Full opinion text
ORDER EDENFIELD, Chief Judge. Currently before the Court are the joint motion of the Defendants and the Plaintiffs for approval of their proposed settlement agreement and a motion by the Intervenors for summary judgment. For the reasons stated herein, both motions are DENIED. Background Plaintiffs filed this action in 1988 challenging several aspects of the Georgia judicial system under the United States Constitution and the Voting Rights Act of 1965 (“VRA”). Specifically, the Plaintiffs challenge the manner in which judges of the state court, superi- or court and court of appeals and justices of the supreme court are elected in Georgia under Section 2 of the VRA and the legality of certain changes made to the Georgia judicial system since the enactment of Section 5 of the VRA. Pursuant to Section 5 of the VRA, Pub.L. 89-110, Title I, § 5, 79 Stat. 4S7 (codified as amended at 42 U.S.C. § 1973c (1988)), a three-judge panel was convened. In December 1989, the Panel held that Section 5 applied to judicial elections, that the Georgia electoral scheme has the potential for discriminating against minority voters, and that the State of Georgia failed to obtain preclearance for numerous changes to their electoral scheme as required by the VRA. Brooks v. State Bd. of Elections, 775 F.Supp. 1470 (S.D.Ga.1989), modified on other grounds, 775 F.Supp. 1490 (S.D.Ga.1991), and aff'd, 498 U.S. 916, 111 S.Ct. 288, 112 L.Ed.2d 243 (1990) (Brooks I). Consequently, the Court enjoined the State from filling judicial positions that had not been pre-cleared. Id. at 1484. Incumbents, however, were allowed to remain in office. Id.; injunction extended by Brooks v. State Bd. of Elections, 775 F.Supp. 1490, 1491 (S.D.Ga.1991); Brooks v. State Bd. of Elections, 790 F.Supp. 1156, 1159 (S.D.Ga.1992). By Order dated February 28, 1994, this injunction was extended until March 1, 1995. In June of 1992, after extensive negotiations spanning the course of two years and under the leadership of Senior Federal District Court Judge Anthony' A. Alaimo, the Plaintiffs and the Defendants reached a proposed settlement agreement. This agreement, presented to the Court in the form of a consent decree, provides, inter alia, that: (1) the Governor will hereafter appoint all judges in Georgia; (2) appointed judges will thereafter be subject only to retention elections; (3) by the end of 1994 there will be at least twenty-five black superior court judges and five additional blacks will be appointed to either the state court or the superior court; (4) in order to realize these numerical requirements, a new category of judgeships, “State Assignment Superior Court Judge-ships” may be created and filled by black candidates to serve by assignment in any of the State’s judicial circuits; and (5) any disputes that arise under this system in the future will be overseen by Senior District Judge Anthony A. Alaimo. (Consent Decree at 10-14.) The Plaintiffs and the Defendants (collectively referred to herein as the Proponents) signed this consent decree with the understanding that it would not become effective until it was approved by the Department of Justice and the Court. On September 25, 1992, the Panel granted a motion for the Intervenors to enter this litigation. The Intervenors, along with ami-cus curiae Thomas Burnside, argue that the terms of the proposed settlement violate their state and federal constitutional rights. On August 30, 1993, the U.S. Attorney General approved the settlement (pre-cleared the changes to the Georgia judicial electoral system contained in the consent decree), conditioned upon approval by this Court, and based on assurances by the Georgia Attorney General that the terms of the plan do not violate the Georgia Constitution. In November of 1993, the Panel denied a motion to allow interim gubernatorial appointments to certain judicial posts pending a final decision by the Court on the proposed settlement agreement. In that Order, the Panel also severed the Section 5 and Section 2 portions of this case, retaining control over the Section 5 claim and directed that the Section 2 claim be addressed by this Court alone. By the authority of that order, this Court must now decide whether or not to approve the proposed settlement agreement. On November 22, 1993, this Court certified the Plaintiff Class as all present and future black registered voters in Georgia and ordered that notice be given to absent class members pursuant to Rule 23(e). A fairness hearing was held January 12, 1994, at which the Court heard evidence and argument from the Plaintiff Class Representatives, the Defendants, the Intervenors, amicus curiae Thomas Burnside, and several objectors, acting in both their individual and representational capacities. It is important to note that this hearing was not a trial of the merits of this case. Nevertheless, the hearing provided an opportunity for interested parties to make their positions on the proposed settlement agreement known to the Court. After hearing this evidence and testimony, the Court is convinced that all points of view regarding the proposed settlement agreement have been represented and expressed to the Court. Analysis 1. The Court’s Role A. Generally Rule 23(e) of the Federal Rules of Civil Procedure provides that “[a] class action shall not be dismissed or compromised without the approval of the court.” Fed. R.Civ.P. 23(e). This rule is designed to ensure that class actions are not settled or compromised to the undue detriment of others, normally absent members of the plaintiff class, who will be affected by the terms of the compromise. In this case, the proposed settlement would affect not only members of the Plaintiff Class, but all persons in the State of Georgia. Thus, in addition to imposing an obligation on the Court to protect the interests of absent class members, Rule 23(e) requires the Court to protect the interests of all persons who would be affected by this settlement agreement and who are not adequately represented by the Parties to this litigation. In evaluating a proposed settlement agreement in a class action, a court must examine the totality of the circumstances and must determine, under that broad inquiry, whether the proposed settlement is fair, adequate, reasonable, and legal. E.g., Bennett v. Behring Corp., 737 F.2d 982 (11th Cir.1984); United States v. Marengo County Comm’n., 731 F.2d 1546 (11th Cir.1984) cert. denied, 469 U.S. 976, 105 S.Ct. 375, 83 L.Ed.2d 311; In re Corrugated Container Antitrust Litigation, 659 F.2d 1337 (5th Cir.1981); United States v. City of Alexandria, 614 F.2d 1358 (5th Cir.1980); Wainwright v. Kraftco Corp., 58 F.R.D. 9 (N.D.Ga.1973). The Proponents bear the burden of persuading the Court of the fairness, adequateness, reasonableness and legality of the agreement’s terms. The courts have formulated a host of factors to be considered in evaluating a proposed settlement in a class action, but several of the factors are geared toward eases in which monetary damages are sought from private enterprise and are thus inapplicable to this case. Those most relevant to a case like this are: (1) the circumstances surrounding the settlement; (2) the complexity, expense, possible duration, and present stage of the litigation; (3) the reaction to the settlement proposal; (4) the strength of the plaintiffs’ case; and (5) the reasonableness of the settlement proposal relative to the law and to the strength of the plaintiffs’ case. See, In re Armored Car Antitrust Litigation, 472 F.Supp. 1357, 1367 (N.D.Ga.1979) rev’d in part on other grounds, 645 F.2d 488 (5th Cir.1981) (citing Miller v. Republic Nat’l Life Ins. Co., 559 F.2d 426 (5th Cir.1977); City of Detroit v. Grinnell Corp., 495 F.2d 448 (2nd Cir.1974); see also, Bennett, 737 F.2d at 896 (citations omitted). In considering these factors, the Court must strike a balance between its need for sufficient information upon which to evaluate the proposed settlement and the limitations on its fact-finding abilities. The “trial court must not reach any dispositive conclusions on unsettled legal [or factual] issues, but at the same time must ‘attempt to arrive at some evaluation of points of law [and undisputed points of fact] on which settlement is based.’ ” Wainwright v. Kraftco Corp., 58 F.R.D. 9 (N.D.Ga.1973) (quoting West Virginia v. Chas. Pfizer & Co., 440 F.2d 1079, 1086 (2nd Cir.1971) cert. denied, 404 U.S. 871, 92 S.Ct. 81, 30 L.Ed.2d 115). B. Difficulties The standard under which the Court is to evaluate the proposed settlement is deceptive in its simplicity and clarity. In reality, this case is amazingly vexing, chocked full of analytical pitfalls. One of the primary difficulties in this case is developing an analytical framework that does not force the cart to lead the horse. This difficulty arises from the inherently circular character of the issues presented and the curious position in which they place the Court. The evaluation required for this case involves many highly interrelated factors that cannot be easily defined without reference to other factors. Consequently, the Court is left with the difficult task of analyzing a complex series of issues that has no logical starting point or ending point. Furthermore, this ease leaves the Court in a judicial quandary, torn between a preference for settlement over litigation and the obligations imposed by Rule 23(e). The Court normally prefers settlement over litigation for the simple reason that it is almost always in everyone’s best interest for parties to resolve their differences without resorting to the Court. This general principle remains true in class actions, subject to the requirements of Rule 23(e). The Court’s preference for settlement, however, is mitigated in this case by several realities including the importance of the issues involved, the far-reaching effects of the proposed settlement agreement, the less than unanimous approval regarding the agreement, the unique posture and function of the Court relative to this litigation, and the disputed factual record. The proposed settlement would undeniably alter the right of the people of the State of Georgia to vote for their judicial officers. The right to vote being one of the most fundamental and important rights in a free society, the Court must ensure that any alteration of that right is accomplished through legal and proper means. Moreover, while this proposed settlement takes the form of a consent decree, consent is not unopposed. The Proponents recommend this settlement as the best compromise to a difficult set of circumstances; however, the Intervenors and Objectors (collectively referred to as the Opponents) vehemently oppose it as violative of one of their most sacred and fundamental rights — the right to vote. The Court must, therefore, take this consent decree in the context that it is presented — a contested proposed resolution to a very difficult and important case. Notably, the Proponents have not presented the terms of this consent decree to the State legislature or to the people for approval, but rather they seek to invoke the injunctive and coercive power of the federal court to give effect to their agreement. Entry of a consent decree is not merely an agreement between litigants, it is a “judicial act” that requires the Court to exercise equitable discretion. League of United Latin American Citizens v. Clements (“LULAC”), 999 F.2d 831, 845-46 (5th Cir.1993) cert. denied, — U.S. -, 114 S.Ct. 878, 127 L.Ed.2d 74 (1994) (citing United States v. Swift & Co., 286 U.S. 106, 115, 52 S.Ct. 460, 462, 76 L.Ed. 999 (1932); Kaspar Wire Works, Inc., v. Leco Eng’g & Mach., Inc., 575 F.2d 530, 538-39 (5th Cir.1978); 1B James W. Moore et al., Moore’s Federal Practice ¶ 0.409[5]). Furthermore, since a consent decree potentially involves the Court’s power to sanction and enforce, the court “must not merely sign on the line provided by the parties,” United States v. City of Miami, 664 F.2d 435, 440 (5th Cir.1981) (en banc), but must instead exercise its equitable powers with caution. Accordingly, the Court must not blindly defer to the consenting parties and approve this consent decree, but must remain keenly aware of its responsibilities in evaluating this case. The Court’s analysis is made more difficult by the lack of an unconditional factual record. Indeed, there has been no trial of the merits of this case and the Defendants have not admitted liability under Section 2. Furthermore, the Proponents’ stipulations are not absolute — several contain disputed facts, others present facts in a form that is difficult for the Court to utilize, and all are expressly limited in their application. The lack of an undisputed factual record makes analysis in this case more difficult. The Court, therefore, must decide between its preference for settlement and its obligation under Rule 23(e) to protect the interests of those opposed to this settlement. This case has no singular hinge. It ultimately depends on the interaction of many parts and the considered judgment of this Court. In evaluating this case, the Court must consider five factors: (1) the circumstances surrounding the settlement; (2) the complexity, expense, possible duration, and present stage of the litigation; (3) the reaction to the settlement proposal; (4) the strength of the plaintiffs’ case; and (5) the reasonableness of the settlement proposal relative to the law and to the strength of the plaintiffs’ case. II. The Circumstances Surrounding Settlement The circumstances surrounding settlement do not pose a problem for the approval of the consent decree. In this analysis, the Court inquires into the circumstances surrounding the settlement to determine “whether the settlement negotiations were conducted in an adversarial or collusive manner.” In re Armored Car Antitrust Litigation, 472 F.Supp. at 1368. There is no evidence of collusion of any kind in this case. The Proponents underwent a long and arduous process of negotiation in reaching this compromise. In January and February of 1990, and again in April of 1991, numerous meetings were held between the Parties to discuss settlement possibilities. In April of 1992, settlement discussions were again convened under the leadership of Senior Federal District Court Judge Anthony A. Alaimo. Despite tremendous effort, it was not until June 17, 1992, that the Proponents reached agreement on the terms of this proposed settlement. All parties to these negotiations acted in good faith and should be commended on their persistence, dedication and attention to detail. This settlement agreement, therefore, does not appear to be the product of collusion and the circumstances surrounding the reaching of this agreement do not pose a problem to Court approval. III. The Complexity, Expense, Possible Duration of the Case, and The Present Stage of The Litigation The complexity, expense, duration, and stage of this litigation do not favor the approval of the consent decree. This case has been and continues to be truly vexing for the Parties, the citizens, the voters, and the Court. The issues presented are complex, the analysis has been long and laborious, and the expense of litigation to date has reached into the hundreds of thousands of dollars. Complexity, expense and long duration normally weigh in favor of the approval of a settlement. However, in this case, most of the complex issues have already been addressed, a large percentage of the total cost that will be associated with this case has already been spent, and the suffering associated with long duration has already been endured. Discovery is complete, a plaintiff class has been certified, the evidence has been largely compiled, and the case is all but ready to proceed to trial. Little time, work, or expense would be saved as a result of a settlement at this late stage. Hence, while the complexity, expense, and potential duration of this case would normally weigh in favor of approval of the proposed settlement agreement, the late stage of this litigation reduces the significance of those factors. IV. The Reaction To The Settlement Proposal The attitude of the people affected by a settlement should be taken into consideration in evaluating the fairness of a proposed settlement. See, Piambino v. Bailey, 610 F.2d 1306 (5th Cir.1980) cert. denied, 449 U.S. 1011, 101 S.Ct. 568, 66 L.Ed.2d 469; Hill v. Art Rice Realty Co., 66 F.R.D. 449 (N.D.Ala.1974) aff'd without opinion, 511 F.2d 1400 (5th Cir.1975). It is not necessary for all members of a plaintiff class to approve a proposed settlement. Indeed, a court may approve a settlement even if a considerable number of members of the plaintiff class disapprove of it. See, Bennett, 737 F.2d at 988. Even so, the number of objectors and the nature of their objections are relevant factors for the Court to consider in evaluating the proposed settlement. The Proponents negotiated and signed the proposed settlement agreement and obviously approve of its content. At the fairness hearing, the Proponents represented to the Court that the vast majority of the Plaintiff Class approved of this settlement. Class Representative Tyrone Brooks provided the Court with copies of letters of endorsement by many black community leaders and testified to the general approval of the settlement by members of the Plaintiff Class. Several other witnesses corroborated that testimony. The Intervenors oppose this settlement agreement on the grounds that it violates both Georgia statutory law and Georgia and federal constitutional law. Moreover, they contend that the Proponents lack the requisite legal authority to settle this case in the manner set forth in the consent decree. In their briefs, the Intervenors urged that because this settlement would affect all Georgia voters, the Court should consider, in addition to objections by members of the Plaintiff Class, any objections by Georgia voters who are not members of the Plaintiff Class. The Court agreed and the Court has received hundreds if not thousands of such objections. The crux of most of these objections is that the proposed settlement violates Georgia voters’ statutory and constitutional right to elect judges of their choice via contested elections. (E.g., Objection of Jane Streeter (filed January 13, 1994)). At the fairness hearing, the Georgia Alliance of African American Attorneys (“GAAAA”), a newly enacted state-wide black bar association represented by HJA Alexander, Michael Harvey and Randal Mangham, challenged the Plaintiffs’ contention that there is widespread approval of the settlement agreement. For instance, Mr. Alexander presented to the Court through the cross-examination of Representative Brooks that the voted position of the GAAAA was in opposition to the settlement agreement (Transcript at 57), that the letter of endorsement from the Gate City Bar Association did not represent the voted position of that organization (Id. at 58-59), that the DeKalb Lawyers Association presently oppose the settlement (Id. at 60), and that the endorsement of the Concerned Black Clergy may have been based on misunderstanding (Id. at 62). These representations were not rebutted and they raise serious questions about the approval of the terms of settlement by the members of the Plaintiff Class. Since the fairness hearing, the GAAAA has filed an objection on behalf of the organization as a whole (filed January 3, 1994) and another objection on behalf of at least one thousand members of the Plaintiff Class (filed January 26, 1994). Generally, these objections contend that the consent decree abrogates the right of Georgia voters to elect judges of their choice through contested elections, a right guaranteed by Georgia law and United States Constitution, and that the consent decree violates the Equal Protection Clause of the Fourteenth Amendment to the United States Constitution by establishing impermissible quotas and second class judges. The Court has also received objections from individual members of the Plaintiff Class. These objections mirror the concerns and arguments contained in the objections of the GAAAA. The Court concludes that the number and nature of objections filed in this case, while not dispositive in either direction, demonstrate that there exists significant opposition to this settlement agreement. In addition to considering the views of people affected by the settlement, the Court should consider and give some weight to the views of counsel. Hill, 66 F.R.D. 449. The Court has been fortunate to work with some of the most experienced voting rights attorneys in the country.. Few lawyers, if any, have the experience in the voting rights field as Plaintiffs’ attorneys Laughlin McDonald and Kathleen L. Wilde. Mr. McDonald is the Director of the Southern Regional Office of the ACLU and has litigated many voting rights cases since 1972. Ms. Wilde is a civil rights attorney with fourteen years of experience, eight of which include voting rights litigation. Both of these distinguished attorneys recommend this settlement to the Court. The Defendants have been represented throughout this litigation by Attorney General Michael Bowers and David Walbert. Mr. Bowers has extensive litigation experience in general and in the voting rights area. Although vehemently denying that the current system violates the Voting Rights Act and obviously concerned about the constitutional validity of the various provisions of the agreement, Mr. Bowers recommends this settlement to the Court. Mr. Walbert, an attorney with over twenty years of legal experience and significant experience in the voting rights field also recommends this settlement. In the other corner, Thomas Burnside, a highly respected attorney and amicus curiae in this case, strongly opposes the settlement as violative of the Georgia and Federal Constitutions. Additionally, Stanley House and Charles Wilkinson, attorneys for the Interve-nors, vigorously oppose the settlement on the same grounds. Finally, Randal Mangham, HJA Alexander, and Michael Harvey, representatives of the GAAAA and individual members of the Plaintiff Class, oppose the settlement as a disenfranchisement of minority voters and as an inadequate remedy to the problems they perceive with the current Georgia judicial electoral system. The weight of experience weighs in favor of this settlement, but that in itself is not overly persuasive or surprising. All of the attorneys who recommend this settlement represent the Parties who negotiated this agreement. Hence, the Court finds that counsel recommendations in this case are not particularly helpful. V. The Strength of The Plaintiffs’ Case The Court cannot conclude, based on the unsettled state of the law and the disputed facts in the record, whether the Plaintiffs would be any more likely to win than they would be to lose if this ease proceeded to trial. One of the most important factors for a court to consider in evaluating a proposed settlement under Rule 23(e) is the strength of the plaintiffs’ case — the probability that the plaintiffs would succeed if the case were litigated to a conclusion. See, Florida Trailer & Equipment Co. v. Deal, 284 F.2d 567 (5th Cir.1960); Parker v. Anderson, 667 F.2d 1204 (5th Cir.1982) cert. denied, 459 U.S. 828, 103 S.Ct. 63, 74 L.Ed.2d 65. The court must make this valuation determination, however, without the benefit of trial. Thus, the court must be careful not to overstep its bounds. “The trial court must not reach any dispositive conclusions on unsettled legal [or factual] issues, but at the same time must ‘attempt to arrive at some evaluation of points of law [and undisputed points of fact] on which settlement is based.’ ” Wainwright, 58 F.R.D. at 11 (quoting West Virginia v. Chas. Pfizer & Co., 440 F.2d 1079, 1086 (2nd Cir.1971)). In evaluating the merits of the Plaintiff’s case, the Court must consider the current Georgia judicial electoral system and the legal principles and factual assertions underlying the Plaintiffs’ case. A. The Present System The Constitution of the State of Georgia, and various statutes promulgated thereunder, provide for a system whereby judges of the state court, superior court and court of appeals and justices of the supreme court are elected in nonpartisan open elections. Ga. Const. Art. 6, § 7, ¶ 1 (1983). “Each superior court judge shall be elected by the electors of the judicial circuit in which the judge is to serve,” O.C.G.A. § 15-6-4.1 (1990), and “judges of the state court shall be elected by the qualified electors of the county or counties in which the court is located,” O.C.G.A. § 15 — 7—20(b) (1990). The procedures to be employed in conducting these elections “shall conform as nearly as possible” to the procedures governing general elections in which state legislators are elected. O.C.G.A. § 21-2-138 (1993). Georgia law, however, also empowers the Governor with general authority to appoint individuals to fill vacancies in public offices, Ga. Const. Art. 5, § 2, ¶ 8 (1983); O.C.G.A. 45-12-50 (1990), and with specific authority to appoint individuals to fill vacancies in the judiciary, Ga. Const. Art. 6, § 7, ¶ 3 (1983); O.C.G.A. § 15-7-23 (1990). The Georgia judicial electoral system involves aspects of both election and appointment. The vast majority of judges in this state have reached the bench via appointment. All judges and justices are subject to challenge in open elections at the expiration of their term of office. In reality, however, few incumbents are actually challenged in contested elections, and, of the few incumbents who are challenged, even fewer are defeated at the polls. Nevertheless, under the current system, qualified individuals can run against incumbent judges or justices in open elections and when that occurs, the voters choose who will serve them directly; the candidate having a majority of the votes in the election or the highest number of votes in a run-off wins. O.C.G.A. § 21-2-285.1 (1993). B. The Plaintiffs’ Claims 1. Generally In their Complaint, the Plaintiffs challenge the current Georgia judicial electoral system under the United States Constitution and the Voting Rights Act of 1965. In their motion for approval of the proposed settlement and in the accompanying briefs and arguments, the Proponents have focused their attention on their strongest claim — the Voting Rights Act claim. The Court will therefore focus on the Section 2 claim as well. The Plaintiffs’ primary challenge to the present system is that various components of the system dilute the strength of black voters in violation of Section 2 of the Voting Rights Act of 1965 (“VRA”). The. Plaintiffs also make a claim under Section 5 of the Voting Rights Act. As noted supra, in November of 1993, the three-judge panel severed the Section 5 and Section 2 portions of this ease and retained the Section 5 claim. Thus, this Court is only authorized to address the Section 2 claim. 2. Section 2 of The Voting Rights Act To make out a vote dilution claim under Section 2, a plaintiff must satisfy three threshold requirements: (1) the minority group, here African-Americans, must be sufficiently large and geographically compact to constitute a majority in a single-member district; (2) the minority group must be politically cohesive; and (3) the white majority must vote sufficiently as a bloc to cause it usually to defeat the minority’s preferred candidate. Thornburg v. Gingles, 478 U.S. 30, 50-51, 106 S.Ct. 2752, 2766-67, 92 L.Ed.2d 25 (1986); see also Nipper v. Smith, 1 F.3d 1171, 1177 (11th Cir.1993); LULAC, 999 F.2d 831 (5th Cir.1993). In addition to the threshold requirements, the Gingles Court provided a list, although “neither comprehensive nor exclusive,” of numerous other factors that may be relevant to the “totality of circumstances” analysis under Section 2. This list includes: (1) the extent of any history of racial discrimination in the state that touched on the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process; (2) the extent to which voting in the elections of the state is racially polarized; (3) the extent to which the state has used unusually large election districts, majority vote requirements, anti-single-shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group; (4) if there is a candidate slating process, whether minority group members have been denied access to that process; (5) the extent to which members of the minority group bear the effects of discrimination in such areas as education, employment, and health, which hinder their ability to participate effectively in the political process; (6) whether political campaigns have been characterized by overt or subtle racial appeals; (7) the extent to which members of the minority group have been elected to public office in the jurisdiction; (8) whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group; (9) whether the policy underlying the state’s use of such voting qualifications, prerequisite to voting, or standard, practice or procedure is tenuous. Gingles, 478 U.S. at 37, 106 S.Ct. at 2759; Nipper, 1 F.3d at 1177-78. To whatever degree the threshold factors and the list of additional factors are clear, that is where clarity ends. From that point forward, the waters are muddied and conflicts abound both among the Circuits and within the Eleventh Circuit itself. Among the Circuits, there is significant conflict regarding the correct evidentiary standards for analyzing Section 2 cases, particularly those involving judicial elections. For example, the Eleventh Circuit has recently held that the proper statistical pool for measuring minority electoral success in a judicial election context is the total percentage of blacks in the state population, not the lower percentage of blacks who meet the minimum requirements to become a judge. Nipper, 1 F.3d at 1183. The Fifth Circuit, in contrast, recently held that the correct statistical pool in a judicial election case is the percentage of black attorneys eligible to become judges. LULAC, 999 F.2d at 866. This conflict is not likely to be resolved in the foreseeable future as the Supreme Court has denied the petition for certiorari filed in the LULAC case. LULAC, — U.S. -, 114 S.Ct. 878, 127 L.Ed.2d 74 (1994). Within the Eleventh Circuit itself, the cases diverge “on the questions of whether plaintiffs can establish a Section 2 violation by proving the threshold Gingles factors, see, Solomon v. Liberty County, 899 F.2d 1012, 1017 (11th Cir.1990) (en banc) (per curiam) (Kravitch, J., specially concurring), cert. denied, 498 U.S. 1023 [111 S.Ct. 670, 112 L.Ed.2d 663] (1991), and whether defendants can thereafter raise a defense under the totality of circumstances. Id. at 1033 (Tjoflat, C.J., concurring).” Nipper, 1 F.3d at 1178, n. 11 (citing as recognizing division, Hall v. Holder, 955 F.2d 1563, 1568, n. 9 (11th Cir.1992)). Obviously, the Gingles factors have not been applied uniformly in all cases in the Eleventh Circuit. As a further indication of the uncertainty surrounding this analysis within the Eleventh Circuit, the mandate has been withheld in the Nipper case and a motion for rehearing is pending. In short, the state of the law surrounding Section 2 is in flux. To fulfill its role under Rule 23(e), however, the Court need not reach an absolute conclusion regarding whether the current system violates Section 2; rather, the Court must evaluate the Plaintiffs’ probability of success. This is not a precise analysis, particularly where the law is unclear — indeed, the Court is not to resolve conflicting legal issues at this time. Nonetheless, the Court can distill from the above discussion the basic principles underlying a Section 2 claim and it is based on those principles that the Court must proceed. Even if the law were clear, the factual record is deficient for the purpose of evaluating the merits of this case under Section 2. As noted supra, there has been no trial of the merits of this case, there has been no admission by the Defendants of Section 2 liability, and the Proponents’ stipulations are not unconditional or undisputed. Generally, the Proponents’ stipulations set out the history of Georgia voting law, the history of voting rights cases and racial issue litigation in Georgia, the history of Georgia public schools, the history of Georgia laws involving public accommodations, general demographic information and demographic information specific to the legal and judicial field, and an analysis of Georgia election results. The Court has meticulously studied the Proponents’ stipulations. It is wholly unnecessary, however, to recount the voluminous details of Georgia’s history in this Order. The history of the states of segregation practice and laws at all levels has been rehashed so many times that the Court can all but take judicial notice thereof. Generally, Georgia has a history chocked full of racial discrimination at all levels. This discrimination was ratified into state constitutions, enacted into state statutes, and promulgated in state policy. Racism and race discrimination were apparent and conspicuous realities, the norm rather than the exception. Yet, Georgia has come a long way since the adoption of the Voting Rights Act of 1965 and many of the evils of Georgia’s discriminatory history have been corrected. Unfortunately, some remnants remain. The Proponents’ stipulations are sufficient to raise a factual issue regarding what the Defendants have characterized as the “central disputed issue in this litigation” — whether the current Georgia judicial electoral system violates Section 2. (Defs.’ Resp. to In-tervenors’ Statement of Undisputed Facts at ¶ 14.) Specifically, the stipulations leave genuine issues of fact regarding two of the three Gingles threshold requirements and several of the other considerations set forth in Gingles. There remains an issue of fact regarding the satisfaction of the first Gingles threshold requirement. The first Gingles threshold requirement is that the minority group, here African-Americans, must be sufficiently large and geographically compact to constitute a majority in a single-member district. Gingles, 478 U.S. at 50-51, 106 S.Ct. at 2766. Stipulation number 140 addresses this requirement: 140. Plaintiffs would offer evidence at trial that if judges were elected by subdis-tricts created within the boundaries of existing [judicial] circuits, it would be possible to create districts which had a majority African-American population in 20 circuits which elected 38 of the 159 (23.9%) judges. Said evidence would show, however, that only 27 of those judgeships (17.0% of 159) in 12 circuits would have an African-American voting age population in excess of 55%, and only 20 (12.6%) in 6 circuits would have an African-American voting age population in excess of 60% (each of those six circuits including a major metropolitan area — Atlanta, Albany, Columbus, Decatur, Savannah, and Valdosta). (Pis.’ and Defs.’ Stipulations at ¶ 140). Stipulation number 141 also concerns this requirement: 141. Defendants at trial would offer evidence disputing and opposing that described in the preceding stipulation [number 140] on the grounds, inter alia, that the proposed subdistricts violate the principles recognized in Thornburg v. Gingles, 478 U.S. 30 [106 S.Ct. 2752, 92 L.Ed.2d 25] (1986) and Shaw v. Reno, [— U.S. -], 113 S.Ct. 2816 [125 L.Ed.2d 511] (1993), and that as a practical matter plaintiffs’ subdistricting plan would not enhance materially the opportunity for African-Americans to be elected as judges. (Pis.’ and Defs.’ Stipulations at ¶ 141). Clearly, the above two stipulations create a factual issue regarding the first Gingles threshold requirement. Compounding this factual uncertainly, the Intervenors and some of the objectors have represented to the Court that they will contest the Plaintiffs’ factual contentions if this ease is allowed to proceed to trial. There also remains a material issue of fact regarding the satisfaction of the third Gin-gles threshold requirement. That provision requires that the white majority must vote sufficiently as a bloc to cause it usually to defeat the minority’s preferred candidate. Gingles, 478 U.S. at 50-51, 106 S.Ct. at 2766. Stipulation number 73 provides that the courts have found racial bloc voting in some elections in eight Georgia counties. These findings are some evidence of racial bloc voting. However, evidence of some racial bloc voting in eight of the one hundred fifty-nine Georgia counties, alone, is "probably insufficient to sustain a conclusion regarding racial bloc voting in the State of Georgia as a whole. In addition to the evidence of racial bloc voting in eight counties, “plaintiffs would offer evidence that voting rights challenges resulting in changes to at-large election systems have been brought in approximately 38 cities and 56 counties in Georgia between 1974 and 1990” (Pis.’ and Defs.’ Stipulations at ¶ 74), and “that forty-two (42) other voting cases have been brought in the thirty judicial circuits covered by the Department of Justice’s objection letter of April 25, 1990” (Id. at ¶75). In contrast, “[defendants would offer evidence that the aforesaid cases include cases not based on the Voting Rights Act and cases not alleging racial discrimination.” (Id.) Again, these stipulations present factual conflicts. Thus, there remains an issue of fact regarding whether there is enough evidence of white bloc voting to satisfy the third Gingles threshold requirement. The stipulations leave questions, too, regarding several other relevant considerations set out in Gingles. See, Gingles, 478 U.S. at 37, 106 S.Ct. at 2759; Nipper, 1 F.3d at 1177-78. In particular, the stipulations do not answer, or their answers conflict, with regard to: the extent to which voting in the elections of the state is racially polarized; the extent to which the state has used unusually large election districts; whether political campaigns have been characterized by overt or subtle racial appeals; whether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group; and whether the policy underlying the state’s current voting system is tenuous. 3. Peripheral Issues The Proponents argue that there have been several determinations adverse to the Defendants that justify settlement in this case irrespective of Section 2 — namely, this Court’s ruling that Section 5 applied to all superior court judgeships that were created after November 1, 1964, and the U.S. Attorney General’s objections regarding those same judgeships. This Court’s previous ruling, however, did not establish any Section 2 liability. Brooks I, 775 F.Supp. 1470, aff'd, 498 U.S. 916, 111 S.Ct. 288. Moreover, the Attorney General’s objections were based largely on the State’s failure to provide adequate information to the Department of Justice regarding the positions in question. The Attorney General ultimately found that the State failed to satisfy its burden under Section 5 of proving that the new judgeships would have neither the effect nor the purpose of discriminating on account of race. That finding, however, does not establish a Section 2 violation. Therefore, neither of these adverse determinations expressly found that the current Georgia judicial electoral system violates Section 2, and neither are the equivalent of such a determination. As noted above, the Court must proceed in this case based on the reality that there has been no determination that the current system violates Section 2 and that the Court cannot make such a determination at this point. Yet, even if the Court could make such a determination, or if such a determination were made after a trial of the merits, the Court would not be free then to fashion any remedy. Rather, the Court would be required, with the aid of the parties, to devise a remedy narrowly tailored to cure the problem that resulted in Section 2 liability. Given the discussion and holdings herein, the proposal embodied in the consent decree would not satisfy that requirement. C. Conclusion In conclusion, there are a myriad of disputed factual issues that must be resolved before the Court can reach an answer to the ultimate question of whether the current system of electing judges in Georgia violates Section 2. The Court cannot resolve these factual issues in the context of this Order. Thus, given the uncertainty of the law and the disputed state of the facts, the Court cannot say with any certainty or comfort that the Plaintiffs would be more likely to prevail at a trial of the merits than to lose. As a practical matter, therefore, the Court must proceed based on the reality that there has been no determination that the current system in Georgia violates Section 2 and that the likelihood of that conclusion, should this case proceed to trial, is questionable. VI. The Reasonableness of The Settlement Proposal In this section, the Court’s role comes full circle. The Court is left, in essence, with the inquiry with which it began — whether the proposed settlement is fair, adequate, reasonable, and legal under the circumstances. In making this inquiry, the Court, now with the benefit of the above analysis, must examine the virtues and downfalls of the proposed settlement agreement. Thus, after examining the power of the Governor and Georgia Attorney General to settle this case, the Court will present the proposed settlement agreement, analyze its terms under state and federal law, and ultimately determine whether the Court should approve it. A. The Power of The Defendants to Settle The Governor and the Attorney General have the power to settle this case provided that the settlement terms do not violate the law. The Attorney General of the State of Georgia is the chief legal advisor to the executive department of the State and is charged with the duty to represent the State in civil and criminal eases. GA. CONST. Art. V, See. 3, Par IV (1983); O.C.G.A. §§ 45-15-3, 45-15-34, 45-12-26 (1990). The Governor is the chief executive of the state. GA. CONST. Art. V, Sec. 2, Par. I (1983). The Parties have not presented, nor has the Court uncovered, any specific constitutional, statutory, caselaw, or other authority, that expressly grants to the Georgia Attorney General the power to settle a law suit on behalf of the State. Nonetheless, the mere absence of an express grant of power is not always determinative. The Court finds that as a general principle, inherent in the duty of the Attorney General to represent the State is the power to settle a law suit against the State. To hold otherwise would be to require every law suit against the State to be litigated, as Mr. Bowers expressed at the fairness hearing, “to the hilt of the sword.” (Tr. at 236.) Nothing in the law or in common sense suggests that requirement. The Attorney General’s power to settle, however, is no different than that of any other attorney. No lawyer can settle a case in contravention of the wishes of his clients or in contravention of the law. This is not a case where the Attorney General seeks to settle over the objection of his clients and to thereby impose his policy preferences over those of his state official clients. See LULAC, 999 F.2d at 840-41. In this case, the clients, namely the state Defendants and the Governor, approve of and support the consent decree. The inquiry therefore turns to the legality of the provisions of the consent decree. Clearly, no party, including the1 consenting Parties in this case, and no attorney, including the Attorney General and the other attorneys involved in this case, can settle a ease where the terms of the settlement violate the law. •Mr. Bowers argues that in a VEA case like this one, the Attorney General should be able to- settle even if the terms of the settlement violate state law, provided that, in good faith, he perceives a significant risk of losing the litigation if it proceeded to trial. Mr. Bowers’ reasoning is as follows. The Georgia Attorney General is sworn not only to support Georgia’s Constitution and laws, but the Constitution and laws of the United States as well, and, in the event of a conflict between federal and state.law, the Supremacy Clause would require his allegiance to federal law over state law. From that premise, Mr. Bowers argues that if the Attorney General has a good faith belief that a Georgia law violates federal law, he should not have to endure the added time and expense of a trial before giving effect to the federal law. Mr. Bowers concludes that if the Attorney General has a good faith belief that the State has a significant possibility of losing an action challenging a state law under a federal law, he should also not have to endure the expense and time of trial, but should be allowed to settle the case. Mr. Bowers is a highly principled public servant and .the Court does not question his sincerity. The Attorney General’s argument fails in its leap of logic between its final premise and its conclusion. It is true that when there is a direct conflict between state and federal law the Supremacy Clause requires state law to yield. The Supremacy Clause, however, does not operate until a direct conflict is shown. As indicated supra, there has been no determination to date that the current Georgia system for the election of judges and justices violates Section 2. Without such a determination, the Supremacy Clause does not operate to invalidate the Georgia judicial electoral system. A perceived risk of failure at trial, even a good faith perceived risk of 25-40% as Mr. Bowers estimated at the fairness hearing, (Tr. at 285), is simply insufficient to establish a direct conflict between state and federal law. The Supremacy Clause, therefore, does not enlarge the power of the Attorney General regarding settlement at this time. Ultimately, therefore, the power of the Attorney General, and the power of all parties for that matter, to settle a case such as this one depends on the legality of the provisions of the settlement agreement. The attention of the Court now turns in that direction. B. The Consent Decree The proposed settlement, which takes the form of a consent decree, would work fundamental changes in the Georgia judicial electoral system. Under the consent decree, all current judges and justices who seek additional terms would not be subject to challenge in open elections, but would, instead, be subject only to non-partisan ■ “retention” elections. (Consent Decree p. 10, ¶ 3.) These “retention” elections would be held at the same time as the general election and would require an affirmative vote of a majority of those voting on the question to retain a judge. (Id.) All current vacancies and all vacancies that arise as a result of the failure of a judge to be retained would be filled by gubernatorial appointment from a list of qualified individuals selected by the Judicial Nominating Committee (“JNC”). (Id. at p. 11, ¶ 6-7.) These appointed judges, after serving a minimum of two years, would be subject to regular “retention” elections. (Id. at p. 10, ¶ 3.) Additionally, by December 31, 1994, the State would be required to have no fewer than 25 black superior court judges and would also be required to appoint five other black judges to either the state or superior court. (Id. at p. 10, ¶ 2.) In order to assist in meeting these numerical requirements, a new category of judgeships, “State Assignment Superior Court Judgeships” would be created and qualified black candidates would be appointed to these judgeships to serve in any of the State’s judicial circuits by designation. (Id. at p. 11, ¶ 8.) Subsequent to January 1, 1995, the JNC would recommend judicial candidates to the Governor without regard to race and both the JNC and the Governor would be prohibited from discriminating on the basis of race. (Id. at ¶¶ 11, 12.) The goal of the State, however, would remain a diverse judiciary reflective of the population of the State as a whole. (Id. at ¶ 15.) C. The Consent Decree and Georgia Law The consent decree would violate both the language and spirit of the 1983 Georgia Constitution and the laws promulgated thereunder. Specifically, the system created under the consent decree would violate the Georgia Constitution’s requirement that judges be elected, impermissibly alter the structure of power currently embodied in the 1983 Georgia Constitution regarding the election of judges, and violate several fundamental Georgia statutes. 1. “Election” Versus “Retention Election” The consent decree calls for a change from the current system of potential contested elections to one of “retention elections.” This change cannot be accomplished under the current Georgia Constitution. The 1983 Georgia Constitution provides, in relevant part: All superior court and state court judges shall be elected on a nonpartisan basis for a term of four years. All Justices of the Supreme Court and Judges of the Court of Appeals shall be elected on a nonpartisan basis for a term of six years. Ga. Const. Art. 6, § 7, ¶ 1 (1983). The language “elected on a nonpartisan basis” is not defined in the Georgia Constitution, statutes, case law, or any other binding or persuasive source. Nonetheless, the Georgia Code does provide some relevant definitions from which it can be inferred that “election” does not contemplate a “retention election.” Common usage, a previous unofficial opinion of the Georgia Attorney General and several Georgia cases support that conclusion. The Georgia Code does not provide for retention elections. The term “election” is defined to mean “any general or special election.” O.C.G.A. § 21-2-2(4) (1993); see also O.C.G.A. § 21-5-3(8). “General election” means “an election recurring at stated intervals fixed by law,” O.C.G.A. 21-2-2(7), and normally refers to an election held for members of the General Assembly. Brackett v. Etheridge, 190 Ga. 216, 9 S.E.2d 275 (1940). “ ‘Special election’ means an election that arises from some exigency or special need outside the usual routine.” O.C.G.A. § 21-2-2(28). Nowhere in Georgia law is the term “retention election” present or defined. To read what the term “retention election” implies into the language of the 1983 Georgia Constitution or any of the statutes enacted thereunder violates basic rules of statutory and constitutional interpretation. Furthermore, the plain meaning of the term “election” does not include the rather idiosyncratic concept of a “retention election.” Retention elections are just not what one thinks of when she hears the word “election.” Black’s Law Dictionary defines the term “election” as: the act of choosing or selecting one or more from a greater number of persons, things, courses, or rights. The choice of an alternative.... The selection of one person from a specified class to discharge certain duties in a state, corporation or society. Black’s Law DICTIONARY 464 (5th ed. 1979). The operative part of that definition was adopted by the Georgia Supreme' Court in Poythress v. Moses, 250 Ga. 452, 453, 298 S.E.2d 480 (1983). Interestingly, within the same definition, Black’s distinguishes between “election” and “appointment”: With respect to the choice of persons to fill public office ... the term means in ordinary usage the expression by vote of the will of the people or of a somewhat numerous body of electors. “Election” ordinarily has reference to choice or selection by electors, while “appointment” refers to a choice by an individual. BlaCk’s Law DICTIONARY 464 (5th ed. 1979). These definitions and their adoption by the Georgia Supreme Court support the conclusion that the term “election” does not, in plain language, common usage, or in the 1983 Georgia Constitution, embrace the concept of a “retention election.” In 1983, Attorney General Bowers himself reflected on the question of whether retention elections would satisfy the 1983 Georgia Constitution. He concluded as follows: As the Missouri State Plan does not reserve to the voter the opportunity of choosing “one or more from a greater number of persons,” it does not constitute an “election” in the ordinary signification of that term. Rather, the submission of such a question [to retain a judge or not] is more closely akin to a “referendum.” Historically, the terms “election” and “referendum” are not synonymous. Accordingly, it is my unofficial opinion that the constitutional requirements of Art. VI, Sec. VII, Par. I, could not be met by the adoption of the Missouri state plan by statute, but would require specific constitutional authorization. 1983 Op.Atty. Gen.Ga. 205 (Jan. 11, 1983). Notably, Mr. Bowers incorporated and partially relied on the Black’s Law Dictionary definition of the term “election” discussed above in his Opinion. When the Court questioned Mr. Bowers about that Opinion at the fairness hearing, he responded: ‘Tes, sir. I stand by that.” (Tr. at 254.) The Attorney General’s Unofficial Opinion supports the Court’s conclusion that the term “election,” as the term exists in the 1983 Georgia Constitution, is not broad enough to encompass a retention election system. Moreover, some Georgia case law indicates that retention elections would not satisfy the requirements of Georgia law because of the restrictive ballot used in those elections. For example, in a case challenging part of a municipal charter restricting electors to choosing only names of candidates that appeared on the ballot, the Georgia Supreme Court held that such a restriction violated the Constitution: “the individual elector has the unshackled right to write on the ballot any person he wishes to vote for, and can not be restricted to a choice between those whose names are provided on the ballot.” Thompson v. Willson, 223 Ga. 370, 373, 155 S.E.2d 401 (1967) (citing Stewart v. Cartwright, 156 Ga. 192, 198, 118 S.E. 859 (1923); Griffin v. Trapp, 205 Ga. 176, 182, 53 S.E.2d 92 (1949)). While that ease was decided under the 1945 Georgia Constitution, and in a somewhat different context than the present, its holding remains the law of this State. Under the consent decree, the electorate would not only be “restricted to a choice between those whose names are provided on the ballot,” but would be restricted to one name on the ballot. This restriction would not satisfy Georgia law. Hence, while the term “election” as used in the 1983 Georgia Constitution is not expressly defined, the Georgia Code, common usage, the Black’s Law Dictionary definition, an unofficial Attorney General opinion, and several Georgia cases all indicate that the term is not broad enough to encompass the retention election system set forth in the consent decree. The Proponents argue, however, that the language “elected on a nonpartisan basis” is broad enough to encompass retention elections. Essentially, the Proponents support their assertion with two arguments. First, the Proponents point to the historical variation in what constituted judicial elections in Georgia. Historically, judicial elections have not always meant a contest between multiple candidates voted on by the people. In some instances, the 1877 Georgia Constitution, for example, judicial elections were conducted by majority vote of the General Assembly. Ga. Const. Art. VI, Section II, Para. IV, IX (1877). In other instances, the General Assembly would provide by statute how new judgeships would be filled. E.g., Ga. Const. Art. V, Section II, Para. I (1976); Ga Const. Art. VI, Section III, Para. I (1945). In still other instances, judges were appointed by the governor. E.G., Ga. Const. Art. V, Section II, Para. Ill (1945). Based on this varied past, the Proponents contend that there is no deep-rooted right recognized in Georgia constitutional law for the people to elect their judges in contested elections. This argument is unavailing. Although the term “election” is not defined by the 1988 Constitution, and while Georgia has not always selected judges in the manner it does so now, the term “election” has never — expressly or impliedly — meant “retention election.” Elections throughout Georgia history have always involved a choice between known candidates. Additionally, Georgia’s varied history of judicial elections does not indicate that Georgia voters do not consider this right important or that Georgia voters are somehow not vested in their own system. To the contrary, Georgia voters have a nearly century long history of electing their judges and justices. The details have not always remained constant, but the citizens of Georgia have created a system, embodied it in a constitution, and ratified it in accordance with the law. The citizens of Georgia are entitled to live under the system they designed and ratified until that system is declared unconstitutional, contrary to federal statutory law or is modified through the amendment process. Second, the Proponents assert that the term “election” is unrestrictive and thus should be read broadly, at least broadly enough to include retention elections. This argument is also not persuasive. The “Missouri Plan” existed for quite some time before the 1983 Georgia Constitution was drafted. In fact, there is evidence that the people involved in the drafting of the 1983 Constitution considered the “Missouri Plan” and disagreed over its merits. See, Tr. of Meetings, Committee to Revise Article VI, volumes I through III, Aug. 22,1980, at 81-82, reported in, Ekrhart, supra n. 11, at 32-34. Since there is evidence that retention elections were considered and, after consideration, not specifically enacted, there is no reason to read the term “election” more broadly than its common definition. Moreover, as indicated above, the Georgia Code, common usage, the Black’s Law Dictionary definition, an unofficial Attorney General opinion, and Georgia cases all cut against the Proponents’ argument. In conclusion, a retention election cannot satisfy the current constitutional requirement that judges be elected. “Under our system of government the method of expressing the will of the people is by voting in a legally held election.” Wheeler v. Board of Trustees, 200 Ga. 323, 334, 37 S.E.2d 322 (1946). Under the present Georgia constitution, “retention elections” would not be legally held elections. The consent decree system, therefore, would violate Georgia law. 2. The Power Structure The consent decree also violates Georgia law by impermissibly altering the structure embodied in the Georgia Constitution that defines the rights and power of the Governor, the voters, and qualified potential candidates. These alterations remove power from the people, both as voters and as qualified potential candidates. a. The Power of the Governor The consent decree modifies the role of the Governor in the Georgia judicial electoral system. The Opponents submit that this change violates the Georgia Constitution. Historically, the power of the Governor to appoint judges and justices has varied greatly. Under some previous Georgia constitutions, the 1877 and the 1945 Constitutions for example, the Governor was afforded broad powers of appointment, while under other constitutions, the Governor’s power was more restricted. The Georgia courts have strictly construed the powers of appointment granted to the Governor. E.g., Hooper v. Almand, 196 Ga. 52, 58-60, 25 S.E.2d 778 (1943); Patten v. Miller, 190 Ga. 123, 124, 8 S.E.2d 757 (1940); Mulcay v. Murray, 219 Ga. 747, 755, 136 S.E.2d 129 (1964). The 1983 Constitution empowers the Governor with broad general authority to appoint individuals