While the court may draw boundaries of districts in a manner that might be similar, or even congruent, to those drawn by the legislature, it may also choose to depart substantially from the handiwork of the legislature. The point is not that one or the other method of apportioning is incumbent upon the court; rather, the clear principle is that the court must act in a manner wholly free from any taint of arbitrariness or discrimination.

Further, even if S.B. 1 is to be accorded some presumption of validity with respect to the districts which survived the review of the Department of Justice, the court has an ineluctable responsibility to evaluate the apportionment plan under prevailing constitutional standards. Certainly the action of the Attorney General in entering a limited objection to the legislative plan does not abrogate the force of the Constitution as a constraint on the actions of the legislature, and the ability of federal courts to enforce its principles. A contrary result would patently violate all established principles of judicial review under Article III of the Constitution.

The standard of review of voting rights challenges includes a close analysis of the effects of a given apportionment. The Supreme Court's delineation of evidentiary sources for the indirect inquiry into intent includes "[t]he impact of the official action" as "an important starting point." Arlington Heights v. Metropolitan Housing Corporation, 429 U.S. 252, 266, 97 S. Ct. 555, 563, 50 L. Ed. 2d 450 (1977). Therefore, whether court-ordered plans are to be reviewed under a "more stringent standard", as seems plainly required by Wise v. Lipscomb, supra, or, alternatively, surveyed under prevailing constitutional standards for legislative plans (to the extent the court-order plan incorporates legislative boundaries), a close appraisal of the impact of the plan is obviously demanded.

Regardless of which analytical structure is to be used in the evaluation of the court-ordered plan, deference to legislative judgments has only limited validity. To the extent that the plan adopted by the legislature is constitutionally permissible, then the work of the legislature may be respected. However, the evidentiary factors recommended by Arlington Heights mandate that the procedural and substantive aspects of the apportionment process be scrupulously evaluated. This acute attention to detail must inform the judicial review of legislative plans; for when the court adopts large portions of the legislative plan as its own, on the basis of a deferential attitude of questionable validity, undeniably it must be acutely sensitive to any infirmities in the legislative process, as well as in the final product.

The plain thrust of Supreme Court decisions is that court-ordered plans should be evaluated more stringently than legislative plans. The contours of this close examination entail an exacting analysis of the impact of the plan, incorporating stringent proscriptions against fragmentation, dilution, packing, and other historic methods of depriving minorities of their constitutional right to participate in the process of democratic self-government on equal terms with all other persons. In drawing the boundaries *964 of the congressional districts in Harris County and West Texas, the court has deferred to the legislature. In my judgment, this premise of legislative validity is inappropriate. However, since I find that the legislative plan itself is constitutionally impermissible, even under the lower standard to be applied to legislative apportionments, I will not venture into the uncharted territory of determining the appropriate standard of review. Rather, this dissent will apply to the court-ordered plan the framework for analysis prescribed by the Supreme Court for legislative plans.[1] Even under this less exacting standard, the plan implemented by the court is constitutionally deficient with respect to Harris County and West Texas.

 
II. 
In Search of a Standard

The American ideal of participatory democracy, which undergirds our political and social order, depends on the irrefutable premise, embodied in the United States Constitution, that a citizen enjoys the right to participate in elections on an equal basis with other citizens. Dunn v. Blumstein, 405 U.S. 330, 336, 92 S. Ct. 995, 999, 31 L. Ed. 2d 274 (1972). In 1962, the Supreme Court established the jurisdictional basis upon which federal courts may review claims that this fundamental right has been unconstitutionally denied or abridged by the process of legislative reapportionment. Baker v. Carr, 369 U.S. 186, 82 S. Ct. 691, 7 L. Ed. 2d 663 (1962). Since that time, many federal courts have worked at rough-hewing the contours of the right to vote, and at determining when legislative action, in apportioning representation or in otherwise affecting participation in the electoral process, unconstitutionally offends that right. Constitutional theory in this area, however, lacks clear principles for the analytical assessment of legal claims. Rather, the doctrines which have evolved in the past two decades have about them the quality of landmarks on a barren terrain, by which courts may chart their course toward resolution of the complicated questions that arise in this area.

As the Supreme Court has noted, "[p]olitics and political considerations and inseparable from districting and apportionment." Gaffney v. Cummings, 412 U.S. 735, 753, 93 S. Ct. 2321, 2331, 37 L. Ed. 2d 298 (1973). Additionally, because of the myriad legal standards which have been developed, the legislative processes of reapportionment have become increasingly sophisticated and self-conscious. In some measure, this development is salutary; but to the extent that illicit purposes may infect the often ambiguous legislative procedures, this sophistication works to render judicial inquiry into the process exceedingly difficult. Yet, while these legislative actions present inescapable enigmas, federal courts are still charged with the crucial responsibility of scrutinizing such operations, to insure that they do not result in an unconstitutional denial or abridgement of the right to vote, since it is beyond question that the Constitution "nullifies sophisticated as well as simple-minded modes of discrimination." Lane v. Wilson, 307 U.S. 268, 275, 59 S. Ct. 872, 876, 83 L. Ed. 1281 (1939). See also Gomillion v. Lightfoot, 364 U.S. 339, 342, 81 S. Ct. 125, 127, 5 L. Ed. 2d 110 (1960).

In a variety of contexts, the Supreme Court has furnished guidance to lower courts on how to proceed with the complicated inquiry into the motivation which informs legislative reapportionment. Essentially, this guidance takes the form of delineation of a series of factors which courts are to look at in the process of evaluating the disputable question of purpose. Nowhere are these elements set forth in a *965 manner which purports to be exhaustive. See, e.g., Village of Arlington Heights v. Metropolitan Housing Development Corp., 429 U.S. 252, 268, 97 S. Ct. 555, 565, 50 L. Ed. 2d 450 (1977). On occasion, Supreme Court reviews of inquiries by district courts into legislative purpose are not explicitly tied to enumerated factors. Instead, they appear as implicit ratifications of common sense approaches by district courts. See, e.g., White v. Regester, 412 U.S. 755, 765-770, 93 S. Ct. 2332, 2339-2341, 37 L. Ed. 2d 314 (1973). In the absence of clear guidance for the required review, the body of Supreme Court decisions on challenges to apportionment must be searched for some central organizing principle which may be used in the judicial scrutiny of claims raised in a given action.

From the Court's simple rejection of the "uncouth twenty-eight sided figure" drawn for the City of Tuskegee by the Alabama State Legislature, Gomillion v. Lightfoot, 364 U.S. 339, 81 S. Ct. 125, 5 L. Ed. 2d 110 (1960), to the recent, and more subtle, complex analyses of electoral changes, City of Rome v. United States, 446 U.S. 156, 100 S. Ct. 1548, 64 L. Ed. 2d 119 (1980); City of Mobile v. Bolden, 446 U.S. 55, 100 S. Ct. 1490, 64 L. Ed. 2d 47 (1980), the hallmark of voting rights analysis has been its close attention to local detail. The cases do not always focus upon an identical list of particulars, nor do they always apply to the chosen factors a systematized scheme of analysis. However, whatever variations may exist in the approach of the Court to the claims under review, the inquiry always revolves around a close look at the history, demographics, procedure, and substance which form the backdrop of the particular legislative action.

Close examination of detail is not to be mistaken for what has come to be termed "strict scrutiny" within the parlance of Equal Protection analysis, though certainly there are similarities, which will be discussed later. It is, more truly, a form of intimate attention to particulars; i.e., a searching appraisal of all evidence, both direct and circumstantial, relating to the case at hand. This intense analysis is caused by the presence, in voting rights cases, of a number of concerns which, by their combined force, may call for traditional close scrutiny of the substantive and procedural details respecting the process in question.

First, an exacting inquiry is mandated by the presence, in these cases, of a fundamental interest protected by the Constitution. For nearly a century, the Supreme Court has recognized that "[t]he political franchise of voting is ... a fundamental political right, because preservative of all rights." Yick Wo v. Hopkins, 118 U.S. 356, 370, 6 S. Ct. 1064, 1071, 30 L. Ed. 2d 220 (1996). See also Wesberry v. Sanders, 376 U.S. 1, 17, 84 S. Ct. 526, 534, 11 L. Ed. 2d 481 (1964) ("no right is more previous in a free country"); Reynolds v. Sims, 377 U.S. 533, 561-62, 84 S. Ct. 1362, 1381, 12 L. Ed. 2d 506 (1964) ("right of suffrage is a fundamental matter in a free and democratic society"). Under well-established constitutional doctrine, a legislative action which "impinges upon a fundamental right explicitly or implicitly protected by the Constitution ... requir[es] strict judicial scrutiny." San Antonio School District v. Rodriguez, 411 U.S. 1, 17, 93 S. Ct. 1278, 1288, 36 L. Ed. 2d 16 (1973). Thus, the Court has applied orthodox strict scrutiny to claims that the right to vote is abridged by the use of poll taxes, Harper v. Virginia Board of Elections, 383 U.S. 663, 86 S. Ct. 1079, 16 L. Ed. 2d 169 (1966), property ownership requirements, Kramer v. Union Free School District No. 15, 395 U.S. 621, 89 S. Ct. 1886, 23 L. Ed. 2d 583 (1969), and residency requirements, Dunn v. Blumstein, 405 U.S. 330, 92 S. Ct. 995, 31 L. Ed. 2d 274 (1972). These cases demonstrate that when state action absolutely abridges the right to vote by depriving citizens victimized by the challenged classification of the opportunity to cast a ballot, the Court will scrupulously examine the action, to determine whether it is necessary to promote a compelling state interest. E.g., Kramer v. Union Free School Dist., 395 U.S. at 627, 89 S. Ct. at 1889.

The nature of the exacting and critical observation that has been brought to bear *966 in other types of voting rights claims is less clear. Starting with Reynolds v. Sims, 377 U.S. 533, 84 S. Ct. 1362, 12 L. Ed. 2d 506 (1964), the Supreme Court has reviewed numerous claims that legislative reapportionments passed by state legislatures abridge the right to vote, not by wholly abrogating that right, but by diminishing the effectiveness of the franchise. Clearly, these claims of vote dilution have been viewed with enormous interest, and have provoked a form of conscientious and pragmatic inquiry which, if not technically confined to the rigors of "strict scrutiny", certainly involves a thorough inspection of the claim and its factual underpinnings, to insure that the right to vote has not been effectively eliminated, even while it is superficially preserved. This principle is implicit in United States v. Saylor, 322 U.S. 385, 64 S. Ct. 1101, 88 L. Ed. 1341 (1944) (proscribing ballot-box stuffing), United States v. Classic, 313 U.S. 299, 61 S. Ct. 1031, 85 L. Ed. 1368 (1941) (outlawing alteration of ballots), and United States v. Mosley, 238 U.S. 383, 35 S. Ct. 904, 59 L. Ed. 1355 (1915) (requiring that votes, when cast, be counted). More importantly, it is the explicit foundation for the entry of federal courts into the "political thicket" of legislative reapportionment, effected by Reynolds v. Sims and its progeny. For in these cases, there is no claim that the aggrieved citizens are wholly denied the opportunity to participate in elections which affect them. The claims in these cases are, correctly speaking, generated by a practical reality: by virtue of malapportionment, the effectiveness of the franchise may be diminished in a manner that offends the Constitution. As the Court noted in Reynolds v. Sims, "the right to suffrage can be denied by a debasement or dilution of the weight of a citizen's vote just as effectively as by wholly prohibiting the free exercise of the franchise." 377 U.S. at 555, 84 S. Ct. at 1378. (Emphasis supplied.) The Supreme Court has thus explicitly recognized that the right to suffrage itself encompasses the right of effective participation in the electoral processes. In Dunn v. Blumstein, 405 U.S. 330, 336, 92 S. Ct. 995, 999, 31 L. Ed. 2d 274 (1972), this right was expressed as the "right to participate in elections on an equal basis with other citizens in the jurisdiction." In White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973), the Court held that, in order to prevail on a claim of vote dilution, plaintiffs must demonstrate "the political processes leading to nomination and election were not equally open to participation by the group in question". 412 U.S. at 766, 93 S. Ct. at 2339. Hence, in this context, the Court has made abundantly clear that the claims which are adjudicated in these cases arise under the Constitution, and merit the heightened attention to detail which is routinely triggered when fundamental rights are placed in jeopardy by the state action.

Second, the rationale for this close attention to detail is augmented by the presence, in virtually all contemporary voting rights cases, of a claim that the action of the state legislature involves some racial classification. As a simple historical fact, the intensity of the Court's concern with preservation of the right to vote has been brought about, in large measure, by the relentless attempts of some states to fence minorities out of the political process. Thus, vulgar racial gerrymanders such as the one involved in Gomillion v. Lightfoot, 364 U.S. 339, 81 S. Ct. 125, 5 L. Ed. 2d 110 (1960), are particularly suspect, both because the drawing of jurisdictional boundaries fundamentally affects the right to vote, and also because the lines drawn in that instance manifestly involved the type of racial classification proscribed by the Constitution.

It is, by now, axiomatic that, under the Equal Protection Clause of the Fourteenth Amendment, classifications explicitly based on race are "immediately suspect." Korematsu v. United States, 323 U.S. 214, 216, 65 S. Ct. 193, 194, 89 L. Ed. 194 (1944). See also Bolling v. Sharpe, 347 U.S. 497, 499, 74 S. Ct. 693, 694, 98 L. Ed. 884 (1954). Indeed, racial classifications which operate on the face of legislative enactments to deprive persons of some benefit solely on the basis of race are archetypal examples of instances in which the Court will apply strict scrutiny. This stringent standard of review is *967 applied to racial classifications, even if they do not affect an independently protected constitutional right.

This clear doctrine has been complicated by the evolution of "benign classifications", which, though rooted in race, are rendered less pernicious by virtue of their purportedly beneficent purpose. How such classifications are to be treated analytically is not yet clear. Compare Regents of University of California v. Bakke, 438 U.S. 265, 98 S. Ct. 2733, 57 L. Ed. 2d 750 (1978), with Fullilove v. Klutznick, 448 U.S. 448, 100 S. Ct. 2758, 65 L. Ed. 2d 902 (1980). Though no formal standard of review may be deduced from the pronouncements of the Supreme Court on the subject of benign racial classifications, certain trends may be discerned. The Equal Protection Clause does not forbid legislatures from ever taking cognizance of race, especially in matters in which issues of race and remedial relief for established prior discrimination are inextricably entwined with the substantive issue in dispute. Indeed, legislatures may be commanded, by statute or decisional law, to apprise themselves of the breadth of available options, and to insure that, in choosing among them, the vestiges of prior discrimination are eliminated, or at least minimized. E.g., United Jewish Organizations v. Carey, 430 U.S. 144, 97 S. Ct. 996, 51 L. Ed. 2d 229 (1977) (Voting Rights Act requires race consciousness); White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973) (history of discrimination in voting justifies requirement that state legislature affirmatively remedy existing denials of access to political process); see also Green v. County School Board, 391 U.S. 430, 88 S. Ct. 1689, 20 L. Ed. 2d 716 (1968) (continuing affirmative duty to eliminate vestiges of segregation in education).

On the other hand, it appears that, in most contexts, race-conscious action, even if generated by an avowedly "benign" purpose, will receive the most exacting perusal. In these circumstances, the asserted purpose of the legislative body will be closely examined, to insure that it is in some sense "compelling"; and the classification will be inspected to insure that its confines actually correspond to what is logically necessary to further the expressed design. Cf. Bakke, 438 U.S. at 287-291, 98 S. Ct. at 2746-2748, (Opinion of Powell, J.) 355-362, 98 S. Ct. 2781-2784 (Opinion of Brennan, J., et al.).

As noted, United Jewish Organizations v. Carey, 430 U.S. 144, 97 S. Ct. 996, 51 L. Ed. 2d 229 (1977), makes plain that, at least in the context of voting rights, classifications which involve explicit race consciousness are not inherently suspect. The clearly documented need for remedial action in this area, and the requirements of the Voting Rights Act, demand cognizance of race. 430 U.S. at 159-161, 97 S. Ct. at 1006-1007. See also Beer v. United States, 425 U.S. 130, 96 S. Ct. 1357, 47 L. Ed. 2d 629 (1976) (requiring legislatures to avoid retrogression in the position of racial minorities). However, the legislature's benign attention to race does not immunize the state action from legal challenge, either on the ground that the action worked to the deprivation of other groups, United Jewish Organizations v. Carey, supra, and Bakke, supra, or on the ground that the supposed beneficiaries of the action actually were harmed by it.

In Carey, a group of Hasidic Jews challenged a race-conscious reapportionment passed by the New York state legislature, on the ground that, in attempting to enhance the electoral strength of non-white voters, the legislature had diluted the value of the votes of Hasidic Jews. The Court rejected the claim, and found that race-conscious action benefiting an acknowledgedly disadvantaged class was permissible, if the action in truth advanced the interests of that class and was taken in an attempt to comply with established statutory requirements.

In analyzing the constitutional claims of vote dilution under the Fourteenth and Fifteenth Amendments, the Court, in Carey, acknowledged that explicit racial consideration had engendered the action of the state legislature. 430 U.S. at 151-52, 97 S. Ct. at 1002-03. However, the undisputed fact that the action of the reapportionment statute had been passed in an attempt to satisfy *968 the requirements of Section 5 of the Voting Rights Act cast the state's action in a wholly different light, and considerably dulled the typically exacting scrutiny which race consciousness ordinarily activates, even in cases of "benign classifications." Indeed, neither the rationale proffered by the state as justification for its action, nor the claims of the plaintiffs in Carey were given more than cursory attention. Hence, the analytical scheme by which Carey was decided bears great similarity to what is called in orthodox Equal Protection language "rational relation" analysis.

However, Carey offers no guidance for the analysis of claims brought by the purported beneficiaries of race conscious legislative action; e.g., had the non-white voters in whose behalf the New York State legislature had passed its reapportionment statute challenged the action on the ground that it unconstitutionally diluted their vote, the level of scrutiny such a claim would receive is unclear. Essentially such a legal action would involve the factual claim that, in actuality, the decision of the state legislature was in error, in that the action taken to benefit a particular class will actually work to that class' detriment. Such a case might involve an allegation that the overt and allegedly benign consideration of race conducted by the state legislature was, in reality, a sham, designed to cover up some hidden and forbidden purpose. More charitably, such a suit might claim that the legislature had committed a profound error of judgment, in assessing the impact its action would have on the groud toward which the solicitude of the state was supposedly directed. In either case, a minority protected from invidious classifications by the analytical shield of strict judicial scrutiny would assert that an explicit racial classification by a state legislature had unconstitutionally denied the class equal protection of the law. The question left unanswered by the body of Supreme Court decisions dealing with benign racial classifications, then, is what level of scrutiny will be applied to claims of this type.

Cases such as the one sub judice involve a confluence of two separate factors, either of which, standing independently, would warrant a close and searching review of the factual and legal components of the state action. First, a fundamental right here, the right to participate in elections on an equal basis with other citizens is directly affected by an action of the state. The substantive claim is that, by virtue of the state's action, the right to vote has been denied (see supra at p. 966), because the plaintiffs' electoral impact has been diluted. Second, the challenged action of the state legislature involves an explicit racial classification which allegedly deprives the suspect class of equal protection. The presence, in one action, of these two separate, though related, claims, each of which stands at the core of social order established and protected by the Constitution, is a matter of paramount importance, regardless of the formal analytical framework applied to the claim.

While plaintiffs' claims in this action should not, perforce, receive orthodox strict scrutiny, by reason of the presence of these two triggering considerations, nevertheless, the conjunction of these two factors appears to be without explicit precedent and, so, may be said to defy orthodoxy. The shifting tides of voting rights law in the past two decades have left in their wake a broad framework of analysis rather than a precise formula for review. The Court's most recent decision in this area, City of Mobile v. Bolden, 446 U.S. 55, 100 S. Ct. 1490, 64 L. Ed. 2d 47 (1980), offers the most guidance for assessment of the claims at issue, though the lack of a majority opinion in Bolden substantially undermines its clarity. However, before attempting to extricate from Bolden the proper structure for review of the claims in this action, it is crucial to note that the state action challenged in Bolden did not involve an explicit racial classification. Instead, the claim was that the facially neutral statutory provisions mandating an at-large system of municipal elections, as employed by the City of Mobile, violated the constitutional and statutory rights of black residents of the city. 446 U.S. at 58, 100 S. Ct. at 1495. Since the *969 state and local laws of Alabama and Mobile which established the at-large Commission form of government for the city did not contain explicit racial classifications, the challenge to these laws was treated in accordance with established doctrines of Equal Protection law pertaining to challenges to ostensibly unbiased laws. Within this framework, "[p]roof of racially discriminatory intent or purpose is required to show a violation of the Equal Protection Clause." 446 U.S. at 67, 100 S. Ct. at 1500, quoting Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U.S. 252, 97 S. Ct. 555, 50 L. Ed. 2d 450 (1977).

There was no evidence in Bolden, nor did plaintiffs claim, that the establishment of at-large elections in Mobile involved explicit racial classifications. In Bolden, the plaintiffs relied on a claim of vote dilution under the Fifteenth Amendment and a claim of denial of equal protection, by a facially neutral enactment, under the Fourteenth Amendment. The plurality opinion clearly indicates that the latter claim was treated specifically within the mode of analysis established for challenges to seemingly impartial enactments. 446 U.S. at 66, 100 S. Ct. at 1499. See Washington v. Davis, 426 U.S. 229, 96 S. Ct. 2040, 48 L. Ed. 2d 597 (1976). Additionally, the opinion interpreted the body of Fifteenth Amendment case law as standing for the proposition that claims of vote dilution under that Amendment must meet these standards as well. 446 U.S. at 63, 100 S. Ct. at 1497. Thus, the guidance furnished by City of Mobile v. Bolden, though valuable, is not ultimately definitive, for that case did not involve the two activating strands of "strict scrutiny" claims involved in the action at issue. The various opinions leave wholly open the question of what effect the additional presence of a racially conscious classification would have on the standard of review accorded claims which are brought under the Fourteenth and Fifteenth Amendments. In this dissent, it is assumed that this additional presence does not in any way vitiate the intent requirement. In keeping with the pattern of Supreme Court decisions in this area, the race consciousness of the legislature will serve only to intensify the appraisal of the design and effect of the reapportionment plan.

In the wake of Bolden, certain things are manifest. Even assuming that race-consciousness on the part of the state legislature would not alter the proper approach to claims of this type, the opinions which form the majority in Bolden incorporate large portions of the antecedent Supreme Court decisions relating to claims of vote dilution. The plurality opinion contains a highly self-conscious analysis of the precedents, and it places the Bolden ruling firmly within the evolution of that doctrinal structure. 446 U.S. at 61-66, 100 S. Ct. at 1496-1499. See also 466 U.S. at 83-87, 100 S. Ct. at 1508-1510 (Stevens, J., concurring in judgment). Therefore, the central thrust of Bolden the application of the intent requirement to claims of voting dilution under the Fourteenth and Fifteenth Amendments simply makes explicit the burden of proof under which previous voting rights claims had been resolved.[2] Given the highly contextual nature of voting rights claims, as well as the detailed process by which they are to be resolved, the broad doctrinal categorization which is central to City of Mobile v. Bolden is ultimately of less value to lower courts faced with claims of vote dilution than are guidelines, either explicit or implicit, as to how properly to analyze the specific factual and legal claims raised by an action.

As noted, the Supreme Court has, on occasion, offered specific, explicit criteria for the assessment of these claims. At other times, the standards must be gleaned from a close reading of the evidentiary review conducted by the Court. In the course of *970 applying the intent requirement to claims of vote dilution, the plurality opinion in Bolden explicitly incorporated the evidentiary prototype for proving intent, first set forth in Village of Arlington Heights v. Metropolitan Development Corp., 429 U.S. 252, 266-68, 97 S. Ct. 555, 563-565, 50 L. Ed. 2d 450 (1977). Bolden, 446 U.S. at 70, 100 S. Ct. at 1501. The Arlington Heights factors will be discussed at length later in this dissent. For now, it should be noted that the identification of relevant factors in Arlington Heights did not "purpor[t] to be exhaustive." 429 U.S. at 268, 97 S. Ct. at 565. Therefore, while those factors offer a sense of the nature of inquiry, they should not be taken as establishing boundaries of review. Indeed, in light of the affirmation of previous voting rights cases involving complicated factual inquiries, Bolden may fairly be seen as offering a means of augmenting the Arlington Heights inquiry. Most specifically, the unequivocal endorsement by the Bolden court of the Court's decision in White v. Regester suggests that the meticulous inspection of evidentiary factors conducted by the district court in that case, Graves v. Barnes, 343 F. Supp. 704 (W.D.Tex.1972), is precisely what is required in the search for legislative intent. In sum, claims of vote dilution brought under the Fourteenth and Fifteenth Amendment must establish that the allegedly discriminatory action was the product of an invidious purpose and an investigation of that purpose involves an inquiry which must proceed along a path demarcated by Arlington Heights, and further clarified by previous voting rights cases such as White v. Regester.

To the extent that the criteria adopted in White v. Regester broaden and deepen the inquiry, this extension may be accounted for by the presence, in that case and other vote dilution cases, of the confluence of the two precipitating elements mentioned above fundamental rights and race consciousness. This is not to suggest that the "intensely local appraisal" commended by the Court in White v. Regester, 412 U.S. at 769, 93 S. Ct. at 2341, in any way traduces the spirit of Arlington Heights. Indeed, the latter case itself recognizes that scrutiny of intent "demands a sensitive inquiry into such circumstantial and direct evidence ... as may be available." 429 U.S. at 266, 97 S. Ct. at 563. The type of evidentiary investigation required under White v. Regester is wholly consonant with this recognition. However, it is crucial to note that claims of vote dilution brought against state actions which, as in the case under consideration, are concededly race conscious, differ in a crucial respect from the claim at stake in Arlington Heights, by virtue of the combined presence of fundamental rights and overt race consciousness. To the extent, then, that this additional burdening of the claim warrants even closer attention within the established canons of Constitutional law, it is entirely appropriate that the focus of the court should be sharpened by the additional guidance afforded by the White factors.

The necessity for the "intensely local appraisal of the design and impact", White v. Regester, 412 U.S. at 769, 93 S. Ct. at 2341, of legislative redistricting schemes is further supported by additional elements which are present in cases such as the one at bar. First, by definition, the redistricting process is highly political. These political considerations are "inseparable" from the process; indeed, in many ways they provide the practical and semantic framework for the entire process. As observed, the Supreme Court, in Baker v. Carr, 369 U.S. 186, 82 S. Ct. 691, 7 L. Ed. 2d 663 (1962), held that the presence of these "political questions" could not serve to shield the apportionment process from judicial inquiry. However, beyond that recognition, it is clear that the admixture of constitutional and political issues which shrouds the apportionment process fundamentally affects the manner in which courts must investigate charges of vote dilution.

As even a cursory review of the record in this case will demonstrate, legislative debate on redistricting provides a forum in which elected representatives relentlessly bring to the fore the full range of issues which affect their constituencies. (The *971 multitudinous variety of these concernments is demonstrated by the vast body of case law involving claims of voting rights, and by the justifications proffered by the state as supporting the scheme under attack, as well.) Certain of the matters typically advanced in the course of legislative debate on redistricting have been explicitly approved as relevant by courts reviewing the process. Other interests have been expressly found to be inappreciable. Still others stand in some intermediate zone, in that they may legitimately be considered, but may not be treated as dispositive. But while the relevance of certain of these concernments has been adjudged, the legislative record is replete with references to other matters, all undeniably inherent in localized circumstances, which may severally stand at various levels on the gauge of relevance. The subordinate details of these concernments must be accorded close examination, and their factual accuracy and legal validity must be assessed. Deference to the legislature's choice, whether as to identification or relevant issues or regarding their final resolution, is inappropriate. The entire inquiry must be searching, and rooted in local reality, political and otherwise. Cf. White v. Regester, 412 U.S. at 770, 93 S. Ct. at 2341.

Second, as mentioned, the issues of voting rights, voting dilution, and all its corollary claims, have been subjected to elaborate judicial inquiry over a period of decades. As a matter of law, state legislatures are presumed to be on notice of these decisions. More importantly, as a matter of practical fact, state legislatures have, as a result of prolonged and repeated exposure to litigation, become thoroughly aware of the legal consequences of their actions in the realm of reapportionment. Additionally, the frequent pronouncements of federal courts in this area have presented detailed, if not always lucid, standards by which their actions will be judged and what factors may legitimately motivate their decisions. Surely the intent of the Constitutional proscription of denial of the right to vote, and vigilant application of that provision by federal courts, has been to eliminate the depressing legacy of electoral discrimination in this nation. However, the very volume of meritorious voting rights litigation in the past twenty years which appears, in fact; to be increasing stands as a testament to the guile of those in power, as well as to their intransigence. Yet, as already alluded to, it is well-established that the Constitution "nullifies sophisticated as well as simple-minded modes of discrimination." Lane v. Wilson, 307 U.S. 268, 275, 59 S. Ct. 872, 876, 83 L. Ed. 1281 (1939). Specifically, the common awareness that explicit racist classifications will not survive judicial inspection may safely be assumed to have the effect of driving illicit motives into some darkened sanctuary of hidden purpose, from which they will emerge in the guise of some legitimate concernment. This simple, intuitive axiom stands behind the acknowledgment by the Court in Arlington Heights that the inquiry into intent will involve a survey of circumstantial as well as direct evidence. The Fifth Circuit has stated that "[i]n a voting dilution case in which the challenged system was created at a time when discrimination may or may not have been its purpose, it is unlikely that plaintiffs could ever uncover direct proof that such a system was being maintained for the purpose of discrimination." Lodge v. Buxton, 639 F.2d 1358, 1363 (5th Cir. 1981).

Third, the fact of increasing sophistication involves a correlative problem in the assessment of legislative motivation. In the case at hand, the state constructed an elaborate apparatus for the drafting of reapportionment legislation. Indeed, those who spoke for the state pointed to this mechanism with pride throughout this action. Computerized demographics of great detail were readily available, to supplement the common knowledge of the distribution of racial minorities across the state. The entire process is freighted with a keen awareness of race. Yet this combination of facts may not deflect the inquiry into motive. Indeed, to the same extent that the elaborate technology and racial data may, as the state claims, have facilitated benign action of the state legislature, so, too, it *972 may have served as an ornate barricade, to deflect scrutiny from some fundamentally invidious purpose. Common sense dictates that courts not allow the imposing precision of technology to frustrate the inquiry into the legislative process. Nor should the mere fact that the state openly acknowledges the racial component of its decision, but professes benign purpose, somehow diminish the intensity of the scrutiny given claims brought by minorities alleging that, in reality, their interests were undermined.

In conclusion, claims such as those brought by the various plaintiffs in this action must be accorded a searching and detailed analysis of their factual and legal bases. First, the state action challenged affects the ability of citizens to exercise the fundamental right of participation in the electoral process. Second, the classification involved in the state action was made with a high degree of awareness of its racial impact. Third, by their nature, voting rights claims involve highly politicized decisions. Fourth, the legislative decisions in this area are made on the basis of a plethora of localized concernments, with an eye to the effect the decision will have on specific classes of citizens. Fifth, the history of voting rights litigation has created a situation in which all actors are acutely self-conscious, and highly aware of the legal standards under which their actions will be scrutinized.

In light of this complex of factors, rigid adherence to orthodox standards of equal protection analysis is inappropriate. It is beyond doubt that, as a matter of ultimate resolution, the standards advanced in City of Mobile v. Bolden must be met, in order for plaintiffs to prevail on their claims. However, the full body of Supreme Court decisions stands in direct contrast to the assertion that claims such as the one sub judice may be reduced to simple equal protection claims, and treated in a formulaic manner. Nor does Bolden itself require this sterile and diffident treatment of claims which lie so close to the heart of our constitutional order. Indeed, as noted, Bolden contains two related suggestions concerning how vote dilution claims are to be analyzed. First, Bolden incorporates the criteria set forth in Arlington Heights. Second, Bolden affirms White v. Regester, and the use of other factors in the course of the inquiry into intent. The plurality opinion in Bolden places White v. Regester in a line of cases holding the plaintiffs in voting rights action must establish invidious intent. Given the prevalence of plaintiffs in White v. Regester on their claims of vote dilution, the evidentiary review conducted by the Court in that case, a fortiori, stands as a sufficient inquiry into the intent of the legislature. Accordingly, the factors analyzed in White v. Regester may, perhaps must, be used to discern the intent of the Texas state legislature in passing Senate Bill No. 1.

To recapitulate, plaintiffs' claims in this action demand an intensely local appraisal of the design and impact of the Congressional reapportionment bill passed by the Texas state legislature. This appraisal must blend an understanding of the historical background of the decision, with close examination of present reality, political and otherwise, as it affected, and in turn will be affected by, Senate Bill No. 1.

 
III. 
Proving Discriminatory Intent

The analysis of legislative purpose is inclusive of the following factors:

 
1. Historical Background of the Decision.

Scrutiny of this factor is explicitly approved in Arlington Heights, 429 U.S. at 267, 97 S. Ct. at 564. Additionally, in White v. Regester, the Court, at three separate places, sanctioned the careful historical approach taken by the lower court in that case. 412 U.S. at 766, 767, 769, 93 S. Ct. at 2339, 2340, 2341. See Graves v. Barnes (Graves I), 343 F. Supp. 704 (W.D.Tex.1972). Of course, state actions which provoke charges of racially motivated discrimination rarely arise in an historical vacuum, in which an allegedly discriminatory action is without precedent or which is not rooted in *973 prior discrimination. The long and tragic course of litigation affecting race stands as an ample testament to the virtually inextricable enmeshment of the legacy of segregation and legally mandated discrimination with current patterns of inequality and enduring deprivation.

In the context of school desegregation, the Supreme Court has acknowledged that the legal assessment of claims of racial discrimination, and any subsequent remedial measures, must be conducted in light of established history. E.g., Green v. County School Board, 391 U.S. 430, 88 S. Ct. 1689, 20 L. Ed. 2d 716 (1968); Keyes v. School District No. 1, Denver, Colo., 413 U.S. 189, 93 S. Ct. 2686, 37 L. Ed. 2d 548 (1973). (The analogy between school desegregation and voting rights cases will be developed later. See pp. 973, 974, 976, infra.) The obligatory requirement that the starting point of legal analysis be a scrupulous consideration of historical circumstances is built on the common-sense recognition that present social patterns educational, demographic, and economic and the legislative responses to them, are the product of experience and changes in attitudes of the people, as they occur or develop through time.

In no area of jurisprudence is this diachronic process more justified than in the law pertaining to voting rights. The history of exclusion of minorities from the democratic process is so well-documented as to be beyond question. Indeed, constitutional history since the Civil War may be seen as a "movement toward democracy for more than a few." City of Mobile v. Bolden, 446 U.S. at 104, 100 S. Ct. at 1520 (Marshall, J., dissenting); United States Constitution, Amendments 15, 17, 19, 23, 24, and 26. Specifically, the Fifteenth Amendment placed in the text of the Constitution the nation's firm resolve to eliminate the discriminatory structure of democratic self-governance. This Amendment has met with "unremitting and ingenious defiance" which has served to maintain the "insidious and pervasive evil" of racial discrimination in voting. South Carolina v. Katzenbach, 383 U.S. 301, 309, 86 S. Ct. 803, 808, 15 L. Ed. 2d 769 (1966). The inventiveness of legislatures which, rather than comply with the clear command of the law, choose instead to devise clever means of circumventing it, has led the Supreme Court, in its scrutiny of apportionment plans, to insure that a given plan does not "perpetuate a `ghost of prior malapportionment.'" Burns v. Richardson, 384 U.S. 73, 93, 86 S. Ct. 1286, 1297, 16 L. Ed. 2d 376 (1966), quoting Buckley v. Hoff, 243 F. Supp. 873, 876 (D.C.Vt. 1965). Though delivered in the contextual relation of a challenge to actual numerical apportionment, this caution, in substance, speaks to the practical reality that a legislature which in the past has acted in a purposefully discriminatory manner may do so again, albeit in a more subtle form.

Of course, even a clear demonstration of historical discrimination does not prove present intent. Indeed, such a manifestation would not even rise to the level of a formal presumption of purpose. The law does not contemplate any theory of original sin, under which a state, once guilty of legally mandated racial discrimination, may never purge itself of such sinister motivations. However, as the de jure/de facto distinction in school desegregation law makes clear, proof of a prior discriminatory legal system may drastically alter the manner in which formal legal obligations are applied to particular units of government.

In the context of school desegregation law, the objective of the entire line of cases descendent from Brown v. Board of Education is "to eliminate from the public schools all vestiges of state-imposed segregation." Swann v. Charlotte-Mecklenburg Board of Education, 402 U.S. 1, 15, 91 S. Ct. 1267, 1275, 28 L. Ed. 2d 554 (1971). In light of this objective, school boards which, in the past, had operated "state-compelled dual systems were ... charged with the affirmative duty to take whatever steps might be necessary to convert to a unitary system in which racial discrimination would be eliminated root and branch." Green v. County School Board, 391 U.S. 430, 437-38, 88 S. Ct. 1689, 1693-94, 20 L. Ed. 2d 716 (1968). Even if at the time the actual legal challenge to school assignment plans is brought, the governmental *974 unit does not maintain a system of racial segregation, any contemporary assignment plan is held, in effect, to be tainted by the legacy of prior segregation. The historical background, then, is much more than a valid point of departure for judicial inquiry. In school desegregation cases, the historical context stands as the necessary predicate to any subsequent action, by the school board in the first instance, and, later, by the court when faced with a challenge to a system of public education.

The Court of Appeals for the Fifth Circuit has suggested that an analogous doctrinal structure is applicable to voting rights cases where the challenge is brought against a jurisdiction which, in the past, has maintained a system of enfranchisement or apportionment that denied minorities effective access to the electoral process. In Kirksey v. Board of Supervisors of Hinds County, Miss., 554 F.2d 139 (5th Cir. 1977) (en banc), cert. denied, 434 U.S. 968, 98 S. Ct. 512, 54 L. Ed. 2d 454 (1977), the Fifth Circuit, sitting en banc, reviewed a challenge to a county apportionment plan on the ground, inter alia, that the plan unconstitutionally diluted the voting strength of black citizens in the county. The district court found that the plan was racially neutral, and so constitutional. Kirksey v. Board of Supervisors, 402 F. Supp. 658 (S.D. Miss.1975). The initial panel of the Court of Appeals affirmed. 528 F.2d 536 (5th Cir. 1976). Upon rehearing, en banc, the court reversed the trial court finding.

As a foundation for its review, the Court of Appeals noted that the plaintiffs were required to demonstrate an invidious purpose in order to prevail on their claims. Further, it pointed out that the lower court had engaged in the required inquiry into intent. However, because this inquiry conducted by the trial court was unduly truncated, its finding of racial neutrality was incorrect. The court stated:

 
The court found that the motives were neutral with respect to drawing of the plan .... In the narrow context of the drawing of the plan, this was not plainly erroneous, but it was too narrow. The court did not address whether the plan perpetuated existent purposeful and intentional discriminatory denial of access, because it had erroneously concluded that the past denial of access had attenuated and that no present denial existed.

554 F.2d at 146.

Thus, the Court of Appeals found that the lower court had erred fundamentally, in failing to consider the present plan in light of prior actions by the governing body. In a situation in which the state has, in the past, overtly and consciously discriminated on the basis of race, later actions by the state must be closely scrutinized to insure that they do not bear the stamp of this prior discrimination, even if the actions are, nominally, racially neutral. The court noted that

 
[w]here a plan, though itself racially neutral, carries forward intentional and purposeful discriminatory denial of access that is already in effect, it is not constitutional. Its benign nature cannot insulate the redistricting government entity from the existing taint. If a neutral plan were permitted to have this effect, minorities presently denied access to political life for unconstitutional reasons could be walled off from relief against continuation of that denial. The redistricting body would only need to adopt a racially benign plan that permitted the record of the past to continue unabated. Such a rule would sub silentio overrule White v. Regester.

554 F.2d at 146-147. The analytical approach relied upon by the Court of Appeals in this instance is entirely consistent with, and in fact is founded on, an inquiry into legislative motivation. In essence, this manner of review merely stands for the proposition that a prior history of discrimination undermines the prima facie validity of a facially neutral apportionment plan. The court in Kirksey stated, with great force, that nothing in the Supreme Court cases which establish and refine the intent requirement can be taken to hold that "where purposeful and intentional discrimination already exists, it can be constitutionally perpetuated into the future by neutral *975 official action." Id. at 148. Stated in the obverse, then, a redistricting plan is constitutionally impermissible as racially discriminatory, if it perpetuates an existent denial of access by the racial minority to the political process. Id. at 142. See also Robinson v. Commissioner's Court, 505 F.2d 674, 677 (5th Cir. 1974); Moore v. Leflore County Board of Election Comm., 502 F.2d 621, 623, 624 (5th Cir. 1974).

This summary of contemporary voting rights law is wholly consonant with, and in fact, relies for its validity, on the Supreme Court's holding in White v. Regester. The central doctrine of that case was that, in order to prevail on a claim of vote dilution, plaintiffs must

 
produce evidence to support a finding that the political processes leading to nomination and election were not equally open to participation by the group in question that its members had less opportunity to participate in the political processes and to elect legislators of their choice.

412 U.S. at 766, 93 S. Ct. at 2339, 29 L. Ed. 2d 363 citing Whitcomb v. Chavis, 403 U.S. 124, 149-50, 91 S. Ct. 1858, 1872, 29 L. Ed. 2d 363 (1971). In holding that the plaintiffs in White v. Regester had met this burden, the court relied upon evidence that demonstrated a long history of official discrimination against minorities, and concomitant indifference to their needs, on the part of white officials elected under the auspices of a biased and undemocratic electoral system. 412 U.S. at 766-769, 93 S.Ct. at 2339-2341; see also City of Mobile v. Bolden, 446 U.S. at 69, 100 S. Ct. at 1501.

There was no suggestion in White v. Regester that the most recent re-enactment of multi-member districts in Texas was, when viewed in isolation, the product of an invidiously discriminatory purpose. Rather, in White, it was found that the districting system was fatally infected by the history of discrimination, and that the existent apportionment of representation perpetuated a system in which the political processes leading to nomination and election were not equally open to minorities. As the Supreme Court has made clear in City of Mobile, this holding was squarely within the line of cases which require plaintiffs to demonstrate invidious intent. 446 U.S. at 68-69, 100 S. Ct. at 1500-1501. White v. Regester, and its affirmation in Bolden, stand for the proposition advanced by the Fifth Circuit in Kirksey, that "[w]here a plan, though itself facially neutral, carries forward intentional and purposeful discriminatory denial of access that is already in effect, it is not constitutional." 554 F.2d at 146.

The terms of the analogy to school desegregation cases may now be stated with more precision. In cases such as Green v. County School Board, 391 U.S. 430, 88 S. Ct. 1689, 20 L. Ed. 2d 716 (1968), the Court held that school districts which have intentionally maintained racially segregated school districts are under an affirmative duty to dismantle the dual system. Plans adopted by school districts in an attempt to accomplish that objective must be shown to do so in fact, the beneficence of the school board notwithstanding. In Green, the school board rejected the "freedom-of-choice" adopted pursuant to court order, because it did not meet the obligation incumbent upon the district to eliminate segregation root and branch. The plan adopted by the school board was not per se unconstitutional, nor, when viewed in isolation, could invidious intent be inferred from its adoption. However, it was held unconstitutional in light of the history of educational discrimination in the district, and the fact that the plan perpetuated the effects of that discriminatory legacy.

In correlative terms, White v. Regester may be seen as holding, in essence, that when a jurisdiction which, in the past, has practiced de jure discrimination in voting adopts a reapportionment plan, the plan must remedy existing denials of access or it shall be deemed to be intentionally discriminatory. See Kirksey v. Board of Supervisors, Hinds County, Miss., 554 F.2d 139, 148 n. 16; see also Sims v. Amos, 365 F. Supp. 215, 220 n. 2 (M.D.Ala.1973) (three-judge court). This analogy is pursued, not to suggest that the prevailing structure of voting *976 rights law creates on states an affirmative obligation to remedy existing denials of access, as a formal presumptive matter, but as a highly instructive comparison, which is relevant in attempting to clarify the process of blending historical and contemporary evidence in the perscrutation of intent. First, it should be noted that the underlying justification for the elevation of equal opportunity to the forefront of the concerns of the Supreme Court was the effect education has on the ability of citizens to participate in the process of self-government. Brown v. Board of Education, 347 U.S. 483, 74 S. Ct. 686, 98 L. Ed. 873 (1954). Hence, the affirmative duty which is the cornerstone of the remedial enforcement of the proscription against segregated schools has been implemented in pursuit of the ideal of participatory democracy. Second, the Court has held that the right to education is not a fundamental right, and so does not warrant the rigorous protection of strict scrutiny. San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 93 S. Ct. 1278, 36 L. Ed. 2d 16 (1973).

Viewed in combination, these two facts suggest that the importation of "affirmative duty" analysis into the area of voting rights law, at least as an evidentiary matter, is entirely appropriate. Surely, it cannot be suggested that the Supreme Court is more vigilant in the protection of non-fundamental rights than those which are deemed fundamental. Nor does logic suggest that fundamental rights are best preserved at a secondary, rather than primary stage. The relationship between education and voting is reciprocal and dynamic, and both components of the equation are worthy of the close attention which is mandated by the holdings of the Supreme Court. These holdings command that all lingering vestiges of legally enshrined segregation be eliminated, so that the legacy of racial discrimination no longer poisons our contemporary systems of education and democracy. As a practical matter, when confronted with claims of the kind presented here, courts must appraise the legislative action, to insure not only that the instant action is free of invidious purpose when viewed in isolation, but also to make certain that, though neutral on its face, the action does not perpetuate the unlawful effects of prior purposeful discrimination. This appraisal is precisely the kind that was conducted in White v. Regester.

 
2. The racially discriminatory impact of the action

In Arlington Heights, the Court noted that "[t]he impact of the official action ... may provide an important starting point" for the "sensitive inquiry" into intent. 429 U.S. at 266, 97 S. Ct. 563. The realistic assessment of the effect of a particular legislative action is merely a logical underpinning for an inquiry that is intensely practical. As the court stated in Personnel Administrator v. Feeney, 442 U.S. 256, 279 n. 24, 99 S. Ct. 2282, 2296 n. 24, 60 L. Ed. 2d 870 "[w]hat a legislature, or any official entity is `up to' may be plain from the results its actions achieve, or the results they avoid."

"Racially discriminatory impact" operates at two different stages, in the process of evaluating claims of this nature. Though these phases may be, as a technical matter, analytically distinct, they are intrinsically related. First, for a claim to be cognizable under the Equal Protection Clause, the challenged state action must, in fact, deny equal protection of the laws. For example, in Washington v. Davis, 426 U.S. 229, 96 S. Ct. 2040, 48 L. Ed. 2d 597 (1976), had the plaintiffs failed to establish that the test used by the District of Columbia Police Department had a racially discriminatory impact, the claim would have failed, essentially for lack of a prima facie case of race discrimination. That is, actual discriminatory effect is a threshold matter for a claim under the Equal Protection Clause. Second, if that threshhold is crossed, and plaintiffs establish disproportionate impact, a closer examination of the actual impact will be conducted, as an evidentiary source in the intent inquiry. Arlington Heights, supra.

The justification for evaluation of impact at this second level of inquiry lies in the *977 application, to this body of law, of the legal principle that "a person may be presumed to intend the logical consequences of her actions." This maxim has been a cornerstone of the Anglo-American legal system for centuries. See, e.g., Townsend v. Wathen, 103 Eng.Rep. 579, 580-81 (K.B.1808). It has been applied in a broad variety of contexts in civil and criminal law, and stands as an indispensible component of our system of jurisprudence.

Its application in the context of criminal law is enlightening. In a long line of cases, the Supreme Court has upheld statutory and common law presumptions which allow juries to convict criminal defendants on the basis of an inference of intent, or mens rea. See, e.g., Ulster County Court v. Allen, 442 U.S. 140, 99 S. Ct. 2213, 60 L. Ed. 2d 777 (1979); Leary v. United States, 395 U.S. 6, 89 S. Ct. 1532, 23 L. Ed. 2d 57 (1969). In effect, this presumption allows juries to examine what was actually done, and also the effect it had. From this analysis, the jury is permitted to make the inferential leap that, if the result flowed naturally from the action, the actor intended the consequences. The logical connection upon which this inference is based has been held to be sufficiently strong to convict a person under the "reasonable doubt" standard. (This is not to say that the presumption itself must be valid beyond a reasonable doubt. See Ulster County Court, supra, 442 U.S. at 163-167, 99 S. Ct. at 2227-2229.) Thus, it should be noted that the inference of intent from effect has been upheld in a context in which the party relying on the inference (the State) is held to the most stringent standard of proof, and the party against whom the presumption is applied (the defendant) is, as a matter of constitutional law, entitled to a presumption of innocence, which may only be overcome by proof, beyond a reasonable doubt, of all necessary elements of the offense charged.

Arlington Heights does not suggest that the relationship between effect and intent rises to the level of a formal mandatory presumption. See, 429 U.S. at 266, 97 S. Ct. 563. And in Feeney, the Court made clear that a demonstration of foreseeable impact, may stand only as a strong inference of intent, "a working tool [which is] not a synonym for proof." 442 U.S. at 279 n. 25, 99 S. Ct. at 2296 n. 25. Yet the Court's discussion in Feeney serves to emphasize the strength of the permissive presumption. In Feeney, the inference of intent from the clear demonstration of foreseeable, indeed, inevitable impact made in that case "fail[ed] to ripen into proof" because the state was able to establish that "the impact [was] essentially an unavoidable consequence of a legislative policy that has in itself always been deemed to be legitimate, and when, as here, the statutory history and all of the available evidence affirmatively demonstrate the opposite [of the inference]." Id. (emphasis supplied).

What emerges from the discussion of these principles is a type of escalating inferential standard. The strength of the inference to be made from established discriminatory impact will correspond to the presence or absence of certain crucial variables: the severity of the impact, the degree of foreseeability, the validity of alternative explanations for the state action, and the credibility of the legislative record as support for the final action. These factors, in turn, incorporate a variety of subsidiary inquiries, notably, the extent to which the legislative record reveals that the legislature was actually apprised of the disproportionate impact of the contemplated action, the responses to these warnings which were elicited, and the validity of state policies offered in support of a final action which had a foreseeable and foreseen effect.

Additionally, the foregoing discussion of the existence of an "affirmative duty" to remedy existing denials of access to the political process has a degree of application to this phase of the inquiry as well. Prior findings of historical and actual discrimination in voting may create on state legislatures, as a category of legal obligation, a duty to eliminate from future plans any vestige of racial discrimination. (See supra, at pp. 973-974.) Moreover, the fact that a state has been found to have maintained discriminatory voting practices in *978 the past assuredly stands as notice to the legislature that racial considerations must be at the forefront of the entire apportionment process. Thus, the prevalence of racial concerns in the process of apportioning congressional representation justifies the conclusion that the legislature was acutely aware of the impact its actions would have on the ability of minorities effectively to exercise the franchise. Hence, the inference of intent from impact is strengthened by the practical reality that, in this instance, consideration of foreseeability is not speculative, nor based on objective assumptions. Instead, it is rooted in a clear legislative record which demonstrates that the consequences of Senate Bill No. 1 were actually foreseen, and were actually contemplated as one of the two fundamental concerns of the entire process. See, e.g., DX No. 52, at 11, 15 (July 16, 1981); DX 4.4.1 at 2-4; DX 4.4.4; Deposition of Lynn Moak, at 150; Deposition of Kenneth Shepardson at 13; Deposition of Peyton McKnight at 11.

These first two elements of the inquiry historical background and actual impact have been discussed at great length. This elaborate consideration of these two factors is generated by an evaluation of the combined import of the Supreme Court's decisions in White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973) and City of Mobile v. Bolden, 446 U.S. 55, 100 S. Ct. 1490, 64 L. Ed. 2d 47 (1980).

As the Court noted in Bolden, White v. Regester is the only case in which the Court has upheld a constitutional challenge to multi-member districts, on the basis that the apportionment entailed by those districts unconstitutionally diluted the voting strength of a discrete group. Bolden, supra, 446 U.S. at 68, 100 S. Ct. at 1500. In prevailing in White v. Regester, plaintiffs demonstrated that "the political processes leading to nomination and election were not equally open to participation by the groups in question that its members had less opportunity than did other residents in the district to participate in the political processes and to elect legislators of their choice." 412 U.S. at 766, 93 S. Ct. at 2339. In Bolden, the Court affirmed this holding, and found it entirely consistent with the evidentiary standards required by the Fourteenth and Fifteenth Amendments for claims of vote dilution. 446 U.S. at 69, 100 S. Ct. at 1501.

Common sense dictates that, when the formal and theoretical explanations of standards of proof are insufficiently precise to allow useful exegesis and application to actual cases, attention may profitably be focused on other actual cases in which plaintiffs have prevailed under the given standard. Hence, in light of the ambiguity of the evidentiary review recommended by Arlington Heights, White v. Regester must be examined carefully to determine what factual showing has gained the imprimatur of the Supreme Court.

In Graves v. Barnes (Graves I), 343 F. Supp. 704 (W.D.Tex.1972) (three-judge court), the district court conducted a thorough, elaborate, and searching examination of the history of racial discrimination in Texas in relation to contemporary electoral practices. Additionally, the court reviewed, in great detail, the practical effect of the maintenance of multi-member districts in Bexar and Dallas counties. On the basis of this review, the district court concluded that the system of multi-member representation in Dallas and Bexar County unconstitutionally abridged the rights of blacks and Hispanics to vote and participate in the electoral process.

These findings were affirmed in White v. Regester. The Court unanimously approved the "blend of history and [the] intensely local appraisal of the design and impact" of the apportionments in Dallas and Bexar counties. 412 U.S. at 767, 93 S. Ct. at 2340 (emphasis supplied). As made clear in Bolden, the Court, in White v. Regester, found that the "multi-member districts [were] being used invidiously to cancel out or minimize the voting strength of racial groups." 412 U.S. at 765, 93 S. Ct. at 2339, quoted in Bolden. In relying heavily on evidence of the historical background of an apportionment scheme and on its actual *979 impact on the functioning of the electoral system, the district court in Graves v. Barnes properly applied the evidentiary standards for establishing intent to discriminate on the basis of race in voting rights matters.

The elaborate explication of these two factors in this dissent, and the reliance on them in the subsequent evaluation of the evidence adduced in this action, is the product of the clear force of the Supreme Court's holdings concerning the burden of proof in voting dilution cases brought under the Fourteenth and Fifteenth Amendments.

 
3. Departures from established substantive criteria

A departure from established substantive criteria is cited as a relevant evidentiary source in Arlington Heights, 429 U.S. at 267, 97 S. Ct. at 564. See also Kirkpatrick v. Preisler, 394 U.S. 526, 533, 89 S. Ct. 1225, 1230, 22 L. Ed. 2d 519; Graves v. Barnes (Graves I), 343 F. Supp. 704, 723 (W.D.Tex. 1972) (three-judge court).

In the process of resolving issues such as legislative reapportionment, which implicate a broad range of factors, the legislature inevitably will identify criteria which must be satisfied, or, alternatively, should be avoided. Typically, conformance to these norms gives rise to a presumption of validity. This deference is a function of the Constitutional doctrine of separation of powers, which is premised on the assumption that legislative bodies, by virtue of their close ties with popular sentiment, are best able to sift competing interests and achieve an equitable solution to complicated matters of policy.

In terms of judicial review of legislative acts, this respect for the integrity of the legislative process is manifested in the disinclination of federal courts to invalidate legislative acts which are rationally related to a legitimate government interest. Provided that the legislature has explicated its objectives, and the action taken seems reasonably designed to achieve them, the legislation is deemed valid. Implicitly, this style of review sanctions the legislature's choice of goals, as well as its sense of priorities.

This facade of legitimacy is undermined, however, when, in making substantive choices, the legislature appears to apply varying or contradictory criteria, without sound justification. In San Antonio Independent School District v. Rodriguez, 411 U.S. 1, 93 S. Ct. 1278, 36 L. Ed. 2d 16 (1973), the Court upheld the Texas school financing system against an equal protection challenge. The state policy under attack had essentially two components: (1) the state provided aid to local school districts on basis of a formula which was applied equally to schools throughout the state; (2) this state grant was supplemented by local tax revenues derived from property taxes. Hence, with respect to total revenue applied to public education through local school boards, the amount varied from locality to locality according to local wealth and tax rates. Plaintiffs challenged this system on the ground that it violated the Equal Protection Clause.

The Supreme Court upheld the state scheme. In substance, the Court held that the state system was applied neutrally across the state, and that the policy supporting it preference for local autonomy was sound. However, had the state professed a desire for local autonomy in school finance and yet varied the amount of its education grants to the various localities, the validity of the policy would be marred by its inconsistent application, and its maintenance would have violated the Constitution.

This example exemplifies a situation in which departures from established substantive criteria may, in themselves, demonstrate a violation of the Equal Protection Clause. Such a case is analogous to the recognition, in Arlington Heights, that, in special instances, racially disparate impact alone might prove intent to discriminate. 429 U.S. at 266, 97 S. Ct. 563. E.g. Gomillion v. Lightfoot, 364 U.S. 339, 81 S. Ct. 125, 5 L. Ed. 2d 110 (1960); Sims v. Georgia, 389 U.S. 404, 88 S. Ct. 523, 19 L. Ed. 2d 634 (1967). In most instances, however, discriminatory impact is a crucial but insufficient element *980 of the evidentiary inquiry into intent. Similarly, in most cases, substantive departures will provide an important window into the hidden purposes of the legislature, but, standing alone, will not prove invidious intent.

As a practical matter, in the instant action, the Texas state legislature was highly self-conscious in establishing its priorities for the redistricting process. Members were thoroughly informed, and constantly reminded as well, of the legal standards under which the apportionment scheme ultimately adopted would be reviewed. In recognition of the constitutional and statutory proscriptions against retrogression and dilution, various participants in the process set forth subsidiary criteria against which the final scheme would be measured. There is, of course, the initial question of the validity vel non of these standards. Beyond that inquiry, Arlington Heights indicates that the application of these criteria, if valid, should be consistent; and deviations should be thoroughly and convincingly explained.

The legislature was apprised of this requirement by its legislative counsel, in a study prepared specifically for the purpose of informing legislators of the legal constraints on their action. DX 4.4.11 at 203. See also DX 4.4.4 at 670-71. To the extent that variations in the application of substantive criteria affects the manner in which minority interests were treated by Senate Bill No. 1, such departures are particularly suspicious. The legislature was keenly aware of the importance of minority considerations in the entire process, and of the way in which particular actions might affect those interests. This knowledge supports and strengthens the inference of intent that may be drawn from inconsistencies in the application of established standards.

 
4. Procedural Irregularities

As the Court stated in Arlington Heights, "[d]epartures from the normal procedural sequence also might afford evidence that improper purposes are playing a role." 429 U.S. at 267, 97 S. Ct. at 564.

In a manner similar to that outlined above with respect to substantive deviations, procedural irregularities may cast a pall of illegitimacy over a legislative enactment. First, extreme and grotesque procedural flaws may, of themselves, involve an independent constitutional violation. For instance, if, in the course of deliberation, a legislature, by means of a special parliamentary rule, prohibited representatives of minority districts from participating in the debate, this bizarre measure might well amount to a per se finding of intentional discrimination.

The mention of this evidentiary source in Arlington Heights assuredly does not contemplate abnormalities of this magnitude. It commands, instead, a review of the procedure by which certain measures were taken, to determine whether the interests which plaintiffs in a given case assert were denied equal protection of the law were, as a procedural matter, treated differently than other interests or were at variance with established methods.

 
5. The legislative record

The Court has stated that "[t]he legislative ... history may be highly relevant, especially where there are contemporary statements by members of the decisionmaking body, minutes of its meetings, or reports." 429 U.S. 268, 97 S. Ct. at 565. The evidence presented in this case includes a voluminous legislative record documenting the process by which the Texas state legislature passed the congressional reapportionment bill. Without more, the evidentiary sources previously described would require a thorough search of the legislative record for evidence of intent. Of course, as noted, the official record is unlikely to contain explicit acknowledgement of invidious intent. See Lodge v. Buxton, 639 F.2d 1358 & n. 82 (5th Cir. 1981); McMillan v. Escambia City, Fla., 638 F.2d 1239, 1246-47 n. 15 (5th Cir. 1981). However, the "contemporary statements of members of the decisionmaking body" reveal a great deal about their sense of priorities, their evaluation of competing concerns, and their commitment *981 to established substantive goals. The official legislative record itself is a valid source of "circumstantial" evidence of intent as defined under Arlington Heights. Hence, the legislative record of the consideration and adoption of Senate Bill No. 1 must be carefully searched, as an evidentiary source of independent value, and also as a body of circumstantial evidence of intent of the various kinds suggested by Arlington Heights and White v. Regester.

 
IV. 
Plaintiffs' Claims

The plaintiff and plaintiff-intervenors in this action allege that Senate Bill No. 1, the congressional reapportionment plan adopted by the Texas State legislature, intentionally discriminates against Black and Mexican-American citizens. Specifically, plaintiffs contend that S.B. 1 impermissibly dilutes the voting strength of minorities, by "packing" them in certain congressional districts and, consequently, fencing them out of adjoining districts. Plaintiffs assert that these effects were intended by the legislature. Therefore, they claim Senate Bill No. 1 constitutes an impermissible racial gerrymander, in violation of the Fourteenth and Fifteenth Amendments of the United States Constitution, the Voting Rights Act of 1965, 42 U.S.C. § 1973, and the Civil Rights Act of 1964, 42 U.S.C. § 1983.

The Supreme Court has plainly held that claims of vote dilution are cognizable under both the Fourteenth and Fifteenth Amendments. City of Mobile v. Bolden, 446 U.S. 55, 100 S. Ct. 1490, 64 L. Ed. 2d 47 (1980). However, it appears that the elements of the two separate constitutional claims are identical. Id. Because I believe that plaintiffs have clearly established their claim under settled doctrines of constitutional law, I would pretermit decision of the unresolved question of whether Section 2 of the Voting Rights Act, 42 U.S.C. § 1973, provides a private right of action.

As observed previously, in order to prevail on their claims, plaintiffs must establish Senate Bill No. 1 intentionally discriminates on the basis of race. The evidentiary sources which will be reviewed to determine legislative purpose have been outlined at some length. However, as was also stated earlier, the necessity for this "sensitive inquiry" into intent commanded by Arlington Heights only arises if, as a preliminary matter, plaintiffs have established that the challenged state action actually has a racially discriminatory impact. A claim of vote dilution plainly alleges such a discriminatory impact.

In broad terms, the Supreme Court has entertained two separate types of voting rights cases. First, the Court has considered allegations that the right to vote has been denied absolutely by virtue of state action; that is, that minority citizens have not been allowed to participate at all in the electoral process. Examples of claims of this nature are Gomillion v. Lightfoot, 364 U.S. 339, 81 S. Ct. 125, 5 L. Ed. 2d 110 (1960) (racial gerrymander of city boundaries prevented blacks from voting in municipal elections); and Harper v. Virginia Board of Elections, 383 U.S. 663, 86 S. Ct. 1079, 16 L. Ed. 2d 169 (1966) (proscribing the use of poll taxes).

Second, the Court has examined allegations that a particular electoral system maintained by a governmental unit denies the right to vote by means of diluting the effectiveness of the franchise; that is; minorities have been denied the right to participate in the processes of self-government on an equal basis with other citizens. Examples of this form of voting rights cause of action are White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973), and Wright v. Rockefeller, 376 U.S. 52, 84 S. Ct. 603, 11 L. Ed. 2d 512 (1964). In these cases, the claim is not that minority citizens are denied access to the ballot; the complaint, instead, is that, because of intentional discrimination, the electoral system has been designed in a way that has the effect of diluting the weight of minority votes. See also United Jewish Organizations v. Carey, 430 U.S. 144, 155, 161-62, 165-68, 97 S. Ct. 996, 1004, 1007-08, 1009-11, 51 L. Ed. 2d 229 (1977) (Opinion of White, J.).

*982 An allegation of vote dilution, then, is cognizable under the Constitution. Nonetheless, the Court has not delineated with specificity what elements must be established by a plaintiff claiming an electoral system has the effect of diluting minority votes. (Again, it must be emphasized that this demonstration of discriminatory effect is a prima facie element of a claim of racial gerrymandering. Proof of discriminatory purpose is an additional required element.)

White v. Regester contains the most explicit formulation yet offered by the Supreme Court. In that case, plaintiffs contended that the multimember legislative districts maintained in certain counties in Texas violated the Constitution. Plaintiffs in White v. Regester obviously were allowed to participate in elections, so their contention did not involve absolute deprivation of the right to vote. Their claim was that multimember districts were "being used invidiously to cancel out or minimize the voting strength of racial groups." 412 U.S. at 765, 93 S. Ct. at 2339. There is nothing in White v. Regester or any other Supreme Court case to suggest that such a claim may be brought only in opposition of multimember districts. Indeed, both Wright v. Rockefeller and United Jewish Organizations v. Carey involved challenges to single-member congressional and state legislative districts, respectively. Accord: Kirksey v. Board of Supervisors of Hinds County, 554 F.2d 139 (5th Cir. 1977) (en banc), and cases cited therein at 143.

As a prelude to a review of the factual evidence relied on by the district court in that case, the Court in White v. Regester stated:

 
The plaintiffs' burden is to produce evidence to support findings that the political processes leading to nomination and election were not equally open to participation by the group in question that its members had less opportunity than did other residents in the district to participate in the political processes and to elect legislators of their choice.

412 U.S. at 766, 93 S. Ct. at 2339 (citing Whitcomb v. Chavis, 403 U.S. 124, 149-50, 91 S. Ct. 1858, 1872, 29 L. Ed. 2d 363 (1971)).

Beyond this instructive though certainly not definitive statement, the Supreme Court has not ventured to provide a textbook formulation of the elements of a claim of vote dilution.

In light of the substantial number of occasions on which the Court has dealt with claims of vote dilution, it may be assumed that there has been ample opportunity for such an elucidative pronouncement. The fact that one has not been forthcoming may fairly be taken to indicate not the Court's inability, but, rather, its unwillingness to do so. The corpus of voting rights cases presents a hodgepodge of analytical approaches which suggests that, at bottom, the inquiry is intensely practical and pragmatic. This impression is confirmed by the oft-quoted language at the conclusion of White v. Regester approving the "intensely local appraisal of the design and impact of the Bexar County multimember district in the light of past and present reality, political and otherwise." 412 U.S. at 768-69, 93 S. Ct. 2340-41.

Therefore, the evaluation of plaintiffs' claims of vote dilution must involve a close study of the effect of S.B. 1 on the ability of minorities to participate in the electoral process on equal terms with others. Their access to the process of nominating and slating candidates, their ability to influence the selection of issues to be placed on the agenda of public debate, and their continuing ability to induce legislative action that is responsive to minority concernments are examples of the indicia of legal and effective participation in the political process. There is, of course, no constitutional right to elect a representative of one's race, nor to legislative representation in proportion to the racial composition of the state population. Indeed, in drawing legislative districts, the legislature is under no duty to maximize the political strength of minorities. See Beer v. United States, 425 U.S. 130, 136 n. 8, 96 S. Ct. 1357, 1361 n. 8, 47 L. Ed. 2d 629; Whitcomb v. Chavis, 403 U.S. 124, 156-160, 91 S. Ct. 1858, 1875-1877, 29 L. Ed. 2d 363; City of Richmond v. United *983 States, 442 U.S. 358, 370-72, 95 S. Ct. 2296, 2303-04, 45 L. Ed. 2d 245. Yet it is plain that the ability of minorities to elect candidates of their own choosing, to influence the legislative activities of their representatives, and to form coalitions which can adequately voice their concerns in the political forum are indispensible cornerstones to the validity of representative government. The Supreme Court's vigilant preservation of the integrity of that process has been a hallmark of its defense of the Constitution, and especially its application of the Fourteenth and Fifteenth Amendments to claims of vote dilution. These issues help define the content of "equal access to the political process" and provide guidance for the evaluation of claims of vote dilution. This evaluation is acutely practical, and ultimately must be tied to the rich blend of historical and contemporary factors which, in combination, constitute the dense web of political and social reality.

 
V. 
An Overview of the Claims

Plaintiffs' claims of racial gerrymandering focus on several discrete areas of the state. Of course, to the extent that alterations in one congressional district inevitably affect the shape of contiguous districts, no specific claim of improper line-drawing may be quarantined from scrutiny. Nonetheless, the areas of the state which are fundamentally in controversy may be isolated, and the central issues relating to each identified.

 
A. Dallas and Dallas County.[3]

Under the apportionment plan in operation since 1973, the metropolitan area of Dallas was split into two congressional districts, numbers 5 and 24. Based on 1980 census figures, the population of District 5 is 18.3% Black and 10.8% Hispanic. Total minority percentage is 29.1%. The population of District 24, as presently configured, is 25.5% Black and 11.9% Hispanic. Total minority percentage is 37.4%.

Senate Bill No. 1 re-drew the line between these two districts, thereby substantially altering the racial composition of the districts. Under S.B. 1, the following minority racial percentages would be present: District 5 would be 5.1% Black and 7.0% Hispanic; overall minority percentage would be 12.1%. District 24 would be 46.5% Black and 17.3% Hispanic, for a total minority percentage of 63.8%. (DX 1-App. C-1).

Plaintiff-intervenors Juanita Craft et al. allege that S.B. 1 includes an impermissible racial gerrymander with respect to these two districts, it being their claim that the bill packs minority citizens into one district (the 24th) and thereby dilutes their voting strength. Under the judicially mandated 1973 congressional apportionment plan, minority voters exert a formidable influence on the election of Congressional representatives from both District 5 and District 24. Thus, Craft, et al. claim that S.B. 1 has the intended effect of diminishing the influence minority residents of Dallas County have on the election of congressional representatives.

 
B. Houston and Harris County.

Harris County is a metropolitan county dominated by the City of Houston. Under the 1973 apportionment plan, parts of Harris County are contained in four separate congressional districts, numbers 7, 8, 18 and 22. The racial composition of these districts is as follows:

 
*984
                                         Total
District     Black %    Hispanic %     Minority %
   7           5.9         8.1             14.0
   8          21.9        18.2             40.1
  18          43.1        30.9             74.0
  22          17.3        13.9             31.2

During the decade of the 1970's, Harris County experienced rapid population growth, and especially in terms of minority population. Consequently, not only had existing districts become malapportioned in terms of total population per district, which required that the lines to be redrawn, but population growth also mandated placing one of the three new congressional districts allotted to Texas somewhere in the Harris County metropolitan complex. The new district in the area was designated as Number 25. The composition of the five districts in Harris County under S.B. 1 is:

 
                                         Total
District     Black %    Hispanic %     Minority %
   7           3.2         7.1             10.3
   8          16.7        12.5             29.2
  18          40.8        31.2             72.0
  22           9.6        13.6             23.2
  25          25.0        13.7             38.7

Plaintiffs claim that the configuration of districts within Harris County contained in S.B. 1 unconstitutionally dilutes the strength of minority voting, noting that S.B. 1 maintains District 18 as a minority district, and distributes the remaining minority population among the other districts in a manner that dilutes their voting strength. The total minority population in Harris County is of sufficient size, plaintiffs claim, to warrant the creation of an additional minority district, as well as maintaining District 18 as a "safe" district. Rather than pursue this course, plaintiffs assert, the legislature adopted an apportionment plan that fragments the minority community and diminishes its influence in electoral politics.

 
C. South Texas

The area in the southern part of the state which is implicated by plaintiffs' claims lacks the clear definition of locale provided by the two metropolitan areas previously mentioned. Nonetheless, "South Texas", in the political parlance of the state, connotes a comprehensible geographical area.

Under the existing congressional apportionment scheme, District 15 encompasses most of south Texas. It includes the two most populous counties, Hidalgo and Cameron, which are in the southern-most part of the state, eleven other counties completely, and parts of two others. These counties contain heavy concentrations of Hispanic residents. District 15, as defined under the 1973 apportionment scheme, is 77.3% Hispanic. (It also includes a negligible number of Black residents.)

In the 1970's the entire south Texas region experienced sizeable population growth. The large majority of new residents were Hispanic. Accordingly, in considering where to place the three additional congressional districts which were allocated to Texas on the basis of the 1980 census, the legislature ultimately decided that one of the new districts should be placed somewhere in south Texas. (The broad variety of suggestions for placement will be carefully studied, infra.) The plan adopted by the legislature creates two districts in South Texas: the 15th, which is wholly contained within the boundaries of the existing 15th; and the 27th, a new district which, under S.B. 1, contains counties currently included in the 14th and 23rd Districts. Under S.B. 1, District 15 would be 80.4% Hispanic (with a negligible Black population) and District 27 would be 52.9% Hispanic and 3.0% Black.

Plaintiffs claim that the overwhelmingly Hispanic composition of District 15 constitutes an impermissible packing of Mexican-Americans into one district, at the expense of Hispanic influence in adjoining districts. District 27, as drawn, does contain a bare majority of Hispanic residents. However, plaintiffs contend, in light of the large concentration of Hispanic residents in South Texas, S.B. 1 has the effect of denying these persons full opportunity to participate in the electoral process. The packing of minorities into District 15 dilutes the influence of Hispanics living in South Texas, by *985 diminishing their ability to influence the political processes in that part of the state.

 
D. San Antonio, Bexar County and West Texas

The claims involving these areas may more easily be identified by reference to the congressional districts which are implicated. District 20 is wholly contained within Bexar County, and is focused on the metropolitan area of San Antonio. District 21, as presently configured, includes the northern suburban portions of San Antonio, and the northern part of Bexar County. Additionally, it stretches west from this urban area, to include a large part of West Texas. District 23 includes the southern portions of San Antonio and Bexar County. Moreover, it includes counties to the east of Bexar County, and the region stretching west from Bexar County to the lower Rio Grande Valley.

As under the 1973 plan, the three districts have the following racial composition:

 
                                         Total
District     Black %    Hispanic %    Minority %
    20       9.8         67.8         77.6
    21       2.5         26.0         28.5
    23       4.7         53.1         57.8

The lines of all three districts were changed. Though the boundaries of District 20 were shifted only a short distance in terms of area, the density of population in San Antonio renders even the slightest shift significant in terms of total population and racial proportion. District 21 was not substantially altered, either in terms of geographical area or racial composition. The boundaries of District 23 were substantially altered. However, the racial demographics of the district remained virtually identical.

Under S.B. 1, the makeup of the districts is as follows:

 
                                      Total
District   Black %    Hispanic %    Minority %
    20        8.8        61.7         70.5
    21        2.9        22.2         25.1
    23        4.1        53.1         57.2

Plaintiffs claim that the configuration of Senate Bill No. 1 with respect to these counties has the effect of denying minorities in San Antonio and West Texas the right to participate in the political process on equal terms with others. Specifically, they contend that the lines were drawn in a manner which deprives certain San Antonio residents of effective access to the political process; and that the boundaries established between Districts 21 and 23 fragments an identifiable community of interest, as defined by the large concentration of Hispanics living along the Rio Grande Valley from Presidio to Laredo. A segment of this community is placed in District 21, and a portion in District 23. Plaintiffs assert that this separation constitutes a racial gerrymander, which dilutes the political influence of a recognizable minority group.

 
VI. 
Overview of the Process

The Texas Legislature began to prepare for the reapportionment process in 1980, well before the census data became available. At that time, the House Committee on Regions, Compacts, and Districts ("House Committee") began holding a series of hearings throughout the state to allow citizens to express their desires concerning the placement of congressional district boundaries. These hearings continued through the Spring of 1981. DX. 89. In April of 1981, the House Committee began extensive hearings in Austin which continued into May of 1981. DX. 89, House Committee Tr., Sept. 10, 1981, through April 29, 1981. The Senate Subcommittee on Redistricting (hereinafter Senate Subcommittee) began similar hearings in the Winter of 1981, which continued into April 1981. DX. 5 through 20.

The House Committee began formal sessions for the purpose of voting on proposed plans of redistricting on May 11, 1981. DX. 89, House Committee Tr., May 11, 1981. On May 14, 1981, after two full days of debate, the House Committee passed to the floor of the House a proposed redistricting plan. Dx. 44, House Tr., May 18, 1981, p. 67. That plan, denominated as H.B. 1400, passed to third reading on May 18, and was sent to the Senate. DX. 44.

*986 On May 11, 1981, the Committee of the Whole Senate received the congressional apportionment plan devised by Kenneth Shephardson, Esq., aide to Sen. Peyton McKnight, and the subcommittee staff. DX. 22. The Senate Subcommittee considered the plan on May 16, held a mark-up session on May 18, DX. 23, 24, 25, and 26, and then a full Subcommittee hearing on May 19. DX. 27. The Committee of the Whole Senate again met to consider the apportionment bill on May 20, 1981. DX. 28. During that meeting, both the House plan, H.B. 1400, and the Senate plan, S.B. 799, were laid out for consideration. DX. 28, Committee of the Whole Senate, Tr., May 20, 1981, p. 357. The Committee of the Whole amended S.B. 799, and then voted to substitute S.B. 799 for H.B. 1400. Id., p. 107-108. The Committee of the Whole then voted to send H.B. 1400, as substituted, to the Senate, with the recommendation that it be passed. Id., p. 108. The Senate promptly passed the bill.

The House and Senate versions of H.B. 1400 were dissimilar, especially in their treatment of Dallas-Fort Worth, Houston, and South Texas. Therefore, a conference committee was called to attempt to formulate a compromise plan. The Conference Committee met for three days: May 30 and 31, and June 1, 1981. DX. 30, 31, and 32. However, the Conference Committee failed to agree, and the regular session adjourned, automatically, on June 1. DX. 32. Because the House and Senate were unable to reach an accord on a congressional apportionment plan during the regular session, Governor William P. Clements, Jr., called a special session of the legislature, to begin July 13, 1981, to consider congressional apportionment, and several other matters. The Senate began meeting as the Committee on the Whole on July 13, 1981, DX. 34, and passed a bill on July 15. DX. 36. The Senate met in floor session on July 20, 1981, and approved a plan of apportionment, S.B. 1, on July 21. DX. 37 and 38. S.B. 1 was then sent to the House.

The House Committee heard testimony on July 16, 21, and 22, and convened in formal session on July 27, to vote on prospective plans. DX. 52 and 53. At that time, the Chairman, Rep. Tim Von Dohlen, set out S.B. 1 for the committee's consideration. DX. 53, House Committee Tr., July 27, 1981, p. 2. The committee added several amendments to the Senate bill, and then passed the bill, as amended, to the floor of the House. Id., p. 34. The House, after considering the committee version of S.B. 1 for two days, voted, on August 3, to return the bill to committee. DX. 55, House Tr., August 3, 1981, p. 215. The House Committee met on August 5 and 6, but was unable to agree on a bill to report to the House. DX. 54; DX. 53, House Committee Tr., August 6, 1981. The House Committee again met on August 7, and while the House stood at ease, finally reported out a bill. DX. 53, House Committee Tr., August 7, 1981. The next day, August 8, 1981, the House began deliberation on the plan passed by the Committee. DX. 56. Early in that legislative day, Speaker Clayton introduced a complete substitute to the bill passed by the Committee. Id., p. 21-22. Debate on the Clayton substitute, and other substitutes, continued until the late evening. Tr., Vol. II, p. 601-602. Rep. Bill Messer then called the previous question, and a vote was taken on the pending substitute, and on the Clayton substitute. The Clayton substitute passed to third reading. The House then recessed, and returned to order in the early morning of August 9, a procedural ploy which enabled the House to vote on the Clayton substitute on third reading. TR., Vol. I, at p. 95. All minority members of the House boycotted the August 9 session in an attempt to break the quorum, but a call was placed on the House, and a quorum was obtained. DX. 56, House Tr., August 9, 1981, p. 203-208. With a bare quorum present, the House passed the Clayton substitute. Id., p. 209.

The House version of S.B. 1, the Clayton substitute, varied somewhat from the Senate version. Nonetheless, the Senate passed the House version of S.B. 1, with relatively little debate, on August 10, 1981. DX. 39, Senate Tr., August 10, 1981, p. 30.

 
*987 VII. 
Historical Background

To a certain extent, judicial notice may be taken of the fundamental premise that, in the past, an official policy of racial discrimination operated in Texas at all levels of social and political life. This program effectively denied blacks in Texas the opportunity to participate in any facet of life economic, social, educational, political, and personal on equal terms with whites. Of course, these nominally separate categories of experience dovetail in actuality, and in the instance of race relations in Texas, formed an entire cultural pattern which worked, at every point, to segregate blacks and remove them from the dominant institutions of Texas life. See generally, C. Vann Woodward, Origins of the New South, 1877-1913 (1972).

Of concern in the present case is official discrimination against minorities in the realm of political life. That political involvement is integrally related to participation in other elements of social life is beyond question. However, even when official discrimination in politics and the political process is viewed in isolation, the legacy is long and almost overwhelming. No one can seriously contend that a catalog of legal actions pertaining to discrimination in voting adequately captures the harsh reality of political racism in Texas. Nonetheless, a roll-call of major cases involving constitutional challenges to various devices employed in Texas to fence blacks out of the political process may provide a silhouette that at least hints of the outlines of the historical patterns.

 
Nixon v. Herndon, 273 U.S. 536, 47 S. Ct. 446, 71 L. Ed. 759 (1926)
 
The Supreme Court held unconstitutional a Texas statute which prohibited blacks from voting in the Democratic primary.
 
Nixon v. Condon, 286 U.S. 73, 52 S. Ct. 484, 76 L. Ed. 984 (1932)
 
The Supreme Court struck down Texas statute which permitted the executive committee of a political party to prescribe membership qualifications.
 
The challenged statute was passed within five days of the Court's action in Nixon v. Herndon. Pursuant to the statute, the Democratic Party adopted qualifications which prohibited non-whites from participation in the party primary.
 
The Court's opinion did not flatly ban the delegation of power to the executive committee, nor did it hold that the exclusionary action of the committee was per se unconstitutional. Rather, the decision rested on a finding that action taken by the State Executive Committee was state action, and so, in this instance, violated the 14th Amendment. The opinion reserved the question whether the discriminatory action involved would be proper if taken by the entire party, through a convention.
 
Grovey v. Townsend, 295 U.S. 45, 55 S. Ct. 622, 79 L. Ed. 1292 (1935)
 
Not surprisingly, the Democratic Party of Texas, in response to Nixon v. Condon, adopted at a state Convention a resolution which limited membership in the Democratic Party to whites. The Supreme Court upheld this resolution against constitutional challenge, on the ground that a resolution of a political party does not constitute state action, and so cannot violate the Fourteenth Amendment.
 
The Court's holding with respect to the nature of action taken by a political party rested on deference accorded to two Texas Supreme Court decisions, Love v. Wilcox, 119 Tex. 256, 28 S.W.2d 515 (1930), and Bell v. Hill, 123 Tex. 531, 74 S.W.2d 113 (1934). The combined weight of these two state decisions was that, under the Constitution of the State of Texas, political parties are voluntary associations formed by the free will and liberty of citizens. Article 1, §§ 2 and 27 of the State Constitution grant to citizens the liberty of forming political associations, and action taken pursuant to this grant of political freedom cannot be categorized as state action. Accordingly, the state Supreme Court held that efforts of the political parties, when taken by vote of the full convention, rather than by the *988 executive committee, is immunized from challenge under the Fourteenth Amendment.
 
Smith v. Allwright, 321 U.S. 649, 64 S. Ct. 757, 88 L. Ed. 987 (1944)
 
The Supreme Court overruled Grovey v. Townsend, and held that the exclusion of non-whites from participation in primary elections was unconstitutional.
 
This change of positions derives from the Court's holding in United States v. Classic, 313 U.S. 299, 61 S. Ct. 1031, 85 L. Ed. 1368 (1941), that primary elections are an integral part of the entire electoral process, and so action affecting the ability of citizens to participate in primary elections is, by definition, state action. With that premise established, the action of the democratic party excluding non-whites from participation in primary elections was unquestionably a denial of equal protection.
 
Terry v. Adams, 345 U.S. 461, 73 S. Ct. 809, 97 L. Ed. 1152 (1953).
 
The Supreme Court held that a "pre-primary" conducted by a private association which excluded non-whites from membership was unconstitutional, in that the record clearly demonstrated that the victor in this pre-primary invariably secured the nomination of the Democratic Party and subsequently won the election. This case involved the last overt attempt in the state of Texas to maintain the white-primary system.
 
Texas v. United States, 384 U.S. 155, 86 S. Ct. 1383, 16 L. Ed. 2d 434 (1966), affirming United States v. State of Texas, 252 F. Supp. 234 (W.D.Tex.1966)
 
The Supreme Court summarily affirmed a lower court ruling invalidating the use of a poll tax as a precondition to voting.
 
The lower court found that the poll tax violated the due process clause, since it is "in fact a restraint and a charge on the exercise of the fundamental right to vote." 252 F. Supp. at 236. The court specifically held that the tax did not violate either the Fourteenth or Fifteenth Amendments.
 
The Supreme Court's summary affirmation was explicitly based on its holding in Harper v. Virginia State Board of Elections, 383 U.S. 663, 86 S. Ct. 1079, 16 L. Ed. 2d 169 (1966), which held poll taxes violative of the Fourteenth Amendment (overruling Breedlove v. Suttles, 302 U.S. 277, 58 S. Ct. 205, 82 L. Ed. 252 (1937)). The poll tax had been used in Texas since 1902, and, the district court found, had been imposed originally with the intent of disenfranchising blacks. 252 F. Supp. at 245.

This impressive string of Supreme Court decisions protecting the right of non-whites in Texas to vote stands as a disheartening testament to the perseverance and craft of the dominant political leaders, in their attempt to exclude minorities from participation in the process of self-government. White primaries and poll taxes are notorious and blatant manifestations of the racism that has infected Texas politics. Yet the proscription of these tactics did not end the tale of discrimination in voting. In light of the Supreme Court decisions indicating its refusal to countenance such overt tactics, other, more subtle means of disenfranchising disfavored minorities were devised. Though certain of these tactics have about them a superficial cast of legitimacy, federal courts have frequently held them to be motivated by an invidious intent to discriminate on the basis of race.

In the wake of the Supreme Court's affirmation of United States v. State of Texas, the Texas state legislature proposed, and the electorate ratified, a constitutional amendment providing for annual registration of voters in the state. This enactment was challenged as violative of the Fourteenth Amendment to the United States Constitution. In Beare v. Smith, 321 F. Supp. 1100 (S.D.Tex.1971), aff'd sub nom. Beare v. Briscoe, 498 F.2d 244 (5th Cir. 1974), the challenge was upheld. The court explicitly traced the enactment of the constitutional amendment to the history of racial discrimination in Texas, and found that the amendment, though facially neutral, unconstitutionally deprived minorities of the right to participate in the electoral *989 process on an equal basis with other citizens.

In Robinson v. Commissioner's Court, Anderson County, 505 F.2d 674 (5th Cir. 1974), the Court of Appeals for the Fifth Circuit affirmed a district court ruling that a 1969 reapportionment plan adopted by the Commissioner's Court, which divided the county into precincts for purposes of county elections, constituted an impermissible racial gerrymander. Specifically, the court found that the division of a particular concentration of black voters into three separate precincts operated to minimize or cancel out the voting strength of those citizens. In affirming the district court's finding, the Court of Appeals noted that "[u]nfortunately, the disrespect of voting rights is not a recent innovation in county government in Texas.", 505 F.2d at 676, and viewed the facially neutral plan in light of the history of overt and subtle attempts to fence minorities out of the process of self-government.

These two cases are mentioned only to provide examples of the use of facially neutral techniques of voting discrimination by governmental units in the state of Texas. A full listing of civil actions brought on behalf of blacks and Hispanics in Texas challenging various measures which deprive citizens of the right to vote on the basis of race is beyond the scope of this dissent, and is not necessary to establish the fundamental point that the history of racial discrimination in voting in Texas has endured the tidewaters of the Voting Rights Act, and the initial series of Supreme Court decisions invalidating the use of poll taxes and white primaries. (See Brief of M. Garcia, App. at 3-4, n. 3.) Indeed, the legacy extends to the present, with a force that is surprising.

In 1975, Congress extended the special pre-clearance provisions of the Voting Rights Act of 1965 to Texas. This decision was made on the basis of extensive hearings into the history of voting discrimination in the state. See Hearings Before the Senate Committee on the Judiciary, Sub-Committee on Constitutional Rights, 94th Congress, 1st Session (April 8, 9, 10, 22, 29 and May 1, 1975). Since the pre-clearance provisions were extended to Texas in August of 1975, the Department of Justice has lodged far more objections to governmental actions affecting voting rights in Texas than any other covered state. Between August 15, 1975, and September 18, 1981, the State and its various political sub-divisions received 91 letters of objection. P-1 A. Garcia Exh. No. 76. Congressional Record, October 5, 1981, p. H 6943. In this same period, no other covered state had more than 50 objections, and only three had more than thirty. Id. at H 6941-6944. The election changes objected to by the Department of Justice include the movement of polling places, proposed annexations, alteration of district lines, and a state-wide purge of voter registration lists.

One objection lodged by the Department of Justice deserves special note. In 1975, the state legislature adopted a measure which altered the standards under which the eligibility of a party for State primary election financing were determined. Eligibility for state funds is conditioned on a given party receiving a certain percentage of the total votes cast for governor in the preceding election. In 1975, the state legislature raised the requisite percentage from two percent to twenty percent. (This figure was also used to determine which candidates would be placed on the ballot.) The effects of this change would have been to reduce the visibility of third-parties, by excluding them from primary ballots, and also to deprive those parties of valuable funding for primary election campaigns. The Attorney General objected to the change, and thereby rendered the enactment a nullity.

Contemporaneous political events shed interesting light on the possible motivation for the legislative action. In the 1972 election for governor, La Raza Unida Party a primarily Mexican-American party had received approximately 6.3% of the total votes cast for governor. Under the old formula for determining eligibility for state financing of primary elections, La Raza Unida would have qualified for such funds. However, under the 1975 change, the party *990 would not have received reimbursement, and so would have had to fund, out of its own resources, the full expense of primary election campaigns. The plain effect of the proposed change would be severely to cripple the ability of La Raza Unida to participate on equal terms with other parties in the electoral process. Given the fundamentally Hispanic nature of the party, and its role as a conduit for Hispanic political sentiment, the alteration would have worked a severe hardship on Mexican-American political activism.

Despite the extension of the Voting Rights Act to Texas, and the requirement that all changes in laws, practices, and procedures affecting voting be submitted for preclearance, a wide variety of election abuses continue to frustrate the attempts of minorities to participate in the democratic process. The basic reason for this continuing inequality is that the Voting Rights Act, for all its breadth, simply cannot reach certain "irregularities" in the manner in which approved voting practices are applied at a local level. Vigilant application of the terms of the Voting Rights Act may assure that election changes which constitute impermissible gerrymandering, or undue restraints on the ability of minorities effectively to exercise the franchise, will not be precleared, and so will not be implemented. However, the act of voting in an election is, ultimately, an acutely local enterprise, involving a trip to a designated place, first to register, and later to vote, and the casting of a ballot, which is then processed by local officials. Though this process may be formalized by an approved election code, it is susceptible to abuse at the local level in a variety of ways, which may escape detection. The former Secretary of State for the State of Texas, George Strake, expressed his concern over "voting irregularities" in the state which hampered the ability of minorities to participate in the electoral process. P-I M. Garcia, Exh. No. 13, pp. 7-31. At a hearing on election abuses conducted by the Southwest Voter Registration Education Project, former Secretary Strake stated that, "I am fully aware that abuses, irregularities and illegalities have existed, and I have every reason to suspect that they continue to exist." Id. at 9. Despite his repeated attempts to combat these abuses, former Secretary Strake stated that his office "had experience with cases where we felt there was an unwarranted delay and a reluctance to prosecute [violations] on the part of some local officials." Id. at 10.

Former Secretary Strake's testimony on this occasion highlights a familiar and critical problem in the preservation of the right of minorities to participate in self-governance. The enormous scope of the process in Texas renders effective monitoring of local elections vastly difficult. As a practical matter, the federal government cannot insure that local elections throughout Texas are conducted in an equitable manner. Consequently, a large portion of that burden devolves upon state officials, whose commitment to equal political participation is seriously open to question. See P-I M. Garcia, Exh. Nos. 36A & 13. As an institutional matter, then, the reluctance of state officials to enforce the neutral provisions of state and federal election regulations allows local authorities to apply those provisions in a discriminatory manner. By their nature, these localized "irregularities" defy ready quantification. Yet the history of voting discrimination in Texas is replete with charges of vote fraud, ballot stuffing, ballot tampering, unpublicized re-location of polling places on the day of election, alterations in registration requirements and intimidation of voters. See e.g., Hearings on Extension of Voting Rights Act in 1975, April 8, 9, 10, 22, 29 and May 1, 1975; H.R.Rep. No.97-227 (Committee Report on Voting Rights Act Extension) (September 15, 1981); P-I M. Garcia Exh. Nos. 13 & 36A; Texas Advisory Committee, U. S. Commission on Civil Rights. A Report on the Participation of Mexican-Americans, Blacks and Females in the Political Institutions in Texas, 1968-1978 (January, 1980).

The attempts of Hispanic voters to register and vote meet with the additional obstacle of language barriers. Though the Voting Rights Act now provides for language assistance to insure that language minorities *991 are not effectively excluded from participation in the democratic process, the statute has had only a limited practical effect in Texas. Inadequate resources and insufficient commitments on the part of local officials have led to severe inequalities in the way election provisions affect the attempts of Texans to register and vote. The Federal Election Commission recently found that, despite the provisions of the Voting Rights Act which attempt to remedy the discrimination visited upon Spanish-speaking voters, jobs at registration offices are routinely filled by persons who are unable to assist Spanish-speaking voters with even the most rudimentary information. The Commission concluded that election administrators in Texas, taken as a whole, were the least adequately prepared, among the states covered by the Act, to implement the bi-lingual provisions of the 1975 Voting Rights Act amendments. Additionally, the Commission found that the state was least responsive to the needs of Spanish-speaking voters. Federal Election Commission, Bilingual Services, Vol. III, Report on the State of the Art. The Supreme Court has approved the invalidation of literacy tests which have the effect of excluding language minorities from the electoral process. Katzenbach v. Morgan, 384 U.S. 641, 86 S. Ct. 1717, 16 L. Ed. 2d 828 (1966). However, Texas state registration procedures, which have the effect of excluding language minorities, by virtue of the formidable hurdle created by the failure to provide bi-lingual assistance, frequently remain in place, because they are not reviewed under Section 5 of the Voting Rights Act. The practical effect is far-ranging disenfranchisement of language minorities.

In light of the difficulty of empirical demonstrations of these insidious practices and effects, a more effective manner of establishing effective exclusion of minorities from participation in the electoral process is a survey of actual election results. By now, the disheartening litany of under-representation of blacks and Hispanics in elected positions in Texas is familiar, and need not be repeated in detail. A brief survey will suffice. From 1968 to 1978, 98.4% of the persons elected to serve as county judges were white; 94% of the county commissioners were white; 96% of the mayors were white; 95% of the members of City Councils were white; 94% of the members of the most influential state regulatory bodies were white. Texas Advisory Committee, United States Commission on Civil Rights, "Texas: The State of Civil Rights, Ten Years Later, 1968-1978."

Since the Civil War, Texas has not elected a black or Hispanic United States Senator. Of the 24 congressional representatives elected under the existing apportionment plan, one is black and two are Hispanic. At no time in Texas history has there been more than one black in the congressional delegation, nor more than two Hispanics. Since the state executive branch was reorganized under the 1876 Constitution, there has never been a black or Hispanic elected to any executive office. (One Hispanic was appointed Secretary of State, in March 1968.)

In the State Legislature, blacks and Hispanics are under-represented as well. The state legislature consists of 31 Senators and 150 State representatives. Since 1973, there has not been a black state senator. There are currently four Hispanic state Senators. In the present State House of Representatives, there are 13 black representatives (8.7%) and 18 Hispanic representatives (12%). Each of these figures constitutes disproportionately low representation of these two ethnic groups. (DX 1, App. C-5.)

A variety of factors contribute to the skewed results of political races in Texas. Blacks and Hispanics traditionally participate in elections at a lower rate than do Anglos. Both registration and turnout rates tend to be lower for these two ethnic groups. This central fact of political life is common knowledge in political circles, and informs political debate at all levels. In attempting to assure a "safe" minority district for any given election, those drawing the district lines typically use 65% minority population as a benchmark figure, in recognition of the relatively low propensity of *992 minorities to register and vote. See, e.g., DX 3, p. 21; DX 37.3, p. 17; Deposition of K. Sheperdson, p. 55; Dep. of Gov. W. Clements, p. 43; Veto Proclamation of Gov. Clements, P-I M. Garcia, Exh. 34 at p. 4. Studies conducted by Dr. Frank D. Bean for the purpose of submission to the Attorney General for pre-clearance of the Congressional Redistricting bill reached the following conclusions:

 
1. The greater the percentage of Hispanics or of Blacks in a given area, the lower the percentage of the population registered to vote and the lower the percentage of registered voters that turned out for the 1980 Presidential election. Just the opposite occurs in the case of the percentage of Whites and Others in an area.
 
2. The percentage of Blacks in an area is more negatively related to the turn-out percentage of that area than is the percentage of Hispanics.
 
3. The relationships involving turn-out percentage are stronger than those involving registration percentage.

DX No. 1, App. C-4, p. 2.

This disinclination of blacks and Hispanics to participate in the political processes which affect them is part of an entire cultural pattern that reinforces in these groups a sense of alienation. The state legislature has been notoriously unresponsive to the concerns of minority communities. Representative Matt Garcia, a plaintiff-intervenor in this action, identified major areas of state policy which dramatically affect and deeply concern Texas minorities: reapportionment, bilingual education, welfare, education, and prison reform. TR. at 27-28. Judicial notice may be taken of the fact that these areas have been the subject of extensive litigation, resulting in court orders preserving the guarantees of the Constitution vis-à-vis the intransigence and insensitivity of the state legislature. See generally, Texas Advisory Commission, United States Commission on Civil Rights, Status of Civil Rights in Texas. Vol. I & III (January 1980), esp. Vol I. pp. 162-180, and Vol. III, pp. 41-111, 142-179. Recent legislative activity involving state expenditure on bilingual education revealed a profound antipathy on the part of state legislators to funding these programs at a level that might assure their effectiveness. House Journal, May 29, at 3945 ff.

The district court in Graves v. Barnes (Graves I), 343 F. Supp. 704 (W.D.Tex.1972) (three-judge court), found a close connection between deprivations visited on Hispanics in matters such as education, and their alienation from state politics:

 
There can be no doubt that lack of political participation by Texas Chicanos is affected by a cultural incompatibility which has been fostered by a deficient educational system.

343 F. Supp. at 731.

As stated earlier, the Supreme Court's decision in Brown v. Board of Education was premised on the inherent connection between quality education and effective participation in the political process. Deprivation in the former area inevitably entails ignorance, ennui, and apathy in the latter. The state system of public school financing allows the amount of money spent in a given locality to depend on the relative wealth of that district, and the desire of local citizens to tax themselves to pay for education. As a practical matter, this system creates wide disparities in perpupil expenditure. The Supreme Court has upheld the constitutionality of such a financing scheme, San Antonio Independent School District v. Rodriguez, 411 U.S. 1, 93 S. Ct. 1278, 36 L. Ed. 2d 16, (1973), in an opinion which deferred to the state legislature's policy preference for local autonomy in school finance. This deference was appropriate in analytical terms, because the Court found that wealth was not a suspect classification, and that education is not a fundamental right. This decision notwithstanding, the Supreme Court has on other occasions recognized the direct connection between education and political participation. Brown v. Board of Education, supra; White v. Regester, supra.

Other socio-economic indicia support the conclusion that minorities in Texas are subject *993 to discrimination in the provision of essential services, and that these policies work to reinforce their alienation from the political process. Private and public employment discrimination in Texas is welldocumented in legal opinions. See, e.g., Fisher v. Procter & Gamble Mfg. Co., 613 F.2d 527 (5th Cir. 1980); Sabala v. Western Gillette Inc., 516 F.2d 1251 (5th Cir. 1975); Garcia v. Victoria I.S.D., 17 E.P.D. 8544 (S.D.Tex.1978).

The Texas Advisory Committee to the United States Civil Rights Commission found extreme underrepresentation of blacks and Hispanics at all levels of public employment in Texas. Whether employment opportunity is measured by numbers of persons employed, or, more tellingly, by salary levels of those persons actually employed, the conclusion of the Committee was clear: blacks and Hispanics are not employed in the public sector on an equal basis with Anglos. "The State of Civil Rights, Ten Years Later, 1968-1978", pp. 5-16. See also, "Status of Civil Rights in Texas, Vol. III.", supra, at pp. 37-159.

Discrimination in employment obviously leads to lower income levels for individuals and families victimized by that discrimination. According to the 1970 census, 37.9% of all Mexican-American families in South Texas had incomes below the poverty level. In certain counties, the percentage was in excess of 60%. The mean income of Hispanic families is 56.7% of the mean income of Anglo families. Hispanic per capita income is only 40% of Anglo per capita income. United States Census Data, 1970. See also, Lyndon B. Johnson School of Public Affairs, "The Health of Mexican-Americans in South Texas", pp. 7-15; Office of the Governor, "Summary of Selected Demographic Characteristics from Census Data-Fourth Count" (August, 1972), P-I M. Garcia, Exh. No. 24, passim. Additionally, poverty and unemployment prevent minorities from obtaining adequate housing, and sufficient nutrition. These deprivations are severe in themselves, and also lead to inferior health. Yet, despite this relatively higher level of demand for health care, minorities in Texas are typically unable to secure adequate medical services because of their inability to pay for such care, and also the lack of adequate services in the relevant geographic areas. The connection between abject poverty levels such as those described above, and inferior housing, health, and medical care is well documented. Id.

There is in Texas, then, a staggering correlation between race and inferior education, unemployment and underemployment, poverty, inadequate housing, poor health, and insufficient medical care. For those who perceive themselves trapped within this complex of deprivation, the orthodox political process may often seem hopelessly remote, depressingly cumbersome, and irretrievably biased. A century of neglect, sometimes benign, frequently malign, has created a climate of alienation and apathy, mixed with deep-seated resentment in many of the minority communities in Texas. In light of this reality, low voter registration rates are not surprising. Voter registration patterns, in which blacks and Hispanics register and vote at rates substantially lower than Anglos, were established at a time when the legal regime of official racial discrimination in voting was wholly intact. And as the court in Graves v. Barnes (Graves I) noted,

 
[t]he fact that some of those laws have been changed within the past few years or months does not answer or offset the fact that the voting patterns remain firm. Because they were denied access to the political processes through years of discrimination, the Mexican-Americans do not now register and vote in overwhelming numbers.

343 F. Supp. at 733. See White v. Regester, 412 U.S. at 768-69, 93 S. Ct. at 2340-41.

Since the time of this finding, the Voting Rights Act has been applied to Texas. Federal monitoring of election changes has had a salutary effect, and a wide variety of Texas political figures have strongly urged that the provisions be extended during recent congressional debates on the subject. See H.R.Rep. 97-227 (Committee Report on Voting Rights Act Extension) (September *994 15, 1981). However, these safeguards notwithstanding, minority interests continue to be largely ignored by the state legislature, and blacks and Hispanics continue to be victimized by discrimination in fact in a broad range of socio-economic matters. An additional ten years of deprivation and exclusion from full participation in society has served to solidify the reality of alienation addressed by the court in Graves v. Barnes.

In summary, as an historical matter, blacks and Hispanics have been subjected to official discrimination in politics and in other forms of social life. Though a number of these formal legal barriers have been removed, typically as a result of court order, the reality of contemporary life in Texas for minorities still bears the stamp of this legacy of discrimination. The past and present reality, political and otherwise, of life in Texas for blacks and Hispanics is marked by poor education, inferior housing, poverty, unemployment, inadequate social services, and political isolation. In terms of registration and voting, and in terms of influence on the legislative process, minorities in Texas have suffered, since the Civil War, from an acute denial of effective access to the political process. The insensitivity of the state legislature to the concerns of blacks and Hispanics have left these groups on the outside looking in, through a window clouded by racism and its corollaries: powerlessness, destitution, segregation, estrangement, and despair. It is against this background that a congressional reapportionment plan must be judged.

 
VIII. 
Senate Bill No. 1 Effect and Legislative Motivation

In the overview of the plaintiffs' claims presented in Section IV, the effect of S.B. 1 on the racial composition of the various districts in dispute was set forth. These ethnic percentages provide little more than a starting point for the analysis of the impact of a redistricting plan on the ability of minorities effectively to participate in the electoral process. For a variety of reasons, bare statistics of racial composition reveal little about the actual nature of political representation and power. Most obviously, to the extent that registration and turn-out figures differ from population figures, the percentages distort the strength of minority voting in a given area. In other respects as well, these figures hide as much as they reveal. The political cohesiveness of minorities; the degree to which minority communities have established durable and respected means of conveying their political positions to persons in power; the ability of blacks and Hispanics to form solid coalitions which successfully bridge the cultural and political breaches which, on occasion, may separate those two groups these practical facts of political life have an immensely significant bearing on the real nature of minority representation.

The "Retrogression Study" submitted by the State to the Department of Justice for purposes of review of S.B. 1 under § 5 of the Voting Rights Act relies exclusively on statistical summaries of the minority composition of the various districts. The charts which form the "Study" provide a comparison between the number of districts under the existing apportionment plan and the number under S.B. 1, which have a given percentage of Hispanic and black population. Additionally, the charts show for districts with a stated minority population, how many were represented in 1980-81 by a person of that minority group. DX No. 1, App. C-2. As a preliminary matter, determination of the issue of "retrogression" is reserved for the Department of Justice. The standards under which that determination is to be made are the province of the Department. Beer v. United States, 425 U.S. 130, 141, 96 S. Ct. 1357, 1363, 47 L. Ed. 2d 629 (1976). However, it seems clear that the issues of fragmentation, dilution, and packing are not necessarily co-extensive with retrogression. Certainly, the retrogression study submitted by the State to the Department of Justice affords insufficient information to determine the constitutionality of S.B. 1 under the standards which have been set forth by the Supreme Court. Ultimately, the constitutional judgment *995 hinges on analysis of highly localized information involving communities of interest, political strength, polarized voting, and the quality of access available to minorities in a given region.

 
A. Dallas County

Dallas County is a metropolitan county dominated by the City of Dallas. Census figures for 1980 reveal that the total population of Dallas County is 1,556,549. Of this total, 287,613 are black (18.5%) and 154,560 are Hispanic (9.9%). DX 40 D. 5. Congressional Districts 5 and 24 are wholly contained within the boundaries of Dallas County. Additionally, portions of districts 3 and 6 fall within that county. As noted in section V, plaintiffs in this action challenge only the configuration of Districts 5 and 24. Under the 1973 and the proposed apportionment plans, the minority impact on elections in the other two districts is, and would be, negligible. Accordingly, in the ensuing discussion, the analysis of the effect of S.B. 1 on minority representation in Dallas County will be confined to Districts 5 and 24.

The 1973 congressional apportionment plan, as it affects Dallas County, is itself the product of constitutional litigation. White v. Weiser, 412 U.S. 783, 93 S. Ct. 2348, 37 L. Ed. 2d 335 (1973). The line which currently separates Districts 5 and 24 was adopted by the district court upon remand by the Supreme Court. In the opinion remanding the case to the district court, the Supreme Court directed the lower court to insure that the plan ultimately adopted was constitutional in all respects. 412 U.S. at 797, 93 S. Ct. at 2355. Since White v. Weiser was decided the same day as White v. Regester, and immediately follows that case in the official United States Reports, the inference is clear that the constitutional requirements of White v. Regester were to guide the district court in adopting a congressional reapportionment plan upon remand. The conclusion is plain: at least at the time it was adopted, the congressional apportionment plan in Dallas County was constitutional, not only with respect to one-person, one-vote constraints, but, additionally, with relation to minority representation.

Demographic changes in Districts 5 and 24 during the decade of the 1970's did not substantially alter the racial composition of the two districts. Census data for 1980 show that existing Districts 5 and 24 had a marginal decrease in percentage of black voters, and a slightly larger increase in the percentage of Hispanic voters. DX No. 1 App. C-1. Therefore, neither a concern for avoiding retrogression nor a desire to eliminate existing fragmentation and dilution would dictate that the congressional boundaries in Dallas County be altered, when those issues are analyzed purely on the basis of census data concerning the racial composition of those districts.

District 5 is presently represented by Congressman James Mattox, a white Democrat; Congressman Martin Frost, a white Democrat, is the elected representative from District 24. Mattox presently is serving his third term in Congress, Frost his second. The preponderance of the evidence adduced at the trial was that both Mattox and Frost are considered responsive to minority concernments. (Tr. at 234; DX 35.) Indeed, knowledgeable observers of Dallas politics testified that a candidate would have grave difficulty winning an election in either District 5 or 24 without the support of minority political groups. This type of political influence is essentially veto power. In neither district is the minority presence of sufficient strength that ethnic groups in Dallas County can elect a member of Congress of their choosing, without support from other political groups. However, at the same time, any candidate hoping to prevail in a congressional race must be sensitive to minority needs, and gain the support of that constituency. Id. The record conclusively establishes that the black community in Dallas County, in both congressional districts, has been able to exert effective political influence.

The demonstration of this political leverage exerted by black voters in Dallas is, of course, not equivalent to a showing that *996 racial bloc voting that is, racially polarized voting exists in Dallas County. In fact, election returns establish convincingly that, at least with respect to Congressional races, no such polarization exists. In 1980, Congressman Frost defeated a Black Republican opponent by a substantial margin, and in doing so, garnered the overwhelming support of the black community.

On the basis of these various measurements responsiveness of elected representatives, capacity of minority voters to form effective coalitions and exert substantial influence on the electoral process, and the absence of polarized voting the existing congressional districts appear to be constitutional, and, indeed, in some respects, exemplary. Nonetheless, the line between districts 5 and 24 was drastically altered. Congressional representation in Dallas County was isolated for intense scrutiny in the reapportionment process, in a striking and singular manner. It became the central focus of some of the major actors in the drama. Much about this special treatment is of profound interest, and a great deal of it is highly probative of the intent of the persons who designed S.B. 1. Hence, the issue merits the closest attention.

The line between Districts 5 and 24 follows the "historic boundary" of the Trinity River. That river runs through Dallas, and actually splits the concentrated black residential community in South Dallas into roughly equal halves. As stated, this dividing line was drawn by a federal court acting within the constraints of the Constitution, and the resulting split of the minority community is thus presumptively constitutional.

However, in the course of the legislative debate over congressional reapportionment, sentiment in support of creating a "safe minority district" in Dallas County developed among various members of the legislature, purportedly supported by the popular will of the minority community in Dallas. Untangling the several roots of this preference is essential to an understanding of the manner in which the debate over Dallas County proceeded, and in determining the motivation of the ultimate resolution.

The record plainly reveals that the legislative sympathy for creation of a minority district in Dallas derived from two quite separate sources of influence. First, the Coalition for Minority Representation, a Dallas-based political committee, urged the legislature to create a district in Dallas which would be predominantly black. See, e.g., DX 24.3, pp. 106ff; DX. 86; TR. at 475-76 (Sen. Wilson). Second, the Governor of Texas, the Honorable William P. Clements, Jr., made clear to the legislature that he would veto any plan which did not include the "safe minority district" for Dallas. Dep. of Gov. Wm. Clements, at pp. 11-12; Exhs. to Dep., 3-8; Tr. at 280 (Sen. Mauzy).

These two sources of the impetus for a minority district in Dallas flow together in the legislative record. In open debate, each is referred to as a forceful consideration in resolving the difficult question of Dallas apportionment. Certainly, a coherent articulation of minority support for a given apportionment plan which would enhance their ability to participate in the electoral process is a valid source of guidance for the legislature. Yet a number of factors emerge from the legislative record which dissolve the clarity of minority support for the creation of a minority district in Dallas. These factors were made plain before the legislature, and hence its reliance on the expression of support advanced by blacks in Dallas must be studied closely. On the other hand, a threatened veto by the Governor of any plan that did not accord with his particular sympathies is of little independent weight as a valid legislative consideration. To the extent that the Governor's preferences are founded on a definitive legal principle, or, alternatively, are premises on an interpretation of popular sympathy, then his individual expressions deserve some weight. However, the source of that influence lies in the role the Governor might be playing at such a time, as an influential conduit of public opinion. In this instance, the record reflects that the Governor's support arose largely in a vacuum, *997 detached from close alliance with the preferences of Dallas residents. Additionally, other coincidental evidentiary factors serve to dilute the purity of the Governor's advocacy of minority interests. In sum, confused motivations and influences informed the legislative decision with respect to Dallas apportionment.

 
1. The coherence of minority support for S.B. 1.

For a number of reasons, the purported support of minorities in Dallas for a safe minority congressional district will not withstand close analysis. It is not questioned that a substantial number of minority Dallas residents supported the idea of creating a congressional district in Dallas, from which a black legislator might be elected. And it is beyond doubt that this desire was conveyed to members of the legislature. Yet the record reveals plainly that this expression of minority support was blended with countervailing sentiments articulated contemporaneously.

First, defendants in this action rely heavily on a petition submitted to the legislature on the letter-head of the Coaliton for Minority Representation. The petition is signed by a number of persons professing to be eligible voters in Dallas County, who, by virtue of their signatures, express their "support for the Coaliton's ... redistricting plans for the United States Congress." DX 86. The face of the petition reveals nothing more about the nature of these persons' support. As a source of substantive guidance, and as an expression of an intelligible political position, the petition is something short of wholly reliable. Initially, petitioning is itself a fractured and uncertain process. There is no record whatever of what the individuals signing the petition were told regarding what was being advocated, nor was there evidence as to what the signers may have subjectively believed they were supporting. In a matter as exceedingly complicated as apportionment, it strains credulity to suppose that the persons sponsoring the petition explained the issue in fashion which so much as hinted at its actual scope. One might be willing to suppose that the petition stands for the proposition that a number of voters in Dallas think that the congressional lines in Dallas should be drawn in a fashion that would enhance minority influence on the election. Alternatively, it is conceivable that the persons signing the petition thought that the election of a black member of Congress from Dallas would be a salutary step in the development of a black political presence in Dallas. Beyond these tentative and essentially irrelevant conclusions, little of substance may be deduced from the petitions.

Second, even if the validity of the petition as an expression of minority support for a minority district in Dallas is conceded, the nature of that support is seriously qualified. Allowing political groups and coalitons to exert influence by means of petitions may be one way of assuring broad public participation in a legislative procedure which fundamentally affects the nature of democratic life, but it is not a systematic nor fool-proof way. In the elaborately self-conscious process the legislature went through in articulating its priorities in redistricting and in establishing procedures, there was never a commitment to a systematic polling of popular sentiment. Of course, both the Senate and the House conducted extensive "out-reach hearings" at which voters across the state are allowed to testify, and to express their preferences with respect to the redistricting effort. Transcripts and summaries of these hearings were prepared. Yet this vast amount of "raw data" stands as a sort of oddity in the process, a largely ignored source of primary information about local concerns. See DX Nos. 52-56 (Transcripts of debates in closing days of special session).

The issue of validity of relying on the "wishes of the people" was presented, in a slightly altered context, in Graves v. Barnes, 343 F. Supp. 704 (W.D.Tex.1972.) In that case, the state attempted to justify its disparate treatment of Harris County in state legislative races, by reference to popular support for the established system of elections. The court said:

 
*998 The State argues that it was following the "wishes of the people" when it dealt with its metropolitan areas. The argument simply does not ring true ....
 
[T]here is no evidence in the record that the redistricting board had before it any coherent "wishes of the people" .... It appears to this Court that if a State elects to use the "wishes of the people" to treat metropolitan areas differently and unequally, then it is under some obligation to do a more thorough job of investigating the real "wishes of the people" .... And of course any rationale, including opinion polls, would have to be applied equally to all areas of the state, not followed for one and ignored for others.

343 F. Supp. at 721-22. As stated, the petition submitted by the Coalition for Minority Representation hardly constitutes a coherent expression of the "wishes of the people" of Dallas. Certainly the legislature is entitled to decide that local sentiment will play a role in the decision of how to apportion representation in a given area. Of course, it is plain that no such reference to popular will can supersede the basic constitutional constraints on reapportionment. Id. If all relevant mandatory factors implicated in the redistricting process were to balance in a manner that left the decision in some form of limbo, perhaps local preferences might be a legitimate determinative factor. However, once the legislature decides that its decision will be guided by popular will, the principle expressed in Graves v. Barnes is beyond contravention: public opinion must be tapped in a manner that is systematic, and which reflects that opinion more thoroughly than does a petition signed by a group of county residents under unknown and suspect conditions.

Third, the principle of consistency expressed in Graves v. Barnes is a corollary of the principle of thoroughness. If the legislature chooses to rely on the "wishes of the People" with respect to redistricting in Dallas, it must apply that decision to other areas, or offer a convincing justification for the failure to do so. The record reveals no other locality in which the legislature rested its decision on an expression of popular sympathy. Indeed, in most other areas of substantial controversy, the ultimate decision was based on highly rarified expression of local sentiment. For example, a crucial decision concerning Districts 21 and 23 relied heavily on the testimony of a local state representative, the desire of the incumbent to retain a military base in his district, and a Resolution passed by the City Council of the county seat of one of the counties affected. In no way are these considerations compelling. Indeed, their legal sufficiency will be discussed at length in another portion of this dissent. They are mentioned here only by way of contrast to the justifications offered by the state for its decision with respect to Dallas. This drastic inconsistency seriously undermines the sincerity with which the defendants defend their action as an honest assessment of local sentiment. Inconsistency in the application of substantive criteria is a separate evidentiary factor suggested by Arlington Heights, see supra at 979-980. The entire process of evaluating substantive concerns is marked by inconsistency and contradiction, which will be dissected later. For present purposes, it is sufficient to note that the protest of the state that it was relying on the "wishes of the people" has about it a strong aura of disingenuity.

Fourth, assuming that the petition submitted by the Coalition did accurately assess the feelings of the Dallas black political constituency, the record clearly reveals that the support for a minority district was contingent. Blacks in Dallas were generally aware of the degree of influence they had in congressional elections in both District 5 and 24. The evidence established that the black community was a necessary component of a successful political constituency, for purposes of election from Districts 5 and 24. DX 35.3. At no time did credible representatives of the black community support a congressional apportionment plan which would wholly eliminate that community's influence in one district for the sake of creating a safe black district. Representative Ragsdale, a highly visible supporter *999 of a minority district for Dallas, characterized the ultimate result as "extreme", and he personally found the result unacceptable. DX 55, at p. 165 (July 29, 1981).

Under S.B. 1, the black presence in District 5 is negligible. TR. at 248 (Sen. Mauzy). There was no black support for this result. Indeed, the most accurate expression of black sentiment concerning S.B. 1 is the attitude of black legislators toward the bill. The positions black legislators took in the final stages of debate on S.B. 1 were not based on an abstraction along the lines of creation of a safe minority district, with all the unanswered questions attendant upon such a vague concept. Instead, black legislators in the final days of the special session were confronted with an actual bill which created a minority district which was in no way a "safe" district, and, at the same time, rendered black influence in the other district a nullity. With one exception, all black representatives voted against the bill. TR. at 281 (Sen. Mauzy). Moreover, many of them spoke, with great clarity, about the deficiencies of S.B. 1, and made apparent to the legislature the effect of the bill on the black community. In order to accept the state's assertion that S.B. 1 was passed in homage to the expressed will of the minority community in Dallas, more faith must be placed in the highly suspect petition submitted by the Coalition than is placed in the statements of elected black representatives. This allocation of credibility is flatly at odds with the normal legislative procedure, and is starkly inconsistent with the manner in which other districts were treated. For instance, in the apportionment of Harris County, the legislature accepted, without question, Rep. Craig Washington's secondhand assertions of Congressman Mickey Leland's personal preference in terms of the boundaries of District 18; Rep. Susan McBee's representations of the desires of her local constituency with respect to Districts 21 and 23 were not only accepted as valid (despite considerable evidence to the contrary), but her unsubstantiated declarations were elevated above other concernments which were, in the hierarchy of issues established by the legislature, of a more fundamental nature.

Yet in considering apportionment in Dallas County, the support of minority representatives for certain plans was disregarded, in favor of assertions of minority interests made by persons whose connections with ethnic communities in Dallas County were, at best, tenuous. Several alternative plans for apportioning Dallas County were seriously debated during the Special Session. Two of these plans in particular gained widespread support among minority legislators, who apparently believed that the plans favorably affected the ability of minorities to participate in the political process. Senate Bill No. 3, also known as the McKnight-Mauzy plan, was, for a while, the leading bill in the Senate and the on which served as a vehicle for localized amendments. Sen. Mauzy, who co-sponsored S.B. 3, is a state senator from the Dallas area, whose reputation for responsiveness to minority concerns is widespread. Sen. Mauzy's contribution to S.B. 3 was an alternative proposal for the division of Dallas County. Specifically, S.B. 3 would have apportioned Dallas County in such a way as to increase minority influence in both Districts 5 and 24. Under S.B. 3, the minority population of District 5 would have increased from 29.2% to 31.9%; the composition of District 24 would have changed from 37.4% minority to 45.7% minority.

Sen. John Wilson submitted an amendment to S.B. 3 which, with only minor changes, became S.B. 1. At trial, Sen. Wilson offered two explanations for his dissatisfaction with S.B. 3. First, it divided his home county Williamson County among three congressional district, which was, to Sen. Wilson, "totally unacceptable." Tr. at 476. Second, Sen. Wilson did not believe that S.B. 3 adequately provided for minority participation in Dallas County. Tr. at 477. Sen. Wilson, a White from LaGrange, admitted at trial that Sen. Mauzy was in a much better position than he to evaluate Dallas County politics, as well as the focus of minority sentiment in that community. Tr. at 517. Nonetheless, Sen. Wilson drastically altered S.B. 3 with respect to Dallas *1000 County. His amendment did not win the support of minority representatives. Regardless, he remained undaunted in his zealous championing of minority interests in Dallas. In light of the vigorous opposition to S.B. 1 expressed by virtually all minority representatives, Sen. Wilson's continued assertion that he was merely enacting the will of the minority community in Dallas is either hubristic or in error.

In the latter stages of the debate, another plan the Smith-Ragsdale plan was presented for consideration. It had the support of the minority representatives, at least as a compromise measure to preserve some limited influence in the dominantly white district. The plan was seriously debated as an alternative to S.B. 1. The only substantial objection raised to the plan was pragmatic Governor Clements had promised to veto it. On that representation, the plan was defeated. Tr. 280. See also, Dep. of Gov. Clements, at 15-19. The considerations which were at the bottom of Governor Clements' support of a minority district in Dallas will be discussed hereafter. For present purposes, it is sufficient to note the extent to which this episode qualifies the connection between the expressions of the Dallas minority community and the district lines drawn by S.B. 1.

Fifth, even if the qualifications stated above were absent and the petition were accepted as a coherent expression of local sentiment, legitimately relied on by the legislature in drawing district lines in Dallas, Senate Bill No. 1 does not achieve the goal of a "safe minority district." The commonly accepted benchmark for a minority district is 65% minority population. The figure is used as a sort of triggering device by the Department of Justice in its review procedures under Section Five of the Voting Rights Act, e.g., United Jewish Organizations v. Carey, 430 U.S. 144, 152, 97 S. Ct. 996, 1003, 51 L. Ed. 2d 229, and was commonly used by members of the legislature in debate. DX 3, at p. 21; DX 37.3, at 13, Dep. of K. Shepardson, at 55. Of course, there is nothing talismanic about this figure. Minority representatives have been elected from districts with smaller minority percentages, and districts with more than 65% minority have elected white representatives. Yet it is apparent on the record of this case that the legislature thought the figure to be reliable.

Under Senate Bill No. 1, District 24 has a combined minority population percentage of 63.8%. Blacks constitute 46.5% of the residents, Hispanics 17.3%. Thus, the population is close to the established goal of 65%. However, there is no evidence in the record that District 24 is, in any meaningful sense, a "safe" district. Ms. Lucy Patterson (a Black with extensive experience in Dallas County politics) testified at trial that, in her opinion, a black citizen could be elected from District 24. Tr. at 445. On that thin thread of testimony, the state seeks to establish as a matter of fact that the legislature was successful in creating a safe minority district. However, Ms. Patterson has expressed ambitions to serve in Congress, and her admittedly self-serving statement will not bear that weight.

Several factors contribute to the conclusion that, despite its racial composition, District 24 does not guarantee the election of a minority representation. First, even though the lines of the district were substantially altered, District 24 still has a popular incumbent, the Honorable Martin Frost. As already mentioned, Rep. Frost was challenged in 1980 by a black Republican, Clay Smothers, and prevailed by a considerable margin, 61% to 39%. Rep. Frost has earned the broad support of his constituency in his two terms in office, and is generally viewed as responsive to minority concernments. Tr. at 234, DX 35.

Second, as the result of Rep. Frost's race against Smothers indicates, voting in Dallas County is not clearly polarized along racial lines. The overwhelming weight of the evidence adduced at trial indicated that racial considerations are in no manner determinative of voting patterns in elections in Dallas. Tr. at 234-37 (Sen. Mauzy); DX 35.5.

More important in analyzing the racial composition of District 24 is consideration of the premise involved in the characterization *1001 of District 24 as a minority district. In drawing this picture, defendants group together the black population and Hispanic population into one "minority" category, thereby demonstrating a fundamental insensitivity to the practical reality of minority politics. Coalitions between blacks and Hispanics are certainly part of the political landscape of Texas. However, such political bonding does not take place merely by virtue of skin color or ethnic identity. Rather, mergers of this nature are provoked by commonality and overlap of interest in concrete political situations. A specific history of successful political coalitions may create a climate, in a particular locality, which will foster further cooperation. Ethnic politics in Dallas, however, has been noteworthy for its lack of a legacy of black-brown coalitions. The testimony at trial indicated that blacks and Hispanics frequently find themselves at odds with one another in Dallas politics. Tr. 252-55 (Sen. Mauzy); Dep. of C. Washington, pp. 39-40; Dep. of K. Shepardson, at 51. In discussing Dallas politics, a political scientist told the state Senate that the Black presence in Dallas was of determinative importance in elections in that community. His remarks in no way included discussion of Hispanic influence. DX 35.3 at 1-3. Indeed, evidence adduced in this action indicates that the Hispanic presence in Dallas lacks the coherence of the black constituency, and that in their efforts to establish a political identity, Hispanics have frequently staked out positions quite separate from blacks.

Moreover, the legislature was aware of the strikingly low Hispanic registration and voting figures in Dallas County. Tr. at 237. Members had been informed, as a general matter, that these minority groups tended to register and vote at rates significantly lower than whites. Additionally, Black leaders in Dallas had presented figures concerning the actual rate at which Dallas minority citizens participated in the electoral process. DX 24.3, pp. 106ff. By the end of the special session, actual voter registration information was available which indicated that, in terms of registered voters, District 24 was, if at all, only marginally a minority district, even if black and Hispanic voters were lumped together. DX No. 1, App. C-6.

Finally, as a practical matter, black legislators did not support S.B. 1 because they did not believe that a black could win reelection from District 24. Representative Garcia testified at trial that in his judgment, "the 24th is not a winnable district for a minority." Tr. at 168-69. This judgment is based on consideration of the relatively low registration rates for minorities in Dallas, especially Hispanics, and the difficulty of forming effective political coalitions in that city.

Therefore, the actual result of S.B. 1 with respect to Dallas County is the creation of one district in which blacks will exercise unquestionably powerful influence in the election of a Representative, and one district in which they will have no effective voice. Under the 1973 apportionment plan, the Black community in Dallas exerts substantial influence in both districts, and in recent elections, have been the determinative constituency. This trade-off can scarcely be seen as benefiting the interests of black citizens in Dallas. The record contains no evidence of substantial support of this result. Opposition to it by credible and respected representatives of the Black community is forceful and was known to the legislature.

 
2. Governor Clements' support of a minority district.

This issue does not merit the elaborate analysis accorded the matter of Black community support for a minority district. Gov. Clements described his role in the reapportionment process as "minimal." Dep. of Gov. W. Clements, at 11. Indeed, he took only one position in the whole process: he insisted on the creation of "a Black-Mexican-American minority district in Dallas." Id. Governor Clements made clear to all persons concerned that he would veto any congressional reapportionment bill which did not include such a district.

*1002 This unequivocal position of Governor Clements is noteworthy for several reasons. Most obviously, his particular solicitude for the voting rights of Dallas minorities is unexplained. He testified that it was rooted in his historical knowledge of the community born of his childhood there. Dep., at 14-15. Yet, as Governor, he presumably has developed a substantial familiarity with various areas in the state which have substantial minority population concentrations. For instance, the Hispanic population of Harris County exceeds, in total numbers, the black population of Dallas County by 80,000 persons, DX. 40.D.5. The Governor does not have a special duty to assert the rights of his hometown, nor is there any legitimacy attached to special lobbying efforts undertaken under the impetus of the lingering memories of childhood.

Moreover, his sense that the existing fragmentation of the black community in Dallas was unfair, and so warranted remedial action was admittedly not based on any legal advice. Dep., at 13. Nor was it the product of studies he or his office had conducted concerning the extent and potential influence of Dallas minorities on the Congressional elections. Id. at 14. There is, in the record, a noticeable absence of any meaningful justification for Governor Clements' militancy with respect to the creation of a minority district in Dallas. And, of course, in the end, the "safe district" created by H.B. 1 is certainly not a majority black district, and, in practical terms, is not even a safe minority district. From his self-proclaimed position as advocate of minority interests, however, Governor Clements was ultimately satisfied by the racial composition of District 24, as drawn by S.B. 1. In his testimony, Governor Clements conceded that "[i]t may not be a majority district in the true sense of 50/50-51/50 or 51-49, but it constitutes what was acceptable to me as a minority district which would give the Blacks the representation that I felt like they deserved." Dep., at 20 (emphasis supplied).

This testimony has about it an acutely personal flavor, and to that extent is alarming, in that it tends to belie the altruism professed by Governor Clements and certain other advocates of the minority district. Governor Clements testified that he felt the fragmentation of the black community was "unhealthy" and "wrong." Id. at 36. When asked what fragmentation of minority voting strength meant to him, the Governor responded, "I think the minorities in this district as it is now constituted have been used by the liberal Democratic Party of Texas for their own purposes." Id. at 37. Governor Clements' answer evinces the extent to which his strong sense of the unfairness of the existing apportionment was entwined with his partisan political position as a leader of the Texas Republican Party.

Governor Clements' uncompromising stand on the creation of a safe minority district in Dallas was well-known to the legislature, and it emerges from the record as an immutable fact which dominated the entire process. Thus, lurking awareness of his conviction on the issue ultimately doomed the Ragsdale-Smith compromise plan, even though that plan varied only marginally in racial composition from S.B. 1, and had the support of black representatives whose familiarity with contemporary Dallas politics certainly matched that of the Governor. That Governor Clements would isolate the issue of Dallas County for special attention, and that his concern with that area should be expressed so stridently and single-mindedly, is singular. Upon initial inspection of the census data for 1980, together with an analysis of the changes that demographic changes had worked in existing congressional districts, Dallas would seem to present the least troubling picture of the major areas of controversy. The boundaries of the districts were the product of a court-imposed apportionment plan, and the racial composition of the two districts had changed only marginally in the intervening decade. Governor Clements never professed to be following anything other than his own interpretation of the desires of the minority community in Dallas, and his own sense of racial equity. Moreover, even the elusive well-spring of his commitment to minority representation extended only to *1003 Dallas. Minority residents of Houston, South Texas, and West Texas were denied the zealous advocacy of the Governor who, without explanation, limited his commitment to Dallas County. His entire endeavor, then, provokes misgivings as to his motives.

The impact of S.B. 1 on minority representation in Dallas County may be succinctly expressed: determinative swing-vote influence in two congressional districts was reduced to heavily determinative swing-vote influence in one district. The remaining district is so dominantly white, and so heavily conservative, as to obliterate not only any hope on the part of the minorities remaining in that district that they might influence the outcome of the congressional election, but also the possibility that residents might expect the representative to be responsive to their concerns. This effect was foreseeable, and was actually foreseen, by the legislature. Legislators familiar with Dallas politics kept their colleagues well apprised of the nature of neighborhood politics in Dallas, and of the effect any proposed apportionment plan would have on the congressional elections in that city.

The question remains, as to whether the foreseen effect was intended. As this discussion indicates, the proffered justifications for the final action taken are seriously open to question. They are riddled with inconsistencies, and giving consideration to the legislative record, are something less than compelling. In the apportionment process, Dallas was treated differently from the other major metropolitan area Houston and also from the rural areas which have cognizable minority communities. A later portion of this dissent will examine whether these substantive departures were justified. For now, it is sufficient to note that the effect of S.B. 1 in Dallas was a severe dilution of minority voting strength. The result achieved is wholly consistent with a deliberate, invidious purpose to discriminate against minority residents of Dallas, by packing them into one district and eliminating any influence they might have in another. There is, of course, no expression of this sentiment in the legislative record. However, the alternative justification proffered by defendants is subject to great doubt.

 
B. Harris County

Like Dallas County, Harris County is essentially a metropolitan county. It is dominated by the City of Houston. According to the 1980 census data, the total population of Harris County is 2,409,544. Of this total, 473,698, or 19.7%, are black; 369,075, or 15.3%, are Hispanic. Under the 1973 congressional apportionment plan, three entire congressional districts 7, 8 and 18 and portions of two others 9 and 22 are within Harris County. Districts 7 and 9 are not implicated in this action, and will be discussed only to the extent that the apportionment of the relevant districts tangentially affects them.

Houston and Harris County experienced substantial growth during the 1970's. In that period the county gained 667,632 residents, an increase of 38.3%. DX No. 40.C.5. Accordingly, significant changes were required in the congressional apportionment in Harris County. District 8 was 16% overpopulated, District 18 was 19% underpopulated, and District 22 was 35% overpopulated. DX. No. 40.C.6. Additionally, the racial composition of the county had changed dramatically. The Hispanic population had increased from 186,000 to 368,000, an increase of roughly 98%. By contrast, the Black population had grown 35% and the white population had grown only 30%. Indeed, the tremendous growth of the Hispanic community in Houston was one of the most notable facts in the political life of that area.

Under the existing congressional apportionment, District 8 is 40.1% minority (21.9% black and 18.2% Hispanic). Representative Jack Fields presently represents that district. District 18 is a heavily minority-populated district. Under the 1973 congressional plan, District 18 is 43.1% black and 30.9% Hispanic. Total minority population is 74%. Representative George Thomas "Mickey" Leland is the elected representative *1004 from that district. Representative Leland is a Black. District 22 is 31.2 minority, with 17.3 Black and 13.9% Hispanic. District 22 is represented by Representative Ron Paul.

The minority population in Harris County is concentrated in a small area in the center of the county. Though the entire area may be seen as a "minority community", such a broad description distorts the reality of the ethnic concentrations in Houston. The Black and Hispanic communities are, in large measure, distinct, and may be separately identified. The Black residents in Houston are concentrated in two separate clusters, which are connected in a manner that resembles an hourglass. In other words, one concentration of black residents runs northeast from a central point, and the other runs southwest from that same point, to form roughly an "X" shape. The Hispanic community runs essentially southeast from the point of intersection of that "X". Thus, for present purposes, it may be said that there are three separate minority communities in Houston two Black (north and south) and one Hispanic (east).

The 1973 congressional apportionment splits all three of these areas. District 8 includes portions of the northern Black community and a part of the Hispanic neighborhood. District 18 includes parts of all three communities. District 22 contains a southern part of the Black community in the south part of Houston. By virtue of this division, minority residents of Harris County receive widely varying levels of representation in Congress. Representative Leland is considered an excellent legislator by the minority community, enjoying a reputation as a forceful representative, vigilant in his advocacy of minority concerns. By contrast, minorities do not fare nearly so well in either Districts 8 or 22. Representative Fields is viewed as being not responsive to minority concerns. His opposition to school busing was a major feature of his last congressional campaign. Tr. at 189. Though his district as presently drawn is 40% minority, Representative Fields has been elected without the support of the Black and Hispanic residents of his district. Representative Paul is well-known as an extremely conservative congressman, and he is considered highly insensitive to minority issues, at least insofar as those issues might be addressed by the federal government. Tr. at 504 (Sen. Wilson). Black and Hispanic voters in District 22 effectively have no voice in Congress as a result of Representative Paul's conservative philosophy.

As a result of the enormous population growth in Harris County during the 1970's, the decision was made to place one of the new congressional seats allotted to Texas in Harris County. The new district was designated Number 25. As a result of the reapportionment, District 22 is no longer a significant part of the congressional apportionment of Harris County. In terms of minority representation, it is wholly irrelevant. Under S.B. 1, Districts 8, 18, and 25 provide representation for minority residents of the county.

A number of basic considerations formed the background of the reapportionment of Harris County. First, in light of the strong minority support for Representative Leland, and the obvious necessity of preserving a heavily minority district in Houston to avoid retrogression, maintenance of District 18 as a safe minority district was an unquestioned goal of the reapportionment process. The population of existing District 18 was 99,486 below the ideal population for a district following the 1980 census. Therefore, it was necessary that the lines of District 18 be altered to create a sufficiently large total population. Moreover, this addition of citizens to District 18 had to be done in a manner that did not drastically alter the racial composition of the district.

Second, once the basic decision to add a new district to Harris County had been made, certain important questions remained. The legislature was aware of the racial composition of Harris County, and knew of the substantial increase in the Hispanic population in the county. Under the existing apportionment, in each of the three districts in which there was a significant *1005 minority constituency 8, 18 and 22 Blacks outnumbered Hispanics. Though there is evidence that Blacks and Hispanics in Harris County are able to form effective political coalitions, the effect of the existing apportionment is to subordinate the Hispanic presence to the black presence or, at least, to dilute it into a general "minority" community.

Third, Governor Clements had made known his view that the Black community of Dallas was of sufficient size to warrant creation of a "safe" Black district in Dallas County. According to the 1980 census figures, there are approximately 287,000 Black residents of Dallas County. The census figures indicate that in Harris County there are approximately 369,000 Hispanic residents. In other words, there are 80,000 more Hispanics in Harris County than there are Blacks in Dallas County. On the basic terms involved in the debate over Dallas County, then, logic would dictate that the new district in Harris County be drawn in such a way as to provide Hispanics a "safe" district.

Senate Bill No. 1 does not create a "safe" Hispanic district. In fact, the proposed apportionment creates nothing more than two districts in which there is a significant combined minority presence, in addition to the preservation of Rep. Leland's safe district. Senate Bill No. 1 does accomplish one goal which seemed to be foremost in the minds of the legislators maintaining District 18 as a heavily minority district. The state defends S.B. 1 with respect to Houston by highlighting the great similarity between the boundaries of District 18 as drawn by S.B. 1 and the proposal submitted on behalf of Representative Leland by State Representative Craig Washington. See DX. No. 79. There is some question as to the legitimacy of considering preservation of an incumbent's district in the process of reapportioning legislative districts. However, to the extent that the statements of Representative Washington on behalf of Representative Leland may be taken as expressing the popular will of the Houston minority community, the consideration of those proposals, and the sincere attempt to follow them, have at least partial validity. In no event, however, does the correlation between District 18, as drawn by S.B. 1 and as suggested by Representative Leland, excuse any constitutional deficiency which may exist in the manner in which other districts in Harris County were drawn.

District 18 as drawn by S.B. 1 includes most of the minority communities included in the present boundaries of that district. There are four significant changes in the district, which may be explained by the twin necessities of including new census tracts to make up for the underpopulation of the district, and also to insure that the added tracts did not dilute the minority composition of the district. A thin, finger-shaped extension to the north-east of the district brings within District 18 a small black and Hispanic community. Additionally, a major extension to the north-west of the district brings within District 18 a heavily Black area. Further, a large portion of the Hispanic community, currently in District 8, is included in District 18. Finally, a small black concentration to the south of the district is, under S.B. 1, excluded from District 18 and placed in District 25.

The result of these changes is a slight decrease in the percentage of the minority population, from 74% to 72%. Under this apportionment, a minority candidate is likely to win election from District 18. Dep. of C. Washington, at 8-9. Indeed, the dominant feeling is that Representative Leland will almost certainly win re-election from this district, should he choose to run. Therefore, it appears that the legislature did accomplish one goal of the reapportionment process. However, a substantial number of minority residents are excluded from District 18, and so will not have the benefit of his exemplary representation of minority concerns. Under the 1973 apportionment, those persons are presently denied the benefit of having a responsive representative. Their relative powerlessness was an issue in the course of reapportionment. Taking into account its decision to place another congressional district in the Harris County area, the legislature had ample opportunity *1006 to remedy the existing denial of access to the legislative process.

Under the 1973 congressional apportionment, District 8 includes the northern part of the Houston Black community, on both prongs of the "X". Additionally, it includes a broad swath of the Hispanic community in the eastern part of the city. (District 8, as presently drawn, is highly irregular in its center-city boundaries.) See P-I M. Garcia, Exh. No. 3. DX No. 79. The total population is 40.1% minority.

The lines of District 8 are drastically altered by S.B. 1. The northwestern part of the Black community, and also the section of the Hispanic community to the east of the center-city, are removed from District 8 and placed in District 18. By virtue of this change, the minority population of District 8 is reduced from 40.1% to 29.2%. As stated, Representative Fields, the current representative from District 8 is generally considered non-responsive to minority concernments. Certainly, the minority population of the district does not exercise a decisive influence on the congressional election. Under S.B. 1, the minority population of District 8 is reduced by nearly 11 percentage points. It strains credulity to think that, by means of this reduction, minority residents of District 8 will be able to exert any substantial influence on elections in this district. Though Representative Craig Washington testified that it is possible, though not probable, that a Black or Hispanic candidate could be elected to Congress from this district, Dep. at 11, he offered no support for this somewhat illogical conclusion. Additionally, it should be noted that the composition of District 8 under S.B. 1 is 16.7% Black and 12.5% Hispanic. Whatever the nature of the combined political influence minority voters may exert in District 8, it is clear that Blacks outnumber Hispanics in the district. District 8 as drawn does nothing to cure the problem of submersion of Hispanic interests to Black interests. Even if the majority presence in District 8 is able to exert some influence in the election from that district, the leverage will derive from a coalition, within which the Hispanic voice will lose its distinctive characteristics.

The new congressional district in Harris County is designated as District 25. Under S.B. 1, it runs east and west across the southern part of the county. It thus includes a substantial portion of the southern part of the Harris County Black population. Additionally, it includes the southeastern tip of the Hispanic community. The total minority percentage in District 25 is 38.7%. Of this total, 25.0% is Black and 13.7% is Hispanic. There being no incumbent in the new congressional district, and since its contours are new-made, predicting its political future is intrinsically hazardous. As stated, Representative Craig Washington speculated, without expressing any reasons, that a minority candidate might possible be elected from the 25th Congressional District. Dep. at 12. What appears far more likely is that the minority community will exert influence in the election from that district, though the extent of that influence is, at present, beyond prediction. At most, minority residents of District 25 might hope to have determinative influence, in the sense that their support might be a necessary part of a successful trans-racial political coalition.

Again, it must be noted that Blacks out-number Hispanics in District 25 by almost two to one. The analysis of the effect of this disparity is substantially the same as that advanced above with respect to District 8. By means of this inequality, Hispanics are denied an autonomous voice in Harris County politics.

The overall effect of S.B. 1 on minority political participation in Harris County may be simply stated: Blacks and Hispanics are packed into one district the 18th to insure a continuing "safe" district; the remainder of the minority residents of Harris County are split between District 8 and District 25. Their influence in District 8, already of dubious quality, is seriously diminished, and they are not given potential swing influence in the new district. Under existing apportionment, there is one "safe" district, and there are two districts in which *1007 minorities exercise an indeterminate level of influence. The change in minority voting strength in Harris County may be demonstrated by means of the following chart, in which districts under the 1973 apportionment and under S.B. 1 are listed in order of the minority percentage:

 
1973 APPORTIONMENT                   S.B. 1
District 18        74% minority     District 18   72% minority
District  8        40% minority     District 25   39% minority
District 22[4]       31% minority     District  8   29% minority

When viewed simply in terms of population composition, S.B. 1 may clearly be seen as a dilution of existing influence on Harris County Congressional elections. All three Congressional districts suffer from a decline in percentage, albeit slight, when ranked in descending order, as in the above chart. Alternatively, if viewed geographically, District 25, which substantially replaces District 22 in south Harris County, provides an increase of 8% minority population, while the minority population of District 8 declines 10 percent. This disparity becomes especially egregious, when it is remembered that, during the 1970's, the Hispanic population of Harris County nearly doubled, and grew at three times the rate of the White population. The clear effect of S.B. 1 is to dilute and fragment the minority vote.

S.B. 1, and its insidious effect on minority voting strength in Harris County, are not the product of any necessary and unavoidable congressional apportionment. Not only were there issues of minority representation clearly presented to the legislature and the disastrous effects entailed by S.B. 1 made known, but minority legislators presented alternatives which would have avoided the dilution and fragmentation of S.B. 1, as well as meeting the other goals set out for the process. In particular, the "MALDEF-II" plan, presented by minority legislators on June 30, was a plausible alternative to S.B. 1, because it would have achieved the goals set forth by the legislature and would also have avoided the retrogression entailed by S.B. 1.

The "MALDEF-II" plan only slightly alters the lines in Harris County from the manner in which they are drawn in S.B. 1. However, by acknowledging the presence of minority communities of interest in a manner that is more sensitive to the dual reality of Black and Hispanic interests, it enhances the representation of both groups and avoids fragmenting their influence. The details of the plan will not be set forth here, since it is plain that the legislature is not under a constitutional obligation to adopt a specific plan. However, to the extent that an alternative plan was offered which avoids the pitfalls of the plan that was adopted, the alternative, and also the reasons for its rejection, become highly relevant.

Briefly described, "MALDEF-II" preserves District 18 as a heavily minority district, with 47.6% Black and 17.3% Hispanic residents, for a total minority percentage of 64.9%. Additionally, it creates a majority-minority district in District 25, which would be 29.4% Black and 30.8% Hispanic, for a total minority percentage of 60.2%. Minority population in District 8 would be reduced to 10.8%. These results are achieved by placing Black communities, which are presently in District 8 and so lack effective influence, into District 18, where they will be represented by Representative Leland or his successor, who will unquestionably be sensitive to minority interests. Including these additional Black areas in District 18 allows removal of the Hispanic communities presently in District 18 from that district, and reunification of these areas with the remainder of the Hispanic community in District 25. Therefore, under "MALDEF-II", the Hispanic community is not fragmented between two districts. Instead, it is placed in one district District 25 in *1008 which it is the dominant political presence. In combination with the black population of District 25, the total minority percentage approaches the level considered "safe". And if, as Representative Craig Washington suggests, a minority population of 38% in District 25 makes election of a minority candidate possible, certainly an increase of that percentage to 60% increases the chances considerably.

The state has offered three justifications for the rejection of "MALDEF-II" with respect to Harris County, none of which are convincing. Indeed, they are so unpersuasive as to have about them the air of pretext. First, the state argues that the plan was submitted too late for serious consideration. In fact, the plan is dated, by the state in its brief, as having been submitted on June 30, roughly six weeks before final passage of S.B. 1. The argument defeats itself. Second, the state contends that under "MALDEF-II", District 18 is irregularly shaped. Oddity of shape in such matters is, of course, relative, but a brief comparison of the 1973 apportionment lines in Harris County, the lines as drawn by S.B. 1, and "MALDEF-II" provokes the conclusion that any lines drawn in the county will seem bizarre, if viewed in sharp relief. The objection certainly is insufficient to justify the lines of S.B. 1, which are irregular and which serve to fragment communities of interest.

Third, and most interestingly, the state defends the rejection of "MALDEF-II" by arguing that, had that plan been adopted, the "safety" of District 18 would have been compromised, and, in fact, the decrease in the minority composition of District 18 might have constituted retrogression under the Voting Rights Act. These arguments are curious for several reasons.

First, the commonly used benchmark figure for a safe district was 65% minority. This figure was used with respect to districts in all areas of the state. There was no explanation as to why a 65% minority concentration in District 18 was not sufficiently safe. Second, in the past, a Black woman had been elected to Congress from District 18, when the racial composition of the district was only 60% minority. Third, notwithstanding the changes in the boundaries, District 18 had a strong, popular, respected incumbent, whose support in the district did not appear to be in any way undermined by the reapportionment entailed by "MALDEF-II".

The reasons proffered by the state as justifications for the rejection of "MALDEF-II" in favor of S.B. 1 with respect to Harris County do not withstand analysis, for they simply do not serve to explain the decision (1) to pack District 18 above the level generally considered safe and (2) to split the remaining minority population between two districts in a manner that leaves potential minority influence open to serious question. Neither the state's submission, nor the legislative record, present additional or alternative explanations for the legislature's decision. This absence of convincing explanation strengthens the inference that may be drawn from impact. To the extent that the impact was actually foreseen, the inference is rendered even stronger. In the case of Harris County, minority legislators made clear to the legislature the effect S.B. 1 would have on minority voting strength in Houston. In the face of this knowledge, the legislature, nevertheless, chose an apportionment plan that fragmented the voting strength of these minority citizens not in District 18 between two congressional districts. In doing so, it rejected a reasonable alternative, for reasons which are transparently inadequate as justifications. Additionally, it failed to apply to Harris County the one clear apportionment criterion that had guided the apportionment decision in Dallas County. On the basis of discriminatory impact and a legislative record which reveals that the impact was actually foreseen, and giving consideration to the fact that S.B. 1 was chosen in lieu of valid alternatives, its enactment, with respect to Harris County, warrants a strong inference of discriminatory intent.

 
C. South Texas

The area referred to as South Texas does not have the clear geographical demarcation *1009 offered by the boundaries of a single county, as was the case with Dallas and Harris counties. For purposes of this action, South Texas is defined by the congressional districts alloted to the southern part of the state under the various apportionment plans considered by the state legislature. Briefly described, South Texas, as defined in this action, includes the counties to the south of Bexar County and extends to the Southern part of the state. Additionally, counties along the Gulf Coast, running south from San Patricio County, are included in the region designated "South Texas."

Under the 1973 congressional apportionment plan, District 15 includes the southern-most counties of the state, District 23 contains the counties immediately to the south of Bexar County, and District 14 includes the northern Gulf Coast counties. The 1980 census figures revealed that the existing districts were malapportioned, in some cases to an extreme degree. For example, District 15 was 25% overpopulated. DX. No. 40-C.6. Additionally, a number of districts adjacent to District 15 had been affected by population shifts, and also would be affected by the required reapportionment of other, more remote districts. This "domino effect" required that the apportionment of South Texas be fundamentally altered. A variety of proposals were considered in the course of the legislative debate. A number of the early plans considered in the House of Representatives attempted to remedy existing malapportionment merely by shifting the lines of existing districts. However, from the outset, there was strong support in the Senate for a plan that would place one of the new congressional districts allocated to Texas somewhere in the South Texas region, as a means of alleviating existing malapportionment, and also to secure effective representation for the residents of that area. From the outset of the special session on July 15, 1981, there was no serious opposition to the placement of a new district in South Texas. The prolonged debate over apportionment in South Texas, then, focused on the manner in which the new district would be drawn, and how it would be dovetailed with District 15.

Under Senate Bill No. 1, District 15 includes the four southernmost counties of the state Zapata, Starr, Hidalgo and Cameron. The district runs east and west across the southern border of Texas. District 27, in Senate Bill No. 1, is placed directly to the north of District 15. It includes 16 counties, and runs along the Gulf Coast to San Patricio County, and inland north toward Bexar County.

For purposes of the ensuing discussion, the area of the state circumscribed by Districts 15 and 27, as drawn by S.B. 1, will be treated as a discrete region. Though, in any real sense, South Texas does not exist in such a formal way, such a treatment in this instance is proper. As will become plain, the dispute over the apportionment of South Texas essentially involves a question of the legitimacy of the choice made by the legislature between two alternative apportionment plans. Each of the two plans in question allotted two congressional districts to South Texas, existing District 15 and a new congressional district, designated District Number 27. The area defined by the two districts in each plan is identical. In other words, one plan might have been substituted for the other, without in any way altering the manner in which any other congressional boundaries are drawn.

The claim of the plaintiffs with respect to Districts 15 and 27 is relatively simple, and concentrates largely on the ethnic and racial characteristics of the two districts. Accordingly, the racial and ethnic composition of the area as a whole, and the manner in which it is divided by S.B. 1 should be set forth at the outset:

 
District    Population    % Hispanic   % Black   Total %
  15           526,803       80.36        .25      80.61
  27           527,341       52.90       2.97      55.87
Total        1,054,144       66.62       1.6       68.22

On the basis of these figures, the plaintiffs' claim may be summarized, as follows: The legislature chose to place a new congressional district in South Texas. That new district Number 27 and existing *1010 District 15 were placed together in the context of a statewide apportionment plan. The area encompassed by these two districts has a total minority population of 68.22%. In dividing the area into two congressional districts, the legislature created one district in which minority population exceeds 80%, and another district which is approximately 56% minority. The plaintiffs alleged that this choice concentrates an impermissible "packing" of minorities into one district, at the expense of their influence in the other district.

By way of contrast, plaintiffs argue that an alternative apportionment plan was available to the legislature which would have avoided these deleterious results. The Garcia-Luna amendment to S.B. 1 would have configured South Texas in a manner that would have produced the following composition:

 
District   % Hispanic    % Black   Total
  15           71.4         .4      71.8
  27           62.4        2.3      64.7

The plaintiffs argue that this apportionment avoids the packing of minorities entailed by S.B. 1, and so does not dilute the voting strength of minority residents of South Texas.

At the outset, it must be noted that the percentage of minority residents in District 15 is alarmingly high. The figure is far in excess of the 65% figure used by the legislature as a guidepost for the creation of "safe" districts. It is substantially greater than the next highest minority concentration (72% in District 18 in Harris County). This extreme deviation from the established standards for the apportionment did not escape the notice of legislators. Rep. Von Dohlen, Chairman of the Conference Committee on Redistricting, noted:

 
In a quick perusal of the districts, if you look [at District 15] at 80.36%, you immediately wonder about the packing concept.

DX. No. 30.3, at p. 5. Indeed, the record indicates that a concentration of 80% minority residents in a given district is, for practical purposes, prima facie evidence of packing. See, e.g., Dep. of Gov. Clements, at 64-65; Tr. at 385-86 (Douglas Harlan). The inference to be drawn from an inordinately high minority concentration in one district is especially strong when the "packed" district is adjacent to districts in which the apportionment has the effect of diluting potential minority influence. See DX. No. 4.4.11, at 200; DX. No. 4.4.4 at 652-53.

To a certain degree, then, the racially discriminatory impact of S.B. 1 on minority voting strength in South Texas is established on the basis of the racial composition of the districts. The state has not argued that District 15 is not "packed" with Hispanic voters, nor do they contest the conclusion that, had the Garcia-Luna amendment been adopted, the minority population of South Texas would have been divided more equitably between the two congressional districts. The issue then becomes one of foreseeability, and intent. The state, in essence, defends the action of the legislature by conceding the fact of disproportionate composition, but denying invidious purpose.

The state advances a number of justifications for the choice of the legislature to adopt the configuration of South Texas contained in S.B. 1, instead of the proposal submitted by Representatives Garcia and Luna. In light of the immoderate concentration of Hispanics in District 15 and the pernicious effect it will have on voting strength in South Texas, the issue of motivation is worthy of close attention.

First, the state argues that the configuration of Districts 15 and 27 were drawn in response to a suggestion by the Mexican-American Legal Defense and Education Fund (MALDEF). On April 28, 1981, Joaquin Avila, an attorney for MALDEF, testified before the Senate Subcommittee on Redistricting, in Austin. At that time, he presented, on behalf of MALDEF, a proposal for congressional apportionment, in which District 15 was drawn in precisely the same manner as under S.B. 1. DX. No. 21.4.2. This proposal has been designed "MALDEF I." Again, on May 31, MALDEF presented a plan to the conference *1011 committee, in which District 15 was drawn in the same manner. ("MALDEF II"). DX. 31.4.3. The state admits that, later in the process, MALDEF submitted yet another proposal MALDEF II, modified which drew Districts 15 and 27 in the same manner as the Garcia-Luna amendment. P-1 M. Garcia Exh. No. 16. However, the state appears to argue, MALDEF is bound by its initial suggestion for apportionment of South Texas.

This argument of the state fails to account for several things. The entire process of congressional apportionment is overloaded with amendments, deletions, and modifications. The number of times specific plans were changed to meet objections, and to pursue recently emergent concerns, is simply beyond calculation. In fact, throughout most of the legislative process, specific plans were viewed as vehicles for discussion, inquiry, and analysis, and were in no way considered inviolable. The state offers no justification for its implicit contention that this rule of flexibility is not applicable to MALDEF. Instead, the state argues, MALDEF may legitimately be held to its original characterizations of minority sentiment, however tentatively offered, and however much those sentiments might be altered by additional information and changing circumstance.

Moreover, the great respect with which the legislature treated MALDEF's original suggestion for South Texas was not accorded to MALDEF's proposals for other areas of the state. In fact, the legislative record reveals that MALDEF's plans for West Texas and Harris County were given only cursory consideration, and never gained the serious attention of White legislators. And to the extent that those suggestions were discussed, they were carefully dissected, and liberally amended under the force of competing considerations. No evidence was offered to the effect that MALDEF has an exclusive charter to represent the concerns of Hispanic residents of South Texas; and there is no apparent reason why its representations with respect to that region were treated with such deference, in contrast to the short shrift given MALDEF proposals elsewhere.

Second, the state defends S.B. 1 on the ground that, as drawn, District 15 is geographically compact. Moreover, the total population deviation of District 15 is only .03%. It should be noted, initially, that geographical compactness is a valid criterion for a congressional district only if other more fundamental priorities are met. If a district involves an impermissible racial gerrymander, compactness will not rescue the plan from that judgment. See Kirksey v. Board of Supervisors of Hinds County, 554 F.2d 139 (5th Cir. 1977) (en banc). More fundamentally, though District 15 as drawn by S.B. 1 is relatively compact, District 27 is neither compact nor regularly-shaped. By contrast, under the Garcia-Luna amendment, District 27 is geographically united, consisting of five contiguous counties along the Gulf Coast. Under that plan, District 15 becomes a more sprawling and extended district. The point to be made is not that one plan wins the award for relative compactness of districts, but that the nature of the population distribution in the counties contained in the two districts necessitated the creation of one relatively compact district, and one larger, more diffuse district. Nothing in the record, nor in logic, indicates that, given the necessity of this choice, the proper choice was to compress District 15. As for the numerical balance of the districts, Garcia-Luna achieves closer population equality than does S.B. 1 for Districts 15 and 27.

Third, the state argues that Hidalgo and Cameron counties, the two southern-most counties, have been in the same congressional district for over one hundred years. S.B. 1 preserves this "historic connection." On the other hand, Garcia-Luna divides those two counties, and places Hidalgo in the 15th and Cameron in the 27th. This argument has superficial appeal; however, it is deceptive. Most of the counties in South Texas have historically been in the same congressional district. The possibility that some of them might have to be separated from others does not appear to be the product of some sinister intention on the *1012 part of the sponsors of the Garcia-Luna amendment, but was the result of the decision to place a new congressional district in South Texas. S.B. 1 splits Jim Hogg, Brooks, and Willacy counties from Starr, Hidalgo and Cameron in a manner that disrupts the historical connection between all six of these deep South Texas counties. There is no reason to suspect that the connection between Hidalgo and Cameron is especially sacrosanct, nor that it should have been treated as the one bond that could not be broken.

Fourth, the state argues that a "community of interest" exists among the four southern counties, which warrants their preservation as a congressional "unit". This argument is deeply related to the third argument concerning historical connections. It is equally tenuous. The plea of "community of interest" was often sounded in the course of legislative debate. By the end of the process, mention of the concept provoked a great deal of skepticism. E.g., Dep. of K. Shepardson, at p. 22. Of course, the idea of an extant community of interest which unites the residents of a given area has considerable validity. It is particularly meaningful when used in the context of issues which especially concern a minority group, or a cognizable group united around one unique and acutely local concern, which distinctly separates those persons from others near them. The utility of the concept becomes attenuated when it is used to describe economic similarities or merely geographical boundaries. In any event, community of interest, as frequently used in the course of legislative debate, is not a consideration that operates on the same level of priority as a minority voting strength and other constitutionally mandated concerns.

The record does not contain any indication whatever as to the existence of compelling communities of interest which might dictate drawing the lines of the southern districts in a certain manner. The state contends that the division of the state into designated "planning regions" is probative of extant communities of interest. The concept of localized "planning regions" was developed in the late 1960's during the tenure of John Connally as Governor of Texas. The validity of the divisions with reference to any articulated purpose has not been established, and so the communities of interest identified by given regions are essentially curiosities in the context of the present action. However, several things about the "regions" may be noted.

Hidalgo and Cameron Counties are combined in the Lower Rio Grande Valley region. However, Willacy County is also in this region. The state does not explain why it finds the combination of the former two so telling, yet chooses to ignore the connection in this region between Cameron and Willacy, a bond that is severed by S.B. 1, but preserved by Garcia-Luna. Moreover, Hidalgo County is split from Starr and Zapata Counties, the latter being in the South Texas planning region.

Of additional interest in the inquiry into communities of interest are the divisions of South Texas created by the state legislative districts. DX. No. 73. The state senatorial districts are especially intriguing, because they more closely approximate the size of congressional districts. (The state senatorial districts as drawn in 1981 are currently in litigation in federal court. Accordingly, comparison here will be to the districts enacted in 1971.) Under the 1971 plan, Senatorial Districts 20 and 27 run parallel to one another, north and south through South Texas, in a manner that bears great similarity to the districts created by Garcia-Luna. To the extent that meaningful comparison may be made to the state representative districts, Districts 49 and 58 run north and south as well, and connect portions of the southern-most counties with counties to the north. For example, Starr County is in the same district as Jim Hogg, Brooks, Duvall and Jim Wells Counties. This grouping exists under Garcia-Luna as well. There is no reason to suppose that communities of interest, for purposes of state legislative districting, would be markedly different than for congressional apportionment. See Tr. at 125 (Rep. Garcia).

*1013 In sum, the argument concerning communities of interest is ultimately indeterminate. This ambiguity exists because the terminology is inherently vague, and susceptible to manipulation. Moreover, the placement of two districts in South Texas required that the counties be split in some fashion. Given that reality, reliance on communities of interest as a determinative factor would require not only that such communities be identified, but also that they be placed in some form of hierarchy, in order to establish which, among them, might most easily be severed. The concept simply is not amenable to such fine tuning. Finally, the evidence adduced in this action in no way indicates that S.B. 1 preserves extant communities of interest in any more coherent fashion than does Garcia-Luna. In light of this thoroughgoing muddle, reason suggests that the concept of "community of interest" be avoided as an explanatory device.

Fifth, the state defends S.B. 1 as a valid means of affording Hispanics in South Texas equal access to the electoral process. Specifically, it argues that S.B. 1 maintains District 15 in a form which will assure continued representation by a minority legislator, and that District 27 affords Mexican-Americans the opportunity to elect a person of their choice.

District 15 is currently represented by Representative Eligio "Kika" de la Garza, a Mexican-American. Representative de la Garza has served continuously in the House of Representatives since 1964. In the course of this tenure, he has been an extremely popular figure, as evidenced by the past three general elections, in which he has garnered 74%, 66% and 70% of the vote. There is no reason to believe, nor has the state contended, that the minority percentage of District 15 had to be raised to 80% to insure that Representative Garza, or someone equally sympathetic to Hispanic interests, would be elected.

Additionally, the state contends that District 15, as configured by S.B. 1, preserves a "continuity of representation," in that it retains in Representative de la Garza's district a group of voters who in the past have been part of his constituency. This assertion is accurate, but misrepresents the effect of the apportionment plans on District 15. Of necessity, the boundaries of District 15 had to be altered on the basis of the 1980 census figures. The district was 25% overpopulated. Therefore, it became necessary that approximately 130,000 persons who currently are represented by Representative de la Garza be removed from his district and placed in another district. As noted, both S.B. 1 and the Garcia-Luna plan accomplished that objective. Both plans move a number of counties currently in District 15 and place them in District 27. To that extent, both plans disrupt the continuity of representation that has existed during Representative de la Garza's tenure. In fact, of the counties currently included in District 15, S.B. 1 preserves only the four southernmost, and places the remainder in District 27. Of those four counties in which S.B. 1 preserves continuity, only one Cameron is moved by Garcia-Luna into District 27. And in exchange for that move, a large swath of Representative de la Garza's current constituency is returned to District 15. The state has not argued, nor does the evidence demonstrate, that the residents of Cameron County have a special attachment to Representative de la Garza, nor that those citizens' interest in continuity of representation is more worthy of consideration than the interest of voters in other parts of District 15, as currently apportioned.

District 27, of course, has no incumbent, and its political future is uncertain. Its minority population under S.B. 155.8%is far short of the level presumed sufficient to guarantee minority representation. Beyond doubt, the minority influence in District 27 will be considerable. However, the true nature of that influence is, under S.B. 1, uncertain. By contrast, the Garcia-Luna plan formed District 27 in a manner that created a 65% minority population precisely the level which, throughout the process, was taken as the benchmark figure for minority districts. Moreover, in the entire South Texas area, the minority population percentage is 66%. Far from distorting minority *1014 voting strength in the region, the Garcia-Luna plan closely approximates the Hispanic presence. Indeed, Garcia-Luna is a much more equitable means of apportioning South Texas, given the way the legislature chose to define that area.

Hispanics in South Texas do not have a constitutional right to representation in Congress by a minority legislator. The plaintiffs do not contend that the state legislature was under an obligation to apportion South Texas in a fashion that guaranteed Hispanics two representatives in Congress. However, once the South Texas region was circumscribed, Hispanics residing in that area have a constitutional right to a congressional apportionment plan which fairly and accurately reflects their voting strength. And once the state legislature decided to place two districts in South Texas, the presumption is that those two districts would accurately reflect the minority presence in the region.

As already noted, the region as a whole is 66% minority. Senate Bill No. 1 divides South Texas into two districts, one with 80% Hispanic population, and one with 55% Hispanic population. The impact of this apportionment on minority voting strength is plainHispanics are packed into one district, far in excess of any conceivable justification; and, as a necessary corollary of that packing, the Hispanic influence in the remaining district is diminished. The net effect of this apportionment is to dilute minority voting strength in South Texas. This effect was actually foreseen by the legislature. DX. No. 52, p. 248; DX. No. 55, at 107, 162-65; DX. No. 56 at p. 79.

Finally, the state has, throughout the apportionment process, evinced a serious concern for avoiding retrogression of minority voting strength, as defined under the Voting Rights Act. In the context of Harris County, it became clear that the legislature understood that constraint as prohibiting an apportionment plan which would decrease the minority percentage of any specific district. (See supra at pp. 1004-1005.) Assuming the propriety of this analysis, S.B. 1 presents a harrowing prospect for future apportionments. If, under the principle of non-retrogression, the state legislature cannot draw a district in a manner that would decrease that district's minority population, then, by virtue of S.B. 1, South Texas Hispanics are consigned indefinitely to a congressional district which is severely disproportionate in its Hispanic composition. Senate Bill No. 1, thus, enshrines an apportionment plan which has the clear effect of packing Hispanics into one congressional district, and thereby dilutes their voting strength. This impact is invidiously discriminatory, and it was plainly within the contemplation of the legislature when it adopted Senate Bill No. 1.

 
D. West Texas

As is the case with South Texas, the area referred to as West Texas is a geographically imprecise region. For purposes of this action, West Texas includes that portion of the state circumscribed by the 20th, 21st, and 23rd Congressional Districts, as delineated in S.B. 1. Briefly described, the region includes the San Antonio metropolitan area; Bexar County; the area to the west and southwest of Bexar County, extending west to the vicinages known as the Trans-Pecos and Big Bend; and south along the Rio Grande River to Laredo. Additionally, a number of counties to the northwest of San Antonio are included in this nebulous region.

Under the 1973 congressional apportionment plan, District 20 is wholly contained within Bexar County, and is substantially composed of the residents of San Antonio. District 21 includes the northern suburbs in Bexar County, and extends west and northwest from Bexar County to Brewster County and a portion of Jeff Davis County. District 23 includes the southern portion of Bexar County, and also a number of counties to the southeast and southwest. The district contains four border counties, along the Rio Grande River Valley.

The 1980 census figures revealed that each of these three districts had experienced sizable population shifts during the 1970's. District 20 in San Antonio was underpopulated *1015 by 20%; Districts 21 and 23 were overpopulated, by 29% and 14%, respectively. DX. No. 40.C.6. As a result of these changes and population changes in adjacent congressional districts, a thorough alteration of the congressional boundaries was required. Of particular importance was the decision to place District 27 in South Texas. This decision affected a number of the counties which, under the 1973 congressional plan, were included in District 23. As a result of the proposed placement of District 27, District 23, under the new apportionment plan, is substantially confined to the southwestern portion of the state.

All three of the congressional districts under discussion include portions of Bexar County, the dominant population center in these portions of Texas. A number of counties in West Texas are sparsely populated; the majority include fewer than 10,000 residents. By contrast, Bexar County is densely populated. Certain census tracts of Bexar County include more people than reside in entire West Texas counties. Because of this demographic fact, substantial alterations in the size and shape of West Texas districts are possible. The addition of large numbers of counties to one district may be offset by the removal from that district of a small number of Bexar County census tracts. In other words, the density of population in and around San Antonio provided the legislature with great flexibility in forming the boundaries of Districts 21 and 23. The extensive and heated debate over the apportionment of West Texas focused on the shape and racial composition of those two districts.

Under Senate Bill No. 1, District 21 remained substantially the same as under the 1973 apportionment. District 23, on the other hand, was severely truncated in its geographical scope. Under S.B. 1, District 23 includes a portion of Bexar County which resembles a horseshoe around the urban center of San Antonio; additionally, the district extends straight west from Bexar County to include Medina, Uvalde, Dimmitt, Zavala, Maverick, Kinney and Webb counties. The last three of these counties are on the border between Texas and Mexico. Senate Bill No. 1 creates districts with the following racial composition:

 
                                   Total
District   Black %   Hispanic %   Minority %
  20         8.8       61.7         70.5
  21         2.9       22.2         25.1
  23         4.1       53.1         57.2

The plaintiffs claim may be summarized easily. They assert that S.B. 1 fragments a coherent Mexican-American community which resides in the border counties that run along the Rio Grande River from Webb County at the southern end, north to the Trans-Pecos region, and west to the Big Bend, including Presidio County and Brewster County. S.B. 1 splits this minority community between Districts 21 and 23. District 20 is implicated, in that deficiencies in total district population, or in minority population, may be compensated for by alterations in the contours of District 20.

The plaintiffs claim that the western counties included by S.B. 1 in District 21 Presidio, Brewster, Pecos, Terrell, Crockett and Val Verde actually have more in common with the counties in District 23 than they do with the other counties in District 21. Specifically, plaintiffs assert that these western counties are distinguished by two dominant characteristics: first, they contain substantial and cognizable Hispanic communities; second, they are border counties. As such, plaintiffs argue, they have little in common with the other counties in District 21, which are predominantly white, and are not affected by the special concerns of border communities. The dominant political presence in District 21 is the northern suburban population of Bexar County, commonly referred to as the Silk Stocking area. This community is heavily white, and its political attention is focused on the issues that affect affluent white suburban citizens. The record indicates that these citizens were represented by Representative Thomas G. Loeffler, a white Republican. Representative Loeffler is generally considered responsive to the sentiments of his Bexar County constituency, often to the derogation of the political concerns of other *1016 residents of his district. In this situation, then, the votes of relatively poor, rural Hispanic residents of the western counties are effectively "wasted" in District 21. They constitute only 25% of the total population of the district, and the evidence indicates that they have been unable to gain the attention, much less the political advocacy, of Representative Loeffler.

The record establishes that the western counties have many socio-economic characteristics in common with the Rio Grande Valley counties. First, each of these counties contains an Hispanic population in excess of 40%. P-I, A. Garcia, Exh. No. 105; DX. 40.D.8. Second, this heavy concentration of Hispanic residents creates a substantial ethnic presence in these counties, which extends beyond mere population figures to the creation of a climate of political debate. The Hispanic communities in these counties tend to be politically active; hence, public issues are heavily influenced by minority concernments. P-I, A. Garcia, Exh. Nos. 28, 34, 51, 52, 70, 126; Dep. of F. Cervantes, pp. 40-42; TR. at pp. 341-377 (D. Harlan). Third, the counties are united by education and income levels, which in both respects tend to be alarmingly low. See P-I A. Garcia's Exh. Nos. 2-19; P-I M. Garcia's Exh. No. 24. Dep. of F. Cervantes, at pp. 42-46.

In this context, community of interest has great meaning, unlike the use of that term in the debate over South Texas. In West Texas, the western counties are substantially different, economically, demographically and politically, from the other counties in District 21. In such a situation, a coherent minority political presence can be wholly submerged by a dominant majority presence. Precisely this circumstance exists in District 21, as drawn by S.B. 1. Representative Craig Washington testified that the residents of the western counties in District 21 have two things in common with the residents of Silk Stocking community in Bexar County: first, they are human beings; second, they are in the same congressional district. Dep. of C. Washington, at p. 32. Laying aside this sardonic comment, a vast social and political gulf separates West Texas Hispanics from the remainder of the District 21 constituency.

The legislature was aware of the racial and ethnic composition of the counties involved in the apportionment of Districts 21 and 23, and also was apprised of the various localized issues which characterized the counties in question. DX. Nos. 40.C; 40.D; TR. at 28-50, 145 (M. Garcia); DX. No. 55, at pp. 103-119. It chose, however, to adopt an apportionment plan which effectively buried the Hispanic presence in West Texas beneath the overwhelming Anglo political presence in District 21. Alternatives existed which would have avoided this submersion. On a variety of occasions, plans were presented which would have combined the West Texas counties with the Rio Grande Valley counties in one Congressional district. TR. at 50. The salutary political effect of such a combination was made plain to the legislature. DX. No. 55, at pp. 103-119. Given the availability of this information, the choice to fragment the Hispanic community in West Texas is questionable and must be closely examined.

Before this analysis is conducted, an additional issue which is implicated in the apportionment of West Texas must be set forth. As drawn by S.B. 1, District 23 contains 57.2% minority population. In the debate over West Texas, this district is typically considered a "minority" district. Yet the minority population of the district is insufficient to warrant such an appellation. The percentage is well below the 65% figure; moreover, it appears that the legislature was apprised of the fact that actual minority voter registration in District 23, as delineated in S.B. 1, was 40%. Therefore, the legislature had actual knowledge that they were creating a district which, though superficially minority, was, in fact, quite the contrary. See, TR. at 103 (M. Garcia.); Dep. of F. Cervantes, at p. 122-123; P-I A. Garcia Exh. No. 75.

The alternatives mentioned previously which would have avoided the fragmentation of the Hispanic community inherent in S.B. 1 would also have substantially increased *1017 the Hispanic population of District 23. At various times, the legislature was presented with alternative apportionment plans which would have apportioned District 23 in a manner that included Hispanic percentages from 61 to 66%. See Dep. of K. Shepardson, pp. 44-54; Dep. of P. McKnight at 20, 29; TR. at 50 (M. Garcia); DX. Nos. 2.A; 2.D.4; DX. No. 22.41. The legislature was informed that, generally, these plans would increase actual Hispanic registration rates to approximately 50%. In this instance, then, the legislature had actual knowledge that its evaluation of the "safe" nature of District 23 was false; also, it had available concrete alternatives which would make the district actually safe.

On two separate, though deeply related points, the legislature apparently consciously chose an apportionment plan which worked to the detriment of Hispanic voting strength in West Texas. First, the Hispanic community in West Texas was fragmented by splitting it between two congressional districts; in one of these districts, the Hispanic presence is "wasted." Second, the one West Texas district which was created with an eye toward effective minority representation fails, in reality, to provide even a reasonable assurance of such a result. The reasons for such a conscious and calamitous choice must be explored.

The legislative record offers only the most tenuous rationale for the apportionment of West Texas entailed by S.B. 1. Essentially, two justifications emerge from the record.

First, as a historical matter, the counties in the Trans-Pecos extension have been included in the 21st Congressional district. Therefore, the state has argued, the 21st District as defined in S.B. 1 is a valid means of preserving this historic pattern. Moreover, the state contends, these counties have expressed a desire to remain in that district, and S.B. 1 does no more than enact the popular will of the residents of West Texas. This argument mischaracterizes the evidence.

The most vocal proponent of S.B. 1 with respect to West Texas was Susan McBee, a State Representative from Val Verde County. She strenuously opposed any apportionment plan which placed Val Verde County in District 23. See Dep. of P. McKnight, at 43, 47; Dep. of K. Shepardson at 23, 56; Dep. of S. McBee, passim. In arguing that Val Verde County should not be placed in the 23rd District, Representative McBee presented a resolution by the City Council of Del Rio, the county seat of Val Verde County, that Val Verde remain in the 21st District. DX. No. 71. However, this evidence does not substantiate the far-flung conclusion advanced by defendants, for several reasons.

Initially, the resolution of a city council of a county seat concerning what congressional district it would like to be placed in is of the most marginal weight in assessing the propriety of a legislative reapportionment plan. Moreover, Representative McBee's statements concerning the desires of her constituency are of some value, but should not be accorded determinative weight. Finally, and most importantly, Representative McBee's assertions arose in a context which renders them entirely inapplicable to the ultimate decision made by the state legislature. In opposing the inclusion of Val Verde County in District 23, Representative McBee was speaking in reference to an apportionment plan which moved the dividing line between District 21 and 23 by only one county: from between Kinney and Val Verde Counties to between Val Verde and Terrell Counties. In other words the plan opposed by Representative McBee would have separated Val Verde County from the other counties on the Trans-Pecos extension, by placing only Val Verde County with the Rio Grande Valley counties.

Representative McBee's opposition to this transfer was based on two related points: First, Val Verde had always been in the 21st District, along with the other Trans-Pecos counties; second, Val Verde County shared a community of interests with those counties. Dep. of S. McBee, pp. 7-18. The proposed plan contested by Representative McBee dissolved the historical and contemporary *1018 unity of those counties. For this limited point, then, Representative McBee's expressions are of great value. However, self-evidently, they do not support the decision made by the legislature to reject plans such as the Garcia-Luna plan, which preserved the Trans-Pecos unity by moving the entire region into District 23. Representative McBee's comments simply are inapplicable to that proposal.

The important issue, and the issue that informs Representative McBee's statements, is the existence of a community of interest in the entire Trans-Pecos area; and, after that community of interest is identified, the determination of whether it is more related to the interests of the Rio Grande Valley counties or the counties of District 21 is essential. All the credible evidence introduced at trial indicates that the prevalent community of interest runs along the Rio Grande River, from Presidio to Webb County. See P-I A Garcia Exh. Nos. 2-19; Dep. of C. Washington at p. 32; Dep. of F. Cervantes at 42-46; P-I M. Garcia's Exh. No. 24.

Second, the state contends that any plan which included all the counties of the Trans-Pecos extension in District 23 is unacceptable, because the resulting district would be too large. The issue of compactness has been addressed briefly in the previous section with respect to District 13. The principles advanced there are incorporated here by reference. Briefly put, the concern with size, while relevant, cannot justify an apportionment plan which is otherwise unconstitutional.

Moreover, several matters undermine the validity of the state's concern with size. In large measure, any district drawn in the western part of the state will be relatively large. This problem of size is a function of the sparsity of population of the region, and it simply cannot be avoided. Though precise figures are not in the record, it appears that District 21, as drawn by S.B. 1, is, in truth, substantially larger than District 23, as contained in the Garcia-Luna plan. Finally, the state has advanced only two instances in which a legislator's decision to reject the inclusion of the Trans-Pecos counties in District 23 was based on the size of the district. TR. at 93 (M. Garcia).

Other than these two fragile and ultimately unconvincing reasons, the legislative record and the record of this action are devoid of justifications for the adoption of S.B. 1, in lieu of an alternative which would have consolidated the Hispanic community of West Texas in a district which accurately reflected its voting strength. The state attempts, in its post trial submissions, to defend S.B. 1 on the ground that the legislature had no affirmative duty to adopt a specific apportionment plan advanced by the plaintiffs. This point, though valid, is irrelevant. The duty incumbent on the legislature is not produced by any particular plan submitted by participants in the legislative debate. Instead, it derives from clear constitutional imperatives, which prohibit a state legislature from adopting a plan which fragments an existing minority community, and thereby dilutes the voting strength of that community.

S.B. 1 is deficient with respect to this duty. The apportionment of Districts 21 and 23 fragments an existing minority community which resides along the Rio Grande Valley and the Trans-Pecos extension. By dividing this community between two congressional districts, S.B. 1 impermissibly dilutes Hispanic voting strength. The deleterious impact of S.B. 1 on minority voting strength was made plain to the legislature in the debate over West Texas apportionment. Alternative plans which would have avoided that consequence were presented throughout the process. The choice of S.B. 1 over these alternatives may fairly be said to have been conscious and deliberate. The justifications advanced are patently inadequate. The connection between foreseeable effect and intent becomes strong, in the absence of a competing, legitimate rationale.

 
IX. 
Procedural Irregularities

One of the factors listed by the Supreme Court in Arlington Heights v. Metropolitan *1019 Housing Corp., 429 U.S. 252, 97 S. Ct. 555, 50 L. Ed. 2d 450 (1977), to determine the presence of discriminatory intent in Fourteenth Amendment claims, was the presence of procedural irregularities in the process which produced the action or decision forming the basis of the complaint. While the defendants in the present civil action claim that the legislative process which resulted in the passage of the challenged congressional district plan was marked by no such procedural irregularities, the legislative record, as illuminated by deposition and trial testimony, shows the contrary. Abrupt departures from normal procedures, blatant irregularities, and obvious attempts to deprive the minority legislators and their allies of equal access to the process are starkly revealed in the record.

Prior to the start of the 1981 legislative session, the House created a new committee, designated as the Regions, Compacts, and Districts Committee ("House Committee"), whose appointed task was to hear testimony concerning redistricting plans, and, presumably, to formulate suggested plans for the consideration of the full House membership. State Representative Matt Garcia, of Bexar County, an intervenor in this civil action, volunteered to be placed on the redistricting committee; but he was placed on what he, perhaps unwisely, had denominated as his first choice, the Appropriations Committee. TR. at 23. Bexar County, whose county seat is the city of San Antonio, was the largest metropolitan area that did not have a representative sitting on the House Committee. TR. at 30. When this lack of representation came to his attention, Rep. Garcia personally requested Speaker Billy Wayne Clayton to name him to the committee. Speaker Clayton refused. TR. at 30. While this may not be a procedural irregularity, the lack of representation of Bexar County a major Mexican-American center, 1980 Census, DX. 40.C.6, on the committee indicates an insensitivity to the concerns of South and West Texas, and to their substantial Mexican-American communities.

In response to the criticism that certain areas of Texas were not represented on the House Committee, the House leadership created an Advisory Committee to the Regions, Compacts, and District Committee. TR. at 30. This committee had a diverse ethnic and regional makeup, but it had little or no actual power. The Advisory Committee met only two or three times, and may not have met at all after the crucial census information became available in March, 1981. TR. at 30-31. Thus, at the time that actual plans for carving the state into congressional districts were considered, the Advisory Committee had virtually no role.

The House Committee scheduled extensive hearings throughout the state during the autumn of 1980, and through the winter and spring of 1981. The Senate Subcommittee held similar hearings in the winter and spring of 1981. More testimony was heard by the House in late May and July. However, in drawing plans, the House Committee, and, specifically, Chairman Von Dohlen, did not make direct reference to definitive and persuasive testimony gathered at these hearings.

At the beginning of the May 11, 1981 session of the House Committee, Chairman Tim Von Dohlen laid out a complete plan of congressional apportionment for the committee's consideration. Although Chairman Von Dohlen stated that he had attempted to contact all committee members prior to presenting his plan to the committee in formal session, some committee members were not contacted. DX. 89, Tr., May 12, 1981, at p. 5. Von Dohlen stated that he had been unable to reach Rep. Doyle Willis and Rep. Ragsdale. However, Rep. Ragsdale, a minority member of the committee, stated that he "[had] not been hard to find." When asked who devised the redistricting plan which he offered, Chairman Von Dohlen refused to give an explicit answer, and, instead, stated that everyone, members and witnesses, had offered testimony and comments which were incorporated into the plan. DX. 89, House Committee Tr., May 12, 1981, at pp. 3-6, 51-54; House Committee Tr., May 13, 1981, at pp. 4-5; DX. 44, House Tr. May 18, 1981, at pp. 69-72. It is clear that no minority members *1020 of the House Committee had directly assisted Chairman Von Dohlen in the creation of his plan, nor had they known that he had been preparing it. Id., House Committee Tr., May 12, 1981, at p. 3; DX. 44, House Tr., May 18, 1981, at pp. 73-74.

On May 11, when he first presented his plan, Chairman Von Dohlen had not prepared accompanying demographic information ("demographics") showing the racial and ethnic breakdown of the districts, as required by H.R. 58. DX. 58. Although Rep. Bill Hollowell,[5] Acting Chairman, insisted that the demographics were available in the committee room during the evening of May 11, 1981, minority members who attempted to secure this crucial information were unsuccessful until the next day, in spite of diligent efforts. DX. 89, House Committee Tr., May 12, 1981, pp. 44 and 46.

Prior to the introduction of the Von Dohlen plan, several groups and various individual House members presented plans prepared during the many days of public testimony in the Spring of 1981. DX. 89, House Committee Tr., May 12, 1981, p. 51 (MALDEF plans); House Tr., May 18, 1981, pp. 36-37 (Republican Women's plans). Chairman Von Dohlen laid none of these plans before the committee when it convened in its formal session on May 11, 1981. Chairman Von Dohlen also refused to lay out a plan by Rep. Luna, creating two minority districts in Harris County, which Luna had offered to the Chairman on May 12, the same day that the demographics for the Von Dohlen plan appeared. DX. 44, House Tr., May 18, 1981, p. 78. The Chairman gave two reasons for withholding the Luna plan, viz., that it was submitted too late, and that the Chairman felt that his own plan complied with the Voting Rights Act. Id. The Chairman gave no reasons for his refusal to present other plans submitted by minority groups to the committee.

At the beginning of the May 12, 1981, committee meeting, Chairman Von Dohlen stated that he expected to conduct public testimony, and, on the same day, call for a vote on his recently proposed bill, even though the map of the plan had been available only since the beginning of the May 11 Committee meeting, and the demographics since the late evening of May 11. DX. 89, House Committee Tr., May 12, 1981, at p. 41. Indeed, certain minority groups claim that they did not receive the demographics until the following evening. Id., at p. 44. A representative of the Texas Rural Legal Aid, Raul Noriega, objected strongly to the committee's apparent expectation of intelligent discussion of the Von Dohlen plan, because of the delayed appearance of the required demographics, and he requested that the committee postpone consideration of the Von Dohlen plan until May 16. Id., at pp. 41-51. Noriega pointed out that refusal to allow minority comments on a proposed redistricting plan could be considered evidence of an intent to discriminate against Mexican-Americans. Id., at pp. 50-51. Still, Chairman Von Dohlen, Acting Chairman Hollowell, and Rep. Davis showed considerable hostility toward Noriega's objections, id., at p. 43 (Von Dohlen, from the floor); p. 50 (Davis); p. 59-60 (Hollowell, from the floor), and sought to push the proposal to an early vote.

In response to the comments of Noriega, Rep. Ragsdale made a motion to postpone the consideration of the Von Dohlen plan. Chairman Von Dohlen refused to recognize Rep. Ragsdale for the motion, stating that he, Von Dohlen, would not recognize motions until all testimony had been concluded. DX. 89, House Committee Tr., May 12, 1981, at p. 59. Because Chairman Von Dohlen would not agree to postpone consideration of his plan, minority members of the House Committee were forced to slip away from the meeting room, thus breaking the quorum. This tactic had the effect of blocking the chairman in his attempts to secure adoption of his plan from the committee at a time before adequate comment *1021 could be received, and prior to the formulation and presentation of amendments and substitutes. Since maps and demographics were required for the submission of amendments, plans, and complete substitutes, more than a few hours were necessary for the preparation of additional plans, especially since the delay between submission of the proposed plan to the computer and the final print-out of the requested information could take several days. Tr., pp. 142-143, 178-179. Therefore, if the Von Dohlen plan, with demographics, was available on May 12, 1981, it would have been extremely difficult to formulate major amendments or substitutes in time to present to the committee for consideration on that day.

On May 13, 1981, the House Committee again convened to consider the Von Dohlen plan. During the first portion of the meeting, the committee heard public testimony on the plan before the committee. Rep. Bush then presented a complete substitute plan, which affected only four districts in the Dallas-Fort Worth metroplex. The Bush substitute was adopted, and was further modified by an amendment, offered by Rep. Wright, pertaining to the Houston area. The substitute, as amended, was adopted. At that point, Rep. Berlanga raised a point of order, as to whether the amendments to the substitute could be offered. Chairman Von Dohlen stated that amendments would not be in order, but that complete substitutes could be presented. Rep. Berlanga indicated his intention to present a complete substitute the next day. Von Dohlen then stated that it would be his intent to convene the committee the next day at the noon recess of the House, but the committee was officially recessed until 4 p. m., May 14, 1981. DX. 89, House Committee Tr., May 13, 1981, at p. 65.

Chairman Von Dohlen convened his committee on the floor of the House, at noon, on May 14, 1981. There is no record of this meeting, unlike all other House Committee hearings. Rep. Berlanga, who had received Chairman Von Dohlen's assurance that he could present a complete substitute, was not able to present that substitute. Rep. Berlanga later stated on the floor of the House, that the Committee was "forced in [sic] the position of having to kick out a bill" at the May 14 noon meeting. DX. 44, House Tr., May 18, 1981, at p. 67. There apparently was no discussion and comment upon any substitute bill.

In contrast, the Senate procedure provided a model of deliberate consideration. The Senate counterpart for the House Committee prepared a detailed, informative packet of information for the proposed congressional district configuration, which it distributed to all Senators, and then refrained from consideration for five days, in order to allow members and interested citizens to assimilate fully the impact of the Senate proposals. DX. 22 and 23. The Senate as a whole did not rush consideration of the congressional redistricting bill. The House finally passed the amended version of H.B. 1400; the Senate had not yet acted, and the conference committee session did not begin until twelve days later. See generally DX. 2, IB, Senate Floor Session Tr. May 22, 1981; DX. 30, Conference Committee, Tr. May 30, 1981.

After the House and Senate conferees failed to reach an agreement on a congressional apportionment plan before the adjournment of the regular session, Governor Clements, on June 10, 1981, called a special session of the legislature to begin July 13, 1981. The primary issue on the agenda for the first session was congressional reapportionment, although several apparently unrelated matters were also included.

The Committee of the Whole Senate met on July 13, 1981, to begin consideration of the proposed congressional districts. DX. 34. The Senate approved a plan, S.B. 1, on July 21, 1981, and quickly sent the plan to the House. DX. 38.

The House Committee began hearing public testimony on July 16, 1981, and convened in formal session on July 27, for the purpose of reporting a plan to the House. DX. 53, House Committee Tr., July 27, 1981. At that meeting, Chairman Von Dohlen laid out S.B. 1, and then produced an amendment co-sponsored by Von Dohlen and several *1022 other representatives. Id., p. 3. The Von Dohlen proposal had not been presented during the earlier hearings, at which the public testified, and many members who were not co-sponsors had not seen the amendment prior to the time that it was presented for consideration. Id., pp. 7-8; DX. 55, House Tr. p. 199. The Von Dohlen amendment passed; but a plan presented by Rep. Washington, pertaining to the Houston area, failed merely by one vote. DX. 53, House Committee Tr., July 27, 1981, at p. 23. Consequently, Rep. Washington stated his intention to file a minority report on the Washington Amendment to the full House. Id., at p. 55.

When the House of Representatives met to consider the committee's proposal on July 29, 1981, Speaker Clayton made several rulings adverse to the minority cause. First, he ruled that the minority report submitted by Rep. Washington could not be considered. DX. 55, House Tr., July 29, 1981, at p. 30. Second, he twice ruled that Rep. Garcia's time for speaking expired, when Rep. Garcia was strenuously advocating his proposals for South and West Texas. Id., p. 23, and p. 112. See also, point of order called on C. Smith, p. 90. The speaker sustained points of order against Rep. Garcia, and thereby shortened the already meager debate on the Garcia proposals for the South and West Texas districts. Points of order pertaining to expired time were seldom raised against speakers who advocated amendments that did not benefit minorities. Id., p. 45, and fns. 10 and 11, supra.

The House version of S.B. 1 passed to third reading on August 11, 1981. Id., p. 193. However, many members spoke forcefully in support of a motion to re-submit the proposed redistricting bill to the House Committee for further work, emphasizing that the committee had considered the proposal for only one day, and that few House members seem to be satisfied with the proposal as it stood. Id., pp. 194-200, 208-209. The motion to re-commit passed, and the House Committee was thus relegated to the task of sending out an acceptable bill. Id., at 215.

The House Committee reconvened on August 5, and again on August 6, 1981, but failed to report a bill. DX. 53. The House itself was scheduled to meet to consider reapportionment on August 7; hence, the committee was under severe time pressure when it reconvened on the morning of August 7. Id., House Committee Tr., August 6, 1981, at p. 52. The committee toiled until the House itself met, at which time Chairman Von Dohlen allowed the committee to stand at ease while he attempted to persuade the House to allow the committee to continue its deliberations. Id., House Committee, Tr., August 7, 1981, at p. 33. The House agreed that the House Committee should continue to meet until it could report a bill, and the committee reconvened in a different room. Id., p. 34.

At this second meeting, Chairman Von Dohlen, under pressure, departed from his previously deliberative leadership style, and attempted to force a bill from the committee by means of his procedural rulings. By way of example, he refused to recognize Rep. Semos for a point of order, to the effect that a quorum was not present. Id., pp. 34-35. He did not act on Rep. Clark's suggestion that the committee should recess until all members were able to assemble, with their maps and papers, in the new committee room. Id., p. 38. The Chairman placed a call on the committee, and then allowed a vote to be taken on all amendments while the call was outstanding. Id., p. 41. Although these procedures may have been within the rules, they were certainly partisan, given the unquestioned importance of the subject matter under consideration.

In the second August 7 meeting, Chairman Von Dohlen not only rushed the committee to a vote, but he also refused to enforce H.R. 58, passed during the regular session. H.R. 58 was proposed by Rep. Von Dohlen, and co-sponsored by several committee members. DX. 58. H.R. 58 required all proposed plans of reapportionment to be accompanied by demographics and a map. By its own language, it applied to the legislative process, which presumably includes *1023 committee proceedings. Rep. Davis presented an amendment to the committee that was not accompanied by demographics; but when another committee member objected to the lack of demographics, the Chairman stated that Calendar Committee rules are not applicable in the committee process, and neglected to enforce H.R. 58. DX. 53, House Committee Tr., August 7, 1981, p. 47. If it was the intention of the chair not to put H.R. 58 in operation, he did not make his intention known to other members, in order that they, too, could have avoided computer delays or laborious hand calculations of demographic statistics. Members were thus forced to vote on the Davis proposal, without any knowledge of its exact effect on the racial and ethnic make-up of the affected districts. While Chairman Von Dohlen may have been technically correct in his failure to put into effect H.R. 58, enacted during the regular session, it is clear that the substantial benefits secured by the enactment of H.R. 58 would have greatly assisted the committee members in the special session, just as in the regular session. The Chairman at no time offered a reason for his departure from the committee's requirement.

On the afternoon of August 7, 1981, the Calendars Committee of the Texas House of Representatives, chaired by Rep. Susan McBee of Val Verde County, an opponent of the Garcia plan, DX. 55, House Tr., August 29, 1981, pp. 112-115; see generally, Dep. of Susan McBee, met to consider rules for the debate of the reapportionment scheme on the House floor. Late that afternoon, the Calendar Committee passed an unusual rule allowing only complete substitutes, not amendments, to proffered bills concerning apportionment. P-I M. Garcia, Ex. 19. Special rules are rare in the House, Dep. of Susan McBee, p. 19-20; Tr. at 76 (Rep. Garcia) and 593 (Speaker Clayton), but during the regular session, the Calendars Committee had passed rules requiring maps and demographics for both congressional and state apportionment plans presented in the House. P-I M. Garcia, Exs. 20 and 21. However, there is no contemporary record of the committee deliberations that produced the new, more stringent rule, and no Calendar Committee member or other representative articulated a reason for the passage of the rule on the House floor.[6] Speaker Clayton, in his testimony at trial, and Rep. Susan McBee, in deposition, stated that the rule was passed in an attempt to avoid inaccuracies in assignment of individual census tracts in some apportionment bills. TR. at 596-97; Dep. of Susan McBee at p. 155-59.

The new rule was enacted in violation of a standing House rule that requires special rules to be laid out twenty-four hours in advance of the session. TR. at 594. It was passed at some time after 4:00 p.m. on August 7, whereas the House convened prior to 2 p.m. August 8. Compare, P-I M. Garcia Ex. 19, with Tr., at p. 597. The lapse was twenty-two hours at the maximum. The eleventh hour change in the rules for the August 8 session gave minorities, and others who opposed the House Committee bill or S.B. 1, less time to prepare to offer alternative plans. Rep. Matt Garcia complained that even though the rule passed on August 7, he did not receive notice of its passage until the morning of August 8. TR. at 78. Given that the time lapse between submitting proposed plans to the computer operators and receiving complete maps and demographic information was often greater than twenty-four hours, TR., pp. 56-57, 142-143, and certainly greater than the time between the passage of the rule by the Calendar Committee and the opening of the August 8 session, the new rule substantially hindered the minority groups in preparing plans. While the obvious effect of the rule was to make the amendment process more onerous, the more important effect, discovered later, was that by allowing only complete substitutes, not amendments, any motion to call the previous question effectively shut off debate on *1024 all pending or as yet unintroduced substitutes, and forced a vote, in quick succession, on the substitute, and then on the bill as substituted. DX. 56, House Tr., August 8, at pp. 178-179.

The legislature convened on August 8, 1981, and Rep. Washington immediately challenged the rule requiring complete substitutes, on the basis that the Calendars Committee had no authority to suspend the rules of the House and to forbid amendments to the reapportionment bill. Id., at p. 2. Washington's point of order was overruled. Id., at p. 3. The rule was not challenged on the ground that it had not been laid out twenty-four hours prior to the proceedings to which the rule was to apply.

The first plan addressed by the House was the committee version of S.B. 1, which was laid out and briefly discussed by Chairman Von Dohlen. Immediately thereafter, Speaker Clayton left the chair and presented his complete substitute, with an accompanying map and demographic information. Id., at pp. 21-22. After he introduced his substitute, Speaker Clayton returned to the Speaker's chair to guide the debate on his own bill, which was a highly unusual occurrence. Id., at p. 79. The minority members were caught completely by surprise by this legislative maneuver. They had not had the opportunity to examine the Clayton plan, and were not prepared to present substitutes to it. The complete substitutes that the minority members had hurriedly assembled, by hand, after the Calendar Committee laid out its rule, were designed to serve as substitutes to the House Committee and Senate versions of S.B. 1, not the Clayton substitute. TR., at 602. Moreover, the computer could not spew out substitutes geared to the Clayton plan in time to present to the House on that day. TR., at p. 142-143. Indeed, the computer was so slow that, by the time the minority members became aware of the complete substitute rule, it was very difficult to employ it in drawing maps and computing demographic statistics. TR., at p. 142-143; DX. 50.2.C.

Minority representatives, and other representatives presenting plans and viewpoints favorable to the minorities, sustained defeats in attempting to secure time to present their viewpoints. The Speaker, apparently departing from normal procedure, refused to extend time for remarks; and a motion for the extension of time, which would have allowed each individual to fully debate the merits of redistricting proposals, failed by a very narrow margin. DX. 56, House Tr., August 8, at p. 17-18. Routinely, time was called on members who made persuasive speeches in favor of proposals supported by minority legislators. Id., at pp. 39, 44, 105, 129, 156, 165, 176, 181, 183 and 192. Those members opposing minority proposals overran their time but once. Id., at pg. 29.

Early in the August 8 proceedings, it became clear to several representatives that the rule requiring complete substitutes would work to the procedural disadvantage of those proposing alternatives to the Clayton substitute. After a relatively short debate of the Clayton bill, the Speaker recognized Rep. Messer to close. However, Rep. Bryant intercepted by raising a point of order. Id., at pg. 50. At that time, there were several complete substitutes pending on the desk of the Speaker, which he had not put before the House. Rep. Bryant wanted assurances from the Speaker that, by allowing Messer to close, he was not attempting to prevent consideration of those as yet undisclosed amendments, or substitutes, by the House. Bryant pointed out that the Speaker, in the past eight years, had never failed to show parliamentary courtesy to other members by refusing to present their amendments for the consideration of the entire House. Id., at 52. The Speaker curtly indicated that it was his prerogative to lay out amendments at his leisure, and refused to reply directly to Rep. Bryant's inquiry as to the intentions and effects of recognizing Rep. Messer to close. Id., at 52. Speaker Clayton also later declared his authority to decide the vote spread needed to entertain a motion to verify a vote. Id., at p. 199. Rep. Carlyle Smith then made a parliamentary inquiry *1025 as to the effect of the abandonment of the House by members who wished to see the amendments pending on the desk of the Speaker. Id., at 53.

After this apparent threat, the Speaker laid out several pending amendments, and allowed debate on all of them. Very late in the evening, and after considerable discussion, Rep. Messer moved the previous question. This had the effect of foreclosing debate on all motions still on the Speaker's desk; furthermore, it allowed only position closings on the substitute then before the House, i.e., the Clayton bill, and S.B. 1 itself. Id., at 177. At that time, at least one complete substitute remained on the desk of the Speaker which had not been presented to the House for debate and consideration, specifically, a substitute including a plan by Rep. Garcia. Id., at p. 197. Rep. Messer's motion to call the previous question passed the House, even though some representatives expressed a desire to address the plans still on the Speaker's desk. Id., at pp. 178, 180 and 197. Votes were taken on the substitutes and bills pending, and S.B. 1, as substituted, passed to third reading.

Normally, a bill would be laid out for third reading on the next day, so that members would have the opportunity to consider carefully the effect of the bill. TR., at p. 95. However, the House leadership obviously wished to act very quickly on final passage of the apportionment bill, for instead of waiting until the next day, the House recessed for a period of time, and then reassembled. Id. This procedure produced two legislative days within one twenty-four hour day. Id. In response to this maneuver, minority members of the House and their allies refused to return to the House chambers, hoping to break a quorum. TR., at pp. 95-97. The walkout delayed action on S.B. 1 for a time while security forces seized absent representatives. DX. 56, House Tr. at pp. 203-208. Despite police efforts, none of the minority members were apprehended, and only one or two of them may have returned to the House to vote against the bill, after it became obvious that a quorum had assembled without minority members. TR. at pp. 97-99. The bill then proceeded to passage.

 
X. 
Departures from Established Substantive Criteria

In the legal handbook prepared for the benefit of the Texas Legislature regarding the reapportionment of the congressional districts of the state, the legislators were cautioned:

 
This objective [of allowing minority groups some chance to elect candidates of their choice] should be applied systematically throughout the reapportionment process together with population equality and other objectively verifiable, legitimate state or local goals.

DX. No. 4.4.11, at p. 203; see also DX. No. 4.4.4 at 670-71. This advice is derivative of the basic legal principle that when a legislature identifies the principles which are to guide its policy choices, those principles must be applied consistently and pursued in an equitable manner. A violation of this tenet calls into question the legitimacy of the criteria, as well as the good faith motivation of the legislature.

A survey of the legislative record in this action reveals several major departures from established and articulated substantive criteria. A review of these instances will reveal the extent to which supposedly immutable standards were applied in an inconsistent and contradictory manner.

 
1. The black population of Dallas and the minority district.

Governor Clements testified that, in his judgment, the black community of Dallas County was of sufficient size to warrant creation of a congressional district in which that community could elect a representative of its own choosing. Dep. of Gov. Clements, at pp. 11-20. Governor Clements' unalterable view on this subject fundamentally changed the process of debate on apportionment of Dallas County. His rationale was independently adopted by Senator John *1026 Wilson, architect of the apportionment plan which ultimately became S.B. 1. TR. at 476ff. At a crucial point late in the legislative process, one plan which had received serious consideration was rejected for the explicit reason that, were it passed, it inevitably would be vetoed by Governor Clements. TR. at 280 (Sen. Mauzy).

The total Black population of Dallas County is approximately 280,000 persons. In Harris County, there are approximately 360,000 Hispanic residents. There are 80,000 more Hispanics in Harris County than there are Blacks in Dallas County. Yet the firm commitment of the Governor, and ultimately the entire legislature, to the substantive principle that a minority community of 280,000 persons deserved the political autonomy provided by a "safe" district was not applied to Harris County. In fact, Harris County was apportioned in a manner which makes the Hispanic population secondary to the Black population in each of the three Harris County districts. This striking deviation from an explicit substantive standard is unexplained.

 
2. The 65% standard for a "safe district."

The record of this action is replete with reference to 65% minority population as a benchmark for the creation of a "safe" minority district. E.g., Dep. of K. Shepardson, p. 16. In the ardent attempt of the legislature to create such a district in Dallas County, 65% combined minority population was the unquestioned target. TR. at 477 (Sen. Wilson); Dep. of Gov. Clements at 16. In three instances, this standard was ignored, in a manner which had a detrimental effect on the preservation of minority voting strength.

a. Houston In Harris County, District 18 was drawn so that its total minority percentage was 72%. The figure was justified as necessary to insure that the district would re-elect a minority legislator. No reason was advanced for the conclusion that an additional cushion of 7% minority population was required to insure the safety of the district. In fact, two factors militate in the other direction with respect to District 18. First, in the past, the district had elected a minority legislator when its racial composition was only 60% minority. Second, the incumbent congressman, Representative George Thomas "Mickey" Leland, is a popular figure in Harris County politics, and would not seem to need the additional margin provided in his district by S.B. 1.

b. Deep South Texas The 15th Congressional District, as drawn by S.B. 1, contains a minority percentage in excess of 80%. As a result, the adjacent district, Number 27, contains only 55.9% minority population. Had the 65% criterion been applied in South Texas, two "safe" districts could have been created, which would have accurately reflected the strength of the Hispanic presence in that region.

c. West Texas By virtue of an apportionment plan which splits a minority community between two congressional districts, the 23rd District has a minority population of only 57%. This percentage could have been raised by two alternative means. First, as explained above, the Trans-Pecos counties could have been placed in District 23, thereby increasing the minority percentage to 64.5% (and, not incidentally, unifying the West Texas minority community). Second, the district lines in Bexar County could have been altered slightly, to increase the minority population of District 23. See P-I M. Garcia Exh. No. 1.

 
3. Fidelity to Natural Boundaries

The legislature expressed its intent to adhere, as much as possible, to natural or historic boundaries in its apportionment. In West Texas, the failure to include the Trans-Pecos counties in District 23 has been justified by reference to the historical tradition that Val Verde County has, since 1917, been in District 21. (This fact, ultimately, is irrelevant, as explained above, pp. 119-121.) In South Texas, the refusal to split Hidalgo and Cameron Counties was explained by reference to the fact that those two counties had never been split by a congressional district boundary line. This argument distorts the reality of historical *1027 bonds between counties in deep South Texas. In Harris County, the rejection of proposals which would have increased the minority composition of District 25 was explained by reliance on the Houston Ship Canal as a natural boundary between those two districts. (What constitutes a natural boundary is, of course, problematic.) Yet, in Dallas County, the Trinity River, which is a natural and historic boundary between Districts 5 and 24, was no obstacle to the alteration of those congressional districts, in the single-minded pursuit of a minority district in Dallas County.

These inconsistencies and contradictions undermine the validity of the substantive criteria which have been employed by the state to justify the ultimate apportionment decisions reflected in S.B. 1. To the extent that their authority is confuted, they take on an aura of pretext. This character of suspicion provokes intense curiosity about the motivations that actually underlie the congressional reapportionment plan adopted by the Texas legislature.

 
XI. 
Conclusion

The dominant characteristic of Senate Bill No. 1 is its deleterious effect on minority voting strength in each of the districts challenged in this action. This conclusion emerges from an exhaustive examination of the impact of the apportionment plan on each area, with special attention to local detail and circumstance. This "intensely localized appraisal," White v. Regester, 412 U.S. at 769, 93 S. Ct. at 2341, is required by the complexity of the factual and legal issues implicated in this action. The rationale for the intricate analytical framework employed in this dissent has been elaborately set forth, and need not be reiterated. Essentially, it derives from the confluence, in this action, of two separate streams of social life which warrant the most heightened judicial attention: voting rights and racially motivated legislative action.

In addition to the close attention directed to the effect of the congressional apportionment plan, an intensive inquiry into legislative intent has been conducted. One of the disheartening lessons of recent constitutional history is the perfidy and guile with which those in power sometimes pursue impermissible objectives. Yet the Constitution "nullifies sophisticated as well as simple-minded modes of discrimination." Lane v. Wilson, 307 U.S. at 275, 59 S. Ct. at 876. Therefore, the available circumstantial evidence of intent has been thoroughly sifted.

The inferential connection between effect and intent is often plain, but insufficient to establish a constitutional violation. In the case under consideration, the inference to be drawn from impact is especially strong, in light of the elaborate legislative processes, which produced detailed knowledge of the effects of any plan under consideration before final action was taken. The inference is further reinforced by the combined consideration of the history and lingering vestiges of voting discrimination in the state, the blatant procedural irregularities which permeated the process, and the striking inconsistency with which established substantive standards were applied. Under prolonged scrutiny, these various evidentiary factors form a plain pattern of invidious purpose. In other words, in combination, the indirect evidence of intent presented in this action "ripen[s] into proof." Personnel Administrator v. Feeney, 442 U.S. at 279, n. 25, 99 S. Ct. at 2296 n. 25.

The danger of conducting the "intensely localized appraisal" mandated by the Supreme Court is that, in the course of this microscopic evaluation, the social and political reality implicated in voting rights litigation will somehow get lost. It is well to remember that, in commanding localized appraisal of apportionment plans, the Supreme Court recommended evaluation "in light of past and present reality, political and otherwise." White v. Regester, 412 U.S. at 770, 93 S. Ct. at 2341.

The fundamental reality involved in this action is the extent to which all citizens of Texas will be allowed to participate, on equal terms, in the process of democratic *1028 self-government. The promise of the Constitution is that all members of society are guaranteed equal protection of the law, without regard to race or national origin. At a minimum, this provision assures all persons of the right to participate in the processes of decision-making which affect their lives. The central component of this guarantee is the right to equal political participation. When this right is denied or abridged, as here, the premise of a democratic society is undermined. ROBERT M. PARKER, District Judge, concurring in part and dissenting in part: I concur with the opinions of my fellow Judges, Johnson and Justice, as they relate to the 15th and 27th Districts in South Texas. It is entirely appropriate for this Court to remedy, temporarily, those areas found objectionable under the Voting Rights Act so that congressional elections in Texas can proceed on schedule. I cannot, however, concur in the changes made in S.B. 1 which are not needed to remedy that defect. I also, therefore, concur with Judge Johnson's opinion in those areas of the State in which S.B. 1 is retained. I dissent from those changes made in the Dallas County area to districts 3, 5, 24 and 26. Those changes are not predicated on a Voting Rights objection or a finding that S.B. 1 is unconstitutional under City of Mobile v. Bolden, 446 U.S. 55, 100 S. Ct. 1490, 64 L. Ed. 2d 47 (1980).

In my opinion, the Supreme Court has set forth clear guidelines for courts required to make substantive intrusions into the reapportionment process. The Court has said that "reapportionment is primarily a matter for legislative consideration and determination, and that judicial relief becomes appropriate only when a legislature fails to reapportion according to federal constitutional requisites..." Reynolds v. Sims, 377 U.S. 533, 586, 84 S. Ct. 1362, 1394, 12 L. Ed. 2d 506 (1964), and that "redistricting and reapportioning legislative bodies is a legislative task which the federal courts should make every effort not to pre-empt...." Wise v. Lipscomb, 437 U.S. 535, 539, 98 S. Ct. 2493, 2496, 57 L. Ed. 2d 411 (1978); Connor v. Finch, 431 U.S. 407, 414-415, 97 S. Ct. 1828, 1833-1834, 52 L. Ed. 2d 465 (1977).

In this vein, the Court has indicated that a finding of unconstitutionality in a legislative redistricting plan does not suspend the deference owed the legislative will embodied in those portions which are not offensive. In White v. Weiser, 412 U.S. 783, 93 S. Ct. 2348, 37 L. Ed. 2d 335 (1973), the Supreme Court, in upholding a district court's determination of unconstitutionality in the Texas congressional districts plan prepared by the State, nevertheless, overturned the remedy selected by that district court. In substituting another plan, the Supreme Court stated "The District Court erred in [its] choice... It should have implemented [the plan] which most clearly approximated the reapportionment plan of the state legislature, while satisfying constitutional requirements.... The District Court's preferences do not override whatever state goals were embodied in [the State plan] ... Id. at 796, 93 S. Ct. at 2355. See also, Whitcomb v. Chavis, 403 U.S. 124, 161, 91 S. Ct. 1858, 1878, 29 L. Ed. 2d 363 (1971) ("The remedial powers of an equity court must be adequate to the task, but they are not unlimited. Here the District Court erred in so broadly brushing aside state apportionment policy without solid constitutional or equitable grounds for doing so."

This policy of judicial deference in fashioning equitable remedies recognizes the principle that such remedies should not exceed the scope of the violation for which they are required.[1] To allow otherwise would unjustifiably thwart the clear and legitimate will of the people.

*1029 Recently, the 5th Circuit has recognized and applied this principle in the context of voting dilution challenges such as ours. Addressing what was at that time felt to be the Constitutional test in this area, Judge Wisdom indicated that it was improper for a court, in remedying a constitutional violation, to draw a plan which sought proportional representation. "If a plan passes the dilution test, race is no longer an important factor.... [W]e are not legislatures." Marshall v. Edwards, 582 F.2d 927, 937 (1978) (citations omitted). Further, he stated "[P]roportional racial representation, though attractive, is an abuse of the district court's equitable discretion." Id. at 938.

Absent specific statutory or constitutional violations, any changes to S.B. 1 made in this Court's remedial plan on behalf of the minorities which brought this suit, are merely substitutions by this Court of its preferences for those of the legislature. This can only be understood as an affirmative attempt to enhance minority voting strength in the state beyond the requirements of the Voting Rights Act and the Constitution. Such actions run afoul of the premise that "a court is forbidden to take into account the purely political considerations that might be appropriate for legislative bodies." Wyche v. Madison Parish Police Jury, 635 F.2d 1151, 1160 (5th Cir. 1981).

I am aware that the Supreme Court has indicated that a court-ordered plan will be reviewed under "stricter standards" than a legislative plan. Such standards have only been developed in and applied to two specific aspects of court-ordered plans, neither of which are implicated in this instance.[2]

In those contexts, stricter standards are appropriate because such goals are essentially apolitical. Political ends, such as affirmative action for particular groups, should not and cannot be the subject of a stricter standard without the Court entering the legislative domain. As indicated, this is forbidden.

Further, the Supreme Court makes it clear that the reasons for such stricter standards are precisely those which require judicial deference to the legislative plan in those areas in which it passes constitutional and legal muster.

 
These high standards reflect the unusual position of federal courts as draftsmen of reapportionment plans. We have repeatedly emphasized that `legislative reapportionment is primarily a matter for legislative consideration and determination...,' for a state legislature 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. Connor, supra 431 U.S. at 414-15, 97 S. Ct. at 1833-34.

Thus, these stricter standards only come into play when a court must, in effect, legislate in the absence of legitimate state legislative pronouncement. They are applicable to the changes this Court must make to remedy the Justice Department's objections to the 15th and 27th districts. In the rest of the state, however, the interests which motivate these high standards can best be served by adopting the legislature's plan which has survived constitutional scrutiny.

*1030 In short, it is my opinion that the Justice Department's objection to the two South Texas districts requires this Court to draw an interim plan which corrects the Voting Rights Act violations which exist there so that congressional elections can be held on schedule in Texas. The Justice Department's objections are not, however, an invitation for this Court to redraw other districts according to our preferences when no constitutional violation has been or can be found. To do so is an unwarranted usurpation of the democratic processes of our society.

NOTES

[1] The constitutionality of the Voting Rights Act was affirmatively determined in South Carolina v. Katzenbach, 383 U.S. 301, 86 S. Ct. 803, 15 L. Ed. 2d 769.

[2] The order also postponed the filing deadlines for prospective candidates for the Texas State Board of Education, which is composed of one member elected from each congressional district established by law. Texas Educ.Code Ann.art. 11.21 (Vernon 1972).

The sixteen districts affected by the order were those districts directly affected by the constitutional and statutory challenges to S.B. 1. The remaining 11 congressional districts as defined by S.B. 1 were left unaffected in an effort to avoid intrusion upon the Texas election process any more than necessary.

[3] It should be noted that the Supreme Court "think[s] it clear that Congress intended to provide [jurisdictions covered by the Voting Rights Act] with an expeditious alternative to declaratory judgment actions." Morris v. Gressette, 432 U.S. 491, 97 S. Ct. 2411, 2420, 53 L. Ed. 2d 506 (1977). The unseemly delay, inattention, and inactivity of the office of the Attorney General of the United States provided the State of Texas with anything but an expeditious mechanism for seeking approval of its congressional apportionment plan. Additionally, it should be noted that the three-page objection letter transmitted by the office of United States Attorney General over four and one-half months after the original submission fails to demonstrate that a careful and detailed analysis of the State's submission was made.

[4] Defendants, in their answer to plaintiffs' amended complaints, denied that the mere existence of an apportionment of 27 congressional seats to Texas makes it impossible to utilize the existing 24 congressional districts. In support of their denial, defendants refer to 2 U.S.C. § 2a(c) (2), which provides that "if there is an increase in the number of Representatives, such additional Representative or Representatives shall be elected from the State at large and the other Representatives from the districts then prescribed by the law of such State."

Defendants, however, also indicated in their answer to plaintiffs' amended complaints that they agreed this Court should implement an interim plan of its own for the 1982 elections. This is consistent with the statements made by counsel for defendants at the February 9, 1982 hearing before this Court. At that time, defendants' counsel stated, "the State does not object to the court granting the amended motions or amended petitions and thereby assume jurisdiction of this matter. At that point in time the State maintains that the court does have the jurisdiction to draw an interim plan and the court should undertake that jurisdiction." Record, February 9, 1982, at 12. Counsel later stated, in response to a question from the Court regarding whether the State requested the Court to enter an interim plan, "Yes, Your Honor. It is on the holding of elections under the existing districts, not only are they malapportioned and cause problems on the one hand, but also they necessitate running three congressmen on an at-large basis state-wide which is not preferable." Id. at 13. This sentiment was echoed by counsel for defendants Upham and Clifford.

[5] The order of February 15, 1982 also postponed the filing deadlines for prospective candidates for the Texas State Board of Education.

[6] Article 3, Section 28 of the Texas Constitution establishes and explains the functions of another state body the Texas Legislative Redistricting Board. This Board is given authority, in certain circumstances, to apportion the State Senate and/or House of Representatives. Its authority, however, does not extend to congressional apportionment.

[7] At the outset, this Court expressly determines that the apportionment plan included in this opinion is a court-ordered plan, as opposed to one enacted by an appropriate body of the State of Texas. The State of Texas has not purported to enact an apportionment plan or exercise any legislative judgment since S.B. 1 was made legally unenforceable by the section 5 objection letter of the office of the Attorney General of the United States. Indeed, the representatives of the State of Texas have expressly asked that this Court fashion an interim plan. Although this Court incorporates much of S.B. 1 into its own plan, the completed plan represents the remedial directive of this federal court. As a court-ordered plan, no section 5 preclearance is necessary. See McDaniel v. Sanchez, 452 U.S. 130, 101 S. Ct. 2224, 68 L. Ed. 2d 124 (1981).

[8] This Court, throughout this opinion, makes reference to race or color in the context of the constitutional requirement that the Court implement a fair plan. This reference is made for convenience only. As will be noted in greater detail, this Court recognizes that it is obligated to follow the standards of section 5 of the Voting Rights Act of 1965. Section 5 contemplates protection against discrimination for not only racial minorities, but also language minorities. See 42 U.S.C. §§ 1973c and 1973b(f) (2). Accordingly, this Court recognizes that one of its objectives is to devise and implement a plan that is not only racially fair, but also fair to language minorities.

[9] The difficulties facing this Court in its effort to achieve its coterminous goals was precisely stated by the Fifth Circuit Court of Appeals in Kirksey v. Board of Supervisors of Hinds County, Mississippi, 554 F.2d 139, 152 (en banc):

Achieving one-man, one-vote political democracy without excluding minorities from political life is a complex task that challenges the best of intellects and requires examining many facets of the community, past, present and future. The problem is not susceptible of simplistic solutions, however seductive they may appear. No mechanistic solution is an alchemist philosopher's stone that will turn all the problems of the past and present to future gold.

[10] This Court, in fashioning its remedial plan, relies in large part on the Supreme Court's decisions in cases involving the apportionment of state houses or senates. In doing so, however, the Court acknowledges that the Supreme Court applies stricter standards when evaluating the legitimacy of a congressional apportionment plan than it does when evaluating a plan that apportions a state house or senate. Mahan v. Howell, 93 S. Ct. at 983 citing Kirkpatrick v. Preisler, 394 U.S. 526, 89 S. Ct. 1225, 22 L. Ed. 2d 519 (1969); Gaffney v. Cummings, 412 U.S. 735, 93 S. Ct. 2321, 2325-26, 37 L. Ed. 2d 298 (1973). This Court, accordingly, adjusts for the stricter standards when applying these teachings.

[11] The Supreme Court has also noted in the context of evaluating legislative plans:

The equal protection clause requires that legislative districts be of nearly equal population, so that each person's vote may be given equal weight in the election of representatives.... It was recognition of that fundamental tenet that motivated judicial involvement in the first place in what has been called the "political thicket" of legislative apportionment. Baker v. Carr, [369 U.S. 186] 82 S. Ct. 691 [7 L. Ed. 2d 663] (1962).

Connor v. Finch, 97 S. Ct. at 1834.

[12] To illustrate, if hypothetical Congressional District A has a total population of 100 people and Congressional District B has a total population of ten people, persons living in Congressional District A are underrepresented, while those living in Congressional District B are overrepresented. This is because both districts are allowed a single representative. Since the solitary congressman from District A represents 100 people, the voters in his constituency are casting ballots worth 1/10 the value of those cast by the people living in District B.

If Districts A and B were the only two congressional districts in the state, the ideal population for a congressional district would be 55 people. As a result, District A would have a variance of +45 people from the population norm. It would have a deviation of +81.82%. District B would have a variation of -45 people from the population norm and a deviation of -81.82%.

[13] The de minimis variation requirement is an outgrowth of the stricter standards a federal court faces when it prepares its own remedial decree. The Supreme Court "has tolerated somewhat greater flexibility in the fashioning of legislative remedies for violations of the one-person/one-vote rule." McDaniel v. Sanchez, 101 S. Ct. at 2230. See also Mahan v. Howell, 410 U.S. 315, 93 S. Ct. 979, 35 L. Ed. 2d 320 (1973).

[14] In Wyche, 635 F.2d at 1158, the Fifth Circuit noted that "`[u]nder 10% deviations ... [have been] considered to be of prima facie constitutional validity ... in the context of legislatively enacted apportionments," quoting Connor v. Finch, 431 U.S. 407, 97 S. Ct. 1828, 1835, 52 L. Ed. 2d 465 (disapproving a district court's apportionment of a state legislature with 16.5% deviations in the state senate districts and 19.3% deviations in the state house of representatives districts). See Gaffney v. Cummings, 412 U.S. 735, 93 S. Ct. 2321, 37 L. Ed. 2d 298 (1973) (upholding state apportionment of state senate and house of representatives with 7.83% total deviation from population norm); White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973) (upholding state apportionment of Texas State House of Representatives with 9.9% total deviation from population norm).

While the de minimis threshold that this Court must satisfy in implementing a congressional apportionment plan is undefined and uncertain, this Court recognizes it is more stringent than the requirements faced by a state legislature that is seeking to impose an apportionment scheme. It is also more stringent than the requirements faced by a body seeking to implement an apportionment scheme for a state house of representatives or state senate.

Kirkpatrick v. Preisler, 394 U.S. 526, 89 S. Ct. 1225, 22 L. Ed. 2d 519 (1969) and Wells v. Rockefeller, 394 U.S. 542, 89 S. Ct. 1234, 22 L. Ed. 2d 535 (1969) are considered the leading cases regarding state apportionment of congressional districts. These cases invalidated state apportionment statutes providing for congressional districts having total deviations from the population norm of 5.97% and 13.1% respectively. The Court held in those cases that the Constitution permits only those population variances among congressional districts that "are unavoidable despite a good-faith effort to achieve absolute equality, or for which justification is shown." Kirkpatrick, 89 S. Ct. at 1229. See also White v. Weiser, 93 S. Ct. at 2352. The Supreme Court has indicated that, in the context of state plans, the state must make a good faith effort to achieve precise mathematical equality. In the context of court-ordered plans, the Court "must precisely articulate the interests which justify any deviations from [equal population] standards." Chapman v. Meier, 95 S. Ct. at 766.

[15] Of course, § 5's reference to the guarantees set forth in § 1973b(f) (2) makes it necessary that the appropriate parties are satisfied the apportionment plan does not have a discriminatory purpose or effect regarding language minorities.

[16] It is, of course, the district court's purpose that must be analyzed and not the purpose of the state legislature in proposing possible remedies, although the court may incorporate portions of the legislature's proposals. As it applies to the case sub judice, the State's proposal, S.B. 1, is legally unenforceable due to a § 5 objection of the United States Attorney General. When the Attorney General objected, this Court no longer was in a position to examine the legislature's purpose in enacting the legislation. As a result, the only purpose to be examined is the purpose of this Court in implementing its own plan.

It might be argued that this Court should examine the purpose behind the State's enactment of the apportionment plan that has evolved into its proposed remedy in order to avoid the possibility of aiding the State in discriminating against minorities by doing indirectly what the State could not do directly. While this Court will discuss the § 5 effect standard in greater detail, it should be pointed out that the first element of a plaintiff's claim that an apportionment plan constitutes an impermissible racial gerrymander in violation of the Constitution is to demonstrate the challenged legislation had a disproportionate impact or some injury in fact.

It is clear a demonstration of disproportionate impact or injurious effect, standing alone, is normally insufficient to establish a constitutional violation. City of Mobile v. Bolden, 446 U.S. 55 at 67, 100 S. Ct. 1490 at 1500, 64 L. Ed. 2d 47 (1980); Village of Arlington Heights, 429 U.S. 252, 97 S. Ct. 555, 50 L. Ed. 2d 450 (1977); Washington v. Davis, 426 U.S. 229, 96 S. Ct. 2040, 48 L. Ed. 2d 597 (1976). But see Yick Wo v. Hopkins, 118 U.S. 356, 6 S. Ct. 1064, 30 L. Ed. 220 (1886); Gomillion v. Lightfoot, 364 U.S. 339, 81 S. Ct. 125, 5 L. Ed. 2d 110 (1960). However, an important starting point and necessary element for those complaining of improperly motivated actions by the legislature is to demonstrate the legislation has the impact of diluting minority voting strength.

This Court has no intention of incorporating any part of a state proposal that has a disproportionate impact or injurious effect on minorities. It, therefore, has no intention of incorporating any part of a state proposal that could be subject to a valid constitutional challenge that it dilutes minority voting strength. Accordingly, there is no possibility that a purposefully discriminatory plan may be implemented indirectly.

[17] This Court in no way intends to indicate that it has determined the State of Texas acted with a malevolent purpose in passing S.B. 1. That question is simply not before this Court. This Court seeks only to demonstrate that if the district court possesses proper objectives in the creation of an apportionment plan, it satisfies one of the § 5 standards.

[18] The plurality in City of Mobile stated, "The Fifteenth Amendment does not entail the right to have Negro candidates elected." City of Mobile, 446 U.S. at 65, 100 S. Ct. at 1498-99. The plurality also held that the "right to equal participation in the electoral process does not protect any `political group,' however, defined, from electoral defeat." Id. at 1505. The Court's decision in Whitcomb v. Chavis has been described as standing for the proposition that:

The people of the ghetto were entitled to participation in the political process and to representation as people, and the record did not show that they had been denied that. But they were not entitled to recognition and representation as poor black ghetto dwellers, nor on account of any other characteristic that might be typical of their race.

Taylor v. McKeithen, 499 F.2d at 905. In Whitcomb v. Chavis, the Supreme Court rejected the district court's holding, which it characterized as "expressive of the more general proposition that any group with distinctive interests must be represented in legislative halls if it is numerous enough to command at least one seat and represents a majority living in an area sufficiently compact to constitute a single member district." 91 S. Ct. at 1875.

[19] See Note, Group Representation and Race-Conscious Apportionment: The Roles of States and the Federal Courts, 91 Harv.L.Rev. 1847, 1855 (1978).

[20] Since "the realities of partisan politics" impact upon whether a district can even be characterized as a safe minority district, Wyche implicitly dictates that the federal courts faced with fashioning a remedial decree should not involve themselves in the question of bloc voting. As a result, the court is not allowed to make the assumptions necessary to create a "safe" district.

[21] It has been pointed out that remedial measures aimed at proportional representation in the election process cannot be equated with remedial racial programs in other contexts by virtue of the numerous extraneous factors involved in the political arena. Wyche v. Madison Parish Police Jury, 635 F.2d at 1161; Marshall v. Edwards, 582 F.2d 927.

Without question, this Court deplores the fact of life that some people choose between candidates because of the color of skin. On the other hand, great care must be taken to avoid any so-called remedial measure that, instead of aiding the integration of races, may perpetuate a separation or severance by advocating that one race is entitled to be represented by a member of that race. It has been said that

Racial electoral registers, like religious ones, have no place in a society that honors the Lincoln tradition "of the people, by the people, for the people." Here the individual is important, not his race, his creed, or his color. The principle of equality is at war with the notion that District A must be represented by a Negro, as it is with the notion that District B must be represented by a Caucasian, District C by a Jew, District D by a Catholic, and so on.... The racial electoral register system weights votes along one racial line more heavily than it does other votes. That system, by whatever name it is called, is a devisive force in a community, emphasizing differences between candidates and voters that are irrelevant in the constitutional sense. Of course race, like religion, plays an important role in the choices which individual voters make from among various candidates. But government has no business designing electoral districts along racial or religious lines.

Wright v. Rockefeller, 376 U.S. 52, 84 S. Ct. 603, 611, 11 L. Ed. 2d 512 (1964) (Douglas, J., dissenting) (citations omitted). See also Regents of University of California v. Bakke, 438 U.S. 265, 98 S. Ct. 2733, 2783-84, 57 L. Ed. 2d 750 (1978) (Brennan, J., concurring); City of Mobile, Alabama v. Bolden, 100 S. Ct. at 1511 (Brennan, J., concurring); Carey, 97 S. Ct. at 1020 (Burger, C. J., dissenting); Taylor v. McKeithen, 499 F.2d at 909-910.

[22] "Effective participation does not mean the right to have members of one's race, sex, or group elected to political office." Lodge v. Buxton, 639 F.2d at 1374.

[23] This is the clear lesson to be learned from the Supreme Court's admonitions against a court utilizing multimember districts when fashioning apportionment plans. The warnings are a result of the Supreme Court's view that multimember districts "tend to submerge electoral minorities and overrepresent electoral majorities." Wise v. Lipscomb, 98 S. Ct. 2493, 2497-98 n.5 and cases cited therein. See also Whitcomb v. Chavis, 91 S. Ct. at 1969 quoting Fortson v. Dorsey, 379 U.S. 433 at 439, 85 S. Ct. 498, 501, 13 L. Ed. 2d 401 and Burns v. Richardson, 384 U.S. 73 at 88, 86 S. Ct. 1286 at 1294, 16 L. Ed. 2d 376.

[24] Such a formula conceivably could be fashioned and applied by a legislative body facing the responsibility of apportionment. That body may consider the full panoply of political factors involved. This Court has no such option.

[25] Of course, the Court cannot lose sight of the fact of life that districts characterized as "safe" enhance the access of minority group members to a greater degree than do districts in which minority group members have only a degree of impact such that they may be able to affect the outcome of an election.

[26] A district containing a minority population of sixty-five percent or greater is generally recognized by the Justice Department as a district capable of electing a minority representative. See Beer. As a result, an apportionment change that reduces a district's minority population to some percentage below 65% may reduce the minority group's impact.

[27] Of course, the loss of influence in surrounding districts may be compensated for by enhanced political access elsewhere in the state. A legislature or court must be wary, however, that its scheme "packs" minorities, thereby causing a retrogression in electoral access in one area of the state with no concomitant increase in access elsewhere. Without such a clear accompanying increase, the plan reasonably may be labeled retrogressive.

Of course, this Court, in fashioning a remedial order, is cognizant of how a districting scheme that "fragments" a concentrated minority group may dilute minority voting strength. If this fragmenting occurs at a time and in a manner that augments or diminishes voting strength previously enjoyed, it reasonably may be characterized as a retrogression in access to the political process.

[28] The overall effect of this Court's plan is discussed in greater detail. For the purpose of this analysis, however, District 25 has a minority population of 38.7%. District 26 has a minority population of 7.01%, and District 27 has a minority population of 64.25%. The average population percentage for the three new districts, therefore, is 36.65%.

[29] The Court recognizes that this adjustment for the three new congressional districts smacks of an attempt to achieve "proportional representation." However, the Court in no way attempts to provide proportional representation statewide. The sole purpose for a formula that considers mathematical potential is to provide some rational, reasonable, and equitable basis for comparing two plans that inherently elude mechanical comparison.

The mechanisms utilized by the Court to adjust for an increase of three congressional districts is not thought to be the only method for adjustment. Perhaps they are not the best means. Undoubtedly they have weaknesses. They are, however, two separate mechanisms for comparing election systems that, because of differences in size, are much like apples and oranges. They provide a reasonable and rational method for determining whether the Court's remedial decree satisfies the nonretrogression principle of Beer.

[30] The minority population percentage and the percentage of minorities in the voting age population for the congressional districts as they were defined previously are as follows:

 
   MINORITY CONCENTRATIONS IN PRE-EXISTING
              CONGRESSIONAL
               DISTRICTS                 
                           % Min. in Voting
District    % Min. Pop.    Age Population
    1         20.2              18.01
    2         18.1              16.81
    3          8.7               7.88
    4         16.3              14.26
    5         29.1              26.18
    6         21.5              18.08
    7         14.0              12.86
    8         40.1              36.07
    9         29.6              26.77
   10         27.9              23.90
   11         22.4              20.00
   12         26.0              22.44
   13         13.5              11.00
   14         45.4              40.39
   15         77.8              72.23
   16         60.5              55.20
   17         14.9              11.57
   18         74.1              67.93
   19         28.9              22.97
   20         77.6              72.27
   21         28.5              24.04
   22         31.2              27.99
   23         57.8              52.15
   24         37.4              32.68

[31] As a purely factual matter, these three safe districts did elect minority representatives. Districts 15, 18, and 20 were represented by Congressmen de la Garza, Leland, and Gonzales, respectively.

[32] By expressly setting out the minority percentages in these specific districts, this Court is in no way attempting to state that a fixed numerical percentage provides access. In its evaluation, however, the Court must make reasonable determinations regarding influence. Accordingly, it examines the statewide apportionment plan that previously existed and points to those districts that had significant numerical minority populations as a starting point for evaluating whether the Court's plan causes a retrogression in minority access to the political process.

[33] For analysis of minority influence on congressional elections in pre-existing Districts 5 and 24 see Hearings on Apportionment of Texas Congressional Districts Before the Committee of the Whole Senate, 67th Legislature of Texas, 1st Called Session (July 14, 1981) (testimony of Professor Larry Carlyle).

[34] ELECTION DATA FOR PRE-EXISTING CONGRESSIONAL DISTRICT 24 1978 General Election Martin Frost 39,201 Leo Berman 33,314 1980 General Election Martin Frost 93,690 Clay Smothers (Black) 59,172

[35] ELECTION DATA FOR PRE-EXISTING CONGRESSIONAL DISTRICT 5 1978 General Election Jim Mattox 35,524 Tom Pauken 34,672 James Michael White 397 1980 General Election Jim Mattox 70,892 Tom Pauken 67,848 J. B. Jackson 295

[36] The plan's racial fairness in terms of retrogression is discussed in greater detail. Minority population comparisons of the pre-existing plan and the Court's plan are as follows:

 
             % Min. Pop.
             Pre-Existing    % Min. Pop.
District        Plan         Court Plan      Change
       1        20.2           21.2           +1.0
       2        18.1           18.7           -0.6
 
      3     8.7               7.4         -1.3
      4    16.3              16.7         +0.4
      5    29.1              31.9         +2.8
      6    21.5              16.8         -4.7
      7    14.0              10.3         -3.7
      8    40.1              29.2        -10.9
      9    29.6              29.0         -0.6
     10    27.9              28.9         +1.0
     11    22.4              23.5         +1.1
     12    26.0              27.7         +1.7
     13    13.5              14.0         +0.5
     14    45.4              31.6        -13.8
     15    77.8              72.2         -5.6
     16    60.5              63.9         +3.4
     17    14.9              14.5         -0.4
     18    74.1              72.0         -2.1
     19    28.9              30.4         +1.5
     20    77.6              70.5         -7.1
     21    28.5              25.1         -3.4
     22    31.2              23.2         -8.0
     23    57.8              57.2         -0.6
     24    37.4              45.7         +8.3
     25    ----              38.7         ----
     26    ----               7.0         ----
     27    ----              64.3         ----
 
           VOTING AGE POPULATION IN
            CONGRESSIONAL DISTRICTS 
              % Min. Pop.
              Pre-Existing    % Min. Pop.
District          Plan        Court Plan*    Change
       1          18.01         18.80        +.79
       2          16.81         17.51        +.70
       3           7.88         ----         ----
       4          14.26         14.66         +.40
       5          26.18         28.12        +1.94
       6          18.08         14.79        -3.29
       7          12.86          9.19        -3.67
       8          36.07         25.80       -10.27
       9          26.77         26.16         -.61
      10          23.90         24.48         +.58
      11          20.00         21.13        +1.13
      12          22.44         23.76        +1.32
      13          11.00         11.34         +.34
      14          40.39         27.94       -12.45
      15          72.23         66.61        -5.62
      16          55.20         58.76        +3.56
      17          11.57         11.31         -.26
      18          67.93         66.17        -1.76
      19          22.97         24.26        +1.29
      20          72.27         64.80        -7.47
      21          24.04         20.97        -3.07
      22          27.99         20.93        -7.06
      23          52.15         51.97         -.18
      24          32.68         40.00        +7.32
      25          ----          34.19        ----
      26          ----          ----         ----
      27          ----          58.23        ----
* The Court does not list the percentage that minorities
comprise of the voting age population in
the new Districts 3 and 26. This data was unavailable.
However, it is unnecessary from an
electoral impact standpoint since minorities comprise
such insignificant part of these districts'
total population.

[37] The plan for Dallas County that is utilized by the Court is S.B. 3 or the Mauzy-McKnight proposal. See Defendant's Exhibit 2, C. 2. The plan originated as a committee amendment by State Senator Mauzy during the First Called Session of the Texas Legislature. S.B. 3 was denied enactment by a vote of 15-14. It was replaced by a proposal that ultimately passed through the state senate by a 15-14 vote.

[38] These percentages of the voting age population do not include the voting age population for block groups 2 and 3 of census tract 1 in Dallas County. The total population for these block groups is 0.28% of District 5's total population. Additionally, they do not consider the voting age population for block groups 4, 5, 6, 7, 8, and 9 of census tract 1130 in Tarrant County. The total population for these block groups is 0.31% of District 24's total population.

[39] As noted, this Court attempts to give deference to the choice of the Texas State Legislature, as expressed in S.B. 1, to the degree possible. The standard applied by this Court is that deference is paid except where the Legislature's choice would interfere with this Court's achieving either of its two coterminous constitutional objectives. Consequently, this Court determines that it is unable to incorporate the provisions of S.B. 1 that affect the Dallas County area.

S.B. 1's apportionment scheme for Dallas County severely affected the two minority districts in which minorities had crucial swingvote influence. In District 5, minorities would have comprised a mere 12.1% of the total population and 10.11% of the voting age population. Minorities would have comprised 63.8% of the total population of District 24. However, they would have constituted only 57.96% of the voting age population. Significantly, District 24 under S.B. 1 would not appear to create a safe minority district. At best, it would provide minorities with a greater ability to swing an election to one candidate or another. Of course, minorities already possessed substantial swing-vote influence in District 24.

While not creating a safe minority seat in District 24, S.B. 1 would destroy the effective swing-vote influence minorities previously enjoyed in District 5. Since it is the Court's plan that is being evaluated under strict standards of racial fairness, an apportionment scheme resembling S.B. 1 would result in a severe retrogression in the Dallas County area. Accordingly, this Court determines that deference in this instance would be inappropriate.

This Court recognizes that certain minority group members expressed a desire for a "safe" minority district in Dallas County. After consideration of numerous political factors, and substantial legislative battling, the Texas Legislature decided on the configurations in S.B. 1, and created the "non-safe" District 24. The legislature was at liberty to engage in such considerations. This Court, in fashioning a nonretrogressive apportionment plan does not have that privilege. It must evaluate the new plan without access to questions regarding the ability of separate minority groups to form coalitions or other political concerns. Its concern is whether the new plan reduces the voice of minorities. It is not before this Court to determine whether considerations valid in the legislative context justify simply increasing swingvote influence in one district at the expense of the influence previously enjoyed in a neighboring district. This Court determines, however, that, in the context of a court-ordered apportionment plan, such a trade-off would result in a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise.

[40] This Court is not required to maximize minority potential, even if it could be done. Indeed, it is directed not to engage in race-conscious line-drawing. The Court has been asked, however, to apportion Harris County in a manner that would increase District 25's minority profile to 60.2% of the district's total population. In order to facilitate this objective, the minority population of District 18 would have to be reduced substantially. The minority population would drop from 72.0% under the Court's plan to 64.9%. Such a drop would jeopardize a proven safe minority district in Harris County and arguably only increase the ability of minorities to swing an election one direction or another. Consequently, the Court chooses to defer to the Legislature's choice, particularly since it does not cause a retrogression in minority voting power.

[41] District 27 under the court-ordered plan has a higher minority population than did S.B. 1's District 27. Conversely, District 15 under the Court's plan has a smaller minority population.

This Court determines that its configurations are necessary to adjust for increases in Hispanic population and, thereby, avoid a retrogression in minority voting strength.

This determination, however, should not be perceived as a failure to defer to the State's choice. The problem addressed by the Court adjusting for increases in Hispanic population to avoid retrogression is substantially the same problem that prompted the Attorney General of the United States to interpose an objection to S.B. 1. As a result, the State of Texas, through its counsel, has indicated that its alternate position is "to ask this Court to draw an interim plan incorporating S.B. 1 in its entirety with the exception of the corrective factor for Districts 15 and 27, those being the districts that were objected to and that have been incorporated in the submission we have prepared and offered to the Court." Hearing, February 9, 1982, at 12.

[1] The plan implemented by the court with respect to Harris County and West Texas is identical to Senate Bill No. 1, the congressional reapportionment plan passed by the state legislature. This dissent finds that plan unconstitutional with respect to those two regions, under the standard applicable to legislative plans. Accordingly, this dissent will refer throughout to "Senate Bill No. 1," rather than the "court-ordered plan." Similarly, reference will be made to the claims of the various plaintiff-intervenors raised in their challenge to the constitutionality of S.B. 1.

[2] Justice Marshall, 446 U.S. at 103ff., 100 S. Ct. at 1520ff., and Justice White, 446 U.S. at 94ff., 100 S. Ct. at 1514ff., strenuously challenged this interpretation. Justice White's dissent is especially noteworthy, in that he authored the majority opinion in both Whitcomb v. Chavis, 403 U.S. 124, 91 S. Ct. 1858, 29 L. Ed. 2d 363 (1971), and White v. Regester, 412 U.S. 755, 93 S. Ct. 2332, 37 L. Ed. 2d 314 (1973), which indisputably are major cornerstones in the prevailing analytical structure of voting rights jurisprudence.

[3] S.B. 1 is modified by the court's plan with respect to Dallas County and South Texas. I concur in these modifications, because I believe the plan as passed by the state legislature is unconstitutional as it affects those areas, notwithstanding the letter of objection issued by the Attorney General. Moreover, the effect of S.B. 1 in these regions is highly relevant to the issue of legislative intent with respect to the entire plan. Accordingly, this dissent will include an intensely localized appraisal of the design and impact of S.B. 1 with respect to Dallas County and South Texas. See infra at pp. 995-1003 (Dallas), 1008-1014 (West Texas).

[4] District 22 which, under S.B. 1 is 23% minority, is now substantially outside of Harris County. More importantly, the portion of the district inside Harris County is almost entirely White. The minority population of the district is drawn from Brazoria and Fort Bend County. The reference by the majority to swing influence of Harris County minority voters in District 22 is fundamentally mistaken, and seriously distorts the impact of S.B. 1. It derives from a failure to conduct the sort of intensely localized appraisal required by the Supreme Court.

[5] Chairman Von Dohlen relinquished his seat to Rep. Hollowell during the statement of Raul Noriega, of Texas Rural Legal Aid, in which Noriega objected to the "ramrodding" of the Von Dohlen plan through the committee. Von Dohlen apparently felt it improper to debate Noriega from the chair. DX. 89, House Committee Tr., May 12, 1981, at p. 43.

[6] The Calendars Committee is the only House Committee that does not tape its hearings, because it is the only committee that does not take public testimony. Its only meetings are in formal sessions, at which only committee members participate. Dep. of Susan McBee, p. 154.

[1] "The well-settled principle that the nature and scope of the remedy are to be determined by the violation means simply that federal-court decrees must directly address and relate to the constitutional violation itself. Because of this inherent limitation upon federal judicial authority, federal-court decrees exceed appropriate limits if they are aimed at eliminating a condition that does not violate the Constitution or does not flow from such a violation...." Milliken v. Bradley, 433 U.S. 267, 281-82, 97 S. Ct. 2749, 2757-58, 53 L. Ed. 2d 745 (1977) (citations omitted).

[2] "We have made clear that in two important respects a court will be held to stricter standards in accomplishing its task than will a state legislature: `[U]nless there are persuasive justifications, a court-ordered reapportionment plan of a state legislature must avoid use of multi-member districts, and as well, must ordinarily achieve the goal of population equality with little more than de minimis variation.'" Connor v. Finch, 431 U.S. 407, 414, 97 S. Ct. 1828, 1833, 52 L. Ed. 2d 465 (1976) (citing Chapman v. Meier, 420 U.S. 1, 26-27, 95 S. Ct. 751, 765-766, 42 L. Ed. 2d 766 (1975).