Keyes v. School Dist. No. 1, Denver

Mr. Justice Powell

concurring in part and dissenting in part.

I concur in the remand of this case for further proceedings in the District Court, but on grounds that differ from those relied upon by the Court.

This is the first school desegregation case to reach this Court which involves a major city outside the South. It comes from Denver, Colorado, a city and a State which have not operated public schools under constitutional or statutory provisions which mandated or permitted racial segregation.1 Nor has it been argued that any other legislative actions (such as zoning and housing laws) contributed to the segregation which is at issue.2 The Court has inquired only to what extent the Denver public school authorities may have contributed to the school segregation which is acknowledged to exist in Denver.

The predominantly minority schools are located in two areas of the city referred to as Park Hill and the core city area. The District Court considered that a school *218with a concentration of 70% to 75% “Negro or Hispano students” was identifiable as a segregated school. 313 F. Supp. 61, 77. Wherever one may draw this line, it is undisputed that most of the schools in these two areas are in fact heavily segregated in the sense that their student bodies are overwhelmingly composed of non-Anglo children. The city-wide school mix in Denver is 66% Anglo, 14% Negro, and 20% Hispano. In areas of the city where the Anglo population largely resides, the schools are predominantly Anglo, if not entirely so.

The situation in Denver is generally comparable to that in other large cities across the country in which there is a substantial minority population and where desegregation has not been ordered by the federal courts. There is segregation in the schools of many of these cities fully as pervasive as that in southern cities prior to the desegregation decrees of the past decade and a half. The focus of the school desegregation problem has now shifted from the South to the country as a whole. Unwilling and footdragging as the process was in most places, substantial progress toward achieving integration has been made in Southern States.3 No comparable progress has been made in many nonsouthern cities with large minority populations4 primarily because of the de facto/de jure *219distinction nurtured by the courts and accepted complacently by many of the same voices which denounced the evils of segregated schools in the South.5 But if our national concern is for those who attend such schools, rather than for perpetuating a legalism rooted in history rather than present reality, we must recognize that the evil of operating separate schools is no less in Denver than in Atlanta.

I

In my view we should abandon a distinction which long since has outlived its time, and formulate constitutional principles of national rather than merely regional application. When Brown v. Board of Education, 347 U. S. 483 (1954) (Brown I), was decided, the distinction between *220de jure and de facto segregation was consistent with the limited constitutional rationale of that case. The situation confronting the Court, largely confined to the Southern States, was officially imposed racial segregation in the schools extending back for many years and usually embodied in constitutional and statutory provisions.

The great contribution of Brown I was its holding in unmistakable terms that the Fourteenth Amendment forbids state-compelled or state-authorized segregation of public schools. 347 U. S., at 488, 493-495. Although some of the language was more expansive, the holding in Brown I was essentially negative: It was impermissible under the Constitution for the States, or their, instru-mentalities, to force children to attend segregated schools. The forbidden action was de jure, and the opinion in Brown I was construed — for some years and by many courts — as requiring only state neutrality, allowing “freedom of choice” as to schools to be attended so long as the State itself assured that the choice was genuinely free of official restraint.6

But the doctrine of Brown I, as amplified by Brown II, 349 U. S. 294 (1955), did not retain its original meaning. In a series of decisions extending from 1954 to 1971 the *221concept of state neutrality was transformed into the present constitutional doctrine requiring affirmative state action to desegregate school systems.7 The keystone case was Green v. County School Board, 391 U. S. 430, 437-438 (1968), where school boards were declared to have “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.” The school system before the Court in Green was operating in a rural and sparsely settled county where there were no concentrations of white and black populations, no neighborhood school system (there were only two schools in the county), and none of the problems of an urbanized school district.8 The Court properly identified the freedom-of-choice program there as a subterfuge, and the language in Green imposing an affirmative duty to convert to a unitary system was appropriate on the facts before the Court. There was, however, reason to question to what extent this duty would apply in the vastly different factual setting of a large city with extensive areas of residential segregation, presenting problems and calling for solutions quite different from those in the rural setting of New Kent County, Virginia.

But the doubt as to whether the affirmative-duty concept would flower into a new constitutional principle of general application was laid to rest by Swann v. Charlotte-Mecklenburg Board of Education, 402 U. S. 1 (1971), in which the duty articulated in Green was applied to the *222urban school system of metropolitan Charlotte, North Carolina. In describing the residential patterns in Charlotte, the Court noted the “familiar phenomenon” in the metropolitan areas of minority groups being “concentrated in one part of the city,” 402 U. S., at 25, and acknowledged that:

“Rural areas accustomed for half a century to the consolidated school systems implemented by bus transportation could make adjustments more readily than metropolitan areas with dense and shifting population, numerous schools, congested and complex traffic patterns.” 402 U. S., at 14.

Despite this recognition of a fundamentally different problem from that involved in Green, the Court nevertheless held that the affirmative-duty rule of Green was applicable, and prescribed for a metropolitan school system with 107 schools and some 84,000 pupils essentially the same remedy — elimination of segregation “root and branch” — which had been formulated for the two schools and 1,300 pupils of New Kent County.

In Swann, the Court further noted it was concerned only with States having “a long history” of officially imposed segregation and the duty of school authorities in those States to implement Brown I. 402 U. S., at 5-6. In so doing, the Court refrained from even considering whether the evolution of constitutional doctrine from Brown I to Green/Swann undercut whatever logic once supported the de facto/de jure distinction. In imposing on metropolitan southern school districts an affirmative duty, entailing large-scale transportation of pupils, to eliminate segregation in the schools, the Court required these districts to alleviate conditions which in large part did not result from historic, state-imposed de jure segregation. Rather, the familiar root cause of segregated schools in all the biracial metropolitan areas of our country is essen*223tially the same: one of segregated residential and migratory patterns the impact of which on the racial composition of the schools was often perpetuated and rarely ameliorated by action of public school authorities. This is a national, not a southern, phenomenon. And it is largely unrelated to whether a particular State had or did not have segregative school laws.9

Whereas Brown I rightly decreed the elimination of state-imposed segregation in that particular section of the country where it did exist, Swann imposed obligations on southern school districts to eliminate conditions which are not regionally unique but are similar both in origin and effect to conditions in the rest of the country. As the remedial obligations of Swann extend far beyond the elimination of the outgrowths of the state-imposed segregation outlawed in Brown, the rationale of Swann points inevitably toward a uniform, constitutional approach to our national problem of school segregation.

II

The Court's decision today, while adhering to the de jure/de facto distinction, will require the application *224of the Green/Swarm doctrine of “affirmative duty” to the Denver School Board despite the absence of any history of state-mandated school segregation. The only evidence of a constitutional violation was found in various decisions of the School Board. I concur in the Court’s position that the public school authorities are the responsible agency of the State, and that if the affirmative-duty doctrine is sound constitutional law for Charlotte, it is equally so for Denver. I would not, however, perpetuate the de jure/de jacto distinction nor would I leave to petitioners the initial tortuous effort of identifying “segre-gative acts” and deducing “segregative intent.” I would hold, quite simply, that where segregated public schools exist within a school district to a substantial degree, there is a prima facie case that the duly constituted public authorities (I will usually refer to them collectively as the “school board”) are sufficiently responsible10 to warrant imposing upon them a nationally applicable burden to demonstrate they nevertheless are operating a genuinely integrated school system.

A

The principal reason for abandonment of the de jure/ de jacto distinction is that, in view of the evolution of the holding in Brown I into the affirmative-duty doctrine, the distinction no longer can be justified on a principled basis. In decreeing remedial requirements for the Charlotte/Mecklenburg school district, Swann dealt with a metropolitan, urbanized area in which the basic *225causes of segregation were generally similar to those in all sections of the country, and also largely irrelevant to the existence of historic, state-imposed segregation at the time of the Brown decision. Further, the extension of the affirmative-duty concept to include compulsory student transportation went well beyond the mere remedying of that portion of school segregation for which former state segregation laws were ever responsible. Moreover, as the Court’s opinion today abundantly demonstrates, the facts deemed necessary to establish de jure discrimination present problems of subjective intent which the courts cannot fairly resolve.

At the outset, one must try to identify the constitutional right which is being enforced. This is not easy, as the precedents have been far from explicit. In Brown I, after emphasizing the importance of education, the Court said that:

"Such an opportunity, where the state has undertaken to provide it, is a right which must be made available to all on equal terms.” 347 U. S., at 493.

In Brown II, the Court identified the “fundamental principle” enunciated in Brown I as being the unconstitutionality of “racial discrimination in public education,” 349 U. S., at 298, and spoke of “the personal interest of the plaintiffs in admission to public schools as soon as practicable on a nondiscriminatory basis.” 349 U. S., at 300. Although this and similar language is ambiguous as to the specific constitutional right, it means — as a minimum — that one has the right not to be compelled by state action to attend a segregated school system. In the evolutionary process since 1954, decisions of this Court have added a significant gloss to this original right. Although nowhere expressly articulated in these terms, I would now define it as the right, derived from the Equal Protection Clause, to expect that once the State has as*226sumed responsibility for education, local school boards will operate integrated school systems within their respective districts.11 This means that school authorities, consistent with the generally accepted educational goal of attaining quality education for all pupils, must make and implement their customary decisions with a view toward enhancing integrated school opportunities.

The term “integrated school system” presupposes, of course, a total absence of any laws, regulations, or policies supportive of the type of “legalized” segregation condemned in Brown. A system would be integrated in accord with constitutional standards if the responsible authorities had taken appropriate steps to (i) integrate faculties and administration; (ii) scrupulously assure equality of facilities, instruction, and curriculum opportunities throughout the district; (iii) utilize their authority to draw attendance zones to promote integration; and (iv) locate new schools, close old ones, and determine the size and grade categories with this same objective in mind. Where school authorities decide to undertake the transportation of students, this also must be with integrative opportunities in mind.

The foregoing prescription is not intended to be either definitive or all-inclusive, but rather an indication of the contour characteristics of an integrated school system in which all citizens and pupils may justifiably be confident that racial discrimination is neither practiced nor tolerated. An integrated school system does not *227mean — and indeed could not mean in view of the residential patterns of most of our major metropolitan areas — that every school must in fact be an integrated unit. A school which happens to be all or predominantly white or all or predominantly black is not a “segregated” school in an unconstitutional sense if the system itself is a genuinely integrated one.

Having school boards operate an integrated school system provides the best assurance of meeting the constitutional requirement that racial discrimination, subtle or otherwise, will find no place in the decisions of public school officials. Courts judging past school board actions with a view to their general integrative effect will be best able to assure an absence of such discrimination while avoiding the murky, subjective judgments inherent in the Court’s search for “segregative intent.” Any test, resting on so nebulous and elusive an element as a school board’s segregative “intent” provides inadequate assurance that minority children will not be shortchanged in the decisions of those entrusted with the nondiscriminatory operation of our public schools.

Public schools are creatures of the State, and whether the segregation is state-created or state-assisted or merely state-perpetuated should be irrelevant to constitutional principle. The school board exercises pervasive and continuing responsibility over the long-range planning as well as the daily operations of the public school system. It sets policies on attendance zones, faculty employment and assignments, school construction, closings and consolidations, and myriad other matters. School board decisions obviously are not the sole cause of segregated school conditions. But if, after such detailed and complete public supervision, substantial school segregation still persists, the presumption is strong that the school board, by its acts or omissions, is in some part responsible. Where state action and supervision are so *228pervasive and where, after years of such action, segregated schools continue to exist within the district to a substantial degree, this Court is justified in finding a prima facie case of a constitutional violation. The burden then must fall on the school board to demonstrate it is operating an “integrated school system.”

It makes little sense to find prima facie violations and the consequent affirmative duty to desegregate solely in those States with state-imposed segregation at the time of the Brown decision. The history of state-imposed segregation is more widespread in our country than the de jure/de facto distinction has traditionally cared to recognize.12 As one commentator has noted:

“[T]he three court of appeals decisions denying a constitutional duty to abolish de facto segregation all arose in cities — Cincinnati, Gary, and Kansas City, Kansas — where racial segregation in schools was formerly mandated by state or local law. [Deal v. Cincinnati Board of Education, 369 F. 2d 55 (CA6 1966), cert. denied, 389 U. S. 847 (1967); Downs v. Board of Education, 336 F. 2d 988 (CA10 1964), cert. denied, 380 U. S. 914 (1965); Bell v. School City of Gary, Ind,, 324 F. 2d 209 (CA7 1963), cert. denied, 377 U. S. 924 (1964).] Ohio discarded its statute in 1887, Indiana in 1949, and Kansas City not until the advent of Brown. If Negro and white parents in *229Mississippi are required to bus their children to distant schools on the theory that the consequences of past de jure segregation cannot otherwise be dissipated, should not the same reasoning apply in Gary, Indiana, where no more than five years before Brown the same practice existed with presumably the same effects?” Goodman, De Facto School Segregation: A Constitutional and Empirical Analysis, 60 Calif. L. Rev. 275, 297 (1972).13

Not only does the de jure/de facto distinction operate inequitably on communities in different sections of the country, more importantly, it disadvantages minority children as well. As the Fifth Circuit stated:

“ ‘The Negro children in Cleveland, Chicago, Los Angeles, Boston, New York, or any other area of the nation which the opinion classifies under de facto segregation, would receive little comfort from the assertion that the racial make-up of their school system does not violate their constitutional rights because they were born into a de facto society, while the exact same racial make-up of the school system in the 17 Southern and border states violates the *230constitutional rights of their counterparts, or even their blood brothers, because they were born into a de jure society. All children everywhere in the nation are protected by the Constitution, and treatment which violates their constitutional rights in one area of the country, also violates such constitutional rights in another area/ ” Cisneros v. Corpus Christi Independent School District, 467 F. 2d 142, 148 (CA5 1972) (en banc), quoting United States v. Jefferson County Board of Education, 380 F. 2d 385, 397 (CA5 1967) (Gewin, J., dissenting).14

The Court today does move for the first time toward breaking down past sectional disparities, but it clings tenuously to its distinction. It searches for de jure action in what the Denver School Board has done or failed to do, and even here the Court does not rely upon the results or effects of the Board’s conduct but feels compelled to find segregative intent:15

“We emphasize that the differentiating factor between de jure segregation and so-called de facto *231segregation to which we referred in Swann is purpose or intent to segregate.” Ante, at 208 (emphasis is the Court’s).

The Court’s insistence that the “differentiating factor” between de jure and de facto segregation be “purpose or intent” is difficult to reconcile with the language in so recent a case as Wright v. Council of the City of Emporia, 407 U. S. 451 (1972). In holding there that “motivation” is irrelevant, the Court said:

“In addition, an inquiry into the ‘dominant’ motivation of school authorities is as irrelevant as it is fruitless. The mandate of Brown II was to desegregate schools, and we have said that ‘[t]he measure of any desegregation plan is its effectiveness.’ Davis v. School Commissioners of Mobile County, 402 U. S. 33, 37. Thus, we have focused upon the effect — not the purpose or motivation — of a school board’s action in determining whether it is a permissible method of dismantling a dual system. . . .
“. . . Though the purpose of the new school districts was found to be discriminatory in many of these cases, the courts’ holdings rested not on motivation or purpose but on the effect of the action upon the dismantling of the dual school systems involved. That was the focus of the District Court in this case, and we hold that its approach was proper.” 407 U. S., at 462.

I can discern no basis in law or logic for holding that the motivation of school board action is irrelevant in Virginia and controlling in Colorado. It may be argued, of course, that in Emporia a prior constitutional viola*232tion had already been proved and that this justifies the distinction. The net result of the Court's language, however, is the application of an effect test to the actions of southern school districts and an intent test to those in other sections, at least until an initial de jure finding for those districts can be made. Rather than straining to perpetuate any such dual standard, we should hold forthrightly that significant segregated school conditions in any section of the country are a prima facie violation of constitutional rights. As the Court has noted elsewhere :

“Circumstances or chance may well dictate that no persons in a certain class will serve on a particular jury or during some particular period. But it taxes our credulity to say that mere chance resulted in there being no members of this class among the over six thousand jurors called in the past 25 years. The result bespeaks discrimination, whether or not it was a conscious decision on the part of any individual jury commissioner.” Hernandez v. Texas, 347 U. S. 475, 482 (1954). (Emphasis added.)

B

There is thus no reason as a matter of constitutional principle to adhere to the de jure/de jacto distinction in school desegregation cases. In addition, there are reasons of policy and prudent judicial administration which point strongly toward the adoption of a uniform national rule. The litigation heretofore centered in the South already is surfacing in other regions. The decision of the Court today, emphasizing as it does the elusive element of segregative intent, will invite numerous desegregation suits in which there can be little hope of uniformity of result.

The issue in these cases will not be whether segregated education exists. This will be conceded in most of them. *233The litigation will focus as a consequence of the Court’s decision on whether segregation has resulted in any “meaningful or significant” portion of a school system from a school board’s “segregative intent.” The intractable problems involved in litigating this issue are obvious to any lawyer. The results of litigation — often arrived at subjectively by a court endeavoring to ascertain the subjective intent of school authorities with respect to action taken or not taken over many years — will be fortuitous, unpredictable and even capricious.

The Denver situation is illustrative of the problem. The courts below found evidence of de jure violations with respect to the Park Hill schools and an absence of such violations with respect to the core city schools, despite the fact that actions taken by the school board with regard to those two sections were not dissimilar. It is, for example, quite possible to contend that both the construction of Manual High School in the core city area and Barrett Elementary School in the Park Hill area operated to serve their surrounding Negro communities and, in effect, to merge school attendance zones with segregated residential patterns. See Brief for Petitioners 80-83. Yet findings even on such similar acts will, under the de jure/de facto distinction, continue to differ, especially since the Court has never made clear what suffices to establish the requisite “segregative intent” for an initial constitutional violation. Even if it were possible to clarify this question, wide and unpredictable differences of opinion among judges would be inevitable when dealing with an issue as slippery as “intent” or “purpose,” especially when related to hundreds of decisions made by school authorities under varying conditions over many years.

This Court has recognized repeatedly that it is “extremely difficult for a court to ascertain the motivation, or collection of different motivations, that lie behind a *234legislative enactment,” Palmer v. Thompson, 403 U. S. 217, 224 (1971); McGinnis v. Royster, 410 U. S. 263, 276-277 (1973); United States v. O’Brien, 391 U. S. 367, 381 (1968). Whatever difficulties exist with regard to a single statute will be compounded in a judicial review of years of administration of a large and complex school system.16 Every act of a school board and school administration, and indeed every failure to act where affirmative action is indicated, must now be subject to scrutiny. The most routine decisions with respect to the operation of schools, made almost daily, can affect in varying degrees the extent to which schools are initially segregated, remain in that condition, are desegregated, or- — for the long term f.uture — are likely to be one or the other. These decisions include action or nonaction with respect to school building construction and location; the timing of building new schools and their size; the closing and consolidation of schools; the drawing or gerrymandering of *235student attendance zones; the extent to which a neighborhood policy is enforced; the recruitment, promotion and assignment of faculty and supervisory personnel; policies with respect to transfers from one school to another; whether, and to what extent, special schools will be provided, where they will be located, and who will qualify to attend them; the determination of curriculum, including whether there will be “tracks” that lead primarily to college or to vocational training, and the routing of students into these tracks; and even decisions as to social, recreational, and athletic policies.

In Swann the Court did not have to probe into segre-gative intent and proximate cause with respect to each of these “endless” factors. The basis for its de jure finding there was rooted primarily in the prior history of the desegregation suit. 402 U. S., at 5-6. But in a case of the present type, where no such history exists, a judicial examination of these factors will be required under today’s decision. This will lead inevitably to uneven and unpredictable results, to protracted and inconclusive litigation, to added burdens on the federal courts, and to serious disruption of individual school systems. In the absence of national and objective standards, school boards and administrators will remain in a state of uncertainty and disarray, speculating as to what is required and when litigation will strike.

C

Rather than continue to prop up a distinction no longer grounded in principle, and contributing to the consequences indicated above, we should acknowledge that whenever public school segregation exists to a substantial degree there is prima facie evidence of a constitutional violation by the responsible school board. It is true, of course, that segregated schools — wherever located — are not solely the product of the action or *236inaction of public school authorities. Indeed, as indicated earlier, there can be little doubt that principal causes of the pervasive school segregation found in the major urban areas of this country, whether in the North, West, or South, are the socio-economic influences which have concentrated our minority citizens in the inner cities while the more mobile white majority disperse to the suburbs. But it is also true that public school boards have continuing, detailed responsibility for the public school system within their district and, as Judge John Minor Wisdom has noted, “[w]hen the figures [showing segregation in the schools] speak so eloquently, a prima facie case of discrimination is established.” United States v. Texas Education Agency, 467 F. 2d 848, 873 (CA5 1972) (en banc). Moreover, as foreshadowed in Swann and as implicitly held today, school boards have a duty to minimize and ameliorate segregated conditions by pursuing an affirmative policy of desegregation. It is this policy which must be applied consistently on a national basis without regard to a doctrinal distinction which has outlived its time.

Ill

The preceding section addresses the constitutional obligation of public authorities in the school districts throughout our country to operate integrated school systems. When the schools of a particular district are found to be substantially segregated, there is a prima facie case that this obligation has not been met. The burden then shifts to the school authorities to demonstrate that they have in fact operated an integrated system as this term is defined, supra, at 227-228. If there is a failure successfully to rebut the prima facie case, the question then becomes what reasonable affirmative desegregative steps district courts may require to *237place the school system in compliance with the constitutional standard. In short, what specifically is the nature and scope of the remedy?

As the Court’s opinion virtually compels the finding on remand that Denver has a “dual school system,” that city will then be under an “affirmative duty” to desegregate its entire system “root and branch.” Green v. County School Board, 391 U. S., at 437-438. Again, the critical question is, what ought this constitutional duty to entail?

A

The controlling case is Swann, supra, and the question which will confront and confound the District Court and Denver School Board is what, indeed, does Swann require? Swann purported to enunciate no new principles, relying heavily on Brown I and II and on Green. Yet it affirmed a district court order which had relied heavily on “racial ratios” and sanctioned transportation of elementary as well as secondary pupils. Lower federal courts have often read Swann as requiring far-reaching transportation decrees 17 “to achieve the greatest possible degree of actual *238desegregation.” 402 U. S., at 26. In the context of a large urban area, with heavy residential concentrations of white and black citizens in different — and widely separated — sections of the school district, extensive dispersal and transportation of pupils is inevitable if Swann is read as expansively as many courts have been reading it to date.

To the extent that Swann may be thought to require large-scale or long-distance transportation of students in our metropolitan school districts, I record my profound misgivings. Nothing in our Constitution commands or encourages any such court-compelled disruption of public education. It may be more accurate to view Swann as having laid down a broad rule of reason under which desegregation remedies must remain flexible and other values and interests be considered. Thus the Court recognized that school authorities, not the federal judiciary, must be charged in the first instance with the task of desegregating local school systems. Id., at 16. It noted that school boards in rural areas can adjust more readily to this task than those in metropolitan districts “with dense and shifting population, numerous schools, congested and complex traffic patterns.” Id., at 14. Although the use of pupil transportation was approved as a remedial device, transportation orders are suspect “when the time or distance of travel is so great *239as to either risk the health of the children or significantly impinge on the educational process.” Id., at 30-31. Finally, the age of the pupils to be transported was recognized by the Court in Swann as one important limitation on the time of student travel. Id., at 31.

These factors were supposed to help guide district courts in framing equitable remedies in school desegregation cases.18 And the Court further emphasized that equitable decrees are inherently sensitive, not solely to the degree of desegregation to be achieved, but to a variety of other public and private interests:

“[A] school desegregation case does not differ fundamentally from other cases involving the framing of equitable remedies to repair the denial of a constitutional right. The task is to correct, by a balancing of the individual and collective interests, the condition that offends the Constitution. Id., at 15-16.

Those words echoed a similar expression in Brown II, 349 U. S., at 300:

“In fashioning and effectuating the decrees, the courts will be guided by equitable principles. Traditionally, equity has been characterized by a practical flexibility in shaping its remedies and by a facility for adjusting and reconciling public and private needs.”

Thus, in school desegregation cases, as elsewhere, equity counsels reason, flexibility, and balance. See, e. g., Lemon *240v. Kurtzman, 411 U. S. 192 (1973). I am aware, of course, that reasonableness in any area is a relative and subjective concept. But with school desegregation, reasonableness would seem to embody a balanced evaluation of the obligation of public school boards to promote desegregation with other, equally important educational interests which a community may legitimately assert. Neglect of either the obligation or the interests destroys the evenhanded spirit with which equitable remedies must be approached.19 Overzealousness in pursuit of any single goal is untrue to the tradition of equity and to' the “balance” and “flexibility” which this Court has always respected.

B

Where school authorities have defaulted in their duty to operate an integrated school system, district courts must insure that affirmative desegregative steps ensue. Many of these can be taken effectively without damaging state and parental interests in having children attend schools within a reasonable vicinity of home. Where desegregative steps are possible within the framework of a system of “neighborhood education,” school authorities must pursue them. For example, boundaries of neighborhood attendance zones should be drawn to integrate, to the extent practicable, the school's student body. Construction of new schools should be of *241such a size and at such a location as to encourage the likelihood of integration, Swann, supra, at 21. Faculty-integration should be attained throughout the school system, id., at 19; United States v. Montgomery County Board of Education, 395 U. S. 225 (1969). An optional majority-to-minority transfer program, with the State providing free transportation to desiring students, is also a helpful adjunct to a desegregated school system. Swann, supra, at 26-27. It hardly need be repeated that allocation of resources within the school district must be made with scrupulous fairness among all schools.

The above examples are meant to be illustrative, not exhaustive. The point is that the overall integrative impact of such school board decisions must be assessed by district courts in deciding whether the duty to desegregate has been met. For example, “neighborhood school plans are constitutionally suspect when attendance zones are superficially imposed upon racially defined neighborhoods, and when school construction preserves rather than eliminates the racial homogeny [sic] of given schools.” '20 Keyes v. School District No. 1, 445 F. 2d 990, 1005 (CA10 1971). See also United States v. Board of Education of Tulsa County, 429 F. 2d 1253, 1258-1259 (CA10 1970). This does not imply that decisions on faculty assignment, attendance zones, school construction, closing and consolidation, must be made to the detriment of all neutral, nonracial considerations. But these considerations can, with proper school board initiative, generally be met in a manner that will enhance the degree of school desegregation.

C

Defaulting school authorities would have, at a minimum, the obligation to take affirmative steps of the sort *242outlined in the above section. School boards would, of course, be free to develop and initiate further plans to promote school desegregation. In a pluralistic society such as ours, it is essential that no racial minority feel demeaned or discriminated against and that students of all races learn to play, work, and cooperate with one another in their common pursuits and endeavors. Nothing in this opinion is meant to discourage school boards from exceeding minimal constitutional standards in promoting the values of an integrated school experience.

A constitutional requirement of extensive student transportation solely to achieve integration presents a vastly more complex problem. It promises, on the one hand, a greater degree of actual desegregation, while it infringes on what may fairly be regarded as other important community aspirations and personal rights. Such a requirement is also likely to divert attention and resources from the foremost goal of any school system: the best quality education for all pupils. The Equal Protection Clause does, indeed, command that racial discrimination not be tolerated in the decisions of public school authorities. But it does not require that school authorities undertake widespread student transportation solely for the sake of maximizing integration.21

*243This obviously does not mean that bus transportation has no place in public school systems or is not a permissible means in the desegregative process. The transporting of school children is as old as public education, and in rural and some suburban settings it is as indispensable as the providing of books. It is presently estimated that approximately half of all American children ride buses to school for reasons unrelated to integration.22 At the secondary level in particular, where the schools are larger and serve a wider, more dispersed constituency than elementary schools, some form of public or privately financed transportation is often necessary. There is a significant difference, however, in transportation plans voluntarily initiated by local school boards for educational purposes and those imposed by a federal court. The former usually represent a necessary or convenient means of access to the school nearest home; the latter often require lengthy trips for no purpose other than to further integration.23 Yet the *244Court in Swann was unquestionably right in describing bus transportation as “one tool of school desegregation.” 402 U. S., at 30.24 The crucial issue is when, under what circumstances, and to what extent such transportation may appropriately be ordered. The answer to this turns — as it does so often in the law — upon a sound exercise of discretion under the circumstances.

Swann itself recognized limits to desegregative obligations. It noted that a constitutional requirement of “any particular degree of racial balance or mixing . . . would be disapproved . . . ,” and sanctioned district court use of mathematical ratios as “no more than a starting point in the process of shaping a remedy . . . Id,., at 24, 25. Thus, particular schools may be all white or all black and still not infringe constitutional rights if the system is genuinely integrated and school authorities are pursuing integrative steps short of extensive and disruptive transportation. The refusal of the Court in Swann to require racial balance in schools throughout the district or the arbitrary elimination of all “one-race schools,” id., at 26, is grounded in a recognition that *245the State, parents, and children all have at stake in school desegregation decrees, legitimate and recognizable interests.

The personal interest might be characterized as the desire that children attend community schools near home. Dr. James Coleman testified for petitioners at trial that “most school systems organize their schools in relation to the residents by having fixed school districts and some of these are very ethnically homogeneous.” App. 1549a. In Deal v. Cincinnati Board of Education, 369 F. 2d, at 60, the Sixth Circuit summarized the advantages of such a neighborhood system of schools: 25

“Appellants, however, pose the question of whether the neighborhood system of pupil placement, fairly administered without racial bias, comports with the requirements of equal opportunity if it nevertheless results in the creation of schools with predominantly or even exclusively Negro pupils. The neighborhood system is in wide use throughout the nation and has been for many years the basis of school administration. This is so because it is acknowledged to have several valuable aspects which are an aid to education, such as minimization of safety hazards to children in reaching school, economy of cost in reducing transportation needs, ease of pupil *246placement and administration through the use of neutral, easily determined standards, and better home-school communication.”

The neighborhood school does provide greater ease of parental and student access and convenience, as well as greater economy of public administration. These are obvious and distinct advantages, but the legitimacy of the neighborhood concept rests on more basic grounds.26

Neighborhood school systems, neutrally administered, reflect the deeply felt desire of citizens for a sense of community in their public education. Public schools have been a traditional source of strength to our Nation, and that strength may derive in part from the identification of many schools with the personal features of the surrounding neighborhood. Community support, interest, and dedication to public schools may well run higher with a neighborhood attendance pattern: distance may encourage disinterest. Many citizens sense today a decline in the intimacy of our institutions — home, church, and school — which has caused a concomitant decline in the unity and communal spirit of our people. I pass no judgment on this viewpoint, but I do believe that this Court should be wary of compelling in the name of constitutional law what may seem to many a dissolution in the traditional, more personal fabric of their public schools.

Closely related to the concept of a community and neighborhood education, are those rights and duties parents have with respect to the education of their children. The law has long recognized the parental duty to nurture, support, and provide for the welfare of children, inelud-*247ing their education. In Pierce v. Society of Sisters, 268 U. S. 510, 534-535, a unanimous Court held that:

“Under the doctrine of Meyer v. Nebraska, 262 U. S. 390, we think it entirely plain that the Act of 1922 unreasonably interferes with the liberty of parents and guardians to direct the upbringing and education of children under their control. . . . The child is not the mere creature of the State; those who nurture him and direct his destiny have the right, coupled with the high duty, to recognize and prepare him for additional obligations.”

And in Griswold v. Connecticut, 381 U. S. 479, 482 (1965), the Court noted that in Pierce, “the right to educate one’s children as one chooses is made applicable to the States by the force of the First and Fourteenth Amendments.” I do not believe recognition of this right can be confined solely to a parent’s choice to send a child to public or private school. Most parents cannot afford the luxury of a private education for their children, and the dual obligation of private tuitions and public taxes. Those who may for numerous reasons seek public education for their children should not be forced to forfeit all interest or voice in the school their child attends. It would, of course, be impractical to allow the wishes of particular parents to be controlling. Yet the interest of the parent in the enhanced parent-school and parent-child communication allowed' by the neighborhood unit ought not to be suppressed by force of law.

In the commendable national concern for alleviating public school segregation, courts may have overlooked the fact that the rights and interests of children affected by a desegregation program also are entitled to consideration. Any child, white or black, who is compelled to leave his neighborhood and spend significant time each *248day being transported to a distant school suffers an impairment of his liberty and his privacy. Not long ago, James B. Conant wrote that “[a]t the elementary school level the issue seems clear. To send young children day after day to distant schools by bus seems out of the question.” 27 A community may well conclude that the portion of a child’s day spent on a bus might be used more creatively in a classroom, playground, or in some other extracurricular school activity. Decisions such as these, affecting the quality of a child’s daily life, should not lightly be held constitutionally errant.

Up to this point I have focused mainly on the personal interests of parents and children which a community may believe to be best protected by a neighborhood system of schools. But broader considerations lead me to question just as seriously any remedial requirement of extensive student transportation solely to further integration. Any such requirement is certain to fall disproportionately on the school districts of our country, depending on their degree of urbanization, financial resources, and their racial composition. Some districts with little or no biracial population will experience little or no educational disruption, while others, notably in large, biracial metropolitan areas, must at considerable expense undertake extensive transportation to achieve the type of integration frequently being ordered by district courts.28 At a time when public education generally is suffering serious financial malnutrition, the economic burdens of such transportation can be severe, requiring both initial capital outlays and annual operating costs in the millions of dollars.29 And while constitutional requirements have *249often occasioned uneven burdens, never have they touched so sensitive a matter as wide differences in the compulsory transportation requirements for literally hundreds of thousands of school children.

The argument for student transportation also overlooks the fact that the remedy exceeds that which may be necessary to redress the constitutional evil. Let us use Denver as an example. The Denver School Board, by its action and nonaction, may be legally responsible for some of the segregation that exists. But if one assumes a maximum discharge of constitutional duty by the Denver Board over the past decades, the fundamental problem of residential segregation would persist.30 It is, indeed, a novel application of equitable power — not to mention a dubious extension of constitutional doctrine— to require so much greater a degree of forced school integration than would have resulted from purely natural and neutral nonstate causes.

The compulsory transportation of students carries a further infirmity as a constitutional remedy. With most constitutional violations, the major burden of remedial action falls on offending state officials. Public officials who act to infringe personal rights of speech, voting, or religious exercise, for example, are obliged to cease the offending act or practice and, where necessary, institute corrective measures. It is they who bear the brunt of remedial action, though other citizens will to varying de*250grees feel its effects. School authorities responsible for segregation must, at the very minimum, discontinue seg-regative acts. But when the obligation further extends to the transportation of students, the full burden of the affirmative remedial action is borne by children and parents who did not participate in any constitutional violation.

Finally, courts in requiring so far-reaching a remedy as student transportation solely to maximize integration, risk setting in motion unpredictable and unmanageable social consequences. No one can estimate the extent to which dismantling neighborhood education will hasten an exodus to private schools, leaving public school systems the preserve of the disadvantaged of both races. Or guess how much impetus such dismantlement gives the movement from inner city to suburb, and the further geographical separation of the races. Nor do we know to what degree this remedy may cause deterioration of community and parental support of public schools, or divert attention from the paramount goal of quality in education to a perennially divisive debate over who is to be transported where.

The problem addressed in this opinion has perplexed courts, school officials, other public authorities, and students of public education for nearly two decades. The problem, especially since it has focused on the “busing issue,” has profoundly disquieted the public wherever extensive transportation has been ordered. I make no pretense of knowing the best answers. Yet, the issue in this and like cases comes to this Court as one of constitutional law. As to this issue, I have no doubt whatever. There is nothing in the Constitution, its history, or — until recently — in the jurisprudence of this Court that mandates the employment of forced transportation of young and teenage children to achieve a single interest, *251as important as that interest may be. We have strayed, quite far as I view it, from the rationale of Brown I and II, as reiterated in Swann, that courts in fashioning remedies must be “guided by equitable principles” which include the “adjusting and reconciling [of] public and private needs,” Brown II, 349 U. S., at 300.

I urge a return to this rationale. This would result, as emphasized above, in no prohibition on court-ordered student transportation in furtherance of desegregation. But it would require that the legitimate community interests in neighborhood school systems be accorded far greater respect. In the balancing of interests so appropriate to a fair and just equitable decree, transportation orders should be applied with special caution to any proposal as disruptive of family life and interests— and ultimately of education itself — as extensive transportation of elementary-age children solely for desegregation purposes. As a minimum, this Court should not require school boards to engage in the unnecessary transportation away from their neighborhoods of elementary-age children.31 It is at this age level that neighborhood education performs its most vital role. It is with respect to children of tender years that the greatest concern exists for their physical and psychological health. It is also here, at the elementary school, *252that the rights of parents and children are most sharply implicated.32

IV

The existing state of law has failed to shed light and provide guidance on the two issues addressed in this opinion: (i) whether a constitutional rule of uniform, national application should be adopted with respect to our national problem of school desegregation and (ii), if so, whether the ambiguities of Swann, construed to date almost uniformly in favor of extensive transportation, should be redefined to restore a more viable balance among the various interests which are involved. With all deference, it seems to me that the Court today has addressed neither of these issues in a way that will afford adequate guidance to the courts below in this case or lead to a rational, coherent national policy.

The Court has chosen, rather, to adhere to the de facto/ de jure distinction under circumstances, and upon a rationale, which can only lead to increased and inconclusive litigation, and — especially regrettable — to deferment of a nationally consistent judicial position on this subject. There is, of course, state action in every school district in the land. The public schools always have been funded and operated by States and their local subdivisions. It is true that segregated schools, even in the cities of the South, are in large part the product of social and economic factors — and the resulting residential patterns. But there is also not a school district in the United States, with any significant minority school population, in which the school authorities — in one way or the other — have not contributed in some *253measure to the degree of segregation which still prevails. Instead of recognizing the reality of similar, multiple segregative causes in school districts throughout the country, the Court persists in a distinction whose duality operates unfairly on local communities in one section of the country and on minority children in the others.

The second issue relates to the ambiguities of Swann and the judicial disregard of legitimate community and individual interests in framing equitable decrees. In the absence of a more flexible and reasonable standard than that imposed by district courts after Swann, the desegregation which will now be decreed in Denver and other major cities may well involve even more extensive transportation than has been witnessed up to this time.

It is well to remember that the course we are running is a long one and the goal sought in the end — so often overlooked — is the best possible educational opportunity for all children. Communities deserve the freedom and the incentive to turn their attention and energies to this goal of quality education, free from protracted and debilitating battles over court-ordered student transportation. The single most disruptive element in education today is the widespread use of compulsory transportation, especially at elementary grade levels. This has risked distracting and diverting attention from basic educational ends, dividing and embittering communities, and exacerbating, rather than ameliorating, interracial friction and misunderstanding. It is time to return to a more balanced evaluation of the recognized interests of our society in achieving desegregation with other educational and societal interests a community may legitimately assert. This will help assure that integrated school systems will be established and maintained by rational action, will be better understood and supported by parents and children of both races, and will promote the enduring qualities of an integrated society so essential to its genuine success.

Article IX, § 8, of the Colorado Constitution has expressly prohibited any “classification of pupils ... on account of race or color.”

See, e. g., Swann v. Charlotte-Mecklenburg Board of Education, 402 U. S. 1, 23 (1971):

“We do not reach . . . the question whether a showing that school segregation is a consequence of other types of state action, without any discriminatory action by the school authorities, is a constitutional violation requiring remedial action by a school desegregation decree.” The term “state action,” as used herein, thus refers to actions of the appropriate public school authorities.

According to the 1971 Department of Health, Education, and Welfare (HEW) estimate, 43.9% of Negro pupils attended majority white schools in the South as opposed to only 27.8% who attended such schools in the North and West. Fifty-seven percent of all Negro pupils in the North and West attend schools with over 80% minority population as opposed to 32.2% who do so in the South. 118 Cong. Rec. 564 (1972).

The 1971 HEW Enrollment Survey dramatized the segregated character of public school systems in many nonsouthern cities. The percentage of Negro pupils which attended schools more than 80% black was 91.3 in Cleveland, Ohio; 97.8 in Compton, California; 78.1 in Dayton, Ohio; 78.6 in Detroit, Michigan; 95.7 in Gary, *219Indiana; 86.4 in Kansas City, Missouri; 86.6 in Los Angeles, California; 78.8 in Milwaukee, Wisconsin; 91.3 in Newark, New Jersey; 89.8 in St. Louis, Missouri. The full data from the Enrollment Survey may be found in 118 Cong. Rec. 563-566 (1972).

As Senator Ribicoff recognized:

“For years we have fought the battle of integration primarily in the South where the problem was severe. It was a long, arduous fight that deserved to be fought and needed to be won.
“Unfortunately, as the problem of racial isolation has moved north of the Mason-Dixon line, many northerners have bid an evasive farewell to the 100-year struggle for racial equality. Our motto seems to have been 'Do to southerners what you do not want to do to yourself.’
“Good reasons have always been offered, of course, for not moving vigorously ahead in the North as well as the South.
“First, it was that the problem was worse in the South. Then the facts began to show that that was no longer true.
“We then began to hear the de facto-de jure refrain.
“Somehow residential segregation in the North was accidental or de facto and that made it better than the legally supported de jure segregation of the South. It was a hard distinction for black children in totally segregated schools in the North to understand, but it allowed us to avoid the problem.” 118 Cong. Rec. 5455 (1972).

See, e. g., Bradley v. School Board, 345 F. 2d 310, 316 (CA4 1965) (en banc):

“It has been held again and again . . . that the Fourteenth Amendment prohibition is not against segregation as such. . . . A state or a school district offends no constitutional requirement when it grants to all students uniformly an unrestricted freedom of choice as to schools attended, so that each pupil, in effect, assigns himself to the school he wishes to attend.” The case was later vacated and remanded by this Court, which expressed no view on the merits of the desegregation plans submitted. 382 U. S. 103, 105 (1965). See also Bell v. School City of Gary, Ind., 324 F. 2d 209 (CA7 1963); Downs v. Board of Education, 336 F. 2d 988 (CA10 1964); Deal v. Cincinnati Board of Education, 369 F. 2d 55 (CA6 1966).

For a concise history and commentary on the evolution, see generally A. Bickel, The Supreme Court and the Idea of Progress 126-130 (1970).

See also the companion cases in Raney v. Board of Education, 391 U. S. 443 (1968), and Monroe v. Board of Commissioners, 391 U. S. 450 (1968), neither of which involved large urban or metropolitan areas.

As Dr. Karl Taeuber states in his article, Residential Segregation, 213 Scientific American 12, 14 (Aug. 1965):

“No elaborate analysis is necessary to conclude from these figures that a high degree of residential segregation based on race is a universal characteristic of American cities. This segregation is found in the cities of the North and West as well as of the South; in large cities as well as small; in nonindustrial cities as well as industrial; in cities with hundreds of thousands of Negro residents as well as those with only a few thousand, and in cities that are progressive in their employment practices and civil rights policies as well as those that are not.”

In his book, Negroes in Cities (1965), Dr. Taeuber stated that residential segregation exists “regardless of the character of local laws and policies, and regardless of the extent of other forms of segregation or discrimination.” Id., at 36.

A prima facie case of constitutional violation exists when segregation is found to a substantial degree in the schools of a particular district. It is recognized, of course, that this term is relative and provides no precise standards. But circumstances, demographic and otherwise, vary from district to district and hard-and-fast rules should not be formulated. The existence of a substantial percentage of schools populated by students from one race only or predominantly so populated, should trigger the inquiry.

See discussion in Part III, infra, of the remedial action which is appropriate to accomplish desegregation where a court finds that a school board has failed to operate an integrated school system within its district. Plaintiffs must, however, establish the failure of a school board to operate an integrated school system before a court may order desegregative steps by way of remedy. These are two distinct steps which recognize the necessity of proving the constitutional violation before desegregative remedial action can be ordered.

Indeed, if one goes back far enough, it is probable that all racial segregation, wherever occurring and whether or not confined to the schools, has at some time been supported or maintained by government action. In Beckett v. School Board, 308 F. Supp. 1274, 1311-1315 (ED Va. 1969), Judge Hoffman compiled a summary of past public segregative action which included examples from a great majority of States. He concluded that “[o]nly as to the states of Maine, New Hampshire, Vermont, Washington, Nevada, and Hawaii does it appear from this nonexhaustive research that no discriminatory laws appeared on the books at one time or another.” Id., at 1315.

The author continues:

“True, the earlier the policy of segregation was abandoned the less danger there is that it continues to operate covertly, is significantly responsible for present day patterns of residential segregation, or has contributed materially to present community attitudes toward Negro schools. But there is no reason to suppose that 1954 is a universally appropriate dividing line between de jure segregation that may safely be assumed to have spent itself and that which may not. For many remedial purposes, adoption of an arbitrary but easily administrable cutoff point might not be objectionable. But in a situation such as school desegregation, where both the rights asserted and the remedial burdens imposed are of such magnitude, and where the resulting sectional discrimination is passionately resented, it is surely questionable whether such arbitrariness is either politically or morally acceptable.”

See Bickel, supra, n. 7, at 119:

“If a Negro child perceives his separation as discriminatory and invidious, he is not, in a society a hundred years removed from slavery, going to make fine distinctions about the source of a particular separation.”

The Court today does not require, however, a segregative intent with respect to the entire school system, and indeed holds that if such an intent is found with respect to some schools in a system, the burden — normally on the plaintiffs — shifts to the defendant school authorities to prove a negative: namely, that their purposes were benign, ante, at 207-209.

The Court has come a long way since Brown 1. Starting from the unassailable de jure ground of the discriminatory constitutional and statutory provisions of some States, the new formulation — still professing fidelity to the de jure doctrine — is that desegregation will be ordered despite the absence of any segregative laws if: (i) segregated schools in fact exist; (ii) a court finds that they result from *231some action taken with segregative intent by the school board; (iii) such action relates to any “meaningful segment” of the school system; and (iv) the school board cannot prove that its intentions with respect to the remainder of the system were nonsegregative.

As one commentator has expressed it:

“If the courts are indeed prepared to inquire into motive, thorny questions will arise even if one assumes that racial motivation is capable of being proven at trial. What of the case in which one or more members of a school board, but less than a majority, are found to have acted on racial grounds? What if it appears that the school board’s action was prompted by a mixture of motives, including constitutionally innocent ones that alone would have prompted the board to act? What if the members of the school board were not themselves racially inspired but wished to please their constituents, many of whom they knew to be so? If such cases are classified as unconstitutional de jure segregation, there is little point in preserving the de jure-de facto distinction at all. And it may well be that the difference between any of these situations and one in which racial motivation is altogether lacking is too insignificant, from the standpoint of both the moral culpability of the state officials and the impact upon the children involved, to support a difference in constitutional treatment.” Goodman, De Facto School Segregation: A Constitutional and Empirical Analysis, 60 Calif. L. Rev. 275, 284-285 (1972).

See, e. g., Thompson v. School Board of Newport News, 465 F. 2d 83, 87 (1972), where the Fourth Circuit en banc upheld a district court assignment plan where “travel time, varying from a minimum of forty minutes and a maximum of one hour, each way, would be required for busing black students out of the old City and white students into the old City in order to achieve a racial balancing of the district.” This transportation was decreed for children from the third grade up, involving children as young as eight years of age.

In Northcross v. Board of Education of Memphis City Schools, 466 F. 2d 890, 895 (1972), the Sixth Circuit affirmed a district court assignment plan which daily transported 14,000 children with “the maximum time to be spent on the buses by any child [being] 34 minutes . . . presumably each way. But as Judge Weiek noted in dissent the Sixth Circuit instructed the district judge to implement yet further desegregation orders. Plans presently under consideration by that court call for the busing of 39,085 and 61,530 *238children respectively, for undetermined lengths of time. Id., at 895-896.

Petitioners before this Court in Potts v. Flax, No. 72-288, cert. denied, 409 U. S. 1007 (1972), contended that the implementation of the Fifth Circuit’s directive in Flax v. Potts, 464 F. 2d 865 (1972), would require bus rides of up to two hours and 20 minutes each day and a round trip of up to 70 miles. Pet. for Cert. 14. While respondents contended these figures represent an “astounding inflation,” Brief in Opposition 7, transportation of a significant magnitude seems inevitable.

See United States v. Texas Education Agency, 467 F. 2d 848, 883 (CA5 1972) (Bell, J., concurring in an opinion in which seven other judges .joined):

“In our view the remedy which the district court is required to formulate should be formulated within the entire context of the opinion in Swann v. Charlotte-Mecklenburg Board of Education . . . .” (Emphasis added.)

The relevant inquiry is “whether the costs of achieving desegregation in any given situation outweigh the legal, moral, and educational considerations favoring it. . . . It is clear . . . that the Constitution should not be held to require any transportation plan that keeps children on a bus for a substantial part of the day, consumes significant portions of funds otherwise spendable directly on education, or involves a genuine element of danger to the safety of the child.” Comment, School Desegregation After Swann: A Theory of Government Responsibility, 39 U. Chi. L. Rev. 421, 422, 443 (1972).

A useful study of the historical uses and abuses of the neighborhood school concept is M. Weinberg, Race & Place (1967).

In fact, due to racially separate residential patterns that characterize our major urban areas it is quite unrealistic to think of achieving in many cities substantial integration throughout the school district without a degree of student transportation which would have the gravest economic and educational consequences.

As Professor Bickel notes:

“In most of the larger urban areas, demographic conditions are such that no policy that a court can order, and a school board, a city, or even a state has the capability to put into effect, will in fact result in the foreseeable future in racially balanced public schools. Only a reordering of the environment involving economic and social policy on the broadest conceivable front might have an appreciable impact." Bickel, swpra, n. 7, at 132.

Estimates vary. Swann, 402 U. S., at 29, noted that “[eighteen million of the Nation's public school children, approximately 39%, were transported to their schools by bus in 1969-1970 in all parts of the country.” Senator Ribicoff, a thoughtful student of this problem, stated that “[t]wo-thirds of all American children today ride buses to schools for reasons unrelated to integration.” 118 Cong. Rec. 5456 (1972).

Historically, distant transportation was wrongly used to promote segregation. “Negro children were generally considered capable of traveling longer distances to school and without the aid of any vehicle. What was too far for a white child became reasonably near for a Negro child,” Weinberg, supra, n. 20, at 87.

This deplorable history has led some to argue that integrative bus rides are justified as atonement for past segregative trips and that neighborhood education is now but a code word for racial segregation. But misuse of transportation in the past does not imply neighborhood schooling has no valid nonsegregative uses for the present. Nor would wrongful transportation in the past justify detrimental transportation for the children of today.

Some communities had transportation plans in effect at the time of court desegregation orders. See Swann, supra, at 29 n. 11; Davis v. Board of School Commissioners of Mobile County, 402 U. S. 33, 34-35 (1971). Courts have used the presence or absence of existing transportation in a district as one factor in framing and implementing desegregation decrees. United States v. Watson Chapel School District, 446 F. 2d 933, 937 (CA8 1971); Northcross v. Board of Education of Memphis City Schools, 444 F. 2d 1179, 1182-1183 (CA6 1971); Davis v. Board of Education of North Little Rock, 328 F. Supp. 1197, 1203 (ED Ark. 1971). Where a school board is voluntarily engaged in transporting students, a district court is, of course, obligated to insure that such transportation is not undertaken with segregative effect. Where, also, voluntary transportation programs are already in progress, there may be greater justification for court-ordered transportation of students for a comparable time and distance to achieve greater integration.

The term "neighborhood school” should not be supposed to denote solely a walk-in school or one which serves children only in the surrounding blocks. The Court has noted, in a different context, that “[t]he word 'neighborhood' is quite as susceptible of variation as the word ‘locality.’ Both terms are elastic and, dependent upon circumstances, may be equally satisfied by areas measured by rods or by miles.” Connally v. General Construction Co., 269 U. S. 385, 395 (1926). In the school context, “neighborhood” refers to relative proximity, to a preference for a school nearer to, rather than more distant from, home.

I do not imply that the neighborhood concept must be embodied in every school system. But where a school board has chosen it, federal judges should accord it respect in framing remedial decrees.

Slums and Suburbs 29 (1961).

See n. 21, supra.

In Memphis, for example, which has no history of busing students, the minimum transportation plan ordered by the courts will require, in the School Board’s estimate, an initial capital expenditure *249of $1,664,192 for buses plus an annual operating cost of $629,192. The Board estimates that a more extensive transportation program to be considered by the district court will require initial capital investments of $3,924,000 and annual operating costs of $1,783,490. The most drastic transportation plan before the district court requires estimated annual operating costs of from $2,354,220, $2,431,710, or $3,463,100 depending on the Board's transportation arrangements. Northcross v. Board of Education of Memphis City Schools, 466 F. 2d, at 898 (Weick, J., dissenting).

See n. 9, supra.

There may well be advantages in commencing the integrative experiences at an early age, as young children may be less likely than older children and adults to develop an inhibiting racial consciousness. These advantages should be considered as school boards make the various decisions with the view to achieving and preserving an integrated school system. Supra, at 226-227. But in the balancing of all relevant interests, the advantages of an early integrative experience must, and in all fairness should, be weighed against other relevant advantages and disadvantages and in light of the demographic characteristics of the particular community.

While greater transportation of secondary school students might be permitted, even at this level the desire of a community for racially neutral neighborhood schools should command judicial respect. It would ultimately be wisest, where there is no absence of good faith, to permit affected communities to decide this delicate issue of student transportation on their own.