[170] BALDOCK, Circuit Judge, dissenting.

It is not by accident that the court begins its opinion in the liability phase of this case with a quote from the plaintiffs' expert, Mr. William Lamson. For in deciding that the U.S.D. 501 school board operates a dual school system with respect to student assignment, the court has relied heavily on the plaintiffs' theory of the case as supported by the testimony of Mr. Lamson. In deciding that the board operates a dual system with respect to faculty assignment, the court has adopted the plaintiffs' theory in large part, but impermissibly has made its own factual findings after an ad hoc evaluation of the evidence. See, e.g., Court's Opinion at 870 ("We do not adopt plaintiffs' standard, but instead evaluate the data on its face.").

In this case, the plaintiffs contended that the school board operated a dual school system as evidenced by: 1) racial identifiability in student and faculty/staff assignment, 2) inferior facilities, equipment, curriculum and instruction at schools with higher black enrollments, 3) lower minority test scores, and 4) survey attitudes indicating that some schools were perceived as white schools and others as minority schools. See generally rec. vol. I, doc. 242 (amended complaint). On the other hand, the defendants contended that the plaintiffs' measures of "racial identifiability" alone, given the historically small percentage of minority students and faculty/staff did not indicate a current condition of intentional segregation based upon either 1) present discrimination or 2) a failure to eliminate the vestiges of past state-imposed segregation. Defendants suggested that de facto and racially neutral considerations, rather than de jure segregation, explained the assignment of students and faculty/staff. Defendants also contended that recent school boards were committed to pursuing racial balance in student and faculty/staff assignment, and equal educational opportunity. Finally, defendants relied upon the implementation of two desegregation plans after Brown v. Board of Educ. (Brown I), 347 U.S. 483, 74 S.Ct. 686, 98 L.Ed. 873 (1954), and explained other steps taken to achieve and maintain a unitary system. The district court found in favor of the defendants on all issues and declared the system unitary. Brown v. Board of Educ., 671 F. Supp. 1290 (D.Kan. 1987). On appeal, plaintiffs challenge the district court's judgment concerning student and faculty/staff assignment.

If appellate courts were empowered to make findings of fact, perchance the approach of this court on the issues of student and faculty/staff assignment could be justified. But for reasons which the Supreme Court has made clear, we must accept the factual findings of the district court unless they are clearly erroneous. Fed.R.Civ.P. 52(a); Anderson v. City of Bessemer City, 470 U.S. 564, 573-76, 105 S.Ct. 1504, 1511-13, 84 L.Ed.2d 518 (1985); see also Alexander v. Youngstown Bd. of Educ., 675 F.2d 787, 792-93 (6th Cir. 1982). Although this court acknowledges that the district court's finding of unitariness must be upheld unless clearly erroneous, Court's Opinion at 859, the court has failed to give effect to this principle. Instead, this court has duplicated "the role of the district court" and has impermissibly engaged in "an exercise in appellate factfinding." See Jenkins v. Missouri, 807 F.2d 657, 667 (8th Cir. 1986) (en banc). This is apparent throughout the court's opinion, but is especially salient in the court's characterization of what really occurred in Topeka during each decade subsequent to Brown I, a characterization which totally ignores the district court's findings and forgets that we do not try cases at the court of appeals. See Court's Opinion at 874-77. The same evidence that the district court found wanting this court accepts. This court is "engaging in an original evaluation of the voluminous and ofttimes contradictory record seeking an ultimate conclusion more satisfactory than that reached by the district court." Jenkins, 807 F.2d at 668.

The court claims that the district court erred because it misunderstood the burden of proof in this desegregation case. Court's Opinion at 854, 866-67. If that is true, why does the court not remand the case to the district court for appropriate findings in accordance with the correct presumption and burden-shifting principles, given that this is very much a record case? See infra note 6; Vaughns v. Board of Educ., 758 F.2d 983, 992 (4th Cir. 1985) (court remanded for new findings in light of correct legal principles). Instead, the court makes its own factual findings and remands the case for an appropriate remedy.

Moreover, I do not agree with this court that the district court "failed `to apply the appropriate presumption and burden-shifting principles of law.'" Court's Opinion at 867 (quoting Brinkman v. Gilligan, 583 F.2d 243, 251 (6th Cir. 1978), aff'd, Dayton Board of Educ. v. Brinkman (Dayton II), 443 U.S. 526, 99 S.Ct. 2971, 61 L.Ed.2d 720 (1979)). This court compares the district court's action with that of the errant district court in Dayton II. The comparison misses the mark primarily because, as this court points out, Dayton II, like most of "[t]he Supreme Court desegregation cases[,] involved school systems in which the degree of segregation was sufficiently great that the parties did not seriously dispute on appeal that the plaintiffs had satisfied their burden on" ["the existence of a current condition of segregation."]. Court's Opinion at 859-60 n. 17, 859-60. While the "Dayton public schools [were] `highly segregated by race'[.]" Court's Opinion at 859 n. 17 (quoting Dayton II, 443 U.S. at 529, 99 S.Ct. at 2974), this is not so in Topeka. Here, there exists a genuine issue of fact as to whether there is even a current condition of segregation, particularly in the absence of any one-race minority or virtually one-race minority schools in this district. To be sure, there are eight 90% + white schools (ranging from 6.23% to 9.43% minority); however, those schools have steadily increased in minority enrollment over the years, and this is a system in which 31 out of 34 schools are majority white in student assignment. See infra note 18. Of the three schools that are majority-minority in student assignment, the greatest concentration is 61.86% minority. Id. This court's overinclusive approach to racial identifiability leads it to an incorrect finding that there is a current condition of segregation in Topeka.

The court begins by saying that, in a former de jure system, a school is racially identifiable and part of a current condition of segregation if it deviates, plus or minus (±) 15% or 20% or some other arbitrary amount, from a system-wide average of minority composition. See Court's Opinion at 869-70, 872-73 n. 59, 873; but see Price v. Denison Indep. School Dist., 694 F.2d 334, 360-64 (5th Cir. 1982) (statistical racial identifiability not necessarily indicative of liability). Numbers reflecting the racial composition of student and faculty/staff assignment within a school system are absolutely critical in both the liability and the remedial phase of a school desegregation case. However, statistical measures of racial identifiability are measures of racial balance, a valid sociological objective, but not a constitutional command with respect to student assignment. Nevertheless, this court determines that once a plaintiff has shown statistical racial identifiability in a former de jure system, the plaintiff has established a current condition of segregation. Court's Opinion at 859-60, 869-70.

If a school is not racially identifiable as to student assignment by statistical measures, that certainly does not suggest an absence of segregation to this court. Rather one must look to a "broadly defined," Court's Opinion at 859-60, concept of racial identifiability which considers "demography, geography and the individual history of particular schools and areas of the city," in an effort to find liability. Id. at 861. Although the court acknowledges that one-race schools may evolve from "modern urban demography and geography," id. at 860 n. 18, and that "the mere existence of racially identifiable schools does not violate the Constitution," id. at 886, it completely fails to incorporate these observations in its analysis of this system. Nor does it consider other statistical measures relied upon by the district court which support a finding of no current condition of segregation and compliance with the affirmative duty to dismantle the dual system. See id. at 869-72, 876 n. 71. What the court has done is to apply an absolute presumption against the Topeka schools from the outset because of de jure status in 1954. Even if the presumption applies, the court has given no weight to the substantial evidence concerning de facto influences on racial composition in those schools. Although the court ratifies its conclusion concerning a current condition of segregation with descriptive references to geography, residential population and history, there is no attempt to consider causation in light of the substantial demographic testimony offered by defendants. Id. at 873, 876-77.

The district court implicitly recognized, Brown, 671 F. Supp. at 1295, "that once a court has found an unlawful dual school system," School Bd. v. Baliles, 829 F.2d 1308, 1311 (4th Cir. 1987), the plaintiffs are "`entitled to the presumption that current disparities are causally related to prior segregation, and the burden of proving otherwise rests on the defendants.'" See Court's Opinion at 863 (quoting Baliles, 829 F.2d at 1311); see also Dayton II, 443 U.S. at 537, 99 S.Ct. at 2979 (1979); Keyes v. School Dist. No. 1, 413 U.S. 189, 210-11, 93 S.Ct. 2686, 2698-99, 37 L.Ed.2d 548 (1973); Swann v. Charlotte-Mecklenburg Bd. of Educ., 402 U.S. 1, 25-26, 91 S.Ct. 1267, 1280-81, 28 L.Ed.2d 554 (1971) (discussing remedy once a violation is established). If the system has achieved unitariness, the presumption ends. Baliles, 829 F.2d at 1311; Riddick v. School Bd., 784 F.2d 521, 543 (4th Cir. 1986).

In a former de jure system, plaintiffs may establish the prima facie case by proving that there is a current condition of intentional segregation, that the de jure system or its vestiges remain or were reestablished in part of the school system. Columbus Bd. of Educ. v. Penick, 443 U.S. 449, 464, 467, 99 S.Ct. 2941, 2950, 2951, 61 L.Ed.2d 666 (1979); Keyes, 413 U.S. at 200, 210, 93 S.Ct. 2693, 2698. Even an isolated condition of current segregation, resulting from past or present segregative acts, would violate the affirmative duty to eliminate the dual system and its vestiges and to convert to a unitary system in which racial discrimination is eradicated. Dayton II, 443 U.S. at 537, 99 S.Ct. at 2979; Columbus, 443 U.S. at 459, 99 S.Ct. at 2948; Swann, 402 U.S. at 15, 91 S.Ct. at 1275; see also L. Tribe, American Constitutional Law 1498-1500 (2d ed. 1988) (discussing Supreme Court school desegregation cases).

That critical finding shifts the burden of proof to the defendants because it provides "a sufficient basis for an inferential finding of system-wide discriminatory intent" which in turn will be applied to explain "racial separation in other parts of the school system." Columbus, 443 U.S. at 467-68, 99 S.Ct. at 2951-52. At that point, the defendants must come forward with "proof sufficient to support a finding that segregative intent was not among the factors that motivated their actions." Keyes, 413 U.S. at 210, 93 S.Ct. at 2698. If intent cannot be disproved, the defendants must show that their "past segregative acts did not create or contribute to the current segregated condition," in the other parts of the system. Id. at 211, 93 S.Ct. at 2699; see Dayton II, 443 U.S. at 538, 99 S.Ct. at 2979 ("Board has a `heavy burden' of showing that actions that increased or continued the effects of the dual system serve important and legitimate ends").

The district court correctly cited Keyes and Swann in its discussion concerning the burden of proof and noted that the factual justification in this case for the legal presumption was doubtful. Brown, 671 F. Supp. at 1295, 1295 n. 4 5. I take this to mean that the plaintiffs simply had not proven a current condition of segregation (a constitutional violation) in the Topeka schools which would warrant sustaining the presumption. But the district court went further when it focused on the defendants and held that, in the alternative and regardless of any presumptions in favor of the plaintiffs, the defendants had proven the existence of a unitary system. Id. Merely because the district court properly discussed an absence of discriminatory intent does not negate the fact that in this record there is evidence tending to show that any past segregative acts of this school district are not responsible for schools currently which do not approach the system-wide minority average. See infra pt. VII, p. 935 (analyzing schools which court claims are racially identifiable and segregated). Indeed, this court at various points acknowledges an absence of evidence of segregative intent by current school officials. Court's Opinion at 868 ("we agree with the district court's findings that the current school administration is not presently acting with discriminatory intent"), see also id. at 880, 886; but see id. at 872 (concerning faculty/staff assignment in 1985-86, "it is difficult to posit a [racially] neutral explanation"), 879 id. at 883 ("It is thus clear that Quinton Heights' current boundaries are the result of continued manipulation by the school district."). The district court also considered circumstantial evidence of past segregative intent manifesting itself in the current system, concerning the prima facie case and beyond. Of course, the district court had much quantitative and qualitative evidence to evaluate. Having performed its required task, it found that "the district's student and staff assignment criteria do not have the effect of discriminating against students because of their race." Brown, 671 F. Supp. at 1310 (denying relief based on Title VI regulation) (emphasis added). Because the record supports the district court's determination that there is not unlawful segregation in Topeka and that the system is unitary, albeit on conflicting evidence, I respectfully dissent.

In its opinion, the court dismisses my dissent as a product of misconstruing the principles of school desegregation law. Court's Opinion at 862-63, 867 n. 47, 868 n. 49, 869 n. 51. Apparently, the court views the affirmative duty to eliminate segregation and its vestiges as completely independent from, if not inconsistent with, the intent inquiry. To be sure, those statutory dual systems operating at the time of Brown I automatically assumed an affirmative duty to eliminate not only the dual system, but also the effects of that dual system on present school operations. Keyes, 413 U.S. at 200, 93 S.Ct. at 2693. However, the initial affirmative duty arises from the presence of intentional segregation whether once commanded by statute or practice.

Moreover, the intent inquiry is inclusive of whether a school board has acted affirmatively, not merely neutrally, to eliminate a dual system existing in 1954. A school board's proven failure to heed its affirmative duty and eliminate a dual system and its vestiges, whether that failure be by inaction or by conscious decision, is indicative of segregative intent and compounds the harm of an unconstitutional dual system. See Columbus, 443 U.S. at 461, 99 S.Ct. at 2948 ("Whatever the Board's current purpose with respect to racially separate education might be, it knowingly continued its failure to eliminate the consequences of its past intentionally segregative policies."); see also Green v. County School Bd., 391 U.S. 430, 437-38, 88 S.Ct. 1689, 1693-94, 20 L.Ed.2d 716 (1968). In deciding whether a school system is unitary, the ultimate inquiry must be whether past or present segregative intent has been eliminated from the present school operations.

Thus, a current condition of purposeful separation may arise either from a failure to eliminate 1954 de jure segregation and its effects or by subsequent actions which are segregative in their own right. Stated another way, plaintiffs are entitled to a system-wide desegregation order upon a finding that there is current racial separation which is the result of segregative intent, past or current. Plaintiffs are not entitled to a system-wide desegregation order based upon proven de facto racial separation not attributable to the actions of the school board. Keyes, 413 U.S. at 208, 93 S.Ct. at 2697 ("We emphasize that the differentiating factor between de jure segregation and so-called de facto segregation to which we referred in Swann is purpose or intent to segregate.") (emphasis in original). Thus, contrary to this court's reading of my dissent, I agree that school board action which has the effect of failing to disestablish the dual system violates the school board's affirmative duty. See Court's Opinion at 862 (quoting Pitts v. Freeman, 755 F.2d 1423, 1427 (11th Cir. 1985)); Columbus, 443 U.S. at 449, 99 S.Ct. at 2942.

I must reject however, this court's contention that "[t]his is not a liability case in which plaintiffs must prove that the school board has committed unconstitutional acts; that was established in 1954." Court's Opinion at 862 (emphasis omitted). While this is not a case of initial liability, id. at 868, it is not a case like Swann, see Court's Opinion at 862, in which all parties agreed that the 1969 Charlotte-Mecklenburg school system was not unitary given the Supreme Court's 1968 Green decision, announced three years after the first desegregation plan implemented by that district in 1965. Swann, 402 U.S. at 7, 91 S.Ct. at 1271. In Swann, the issue was whether the district court was correct in selecting among alternative desegregation plans put forth by the parties. Id. at 11, 91 S.Ct. at 1273. Contrary to this court's assertion, Court's Opinion at 862, 868 n. 49, ours is not now a remedy case. Upon the reopening of this case in 1979, some 24 years after the first court-approved desegregation plain, the district court's task was to determine whether liability for the prior de jure system remained. Unlike Swann, the parties in our case hotly contested this issue.

If the liability inquiry in this case was frozen as of 1954 as suggested by the court, id. at 862, we merely would decree a remedy based upon the failure of the school board to have absolute racial balance at every school and much of this court's opinion arguing that (1) there is a current condition of segregation, (2) the school board has failed to fulfill its affirmative duty, and (3) there is a link between present school operations and the prior de jure system, would be superfluous. The inquiry in this case is not so simple because two complex factual issues are involved. The first is whether current disparities in this case comprise a current condition of segregation given that there is not a single one-race minority or virtually one-race minority elementary, middle or high school, in this system. But if there is a current condition of segregation, the burden of proof would shift to the school board as discussed above. The next complex factual issue in the case then becomes whether the effects of any past intentional segregation have been eliminated. An "essential predicate" for a system-wide remedy in this case is that the "Board's purposefully discriminatory conduct and policies, had current, system-wide impact." Columbus, 443 U.S. at 466 n. 15, 99 S.Ct. at 2951 n. 15.

The court also claims that the defendants' experts, the district court and I have erred because we have considered the school board's proof that, having dismantled a dual school system, it is not liable for schools which do not approach the system-wide minority average in student assignment. See Court's Opinion at 876-77. Who could argue with the court's statement that demographic change does not excuse a school board's failure to comply with its affirmative duty to eliminate the effects of a dual system? Id. (relying on Vaughns, 758 F.2d at 988, and Lee v. Macon County Bd. of Educ., 616 F.2d 805, 809-10 (5th Cir. 1980)). In both Vaughns and Lee, the appellate court was concerned with virtually one-race minority schools which had not lost their racial identities. Vaughns, 758 F.2d at 991; Lee, 616 F.2d at 808, 811 (5 one-race minority schools). The disagreement between the court's opinion and my dissent on this point is that I view the school board as having accomplished its affirmative duty and having eliminated the effects that may be charged to a once-dual system. All virtually one-race minority schools have been eliminated, and there are no schools on the threshold of virtually one-race minority status. In any event, the school board is allowed to prove that schools which do not approach the system-wide minority average are not the product of past or present segregative intent or, stated another way, the school board may prove that its recent and remote actions are not the cause of present racial imbalance. In Keyes, the Court recognized that school officials may defend by showing that current racial separation is not a product of past segregative actions. Keyes, 413 U.S. at 211, 93 S.Ct. at 2698. Although the passage of time does not make intentional segregative acts which occurred decades ago any less intentional, id. at 210-11, 93 S.Ct. at 2698-99, the effects of such action may well be attenuated, given the dynamic environment of a school system. The court's absolute refusal to consider this evidence is tantamount to imposing strict liability on the school board in the absence of perfect racial balance.

I.

This action began when the original plaintiffs challenged Topeka's maintenance of a segregated system of schools for grades kindergarten through six (K-6). Brown v. Board of Educ., 98 F. Supp. 797 (D.Kan. 1951), rev'd, 349 U.S. 294, 75 S.Ct. 753, 99 L.Ed. 1083 (1955). At the time, Topeka operated 22 elementary schools, 18 for white children and 4 for black children. Id. Many black children were bused a considerable distance to the 4 black schools. Id. at 798. The challenge involved both the contention that separate schools were inherently unequal and the contention that the physical facilities, curricula, faculty and student services of the black schools were inferior. Id. at 797-98. Circuit Judge Huxman, writing for a three-judge district court, adhered to precedent, Plessy v. Ferguson, 163 U.S. 537, 16 S.Ct. 1138, 41 L.Ed. 256 (1896), and held that separate but equal facilities were constitutional. Brown, 98 F. Supp. at 798-99; see generally In Memoriam, 474 F.2d preface at 13-14 (1973) (discussing judge's difficult decision). The district court entered as a finding of fact that the "physical facilities, the curricula, the courses of study, qualification of and quality of teachers, as well as other educational facilities in the two sets of schools are comparable." Brown, 98 F. Supp. at 798. The judgment was appealed to the Supreme Court, with initial arguments during the 1952 Term.

During September 1953, and prior to the Supreme Court's opinion, the Topeka school board adopted a policy to terminate the maintenance of segregation in the elementary grades. Rec. ex. vol. II at 60. The board adopted the first step of a desegregation plan which terminated the white-only segregation at two elementary schools (Southwest Randolph) and discontinued the board's supplied transportation to 15 black children in those districts. In January 1954, the board adopted the second step of the desegregation plan to become effective in September 1954. Rec. ex. vol. II at 61. The second step terminated the white-only segregation in 12 more elementary schools, eliminated transportation for black students in these 12 districts, and permitted black students to continue at their former schools, if desired. Id.

In May 1954, the historic decision of the Supreme Court reversed the three-judge district court. Brown I, 347 U.S. 483, 74 S.Ct. 686. The Court held that "the segregation of children in public schools solely on the basis of race" denied minority group children equal educational opportunities contrary to the fourteenth amendment. Id. at 493, 74 S.Ct. at 691. With respect to the Kansas case, the Court expressly found that the non-elementary grade levels of the public schools were "operated on a non-segregated basis." Id. at 486 n. 1, 74 S.Ct. at 687 n. 1. The Court did not alter the finding of the district court of substantial equality "with respect to buildings, transportation, curricula, and educational qualifications of teachers." Id. at 486 n. 1, 74 S.Ct. at 687 n. 1.

In February 1955, the Topeka board adopted the third step of the desegregation plan to become effective in September 1955. Rec. ex. vol. II at 63. The plan addressed the four black elementary schools (McKinley, Buchanan, Monroe and Washington). Under the third step, McKinley was closed, and Buchanan, Monroe and Washington were placed within the general framework of the elementary school attendance districts. Id. All elementary students affected were given the option of finishing elementary school in their present schools, with the exception of McKinley students. Children entering kindergarten in school year 1955-56 were given the option of attending the school they would have attended in 1954-55, had they been old enough.

Having decided the question of constitutional liability, the Court ordered further argument on a remedy and issued its next opinion concerning appropriate relief in May 1955. Brown v. Board of Educ. (Brown II), 349 U.S. 294, 75 S.Ct. 753, 99 L.Ed. 1083 (1955). The Court noted that "substantial progress" had been made in Kansas "from the transition to a system of public education freed of racial discrimination." Id. at 299, 75 S.Ct. at 755. Recognizing that full implementation of Brown I might "require solution of varied local problems" by school authorities, the Court remanded this case to the three-judge district court for consideration of whether the actions of school authorities constituted a good faith implementation of Brown I. Id. School authorities were to make a "prompt and reasonable start toward full compliance" with Brown I. Id. at 300, 75 S.Ct. at 756.

Once such a start has been made, the courts may find that additional time is necessary to carry out the ruling in an effective manner. The burden rests upon the defendants to establish that such time is necessary in the public interest and is consistent with good faith compliance at the earliest practicable date. To that end, the courts may consider problems related to administration, arising from the physical condition of the school plant, the school transportation system, personnel, revision of school districts and attendance areas into compact units to achieve a system of determining admission to the public schools on a non-racial basis. . . . They will also consider the adequacy of any plans the defendants may propose to meet these problems and to effectuate a transition to a racially nondiscriminatory school system.

Id. at 300-01, 75 S.Ct. at 756-57. The district courts were to be guided by principles of equity and to retain jurisdiction during the transition period. Id. In June 1955, the Supreme Court mandate ordered the district court to take such actions "as are necessary and proper to admit to public schools on a racially nondiscriminatory basis with all deliberate speed the parties to these cases." Id. at 301, 75 S.Ct. at 757.

On remand, the district court held a hearing on the formulation of a decree and judgment. In October 1955, the panel approved the plan already adopted and partially implemented by the Topeka board "as a good faith beginning to bring about complete desegregation." Brown v. Board of Educ., 139 F. Supp. 468, 470 (D.Kan. 1955) (per curiam). The court characterized the plan as follows:

The central principle of the plan is that hereafter, except in exceptional circumstances, school children irrespective of race or color shall be required to attend the school in the district in which they reside and that color or race is no element of exceptional circumstances warranting a deviation from this basic principle.

Id. at 469. The court noted that in several respects, which were minor and not discussed, the plan did not comply fully with the Supreme Court mandate. Id. One serious objection to the plan was that part which permitted kindergarten children to attend the school which they would have attended had they started school one year previously. But because that part was temporary and not part of the permanent plan, it did not defeat a finding of good faith.

At the hearing, the plaintiffs objected to certain schools, such as Buchanan, remaining all black. The district court rejected this as a defect in the plan, stating:

Desegregation does not mean that there must be intermingling of the races in all school districts. It means only that they may not be prevented from intermingling or going to school because of race or color.

If it is a fact, as we understand it is, with respect to Buchanan School that the district is inhabited entirely by colored students, no violation of any constitutional right results because they are compelled to attend the school in the district in which they live.

Id. at 470. In 1973, the Supreme Court indicated that this limited interpretation of Brown was rejected in 1968 by its decision in Green, 391 U.S. at 437-38, 88 S.Ct. at 1693-94. Keyes, 413 U.S. at 200-01 n. 11, 93 S.Ct. at 2693-94 n. 11. Meanwhile, the district court retained jurisdiction of this case for the purpose of entering a final decree pending full compliance with the Supreme Court's mandate to desegregate. The court's order approving the plan was not appealed.

In December 1956, the Topeka board considered the fourth step of its plan to comply with the Supreme Court's mandate in Brown II as ordered by the district court. Rec. ex. vol. IV at 120. This final step of the plan was to become effective in September 1956, and eliminated the kindergarten option provision criticized by the district court. Rec. ex. vol. II at 66. The plan also required all elementary students moving into districts, not designated as optional districts between two or more elementary schools, to attend the elementary school in the requisite district, subject only to traditional exceptions. Id. at 66. Two board members opposed the plan's optional features as prolonging the time before complete integration. Id. at 67. Concern was expressed about the future of the teachers at the de jure black elementary schools. Id. Notwithstanding, the superintendent, under questioning from the board, indicated that no additional steps were contemplated and that the adoption of the plan brought Topeka "into full compliance with the law." Id..

The fourth step of the plan was opposed by the president of the NAACP, Mr. Burnette. Mr. Burnette told the board that the NAACP only took legal action in this case as a "last resort." Rec. ex. vol. IV at 120. He opposed the plan because "it would take seven long years to terminate racial segregation" and no steps had been taken to integrate black teachers. Id. at 120-21. The fourth step of the plan was approved in January 1956. Id. at 121. The plan had been implemented fully as of the 1960-61 school year, and by 1961-62, all elementary students, who had been permitted to continue at the school which they attended prior to implementation of step four of the plan, had completed their elementary education.

Two observations should be made at this point. First, the school board hardly can be faulted for implementing its plan to desegregate the elementary schools. In its revised opinion, the court has downgraded the school board's initial desegregation efforts from "a remarkably enlightened beginning", in the mid-1950's," see Brown v. Board of Educ., No. 87-1668, slip op. at 42 (filed Jul. 2, 1989 withdrawn Jul. 19, 1989), to "a `good faith' beginning", Court's Opinion at 874 (quoting Brown, 139 F. Supp. at 470). This court then characterizes "[t]he mid-50s to mid-60s," as "no doubt" segregative under modern authority. Court's Opinion at 874-75. The board had the approval of the district court and federal authority well into the 1970's, including that of this circuit, which echoed the district court's conclusion that Brown I and Brown II, while prohibiting segregation, did not require concern with integrative racial balancing in neighborhood schools. See Keyes v. School Dist. No. 1, 445 F.2d 990, 1005 (10th Cir. 1971) ("We never construed Brown to prohibit racially imbalanced schools provided they are established and maintained on racially neutral criteria, and neither have other circuits considering the issue."), modified, 413 U.S. 189, 93 S.Ct. 2686, 37 L.Ed.2d 548 (1973); Board of Educ. v. Dowell, 375 F.2d 158, 166 (10th Cir.) ("It is still the rule in this Circuit and elsewhere that neighborhood schools attendance policies, when impartially maintained and administered, do not violate any fundamental Constitutional principles or deprive certain classes of individuals of their Constitutional rights."), cert. denied, 387 U.S. 931, 87 S.Ct. 2054, 18 L.Ed.2d 993 (1967); Downs v. Board of Educ., 336 F.2d 988, 995-96, 998 (10th Cir. 1964) (neighborhood schools which may result in racial imbalance not actionable; no duty to integrate), cert. denied, 380 U.S. 914, 85 S.Ct. 898, 13 L.Ed.2d 800 (1965); Bell v. School City of Gary, 324 F.2d 209, 213 (7th Cir. 1963) (neighborhood school plan drawn with no intent or purpose to segregate is constitutional even if effect is that some schools will not be racially balanced), cert. denied, 377 U.S. 924, 84 S.Ct. 1223, 12 L.Ed.2d 216 (1964). Second, courts do not speak unless spoken to and cannot decide what is not before them. Here, the school board made hundreds of decisions since 1952 that the plaintiffs contend reflect segregative intent and effect, yet not until 1979 was the court asked to intervene. While this in no way would excuse noncompliance with the constitutional command to desegregate the schools, it does furnish a backdrop against which to understand the liability focus of this case.

II.

The amplification of Brown I and Brown II began with the Court's decision in Green, 391 U.S. 430, 88 S.Ct. 1689, 20 L.Ed.2d 716 (1968). The New Kent County, Virginia system was considerably smaller than the Topeka system, with an enrollment of 1300 students in two schools, approximately 57% black and 43% white. Id. at 432, 88 S.Ct. at 1691. The court considered a plan which allowed a student to choose between schools which were essentially one-race schools with respect to "faculty, staff, transportation, extracurricular activities and facilities." Id. at 435, 88 S.Ct. at 1693. In every way, the schools were a product of de jure segregation. Although students had a choice between schools, attendance patterns largely reflected the de jure system. Id. at 441, 88 S.Ct. at 1696. Eighty-five percent of the black children attended an all-black school. Id. at 441, 88 S.Ct. at 1696. The court made it clear that school boards operating state-compelled dual school systems at the time of Brown II "were . . . clearly charged with the affirmative duty to take whatever steps might be necessary to convert to a unitary system in which racial discrimination would be eliminated root and branch." Id. at 437-38, 88 S.Ct. at 1693-94.

From 1955-1969, the Topeka school district underwent substantial transition. Due to annexation and unification between 1957 and 1964, the size of the district more than doubled, increasing from 16.60 to 37.45 square miles, which "fundamentally changed the system." See rec. vol. XI at 2399 (commenting upon changes from 1955-60). The population in Topeka grew until 1970, and thereafter declined. Rec. vol. XI at 2292-93; ex. vol. I at 12. This trend was reflected in school enrollment which has declined substantially in recent years. The black and hispanic population increased significantly. Rec. ex. vol. I at 12. And the Topeka school board continued with its responsibility to eliminate de jure segregation and its vestiges.

In the 1955-56 school year, the board operated 23 elementary schools, 7 junior high schools, and 1 high school. Brown, 671 F. Supp. at 1293; rec. ex. vol. IV at 38; supp. vol. XI at 186, 225. At the time of Brown I, the Court recognized that Topeka did not operate de jure secondary schools. Brown I, 347 U.S. at 486 n. 1, 74 S.Ct. at 687 n. 1. In the 1955-56 school year, black elementary students attended 18 out of 23 elementary schools. Brown, 671 F. Supp. at 1293. At the elementary school level, the total black student population was 10.4% and three schools were 90 + % black (Buchanan, Monroe and Washington). See supra note 9. 45.6% of the black elementary students attended these three schools. Id. Thirteen elementary schools were 90 + % white. Id. At the junior high school level, the total black student population was estimated to be 11.2%. See supra note 10. There were no 90 + % black junior high schools, but 3 schools were 90 + % white (Holliday, Roosevelt Capper).

In the 1968-69 school year, the school board operated 34 elementary schools, 11 junior high schools and 3 high schools, having added Highland Park High School in 1959 through annexation and Topeka West High School in 1961 through construction. Rec. ex. vol. IV at 64; supp. vol. XI at 225. The system had 25, 737 students. Rec. ex. vol. IV at 65; see supra note 8. Three of the four initially de jure black elementary schools had been closed by the 1968-69 school year. McKinley was closed in 1955, Buchanan in 1959, and Washington in 1962. Brown, 671 F. Supp. at 1293. Monroe, the only remaining de jure black elementary school (Monroe) had an enrollment which was 25% white. Id. At the elementary school level, the total minority student population was 17.39% minority (12.40% black), and one school (Parkdale) was 90+% minority and three others were 50+% minority (Belvoir, Lafayette, Monroe). See supra note 11. Eighteen elementary schools were 90+% white. Id. At the junior high school level, the total minority student population was 16.50% (11.81% black). See supra note 12. There were no 90+% black junior high schools, but five junior high schools were 90+% white (Capper, Eisenhower, Jardine, Landon Roosevelt). Id. There were no 90+% black high schools, but one high school was 90+% white (Topeka West). See supra note 13. Minority students attended every school in the district, but in widely varying numbers. See supra notes 11-13.

In 1971, the Supreme Court provided greater guidance for the remedial phase of desegregation. Swann, 402 U.S. at 18, 91 S.Ct. at 1277. The Charlotte-Mecklenburg school system was considerably larger than the Topeka system, with an enrollment of more than 84,000 and 107 schools. Id. at 6, 91 S.Ct. at 1271. The racial composition was 71% white and 29% black. To support liability, the Supreme Court mentioned that two-thirds of 21,000 black students in Charlotte city schools attended 21 schools which were 99+% black. Id. at 7, 91 S.Ct. at 1271. The district court's objective was to have the schools approximate the racial composition of the system, 71% white and 29% black. Id. at 9-10 n. 4, 91 S.Ct. at 1272-73 n. 4. To that end, the district court endorsed a plan that called for busing to achieve schools 9% to 38% black. Id. at 9, 91 S.Ct. at 1273.

As a remedial phase case, Swann is important because it speaks "to what extent racial balance or racial quotas may be used as an implement in a remedial order to correct a previously segregated system." Id. at 22, 91 S.Ct. at 1279. The school board argued that the district court erred by seeking racial balance in the 71%-29% range for each school. Id. at 23, 91 S.Ct. at 1279. The Court recognized that the district court's "use . . . of mathematical ratios was no more than a starting point in the process of shaping a remedy, rather than an inflexible requirement," and was within the district court's equitable discretion. Id. at 25, 91 S.Ct. at 1280.

The Court spoke to whether there is any constitutional requirement of racial balance.

If we were to read the holding of the District Court to require, as a matter of substantive constitutional right, any particular degree of racial balance or mixing, that approach would be disapproved and we would be obliged to reverse. The constitutional command to desegregate the schools does not mean that every school in the community must always reflect the racial composition of the school system as a whole.

Id. at 24, 91 S.Ct. at 1280. This is of paramount importance for this case because, despite the court's citation of this principle, see Court's Opinion at 861, 886, the court has not allowed it to operate. Rather, even after the Topeka board has implemented fully two desegregation plans and the system "improved dramatically in the last ten years as far as desegregation is concerned," Court's Opinion at 877, the liability focus of the plaintiffs and the court is unmistakably on the principle that more could have been done to achieve greater racial balance, that is more schools could approach the system-wide minority average in student assignment. See Court's Opinion at 876, 877-78, 882-84 (analyzing schools which plaintiffs suggest have "clear racial identities," id. at 882, and essentially criticizing school board decisions which resulted in these schools not approximating the system-wide average of minority students). However, "the fact that a school board's desegregation plan leaves some disparity in racial balance among various schools in the system does not alone make that plan unacceptable." Wright v. Council of the City of Emporia, 407 U.S. 451, 464, 92 S.Ct. 2196, 2204, 33 L.Ed.2d 51 (1972) (footnote omitted).

Equally important for this case is the Supreme Court's discussion in Swann of "whether every all-Negro and all-white school must be eliminated as an indispensable part of a remedial process of desegregation." Swann, 402 U.S. at 22, 91 S.Ct. at 1279. In a section entitled "One-Race Schools," the Court applied a common-sense approach to the problem of racial concentration in various parts of a metropolitan area.

The record in this case reveals the familiar phenomenon that in metropolitan areas minority groups are often found concentrated in one part of the city. In some circumstances certain schools may remain all or largely of one race until new schools can be provided or neighborhood patterns change. Schools all or predominantly of one race in a district of mixed population will require close scrutiny to determine that school assignments are not part of state-enforced segregation.

In light of the above, it should be clear that the existence of some small number of one-race, or virtually one-race, schools within a district is not in and of itself the mark of a system that still practices segregation by law. 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.

Id. at 25-26, 91 S.Ct. at 1280-81. When there has been intentional segregation in the past, there is a presumption against schools that are one-race minority or virtually one-race minority, and the school board has the burden of showing assignment is genuinely nondiscriminatory. Id. at 26, 91 S.Ct. at 1281; Price, 694 F.2d at 353-56. The district court recognized this presumption, although it noted that this case does not "fit the Swann presumption easily." Brown, 671 F. Supp. at 1295. Even though the proportion of minority students in this system is smaller than in many cases and there are no one-race minority or virtually one-race minority schools, the literal Swann presumption could be applied to the 90+% white schools. See Price 694 F.2d at 364-65.

This court acknowledges the critical importance of a finding of "one-race" schools to support a current condition of segregation and liability overall. Court's Opinion at 860-61. The court correctly notes that the significance of "one-race" white schools in this case is limited because of Topeka's large majority white student population (74.05% in 1985). Id. at 869; see also Price, 694 F.2d at 364-65 (recognizing major purpose behind school desegregation cases is to secure rights of minority groups; literal compliance with Swann may not be required when system is heavily white and all schools are majority white). But the court departs from Swann when it imposes liability on the basis of student assignment in the absence of "one-race" minority schools or a district court finding of current intentional segregation. This court imposes liability for a lack of racial balance, while virtually ignoring the demographic testimony of defendants' expert (and other evidence adduced by the defendants), which was credited by the district court. See Court's Opinion at 869 n. 52 ("In a school system with a 26% minority student population, we think the number of schools in Topeka that approach or are over the 50% minority mark constitute persuasive evidence that the school system has not met its duty to desegregate.") (emphasis added).

In 1973, the Supreme Court added an important presumption which expanded the consequences of a current condition of intentional segregation in any part of a school system. Keyes, 413 U.S. at 208, 93 S.Ct. at 2697. The district court in that case had found contemporary and intentional segregation, including student and teacher assignment, with respect to schools in the Park Hill area in Denver, which were attended by almost 38% of the black student population. Id. at 199, 93 S.Ct. at 2692. Blacks comprised 14% of the students in the system, hispanics comprised 20%; for a system minority population of 34%. Id. at 195, 93 S.Ct. at 2690. Three of five elementary schools in the Park Hill area were 89+% minority; one junior high was 76+% minority and the other was 96+% minority. Id. at 199 n. 10, 93 S.Ct. at 2692 n. 10. Near Park Hill were 22 core city schools, all 22 were 70+% minority and 11 were 90+% black. The district court and the court of appeals essentially required independent proof that the core city schools were a product of official segregation. Id. at 205, 93 S.Ct. at 2695. The Court stated:

[W]e hold that a finding of intentionally segregative school board actions in a meaningful portion of a school system, as in this case, creates a presumption that other segregated schooling within the system is not adventitious. It establishes, in other words, a prima facie case of unlawful segregative design on the part of school authorities, and shifts to those authorities the burden of proving that other segregated schools within the system are not also the result of intentionally segregative actions.

Id. at 208, 93 S.Ct. at 2697. The Court emphasized that purpose or intent to segregate is what differentiates actionable de jure segregation from de facto segregation. Id. Thus, plaintiffs would be entitled to the Keyes presumption upon a showing of purposeful segregation in a meaningful part of the school system. Id. at 208-09, 93 S.Ct. at 2697-98. As noted, once the presumption is invoked, the defendant school board must come forward with evidence showing that "the existence of subsequent or other segregated schooling within the same system . . . is not also the result of intentionally segregative acts." Keyes, 413 U.S. at 210, 93 S.Ct. at 2698. The school board must either disprove segregative intent or show "that its past segregative acts did not create or contribute to the current segregated condition." Id. at 211, 93 S.Ct. at 2699.

The framework in Keyes formed the basis for the Court's decisions in Columbus and Dayton II. These cases held that if a school board was operating a dual school system in 1954, even absent a state provision so allowing, a plaintiff may satisfy his initial burden of proof by relying upon evidence of " recent and remote intentionally segregative actions" of the school board. Columbus, 443 U.S. at 463-65, 99 S.Ct. at 2949-51 (emphasis added); see Dayton II, 443 U.S. at 537, 99 S.Ct. at 2979. In both cases, the affirmative duty to eradicate the effects of past segregative conduct not only had been unfulfilled, but also segregative practices persisted. Columbus, 443 U.S. at 460-63, 99 S.Ct. at 2948-49; Dayton II, 443 U.S. at 537, 99 S.Ct. at 2979. Either type of violation would satisfy the intent requirement under the fourteenth amendment; the basis for a system-wide remedy was "purposefully segregative practices with current segregative impact." Columbus, 443 U.S. at 466, 99 S.Ct. at 2951; Dayton II, 443 U.S. at 537-38, 99 S.Ct. at 2978-79.

This court questions the applicability of Keyes and Columbus because those cases involved initial determinations of liability concerning policy-based dual, rather than statutorily dual, school systems. See Court's Opinion at 862. The court explains that the presumption developed in Keyes is limited only "to establish[ing] initial de jure segregation in a case where no statutorily mandated segregation exists." Court's Opinion at 863 n. 27; see also Columbus, 443 U.S. at 458, 99 S.Ct. at 2946 (using Keyes presumption to establish system-wide segregation in 1954); Dayton II, 443 U.S. at 535, 99 S.Ct. at 2978 (same). According to the court, a different presumption "applies . . . after either statutorily-mandated or policy-based de jure segregation has been established." Court's Opinion at 863 n. 27. The court then cites Dayton II in a later discussion for support. Court's Opinion at 863. However, like Columbus, Dayton II involved an initial determination of system-wide liability in Ohio where statutorily mandated segregation ceased after 1888. 443 U.S. at 534-35, 537, 99 S.Ct. at 2977-78, 2979; Columbus, 443 U.S. at 455-56, 455 n. 4, 99 S.Ct. at 2945-46, 2945 n. 4. Segregation by official policy, not by statute, was at issue.

In both Columbus and Dayton II, the Keyes presumption was employed to decree a current system-wide remedy on the basis of recent and remote purposefully segregative action, including the failure to satisfy the affirmative duty to eliminate a dual system. Columbus, 443 U.S. at 467-68, 99 S.Ct. at 2951-52 (discussing use of Keyes); Dayton II, 443 U.S. at 537, 99 S.Ct. at 2979 (citing Keyes), 443 U.S. at 534, 541-42, 99 S.Ct. at 2977, 2981-82 (consequences of 1954 dual system and intentionally segregative effect of actions since 1954 furnished basis of liability). Thus, the Supreme Court not only affirmed lower court findings of officially dual school systems as of 1954, but also approved current system-wide remedies based upon the authority of Keyes. Columbus, 443 U.S. at 467-68, 99 S.Ct. at 2951-52.

III.

In 1973, an action was filed against U.S.D. 501 and various other governmental defendants principally alleging that children in east and north Topeka (with higher concentrations of minority students) received inferior educational opportunities and facilities as compared to children in west and south Topeka. Johnson v. Whittier, No. T-5430 (D.Kan. filed Sept. 10, 1973). The filing of Johnson led to an investigation in November and December of 1973, by the U.S. Department of Health, Education and Welfare (HEW) to determine whether U.S.D. 501 was in compliance with Title VI of the Civil Rights Act of 1964, 42 U.S.C. § 2000d. Rec. ex. vol. II at 30.

In January 1974, HEW released its findings by letter. The findings of HEW are very similar to the plaintiffs' allegations in this case. Specifically, HEW determined four areas of non-compliance. First, HEW claimed that a "substantial number of . . . schools continue to operate with student racial compositions not consonant with a unitary plan of student assignment capable of fully desegregating [the] district." Rec. ex. vol. V at 13. HEW found that five elementary schools "have substantially disproportionate minority student compositions clearly the result of a former dual pattern of operation." Id. The schools and minority percentages were: Monroe (73.98%), Lafayette (61.22%), Lowman Hill (49.08), Parkdale (91.81%) and Belvoir (71.90%). Id.; rec. ex. vol. II at 72. Second, optional attendance zones were criticized. In the "most extreme" (and only) case mentioned "students transferring from areas outside the Lafayette attendance area outside the Lafayette attendance area constitute approximately one-third of the minority enrollment." Rec. ex. vol. V at 13. Thus, HEW was concerned with transfers by minorities to schools with already high minority percentages. Third, HEW found that junior high school facilities in schools with higher minority percentages (Crane, 50.85%; Curtis, 28.90%; East Topeka, 63.32%; and Highland Park, 27.16%) generally were inferior to and older than the four schools with the lowest percentage of minority enrollment (French, 2.97%; Landon, 1.89%, Jardine, 2.13% and Capper, 1.23%). Id. at 13-14; rec. ex. vol. IV at 84 (1973-74 racial inventory). Fourth, HEW found that a larger percentage of minority students, as compared to white students, attended elementary schools with inadequate kindergarten rooms and smaller library media centers. Rec. ex. vol. V at 14; vol. X at 1435.

In February 1974, the board, while not admitting liability, directed its administrative staff to develop a plan for compliance which could include reassigning students, redefining attendance areas, closing schools and constructing new or modifying existing facilities. Rec. ex. vol. II at 72. The plan developed by the staff was presented in April 1974, and called for changing the attendance areas of 22 schools and closing 7 schools. Id. A concerned parents group complained about the lack of parental involvement and sought to be involved in the development of any plan. Id. at 74. Unlike a federal court, a democratically elected school board is charged with listening to parental concerns. The plan prepared by the staff was not adopted. Id. at 80. Instead, the board unanimously rejected the plan as creating inconvenience and hardship while lowering the quality of education. Id. at 80, 82. The board resolved that the "District determine if inequality of educational opportunities exist, and if so, take the remedial steps necessary to rectify the same, retaining the neighborhood school concept wherever possible." Id. at 80.

In June 1974, HEW initiated formal administrative proceedings to terminate federal funds for the district's alleged lack of compliance with Title VI. See In re Topeka Unified School District No. 501, HEW Docket No. S-79. In the Notice of Opportunity for Hearing, HEW's litigating position unfolded. Rec. ex. vol. V at 150-58. HEW alleged that Topeka never fully desegregated its elementary and junior high schools as required by Brown I and that there were a substantial number of schools that were "one-race, virtually one-race, or substantially disproportionate in student racial composition." Id. at 155 (¶¶ 12, 13). In addition to alleging a dual system with respect to student assignment and facilities, HEW claimed that the school district "has continued to assign faculty in a manner which reinforces the racial identity of its substantially disproportionate and one-race schools." Id. at ¶ 14.

The board went to federal court and sought to enjoin HEW from holding the hearing and cutting off federal funds. The board took the position that the entire Topeka system was still under the 1955 order to comply with Brown I. See Brown v. Board of Educ., 84 F.R.D. 383, 390-91 (D.Kan. 1979). The district judge agreed and concluded that the district court was the proper forum for resolving desegregation claims. Id. at 391. The board then developed a short-range facilities plan which resulted in the closing of two elementary schools (Clay, 25.35% minority percentage, and Monroe, 82.73%) and two junior high schools (Crane, 52.85%, and Curtis, 25.64%), all at the end of the 1974-75 school year. Rec. ex. vol. II at 90-92, 96; ex. vol IV. at 87-88; vol. X at 1436. Monroe Elementary was the last formerly de jure black school, and it was closed at the end of the 1974-75 school year. Rec. ex. vol. II at 84, 90. Ironically, the board hired Dr. Gordon Foster, assisted by Mr. Lamson, who presented alternate suggestions. These parties became the plaintiffs' lead expert witnesses in this case.

By May 1975, the board had adopted an affirmative action plan, and approved the creation of a citizens' advisory committee and a specialized learning center to "be operated in a manner intended to encourage integration among minority and majority students."Rec. ex. vol. II at 94; ex. vol. IV at 3-8; vol. X at 1449-52. All of these programs are still in place, including the learning center which is known as the Topeka Adventure Center. See Brown, 671 F. Supp. at 1309 (discussing Adventure Center).

In March 1976, a divided board adopted a long-range facilities plan, after sharing the proposal with and receiving input from the advisory committee and the community. Rec. ex. vol. II at 96, 99; vol. X at 1437. The plan was a major restructuring of the Topeka schools and it required five years for implementation due to budgetary limitations (cost of capital improvements) and to avoid a reduction in force. Rec. ex. vol. II at 97-98; rec. vol. X at 1454-55 (deputy superintendent Henson). The plan had three purposes: economic efficiency resulting in better conditions for students and staff alike, educational efficiency with respect to facilities and improvement of racial composition in the schools. Rec. vol. X at 1439. Board member Stratton noted that financial concerns were important because of declining enrollment — the district had lost almost one-quarter of its enrollment (6,000 students) in five years. Rec. ex. vol. II at 97. Board member Hurd noted the overwhelming support in the community for the neighborhood school concept. Id. at 98. The plan contemplated the construction of new schools, the closing of older facilities, and the redrawing of attendance boundaries, taking into consideration the anticipated effects on the racial composition of the schools. For our purposes, the principal components of the plan, which this court refers to as a "brief flurry of action," Court's Opinion at 874, included: elimination of all remaining optional attendance areas; closure of five junior high schools (Boswell, Capper, East Topeka, Highland Park and Roosevelt); transfer of the ninth grade to the senior high schools in the 1980-81 school year, conversion of Central Park Elementary to a middle school and construction of a new middle school at the Holliday-State Street site; and the closure of six elementary schools (Central Park, Grant, Parkdale, Polk, Rice and Sheldon). Rec. ex. vol. II at 97-99, 100 (elimination of optional attendance zone at high schools); rec. vol. X at 1439-43 (discussion of middle school concept). In April, HEW officials came to Topeka for two days and discussed the long-range facilities plan, the capital improvements plan, the citizen's advisory group, the affirmative action plan and the Adventure Center. Rec. ex. vol. II at 101.

In September 1976, HEW moved to dismiss its administrative enforcement proceedings on the grounds that the school district had "adopted a plan to remedy the violations of Title VI. . . alleged in the Notice of Opportunity for hearing in this matter." Id. at 175. An order of dismissal without prejudice was entered. Id. at 173. In April 1979, HEW received a complaint from Richard Gellar, M.D., alleging that the district's proposed closing of Central Park Elementary (which was to be converted into Robinson Middle School) "could result in the segregated placement of students within the District on the basis of race." Id. at 181. Dr. Gellar then amended his complaint to allege that the implementation of the Long-Range Facilities Plan "has resulted in increased segregation of students in schools since 1976." Id. at 184.

The Office of Civil Rights conducted an investigation and concluded that "there is not sufficient evidence to conclude that the District's pupil assignment practices implemented pursuant to the adoption of its Long Range Facilities Plan violate Title VI." Id. at 185. The Office found that the district was complying with the Long-Range Facility Plan and "reassigning students in a nondiscriminatory manner." Id. at 186. The Office also concluded that an open enrollment policy, which permitted students to attend a school other than their neighborhood school, "had not significantly affected the racial composition of its schools at this time," but continued monitoring was promised. Id. at 187. The concern was that open enrollment would contribute to increased minority enrollment at schools with high minority percentages. During the second year of the open enrollment policy, the board imposed a majority to minority limitation on the program which assured that transfers would only be approved if they enhanced racial balance. See rec. vol. XIII at 2700-01.

By the 1981-82 school year, the Long-Range Facilities Plan was implemented. Due to the conversion to middle schools, approximately 190 teachers were reassigned in 1980 "to achieve a distribution of minority staff members which will comply with the requirements of the law." Rec. ex. vol. IV at 218. In 1981-82, the board operated 26 elementary schools, 6 middle schools, and 3 high schools. See infra note 18. The administrative staff designed the Long-Range Plan to have an integrative effect as the district court apparently recognized, but this court ignores. Rec. vol. X at 1460-1461, XIII at 2719 (deputy superintendent Henson); Brown 671 F. Supp. at 1299-1300. For example, the junior high school areas formerly served by East Topeka (71.4% minority in 1979), were combined with those of Holliday and Curtis (lower minority percentages), to be served by Chase Middle School which opened in 1980 with 38.5% minority students. Rec. vol. IV at 106, 110, X at 1461, XIII at 2722. Rice Elementary (33.6% minority in 1980), which the district court noted had a desirable racial balance, 671 F. Supp. at 1299, was closed, but its attendance area was divided between Lafayette and Belvoir, thereby reducing the minority percentage in these schools. Rec. ex. vol. IV at 109; vol. X at 1461; vol. XIII at 2721. Lafayette went from 61.3% minority to 56%, while Belvoir went from 75.6% to 62.8% minority. Rec. ex. vol. IV at 109, 122. Likewise, the closing of Crane Junior High (52.9% minority in 1974) resulted in most of those students attending Boswell which increased minority students from 14.6% to 28.5%. Id. at 88, 91. Then Boswell (42.5% minority in 1979) was combined with Roosevelt (19.5% minority in 1979), and all students were sent to the new Robinson Middle School which opened with 32.9% minority students. Id. at 106, 110; rec. vol. XIII at 2722-23. Moreover, this court points out that "the target minority percentages approved by HEW in 1976 have either been met or improved on." Court's Opinion at 878 n. 76.

With one exception resulting in the combination of two middle schools (Landon into French), the school boundaries for U.S.D. 501 remained constant from the 1981-82 school year until the end of the 1985-86 school year. Rec. vol. X at 1503-04. All schools remained open and the grade configurations remained the same. Thus, in analyzing whether there is a current condition of segregation, the district court had stable boundaries for a five-year period and could analyze data which was consistent (compiled the same way) and comparable (measuring the same phenomenon). What is apparent is that the racial composition remained stable, and even became more integrated, given slightly increased minority enrollment. Brown, 671 F. Supp. at 1298.

IV.

Having discussed the two desegregation plans completely implemented by the board as the law developed, I now contrast the district court findings with this court's understanding of those findings. As an initial matter, it is useful to consider the scope of the school board's affirmative duty to desegregate and the scope of our review of the district court's findings.

A.

This court tells us that "[w]hat more can and should be done, if anything, is the final component in a determination of unitary status." Court's Opinion at 865. It is up to the school district to "demonstrate that it has done everything feasible." Id. at 866. Topeka has implemented two desegregation plans, one court-ordered and one effectively imposed by the federal government. Apparently, this court does not think that either plan counts for much, repeating in stentorian tone, that there is "no pattern to the changes over the years." Court's Opinion at 878 (emphasis omitted). Past efforts do not conform to the court's vision of that which must be done. See id. at 878 ("The school district has unquestionably had the opportunity to draw up and execute a scheme designed to lead to comprehensive desegregation.") (emphasis added). This court is unable to see the desegregative effect of many changes because it has strayed far beyond the purpose of desegregation law.

To comply with federal constitutional standards, the purpose of a desegregation plan, whether voluntary or decreed, is to eliminate an unconstitutional dual system and its vestiges arising from state-imposed segregation. Swann, 402 U.S. at 15, 28, 91 S.Ct. at 1275, 1282. A desegregation plan should be matched with the nature and scope of the constitutional violation. Id. at 16, 91 S.Ct. at 1276; see also Pasadena City Board of Education v. Spangler, 427 U.S. 424 at 434, 96 S.Ct. 2697 at 2703, 49 L.Ed.2d 599 (1976); Milliken v. Bradley (Milliken I), 418 U.S. 717, 738, 94 S.Ct. 3112, 3124, 41 L.Ed.2d 1069 (1974). The plan should be remedial, designed to restore the victims of discriminatory conduct to the position they would have been in had the discriminatory conduct not occurred. Milliken I at 746, 94 S.Ct. at 3128. And in evaluating the plan, the intrinsic value of state and local control of public education should be considered. Dayton v. Board of Educ. (Dayton I), 433 U.S. 406 at 410, 97 S.Ct. 2766 at 2770, 53 L.Ed.2d 851 (1977). To those three ends, Milliken II, 433 U.S. at 280-81, 97 S.Ct. at 2757-58, a voluntary plan should accomplish the most desegregation feasible. Feasibility includes not only geographic factors, but also funding and transportation factors. Lee v. Anniston City School System, 737 F.2d 952, 957 (11th Cir. 1984). "The effectiveness of a remedy is the question and it makes no sense to construct decrees that do not grasp the real world." United States v. Pittman, 808 F.2d 385, 394 (5th Cir. 1987) (Higginbotham, J., concurring). Or, as stated by the First Circuit: "A realistic approach, moreover, serves the broader objective of ensuring that the courts do not intervene in school affairs any longer than is strictly necessary." Morgan, 831 F.2d at 324.

Thus, unitary or desegregated status does not mean that every school must approach the system-wide average of minority composition. As noted by the Fifth Circuit:

The constitution does not require school districts to achieve maximum desegregation; that the plan does not result in the most desegregation possible does not mean that the plan is flawed constitutionally. "The constitutional command to desegregate the schools does not mean that every school in the community must always reflect the racial composition of the school system as a whole." [ Swann, 402 U.S. at 24, 91 S.Ct. at 1280.]. "The school board's constitutional duty is to cure the continuing effects of the dual system, not to achieve an ideal racial balance." [ Lee v. Tuscaloosa City School System, 576 F.2d 39 41 (5th Cir. 1978).].

Monteilh v. St. Landry Parish School Bd., 848 F.2d 625, 632 (5th Cir. 1988) (footnotes omitted); accord Morgan, 831 F.2d at 325 ("Both the Supreme Court and this court have repeatedly stated that a judicially imposed desegregation remedy goes too far if it attempts to engineer some sort of idealized racial balance in the schools."); Flax v. Potts, 864 F.2d 1157, 1160 (5th Cir. 1989) ("There is no constitutional mandate that each school in the school district reflect the racial composition of the school district as a whole."); Pitts v. Freeman, 755 F.2d 1423, 1427 (11th Cir. 1985) ("We do not hold, however, that the defendants' affirmative duty compels them to adopt the most desegregative alternative available."); Lee, 737 F.2d 956-57.

When the constitutional violation is purposeful separation of the races caused by school officials in student and faculty/staff assignment, "[t]he mix that would have occurred but for the racism is a judicially created hypothetical." United States v. Overton, 834 F.2d 1171, 1176 (5th Cir. 1987). The remedy is to match the school system to a model of unitary or desegregated status. Id. at 1176-77. A major focus of this court's liability analysis is on schools which are predominantly white due to residential migration to the western part of Topeka. But the dominant purpose of desegregation law is to remedy the racial discrimination against minority groups, discrimination which was ineluctably part of a de jure system. See Price, 694 F.2d at 364-65. Merely because some schools are largely white in this predominantly white system is not "per se offensive to the Constitution." Valley v. Rapides Parish School Bd., 702 F.2d 1221, 1226 (5th Cir.), cert. denied, 464 U.S. 914, 104 S.Ct. 276, 78 L.Ed.2d 256 (1983). What would be offensive to the Constitution would be the retention of all-minority or virtually all-minority schools "where reasonable alternatives may be implemented." Id. (quoted in Monteilh, 848 F.2d at 633); see, e.g., United States v. Pittman, 808 F.2d 385, 391 (5th Cir. 1987) (rejecting magnet school plan which did not eliminate 2 one-race black schools serving 40% of the black elementary students). Given the district court's findings, that condition is not present here because: 1) there are no all-minority or virtually all-minority schools in this system, and 2) those schools which have higher than average concentrations of minority students are not due to the effect of segregative actions by the defendants given the well-supported findings of the district court.

The court also indicates its concern with the few schools which have majority-minority or near majority-minority student enrollments, see supra note 18. Court's Opinion at 869 n. 52, 872-73 n. 59. In a system which has achieved unitariness, a "no majority of any minority" requirement would be beyond what the Constitution requires. Pasadena City Bd. of Educ. v. Spangler, 427 U.S. 424, 434, 96 S.Ct. 2697, 2703, 49 L.Ed.2d 599 (1976). But even in a system which has not been declared unitary, the lack of absolute racial balance which is not attributable to the school board is not tantamount to hindering the process of desegregation.

B.

Although this case involves constitutional claims of racial segregation by school officials, it is essential to keep in mind the proper division of functions between this court and the district court. Dayton Board of Educ. v. Brinkman (Dayton I), 433 U.S. 406, 410-11, 97 S.Ct. 2766, 2770-71, 53 L.Ed.2d 851 (1977). The district court is charged with hearing the evidence and rendering findings of fact. Our task is to review those findings under the clearly erroneous standard, with due regard for the power which Congress has invested in the district courts to make such findings. See Fed.R.Civ.P. 52(a). "The reviewing court oversteps the bounds of its duty under Rule 52(a) if it undertakes to duplicate the role of the lower court." Anderson, 470 U.S. at 573, 105 S.Ct. at 1511. Overstep, we have!

"Findings of fact, whether based on oral or documentary evidence, shall not be set aside unless clearly erroneous." Fed.R.Civ.P. 52(a). A finding is clearly erroneous "when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed." United States v. United States Gypsum Co., 333 U.S. 364, 395, 68 S.Ct. 525, 542, 92 L.Ed. 746 (1948) (emphasis added). This standard is not a license to replace the district court's view of the evidence with that of the reviewing court; the reviewing court may not reverse merely because it would have viewed the evidence differently or "given greater weight to certain evidence." 5A J. Moore J. Lucas, Moore's Federal Practice, ¶ 52.03[1] (1989); accord Inwood Laboratories, Inc. v. Ives Laboratories, Inc., 456 U.S. 844, 855-58, 102 S.Ct. 2182, 2189-91, 72 L.Ed.2d 606 (1982); United States v. National Ass'n of Real Estate Bds., 339 U.S. 485, 495-96, 70 S.Ct. 711, 717-18, 94 L.Ed. 1007 (1950). "Where there are two permissible views of the evidence, the factfinder's choice between them cannot be clearly erroneous." Anderson, 470 U.S. at 574, 105 S.Ct. at 1511. Applying these standards, the district court should be affirmed.

There is "great value" in relying on the factual findings of the district court. Dayton II, 443 U.S. at 534-35 n. 8, 99 S.Ct. at 2977-78 n. 8. The intent inquiry in school desegregation litigation almost always concerns objective numerical evidence that is circumstantial, in addition to testimonial evidence by school officials and other desegregation experts. Given the law, the intent inquiry in a desegregation case, including whether segregative intent from 1954 and beyond remains due to non-compliance with the board's affirmative duty, "is an issue that can present very difficult and subtle factual questions." See Columbus, 443 U.S. at 470-71, 99 S.Ct. at 2983-84 (Stewart, J., concurring in result).

This court apparently views its initial inquiry, concerning a current condition of segregation, as largely numerical. But "[w]here numbers alone are insufficient to define racially identifiable schools," the court invokes a broader form of racial identifiability which concerns "demography, geography, and the individual history of particular schools and areas of the city," to virtually assure liability. See Court's Opinion at 861. This approach is the one urged by plaintiffs' expert, Mr. Lamson. See rec. vol. III at 285-286, vol. IV at 459, 493-94, 522. While it is indisputable that racial identifiability may exist without reference to student assignment, and that faculty/staff assignment, facilities or extracurricular activities may result in a school being racially identifiable, see Keyes, 413 U.S. at 196, 93 S.Ct. at 2691 (citing Keyes v. School District No. One, 313 F. Supp. 61 at 74 (D.Colo. 1970); Swann, 402 U.S. at 18, 91 S.Ct. at 1277, the court's expanded definition of racial identifiability is not supported by several of the cases it cites. Court's Opinion at 861 n. 25.

The court justifies its de novo review of the numerical data as follows:

As a general matter, it is important to note that much of the record evidence consists of statistics and other undisputed facts. Our differences with the district court lie mainly in how the essentially undisputed facts are assessed in light of the school district's mandate to dismantle the segregated school system. We believe that the district court's finding of unitariness is flawed by the undue deference it gave to the school district's neighborhood school policy and the court's failure to give the proper weight to its own findings that certain actions and omissions by the school district had a segregative effect.

Id. at 868-69 (footnotes omitted). The court then makes the rather incredible statement that, in reaching its decision, "we did not rely upon the experts' opinions." Id. at 868 n. 50; compare id. at 869 n. 52 ("In a school district with a 26% minority student population, we think the number of schools in Topeka that approach or are over the 50% minority mark constitute persuasive evidence that the school system has not met its duty to desegregate.") with rec. vol. V at 622-23 (Dr. Foster's citing the presence of 3 elementary schools which are over 50% minority and 4 which are over 42% minority when asked his opinion of whether vestiges of the former dual school system exist).

The court contends that its disagreement with the district court merely concerns assessment of "essentially undisputed" facts, id. at 869, and of "basically uncontroverted evidence," id. at 868 n. 50; however, in no way can it be said that the implications of the numerical date in this case were uncontested. Indeed, the month-long trial of this case belies that notion. Moreover, the characterization of the numerical date is inherently the province of the district court.

In 1985, the Supreme Court settled that the clearly erroneous standard applies "even when the district court's findings do not rest on credibility determinations, but are based on physical or documentary evidence or inferences from other facts." Anderson, 470 U.S. at 574, 105 S.Ct. at 1511. In 1985, Rule 52 also was amended to encompass documentary evidence. The Advisory Committee addressed and rejected considerations offered in support of the view that it was not necessary to give special deference to factual determinations based on documentary evidence as opposed to credibility determinations:

These considerations are outweighed by the public interest in the stability and judicial economy that would be promoted by recognizing that the trial court, not the appellate tribunal, should be the finder of the facts. To permit courts of appeals to share more actively in the fact-finding function would tend to undermine the legitimacy of the district courts in the eyes of litigants, multiply appeals by encouraging appellate retrial of some factual issues, and needlessly reallocate judicial authority.

Advisory Committee on Civil Rules, Proposed Amendments to the Rules of Federal Civil Procedure, 105 F.R.D. 218, 223 (1984). There are sound policy reasons for adhering to the clearly erroneous rule in all cases, whether documentary or testimonial evidence is involved.

C.

Another reason given for this court's rejection of the district court's factual findings is that it gave "undue deference" to the school district's neighborhood school policy. Court's Opinion at 869. A neighborhood school policy is not, in and of itself, violative of the fourteenth amendment. Crawford v. Board of Educ., 458 U.S. 527, 537 n. 15, 102 S.Ct. 3211, 3217 n. 15, 73 L.Ed.2d 948 (1982). Just as equal educational opportunity is a national policy, so too is neighborhood schooling. See 20 U.S.C. § 1701. There are significant advantages to neighborhood schooling, such as students walking to school, the absence of long bus rides, easier participation in extracurricular activities and possibly greater parental involvement. Of course, the advantages of neighborhood schools cannot outweigh the school board's duty to comply with the Constitution and "dismantle the dual school system." Swann, 402 U.S. at 28, 91 S.Ct. at 1282. To that end, this court has insisted that a neighborhood school plan be "impartially maintained and administered," and "not used as a mask to further and perpetuate racial discrimination." Dowell, 375 F.2d at 166.

The Supreme Court has spoken in the remedial phase of school desegregation litigation concerning neighborhood school plans:

Absent a constitutional violation there would be no basis for judicially ordering assignment of students on a racial basis. All things being equal, with no history of discrimination, it might well be desirable to assign pupils nearest their homes. But all things are not equal in a system that has been deliberately constructed and maintained to enforce racial segregation. The remedy for such segregation may be administratively awkward, inconvenient, and even bizarre in some situations and may impose burdens on some; but all awkwardness and inconvenience cannot be avoided in the interim period when remedial adjustments are being made to eliminate the dual school systems.

No fixed or even substantially fixed guidelines can be established as to how far a court can go, but it must be recognized that there are limits. The objective is to dismantle the dual school system. "Racially neutral" assignment plans proposed by school authorities to a district court may be inadequate; such plans may fail to counteract the continuing effects of past school segregation resulting from discriminatory location of school sites or distortion of school size in order to achieve or maintain an artificial racial separation. When school authorities present a district court with a "loaded game board," affirmative action in the form of remedial altering of attendance zones is proper to achieve truly non-discriminatory assignments. In short, an assignment plan is not acceptable simply because it appears to be neutral.

Swann, 402 U.S. at 28, 91 S.Ct. at 1282. As the above passage makes clear, it is a present constitutional violation which empowers a federal court to depart from a neighborhood school plan in favor of some other form of assignment in order to dismantle a dual school system. See Davis v. Board of School Commr's, 402 U.S. 33, 37, 91 S.Ct. 1289, 1292, 28 L.Ed.2d 577 (1971) (remedy dependent on having found a violation). If the dual school system has been dismantled and there is no constitutional violation, a neighborhood school policy is constitutionally permissible.

In the liability phase of this case, it was entirely appropriate for the district court to consider whether the school district had maintained and administered its neighborhood school policy in a neutral fashion, particularly because the district court found that "the effects of school segregation on residential choice, if any, are not cognizable today." Brown, 671 F. Supp. at 1297 (emphasis added). There are other reasons to reject the court's criticism of the district court's findings on this point. See Court's Opinion at 867, 867 n. 48. First, the district court was compelled to make its findings that the board maintained its neighborhood school policy in a race-neutral fashion because the major part of the plaintiffs' case was devoted to proving otherwise. See, e.g., rec. vol. II at 23-26; vol. X at 1418; supp. vol. XI at 241 (summaries of Mr. Lamson's findings). That is crystal clear on this 3,000 page plus record. The district court found that the neighborhood boundaries were not gerrymandered. Brown, 671 F. Supp. at 1301. Second, a fair reading of the district court's opinion in this case makes it evident that the district court was well aware of its responsibility to look beyond district evidence concerning the intent of the school board, and also consider a foreseeability test in deciding whether the school board had failed to dismantle the dual system. The district court looked beyond the neighborhood school plan for segregative effect, thoroughly analyzing quantitative measures of racial composition.

D.

Another reason the court gives for rejecting the district court's factual findings is that this court would weigh the evidence differently. Court's Opinion at 869 (discussing "court's failure to give proper weight to its own findings that certain actions and omissions by the school district had a segregative effect" and citing Diaz v. San Jose Unified School Dist., 733 F.2d 660 (9th Cir. 1984) (en banc), cert. denied, 471 U.S. 1065, 105 S.Ct. 2140, 85 L.Ed.2d 497 (1985), for an analogous proposition about clearly erroneous district court findings concerning segregative intent). Regrettably, this court does not give a fair recitation of the district court findings. We are told: "The district court made the following findings:

(1) that the neighborhood school attendance boundaries drawn in 1955 had the effect of maintaining segregation." Court's Opinion at 867 (emphasis added). What the district court said was:

A review of the school boundaries as they have developed over thirty years does not reveal a segregative pattern that remains today. The boundaries set around the former de jure black elementary schools after this case was remanded by the Supreme Court appear to have perpetuated the racial identity of those schools. But, the schools have long been closed and the segregative effects of those boundaries have attenuated entirely.

Brown, 671 F. Supp. at 1300. It was not clearly erroneous for the district court to find that 1955 boundaries for closed schools do not figure large into the current situation, over 30 years later. Though a court must look closely and carefully, conduct which resulted in segregative effect thirty years ago may indeed be so attenuated as to be insufficient to support a finding of intentional segregation caused by the school board. Keyes, 413 U.S. at 211, 93 S.Ct. at 2698; Higgins v. Board of Educ., 508 F.2d 779, 792 (6th Cir. 1974). Moreover, this finding demonstrates that the district court did consider the effect of school board action or inaction.

(2) "that the construction of new schools since that time had the effect of `promot[ing] racial separation;'" Court's Opinion at 867 (emphasis added).

The district court said:

Although, on its face, the construction of schools, particularly on the west side of the district, appears to have promoted racial separation, the court does not believe that the district's school construction policy was intended to maintain or promote segregation. Most of the construction occurred during the peak years of student enrollment. More schools were needed. The location of the schools in areas of residential expansion was consistent with the race-neutral neighborhood school concept. There is no evidence that the students attending these schools could have been accommodated in the existing schools or that land was available in the older, more racially mixed areas of the district to build new schools.

. . . .

In sum, it does not appear that the district's school construction policy has promoted segregated residential patterns or segregated schools.

Brown, 671 F. Supp. at 1300 (emphasis added). This is a key finding, because with it, the trial court rejected Mr. Lamson's theory that the school board could have sent students to inner city schools with higher minority concentrations that were underutilized, rather than construct, new facilities in areas that were expanding and predominantly white. See, e.g., rec. vol. II at 150-51, 179; III at 268-73, 311-14, 334-36; supp. vol. XI at 40-49. Contrary to this court's opinion, the trial court did not find that school construction since 1955 promoted racial separation.

(3) "that the reassignment of students from previous de jure schools to adjacent schools with higher-than-average percentages of minority students had the effect of increasing those percentages;" Court's Opinion at 867 (emphasis added). The district court discussed only one situation involving a de jure school on the page cited by this court, namely the transfer of Monroe students to Quinton Heights elementary school. See 671 F. Supp. at 1301. Monroe was closed at the end of the 1974-75 school year because at the beginning of that school year it only had 102 students. Rec. ex. vol. IV at 84, 90. Students were reassigned to Polk, Highland Park North and Quinton Heights. Rec. supp. vol. XI at 173. The minority percentage at Quinton Heights went from 37% (84 out of 227) in 1974-75 to 39.09% (77 out of 197) in 1975-76. Rec. ex. vol. IV at 87, 94. Thus, after Monroe closed, 7 fewer minority students attended Quinton Heights even after the Monroe students were added. Given that the percentage increase (2.09%) of minority enrollment was so small and was accompanied by a decrease in the absolute number of minority enrollment, it is hard to fault the district court for finding that the school district did not act improperly when it sent Monroe students to three neighborhood schools, including Quinton Heights.

(4) "and that the assignment of faculty had the effect of placing minority faculty disproportionately at schools with higher-than-average minority student percentages." Court's Opinion at 867 (emphasis added).

The district court said:

Plaintiffs' major contention with regard to faculty and staff is that the assignment of faculty and staff has served to racially identify the schools in the district.

. . . .

Examining the school system as a whole, despite the tendency to have more minority staff in schools with a greater than average minority student population, the court does not believe the district's assignment policies serve to identify schools as intended for white or black students. The pattern identified by plaintiffs is not monolithic. The current percentages of minority staff in the district's schools are generally within ranges acceptable to other courts in desegregation litigation. Other facets of faculty and staff policies are nondiscriminatory. In sum, the record before the court with regard to the district's approach to faculty and staff is not indicative of a dual system of education.

Id. at 1304-05. As discussed below, the faculty/staff assignment pattern in this case was almost within the range set by plaintiffs' own expert, Dr. Foster, for the one year in which substantially complete data was available to him. See infra note 31 and accompanying text. The district court's finding that the faculty/staff assignment was not indicative of a current condition of segregation is supported by the record, particularly considering the assignment figures for the 1986-87 year.

Continuing with its description of the lower court findings, this court concludes that (5) "the school district's use of space additions, [(6)] its siting of Topeka West high school, [(7)] its drawing of attendance boundaries, and [(8)] its failure to adopt various reorganization plans did not further the process of desegregation." Court's Opinion at 867 (emphasis added) (citing Brown 671 F. Supp. at 1298-1301, 1308-09). That is only true if desegregation is viewed as replicating the system-wide minority average in every school, in addition to eliminating a de jure system and its vestiges. The emphasis of the plaintiffs' case was on the alleged segregative intent and effect of school board actions and the board's inattention to racial balance. The board came forward with legitimate, non-discriminatory reasons (strong enough to disprove segregative intent) for many of its actions and attributed the racial composition of several schools to demographic forces. More importantly, the school board was able to convince the district court that there was not a current condition of segregation in Topeka, something which is amply supported by the record and not refuted by this court's opinion.

Contrary to the study by the plaintiffs' expert, the district court found that space additions were not "intentionally used to promote segregation or that schools are racially imbalanced today because of space additions." Brown, 671 F. Supp. at 1299.

As for the siting of Topeka West high school, it must be remembered that Topeka never had a dual system of education at the high school level. The site for Topeka West was acquired in the early or mid-50's from a single family because it was big (40 acres) and well-located. Rec. vol. X at 1433, 1506-07. The district court found that the construction of Topeka West was a logical response to the overcrowding of Topeka High and the rapid population growth which was occurring in the western part of the district. Brown, 671 F. Supp. at 1300; see also rec. ex. vol. I at 14 (building patterns indicating that over half of new housing build in Topeka from 1950-70 was in the west area); rec. vol. X at 1432-33 (Topeka High School was overcrowded; Topeka West was built, without a planned athletic facility, only after a second bond election). Topeka doubled in population between 1955 and 1965. Rec. supp. vol. VII at 2094. Indeed, the enrollment at Topeka West was so great that some of the attendance area was transferred back to Topeka High School, furthering integration. Rec. vol. X at 1434. A site in west Topeka made sense because the newly annexed Highland Park High served the east part of the city, while Topeka High was centrally located. Brown, 671 F. Supp. at 1300; rec. ex. vol. I at 14. Moreover, the site was large enough to meet the school board's needs. Rec. vol. X at 1507.

Reorganization plans N and X, proposed in 1984, assumed that no school would have a minority enrollment of greater than 50%, and projections indicated that number would be reduced to 45%. Rec. ex. vol. V at 51; Brown, 671 F. Supp. at 1308-09. In 1985-86, three schools had majority-minority populations: Belvoir (61.86%); Lafayette (56.81%) and Highland Park North (57.93%). Brown, 671 F. Supp. at 1294. The district court was correct in finding that Plans N and X did not constitute a forgone opportunity to improve the racial composition of the schools. Indeed, the failure of Topeka to adopt Reorganization Plans N and X in 1984 should count for nothing in this case. See Court's Opinion at 877 n. 75 (acknowledging opposition to plans). If there is one consistent theme in this record it is that the plans were expensive ($13 million for N; $15 million for X) and that they were opposed by every segment in Topeka. See rec. vol. XII (testimony of former board members) at 2451 (opposition came from every quadrant geographically, as well as every racial segment), 2452 (plans were expensive), 2467 (cost of plans), 2471-72 (plans would have closed many schools and created some very large schools), 2479 (no sector of community would support plans), 2481 (NAACP vigorously opposed plans); vol. XIII (testimony of deputy superintendent) at 2728 (community attached tremendous importance to neighborhood schools and opposed plans).

In determining that there is no current condition of segregation in Topeka, the district court carefully examined 18 factors: statistical measures of student population; student transfer policy; optional attendance zones; space additions; school closings; school openings; school site locations; school boundary locations; history of the schools; facilities; extracurricular activities; curriculum; transportation; faculty and staff; community attitudes; equality of education; opportunities for desegregation; and other factors. The district court properly examined the traditional factors which bear on unitariness, in addition to those factors raised by plaintiffs which might have a bearing on unitariness. Baliles, 829 F.2d at 1312. Its findings in favor of the defendants in each area are supported by the record, and its ultimate finding of unitariness is likewise supported.

V.

This court's criticism of the district court is succinctly stated:

The court evidently believed that if these two criteria, i.e., no intent to discriminate and consistency with a race-neutral neighborhood school plan, were met, the school district's actions would pass constitutional muster.

. . . .

While it did find that the school district had taken some actively desegregative actions, we are convinced that the court's overall conclusion on unitariness was fatally infected by the inadequacy of the burden of proof to which it held the school district.

Court's Opinion at 867, 868. The district court's attention to the neighborhood school plan already has been discussed in part. See supra pp. 913-14. This court misses the mark if it is implying that intent is not relevant in this case. At a minimum, there must be a finding that the condition complained of is the effect of discriminatory intent, whether that discriminatory intent originated in 1954 or since. The district court made permissible inferences given the evidence. It determined that there was not a current condition of segregation and, therefore, the plaintiffs were not entitled to a presumption that the racial imbalance in the school system was a product of past segregative actions.

The whole of desegregation law is predicated upon a constitutional violation. "The school desegregation cases have also adhered to the basic equal protection principle that the invidious quality of a law claimed to be racially discriminatory must ultimately be traced to a racially discriminatory purpose." Washington v. Davis, 426 U.S. 229, 240, 96 S.Ct. 2040, 2048, 48 L.Ed.2d 597 (1976). The Constitution does not employ a no-fault approach to liability when it comes to school board actions. Contrary to the court's opinion, intent is still relevant because it bears on causation. Only intentional state action which brings about or maintains segregation results in a constitutional violation. Columbus, 443 U.S. at 464, 99 S.Ct. at 2950. Even in a system that once was dual, whether by statute or practice, "[d]isparate impact and foreseeable consequences, without more, do not establish a constitutional violation . . . . [But] actions having foreseeable and anticipated disparate impact are relevant evidence to prove the ultimate fact, forbidden purpose." Id. Thus, a district court may consider a school board's adherence to a neighborhood school policy which may lead to racial imbalance, as one factor among many, in deciding "`whether an inference of segregative intent should be drawn.'" Id. at 465, 99 S.Ct. at 2950 (quoting Penick v. Columbus Bd. of Educ., 429 F. Supp. 229, 255 (S.D.Ohio 1977), aff'd in part and vacated in part, 583 F.2d 787 (6th Cir. 1978), aff'd, 443 U.S. 449, 99 S.Ct. 2941, 61 L.Ed.2d 666 (1979)). "The inference is permissive, not mandatory." Higgins, 508 F.2d at 793 ("While it is true that a court may infer such an intent [purposeful segregation] from the circumstances there is no authority for the proposition that such an intent must be inferred in all cases where segregated patterns exist in fact") (emphasis in original).

In Dayton II, the Court said:

We have never held that as a general proposition the foreseeability of segregative consequences makes out a prima facie case of purposeful racial discrimination and shifts the burden of producing evidence to the defendants if they are to escape judgment; and even more clearly there is no warrant in our cases for holding that such foreseeability routinely shifts the burden of persuasion to defendants. Of course, as we hold in Columbus today, ante, [443 U.S.], at 464-65 [ 99 S.Ct. at 2950-51], proof of foreseeable consequences is one type of quite relevant evidence of racially discriminatory purpose, and it may itself show a failure to fulfill the duty to eradicate the consequences of purposefully discriminatory conduct.

Dayton II, 443 U.S. at 536 n. 9, 99 S.Ct. at 2978 n. 9. Merely because Topeka built and operated neighborhood schools that fall outside the statistical measures proposed by plaintiffs' experts does not mean that there is a current condition of segregation in Topeka. After two desegregation plans, this case is, at best, one of "foreseeability" based upon effect, although plaintiffs contended that they had shown more. See, e.g., rec. vol. II at 23-26 (opening statement), III at 381 (plaintiffs' expert cataloged effect of school board actions), IV at 449 (plaintiffs' expert took "racial balance as your starting point," in proposing alternatives to board's administrative choices).

To be sure, "the measure of the post- Brown I conduct of a school board under an unsatisfied duty to liquidate a dual system is the effectiveness, not the purpose, of the actions in decreasing or increasing the segregation caused by the dual system." Dayton II, 443 U.S. at 538, 99 S.Ct. at 2979; Columbus, 443 U.S. at 458-59, 99 S.Ct. at 2946-47; Green, 391 U.S. at 437-38, 88 S.Ct. at 1693-94. That, however, does not bar the district court from considering the school board's evidence that its actions have been effective in dismantling the dual system, that the current racial composition of the school system is not a product of discriminatory intent originating before Brown I and living on in effect to the present day. This court's opinion, by downplaying intent and causation, specifically the school board's evidence that segregative effects of past discriminatory intent no longer remain, has imposed liability based on the inferences it wishes to draw from certain objective evidence and testimony upon which reasonable people could differ as to interpretation. On the surface, this process may appear to have the advantage of simplicity and consistency. But the close evidence in this case was hardly simple and we are not allowed such consistency at the expense of district court findings which have support in the record.

The district court was well aware that quantitative date in school desegregation cases may be indicative of past or present segregative intent. Brown, 671 F. Supp. at 1295. It also was cognizant of the school board's affirmative duty to dismantle the dual system. Id. at 1292-93. Finally, the district court recognized that it could draw an inference of segregative intent or an inference of the lack thereof, based upon the efficacy of the school board's desegregation efforts, including compliance with its affirmative duty:

Evidence of segregative motive or the absence of such intent is relevant but not controlling in determining unitariness. "The measure of the post- Brown I conduct of a school board under an unsatisfied duty to liquidate a dual system is the effectiveness, not the purpose, of the actions in decreasing or increasing the segregation caused by the dual system." Dayton II, supra, 443 U.S. at 538, 99 S.Ct. at 2935.

In sum, a unitary school system is one in which the characteristics of the 1954 dual system either do not exist or, if they exist, are not the result of past or present intentional segregative conduct of defendants or their predecessors.

Brown, 671 F. Supp. at 1293.

The court shall examine many factors to decide whether illegal segregation exists in U.S.D. # 501. But, the statistical measures of racial makeup of the schools' student bodies are of prime importance. Often in school desegregation cases such statistics alone prove a pattern of discrimination.

Id. at 1295.

If a district has consistently dragged its feet on desegregation, then the vestiges of the segregated system may remain.

Id. at 1308. Thus, I vehemently disagree with the court's conclusion that the district court incorrectly considered intent, while ignoring the board's affirmative duty.

The district court in this case discussed those factors which supported an inference of segregative intent or non-compliance with the affirmative duty to desegregate. It then explained why it rejected those inference. See Alexander v. Youngstown Bd. of Educ., 675 F.2d 787, 792-93 (6th Cir. 1982) (trial court may reject inferences of segregative intent). That critical discussion, however, is the departure for this court's reversal. See Court's Opinion at 867 (summarizing district court's findings). This court simply does not agree with the district court's declining to draw those inferences. Although this court claims that the district court did not properly evaluate the evidence for segregative effect, no fair reading of the district court opinion would support that conclusion. The district court's exhaustive discussion of statistical racial identifiability and its implication, as well as its detailed treatment of each school alleged to be racially identifiable, belie this court's conclusion that the district court "failed adequately to weigh the conduct of the school district." Id. at 868.

Perhaps the court has decided to dispense with the intent and causation inquiry because it seeks to ameliorate the effects of "subconscious racial discrimination." See Court's Opinion at 865-66. Indeed, the court's remark concerning "longstanding racism" carries with it an implication which is unwarranted. Id. at 866 ("Courts must assess the school district's achievements with an eye to the possible and practical, but they must not let longstanding racism blur their ultimate focus on the ideal."). The court's explanation of the presumption it invokes, which I maintain is the same presumption discussed in Keyes, conforms with the tenor of the court's opinion. "This presumption ensures that subconscious racial discrimination does not perpetuate the denial of equal protection to our nation's school children." Court's Opinion at 863 (footnote omitted).

The court tells us that a presumption different than the one in Keyes, see supra note 20, applies in this case because the school district was operating a de jure system in 1954. Court's Opinion at 862-63, 863 n. 27. Reliance is placed on a footnote in Keyes, 413 U.S. at 211 n. 17, 93 S.Ct. at 2699 n. 17. That footnote is contained in a discussion of defenses the school board may raise to the prima facie case, including the defense that the effects of past segregative acts have been completely attenuated. The footnote explains that once the presumption is invoked, plaintiffs are not forced to prove non-attenuation because the burden shifts to the school board to prove attenuation. Id. Express language in Keyes makes it clear that the presumption is not limited only "to establish[ing] initial de jure segregation in a case where no statutorily mandated segregation exists." Court's Opinion at 863 n. 27. The Keyes Court stated:

Thus, be it a statutory dual system or an allegedly unitary system where a meaningful portion of the system is found to be intentionally segregated, the existence of subsequent or other segregated schooling within the same system justifies a rule imposing on the school authorities the burden of proving that this segregated schooling is not also the result of intentionally segregative acts.

413 U.S. at 210, 93 S.Ct. at 2698; see also Columbus, 443 U.S. at 457 n. 5, 99 S.Ct. at 2946 n. 5.

The reason for the presumption is that "`the prior doing of other similar acts, whether clearly part of a scheme or not, is useful as reducing the possibility that the act in question was done with innocent intent.'" Keyes, 413 U.S. at 207-208, 93 S.Ct. at 2696-97 (quoting 2 J. Wigmore, Evidence 200 (3d ed. 1940)). Thus, "a finding of illicit intent as to a meaningful portion of the item under question has substantial probative value on the question of illicit intent as to the remainder." Keyes, 413 U.S. at 207-08, 93 S.Ct. at 2696-97. The Supreme Court has yet to incorporate subconscious racial discrimination into the equal protection analysis. While a finding of de jure segregation pursuant to statute in 1954 eases the plaintiffs' evidentiary burden, Columbus, 443 U.S. at 457 n. 5, 99 S.Ct. at 2946 n. 5, intent, whether proven by direct or circumstantial evidence, remains an essential element. This court apparently concedes as much, when in the context of its discussion of the burden of proof and unitariness states:

Finally, objective proof of the school district's intent must be considered. How a district lobbies its patrons and government agencies on issues that affect desegregation, whether it seeks and heeds the desegregation recommendations of others, and the cooperativeness of the district in complying with court orders, for example, bear on the manner in which the district has shaped current conditions in the school district.

Court's Opinion at 865 (emphasis added and footnotes omitted). Several cases, including one of our own, also have considered good faith of the school district in deciding unitariness. Dowell v. Board of Educ., 890 F.2d 1483, 1499 (10th Cir. 1989); Morgan, 831 F.2d at 321; Ross, 699 F.2d at 226. Good faith speaks to intent and causation and making such inquiries is inconsistent with claiming that intent is irrelevant.

The court comments that: "A focus on provable intent alone would deny a remedy to too many Americans." Court's Opinion at 863. While this may be true once there has been a finding of a current condition of segregation in a meaningful portion of a school system with a relevant history of segregation by law, see Keyes, 413 U.S. at 210, 93 S.Ct. at 2698, such a finding should be a prerequisite to invoke the presumption discussed in Keyes and insure that any resultant system-wide desegregation order is linked to past or present racial discrimination. What the court has done in this case is to substitute its judgment for that of the district court concerning the degree of imbalance necessary to invoke and sustain the presumption. Nowhere is this more evident than in the court's discussion of whether there is a current condition of segregation with respect to student and faculty/staff assignment in the Topeka schools.

VI.

It is clear that to sustain the presumption, the plaintiffs were required to prove "a current condition of segregation from intentional state action" with respect to "a meaningful portion of [the] school system." Keyes, 413 U.S. at 205, 208, 93 S.Ct. at 2696, 2697. "Significant racial imbalance" in schools may operate as a "signal" to shift burden of proof, but even that "is a very different matter from equating racial imbalance with a constitutional violation calling for a remedy." Milliken v. Bradley (Milliken I), 418 U.S. 717, 741 n. 19, 94 S.Ct. 3112, 3125 n. 19, 41 L.Ed.2d 1069 (1974). Turning first to student assignment, what the court has done is to decide that if certain schools do not fall within certain percentages of minority enrollment, say 11% to 41% or even 10% to 60%, those schools are racially identifiable and indicative of a current condition of segregation resulting from the prior de jure system. Court's Opinion at 869-70.

A.

Plaintiffs' expert Mr. Lamson testified that a school was racially identifiable as black if it had more than twice the system-wide average of black students. Rec. vol. III at 365. In 1985-86, the black student percentage was 18.4%, so a school with than more than 36.8% black students would have been racially identifiable. On the other hand, a school was racially identifiable white if it had less than one-half the system-wide average of black students. Id. Thus, a school would have been racially identifiable white if it had less than 9.2% black students. Moreover, under Mr. Lamson's view of racial identifiability, a school which had the system-wide average of black students still might be racially identifiable and segregated if neighboring schools had a greater percentage of white students. Rec. vol. IV at 531-32. Thus, according to his view, in 1952-53, Belvoir was segregated when it had 20.8% black students and system-wide black percentage was 11%. Id. at 456-59. Likewise, in 1985-86, Robinson Middle School was racially disproportionate, according to Mr. Lamson, when it had 20.84% black students and the system-wide middle school black percentage was 19.01%. Id. at 486-88. Plaintiffs' expert Dr. Foster did not agree with Mr. Lamson that a school which had the system-wide minority average still could be racially identifiable. Rec. vol. V at 687-88.

Dr. Foster advocated using a measure of ± 15% of the system-wide minority percentage to come up with racially identifiable schools, although he acknowledged using ± 20% in other cases he had worked on. Rec. vol. IV at 549-51; vol. V at 705-10. When the system-wide minority percentage was below 15% or 10%, Dr. Foster used a sliding scale of ± 10% or ± 5%. Rec. vol. IV at 553-54, 575-78. The trial judge seems to have analyzed the plaintiffs' case in part with a ± 15% standard. Brown, 671 F. Supp. at 1295-96; see rec. vol. V at 578. But he came to a very different ultimate finding than this court, which has attached far more significance to the measure: "[W]e believe it perfectly reasonable to use a ± 15% deviation in assessing whether the school system ever achieved unitariness." Court's Opinion at 870 n. 54. Using this standard, there were 15 racially identifiable elementary schools in 1985-86; 7 racially identifiable minority, 8 racially identifiable white. Rec. vol. V at 593, 711; supp. vol. XIII (pl. ex. 155D); see Court's Opinion at 872-73 n. 59. There were 4 racially identifiable secondary schools using the ± 15% standard; 1 racially identifiable minority, 3 racially identifiable white. Rec. supp. vol. XIII (pl. ex. 155G); see Court's Opinion at 872-73 n. 59.

The court discusses the testimony of defendant's expert, Dr. Armor, and gives the impression that Dr. Armor suggested a standard whereby a school would not be racially identifiable (segregated) if it contained minority students within a 10% to 60% range. Court's Opinion at 870, 873; see also id. at 884 (court's discussion of "predominantly white schools."). The court then concludes, using this standard, that there are 9 schools which are racially identifiable. Id. at 870. Yet, Dr. Armor was critical of the concept of basing liability on quantitative measures of racial identifiability.

In the interest of completeness, this court should have summarized Dr. Armor's testimony concerning the implications of his "range of acceptability of 10-60%." Id. at 870. This court should have explained that Dr. Armor, under questioning from the court, offered his conclusion, based on three quantitative measures (his absolute standard, the dissimilarity index and the relative exposure index) that the school board had met its affirmative duty under Brown I and had desegregated the elementary grades as to student assignment. Rec. vol. XIII at 2641-42. Had the court done so, it would not have been able to attribute a finding of nine racially identifiable schools by student assignment which result in a current condition of segregation, to Dr. Armor. See Court's Opinion at 870, 873, 884 (invoking Dr. Armor's approach to justify findings).

Dr. Armor did point out that under his 10-60% range, one elementary school (Belvoir; 61.86) exceeded it and five elementary schools fell below it (Crestview, 8.94%; Gage 9.43%; McCarter, 9.16%; McClure, 7.21%; Potwin 7.73%). Rec. vol. XIII at 2580; ex. vol. IV at 170-74. Applying this 10-60% range to the secondary schools, three schools were below (French, 6.23%; Landon, 9.29% [now closed]; Topeka West, 7.94%). See rec. ex. vol. IV at 175-76. However, Dr. Armor preceded and concluded his list with the caveat that merely being outside his range is not necessarily indicative of a current condition of segregation. Rec. vol. XIII at 2579, 2580. This is in accord with plaintiffs' expert Dr. Foster who admitted that racially identifiable schools under his ± 15% standard are not necessarily unlawfully segregated. Rec. vol. VI at 792-93. Dr. Armor testified as follows:

What I have been talking about in terms of this standard or a definition of integrated school has been primarily from the social science conception of what would constitute, what we ought to define as an integrated school or a desegregated school. It doesn't necessarily mean that a school that falls outside that range is, is in violation of the [C]onstitution or is wrong or is improper. One has, in any liability analysis one has to add to that — even my standard one needs to add the analysis of why a school is 5 percent minority or 70 percent minority outside my ideal range because if it's not, if it's that way because of demographic factors, then I don't, I don't see necessarily a basis for why it has to be changed; but I would still hold that ideally or what you strive for ought to be schools within that range.

Rec. vol. XIII at 2579. The district court's finding that "students are not separated on the basis of race," Brown, 671 F. Supp. at 1298, is supported by Dr. Armor's testimony and the factual findings regarding the above schools contained in the balance of the district court's opinion. Id. at 1294, 1301-04.

Assuming, arguendo, that this court's interpretation of Dr. Armor's testimony concerning the 10-60% range had merit, I submit that it would be anomalous to find a current condition of segregation regarding student assignment based on only one elementary school (Belvoir) which exceeded the target minority percentage and just barely (by 1.86%). Regarding the 8 predominantly white schools mentioned by Dr. Armor, in a system with low minority percentages overall, it is not unexpected to find some schools which will have large percentages of white students. Court's Opinion at 869. More importantly, every single one of those predominantly white schools identified by Dr. Armor, with the exception of McCarter (which went from a 9.20% minority percentage to 9.16%), has increased the percentage of minority enrollment when the 1985-86 figures are compared to those in 1981-82. See supra note 18.

Moreover, the district court considered other measures of integration in deciding that unlawful segregation did not exist. Brown, 671 F. Supp. at 1298. Specifically, Dr. Armor testified concerning the dissimilarity index and the relative exposure index which are measures of system-wide integration. It is passing strange for the court to all but dismiss the significance of these measures in its discussion of whether there is a current condition of segregation, yet report them and rely upon them concerning the positive effects of the board's implementation of its second desegregation plan. Court's Opinion at 870 n. 55 (rejecting measures because they would not identify a hypothetical all-black school), 875-76, 876 n. 71 (relying on measures to indicate success of school district's desegregation efforts from 1975-81). The court's criticism of these measures, that "they say nothing about the racial balance in individual schools," Court's Opinion at 870 n. 55, is really only partially correct because the formulas rely upon data concerning each individual school. Dr. Armor readily indicated that these indices are to be supplemented by looking at individual schools. Rec. vol. XIII at 2655; see also Court's Opinion at 870 n. 55. And so the district court did.

The dissimilarity index measures the departure of a school system from perfect racial balance. Id. at 2581. If every school had the system-wide minority composition, there would be perfect racial balance and the index would have a value of zero. Id. On the other hand, if every school were completely segregated, the index value would be expressed as 1.0 (or 100). Id. The relative exposure index measures the opportunity for interracial contact given the number of whites and minorities in the system. Id. at 2582. A value of zero indicates total interracial contact, whereas a value of 1.0 (or 100) indicates a total separation. Id. Thus, regarding either measure, a decreasing value reflects a desegregative effect. Dr. Armor preferred the exposure index because it is less a measure of racial balance. Id. at 2583.

Dr. Armor calculated the dissimilarity index and relative exposure index for Topeka over time. Regarding the elementary schools, he testified that from 1966 to 1985 both indices dropped significantly indicating that: 1) there has been a "very substantial decline in the level of segregation" over 20 years, and 2) this improvement came at a time when the percentage of minority students increased from less than 20% (16.5%) to its present level of 26%, and 3) the 1985 value of approximately 15 in the exposure index suggests that a "very substantial degree of interracial contact" is occurring at the elementary level. Id. at 2587. Regarding the middle schools, he reported similar trends, though not quite as dramatic: declines in both indices, improvement during a time when the percentage of minority students rose from 15% to its present level of 26%, and "a very substantial degree of desegregation." Id. at 2587-88. As to the high schools, neither index shows appreciable change over time, although both indices have dropped slightly. Id. at 2588. The reason for this is that both indices started out quite low and it is difficult to get below 10 on the exposure index (1985 value for high schools was 7.3). Id. at 2588.

Dr. Armor reasoned that had Topeka done nothing concerning integration, it is very likely that the indices would have remained at their significantly higher 1966 levels. Id. at 2589-90. Contrary to this court's appellate factfinding, see Court's Opinion at 880 (attendance boundaries) 881-82 (school closings), Dr. Armor testified that "the reason why desegregation has improved to this extent was a series of school closings and boundary changes" affecting predominantly minority or white schools. Rec. vol. XIII at 2590. Dr. Armor also testified that the system's transfer policies over the years had no significant effects, either segregative or desegregative. Id. at 2598. The policies were neutral. Id. This evidence favorable to the defendants was corroborated by defendants' other expert, who considered what would have happened had school boundaries been held constant as of 1963.

B.

The plaintiffs' experts disputed Dr. Armor's work, just as he disputed theirs. For example, plaintiffs' expert Dr. Crain testified concerning the limitations of these indices and that other cities had lower dissimilarity and exposure indices and that more could be done as evidenced by Plans N and X that were not adopted. Rec. vol. IX at 1270-73, 1282-84, 1287; but see F. Welch A. Light, New Evidence on School Desegregation-U.S. Comm'n on Civil Rights Clearinghouse Pub. No. 92, 38 (1987) ("The fact that the dissimilarity index is useful in depicting changes within a district does not imply that it is useful in comparing districts."). Dr. Foster indicated that he used the dissimilarity index "to determine the progress made across the State of Florida in school desegregation from year to year when data are available," and explained that it was not used "on a school-to-school basis." Rec. vol. IV at 556, vol. V at 686. In short, the experts disagreed concerning the proper approach and whether there was a current condition of segregation regarding student assignment.

For this reason, we need pay particular attention to the Supreme Court's conclusion in Anderson that a factual finding cannot be clearly erroneous when "there are two permissible views of the evidence," when credibility is at issue, and even when it is not. 470 U.S. at 574, 105 S.Ct. at 1511. We have noted:

The Anderson Court justified this conclusion by asserting that "[t]he rationale for deference to the original finder of fact is not limited to the superiority of the trial judge's position to make determinations of credibility. The trial judge's major rule is the determination of fact, and with experience in fulfilling that role comes expertise." Anderson, 470 U.S. at 574, 105 S.Ct. at 1512. Although even the trial judge may not always be confident that he "knows" what happened, he is in the best position to "determine whether the plaintiff has succeeded in presenting an account of the facts that is more likely to be true than not." Id. at 580, 105 S.Ct. at 1515.

Pitre v. Western Elec. Co., Inc., 843 F.2d 1262, 1266 n. 1 (10th Cir. 1988). If ever there were a case for application of these principles, this is the one.

For this case not only involves documentary evidence, it also involves credibility determinations. At one point, in consistently exercising his discretion to admit evidence that might bear upon the truth of the issues, the trial judge remarked: "Let me say, it is the duty of a court to listen to the experts on both sides, and decide which set of experts to believe. I think it is necessary that I hear everything I can hear on this subject." Rec. vol. IX at 1222. It does not surprise me that the trial judge exercised his prerogative to believe the defendants' experts.

Even on this cold record, it appears that the defendants' experts may have had a patina of believability that comes with acknowledging ambiguity (the evidence is rarely all one way), yet still supporting a position. See M. Chesler, J. Sanders D. Kalmuss, Social Science in Court — Mobilizing Experts in the School Desegregation Cases 90-106 (1988). For example, defendants' expert Dr. Armor testified that the open enrollment plan which operated for two years and then was discontinued in 1980 had a slightly segregative effect. Rec. vol. XIII at 2624; Brown, 671 F. Supp. at 1298. Likewise, defendant's expert Dr. Clark testified that a boundary change for Parkdale between 1963-64 and 1966-67, which resulted in a greater concentration of minority students "could be discussed in terms of having partly segregative effects, partly demographic effects." Rec. vol. XI at 2326; ex. vol. I at 43. The court notes this fact in its opinion, Court's Opinion at 875, while completely ignoring the conclusion of defendants' expert which supports the district court's findings.

The plaintiffs put on much testimony concerning factual matters which the Supreme Court decided as a matter of law years ago, namely the educational and psychological harms of segregation. Brown I, 347 U.S. at 494-95, 74 S.Ct. at 691-92. Plaintiffs' expert Dr. Crain testified that segregated schools are responsible for the following problems as they affect blacks who might attend such schools: 1) depressed performance on standard achievement tests, which cannot be remedied by compensatory education, 2) higher failure rates on minimum competency testing, 3) a propensity to attend black colleges, 4) a propensity to quit college, if a white college is selected, 5) lack of clear career aspirations, 6) employment discrimination, 7) discomfort with white supervisors, 8) more frequent job changes, 9) higher juvenile delinquency, 10) higher teenage pregnancy, 11) greater unhappiness at age 40, 12) a desire to live in an unintegrated area, 13) infrequent home ownership, and 14) greater discomfort around whites. Rec. vol. IX at 1231, 1237, 1243, 1251-52, 1254, 1255, 1257, 1259, 1260, 1261, 1262. Viewing the transcript, I do not think that this remarkable list, which Dr. Crain claimed was applicable to Topeka based on his brief visit, id. at 1264-65, enhanced his credibility. Nor did his definition of a segregated system seem consistent with that term as commonly understood in desegregation cases. Dr. Crain testified that, in his view, a segregated system was "when you have that perception of unequal racial composition," because "the harmful effects are going to start occurring because the whole process of desegregation is going to start hitting." Rec. vol. IX at 1289. Finally, Dr. Crain was quite willing to criticize the work of defendants' expert Dr. Poggio, by equating a cognitive abilities test with the Kansas Minimum Competency test, yet Dr. Crain clearly was unaware of the sequence of the tests and the characteristics of each — regarding the Kansas Minimum Competency test, he did not know its objectives, what it purports to measure, how the test was designed or whether the passing scores change from year to year. Rec. supp. vol. X at 2862-71; supp. vol. IX at 2922-31 (Dr. Poggio's refutation of Dr. Crain's criticisms). Without question, the trial court credited Dr. Poggio's analysis over that of Dr. Crain. Brown, 671 F. Supp. at 1307-08.

Plaintiffs' expert Mr. Lamson was similarly emphatic. He testified that his task was "to identify all of those instances where administrative action would have tended to affect the racial composition of the schools." Rec. vol. III at 363, see also vol. II at 105. After looking at hundreds of physical and administrative decisions (construction, portable building use, attendance boundaries, optional attendance zones) that affect student assignment for thirty years, Mr. Lamson's conclusion is notable for its simplicity and its partiality:

I can confidently state that I was unable to find one instance where any of those physical and administrative educational tools were used as a part of a purposeful and ongoing program of affirmative pupil desegregation at either the elementary, junior high or middle, or senior high school level.

Quite to the contrary . . . all of those administrative tools were used in an ongoing manner to actively separate the Black and white elementary, junior high and middle, and senior high students of the Topeka school system wherever and whenever feasible.

Rec. supp. vol. XI at 240-41. Mr. Lamson's testimony proves a bit much. Analyzing the motivations and actions of multi-membered public bodies over three decades is difficult at best, but concluding that the Topeka system underwent no desegregation 25 years after Brown I blinks reality. Common sense tells us that even a broken clock is right twice a day. Yet too much of Mr. Lamson's methodology has found its way into this court's opinion, given the trial court's rejection of the proposed findings supporting it. This is especially true concerning the court's comments about optional attendance zones, attendance boundaries, and school openings and closings, Court's Opinion at 877-82, because plaintiffs' other lead expert, Dr. Foster, did not do an independent analysis of these factors. Rec. vol. V at 693; vol. VI at 739.

C.

The court also has found a current condition of segregation pertaining to faculty and staff assignment. Court's Opinion at 870-72. The court's reasoning here is questionable because, after rejecting the plaintiffs' standard as "very narrow," id. at 870 n. 56, the court then finds liability when, even under the plaintiffs "very narrow" standard, all that would have been necessary to eliminate racial identifiability regarding certificated faculty would have been to exchange 13 pairs of faculty members in 1981-82. Rec. vol. V at 663-66. That is only 26 certificated faculty out of 911 then employed, or 2.85%. Id.; rec. vol. VI at 773-74. Faculty members are not fungible; whether this could be accomplished depends upon being able to send 13 minority faculty members into different schools, and replacing them with 13 white faculty members, while making sure that all have the right teaching certificates. Rec. vol. V at 664-65.

We begin with the proposition that "principals, teachers, teacher-aides and other staff who work directly with children at a school" may not be assigned so that the racial composition of the staff will indicate that a school is intended for white or minority students. Singleton v. Jackson Mun. Separate School Dist., 419 F.2d 1211, 1217-18 (5th Cir.) (remedy phase), rev'd on other grounds, 396 U.S. 290, 90 S.Ct. 608, 24 L.Ed.2d 530 (1970). Even if the racial composition of a school could be identified by assignment of faculty, that factor, in and of itself, does not mean that the school system is dual with respect to student assignment. Dayton II, 443 U.S. at 536 n. 9, 99 S.Ct. at 2978 n. 9. While a school should not be racially identifiable by its faculty, United States v. Montgomery County Bd. of Educ., 395 U.S. 225, 236, 89 S.Ct. 1670, 1676, 23 L.Ed.2d 263 (1969), the racial composition of the faculty need not be equivalent to that of the student population, Fort Bend Indep. School Dist. v. City of Stafford, 651 F.2d 1133, 1138 (5th Cir. 1981). It is possible for a school district to be unitary with respect to student assignment, but not with respect to faculty/staff assignment. Pasadena, 427 U.S. at 436, 96 S.Ct. at 2704-05; Morgan, 831 F.2d at 318-19.

Topeka is not heavily populated, has not been growing and does not have a large minority population as a percentage of the total population. This is reflected in the data on minority employment, which has remained fairly constant over the last five years. Dr. Foster indicated that faculty composition should approximate the minority percentage enrollment in a school system, something which has not occurred in Topeka. Rec. vol. V at 656. Dr. Foster acknowledged, however, that one would have to look at the relevant labor pool, an analysis that was not performed here. Id. at 657-58; see also Hazelwood School Dist. v. United States, 433 U.S. 299, 308, 97 S.Ct. 2736, 2741-42, 53 L.Ed.2d 768 (1977). But, Dr. Foster did testify that the demand far exceeds the supply of minority faculty members and that he found no fault with the district's recruiting policies. Rec. vol. VI at 757-59.

Dr. Foster analyzed faculty staff assignment, "not as a raw figure of percentage points from the mean," e.g. ± 15% or ± 20% from the system-wide minority average, but instead "as a percentage of the mean majority," say ± 15% of the mean minority percentage. Rec. vol. V at 646. For example, in 1985-86, the percentage of minority faculty staff at the elementary schools was 11.23%. Rec. ex. vol. IV at 261, 15% of 11.23% is 1.68%; thus, a school could vary by ± 1.68% from the minority percentage of 11.23%. Id.; rec. vol. V at 650. This means that a school would be racially identifiable if it had more than 12.9% (11.23% plus 1.68%) or less than 9.6% (11.23% minus 1.68%) minority faculty/staff. Id. Applying this concept to the secondary schools which had a minority faculty staff percentage of 12.65%, a school would be racially identifiable if it had more than 14.6% (12.65% plus 1.90%) or less than 10.8% (12.65% minus 1.90%) minority faculty staff. Rec. ex. vol. IV at 262. Applying this very tight standard is based on the theory that the school district has greater control over faculty staff assignments and that, insofar as possible, we should "get every school just like every other." Rec. vol. V at 646, 650-51. Under this method, 22 out of 26 elementary schools are racially identifiable and 8 out of 9 secondary schools are racially identifiable. Rec. ex. vol. IV at 261-62.

This court has rejected Dr. Foster's standard. The court has also rejected the district court's approach which used a ± 10% variation around the mean because in one year, 1975, the standard would not have classified elementary schools as racially identifiable even if they had no minority faculty and staff. Court's Opinion at 879; see Brown, 671 F. Supp. at 1305 (noting that in 1985 only six schools had minority staff percentages beyond ± 10% of the district-wide average and that all schools have been within ± 10% range in last 10 years). In fairness to the district court, the problem identified by this court with the ± 10% standard would have occurred only for the first of the ten years; in all other years, the district-wide elementary minority staff percentage exceeded 10%. Moreover, it should be noted that in Dowell, Dr. Foster testified that he had "`no problem'" with a ± 10% standard, however, the system-wide average of black faculty in the school district at issue (Oklahoma City) was 36.9%. Dowell v. Board of Educ., 890 F.2d at 1537 (10th Cir. 1989) (Baldock, J., dissenting).

In finding its "distinct pattern of correlation," Court's Opinion at 879, between minority enrollment and assignment of minority staff, the court initially used an ad hoc approach which was difficult to apply objectively. In its revised opinion, the court has indicated that a school is racially identifiable as to faculty/staff assignment if it is racially identifiable minority as to student assignment, using a ± 15% standard, and 17.3% or more of the faculty/staff at the school is minority. Court's Opinion at 872-73 n. 59. Likewise, a racially identifiable white school (using a ± 15% standard) is racially identifiable as to student assignment if 5.6% or less of the faculty is minority. Id. Using this method, there are 14 out of 19 schools racially identifiable by faculty/staff assignment in 1985-86. This means that not all of the schools identified as racially identifiable by the court as to student assignment are racially identifiable as to faculty/staff assignment.

Although the court plainly has adopted a ± 15% standard for analyzing racial identifiability as to student assignment, it is interesting to compare the schools that the court finds racially identifiable as to faculty/staff assignment with the schools identified (as to student assignment) using Dr. Armor's "most generous" method. See Court's Opinion at 870, 872-73, 879; see also supra p. 922 (listing schools identified by Dr. Armor for further study). There are only four schools in common to both lists (Gage, McClure, French and Topeka West). All of these schools have increased minority student representation over the years and in 1986-87, had increased minority faculty and administrative personnel.

The court notes that the data concerning faculty and staff was not as complete as it might be, and the problem seems to be that a breakdown of minority teaching and managerial personnel for each school was not available. Rec. vol. V at 660. Dr. Foster was able to prepare a breakdown of the district by employment category which tends to refute some of the court's statements, however. Dr. Foster broke the data down into two areas overall. I suspect that the first group must be certificated.

1985-86

In that same vein, the court remarks that the "practice of disproportionate assignments also reinforces the irrational notion that minority teachers are inferior and not fit to teach white children." Court's Opinion at 872. There simply is no evidence of that in the current Topeka system based on this record. No one disputes that it is important to have white and minority students exposed to positive role models of different races, see rec. vol. VI at 761, but the court's remark is simply inflammatory. Indeed, in a colloquy between Dr. Foster and counsel for the school board, it was revealed that as late as 1971, representatives of the black community appeared before the board seeking "the assignment of black teachers as black role models in predominantly black schools." Rec. vol. VI at 762. Dr. Foster testified that "blacks consistently across the country have demanded that their schools not be staffed predominantly by white teachers, but that black principals and black teachers also appear before the black children." Id. Of course, the bottom line here is the constitutional imperative that staffing decisions be neutral and "that you don't want an absence of either race or ethnic group in any school if you can avoid it." Id.

The court tells us that the distinction between faculty and staff is important, but never comes to grips with what that distinction is. Court's Opinion at 871-72. On the basis of absolute numbers, it appears that 167 minority employees are involved with the educational function (the first group above plus teacher aides) and 67 minority employees are involved with the service function. The strong implication of this court — that minorities are being hired merely for janitorial and service jobs — is not supported. Moreover, such an implication is inconsistent with the court's observation that the current superintendent is black, as are several principals and managerial personnel. Court's Opinion at 879.

The district court recognized that there was a "tendency to have more minority staff in schools with a greater than average minority student population," yet after a careful review of all the evidence, decided that there was not a current condition of segregation with respect to staff assignment. Brown, 671 F. Supp. at 1305. In deciding that faculty/staff assignment remains segregated, Court's Opinion at 872, this court has relied in part upon the extremes of the data to conclude there is an unlawful pattern. For example, in 1985-86, every elementary school had some minority staff, with the exception of McClure. Only one elementary school had more than 7 minority faculty/staff, Avondale East at 10.45 faculty/staff members. Yet, that is the one the court uses to illustrate its apparent conclusion that there is a segregated system of faculty staff assignment at the elementary level. Court's Opinion at 872; see also id. at 879. When one looks at the elementary data, it is apparent that a perfectly uniform distribution is unlikely given the lack of elementary minority personnel — there were 42 elementary classroom teachers and 74.50 minority faculty/staff members to be assigned to 26 elementary schools. Rec. ex. vol. IV at 261, 268.

Likewise, the court compares Robinson Middle School with 12.20 minority staff members out of 49.35, not with other middle schools, but with Topeka West High School which had 3 minority out of 120. Court's Opinion at 872. Note that in 1984-85 Topeka West had almost three times the minority faculty/staff (8.55 out of 114.30) than it did in 1985-86.

Given the court's interest in complete data which breaks out the professional staff that actually works in the particular schools, it is most surprising that the court's opinion does not cite or discuss the minority staffing report for 1986-87 which finally includes administrative and certificated staff assigned to the schools. Rec. ex. vol. II at 160-61. Although admitted over plaintiffs' objection, rec. vol. XVI at 2523-2525, this more current and more relevant report presents a far less extreme picture of minority assignment for the system. Every school in the system has minority personnel. With a 10.78% minority average for elementary school personnel (55 out of 510), elementary schools range from 4.35% minority staffing (Quincy) to 18.18% (Avondale East and Highland Park North). More significant, because of the small numbers involved, is that every elementary school has from one to four minority personnel. Rec. vol. II at 161.

With a 13.22% minority average for middle school personnel (23 out of 174), schools range from 6.67% (Jardine) to 16.67% (Robinson). Every school has from two to seven minority personnel. Finally, at the high school level there is a minority average of 9.77% for high school personnel, ranging from 5.88% (Topeka West) to 11.84% (Highland Park High). The range is from five to eleven minority faculty members.

D.

The district court recognized that the "pattern identified by the plaintiffs is not monolithic." Brown, 671 F. Supp. at 1305. And it is not. The numerical evidence on this point cuts both ways. But this court's finding that this is "a system that has kept white faculty at primarily white schools and minority faculty at predominantly minority schools," Court's Opinion at 879, appears radical given these numbers. Moreover, I agree with the district court that "as important as the racial percentages of the faculty and staff is the fair assignment of qualified and dedicated faculty throughout the district." Id.; see also Milliken v. Bradley (Milliken II), 433 U.S. 267, 280 n. 15, 97 S.Ct. 2749, 2757 n. 15 (one must not lose sight of educational function of schools). This is a school district that is applying the research concerning effective schools. See U.S. Dep't of Education, What Works-Research About Teaching and Learning 45 (1986) ("The most important characteristics of effective schools are strong instructional leadership, a safe and orderly climate, school-wide emphasis on basic skills, high teacher expectations for student achievement, and continuous assessment of pupil progress."); rec. supp. vol. V at 1584-1608, 1635-36 (district has uniform curriculum guides), 1641-42, 1651-53, 1749; supp vol. VII at 2033; vol. XII at 2521-22. It is a system in which schools throughout the district have received outside recognition. Rec. supp. vol. V at 1617-18 (Highland Park H.S., 1984), 1619-20 (Topeka West H.S., 1985), 1620-21 (Eisenhower Middle School, 1985), 1622 (School District, 1986), 1623, (Topeka West H.S. — National School of Excellence, 1984; Robinson Middle School — National School of Excellence, 1985), 1624 (Bishop Elementary School — State Program of Excellence, 1986), 1732 (Head Start, 1974). Finally, the evidence clearly established that the Topeka schools are committed to an awareness of cultural diversity and equal educational opportunity for all. Rec. supp. vol. V at 1759, 1632, 1653-54, 1657-58, 1686, 1717, 1760-61, 1772-73; vol. XII at 2444-45. Ironically and commendably, this is a district that includes the study of Brown v. Board of Education in its curriculum. Rec. supp. vol. V at 1722.

VII.

At trial, plaintiffs contended that eight schools were "racially identifiable nonwhite or vestige schools." Brown, 671 F. Supp. at 1297. These included Eisenhower Middle School and seven elementary schools: Avondale East, Belvoir, Highland Park North, Hudson, Lafayette, Lowman Hill and Quinton Heights. The court has analyzed the latter six of these schools for "what the district did and did not do," Court's Opinion at 882, totally ignoring the district court's findings and the de facto reasons for the racial composition of these schools. Court's Opinion at 882-85. Part of the court's analysis of these schools is contained in an earlier part of its opinion. See Court's Opinion at 880 (discussing attendance boundaries of Lafayette, Quinton Heights and Lowman Hill). Appendix A contains the 1969-70 elementary school attendance boundaries; appendix B contains those for 1985-86.

Ironically, this court's analysis of these racially identifiable schools is placed in a section entitled "The Causal Link Between De jure Segregation and Current Condition of Segregation." Court's Opinion at 874. The court steadfastly refuses to consider an alternative interpretation of the evidence which would show the absence of a link between the present lack of perfect racial balance and the prior de jure segregation. The court's position is that any consideration of demographic factors which resulted in a lack of racial balance would be to absolve the school board of its affirmative duty. Id. at 876-77, 882. Not so. Courts routinely have considered the presence or absence of demographic factors over which the school board had no control in deciding whether school systems were unitary. See, e.g., Morgan, 831 F.2d at 320 (system unitary; court considered racial concentration in court-drawn school districts); Lawrence, 799 F.2d at 1043 ("In Lawrence County, unlike the situation in Pasadena Board of Education, the School Board's failure to achieve a desegregated school system has not resulted from population migration into, or out of, or within this small rural county."); Ross, 699 F.2d at 226 (system unitary; current housing patterns after desegregation order resulted in several one-race schools). The rationale for this is best stated by this court in a passage concerning one-race schools:

Although virtual one race schools "require close scrutiny," they are not always unconstitutional.[18]

[18] Given modern urban demography and geography, one-race schools may well have evolved for reasons beyond school board control. See Calhoun v. Cook, 522 F.2d 717, 719 (5th Cir. 1975). Thus, where a school system consists of de facto one-race schools, rather than de jure, the system is not unconstitutional.

Court's Opinion at 860, 860 n. 18. The above is equally applicable to schools which depart from the system-wide minority average — "they are not always unconstitutional" for they "may well have evolved for reasons beyond school board control." Id.

Concerning these eight schools that the plaintiffs contend are "racially identifiable," the district court found that "the schools are attended by more white students than black students (1,248 white students, 1,079 black students). Indeed, white students make up close to half (49.3%) of all the students that attend these allegedly non-white schools." Brown, 671 F. Supp. at 1298. As the following reveals, there is no need for the appellate fact finding that is going on here, the district court's finding that these schools do not reflect a current condition of unlawful segregation is not clearly erroneous; indeed, it is amply supported by the record and the weight of the evidence. Moreover, the district court's factfinding is free of several errors which permeate the appellate factfinding.

A.

Belvoir elementary was annexed in approximately 1959, thus plaintiffs cannot claim it was a vestige of the former de jure system. Rec. supp. vol. XI at 148. This court attaches significance to the fact that Belvoir's student population has not been less than 50% minority. Court's Opinion at 882. Yet, the district court carefully explained why this does not create liability:

Belvoir Elementary School is centered in an area with a high minority population. It is surrounded by other schools with higher than average minority populations. It is not the black school in a racially mixed area. It is a school with a higher than average black student population in a residential area with a higher than average black population.

Brown, 671 F. Supp. at 1300 (emphasis in original). The district court made further findings, which certainly are corroborated by the data on racial composition. The district court's findings also are supported by the defendants' expert Dr. Clark, who analyzed Belvoir's hypothetical racial composition, assuming its boundaries remained constant. Rec. ex. vol. I at 28, 34-35.

The district court found that two federally subsidized housing projects constructed in 1963 (Pine Ridge; 210 family units) and in 1965 (Trail Ridge; 148 family units) caused the minority elementary student population to increase to over 50%. Brown, 671 F. Supp. at 1302; rec. ex. vol. IV at 215. Plaintiffs' expert Mr. Lamson did not take these housing projects into account. Rec. vol. IV at 463. The district court also found that closing Rice Elementary in 1981 and reassigning some of those students to Belvoir improved the racial balance at Belvoir. Brown, 671 F. Supp. at 1302. This court characterizes the closing as having "a modest desegregative effect at Belvoir," Court's Opinion at 882, but the data reveal that the minority percentage dropped by approximately 13%. See supra note 43. The Belvoir attendance district now is largely minority and the changed racial mix in the school has been caused by actions other than those of the school board. See Higgins, 508 F.2d at 791 ("the law imposes no affirmative duty upon school officials to correct the effects of segregation resulting from factors over which they have no control").

B.

This court's complaint about Hudson Elementary is that "[w]hile the passage of time has dissipated Hudson's identity as a white school, this change is due solely to the spread of minorities into its attendance area, not to any action by the school district." Court's Opinion at 882. It seems anomalous to find fault with the school board concerning residential movement which altered the original racial composition at Hudson. But it is consistent with this court's view that residential patterns which detract from racial balance are always the responsibility of the school board, while residential migration which improves racial balance should not inure to the benefit of the school board insofar as liability is concerned. See Court's Opinion at 863-64 ("A showing that the school district has not promoted segregation and has allowed desegregation to take place where natural forces worked to that end is insufficient."). Following this logic, residential movement could lead to absolute racial balance and still support a finding of school board liability, especially given the court's view that a school which approaches the system-wide minority average may still be racially identifiable.

Apparently, the defendants are constitutionally liable for Hudson because it now exceeds the system-wide minority average by 20%, with a 46.55% minority student population. See Court's Opinion at 882. The plaintiffs view Hudson as racially identifiable black, notwithstanding that 53.45% of the students were white in 1985-86. Appellants' Brief at 38; Brown, 671 F. Supp. at 1301. Hudson is perhaps the archetype for illustrating the court's refusal to consider the effect of demographic factors which result in changing student racial composition at various schools. The district court gave clear reasons explaining the racial composition at Hudson, reasons which do not encompass liability for these defendants. Hudson was constructed in 1963 and has steadily increased in minority enrollment. The boundaries of Hudson have not changed since 1964. Rec. ex. vol. IV at 13-29. The district court found that "the construction of subsidized housing projects in the Hudson attendance area" in 1965 (Colonial Townhouses; 137 family units), in 1967 (Highland Park Apartments; 200 family units) and in 1970 (Deer Creek; 92 family units), together "with the general increase of minority residents on the east side," was responsible for the increase in minority enrollment at Hudson. Brown, 671 F. Supp. at 1302; rec. ex. vol. IV at 215. Plaintiff's expert Mr. Lamson did not take into account the effect of any of these demographic changes. Rec. vol. IV at 464-65.

Plaintiffs' position concerning Hudson and Avondale is legally untenable. After conceding that the increased minority enrollment at Hudson and Avondale East is "purely [the] result of residential change," plaintiffs continue:

In a de jure school system, the school board has the duty to react to residential change to ensure that no school becomes disproportionately Black. Swann, 402 U.S. at 14 [ 91 S.Ct. at 1275]. The failure of the Topeka School Board to prevent Hudson and Avondale East from becoming disproportionately Black is a constitutional violation.

Appellants' Brief at 38. Plaintiffs' position is not supported by the law or the facts. After two desegregation plans and over thirty years after Brown I and Brown II, it cannot be said that Topeka operates a de jure system. Apparently, plaintiffs' conception of the duty imposed by the Supreme Court is never-ending. The Court, however, has declared to the contrary. In Swann, the Court recognized that once the constitutional violation is eliminated, school authorities are not "constitutionally required to make year-by-year adjustments of the racial composition of student bodies" even if the community experiences demographic changes. Swann, 402 U.S. at 31-32, 91 S.Ct. at 1283-84. Moreover, there is no requirement that every school in the district reflect the system-wide average minority enrollment, either before or after a finding of unitariness. See Pasadena, 427 U.S. at 434, 96 S.Ct. at 2703.

C.

Like Belvoir, Highland Park North Elementary was not part of the Topeka system in 1954. Instead it was annexed in 1959, rec. vol. IV at 394, and is estimated to have had close to the system-wide average of minority students then. There has been a steady increase in minority students attending due to residential patterns. Brown, 671 F. Supp. at 1302. This court's complaint is that when Parkdale closed in 1978, the "reassignment [of Parkdale students] increased Highland Park North's minority percentage at a time when it was already high." Court's Opinion at 882. As the district court points out, however, the school board was able to close Parkdale, a predominantly black school (85.6% minority) while reassigning its students to a nearby school, Highland Park North. Brown, 671 F. Supp. at 1302. Moreover, as footnote 45 indicates, the increased minority percentage for 1978-79 was caused by two factors: an increase of 43 (170-127) minority students and a decrease of 21 (142-121) white students. Considering the court's fleeting approach to some of these schools, see, e.g., Court's Opinion at 882, it is unclear how the court's observation trumps the district court's reasoned analysis rejecting constitutional liability.

D.

Lafayette Elementary began as an all-white de jure school. Brown, 671 F. Supp. at 1302. It has increased in minority enrollment and has remained a majority-minority school, although the composition of the minority groups has varied over the years. Id. (noting that minority composition was 56-59% over the last five years, with a varied composition). This court notes that federally subsidized housing projects were built in the area in the early 1960's. Court's Opinion at 883; rec. ex. vol. IV at 215 (1961-Eastboro Apts.; 28 family units; 1962-Eastboro Apts.; 76 family units). More units were built in 1970. Rec. ex. vol. IV at 215 (Ripley Park Apts., 102 family units).

This court first looks to 1966 and reports the racial compositions of Lafayette's neighboring schools (omitting Highland Park North). Court's Opinion at 883. I have done the same in chart form comparing the minority percentages with the surrounding schools, in 1966 and now.School Lafayette Minority % Minority %

The border between Lafayette and State Street, the school to Lafayette's north, has functioned as the boundary between the mostly white area served by State Street and the school to its north [Lundgren] . . . and the schools to its south and southwest [Highland Park North and Belvoir]. Lafayette has long had twice the minority population of State Street; thus, since 1966, Lafayette's minority population has ranged from 52% to 69% while State Street's has ranged from 27% to 34%.

Id. (footnote omitted). This conclusion is remarkable not only because it is appellate factfinding concerning intent including non-compliance with the affirmative duty on "objective" facts that are far from obvious, but also because it ignores that the surrounding population of the Lundgren and State Street schools (as of 1980) had far less black concentration than did the surrounding population of Lafayette, Highland Park North and Belvoir. See rec. ex. vol. I at 49. Moreover, it conflicts with the district court's finding on this point attributing the increased minority percentage to demographic factors, Brown, 671 F. Supp. at 1303, rather than a "segregative effect" of boundary changes by defendants. The district court's finding is supported by common sense and the defendants' expert, Dr. Clark. See rec. ex. vol. I at 35, 43-45.

Once again, the district court's findings make more sense than ours. Looking at the chart, see supra p. 940, it is apparent that Topeka closed two schools with greater and lesser minority concentrations than Lafayette. The district court noted that the closing of Parkdale resulted in the Lafayette minority percentage increasing by four percent (from 62% to 66%), Brown, 671 F. Supp. at 1302; however, the next year saw the percentage return to its prior level (from 66% to 62%), and decrease thereafter. See supra note 42. The district court then thought that the closing of Rice, which resulted in reassignment of some students to Lafayette, may have contributed to the minority percentage decreasing in future years. Brown, 671 F. Supp. at 1302. Dr. Clark, defendants' demographic expert, using residential population data which had a high correlation (.8 to .9 out of 1.0) with school attendance data, viewed the increased minority percentage at Lafayette as a result of demographic changes. He concluded that the closure of Parkdale, with respect to Lafayette, was overall desegregative. Rec. vol. XI at 2338, 2373, 2388-89; ex. vol. I at 35, 43, 45.

This court's ultimate conclusion, that "[u]nder its neighborhood school plan, the school district simply maintained Lafayette as a school with disproportionate minority percentages," Court's Opinion at 883; see also id. at 880, ignores the demographic factors involved and implies that the school district had an obligation to racially balance Lafayette because it exceeded the system-wide minority percentage.

E.

Like Lafayette, Quinton Heights Elementary began as an all-white de jure school. Brown, 671 F. Supp. at 1303. The district court noted the steady increase in minority student population, and that the minority concentration has remained approximately 50% over the last several years. Id. This court's approach to analyzing the school has been to adopt the conclusion of plaintiffs' expert contained in his report and testimony, while simplifying the analysis and not discussing the effect of allegedly segregative optional attendance zones, as did Mr. Lamson. See Court's Opinion at 883; rec. vol. XI at 139-48, 173-78 (Lamson Report on Quinton Heights, School Closings), vol. III at 248-59 (Lamson testimony concerning Quinton Heights) (cited in Court's Opinion at 883 n. 88), 284, 287, 290-93 (closure of Monroe and Polk and affect on Quinton Heights); see also Court's Opinion at 874-75 (summarizing 1955-65 in Topeka system — "Plaintiffs introduced evidence tending to show that the school district's use of portable classrooms and optional attendance zones served to maintain segregation by concentrating students of one race at certain schools."). Of course, to adopt the full analysis with the discussion of optional attendance zones would conflict with the district court's finding that "any segregative effect of the zones was slight and does not remain today." Brown, 671 F. Supp. at 1298. The district court's finding, which necessarily includes the optional attendance zones discussed by plaintiffs' expert, is supported by the testimony and report of defendants' expert Dr. Clark. Rec. vol. XI at 2304-05, 2309-10, 2318-19, 2321-22, 2327-28. Dr. Clark analyzed the effect of boundary changes in Topeka, including optional attendance zones. See supra note 30; see also rec. ex. vol. I at 19. But concerning school closings, the court has been convinced by the testimony of plaintiffs' expert.

Again, we track this court's odyssey. The court begins by considering two closings: Pierce and Van Buren. In 1957, Pierce was annexed to the district; in 1958 the adjacent schools to the west with which it shared a common boundary were Quinton Heights and Monroe. Rec. supp. vol. XI at 140; rec. ex. vol. IV at 12. In 1958, the enrollment began to decline at Quinton Heights at an average rate of 21 students per year. Rec. supp. vol. XI at 142. In 1959, the 79 remaining black students at Pierce were assigned to three other elementary schools: Quinton Heights, Highland Park North, and Highland Park Central. Id. at 142. Plaintiffs' expert admitted that what effect this had on the racial composition of Quinton Heights is not known. Id. at 143. Mr. Lamson also reported that during 1956-66 Quinton Heights lost white students while gaining black students. Id. at 142.

In 1962, the attendance boundary of Quinton Heights was expanded northward into what had been areas served by Van Buren and Central Park. Id. at 144; rec. vol. III at 251. This resulted in a gain of 45 students for Quinton Heights after it had lost an average of 21 students for the previous 4 years. Rec. supp. vol. XI at 144. Van Buren, however, still had an attendance area and continued to operate. In 1964, Van Buren was closed; most students were sent to Monroe; some were sent to Quinton Heights. Rec. vol. III at 255; supp. vol. XI at 145-46.

This court continues with the observation that by 1966 the minority percentage of Quinton Heights was 36.4% and certain schools surrounding Quinton Heights had greater and lesser concentrations of minority students. Court's Opinion at 883. And so we continue. The following chart provides more detail.School Heights Minority % Minority %

The average minority percentage in 1966 for elementary schools was 16.49%. Rec. ex. vol. IV at 56. Keeping this in mind, it appears that other schools, besides neighboring Monroe, had minority percentages which very nearly met, or exceeded, the system-wide elementary average. As the above chart points out, Highland Park Central, Avondale East, Central Park and Highland Park North simply are not one-race white or virtually one-race white schools. Moreover, I do not see the significance of the court's statement that Quinton Heights had the seventh largest minority population out of 35 schools. How does this fact create liability? Thus, the district court did not err when it rejected the attempt to "characterize the reassignments in the 1960s as part of a process of focusing Quinton Heights as a school for black students." Court's Opinion at 883.

The court next demonstrates its ability to find its own facts by describing the population density maps in relation to a neighboring school of Quinton Heights: "Indeed, on the maps available to us, it is quite noticeable that Polk's boundaries in 1970 correspond closely to a patch of white population surrounded by minority population assigned to other schools, including Quinton Heights." Id. The court's subjective observation demonstrates the limitations of its approach. The map cited by the court, rec. ex. vol. I at 67, tells us the percentage of black population by block. Notwithstanding the intended breadth of the court's analysis, the map does not tell us about other minorities, e.g., hispanic. Nor does it tell us much about population density or anything about student density. See rec. vol. IV at 485. Yet, the court's clear implication from looking at the map is that the boundary around Polk as of 1970 was drawn in such a way as to exclude minorities. The following chart, comparing the minority percentages of Polk with its neighboring schools, including Quinton Heights, as of 1970, the date of the map, undercuts the court's finding.

School Polk Minority %

Continuing with the analysis of Quinton Heights, the court next complains about two adjacent school closures in the 1970's because they resulted in extending the northern boundary of Quinton Heights into northern areas with high minority concentrations. Court's Opinion at 883. Unfortunately, the court does not share with us which closings are involved, id., but I suspect they are Monroe which closed in 1975, and Polk which closed in 1979. When Monroe closed, its attendance area was divided between Polk, Highland Park North and Quinton Heights. Rec. supp. vol. XI at 173. As I previously noted, the effect on Quinton Heights was negligible; although the minority percentage increased from 37 to 39.09%, seven fewer minority students attended in 1975. See supra p. 915 (noting that 2.09% increase largely due to declining white enrollment). Indeed, not until the fall of 1979 did the number of minority students at Quinton Heights exceed what existed immediately prior to the closing of Monroe. Thus, expanding Quinton Heights' attendance area northward when Monroe closed could hardly support a firm and definite conviction that Quinton Heights' boundaries were being manipulated so as to racially identify the school as intended for minority students.

Apparently the closing of Polk resulted in the reassignment of its students to Quinton Heights, Sumner, and Lowman Hill. Rec. vol. III at 291. The minority percentage at Quinton Heights did increase from 38.95% (in 1978-1979) to 50.23% (in 1979-80), with the addition of 44 minority students and 5 white students. This is the only action here which arguably is indicative of inattention to the affirmative duty. The district court, however, rejected that inference. Brown, 671 F. Supp. at 1303. Based on the foregoing, I cannot agree with the statement that: "It is thus clear that Quinton Heights' current boundaries are the result of continued manipulation by the school district." Court's Opinion at 883. The court's analysis is inconsistent with its contention that the district court should not have focused on intent, and that there was indeed none. Court's Opinion at 868.

The court attaches great significance to the fact that Quinton Heights is a relatively long district which extends into minority populations in the north and east. Id. at 883. But what the court does not point out is that the southernmost part includes all of the area for Topeka Country Club and the Kansas National Guard Armory, which easily comprises one-quarter of the school's attendance area. See rec. ex. vol. I at 49; rec. ex. vol. V at 60. These areas probably have few, if any, elementary students; thus from a practical perspective, the Quinton Heights area is not as long as it may appear because its southernmost area is occupied by large institutions. Merely because the balance of the school's attendance boundaries includes minority population, or that the school includes Topeka Avenue or railroad tracks, does not compel a conclusion that the boundaries were drawn with segregative intent or effect. See rec. vol. III at 291-92 (Mr. Lamson commenting about undesirable configuration of attendance area). Indeed, the presence of natural barriers in elementary attendance areas is hardly unique in this system. See rec. ex. vol. V at 225-228 (listing 20 elementary schools with natural barriers, e.g. institutional spaces, major thoroughfares, waterways, railroad tracks).

The court's earlier discussion of "peculiar-looking attendance boundaries" is peculiar in itself because, like the court's subjective approach to map reading, the court of appeals has moved far beyond its review function into the realm of a trial court. Court's Opinion at 880. Still, based on this untethered perception of "peculiarity," the court declares that Quinton Heights' current boundaries are segregative. Id. Put simply, the court seeks to overturn the district court's factual finding that the attendance zones are not gerrymandered. Brown, 671 F. Supp. 1300. Notwithstanding the countless hours spent on this case by the panel, I remain firmly convinced that we have added very little to the district court's amply supported findings. Admittedly, the evidence is not perfectly consistent and all in defendants' favor. But it is not nearly so one-sided or in favor of the plaintiffs' as this court apparently believes. The trial judge had a firm grip on the applicable law and a superb command of the facts in this complex case. We have done no better, even assuming that we were so empowered.

F.

Notwithstanding that Lowman Hill Elementary began as an all-white de jure school, this court finds that "[t]he school appears to have been designed and maintained as a school with a concentration of minority students." Court's Opinion at 884. The court begins by describing plaintiffs' contention concerning optional attendance zones, see Appellants' Brief at 13 n. 20; rec. supp. vol. XI at 96-98; vol. III at 235-41; however, it is unclear whether the court accepts those contentions, in light of the district court's rejection of those same contentions. Court's Opinion at 883-84; see Brown, 671 F. Supp. at 1298 (optional attendance zones (all eliminated in 1976) did not have significant segregative effect; any effect attenuated). The emphasis of this court seems to be on the inclusion of two areas that have greater concentrations of black residents in the northwestern part of central Topeka, within the Lowman Hill attendance boundaries. The district court pointed out that these two areas "are closer to the Lowman Hill Schools than other schools in the area." Brown, 671 F. Supp. at 1301. Indeed, that was the observation of defendants' expert Dr. Clark. Rec. vol. XI at 2344-45.

In 1959, a new and expanded Lowman Hill school was constructed which received black students from Buchanan. Rec. supp. vol. XI at 92. This increased Lowman Hill's minority percentage to 49.63% by 1966, but the school has never had a majority-minority population. Brown, 671 F. Supp. at 1301. The district court gave very close attention to this school and its relationship to surrounding schools.

This court mentions the closings of three surrounding schools and complains that these closings "did not affect the minority student percentage at Lowman Hill." Court's Opinion at 880. The court then provides the answer to "why not?" when it tells us that "the student population it [Lowman Hill] acquired was apparently of the same makeup as its existing population." Id. at 884. The answer is not surprising when one compares the minority percentages of the schools that closed in relation to Lowman Hill.Year Closed Year Closed School Closed

Rec. ex. vol. IV at 86, 101, 105. The above chart indicates that the schools that were closed were very similar in minority concentration to Lowman Hill. If anything, the presence of racially similar neighboring schools that were ultimately closed undercuts the court's finding that Lowman Hill is racially identifiable and used to concentrate minorities.

Looking at the neighboring schools and their racial composition in 1985 hardly compels the conclusion that Lowman Hill is racially identifiable.School Hill Minority %

G.

The court next shifts to "primarily white schools," as racially identifiable. These apparently are schools which this court has determined should have greater minority enrollments, both in the past and the future. Court's Opinion at 884. These schools purportedly have an identity as "schools for white children," id. at 884, which I suspect can only be changed to the court's satisfaction by strict racial balancing, regardless of the de facto reasons for their current racial composition. The purpose of desegregation law is to wipe out the effects of the prior de jure system, not to eliminate every difference in racial composition between schools. See supra p. 909-11. Moreover, many of these schools have experienced increases in minority percentages as a result of residential movement. Brown, 671 F. Supp. at 1303-04. Despite this court's clear view that positive changes in racial composition due to demographic factors are not relevant to determining whether the vestiges of a prior de jure system remain, see Court's Opinion at 874, 875, one would expect that a school district operating a dual system would have taken action to oppose or modify the effects of trends. That is not what has happened in Topeka. Brown, 671 F. Supp. at 1304.

Concerning the location of new schools, this court makes the following statement which could give the wrong impression:

All of the elementary schools and all but one of the secondary schools opened by the school district since Brown II were built in the white outer part of the district. Robinson middle school, which serves the central part of Topeka, is the sole exception. Furthermore, many of these schools were built near to the edges of the district.

Court's Opinion at 881; see also id. at 874-75. The judgment in Brown II was announced May 31, 1955. To be sure, numerous schools were opened in the western part of the district and had substantial white enrollments. The district court identified these as:

McClure, Sheldon, McEachron, McCarter, Bishop and Hudson elementary schools, as well as Jardine [1961], Eisenhower [1961] and Landon Junior High Schools [1963] and Topeka West Senior High School [1961]. These schools were in areas of residential expansion. During the same period, new schools with mixed racial compositions replaced existing facilities at Lowman Hill, Lafayette, Highland Park Central, Central Park and Belvoir elementary schools.

Brown, 671 F. Supp. at 1299; rec. supp. vol. XI at 194, 225. It should be added that Quinton Heights was replaced in 1954. Rec. supp. vol. XI at 40, 50. Thus, several schools which were not on the outer edge of Topeka were included in the district's new school construction during and subsequent to the mid-50's: Lowman Hill, 1959; Quinton Heights, 1954; Lafayette, 1956; Highland Park Central, 1965; Belvoir, 1967; Central Park, 1967. Brown, 671 F. Supp. at 1299; rec. supp. vol. XI at 40, 43, 50, 92, 166, 170.

Based on its erroneous finding that Topeka's neighborhood school plan resulted only in the construction of schools for white children in the far reaches of the district, thereby increasing segregation, the court tells us that "[a] school district conscious of its constitutional duty would have attempted to counteract that effect." Court's Opinion at 881 (citing Diaz, 733 F.2d at 667-69); see also Court's Opinion at 864 and 864 n. 31 (endorsing "castigation" approach). Although Diaz contains helpful language that a neighborhood school policy must be carefully analyzed as to application and enforcement before constitutional liability may be imposed, 733 F.2d at 664, the case is both legally and factually distinguishable. The duty of the school board in Diaz was different than that of these defendants; clearly established California state law required affirmative action to relieve segregation, regardless of the cause. Diaz, 733 F.2d at 666-67. Moreover, according to the majority in Diaz, the board failed to consider state guidelines and rejected alternatives which would have enabled it to comply with its duty. Id. at 669. That is not what occurred here. Noticeably absent from the plaintiffs' evidence is any complaint to the school board or the district court, when the school construction in question occurred.

Finally, I come to Topeka West High School, built in 1961 and considered by the plaintiffs and the court as "the quintessential example of deliberate channeling of white students to schools in the western part of the school district." Court's Opinion at 884; see also id. at 881 (plaintiffs/court finding "that Topeka deliberately placed new schools as far as possible from minority residential areas in order to polarize the district into white outer and minority inner schools."). Once again, this court's focus on "deliberate" conduct is inconsistent with its criticism of the district court for considering intent. Although this court criticizes the location of the school, which is near the western edge of the district, there were valid reasons, which have been discussed elsewhere, for placing the school in that location. See supra p. 916; see also Brown, 671 F. Supp. at 1300. The notion that Topeka West was built to channel white students from Topeka High, as contended by the plaintiffs, is not supported by the facts. In 1961, Topeka High was a winner of the prestigious Bellamy Award. Rec. vol. X at 1500. But it was overcrowded with 2,300 students. Rec. vol. IX at 1442-43. Topeka West opened with 700 students the first year, 900 the second, 1100 the third and continued to grow until 1969 when part of the attendance area was transferred back to Topeka High. Id. at 1434. Topeka High led high school enrollment totals for several years after the opening of Topeka West. Like Topeka High, Topeka West has won a prestigious award: Topeka West was designated a National School of Excellence recently, rec. supp. vol. V at 1622-23. Topeka West has had increased minority enrollment over the years as the western portion of Topeka becomes more residentially integrated.

H.

In addition to school openings, the court has undertaken to overturn the district court's finding that, on the whole, school closings were "an integrative device." Brown, 671 F. Supp. at 1299. Again, the court takes shelter in describing the plaintiffs' contention: "that school closings of highly segregated schools were not desegregative in practice because the school district simply reassigned the students at those schools en masse to a nearby school which then took over the segregated status of the school closed." Court's Opinion at 881; see also supra note 53. The court then adopts this as its finding: "The evidence supports this contention for the 1950s and 1960s." Court's Opinion at 881. Regarding the 1970's on the whole, the court tells us that most closings were not segregative or desegregative. Id.

In its section on school closings, the district court mentioned 13 closings by name. Brown, 671 F. Supp. at 1299. I have discussed several closings in the context of this dissent and see no reason why the district court should be overturned. This court's one paragraph summary analysis of school closings, while giving the appearance of completeness, does not facilitate my review. I cannot respond to the court's position without knowing which closings are covered by its findings and why.

VIII.

The district court recognized that the plaintiffs "hinge their case on this absence of racial balance in all schools and the failure of the defendants to promote racial balance." Brown, 671 F. Supp. at 1295.

At any time, more could have been done to achieve racial balance in the schools. But, it begs the issue of this case to argue that racial balancing must be done today because it was not done yesterday. More should be done to improve racial balance in the schools if the existing imbalance follows from defendants' past intentional segregative conduct.

Id. at 1309. Thereafter, the district court found that the present racial imbalance does not derive from "the de jure system or a foot-dragging segregationist policy." Id.

In 1985-86, Topeka had no one-race or virtually one-race minority schools. Whether we look at percentages or absolute numbers, the district court was not clearly erroneous in deciding that the Topeka system was not currently segregated. Of 34 schools, only three had majority-minority populations. Brown, 671 F. Supp. at 1294. Those three schools (Belvoir, 146 minority students (ms); Lafayette, 217 ms; and Highland Park North, 179 ms) served 542 minority elementary students or one-quarter of the minority elementary population (2,141 ms). See rec. ex. vol. IV at 170, 171, 174; Court's Opinion at 877. Five elementary schools (Crestview, 32 ms; Gage, 25 ms; McCarter, 35 ms; McClure, 23 ms; Potwin, 18 ms) had 90 + % white enrollment, yet they served 6.2% of the minority elementary population. Id. at 170, 172, 174. Two of the middle schools which are now combined (French, 241 white students (ws); and Landon, 205 ws) served 31 percent of the white middle school population (1,446 ws); while serving 6.95% of the minority middle school population (37 ms/532 ms). Id. at 175-176. At the high school level, Topeka West had a 90 + % white enrollment, serving 41.6% (1391 ws/3349 ws) of the white high school population, but also serving 11.48% (120 ms/1045 ms) of the minority high school population. Id. at 176-77. Though this court views the evidence differently, I submit that the district court's finding is supported by these numbers, when considered on this record, and looking back to the prior years. See supra text accompanying notes 9-13; see also supra note 18.

Although the court repeatedly tells us that racial balancing is not the objective, the court does acknowledge that "numbers are usually the focus of desegregation plans." Court's Opinion at 886; see also id. at 860. Without question, numbers resulting in better (strict) racial balance are the only remedy for what the court perceives as "schools that are now and always have been white, schools that are now and long have been heavily minority, and others." Court's Opinion at 886. The court seems to be leaning toward replication of the system-wide minority average in every school when it indicates that even compliance with a ± 15% standard for student assignment would be insufficient. Court's Opinion at 885. In the words of this court:

Dr. Foster also pointed out that such an approach would move schools now just beyond the 15% range to just within it, an improvement he did not find greatly desegregative as would preserve the existence of schools differing by as much as 30% in their minority percentage and thus still marked as minority or white schools.

Id. at 885-86. (footnote omitted). The inherent nature of a ± 15% standard for student assignment means that some schools may vary by as much as 30%, though many probably will not. My research has not disclosed any court finding that compliance with a ± 15% standard as to student assignment was indicative of a current condition of segregation.

When one looks carefully at the most recent data we have on student assignment, see supra note 18, while considering the racial composition of this system (74% white), this court's views are unique, given the wide distribution of minority students throughout the system and the absence of one-race minority schools. Indeed, several courts reviewing desegregation plans have given effect to the language in Swann that some "one-race, or virtually one-race, schools within a district [are] not in and of [themselves] the mark of a system that still practices segregation by law." 402 U.S. at 26, 91 S.Ct. at 1281. See, e.g., Morgan, 831 F.2d at 323 (Boston system with majority-minority student enrollment unitary; there was "substantial degree of integration" despite several schools with 80% + enrollments of one race and several schools out of compliance with court-ordered desegregation plan); Calhoun v. Cook, 522 F.2d 717, 719 (5th Cir. 1975) (Atlanta system with 85% black enrollment unitary; 92/148 schools 90% + black); Ross v. Houston Indep. School Dist., 699 F.2d 218, 219, 226 (5th Cir. 1983) (Houston system with 74% minority enrollment unitary; 55/226 schools were 90% + black and 2 schools all white). I submit that the presence of some predominantly white schools in this system and three schools which have a slight majority of minority students does not violate the equal protection clause. Washington, 426 U.S. at 240, 96 S.Ct. at 2047-48 (predominantly one-race schools not alone violative of equal protection).

Because the numbers are not indicative of a current condition of segregation, and the testimony relied upon by the district court supports the very opposite, it becomes necessary for the court to engage an unyielding presumption against the Topeka system from the outset. The court simply has jumped the gun in invoking the Keyes presumption to arrive at the preliminary and critical conclusion that there is a current condition of segregation in Topeka. Nowhere is this more evident than in footnote 61, Court's Opinion at 873-74, when the court explains that its discussion of the current condition does not conflict with the district court's conclusion that there is no unlawful segregation in Topeka. This court reasons that because the district court "did find the existence of racial disparities in school enrollment and staff/faculty assignment," it should have immediately invoked a presumption against Topeka: "Our disagreement with the district court is chiefly on the significance of these findings in a district with Topeka's history, and bearing the weight of a presumption against it which the district court failed to accord." Court's Opinion at 874 n. 61. This court's understanding of the district court's findings differs from mine. See supra p. 914-17. But even granting the comparatively small departure from perfect racial balance in this system, Topeka's history is one factor, among many, that a court may consider in deciding whether the presumption should be invoked. "While that history of discrimination cannot and should not be ignored, it `cannot in the manner of original sin, condemn governmental action that is not itself unlawful.'" Riddick, 784 F.2d at 539 (quoting City of Mobile v. Bolden, 446 U.S. 55, 74, 100 S.Ct. 1490, 1503, 64 L.Ed.2d 47 (1980) (plurality opinion by Justice Stewart)). To require perpetual application of the presumption would mean that no school district which had a de jure system in 1954, could ever be free of the presumption and achieve unitary status. The requirement of a current condition of segregation resulting from past or present segregative intent (as opposed to segregation resulting from voluntary demographic change), prior to the operation of the presumption, allows a school system to redeem itself and, one day, be free of federal court supervision. See Riddick, 784 F.2d at 538-39 (rejecting argument "that the Norfolk school board must continue to justify all of its actions because of the history of segregation."). That day should come for U.S.D. 501 in Topeka.

Alternatively, even if the presumption applies, the school board may adduce proof that the effects of past segregative intent have attenuated completely. Keyes, 413 U.S. at 211, 93 S.Ct. at 2699. This was recognized in Dayton II when the Court indicated that the Dayton school board "had failed to come forward with evidence to deny `that the current racial composition of the school population reflects the systemwide impact' of the Board's prior discriminatory conduct." Id. 443 U.S. at 537-38, 99 S.Ct. at 2979-80 (quoting Brinkman, 583 F.2d at 258). Here, "the Board was given ample opportunity to counter the evidence of segregative purpose and current, systemwide impact," Columbus, 443 U.S. at 467-68, 99 S.Ct. at 2951-52, and such evidence was credited by the district court.

In repeatedly castigating the defendants for noncompliance with the affirmative duty, the court relies upon the same incomplete approach that plaintiffs' expert took — not considering regional and local migration and other demographic factors which necessarily influence school board decisions. See rec. vol. IV at 510-511. For example, the court criticizes the defendants for building new schools in the late 1950's and 60's to accommodate population growth in western Topeka and concludes that "the erection of so many schools devoted to educating white children had a significant segregative effect on the district." Court's Opinion at 881. In arriving at this conclusion, the court apparently relies upon the following: "With respect to the new schools themselves, [the plaintiffs] argue that Topeka should have taken advantage of the underutilized inner schools; that is, that Topeka should have voluntarily bused white children from outer Topeka to the inner city schools instead of putting up new buildings." Court's Opinion at 881; see also rec. supp. vol. XI at 158-72 (Lamson theory). It is one thing to theorize that students could have been bused to a slightly underutilized seven grade (K-6) elementary school (e.g. Lowman Hill with an average of 43 spaces unused from 1957-66, or Quinton Heights with 28 spaces unused in 1966), but it is quite another to match the grade levels of spaces available each year with the grade levels of a hypothetical group that would be assigned to the school annually. See rec. vol. IV at 448-49 (underutilization analysis did not consider grade level of students); vol. X at 1538-41; supp. vol XI at 169. Moreover, this suggestion ignores the complaint that HEW had in 1974 concerning educating students in older schools. See supra p. 905.

Another of the administrative possibilities suggested by Mr. Lamson to improve racial balance would have been to combine Highland Park Central and Highland Park South in 1968, and build a school between the two. This proposal would have entailed abandoning two existing buildings with large enrollments, and the expense of building a new school. Rec. vol. X at 1475-76. The combined enrollment of the new school would have been 1,334 students, something which has never occurred in Topeka, and what deputy superintendent Henson thought was "too large an elementary school for this community or any community." Id.; rec. vol. IV at 512-13; see also supp. vol. XI at 170. Moreover, in suggesting alternatives for the 1950's and 1960's decisions concerning elementary boundaries, some of these alternatives would have required busing and lunch provisions for students, but Mr. Lamson did not consider that the Topeka system lacked a hot lunch program until the mid-60's. Rec. vol. IV at 443-448, 510; supp. vol. XI at 34-35, 56.

Unfortunately, one could get the wrong impression concerning the board members' attitude about desegregation. This court seems to fault the board members for not testifying that they regularly took desegregation concerns into account. Court's Opinion at 878. I do not read their testimony as precluding such concerns. Mr. Dennis C. Payne, board member from 1957-65, explained that the neighborhood concept was in keeping with the desegregation required by Brown because, previously, black children were not allowed to attend the schools closest to their homes. Rec. vol. XII at 2417 ("So we felt the neighborhood concept was what everybody wanted and what we wanted."). Mr. Payne testified to an absence of gerrymandering for racial segregation. Id. He also testified that he was aware of white flight due to the original four step plan and adherence to the neighborhood school plan. This court quotes his response: "What could we do? We can't make people move, no way." Id. at 2423 (quoted in Court's Opinion at 878). Considering the time period to which this statement refers, and the school district's then-recent adoption of a plan which met with the approval of this court, I would suggest that the school board could not prohibit people from moving, although perhaps it could have changed the consequences of a move with respect to school assignment. Cf. Ross, 699 F.2d at 225 (when district has successfully implemented plan to achieve unitary status, there is "no affirmative fourteenth amendment duty to respond to the private actions of those who vote with their feet").

Mr. Joe Douglas, Jr., the fire chief of the City of Topeka, and a board member from 1977-85 (board president for two years), testified that he and the other black member of the board opposed the open enrollment plan, which was adopted due to pressure from working parents who wanted their children to attend schools close to day care arrangements. Rec. vol. XII at 2441; see Court's Opinion at 878 n. 81. The plan was soon discontinued because of its effect on minority percentages. Rec. vol. XII at 2442-43. It was replaced by a majority to minority transfer plan. Id. at 2443-44; Brown, 671 F. Supp. at 1298. While he was on the board, Mr. Douglas advocated that no school in the district should exceed 50% minority, but the fact that some schools exceeded that percentage was not indicative of segregation, which he termed "as in the eye of the beholder." Rec. vol. XII at 2454-55. Contrary to the court's assertion that board members did not consider desegregation concerns, Mr. Douglas testified that the school board that he served on "took as a priority the lowering of minority percentages." See Court's Opinion at 878; rec. vol. XII at 2456. The current superintendent of schools, Dr. Edwards, supported the commitment to improve the racial composition of the schools. Rec. vol. XII at 2530.

During trial, there was testimony about concerns identified by the National Commission on Excellence in Education which prepared A Nation at Risk. Rec. supp. vol. VII at 1996-99. The Commission emphasized that improved education is essential not only to assure that the United States will remain competitive internationally, but also to assure "the intellectual, moral and spiritual strengths of our people which knit together the very fabric of our society." National Commission on Excellence in Education, A Nation at Risk: The Imperative For Educational Reform, at 7. It appears that Topeka has heeded the call to provide "all, regardless of race or class or economic status, . . . a fair chance and . . . the tools for developing their individual powers of mind and spirit to the utmost," for "individuals in our society who do not possess the levels of skill literacy, and training essential to this new era will be effectively disenfranchised, not simply from the material rewards that accompany competent performance, but also from the chance to participate fully in our national life." A Nation at Risk at 4, 7. "No person of goodwill toward his fellow man can logically argue" with the naturally correct premise of Brown I: equal educational opportunity in a unitary system for all children regardless of race. Keyes v. School Dist. No. 1, 521 F.2d 465, 490 (10th Cir. 1975) (Barrett, J., concurring), cert. denied, 423 U.S. 1066, 96 S.Ct. 806, 46 L.Ed.2d 657 (1976). But neither this court's opinion nor the record convinces me that the equal opportunity and benefits provided to all by Topeka's neighborhood schools will be improved by the protracted federal court supervision which is sure to follow at the expense of valued local control. See Milliken I, 418 U.S. at 741, 94 S.Ct. at 3125 (recognizing virtues of local control of public education).

Like the district court, I do not think that the constitutional command to eliminate the dual system goes unheeded 35, or even 30, years after Brown I. Within that time, tremendous demographic and social change has affected Topeka and its school system. The school board has demonstrated concern for improving the racial composition in the Topeka schools while retaining the neighborhood school concept. In some cases, the two are not compatible, but in this system substantial integration has been achieved, and the system is now unitary, although not perfect. Because I am not persuaded of a remaining constitutional violation, and I am persuaded that the appellate court has retried the case for a different outcome, I would affirm the district court's judgment declaring the Topeka system unitary at last. Brown, 671 F. Supp. at 1311.

Total Enrollment U.S.D. 501

Racial Inventory of U.S.D. 501 Elementary Students School Year 1955-56

Racial Inventory of U.S.D. Junior High Students School Year 1955-1956

Racial Inventory of USD 501 Elementary Students School Year 1968-69

Racial Inventory of USD 501 Junior High Students School Year 1968-69

Racial Inventory of USD 501 Senior High Students School Year 1968-69

Racial Inventory of USD 501 Students School Years 1985-1986 compared with 1981-82

Elementary Schools 1985-86

Middle Schools 1985-86

Senior High Schools 1985-86

Topeka Desegregation Indexes Based on Actual Enrollment Data Selected Years

Racial Inventory Employees

Elementary Minority Staff Percentage

Elementary School Faculty/Staff Assignment 1985-86

Staff

Middle School and High School Faculty/Staff Assignment 1985-86

Staff

Minority Staffing Report Elementary Schools Includes Administrative and Certificated Staff 1986-87

Minority Staffing Report Middle Schools Includes Administrative and Certificated Staff 1986-87

Minority Staffing Report High Schools Includes Administrative and Certificated Staff 1986-87

Racial Inventory-Belvoir Elementary 1952-1985

Racial Inventory-Hudson Elementary 1966-1985

Racial Inventory-Highland Park North Elementary 1952-1985

Racial Inventory-Lafayette Elementary 1953-1985

Racial Inventory-Quinton Heights Elementary 1953-1985

Racial Inventory-Lowman Hill Elementary 1953-1985

Racial Inventory-Topeka West High School 1966-1985