Full opinion text
MEMORANDUM OPINION AND ORDER MYRON H. THOMPSON, Chief Judge. Irl this lawsuit, three African-American plaintiffs — Hoover White, John A. Dillard, and Glenn Moody — claim that the current at-large system of electing Alabama appellate judges violates § 2 and § 5 of the Voting Rights Act of 1965, as amended, 42 U.S.C.A. §§ 1973, 1973c (West 1994). They name as defendants the State of Alabama and its Secretary of State. The plaintiffs invoke the jurisdiction of the court pursuant to 28 U.S.C.A. §§ 1331,1343(3) (West 1993) and 42 U.S.C.A. § 1971(d) (West 1994). The plaintiffs represent all African-American electors and resident citizens in Alabama. In settlement of this litigation, the plaintiffs and the defendants have submitted to the court a proposed final judgment for approval under Rule 23(e) of the Federal Rules of Civil Procedure. For the reasons that follow, the court concludes that it should approve and adopt the proposed judgment. I. A BRIEF OVERVIEW The proposed final judgment provides for temporary relief that promises to afford to the black citizens of this state their right to an equal opportunity to participate in the political process and elect candidates of their choice to appellate judgeships. At the same time, the proposed judgment will serve two important and substantial state interests: first, it will preserve the state’s at-large system of electing appellate judges; and, second, it will allow the 1994 elections and future elections for all existing appellate judge-ships to proceed even though some of the judgeships were created without the required federal approval. In broad terminology, the proposed judgment achieves these ends by modifying the already existing and frequently used procedure under which persons, and almost exclusively white persons, have been appointed by the governor to appellate judgeships and thereby have enjoyed the privilege and advantage of incumbency before having to stand for election. The proposed judgment extends to minority-preferred candidates— who are the candidates of choice for blacks but need not themselves be black — a special and expanded opportunity to enjoy this privilege and advantage before having to stand for election. Thus, under the proposed final judgment, the governor — based on recommendations from a special nominating committee, composed in a manner to attempt to reflect the interests of most African-American Alabamians — would appoint minority-preferred candidates to a limited number of appellate judgeships. These appointed judges would then, like all other judges, have to stand for the approval of the electorate at large. In order to make this expanded opportunity of incumbency realistically available to minority-preferred candidates within a reasonable period of time, the proposed judgment provides for an increase in the size of the courts of criminal and civil appeals. It also provides for a possible temporary increase in the size of the state supreme court. The proposed final judgment is not without opposition. On the one side, there are those who claim, among other things, that modifying the at-large system is inadequate. Three plaintiff-intervenors, on behalf of all Alabama Republicans, and one African-American in-tervenor claim that the only appropriate relief would be to transform the state’s scheme for electing appellate judges from an at-large system to single-member districts, in which each appellate judge or justice would represent only a geographic portion of the state and in which district lines would then be drawn according to the race of voters. According to these objectors, this single-member scheme would provide African-Americans a greater assurance of being able to elect candidates of their choice. On the other side, there are those who claim that the court should not modify the at-large system in any manner. Yet all the objectors must recognize that without the temporary relief now before the court, the 1994 elections would probably not be able to proceed for some of the appellate positions and, as a result, these positions would be removed from the electorate with the current incumbents holding over. For the reasons given in this opinion, the court concludes, first, that the evidence supports a settlement of the plaintiffs’ claims and, second, that the court should proceed cautiously at this time and adopt the relief suggested by the plaintiffs and the defendants. Their relief holds out a substantial promise of success and yet is minimally race-conscious and governmentally intrusive. This relief, like the relief for which the three Republican intervenors and the one African-American intervenor want to hold out, would afford a substantial promise to open up fully the state’s political process to its African-American citizens; however, this relief, unlike the relief the intervenors want, would be only temporary, would preserve the state’s at-large system, and would with certainty allow the 1994 elections to go forward for all appellate judgeships. The proposed settlement would also make only limited and necessary changes to state law and would not violate federal statutory or constitutional law. It may be, as feared by some of the objectors, that the proposed judgment will not open up the political process. But, as is shown below, the proposed judgment — which has the express support of many prominent black political officials in the state and the implicit support of the overwhelming majority of black citizens in the state — does not foreclose the opportunity for the court to revisit, after a reasonable period of time, the question of the adequacy of relief. Admittedly, some, but not all, judges appointed under the proposed judgment would enjoy incumbency for six years before having to go before the voters to retain their offices. However, this period of incumbency, while extended, is not greatly out of line with state practice, when that practice is viewed over the entire period in which blacks have allegedly been denied an equal opportunity to elect candidates of their choice. For example, within the last 30 years, one appellate judge served two successive appointive terms totalling almost four years and three other appellate judges served appointive terms of approximately two and a half years. Indeed, one of the objectors to the proposed settlement — a white Republican, who was first appointed to the trial bench and then appointed to the appellate bench — is now eligible for retirement benefits for his years on the appellate bench without ever having won election to a judicial seat. Moreover, as will be shown below, the evidence reflects that current racial barriers “ “will not wither away of their own accord,’” United States v. Paradise, 480 U.S. 149, 163, 107 S.Ct. 1053, 1062, 94 L.Ed.2d 203 (1987) (plurality opinion) (quoting Paradise v. Prescott, 585 F.Supp. 72, 75 (M.D.Ala.1983)), and that, if these barriers are to be not only overcome but effectively eliminated over time, this extended period of incumbency is necessary, albeit only as part of a temporary measure. It is also significant that, if the court were to reject the settlement and, as appears reasonably likely, elections could not proceed for a number of federally-unapproved appellate positions, the judges who occupy those positions could enjoy uneleeted incumbency for up to six years. For example, in the neighboring State of Georgia, it has been almost six years since many judgeships, for which the required federal approval had not been given, have been up for election. Brooks v. State Bd. of Elections, 848 F.Supp. 1548, 1551 (S.D.Ga.1994). The Georgia judiciary was recently described as “an overworked judiciary frozen in its current form.” Id. at 1573. Therefore, with or without the proposed settlement, the State of Alabama will probably face judges with extended unelected incumbency. Indeed, it appears that one of the Alabama judges who might enjoy this extended incumbency in the absence of the settlement received his position by appointment. In order to avoid the situation in Georgia and in order to open up further the political process to Alabama’s black citizens, the court today affirms the State of Alabama’s choices as embodied in the proposed settlement. The court therefore finds that the relief is fair, adequate, and reasonable, as well as legal and good public policy. II. BACKGROUND In the last 125 years, only two African-Americans have served as appellate judges in Alabama. As of the 1990 census, however, blacks comprised 25.26% of the population of Alabama, 22.73% of the voting age population, and 20.29% of the total registered voters. The stark disparity between the small number of black appellate judges and the large black population is the essence of this lawsuit. Alabama’s history of racial discrimination generally, and in the area of voting in particular, is well-documented in decisions of this court and others. African-Americans have been discriminated against in almost every area of Alabama public life. Dillard v. Crenshaw County, 640 F.Supp. 1347, 1359-60 (M.D.Ala.1986) (citing numerous cases finding racial discrimination in Alabama). The state has had an “unrelenting historical agenda, spanning from the late 1800’s to the 1980’s, to keep its black citizens economically, socially, and politically downtrodden, from the cradle to the grave.” Id. at 1357. No area has been more problematic for African-Americans than achieving voting rights and access to the political process. See id. at 1356-59 (describing state’s history of voting discrimination). This case presents the first voting rights challenge to Alabama’s appellate court system. Alabama’s judicial system contains three appellate courts: a supreme court, a court of criminal appeals, and a court of civil appeals. Ala. Const, art. VI, § 6.01(a). The Alabama Constitution further states that the number of judges on each of the three appellate courts is determined by the legislature. Ala. Const, art. VI, §§ 6.02(a), 6.03(a)-(b). Currently, there are nine justices on the supreme court. Ala.Code § 12-2-1 (1986). Each of the courts of appeals is composed of five judges. Ala.Code § 12-3-1 (Supp.1994). Appellate judges are elected statewide for a term of six years. Ala. Const, art. VI, §§ 6.13, 6.15. Vacancies are filled by gubernatorial appointment. Ala. Const, art. VI, § 6.14. Of the seventeen appellate court judges currently sitting, only one, Associate Justice Ralph Cook, is black. He was appointed to replace the only prior black justice, Associate Justice Oscar Adams, who was also appointed. No other black has been an appellate judge since 1865. A. Procedural History The plaintiffs brought this lawsuit on January 27, 1994. Their complaint, as amended, asserts two claims under the Voting Rights Act. The first claim is that four acts of the state legislature regarding the state appellate courts violate § 5 of the Voting Rights Act, which provides that a covered jurisdiction shall not “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 1,1964” without obtaining'preclearance from the federal government. 42 U.S.C.A. § 1973c (West 1994). In 1964, the appellate court system in Alabama consisted of a three-judge court of appeals and a seven-judge supreme court. All of these judges were elected statewide. The § 5 claim centers on four post-1964 acts of the Alabama legislature concerning appellate judges. Those four acts adopted, among other provisions, the following changes in state law: (1) the supreme court was expanded to nine judges, 1969 AlaActs 602; (2) a three-judge court of criminal appeals and a three-judge court of civil appeals replaced the former court of appeals, 1969 AlaActs 987; (3) the number of judges on the court of criminal appeals was increased to five, 1971 Ala.Acts 3d Spec.Sess. 75; and (4) the court of civil appeals was expanded to five judges, 1993 Ala.Acts 346. The plaintiffs sought a declaratory judgment that the four acts are void because they were not precleared. They also sought injunctive relief against the enforcement of the acts. The second claim is that the current system of statewide election of Alabama appellate judges violates § 2 of the Voting Rights Act, which forbids the “denial or abridgement of the right of any citizen of the United States to vote on account of race or color.” 42 U.S.C.A. § 1973 (West 1994). The plaintiffs allege that the current statewide method of electing appellate court judges dilutes black voting strength, thereby denying African-Americans an equal opportunity to participate in the political process and elect candidates of their choice. On April 15, 1994, in settlement of this litigation, the plaintiffs and the defendants submitted to the court a proposed final judgment. The day before, on April 14, the United States Attorney General had interposed an objection under § 5 to the four legislative acts in question. The Attorney General further stated, however, that if this court approved and adopted the proposed judgment, she “would be prepared to grant the requisite preclearance.” In other words, the four acts would be unobjectionable if the additional changes contemplated by the proposed judgment were approved and adopted. The Attorney General further noted that a three-judge court had been convened pursuant to 42 U.S.C.A. § 1973c (West 1994) and 28 U.S.C.A. § 2284 (West 1978 & Supp.1994) to hear the § 5 claim and thus was “poised to address the question whether injunctive relief should be granted based on the unpre-cleared status of appellate court judgeships that are up for election this year.” She therefore added that, because the “proposed consent judgment contemplates that elections will go forward this year under the at-large election system” and because “the Attorney General has precleared the changes occasioned by the proposed judgment and is prepared to preclear the legislative changes if the court grants its approval to the judgment,” she “believe[d] that it would be appropriate to defer granting injunctive relief and thus allow the primary election for the un-precleared positions to be conducted.” The Attorney General advised, however, that, “Should the court not approve the judgment before this year’s general election [in November 1994], the issue of granting injunctive relief should be revisited.” On the basis of the Attorney General’s conditional preclearance of the proposed judgment, the three-judge court issued an order on April 15, 1994, allowing primary elections for unprecleared state appellate judicial positions to go forward. The court agreed with the Attorney General that absent an “extreme circumstance,” a district court must enjoin elections that would be conducted in violation of § 5’s prohibition against implementation of unprecleared changes. Clark v. Roemer, 500 U.S. 646, 654, 111 S.Ct. 2096, 2102, 114 L.Ed.2d 691 (1991). Such a circumstance was presented because “the Attorney General has pre-cleared the changes that would be effected by the proposed consent judgment and is prepared to preclear the legislative changes if the proposed judgment receives judicial approval” and the fact that “the proposed judgment contemplates that elections will go forward this year under the at-large election system.” As suggested by the Attorney General, however, the three-judge court cautioned that “if the proposed judgment does not receive judicial approval before this year’s general election, the court should revisit the issue of whether injunctive relief would be appropriate.” Subsequently, in an order entered on May 11,1994, the three-judge court stayed proceedings before it, severed the § 2 claim from the § 5 claim, and referred the proposed settlement to the single-judge court. White v. State of Alabama, 851 F.Supp. 427, 430 (M.D.Ala.1994) (three-judge court). Over the course of the proceedings a number of parties, in addition to the plaintiffs and the defendants, have become involved. The three-judge court granted the Department of Justice’s motion for leave to participate as amicus curiae. The three-judge court also granted a motion to intervene as a plaintiff by Ralph E. Bradford, an African-American seeking to remedy the alleged § 2 violation by replacing the at-large election system with single-member districts. The single-judge court allowed the following persons, who are Republicans and not African-American, to intervene as plaintiffs in support of single-member districts: Mark Montiel, a judge on the court of criminal appeals and Republican candidate for supreme court in 1994; Johnny Curry, a state legislator and chairman of the Jefferson County Republican Executive Committee; and Jack Williams, executive director of the House Republican Caucus. The court certified a class consisting of all Alabama electors who are Republican and a subclass consisting of all Alabama electors who are Republican and are not African-American, both classes to be represented by Judge Montiel, Curry, and Williams. The single-judge court also ordered that Christopher Boehm, who is not African-American, be allowed to intervene as a defendant supporting the current system of at-large elections. The court certified a class consisting of all citizens who are qualified electors of the State of Alabama and are not African-American or black, which would be represented by Boehm. Meanwhile, concurrently with filing the proposed final judgment on April 15, 1994, the plaintiffs and the defendants requested approval of that agreement. The court subsequently entered an order preliminarily and conditionally approving the proposed settlement, subject to objections at a fairness hearing to be held on July 29, 1994. The court also approved the proposed notice to the public, which the state was required to have published in major Alabama newspapers. Twenty-nine people filed written objections and three of those objectors spoke at the July 29 fairness hearing. One of the written objections came from a named inter-venor. In addition to the written and oral objections, all of the intervenors object to at least portions of the settlement for reasons that overlap with those of the objectors. The objections break down into ten categories. First, several people supported single-member districts as a remedy. Second, some people objected to using an appointment procedure rather than electing judges. Third, a number of objectors decried what they saw as an attempt to place black judges on the bench through affirmative action procedures. Fourth, some people believed the settlement violates Alabama law and objected that the state attorney general does not have the power to make such a settlement. Fifth, some people felt the settlement is collusive or politically motivated. Sixth, four prisoners objected, stating that any settlement should include restoration of direct appeal rights, a new appeal, and the right to file a motion for a new trial. Seventh, two people had procedural objections to the court’s jurisdiction. Eighth, one person objected to the possibility of the chief justice appointing associate justices. Ninth, one person said the judgment lasted too long. And tenth, one person said the judgment should last longer. The court also received a joint evidentiary record for the fairness hearing that was prepared by all of the parties, including the intervenors, and that was filed with the court in parts on July 5 and September 13, 14, 16, and 21, 1994. After the hearing, the court certified a plaintiff class consisting of all black resident citizens and electors of the State of Alabama, with the class represented by White, Dillard, and Moody. On September 2, 1994, the court held a subsequent hearing with expert witnesses for the purpose of obtaining clarification on a number of issues. At the court’s request, the parties submitted additional evidence on some of these issues. B. Provisions of the Settlement In general terms, the proposed judgment would implement a remedy that remains within the framework of statewide elections historically used for selecting appellate judges in Alabama, while allowing minority-preferred candidates, who can be either black or non-black, to run for election with the benefit of incumbent status. It would allow the 1994 elections to go forward under existing state law, even though portions of that law have not been precleared. The proposed judgment would terminate in the year 2019, though the court could terminate it earlier if the objectives have been achieved. The judgment could also be extended if necessary. If, in the year 2003, the objectives of the proposed settlement are not being met, the parties must first attempt to agree on a further remedy; failing that, the plaintiffs may petition the court for additional relief. The specifics of the proposed judgment are different for the courts of appeals and the supreme court. 1. Courts of Appeals The proposed settlement would make the following changes to the current method of selecting judges for the courts of appeals: Prior to 1997, the state will add two judges to each of the court of criminal appeals and the court of civil appeals, for a total of seven judges on each court. The four new judges, who will be seated in January 1997, will be chosen through procedures set forth in the proposed settlement. These procedures provide for the creation of a nominating commission composed of one member selected by the Alabama Lawyers Association; one member selected by the Alabama State Bar Association; two members selected from the plaintiff class by class counsel; and one member selected by majority vote of the other four members or, in the event of deadlock, by the Alabama Black Legislative Caucus. The commission will select three attorney applicants to nominate for each judgeship. The governor will then choose an appointee for each judgeship from the three nominees for that judgeship. Each of the four judges will be appointed for a term of six years. After the new appointments, if there is a vacancy on either the court of criminal appeals or the court of civil appeals and there are fewer than two members of such court who are black or who were appointed through the nominating process, then the vacancy will be filled by the same method of nomination. Rather than serve a full six year term, however, such appointees will be subject to election after serving one year, in accordance with the constitutional provision for filling vacancies. Ala. Const, art. VI, § 6.14. 2. Supreme Court The nominating commission will also be used to fill positions on the supreme court, but in a different way. Starting in 1995, if there are fewer than two associate justices of the supreme court who are black or who were appointed through the nominating process, then any vacancies will be filled through the same nominating process used for vacancies on the courts of appeals. Appointees to vacant positions will be subject to election after one year. For each of the elections in 1996,1998, and 2000, if there are fewer than two associate justices of the supreme court who are black or who were appointed through the nominating process, then if any sitting associate justice does not qualify for reelection, that open seat will be filled through the nominating process for a term of six’ years. In each of 1998 and 2000, if there are fewer than two associate justices who are black or who were appointed through the nominating process and no sitting associate justice fails to qualify for reelection, then the state will create an additional seat on the supreme court, which will be filled through the nominating process for a term of six years. If there are more than eight associate justices on the supreme court, any seat vacated by an associate justice who is not black or appointed through the nominating process will not be filled. 8. Evolution of the Settlement The proposed final judgment currently before the court is not the first settlement to be submitted by the plaintiffs and the defendants. Their first proposal was submitted on February 24, 1994. The current proposal and the original proposal differ in many important respects. First, the original proposal provided that only blacks would be eligible for appointment. The current agreement removes all racial prerequisites to appointment. Second, the original proposed judgment provided that appointments would be triggered if two African-Americans were not sitting on each appellate court. The current proposal provides for appointments if there are fewer than two judges or justices who are black or who were appointed through the nominating process. Because there will be two initial appointments to both of the courts of appeals and because these appointments need not be black, future appointments to vacancies on these courts will not depend wholly on the race. The trigger for initial appointments to the supreme court will remain dependent on race because there will be no initial appointments like the courts of appeals. Third, if the number of associate justices is increased pursuant to the provisions of the judgment, the original proposal abolished a seat if it was vacated by a white justice. The current proposal abolishes the seat if the vacating justice is white or if the seat was not filled by the judicial nominating commission. Fourth, the original proposal provided for discussions about further remedial measures if there are fewer than two African-Americans on any of the appellate courts for more than a year after 2003. The current proposal triggers this provision only if there are fewer than two judges or justices who are black or who were appointed through the nominating process. Fifth, the original settlement provided that the first white supreme court justice who vacated office would be replaced by a black, irrespective of the number of black justices then serving on the court. This provision has been eliminated in the current proposal. Sixth, the original proposal was of unlimited duration. The current proposal is limited to 24 years unless extended by the court. It also explicitly allows the state to petition for earlier termination of the judgment. Subsequently, the parties made two further changes. The first change was in response to concerns raised by the Boehm defendant class on behalf of all white electors of the state. Counsel for the Boehm class wrote that, “After evaluating the current state of the record before this Court and evaluating the prospects of litigation if the proposed consent judgment is not approved, the ‘Boehm Class’, in large measure, supports the proposed consent judgment.” Counsel added, however, “that this Court can and should use its equitable power to make minor adjustments to the proposed consent judgment to ensure current and future constitutional protection for the ‘Boehm Class’.” Counsel explained that the nominating commission “would contain a minimum of three and, possibly, a maximum of five African-American members,” with the result that there is the “potential of having no non-African-American and the probability of having only one” on the commission. Similarly, the Alabama State Bar passed a resolution taking issue with the proposed settlement only with regard to the nominating commission in two respects: first, that the members of the commission should be chosen by and not necessarily have to come from certain groups; and, second, that the members should be chosen by predominantly black groups other than Alabama Black Legislative Caucus and other than counsel for the named plaintiffs. The court agreed with some of the concerns raised by the Boehm class and the State Bar, and, therefore, at the urging of the court, the plaintiffs and the defendants modified the requirements for the composition of the nominating commission. There will still be two members selected from the plaintiff class by class counsel. The other three members, however, will no longer be selected by and from, the predominantly black Alabama Lawyers Association; by and from the Alabama State Bar Association; and by majority vote of the other four members or, in the event of deadlock, by and from the Black Legislative Caucus. Each of the three members will be selected only by these organizations, further eliminating race-based measures because the members of the commission will not have to be members of the Alabama Lawyers Association and, in the event of deadlock, the Black Legislative Caucus. Each of these three members may therefore be of any race. The second modification suggested by the court was made to avoid unnecessary changes in state law. The plaintiffs and the state agreed to eliminate a provision that allowed the chief justice to make appointments should the governor fail to do so. The final version of the proposed judgment was filed on September 15 and precleared on September 20, 1994. It is, therefore, the result of an extended process in which all concerned — the plaintiffs, the defendants, the intervenors, and other interested persons and groups — were invited to work together to find a remedy to racial discrimination that is dominantly race neutral and fits as much as possible into the historical framework for statewide elections, and that, to the extent it is race-conscious and governmentally intrusive, is so in only a temporary and reasonably narrow way. III. STANDARDS FOR REVIEW The court has previously explained the standards for deciding whether to approve a settlement in a voting rights class-action case. Dillard v. Crenshaw County, 748 F.Supp. 819, 823 (M.D.Ala.1990). Judicial policy favors voluntary settlement as the means of resolving class-action eases. Cotton v. Hinton, 559 F.2d 1326, 1331 (5th Cir.1977). The court has applied this policy to voting rights class actions. See, e.g., Crenshaw County, 748 F.Supp. at 823; Dillard v. Town of Louisville, 730 F.Supp. 1546, 1548 (M.D.Ala.1990); Harris v. Graddick, 615 F.Supp. 239, 241-42 (M.D.Ala.1985). It is also established, however, that the settlement process is subject to abuse and, therefore, courts must independently evaluate whether a settlement is fair, adequate, and reasonable. Piambino v. Bailey, 757 F.2d 1112, 1139 (11th Cir.1985), cert. denied, 476 U.S. 1169, 106 S.Ct. 2889, 90 L.Ed.2d 976 (1986); Pettway v. American Cast Iron Pipe Co., 576 F.2d 1157, 1169 (5th Cir.1978), cert. denied, 439 U.S. 1115, 99 S.Ct. 1020, 59 L.Ed.2d 74 (1979). For instance, the interests of the class and its lawyer may diverge, or some members of the class may be “sold out” by other members. Pettway, 576 F.2d at 1169. As part of determining fairness, adequacy, and reasonableness, the court must ensure that the settlement is not collusive. Piambino, 757 F.2d at 1139. Finally, the court has the duty of ensuring that the settlement is not illegal or against public policy. United States v. City of Alexandria, 614 F.2d 1358, 1362 (5th Cir.1980); Harris, 615 F.Supp. at 241-42. IV. WHETHER THE SETTLEMENT IS FAIR, ADEQUATE, AND REASONABLE In deciding whether a settlement is fair, adequate, and reasonable, a court may examine the following factors: (1) the views of the class members; (2) the views of class counsel; (3) the substance and amount of opposition to the settlement; (4) the possible existence of collusion behind the settlement; (5) the stage of the proceedings; (6) the likelihood of success at trial; (7) the complexity, expense, and likely duration of the lawsuit; and (8) the range of possible recovery. See Leverso v. Southtrust Bank of Alabama, Nat. Assoc., 18 F.3d 1527, 1530 n. 6 (11th Cir.1994); Bennett v. Behring Corp., 737 F.2d 982, 986 (11th Cir.1984); Crenshaw County, 748 F.Supp. at 823. A. Views of Class Members The first place a court should look to determine whether a settlement is fair, adequate, and reasonable is to the views of the class. Shuford v. Alabama State Bd. of Educ., 846 F.Supp. 1511, 1517 (M.D.Ala.1994). As a preliminary matter, the court must ensure that class members have been given notice of the settlement, as required by Rule 23(e) of the Federal Rules of Civil Procedure. In this case, a court-approved notice was published twice weekly for three weeks in major Alabama newspapers. The notice summarized the proposed settlement and advised interested persons about how to object in writing and at the fairness hearing. The court held a fairness hearing to obtain the views of both class and non-class members. The court finds that the notice and fairness hearing were adequate to inform the plaintiff class and other interested parties about the proposed settlement and to seek and ascertain their opinions. Based on the lack of objections, the court concludes that the plaintiff class overwhelmingly supports the proposed settlement. Cf. Reynolds v. King, 790 F.Supp. 1101, 1109 (M.D.Ala.1990) (majority’s silence may be, but is not always, indicative of class-wide support). Only two objections came from people who identified themselves as African-American. One of these objectors argued for a remedy of single-member districts, but supported the nominating commission in the alternative. The second black objector is plaintiff-intervenor Bradford, who opposes the settlement on a number of grounds, including that the settlement is illegal and that the remedy should be single-member districts. The court also received evidence from the plaintiff class in regard to the proposed settlement. A number of prominent African-Americans support the proposed settlement. Among them are Oscar Adams, the first black associate justice on the Alabama Supreme Court and the only black person ever elected statewide to a constitutional office in Alabama, albeit after a period of incumbency; Fred Gray, a prominent black attorney and former state representative; Richard Arrington, Jr., the African-American mayor of Birmingham; and Joe L. Reed, chair of the Alabama Democratic Conference, a black political organization, and a veteran of Alabama politics. These four men cited a number of reasons for supporting the proposed settlement. Mayor Arrington indicated that the nominating commission is a good way of identifying minority-preferred candidates who can be elected statewide after appointment. Gray and Reed suggested that black attorneys are reluctant to run for appellate judgeships without the advantage of incumbency. Justice Adams emphasized the importance of incumbency in allowing him to win statewide races He stated that, “Incumbency is the best method ... to overcome the reluctance on the part of many white voters to vote for a black candidate.” Adams believes that the best “way of remedying any underrepresentation of blacks in appellate judgeship positions would be through an appointment process comparable to that contained in the proposed consent judgment.” He further added that “the settlement in this case does the best possible job of preserving the interests of the State in the judicial system.” Particularly because there have been only two objections by members of the plaintiff class, the court does not believe that the proposed final judgment is unfair to any segment of the black population No money damages are at stake and all black voters are treated equally. Because there is no conflict of interest within the class which would elevate in importance the views of a small segment of the class, majority opinion is compelling. Pettway, 576 F.2d at 1217. Based on the paucity of objections and the strong support for the proposed judgment among African-American leaders, the court finds that the plaintiff-class favorably views the proposed settlement. B. Views of Class Counsel In considering the fairness, adequacy, and reasonableness of a proposed settlement, a court should also consider the views of counsel for the class. Pettway, 576 F.2d at 1215. Class counsel for the plaintiffs have argued that the proposed settlement is fair, adequate, and reasonable. These attorneys are experienced civil rights and voting rights lawyers and the court respects their views. C. Substance and Amount of Opposition to the Settlement A number of objections have been raised to the proposed settlement. The bulk of the objections relate to the legality of the proposed settlement and are dealt with in a later section. The court will, therefore, limit this section to the objections relevant to fairness, adequacy, and reasonableness. One objection is that the proposed remedy is not the most appropriate one. Most people holding this view propose single-member districting as a solution. The court is persuaded, however, that the proposed remedy is at least as good for African-Americans as any alternative remedy. According to Mayor Arrington and Reed, the proposed settlement is superior to single-member districts for appellate courts because at-large seats allow blacks to have a significant influence on all appellate judges, rather than have their dominance limited to a small number of districts with little presence in the majority of the districts. Gray also supports the proposed settlement rather than a single-member districting scheme. He reasons that the proposed settlement would allow African-Americans “to have a voice in the initial appointment of at least two persons of choice on each appellate court be they Black or white. At the same time the proposed judgment retains the rights of Blacks to equally share in the election of all of the other members of the court.” This evidence mirrors a debate among courts and scholars over the best remedy for voting discrimination. In response to the predominance of single-member districts as a voting rights remedy, some courts and scholars have argued that districting is not the best method of ensuring minority interests are taken into account. See, e.g., League of United Latin American Citizens v. Clements, 999 F.2d 831, 872-73 (5th Cir.1993) (en banc), cert. denied, — U.S. -, 114 S.Ct. 878, 127 L.Ed.2d 74 (1994); Lani Guinier, No Two Seats: The Elusive Quest for Political Equality, 77 Ya.L.Rev. 1413, 1447-57 (1991). As the Supreme Court summed up the dilemma, the creation of majority-minority districts “does not invariably minimize or maximize minority voting strength.” Voinovich v. Quitter, — U.S. -, -, 113 S.Ct. 1149, 1156, 122 L.Ed.2d 500 (1993). “Instead,” the Court continued, “it can have either effect or neither.” Id. “On the one hand, creating majority-black districts necessarily leaves fewer black voters and therefore diminishes black-voter influence in predominantly white districts,” id., while, “On the other hand, the creation of majority-black districts can enhance the influence of black voters” by ensuring they can elect candidates of their choice. Id. One important question is whether minority voters are better off when they have “influence over a greater number of seats, or control over a lesser number of seats.” Holder v. Hall, — U.S. -, -, 114 S.Ct. 2581, 2595, 129 L.Ed.2d 687 (1994) (Thomas, J., dissenting). But see Guinier, supra, at 1457-58 (rejecting both alternatives). The court need not resolve this dilemma because the parties have expressed their preferences in the proposed settlement. The court need only observe that single-member districts are not necessarily the best solution in all cases. It is true that minorities have often sought single-member districts as a voting rights remedy. Nevertheless, the court finds it entirely reasonable that in this case, involving a statewide appellate judicial system rather than a legislative body, single-member districts are not in the best interests of blacks where, as it appears under the proposed final judgment, there exists an alternative scheme that in substantial measure achieves the goals of districting without sacrificing minority influence in the state as a whole. The proposed settlement might not be optimal for some groups of people, but it appears to be a fair way of remedying past discrimination against African-Americans. Further, the appointment remedy avoids “marginalizing” those black and white voters who would end up a minority in race-based districts. Another objection is that it is unfair to use an appointment procedure rather than electing judges. Admittedly, under the proposed judgment four to six appellate judgeships will be appointed. However, the settlement partially disperses the effect of this procedure on candidates and on the voters at large by adding four judges to the courts of appeals. Therefore, either with or without the settlement, candidates will still be able to run for and voters will still be able to vote for the same number of positions on those two courts, even though with the settlement the absolute right to vote for and run for every position may be hampered. Additionally, because the need for any appointments to open or created seats on the supreme court is only conditional, it is possible that all supreme court seats can be filled without impinging on the right to election. Further, each Alabama citizen will still be electing more judges than if single-member districts were imposed. Finally and most importantly, each of the appointed positions will ultimately come up for election. Citizens will have the last word on judges through the ballot. To be sure, some appointed judges would enjoy incumbency for six years before having to stand for election. This period of incumbency is not greatly out of line with state practice, when that practice is viewed over the entire period in which blacks have allegedly been denied an equal opportunity to elect candidates of their choice. The evidence reflects that within the last 30 years, one appellate judge served two successive appointive terms totalling almost four years and three other appellate judges served appointive terms of approximately two and a half years. Indeed, one objector to the proposed settlement — Judge Montiel, who was first appointed to the trial bench and then appointed to the appellate bench— is now eligible for retirement benefits for his years on the appellate bench without ever having won election to a judicial seat. Moreover, as discussed later, if the current racial barriers to black participation in the election of appellate judged are to be not only overcome but effectively eliminated over time, this extended period of incumbency is necessary. It is also significant that, with or without the proposed settlement, the State of Alabama will probably face judges with extended unelected incumbency. That is, if the court were to reject the settlement and, as appears reasonably likely, elections could not proceed for a number of unprecleared appellate positions, the judges who occupy those positions could enjoy uneleeted incumbency for up to six years. In the neighboring State of Georgia, it has been almost six years since many unprecleared judgeships have been up for election. Brooks v. State Bd. of Elections, 848 F.Supp. 1548, 1551 (S.D.Ga.1994). The Georgia judiciary was recently described as “an overworked judiciary frozen in its current form.”. Id. at 1573. The effect on Alabama’s appellate system could be comparable. In the 1994 election alone, the election for the supreme court seat held by Justice Hugh Maddox and arguably for the seats held by Justices Ralph Cook and Mark Kennedy would be subject to stay; the election for the court of criminal appeals seat held by Judge Mark Montiel and arguably for the seat held by Judge Bill Bowen would be subject to stay; and the election for the court of civil appeals seat held by Chief Judge William Robertson and the two seats added in 1993 would be subject to stay. Future elections for at least one supreme court seat, one court of criminal appeals seat and two court of civil appeals seats would also be subject to stay. This potential disruption absent the settlement cannot be ignored. Some people have expressed concern about the length of the proposed settlement. The court finds that the 24-year time period is reasonable. The proposed judgment can be extended or shortened by the court if circumstances warrant. Moreover, as is explained later in this memorandum opinion, 24 years are needed to effect a change in the composition of the court. For the judgment to be fully successful, changes must take place in voting behavior. Further, of the 26 appointees to appellate courts since 1968, only two have been black. Twenty-four years is therefore a reasonable period of time to allow for the political process to open up. Finally, four prisoners objected, maintaining that, because legislative acts concerning the court of criminal appeals had not been precleared, the affirmance of their convictions was invalid. They argued that any settlement should include restoration of direct appeal rights, a new appeal, and the right to file a motion for a new trial. This argument is unpersuasive. Action taken by courts does not become retroactively .invalid because the courts have not been precleared. “Granting such a remedy would result in the reversal of years of convictions, the overturning of myriad judgments; in short, it would create genuine chaos.” Brooks v. State Bd. of Elections, 775 F.Supp. 1470, 1482 (S.D.Ga.1989) (three-judge court), aff'd mem., 498 U.S. 916, 111 S.Ct. 288, 112 L.Ed.2d 243 (1990). D. Existence of Collusion Objections have been raised that the settlement was a product of collusion between counsel for the plaintiffs and the defendants. It should be noted that the charge is not the typical claim of collusion in a class case: that the class counsel or the named plaintiffs benefit from the settlement at the expense of the class or some members of the class. See, e.g., Holmes v. Continental Can Co., 706 F.2d 1144, 1147-48 (11th Cir.1983); Reynolds, 790 F.Supp. at 1105-08. Rather, the charge of collusion seems to be based on the presumption that, because the settlement was reached quickly, the parties did not conduct settlement talks as adversaries. There is no evidence, however, that the parties engaged in illegal negotiations to reach the proposed settlement. To the contrary, the parties maintain that they engaged in arms-length negotiations. Although the settlement does not include the typical relief for a § 2 violation—that is, single-member dis-tríete — the plaintiffs did not ask for this remedy in their complaint, as amended. Though polities may have been a factor in the proposed judgment, which arguably preserves the strength of Democrats to a greater degree than would single-member districts, there is no suggestion that the state wanted the suit to be brought. Had some state officials encouraged this suit to avoid the legislative process and achieve a result they desired, the court would be concerned. Cf. United States v. Johnson, 319 U.S. 302, 63 S.Ct. 1075, 87 L.Ed. 1413 (1943) (per curiam) (suit collusive if it is not adversary). All the evidence suggests, however, that the state was satisfied with the current at-large system of electing appellate judges and that the proposed settlement is the way the state attorney general sought to preserve that system to the greatest extent possible. Furthermore, this lawsuit did nothing to prevent Republicans from raising claims against at-large elections as currently organized and as proposed under the settlement. In a separate opinion issued today, White v. State of Alabama, 867 F.Supp. 1571 (M.D.Ala.1994), the court discusses in depth the Republican claims of political discrimination under Davis v. Bandemer, 478 U.S. 109, 106 S.Ct. 2797, 92 L.Ed.2d 85 (1986). E. Other Factors The remaining four factors are interrelated: the stage of the proceedings; the likelihood of success at trial; the complexity, expense, and likely duration of the lawsuit; and the range of possible recovery. This litigation has been ongoing for less than a year. An early settlement has the potential for substantial cost savings. On the other hand, without a full understanding of a ease’s potential, parties can settle for less or more than they should. This possibility does not appear to be a problem in this case. Voting rights actions are frequent enough and counsel in this case are experienced enough, that the potential gains of this action are well known. Future revelations are unlikely. The parties are fully aware that the plaintiffs have a substantial likelihood of prevailing on the merits of the § 2 claim. But this result would probably come at quite a cost in time and effort to the parties, as well as disruption to the state court system. Two current challenges to the at-large election of judges that have not settled have gone on for over half a decade with no end in sight. Southern Christian Leadership Conference of Alabama v. Evans, 785 F.Supp. 1469 (M.D.Ala.1992), judgment vacated, 18 F.3d 897 (11th Cir.), vacated and reh’g en banc granted, 18 F.3d 897 (11th Cir.1994) (Alabama trial judges) (filed 1988 and still ongoing); Brooks v. State Bd. of Elections, 848 F.Supp. 1548 (S.D.Ga.1994) (Georgia judicial system) (filed 1988 and still ongoing). In contrast, the proposed settlement allows the 1994 election and future elections to continue on schedule. Any relief the plaintiffs eventually received through trial would likely come after a long struggle and would probably not be significantly better than that afforded by the proposed settlement. In light of the above considerations, the court has independently evaluated the fairness, adequacy, and reasonableness of the proposed settlement. “A settlement is in large measure a reasoned choice of a certainty over a gamble, the certainty being the settlement and the gamble being the risk that comes with going to trial.” Paradise v. Wells, 686 F.Supp. 1442, 1446 (M.D.Ala.1988). As a result, the question is not “whether the proposed consent decree is the best deal possible” but whether it is “at a minimum, fair, adequate, and reasonable.” Id. at 1448. Based on the views of class counsel and prominent members of the class, the court believes that the proposed settlement will benefit African-Americans. This conclusion springs from the court’s beliefs, based on evidence, that the appointment process will produce quality minority-preferred candidates who would otherwise not choose to run for statewide office and that the advantages of incumbency will allow minority-preferred candidates to achieve election after appointment, thereby allowing black voters greater influence. The court realizes that this result is by no means guaranteed. The proposed settlement risks the possibility of appointees consistently losing elections. On the other hand, this system has the potential for achieving a result that is much more satisfying than single-member districts: as whites become used to seeing blacks in important statewide positions, race will hopefully decrease as a factor in voting. Single-member districts do not necessarily hold out this hope in as short a time period. Thus the proposed judgment is fair to the state as a whole. It represents an answer to the conflicting views found in a recent redistricting case, Johnson v. Miller, 864 F.Supp. 1354 (S.D.Ga.1994) (three-judge court) (per curiam). The majority of the court in that case held the Georgia congressional districting plan unconstitutional and stated: “The time has come to contemplate more innovative means of ensuring minority representation in democratic institutions.” Id. at 1393. The dissent, on the other hand, argued that “the Constitution does not condemn all race-conscious districting.” Id. at 1397 (Edmondson, J., dissenting). The court is presented with a proposed settlement that, while race-conscious, achieves in an innovative manner minority access to the political process without the drawbacks of single-member districts. Finally, the court, as will be discussed below, finds that the plaintiffs have a good chance of proving a § 2 violation in the method of electing appellate judges. However, the court finds it entirely reasonable that the plaintiffs wish to settle now in light of the lengthy litigation currently ongoing in Alabama and elsewhere over elected judges. This settlement avoids the expense and delay of complex voting rights litigation. Based on the above considerations, the court finds that the proposed final judgment is fair, adequate, and reasonable. V. WHETHER THE SETTLEMENT IS LEGAL AND GOOD PUBLIC POLICY The court must also address whether the proposed judgment is legal. Opponents contend that the settlement is illegal because it contravenes state law, § 2 of the Voting Rights Act, and the equal protection clause. The court does not agree and concludes upon review that the proposed settlement is legal and good public policy. A Whether the Settlement Violates State Law The court turns first to the contention that the proposed settlement should not be approved because it conflicts with state law. The first issue for analysis is whether the court has the authority to examine state law issues at all. After finding that it does, the court turns to the state attorney general’s authority to enter into a settlement modifying state law absent a finding of liability. After satisfying itself that the settlement is properly before it, the court examines its power to approve a settlement that changes state law. Finally, the court analyzes the specific modifications to state law contemplated by the proposed settlement. 1. Whether the Court has Authority to Consider State Law Before analyzing the extent to which any invalidation of state law affects the proposed judgment, the court must address a threshold issue of its authority to conduct such an examination. The plaintiffs and the defendants argue that the United States Constitution’s eleventh amendment guarantee of sovereign immunity restricts the court’s inquiry into the extent to which the proposed settlement conforms to state law as a basis for denying approval to the proposed judgment. This contention is untenable. It is true that “a claim that state officials violated state law in carrying out their official responsibilities is a claim against the State that is protected by the Eleventh Amendment.” Pennhurst State School & Hosp. v. Halderman, 465 U.S. 89, 121, 104 S.Ct. 900, 919, 79 L.Ed.2d 67 (1984). The word “claim” in this holding, however, refers to a cause of action in a suit, not just any assertion about the law by a party. See id. Pennhurst does not hold that the eleventh amendment prevents federal courts from considering state law as a factor in evaluating a settlement, even if those same issues could not be raised as a cause of action. Even assuming that the court is unable to examine state law issues in the course of evaluating a settlement generally, under eleventh amendment jurisprudence Congress may still abrogate sovereign immunity in certain situations. For instance, because the enforcement powers given to Congress in the fourteenth amendment limit state authority, laws enacted pursuant to the fourteenth amendment have been viewed as falling outside the limits of the eleventh amendment. Thus, in Fitzpatrick v. Bitzer, 427 U.S. 445, 456, 96 S.Ct. 2666, 2671, 49 L.Ed.2d 614 (1976), the Supreme Court allowed an action against a state under Title VII of the Civil Rights Act of 1964, as amended, 42 U.S.C.A. §§ 2000e to 2000e-17; see also Hutto v. Finney, 437 U.S. 678, 693-98, 98 S.Ct. 2565, 2575-77, 57 L.Ed.2d 522 (1978) (attorney’s fees under Civil Rights Attorney’s Fees Awards Act of 1976, 42 U.S.C.A § 1988, upheld against state). Like the fourteenth amendment, the fifteenth amendment affords Congress enforcement power. The Voting Rights Act was passed pursuant to the enforcement authority of the fifteenth amendment. State of South Carolina v. Katzenbach, 383 U.S. 301, 308, 86 S.Ct. 803, 808, 15 L.Ed.2d 769 (1966). Just as laws such as Title VII passed pursuant to the fourteenth amendment trump the eleventh amendment, so does the Voting Rights Act. See id. at 325, 86 S.Ct. at 817 (fifteenth amendment “supersedes contrary exertions of state power”); City of Rome v. United States, 446 U.S. 156, 178-80, 100 S.Ct. 1548, 1562-63, 64 L.Ed.2d 119 (1980) (federalism overridden by legislation enforcing fifteenth amendment). There is no eleventh amendment bar to a federal court examining the extent to which a voting rights settlement conforms to existing state law. 2. Whether the State Attorney General Has Authority to Enter into a Settlement Modifying State Law Opponents of the proposed settlement contend that the attorney general cannot agree to modifications in state law to settle this case absent a finding by the court that the state has violated federal law. See Brooks, 848 F.Supp. at 1563 (attorney general permitted to settle in violation of state law only if federal liability determined); see also League of United Latin American Citizens, 999 F.2d at 845-47. The Alabama Supreme Court has held that “the attorney general has the power to manage and control all litigation on behalf of the State of Alabama.” Ex parte Weaver, 570 So.2d 675, 684 (Ala.1990). This power at a minimum includes the ability to settle a case if the settlement does not conflict with state law. State ex rel. Carmichael v. Jones, 252 Ala. 479, 41 So.2d 280, 285 (Ala.1949). Assuming that the proposed settlement does conflict with some aspects of state law, the court must conduct further analysis into the attorney general’s power to settle cases. It is important to note that Jones dealt with a state law claim rather than a federal one. In a federal suit, the attorney general must be cognizant not only of violations of state law but federal law as well. It is basic constitutional law that “the Supremacy Clause invalidates all state laws that conflict or interfere with an Act of Congress.” Rose v. Arkansas State Police, 479 U.S. 1, 3, 107 S.Ct. 334, 335, 93 L.Ed.2d 183 (1986) (per curiam). Therefore, if the attorney general believes that state law violates federal law, his duty is to uphold federal law. See Delchamps, Inc. v. Alabama State Milk Control Bd., 324 F.Supp. 117, 118 (M.D.Ala.1971) (three-judge court) (per curiam). If a state admits it is violating federal law, it would be a waste of time and resources for the court to hold a trial merely to give the attorney general the power to settle on a remedy changing state law. Since most § 2 remedies involve changes to state or local law, requiring a court finding of liability would effectively destroy the strong policy in favor of settlement in such eases. For this reason, the attorney general must have the power to enter into a settlement that alters state law if liability is admitted when the purpose is to correct federal law violations. This lawsuit has the added complication that the state does not admit liability. There are numerous cases that hold, however, that a jurisdiction is not required to admit liability to settle litigation. Recent Eleventh Circuit Court of Appeals decisions in the employment discrimination area suggest that it is not required that “a city settling litigation by consent decree declare itself to have violated the law.” In re Birmingham Reverse Discrimination Employment Litig., 20 F.3d 1525, 1539 (11th Cir.1994); see also Ensley Branch, NAACP v. Seibels, 31 F.3d 1548, 1557-58 (11th Cir.1994). Likewise, requiring a jurisdiction to admit liability before settling a voting rights action would thwart the strong policy in favor of settlement. The attorney general need not admit liability in order to settle a § 2 case that modifies state law as long as there is sufficient evidence to support the alleged § 2 violation. See Moch v. East Baton Rouge Parish School Bd., 533 F.Supp. 556, 559-60 (M.D.La.1980) (§ 2 consent decree in case that modified state law approved with provision denying liability). Finally, there is no state law prohibiting state