Full opinion text
OPINION Before GIBBONS, Circuit Judge, WRIGHT and LAYTON, Senior District Judges. CALEB M. WRIGHT, Senior District Judge. This case arises under the continuing jurisdiction of this Court to implement prior opinions finding unconstitutional segregation in the public schools in Delaware. The instant opinion concerns the choice of an appropriate remedy for constitutional violations in the operation of the schools of Wilmington and the surrounding suburban districts and the requirement of an inter-district remedy. In prior opinions, this Court ruled that the segregation of the Wilmington schools was never erased; and that this segregation resulted from a combination of factors, including demographic and housing patterns initiated and supported by state action; and the redrawing of school district lines during a period of consolidation and reorganization under the Educational Advancement Act, 14 Del.C. §§ 1001, et seq Having found a violation which included inter-district effects, we ruled that under the guidelines laid down in Milliken v. Bradley, 418 U.S. 717, 94 S.Ct. 3112, 41 L.Ed.2d 1069 (1974), [hereinafter “Milliken ”] the Court could consider both inter- and intra-district remedies for the constitutional deprivation. 393 F.Supp. at 446-47, aff’d. per curiam, 423 U.S. 963, 96 S.Ct. 381, 46 L.Ed.2d 293, 44 U.S.L.W. 3299 (1975). Following the ruling of this Court on the liability phase, the parties were directed to develop plans to remedy the violations found, using both inter- and intra-district methods. These plans were to be submitted through the State Board of Education and its staff, the Department of Public Instruction, for professional comment and evaluation. The State Board was then to recommend to the Court the proposals it thought best. These original suggestions, some nineteen in number were submitted to the Court in August, 1975. Three weeks of evidentiary hearings followed. During the course of those hearings, additions, changes and shifts in emphasis were made to many of the plans. Those presently before the Court, therefore, differ in several respects from those originally submitted. All the factual and legal issues have been briefed, and the matter is ready for decision. I. Factual Background There is no need to discuss here either the prior history of segregated schools in Delaware or New Castle County, nor to recount in detail the prior findings of this Court, since they are set out in prior opinions. Nonetheless, a short discussion of some of the facts found in prior opinions will aid in the understanding of the opinion. New Castle County, Delaware, is some four hundred forty-three square miles in area, and northern New Castle County, the area primarily concerned in the opinion, is some two hundred fifty-one square miles in area. This area of Northern New Castle County has a public school population of 80,678 of whom 63,370 or 78.5% are white; 15,722 or 19.4% are black; and the remainder of whom are other minorities including American Indian (89 or .1%), Hispanic (1,120 or 1.38%) and Oriental (377 or .46%). Within the area, 11,733 or 74.6% of the black students attend school in the Wilmington District. If the only other majority black school district in the area, DeLa-Warr, is added to Wilmington, 13,473 or 85.6% of the black students in the northern County area attend school in those two districts. The apparent contrast can be made even plainer by comparing the Wilmington and DeLaWarr enrollments to the other Northern New Castle County districts. Wilmington Public Schools are 84.7% black; DeLaWarr Public Schools are 54.9% black. No other Northern New Castle District is less than 90% white, and most are significantly higher. Delaware has a long history of using small school districts, and it became apparent during the post-war period that many of these districts could not support a full educational program. In common with many other states, Delaware has, over the years, consolidated many of these smaller districts, some for financial reasons, others for purposes of required desegregation. A comparison of a map of the school districts comprising the northern County area shows some nineteen school districts in 1959 to twelve districts in the same area today (excluding the separate Vocational-Technical District). The last major reorganization, carried out under the Educational Advancement Act, as described in the last opinion, 393 F.Supp. at 438 passim, included the consolidation of several small districts into the presently existing districts. This Court ruled that the exclusion of Wilmington from the process of reorganization by statute was an unconstitutional racial classification, and “ ‘contributed to the separation of the races by . redrawing school district lines.’ ” The variety of plans submitted as remedies may be grouped for analytical convenience into three broad categories. Certain of the plans submitted are “voluntary plans” of varying scope. These include so-called free transfer provisions, and magnet schools. A second category of plans calls for the reorganization of the districts in the area, dividing the black population among the new districts or attendance area and having new school boards make assignments of the students within the area. These plans range from one proposed by the State Board which would divide the area into five new districts; several variations which would include smaller or greater areas; and a county-wide plan designed to consolidate the whole area into one district. The last category is a set of mandatory assignment plans providing for the transfer and transportation of students among the existing districts. The plans vary in the area to be included, in the transportation, and in the amount of time students would actually spend in desegregated experiences. The Court has considered all these plans as well as the testimony adduced at the hearings, in the light of the requirements of Milliken and Swann, and on that basis has determined the remedy to be followed here. II. Class Representation The State Board of Education urges that the present class plaintiffs are not properly before the Court. The motion was brought on two grounds: first, that the named representatives of the class had left or been graduated from the Wilmington School System, and therefore the issues were mooted as to them; second, that in any event, at least insofar as remedy was concerned, the named plaintiffs and their counsel did not fully, fairly and adequately represent the interests of the class, since (it was alleged) not all black school children and their parents in Wilmington desired particular forms of remedy favored by the representatives. Professor Moore lists four categories which the Court must consider before determining that the proposed representation is adequate: (1) whether the interests of all parties are “coextensive”; (2) whether the interests are “antagonistic”; (3) the proportion of the named representatives to the class as a whole; and (4) any facts bearing upon the ability of the named representative to speak for the class as a whole. In addition to those factors, the Court notes that the present action must continue in any event, since the Wilmington School Board intervened as a party plaintiff and has full right to pursue the action. Moreover, even if the named representatives were to be found to be “unrepresentative” of a 11 Wilmington school children, they might well represent a substantial sub-class of the group. See Rule 23(c)(4). Here it would appear that the interests are “coextensive”, since all the potential plaintiffs have made clear that they desire an end to segregatory actions on the part of the State. The point pressed by the State is that because some Wilmington parents of black children do not support the plan favored by the named plaintiffs, “antagonistic” interests are present, and therefore, the class is improperly represented. First, it is not clear that in these circumstances, the exact nature of the remedy proposed is an interest which cuts to the subject matter of the suit. See generally, 3B Moore’s Federal Practice ¶ 23.07[3] at 23-404. All potential representatives of plaintiffs who have appeared before the Court have agreed that some remedy is required, and all seek relatively broad remedies. The remedy ordered by this Court must of necessity apply to all black children within the City. Since any remedy ordered in favor of the purported sub-group represented by the named plaintiffs would be determinative of the rights of all, the question is really not the antagonism of interests, but whether the Court has had a full and fair presentation of all possible views on the matter. We, therefore, hold that the named representatives are at the very least part of a class whose rights have been violated. Through the presentations of all parties to this suit, and that of all amici, the Court has been thoroughly informed of the differing views on remedy. The action may therefore continue as a class action, and the class will continue to be that class whose rights were violated, the school children of Wilmington. Should it, anytime in the future, appear that antagonistic interests would prevent fair adjudication, the Court may take the necessary steps of requiring the addition of new parties to fill any representation “gaps”. See Rule 23(d) Fed.R. Civ.Proc. III. The Legal Standard of Milliken and Swann Brown v. Board of Education, 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954), was a major step in American constitutional law, and it is from that case primarily, and its progeny Green v. New Kent County Board of Education, 391 U.S. 430, 88 S.Ct. 1689, 20 L.Ed.2d 716 (1968), and Keyes v. Denver School District No. 1, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548 (1973), that the determination of the existence of a violation of the rights of children in segregated schools is made. That determination has been made in this case, in opinions that date from 1956 to 1975. Milliken and Swann deal not with the existence of a violation, but with the standards which a court must follow in remedying the violation which it has found. They are expressions not of constitutional law, but of the proper scope of equitable discretion. Moreover, Milliken and Swann must not be read in isolation. As the Court in Milliken noted, harking back to both Swann and Brown II, the flexibility and pragmatic judgment of the chancellor of equity have been the consistent pattern of determining remedy in desegregation cases. Milliken and Swann are not, therefore, substantial departures from prior case law, nor steps backward in the development of remedies for desegregation. Rather, they establish the limits of the equitable discretion to be exercised by the court. Thus, Milliken makes plain that the remedy to be ordered must be commensurate with the scope of the violation which has been found. And insofar as the remedy to be applied is the desegregation of schools, the violation which was found must be proximately related to the operation of the school system, or to its enrollment. Further, if it is shown that the substantial disparity in enrollment patterns between districts was substantially caused by governmental activity, an inter-district remedy may be considered. The mere fact that an inter-district violation occurred does not necessarily require an inter-district remedy. It is too longstanding a rule of equity to require citation that although equity will give complete relief, it will limit the exercise of its power to a remedy which is reasonably necessary and likely to succeed. Moreover, an inter-district violation having only de minimis effects will not require school desegregation across the district lines. See 418 U.S. at 750, 94 S.Ct. at 3130, 41 L.Ed.2d at 1094. We need not catalogue the set of violations with regard to housing and zoning set forth in the last opinion affirmed by the Supreme Court. Nor need we rehearse the reasons why we held the Educational Advancement Act to be an unconstitutional “redrawing of district lines.” It suffices to say that the acts described in the prior opinions were the acts of the State and its subdivisions, and had a substantial, not a de minimis, effect on the enrollment patterns of the separate districts. The suburban districts have attempted to foreclose the application of an inter-district remedy including them by citing the prior finding of this Court that each of them was at present operating a unitary system, and urging that they had committed no constitutional violation. Such a defense is inadequate where, as here, the local boards are creatures of the State, and it was the State Legislature and the State Board of Education which acted in a fashion which is a substantial and proximate cause of the existing disparity in racial enrollments in the districts of Northern New Castle County. The fact that birth rates, or population shifts, or other factors also contributed to a degree will not relieve the State from its obligation to desegregate The remedy for the violation must include school districts which are its instrumentalities and which were the product of one of the violations. The remedy for the acts of the State may be inconvenient, burdensome, and expensive to some of those instrumentalities, but neither inconvenience, burden nor expense can negate the duty of the Court to order effective relief when a not insubstantial violation has been shown. The suburban districts have urged that under the standard laid down by the plurality opinion in Milliken, no inter-district remedy can be ordered to include them, unless they themselves have been guilty of a violation. The suburban districts emphasize that it was not their actions which had any segregatory effect in Wilmington, and that we did not in the last opinion hold that the Educational Advancement Act was drawn with an actual segregatory intent. See, Milliken, supra, 418 U.S. at 745, 94 S.Ct. at 3127, 41 L.Ed.2d at 1091. That claim, however, must fail, because it misread both the language of Mil-liken and the prior holding of this Court. First, the Chief Justice’s opinion in Milliken made clear that action by the State which caused inter-district segregation would be sufficient to allow an inter-district remedy. The further specification of the deliberate drawing of lines to achieve segregation was by way of example, not limitation. As the. Supreme Court made clear in Wright v. Council of Emporia, it is not the purpose for which the lines are drawn that is determinative of whether they work an impermissible classification, but whether the effect is such that they do. See 407 U.S. 451, at 461-62, 92 S.Ct. 2196, at 2204, 33 L.Ed.2d 51, at 60 (1972). The concurring opinion of Mr. Justice Stewart in Milliken, necessary for the majority holding, makes it even clearer that an inter-district remedy may follow wherever it was shown that “state officials had contributed to the separation of the races by drawing or redrawing school district lines.” 418 U.S. at 755, 94 S.Ct. at 3132, 41 L.Ed.2d at 1097. The activities found in the prior opinion clearly met that standard, and those findings have been affirmed. Nor is the liability of the State merely derivative. See Milliken, 418 U.S. at 748, 94 S.Ct. at 3129, 41 L.Ed.2d at 1093. Rather, here the State actively contributed to the separation of the races. Nor do we find Mr. Justice Stewart’s opinion in Gautreaux to be a significant change from his earlier view. Where the State has contributed to the separation of races by redrawing school lines, necessarily the districts on both sides of the lines are part of the violation itself, and exclusion of the suburban districts cannot be predicated on their own purported innocence when their present lines were drawn or redrawn in the course of a violation. Moreover, the last opinion also found that although the suburban districts now provided a unitary system for all children within their districts, past activity on their part had not been so confined. The actions of the suburban districts therefore meet the test laid down in Gautreaux for inter-district relief, independently of the State’s actions For present purposes, it is sufficient to point out, as we found in the last opinion, that despite the separate operation of the systems since the 1950’s, the racial characteristics of the city and the suburbs are still inter-related, and the actions of state officials and local officials were sufficient to create an inter-district effect under Milliken. Our duty is to order a remedy which will place the victims of the violation in substantially the position which they would have occupied had the violation not occurred. Prior opinions of the Supreme Court have held unequivocally that where the violation found resulted in the operation of a dual school system, the Court must order the “greatest possible actual degree of desegregation”, consistent with the practicalities of the situation, Davis v. Mobile County Bd. of School Comm’rs., 402 U.S. 33, 37, 91 S.Ct. 1289, 1292, 28 L.Ed.2d 577, 580 (1971); and it must do so by a plan which is reasonably certain to achieve desegregation now In light of this standard, we must determine whether sufficiently remedial actual desegregation is possible under a plan which would be confined to the boundaries of the present Wilmington district; what level of desegregation will at a minimum establish that a dual system is no longer operating; how wide a geographical area must be included to accomplish properly that result; and by what method and with what safeguards this result can be accomplished. It is to these questions that we turn next. IV. Wilmington Only Plan All of the defendants in the instant phase of the case have urged vigorously that the Court has no power to order an inter-district remedy absent further findings of inter-district violation. They have also urged that a plan which involves only the Wilmington schools will be adequate to remedy the violations found in the prior opinions. Since a finding that such a plan was adequate would relieve the Court of the necessity of dealing with many vexing questions, we deal with this issue first. The Wilmington School District, the boundaries of which are coterminous with the city lines, presently serves 13,852 students, of whom 11,733 or 84.7% are black and an additional 746 or 5.4% of whom are Hispanic. The system as a whole is therefore 90% minority enrollment. Only two schools within the system are presently majority white: Cedar Hill Elementary School which is 73.4% white; and Highlands Community School which is 59.6% white. All of the other schools are not only majority black, but heavily so. As we noted in an earlier opinion, some of those schools which were black schools de jure in 1954 have continued to have enrollments over 91% black. The plan proposed to cluster certain of the elementary schools in an effort to re-' duce the black enrollment; and to redraw the “feeder” lines for the middle schools, and the attendance boundaries of the two high schools. It was drawn in great part by the Wilmington School Board, although some changes and additions were made by the staff of the State Department of Public Instruction. As submitted, the plan is necessarily limited in its effect by the population characteristics of the area it would cover. Nonetheless, it would result in the schools which were formerly de jure minority schools being desegregated to the extent of bringing their enrollments below 90% black for the first time. Other schools would, however, remain more than 90% minority. We must, therefore, determine whether such a level of minority enrollment in the context of the present case would be sufficient to constitute a remedy for the violations found. The Supreme Court has made it clear that there is no magic in percentages of minority enrollment. See Swann, 402 U.S. at 25-26, 91 S.Ct. at 1280, 28 L.Ed.2d at 571; Milliken, 418 U.S. at 740 — 41 & n. 19, 747 n. 22, 756, 94 S.Ct. at 3125, 3128, 3132, 41 L.Ed.2d at 1088, 1092, 1097 (Stewart, J., concurring). Such percentages are merely the starting point in the formulation of a remedy and the determination of whether the remedy is complete. Swann, 402 U.S. at 25, 91 S.Ct. at 1280, 28 L.Ed.2d at 571. Moreover, such figures are meaningless unless seen in the proper context: they must be weighed in light of the characteristics of the community. Id.; and see Wright v. Council of City of Emporia, 407 U.S. at 464-65, 92 S.Ct. at 2204, 33 L.E'd.2d at 62 (1974). If the community whose characteristics would be determinative of the weight to be given to the enrollment figures were seen as Wilmington alone, such a plan might afford the relief which the Constitution requires. The fact that the schools would continue to be heavily black while the suburbs would be white would not alone require an inter-district remedy. Milliken, 418 U.S. at 756, 94 S.Ct. at 3133, 41 L.Ed.2d at 1097 (Stewart, J.,” concurring). This Court, however, cannot use such a narrow view of the community, in light of the scope of the prior violations. Gautreaux, - U.S. at-, 96 S.Ct. at 1547, 47 L.Ed.2d at 805, 44 U.S.L.W. at 4485. The Supreme Court has made it clear that the determination of the violation is a key factor in determining remedy. Swann, 402 U.S. at 16, 91 S.Ct. at 1276, 28 L.Ed.2d at 566; Milliken 418 U.S. at 744, 94 S.Ct. at 3127, 41 L.Ed.2d at 1090. We have already determined that the State had not fulfilled its mandate to operate a unitary school system; and that in the past the area comprising the suburban and city districts was treated jointly for many school purposes, including the transportation of black students to de jure segregated schools. In these circumstances, it is apparent that the entire northern New Castle area must be treated as one community in terms of its population characteristics, because that is the way it was perceived and treated by the State and its citizenry. Gautreaux, -U.S. at-, 96 S.Ct. at 1547, 47 L.Ed.2d at 805, 44 U.S.L.W. at 4485; Keyes v. Denver School District No. 1, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548 (1973). While other districts in this area were being consolidated or considered for consolidation, the Wilmington District was expressly reserved from the exercise of discretion by the State Board, and that reservation “played a significant part in maintaining the racial identifiability of the Wilmington and suburban New Castle County school districts.” Moreover, state action was also found to be responsible here for the racial identifiability of the suburbs and the city, through enforcement of racial covenants, zoning and encouragement and support of private discrimination in housing. For these reasons, the effectiveness of any plan must be judged by how well it terminates racial identifiability of the schools in light of the population characteristics of the northern county area. Gautreaux,-U.S. at-, 96 S.Ct. at 1547, 47 L.Ed.2d at 805, 44 U.S.L.W. at 4485. Cf., Swann, 402 U.S., at 25, 91 S.Ct. at 1280, 28 L.Ed.2d at 572. To hold otherwise in the present context would mean that state officials could, by their actions, effectuate a pattern of discrimination in schools and housing, then argue that the area had become divided into separate communities identified by them, and that only the population characteristics of their selected area need be taken into account. The school population of Northern New Castle County is presently 78.5% white. A plan limited to the confines of Wilmington would result in city schools that would be, for the most part, 85% to 95% black, while the suburban schools, other than those of DeLaWarr, remained overwhelmingly white. Thus, a Wilmington-only plan would not significantly affect the present racial identifiability of the Wilmington or suburban schools. It is the duty of the State to bring forward a plan that will achieve desegregation, to the “greatest possible actual degree”, Swann, 402 U.S. at 26, 91 S.Ct. at 1281, 28 L.Ed.2d at 572; and here that duty will not have been carried out when the schools will maintain their former racial identity. We do not here decide that the fact that Wilmington would be a black system surrounded by white systems is sufficient to call for an inter-district remedy. See Milliken, 418 U.S. at 735, 747 & n. 22, 94 S.Ct. at 3122, 3128, 41 L.Ed.2d at 1085, 1092. Nor do we determine that a majority black system is of necessity segregated. See United States v. Scotland Neck Board of Education, 407 U.S. 484, 92 S.Ct. 2214, 33 L.Ed.2d 75 (1972). Rather, we hold only that insofar as the Wilmington-only plan is concerned, where inter-district violations have been found, it is appropriate to look at the population of the area over which the violations occurred to determine in the first instance whether the plan submitted results in actual desegregation. See Gautreaux, - U.S. at -, 96 S.Ct. at 1547, 47 L.Ed.2d at 805, 44 U.S.L.W. at 4485. Where the plan would result in the maintenance of the traditional racial identity previously established by State action, and that disparity in racial enrollments remains substantial, it cannot be said that it results in the disestablishment of a dual system. See United States v. Scotland Neck Board of Education, 407 U.S. 484, 490, 92 S.Ct. 2214, 2217, 33 L.Ed.2d 75, 80. We, therefore, hold that under the proof developed before us, a Wilmington-only plan would not remedy the violations previously found, and that we must go on to consider an inter-district remedy. V. Proposed Inter-district Remedies The Court has had before it several different types of plans for remedying the violations found, some of which have been recommended in varying degrees by the parties, and others of which were proposed by various individuals and submitted through the State Board. We will not take the time to list seriatim all the various proposals, which are part of the record, but it is appropriate to note that not one of the plans submitted has been found to be completely acceptable taking into consideration all the practicalities of the situation. Davis v. Mobile County Bd. of School Comm’rs., 402 U.S. 33, 37, 91 S.Ct. 1289, 1292, 28 L.Ed.2d 577, 580 (1971). We make the following findings with regard to the various types of inter-district plans. A. Voluntary Plans The category of voluntary plans includes several different proposals. A proposal on which much evidence was adduced calls for the creation by the suburban districts and Wilmington of so-called “magnet schools” to serve five “zones” which would be drawn up by combining a portion of Wilmington with one or more suburban districts. The racial make-up of the school population of each zone would approximate the racial characteristics of the county as a whole. All of the presently existing school districts would be retained, and, in addition, each “zone” would have an “Advisory Committee” which would be responsible for implementation of magnet proposals. The actual operation of individual magnet programs in schools would be the responsibility of the district in which the building was physically located. D.DPI.Rem. # 8 at 5-6. The central concept in all of these voluntary proposals is that some schools are or can be made so attractive that students will enroll in those schools for the particular program rather than in the neighborhood school to which they would ordinarily have been assigned. The magnet schools proposed are characteristic of the concept: every school in every district would eventually house some sort of magnet program thought by the residents to be both educationally desirable and desegregatory in its attraction. As originally advanced, no binding racial quotas were to be included, although the State Board agreed at trial that they might be used in addition to the program designs. Educators who testified in favor of the plan all agreed that they hoped the plan would work but that there was no way that success could be guaranteed. They sought instead, time to try the plan. In determining whether any voluntary plan developed in the testimony meets the requirements of a desegregation plan, the Court has had to consider the goals which any plan ordered by the Court would be required to meet; and then determine whether the plan proposed offers adequate assurance that the goals would be met. The Supreme Court has made clear that time is a commodity whose place has become restricted in desegregation cases. “ ‘The time for mere “deliberate speed” has run out,’ . . .. The burden on a school board today is to come forward with a plan that promises realistically to work, and promises realistically to work now.” Moreover, the plan must not offer merely a hope of some desegregation at some time in the future. “The district judge or school authorities should make every effort to achieve the greatest possible degree of actual desegregation, and will thus necessarily be concerned with the elimination of one-race schools.” The magnet plans proposed here simply do not measure up to those requirements. We do not hold that magnet plans are incapable of achieving reasonable levels of desegregation. In a different context, with different safeguards, such a plan might be sufficient. See Hart v. Community School Board, 512 F.2d 37 (2d Cir. 1975). However, in other contexts, magnet plans similar to the one here have been disapproved. See, Brinkman v. Gilligan, 503 F.2d 684, 704 (6th Cir. 1974); Spangler v. Pasedena Board of Education, 375 F.Supp. 1304, 1307 (C.D.Cal.1974), aff’d, 519 F.2d 430, 438-39 (9th Cir. 1975), cert. granted, 423 U.S. 945, 96 S.Ct. 355, 46 L.Ed.2d 276, 44 U.S.L.W. 3279 (1975). The magnet system called to the attention of the Court by the State Board, that beginning operation in Houston, shows relatively little success in actually desegregating schools: it has in fact resulted in some schools receiving an even greater minority enrollment. The suburban districts which brought forward “expansions” of the outline made by the State were unable to assure the Court that it would actually result in significant desegregation. Moreover, the State Board failed to cost out its proposal adequately. In light of the cost of operating programs in other cities, the cost figures given to the Court in the “expansions” seem unreliable indeed. The magnet program is heavily dependent upon the unique drawing power of particular programs and faculty to attract and hold students. There is necessarily a limited market for special programs. While magnets might be used to desegregate individual schools otherwise not part of a segregated system, their use as the sole means of system-wide desegregation is decidedly unpromising. See Bradley v. Milliken, 402 F.Supp. 1096, 1147 (E.D.Mich.1975). Absent a showing that significant desegregation must occur In fact as a part of the operation of magnet schools, the Court cannot accept the plan. No such showing has been made. The other voluntary plans proposed to the Court suffer from similar defects. As Green, supra, made clear, voluntary programs are unacceptable where there are “reasonably available other ways . . . promising speedier and more effective conversion to a unitary, nonracial school system . . ..” None of the other voluntary plans proposed promises realistically to work effectively now, and those plans lack even the modicum of attraction inherent in the magnet system. Thus, assuming that a voluntary program could be developed in an effort to meet reasonable desegregation goals, the proposals presented to this Court are not adequate to assure meeting that goal, and are therefore inadequate as remedies. B. Cluster and Center Plans The Wilmington School Board, as Intervening Plaintiff, submitted a major proposal [“the Wilmington Metro Plan”] which would require the transportation of a sufficient number of students across the district lines to desegregate some of the area schools. See, Int.P.Rem. # 1, 8. The enrollments would be determined by “clustering” particular city schools with given suburban schools, so that each school within the cluster would have approximately the same racial characteristics. In order to accommodate all the children involved and to spread the burden among suburban and city children, grade patterns would be altered so that schools would serve part of a given elementary or high school level. The Wilmington Metro Plan proposes that all of the existing districts be retained, at least initially, to insure the fair treatment of the students being transferred, and to assist in the organizational transition from a dual to a unitary system. The State Board and the suburban districts are very strongly opposed to the cluster plan, denominating it “forced busing” which they call a “bankrupt concept”. The Wilmington School Board takes the position that it is the only plan which actually insures the re-assignment of students to desegregated schools, and is therefore the only plan presented to the Court which meets the requirements of Green. The only other plan submitted to the Court which effectively reassigns students is the so-called “Center Plan” submitted by the Alfred I. DuPont School District. The Center Plan involves the part-time re-assignment of elementary children, so that those in grades one through six would attend neighborhood schools for all of their academic subjects, and for one day per week would meet in desegregated “Centers” for training in e. g., physical education, music and art. Children in the seventh through the twelfth grades would be mandatorily assigned full time to desegregated schools, all seventh graders in the area being taught in the present city schools, and all eighth through twelfth grades being taught in present suburban schools. Where desegregation is to take place in one school district, the use of clusters and pairings, as well as other sorts of re-assignment and transportation schemes offer the best guarantee that actual desegregation will take place. The Supreme Court has made it clear that such plans are a proper part of an overall remedy. See, Swann, 402 U.S. at 29-30, 91 S.Ct. at 1282, 28 L.Ed.2d at 574. But the Court has also made it clear that all of the circumstances of a particular locality are to be taken into account when the District Court makes its determination of what plan to order. The cluster and center plans both include inter-district assignments. We must agree with the Supreme Court that an inter-district transportation plan standing alone is difficult to administer, and fraught with complex problems unsuited for judicial determination. In particular, we note that in Delaware local districts have had a long history of control over curriculum and text choice, as well as the common governmental functions of setting and levying taxes, and maintaining communication and accountability between administrators and parents. The transportation schemes would endanger all of these functions, or make their performance much more difficult. The mere fact of extra administrative burden is not, of course, sufficient to make a plan so impractical or burdensome as to be beyond the equity power of the Court. Nonetheless, where methods are available which will accomplish the result without the associated problems, the Court, as a matter of equitable discretion, should follow that course which will require its intervention in the least possible degree which will insure compliance. See generally, Milliken, 418 U.S. 740-45, 94 S.Ct. 3125, 41 L.Ed.2d 1088; Gautreaux, - U.S. at -, 96 S.Ct. at 1543, 47 L.Ed.2d at 797, 44 U.S.L.W. 4483-85. The Wilmington Metro Plan, though containing many admirable features and safeguards with regard to minority rights, would place the Court in the ongoing position of general supervisor of education in New Castle County. In the event of disagreements over curriculum patterns or textbooks, the Court or a master would have to step in. Moreover, the relationship of párents to the school their child attends, and the interests which they take in the operation and policies of that school can be an important determinant of the success of the child in education. Except in the Wilmington and Alexis I. DuPont Districts, school board members are elected. See 14 Del.C. §§ 1051, 1062, 1063. The Wilmington Metro Plan would make it much more difficult for individual parents to require accountability from teachers and administrators who are employed by districts other than that of their voting residence.. In light of all the circumstances here, the Court cannot in the exercise of equity order the cluster plan as proposed. Transportation schemes, as well as the redrawing of attendance lines and other shifts in present patterns of attendance, will undoubtedly be required to desegregate. But such plans are better drawn where the greatest possible degree of control is in the hands of local leaders acting in accordance with constitutional limitations. The Center Plan suffers from the same defects, at least insofar as it would assign students from one district to schools under the supervision of a district over which their parents have little control or political voice. Moreover, the Center Plan, whatever the purpose of the originators, would place an inequitable burden on the present staff of the Wilmington schools, requiring, for example, that present high school teachers switch to seventh grade, or scramble for the positions which might open up in their field in the suburban districts. In addition, while accomplishing the desegregation of the secondary level schools, the mere part-time desegregation for the elementary grades is inadequate as a final result of a desegregation plan under the standard established by Green and Swann. Keyes v. Denver School District # 1, 521 F.2d 465, 477-79 (10th Cir. 1975). No cluster or center plan which has been proposed would meet the test of administrative feasibility which is, in our view, an absolute prerequisite to the long-range success of any remedy. C. Reorganization Plans The reorganization plans submitted by various parties propose correcting the prior violations by redrawing the boundary lines found to have been drawn improperly. That is, they attempt to combine Wilmington and various of the suburban districts into new consolidated districts. It is expected that the populations of these districts will reflect the racial character of the area as a whole, and that following the effectuation of the remedy, the new districts could assign pupils within their localities on a desegregated basis and otherwise operate a unitary system. The differences among the reorganization plans proposed pose two issues: (1) whether the area should be split into several new districts or consolidated into one large district; and (2) the extent of the geographical area to be included in the reorganization. The State Board and the suburban districts have introduced a reorganization plan in which the Wilmington District is split into five parts. Five new districts are established, each consisting of one or more suburban districts and one-fifth of Wilmington. Two of the suburban districts have sought to include in the five district reorganization plan the Newark District. Other than changing the pupil and geographical ranges, the inclusion or exclusion of Newark would not affect the operation of the plan. We will, therefore, discuss the’ plan in light of its other characteristics, since we deal infra with the extent of geographical area to be required. After reorganization, the operation of the proposed five districts would be as present Delaware law requires. New elections for board members would eventually have to be held, but in virtually all respects, the proposal follows the reorganization scheme approved by the State Legislature and described in the last opinion. The only departures from the existing state scheme set forth in 14 Del.C. §§ 1001 et seq., would be that in contravention of provisions of the Act, Wilmington would be split; enrollments in certain of the reorganized districts would exceed 12,000; and the salaries of teachers and staff in the reorganized district would not be “leveled up”. The only change required by the inclusion of Newark would be that one or more districts other than Wilmington would be split. The other major proposal for reorganization was that submitted by the DeLa-Warr District, which would require that the whole of New Castle County be included in one school district which would be further divided into four or more attendance areas. The area or district would be supervised by an elected board, and each of the attendance areas would have elected or appointed advisory groups which would maintain contact with the citizenry. It is the position of DeLaWarr that by including the entire county, several educational benefits would be achieved as well as desegregation of the schools. The tax assessments and collections across the county would be equalized for all the schools, and the population of the entire county would necessarily be reflected in the population of the school district and the schools. If the population were to shift in ways not now foreseen, the county-wide district could shift attendance zones to ensure the continued operation of a unitary system over the whole county. The suburban districts other than DeLa-Warr do not favor a reorganization, but would prefer it to transportation plans implemented without such shift in control. All of the parties, except DeLaWarr, have generally opposed the county-wide plan as creating a major shift in the way Delaware has historically operated its schools. De-LaWarr’s preference for its plan is at least partially based on its own unique position. Among the districts in New Castle County, DeLaWarr is relatively poor since not only is its population of a substantially lower income level, but a major part of the district’s tax base is exempt from taxation because of public uses of the property, including the approaches to the Delaware Memorial Bridge, etc. The change to a county-wide system would enable DeLaWarr to benefit from the better tax assessment and property ratios of the other suburban districts. The power of the Court to order a reorganization would not appear to be in doubt. Justice Stewart’s concurring opinion in Milliken makes explicit reference to a decree calling for the restructuring of districts as an appropriate remedy for certain kinds of violations. See 418 U.S. at 755, 94 S.Ct. at 3132, 41 L.Ed.2d at 1097; and see Gautreaux, - U.S. at -, 96 S.Ct. at 1545, 47 L.Ed.2d at 800, 44 U.S.L.W. 4484 & n. 12. The Supreme Court has made clear in other contexts that the courts have the power to prevent a reorganization or creation of new districts when to allow the shift would recreate dual systems. See Wright v. Council of City of Emporia, 407 U.S. 451, 92 S.Ct. 2196, 33 L.Ed.2d 51 (1972); United States v. Scotland Neck Board of Education, 407 U.S. 484, 92 S.Ct. 2214, 33 L.Ed.2d 75 (1972). Moreover, as noted by the state defendants, reorganization may well be peculiarly suited as the remedy in the instant case, where one of the violations which was found was the prior improper reorganization of districts including some of those now before the Court. The standard formulated by the Court is that the remedial decree should be directed toward placing the victims of the violation in the position they would have occupied had the violation not occurred. Where one of the violations was the isolation of Wilmington from the possibility of union with other districts, prima facie an appropriate remedy would be ordering of the union to take place. We have determined that the violations found in the instant case present a situation where some sort of consolidation or reorganization is required. There is ample testimony that without reorganization of some kind, no plan will be able to function in an administratively feasible manner. We will, therefore, require the implementation of a reorganization plan. Nonetheless, for the reasons which follow, the Court cannot order any of the reorganization plans as proposed. The Supreme Court in Milliken mentioned some of the major problems in consolidating districts, including assessing taxes and setting of election periods. Those decisions are difficult, but in the main, standards exist either in state statutory materials or court decisions which could be used by the equity court to determine some of these issues, and whether proper discretion had been exercised by the state and local officials. The decisions we are asked to make here are much more difficult, and much more open to question with little guidance from state law or federal constitutional guarantees. The State Board and the suburban districts would have us determine, e. g., where present districts should be cut, and what weight should be given to equalizing tax assessments or populations. Although the State Board has presented a plan, during testimony it admitted that, in light of more recent figures, it would probably recommend somewhat different lines be drawn. In view of the major differences in opinion with regard to the inclusion or exclusion of Newark, the issues of transportation time and population levels of appropriately sized districts ought to be dealt with explicitly by State education officials. Had the State Board’s plan or any other plan been drawn with reference to specific criteria formulated after consideration of the task to be accomplished, and was therefore a reasonable method of accomplishing the goal by persons given such power in the state system, the Court could then view the matter as requiring only a determination of whether discretion was properly exercised in drawing the lines, etc. Absent such criteria, we feel that the more proper course is to create a situation which will not freeze the district lines by court order, but will create a framework within which the State can make a future determination of proper districts for the area, while insuring that actual desegregation will take place. The State Board has claimed that its plan for reorganization does nothing but follow the method approved by the State Legislature in the Educational Advancement Act, and that therefore some weight should be given to it. First, as we noted above, the proposal varies from the statute in certain respects, and we cannot now determine by hindsight what weight was attached to those portions of the statute with which the proposal differs. The State Board’s proposal splits at least one district, despite the bond problems which might be caused thereby. We do not know and cannot now guess how the State Legislature would have desired to handle such a problem. The State Board treats rather lightly the size limitation of reorganized districts imposed by the Legislature. It may well be that size is no longer the problem it was once thought to be, but we have no evidence upon which we can say with finality that the State Legislature would treat it as such. Similarly, to impose a county-wide system although it would offer significant advantages not only to DeLaWarr but also to the Court, would be to order a major shift in Delaware school policy. If such a shift were necessary to remedy a constitutional violation, it would be permissible, but as we find in our treatment of geographic area, infra, pp. 353-354, no such showing has been made here. Nonetheless, as we noted supra, some reorganization is required. The Court must at a minimum determine the districts which will be included in such a reorganization, and make provisions for the governance of the area in the event that the State officials fail to act. We note that our opinion in this regard is not a final determination of the organization of the area and of the lines to be followed in setting up such an area, as would be the ease if we were to order one of the reorganization plans proposed to us. Rather, the reorganization outlined infra is effective only in absence of proper state action to change it. Of course, if the state or local officials were to act in such a manner as to defeat or block desegregation under the guise of a shift in the reorganization plan, the Court would be forced to review the State’s action in light of the requirements set out in Wright v. Council of City of Emporia, 407 U.S. 451, 92 S.Ct. 2196, 33 L.Ed.2d 51 (1972); and United States v. Scotland Neck Board of Education, 407 U.S. 484, 92 S.Ct. 2214, 33 L.Ed.2d 75 (1972). In determining how such a reorganization would best be accomplished in light of our limited objectives, we have, whenever possible, attempted to comply with state law. It would seem clear that we are not bound by state law, for this is a matter of a federal remedy for a violation of federal rights. See Louisiana v. United States, 380 U.S. 145, 154-56, 85 S.Ct. 817, 822, 13 L.Ed.2d 709, 715 (1965); Haney v. Board of Education of Sevier County, 429 F.2d 364, 368 (8th Cir. 1970). However, we read Mil-liken as requiring that the equity court in ordering desegregation remedies give proper deference to the traditions and acts of the states in setting up educational units. 418 U.S. at 741-42, 94 S.Ct. at 3125, 41 L.Ed.2d at 1089; and see Gautreaux, - U.S. at-, 96 S.Ct. at 1543, 47 L.Ed.2d at 797, 44 U.S.L.W. at 4483-84. Such decisions are far better left to legislators and the process of compromise than to the rigors of judicial determination. 418 U.S. at 744, 94 S.Ct. at 3126, 41 L.Ed.2d at 1090. We have, therefore, followed state law in determining the extent of reorganization and the method by which districts will be included, except where those provisions would prevent a remedy from becoming effective. Determinations of methods of governance, and the day-to-day operations of the schools will be left in the hands of appropriate local officials. This Court should have no need to interfere in those decisions, unless they violate federal law or constitutional provisions. In any event, the initial determination will be by those who are by training, expertise and experience qualified to make such judgments. See Milliken, 418 U.S. 744, 94 S.Ct. 3126, 41 L.Ed.2d 1090. VI. Remedy As we have made clear, reorganization or consolidation of some portion of the existing districts is required, since otherwise an inter-district plan would fail by reason of its administrative burden. While there are undoubtedly problems in the management of large school districts, the problems of curriculum coordination, planning, teacher and staff control and supervision, policy implementation, and coordination of school programs with the needs of the community can be better met by a district with sufficient power to meet the problems which will arise, rather than by having 11 or 12 different districts arrange separate solutions. Moreover, it will allow the parents of children attending the schools to have some voice and control over their child’s education by the power to elect school board members. In reviewing the evidence, it has become plain that the Court must choose between following the provision of state law which confined reorganized districts to enrollments of fewer than 12,000; and that provision which sought to limit reorganization to whole districts. The reasons for the first limitation were several, including a desire to maintain small units which would require a lower administrative overhead, and to continue the use of small locally elected school boards. We have already found that there is substantial professional disagreement over whether the 12,000 population figure is necessary to achieve the administrative economies sought, and the present testimony has reinforced that view. Moreover, as we have already noted, even the State Board in its suggestions to us did not find the 12,000 figure a controlling limitation. Local community control is, of course, an important feature of American education, and we are required to give deference to it unless circumstances dictate otherwise. Milliken, 418 U.S. at 741, 94 S.Ct. at 3125, 41 L.Ed.2d at 1089, and see Gautreaux, - U.S. at -, 96 S.Ct. at 1545, 47 L.Ed.2d at 800, 44 U.S.L.W. at 4484-85. We find that circumstances here indicate that the required change in local control need not be so substantial as to give rise to major problems. The 12,000 pupil limitation would require us to split existing districts, and would for all practical purposes make any reasonable reorganization impossible. The change which we require infra, although initially setting up a large district, is not only subject to appropriate subdivision for local control over issues of policy in particular schools, or local initiative with regard to curriculum, etc., but is also subject to redivision into smaller governmental units by action of the State, so long as such subdivision does not result in the frustration of the desegregation objective. Even if the State should decline the opportunity, experience in other states has shown that districts of the size proposed infra can be effectively administered. The alternative in the instant case would be for the Court to attempt to redistribute the population and tax ratables of the area by drawing new district lines. Although such an operation is theoretically possible, and might be accomplished in reliance on the evidence presented by the State Board and other parties, we are unwilling to impose our view of the most reasonable distribution of tax base and population levels. Even if we were to assume, as the State Board evidently did, that Wilmington should be treated separately as the source of the “problem” and that no other district lines need be shifted, we would be required to determine which alternative presented by the parties was the “fairest” way to split that district. For the Court to become involved in such a task in the first instance, would be to raise many of the problems which the Supreme Court in Mil-liken found to be unsuited for judicial determination. 418 U.S. at 743-44, 94 S.Ct. at 3126, 41 L.Ed.2d at 1090. And see San Antonio School District v. Rodriguez, 411 U.S. 1, 41, 93 S.Ct. 1278, 1301, 36 L.Ed.2d 16, 47 (1973). The problem of splitting the liability on bonds and other outstanding obligations also raises difficult questions of how much of a tax “break” to allow to some of the districts, and an examination of the Educational Advancement Act makes clear that the problem is not one previously determined by the State Legislature. For these reasons, the proper course for this Court is to consolidate whole districts, subject to the power of the Legislature to redivide the areas in the future, using appropriate lines which are non-racial in their purpose and effect. A. Extent of Area Having determined that only whole districts are to be included in the plan; that the 12,000 pupil limitation on enrollments does not apply; and that this Court is not the proper agency to subdivide the area into new districts, we turn to the question of which districts must be included in the remedy to be ordered. It is apparent, of course, that this will affect the question of what will constitute a prima facie desegregated school, and will affect the administrative and other operational burdens. This determination is primarily a question of whether two outlying districts, Appoquinimink and Newark are to be included within the scope of the plan. The decision as to Appoquinimink is relatively easy. Only DeLaWarr’s plan sought to include Appoquinimink, and then only for the sake of logical consistency in urging a “County-Wide” plan, since Appoquinimink would be treated as a separate “attendance area”. Another district partially within the county was excluded since it has no school buildings in the area. As Milliken emphasized, the task before the Court is to remedy the violation found. 418 U.S. at 738, 94 S.Ct. at 3124, 41 L.Ed.2d at 1087. The Court cannot conclude on the present evidence that Appoquinimink would in any way have properly been part of the exercise of discretion of the State Board in eliminating the dual schools at the time of the Educational Advancement Act. The district was created by the reorganization, as a combination of the former Odessa and Middletown districts. Moreover, the district is apparently operating a unitary system which is approximately 70% white. Its schools are located in the approximate center of the district, a substantially greater distance from the major black population centers in Wilmington and DeLaWarr than any other district. Since its inclusion would in any event be of very little impact on the existence of predominately white or black schools in other areas of the county, we have determined that Appoquinimink need not be included. Newark School District has also urged that it is too far distant from Wilmington to make its inclusion proper, saying inter alia, that transportation time would increase greatly. See, Swann, 402 U.S. at 30-31, 91 S.Ct. at 1282, 28 L.Ed.2d at 574. It has also urged that Mount Pleasant and Conrad seek to include it for an impermissible reason, viz, the prevention of white flight to Newark following the desegregation of other schools in the area. The mere existence of racial disparity in the enrollments of neighboring districts is not a constitutional violation; and thus the existence of such a disparity is no grounds for ordering a remedy. It is clear from other cases, however, that the potential of white flight may be included in the exercise of the Court’s informed discretion on what would constitute an appropriate remedy. We have already determined that a constitutional violation existed at the State level, and moreover, we have said that the effects of the pre-Brown segregation to which Newark was a party have not yet been dissipated. Based on the evidence presented to us, we think it clear that Newark is not so far from some portions of Wilmington that a workable desegregatory school attendance plan could not be implemented. In fact, transportation times even on the longest and most unlikely routes were not so significantly long as to endanger health or welfare of students. See, Swann, 402 U.S. at 30-31, 91 S.Ct. at 1282, 28 L.Ed.2d at 574. The witnesses from Newark agreed that the present bus route times were not dispositive because the routes would have to be redesigned in light of whatever new assignments were made, and that such assignments would most likely be made to include only those schools which were geographically near. It is difficult to say with any certainty that Newark would have been included in any reorganization had the State Board been entitled to exercise its discretion in 1968. Since Newark at that time had close to 12,000 students, the effect of the enrollment limitation may have been to foreclose Newark’s inclusion. On the other hand, had the Legislature or the State Board considered desegregation as one of the appropriate goals to be accomplished in the course of reorganization, very different criteria might have led to the consolidation of part of either Wilmington or DeLaWarr with part of the present Newark district. We do not, however, rest our holding on such post hoc rationalizations, and on what might have been. Rather, uncontradicted testimony indicates that the stability of any desegregation plan is enhanced by the inclusion of larger geographical areas and higher white populations. The Court cannot ignore the fact brought so forcefully to its attention that desegregation is costly, in ways beyond dollars spent on additional equipment and training. The difficulties of declining tax bases, and the problem of preventing growth areas from maintaining the duality of schools in the Northern New Castle County area require the inclusion of Newark. We note for the sake of clarity that the inclusion of Newark in the area to be desegregat