Parents Involved in Community Schools v. Seattle School Dist. No. 1, 551 U.S. 701 (2007)

OPINION OF THE COURT
PARENTS INVOLVED IN COMMUNITY SCHOOLS V.SEATTLE SCHOOL DIST. NO. 1
551 U. S. ____ (2007)

SUPREME COURT OF THE UNITED STATES
NOS. 05-908 AND 05-915

PARENTS INVOLVED IN COMMUNITY SCHOOLS, PETITIONER

05–908   v.

SEATTLE SCHOOL DISTRICT NO. 1 et al.

on writ of certiorari to the united states court of appeals for the ninth circuit

CRYSTAL D. MEREDITH, custodial parent and next friend of JOSHUA RYAN McDONALD, PETITIONER

05–915   v.

JEFFERSON COUNTY BOARD OF EDUCATION et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 28, 2007]

   Chief Justice Roberts announced the judgment of the Court, and delivered the opinion of the Court with respect to Parts I, II, III–A, and III–C, and an opinion with respect to Parts III–B and IV, in which Justices Scalia, Thomas, and Alito join.

   The school districts in these cases voluntarily adopted student assignment plans that rely upon race to determine which public schools certain children may attend. The Seattle school district classifies children as white or nonwhite; the Jefferson County school district as black or “other.” In Seattle, this racial classification is used to allocate slots in oversubscribed high schools. In Jefferson County, it is used to make certain elementary school assignments and to rule on transfer requests. In each case, the school district relies upon an individual student’s race in assigning that student to a particular school, so that the racial balance at the school falls within a predetermined range based on the racial composition of the school district as a whole. Parents of students denied assignment to particular schools under these plans solely because of their race brought suit, contending that allocating children to different public schools on the basis of race violated the Fourteenth Amendment guarantee of equal protection. The Courts of Appeals below upheld the plans. We granted certiorari, and now reverse.

I

   Both cases present the same underlying legal question—whether a public school that had not operated legally segregated schools or has been found to be unitary may choose to classify students by race and rely upon that classification in making school assignments. Although we examine the plans under the same legal framework, the specifics of the two plans, and the circumstances surrounding their adoption, are in some respects quite different.

A

   Seattle School District No. 1 operates 10 regular public high schools. In 1998, it adopted the plan at issue in this case for assigning students to these schools. App. in No. 05–908, pp. 90a–92a.[Footnote 1] The plan allows incoming ninth graders to choose from among any of the district’s high schools, ranking however many schools they wish in order of preference.

   Some schools are more popular than others. If too many students list the same school as their first choice, the district employs a series of “tiebreakers” to determine who will fill the open slots at the oversubscribed school. The first tiebreaker selects for admission students who have a sibling currently enrolled in the chosen school. The next tiebreaker depends upon the racial composition of the particular school and the race of the individual student. In the district’s public schools approximately 41 percent of enrolled students are white; the remaining 59 percent, comprising all other racial groups, are classified by Seattle for assignment purposes as nonwhite. Id., at 38a, 103a.[Footnote 2] If an oversubscribed school is not within 10 percentage points of the district’s overall white/nonwhite racial balance, it is what the district calls “integration positive,” and the district employs a tiebreaker that selects for assignment students whose race “will serve to bring the school into balance.” Id., at 38a. See Parents Involved VII, 426 F. 3d 1162, 1169–1170 (CA9 2005) (en banc).[Footnote 3] If it is still necessary to select students for the school after using the racial tiebreaker, the next tiebreaker is the geographic proximity of the school to the student’s residence. App. in No. 05–908, at 38a.

   Seattle has never operated segregated schools—legally separate schools for students of different races—nor has it ever been subject to court-ordered desegregation. It nonetheless employs the racial tiebreaker in an attempt to address the effects of racially identifiable housing patterns on school assignments. Most white students live in the northern part of Seattle, most students of other racial backgrounds in the southern part. Parents Involved VII, supra, at 1166. Four of Seattle’s high schools are located in the north—Ballard, Nathan Hale, Ingraham, and Roosevelt—and five in the south—Rainier Beach, Cleveland, West Seattle, Chief Sealth, and Franklin. One school—Garfield—is more or less in the center of Seattle. App. in No. 05–908, at 38a–39a, 45a.

   For the 2000–2001 school year, five of these schools were oversubscribed—Ballard, Nathan Hale, Roosevelt, Garfield, and Franklin—so much so that 82 percent of incoming ninth graders ranked one of these schools as their first choice. Id., at 38a. Three of the oversubscribed schools were “integration positive” because the school’s white enrollment the previous school year was greater than 51 percent—Ballard, Nathan Hale, and Roosevelt. Thus, more nonwhite students (107, 27, and 82, respectively) who selected one of these three schools as a top choice received placement at the school than would have been the case had race not been considered, and proximity been the next tiebreaker. Id., at 39a–40a. Franklin was “integration positive” because its nonwhite enrollment the previous school year was greater than 69 percent; 89 more white students were assigned to Franklin by operation of the racial tiebreaker in the 2000–2001 school year than otherwise would have been. Ibid. Garfield was the only oversubscribed school whose composition during the 1999–2000 school year was within the racial guidelines, although in previous years Garfield’s enrollment had been predominantly nonwhite, and the racial tiebreaker had been used to give preference to white students. Id., at 39a.

   Petitioner Parents Involved in Community Schools (Parents Involved) is a nonprofit corporation comprising the parents of children who have been or may be denied assignment to their chosen high school in the district because of their race. The concerns of Parents Involved are illustrated by Jill Kurfirst, who sought to enroll her ninth-grade son, Andy Meeks, in Ballard High School’s special Biotechnology Career Academy. Andy suffered from attention deficit hyperactivity disorder and dyslexia, but had made good progress with hands-on instruction, and his mother and middle school teachers thought that the smaller biotechnology program held the most promise for his continued success. Andy was accepted into this selective program but, because of the racial tiebreaker, was denied assignment to Ballard High School. Id., at 143a–146a, 152a–160a. Parents Involved commenced this suit in the Western District of Washington, alleging that Seattle’s use of race in assignments violated the Equal Protection Clause of the Fourteenth Amendment,[Footnote 4] Title VI of the Civil Rights Act of 1964,[Footnote 5] and the Washington Civil Rights Act.[Footnote 6] Id., at 28a–35a.

   The District Court granted summary judgment to the school district, finding that state law did not bar the district’s use of the racial tiebreaker and that the plan survived strict scrutiny on the federal constitutional claim because it was narrowly tailored to serve a compelling government interest. 137 F. Supp. 2d 1224, 1240 (WD Wash. 2001) (Parents Involved I). The Ninth Circuit initially reversed based on its interpretation of the Washington Civil Rights Act, 285 F. 3d 1236, 1253 (2002) (Parents Involved II), and enjoined the district’s use of the integration tiebreaker, id., at 1257. Upon realizing that the litigation would not be resolved in time for assignment decisions for the 2002–2003 school year, the Ninth Circuit withdrew its opinion, 294 F. 3d 1084 (2002) (Parents Involved III), vacated the injunction, and, pursuant to Wash. Rev. Code §2.60.020 (2006), certified the state-law question to the Washington Supreme Court, 294 F. 3d 1085, 1087 (2002) (Parents Involved IV).

   The Washington Supreme Court determined that the State Civil Rights Act bars only preferential treatment programs “where race or gender is used by government to select a less qualified applicant over a more qualified applicant,” and not “[p]rograms which are racially neutral, such as the [district’s] open choice plan.” Parents Involved in Community Schools v. Seattle School Dist., No. 1, 149 Wash. 2d 660, 689–690, 663, 72 P. 3d 151, 166, 153 (2003) (en banc) (Parents Involved V). The state court returned the case to the Ninth Circuit for further proceedings. Id., at 690, 72 P. 3d, at 167.

   A panel of the Ninth Circuit then again reversed the District Court, this time ruling on the federal constitutional question. Parents Involved VI, 377 F. 3d 949 (2004). The panel determined that while achieving racial diversity and avoiding racial isolation are compelling government interests, id., at 964, Seattle’s use of the racial tiebreaker was not narrowly tailored to achieve these interests, id., at 980. The Ninth Circuit granted rehearing en banc, 395 F. 3d 1168 (2005), and overruled the panel decision, affirming the District Court’s determination that Seattle’s plan was narrowly tailored to serve a compelling government interest, Parents Involved VII, 426 F. 3d, at 1192–1193. We granted certiorari. 547 U. S. __ (2006).

B

   Jefferson County Public Schools operates the public school system in metropolitan Louisville, Kentucky. In 1973 a federal court found that Jefferson County had maintained a segregated school system, Newburg Area Council, Inc. v. Board of Ed. of Jefferson Cty., 489 F. 2d 925, 932 (CA6), vacated and remanded, 418 U. S. 918, reinstated with modifications, 510 F. 2d 1358, 1359 (CA6 1974), and in 1975 the District Court entered a desegregation decree. See Hampton v. Jefferson Cty. Bd. of Ed., 72 F. Supp. 2d 753, 762–764 (WD Ky. 1999). Jefferson County operated under this decree until 2000, when the District Court dissolved the decree after finding that the district had achieved unitary status by eliminating “[t]o the greatest extent practicable” the vestiges of its prior policy of segregation. Hampton v. Jefferson Cty. Bd. of Ed., 102 F. Supp. 2d 358, 360 (2000). See Board of Ed. of Oklahoma City Public Schools v. Dowell, 498 U. S. 237, 249–250 (1991); Green v. School Bd. of New Kent Cty., 391 U. S. 430, 435–436 (1968).

   In 2001, after the decree had been dissolved, Jefferson County adopted the voluntary student assignment plan at issue in this case. App. in No. 05–915, p. 77. Approximately 34 percent of the district’s 97,000 students are black; most of the remaining 66 percent are white. McFarland v. Jefferson Cty. Public Schools, 330 F. Supp. 2d 834, 839–840, and n. 6 (WD Ky. 2004) (McFarland I). The plan requires all nonmagnet schools to maintain a minimum black enrollment of 15 percent, and a maximum black enrollment of 50 percent. App. in No. 05–915, at 81; McFarland I, supra, at 842.

   At the elementary school level, based on his or her address, each student is designated a “resides” school to which students within a specific geographic area are assigned; elementary resides schools are “grouped into clusters in order to facilitate integration.” App. in No. 05–915, at 82. The district assigns students to nonmagnet schools in one of two ways: Parents of kindergartners, first-graders, and students new to the district may submit an application indicating a first and second choice among the schools within their cluster; students who do not submit such an application are assigned within the cluster by the district. “Decisions to assign students to schools within each cluster are based on available space within the schools and the racial guidelines in the District’s current student assignment plan.” Id., at 38. If a school has reached the “extremes of the racial guidelines,” a student whose race would contribute to the school’s racial imbalance will not be assigned there. Id., at 38–39, 82. After assignment, students at all grade levels are permitted to apply to transfer between nonmagnet schools in the district. Transfers may be requested for any number of reasons, and may be denied because of lack of available space or on the basis of the racial guidelines. Id., at 43.[Footnote 7]

   When petitioner Crystal Meredith moved into the school district in August 2002, she sought to enroll her son, Joshua McDonald, in kindergarten for the 2002–2003 school year. His resides school was only a mile from his new home, but it had no available space—assignments had been made in May, and the class was full. Jefferson County assigned Joshua to another elementary school in his cluster, Young Elementary. This school was 10 miles from home, and Meredith sought to transfer Joshua to a school in a different cluster, Bloom Elementary, which—like his resides school—was only a mile from home. See Tr. in McFarland I, pp. 1–49 through 1–54 (Dec. 8, 2003). Space was available at Bloom, and intercluster transfers are allowed, but Joshua’s transfer was nonetheless denied because, in the words of Jefferson County, “[t]he transfer would have an adverse effect on desegregation compliance” of Young. App. in No. 05–915, at 97.[Footnote 8]

   Meredith brought suit in the Western District of Kentucky, alleging violations of the Equal Protection Clause of the Fourteenth Amendment. The District Court found that Jefferson County had asserted a compelling interest in maintaining racially diverse schools, and that the assignment plan was (in all relevant respects) narrowly tailored to serve that compelling interest. McFarland I, supra, at 837.[Footnote 9] The Sixth Circuit affirmed in a per curiam opinion relying upon the reasoning of the District Court, concluding that a written opinion “would serve no useful purpose.” McFarland v. Jefferson Cty. Public Schools, 416 F. 3d 513, 514 (2005) (McFarland II). We granted certiorari. 547 U. S. __ (2006).

II

   As a threshold matter, we must assure ourselves of our jurisdiction. Seattle argues that Parents Involved lacks standing because none of its current members can claim an imminent injury. Even if the district maintains the current plan and reinstitutes the racial tiebreaker, Seattle argues, Parents Involved members will only be affected if their children seek to enroll in a Seattle public high school and choose an oversubscribed school that is integration positive—too speculative a harm to maintain standing. Brief for Respondents in No. 05–908, pp. 16–17.

   This argument is unavailing. The group’s members have children in the district’s elementary, middle, and high schools, App. in No. 05–908, at 299a–301a; Affidavit of Kathleen Brose Pursuant to this Court’s Rule 32.3 (Lodging of Petitioner Parents Involved), and the complaint sought declaratory and injunctive relief on behalf of Parents Involved members whose elementary and middle school children may be “denied admission to the high schools of their choice when they apply for those schools in the future,” App. in No. 05–908, at 30a. The fact that it is possible that children of group members will not be denied admission to a school based on their race—because they choose an undersubscribed school or an oversubscribed school in which their race is an advantage—does not eliminate the injury claimed. Moreover, Parents Involved also asserted an interest in not being “forced to compete for seats at certain high schools in a system that uses race as a deciding factor in many of its admissions decisions.” Ibid. As we have held, one form of injury under the Equal Protection Clause is being forced to compete in a race-based system that may prejudice the plaintiff, Adarand Constructors, Inc. v. Peńa, 515 U. S. 200, 211 (1995); Northeastern Fla. Chapter, Associated Gen. Contractors of America v. Jacksonville, 508 U. S. 656, 666 (1993), an injury that the members of Parents Involved can validly claim on behalf of their children.

   In challenging standing, Seattle also notes that it has ceased using the racial tiebreaker pending the outcome of this litigation. Brief for Respondents in No. 05–908, at 16–17. But the district vigorously defends the constitutionality of its race-based program, and nowhere suggests that if this litigation is resolved in its favor it will not resume using race to assign students. Voluntary cessation does not moot a case or controversy unless “subsequent events ma[ke] it absolutely clear that the allegedly wrongful behavior could not reasonably be expected to recur,” Friends of Earth, Inc. v. Laidlaw Environmental Services (TOC), Inc., 528 U. S. 167, 189 (2000) (quoting United States v. Concentrated Phosphate Export Assn., Inc., 393 U. S. 199, 203 (1968) (internal quotation marks omitted)), a heavy burden that Seattle has clearly not met.

   Jefferson County does not challenge our jurisdiction, Tr. of Oral Arg. in No. 05–915, p. 48, but we are nonetheless obliged to ensure that it exists, Arbaugh v. Y & H Corp., 546 U. S. 500, 514 (2006). Although apparently Joshua has now been granted a transfer to Bloom, the school to which transfer was denied under the racial guidelines, Tr. of Oral Arg. in No. 05–915, at 45, the racial guidelines apply at all grade levels. Upon Joshua’s enrollment in middle school, he may again be subject to assignment based on his race. In addition, Meredith sought damages in her complaint, which is sufficient to preserve our ability to consider the question. Los Angeles v. Lyons, 461 U. S. 95, 109 (1983).

III

A

   It is well established that when the government distributes burdens or benefits on the basis of individual racial classifications, that action is reviewed under strict scrutiny. Johnson v. California, 543 U. S. 499, 505–506 (2005); Grutter v. Bollinger, 539 U. S. 306, 326 (2003); Adarand, supra, at 224. As the Court recently reaffirmed, “ ‘racial classifications are simply too pernicious to permit any but the most exact connection between justification and classification.’ ” Gratz v. Bollinger, 539 U. S. 244, 270 (2003) (quoting Fullilove v. Klutznick, 448 U. S. 448, 537 (1980) (Stevens, J., dissenting); brackets omitted). In order to satisfy this searching standard of review, the school districts must demonstrate that the use of individual racial classifications in the assignment plans here under review is “narrowly tailored” to achieve a “compelling” government interest. Adarand, supra, at 227.

   Without attempting in these cases to set forth all the interests a school district might assert, it suffices to note that our prior cases, in evaluating the use of racial classifications in the school context, have recognized two interests that qualify as compelling. The first is the compelling interest of remedying the effects of past intentional discrimination. See Freeman v. Pitts, 503 U. S. 467, 494 (1992). Yet the Seattle public schools have not shown that they were ever segregated by law, and were not subject to court-ordered desegregation decrees. The Jefferson County public schools were previously segregated by law and were subject to a desegregation decree entered in 1975. In 2000, the District Court that entered that decree dissolved it, finding that Jefferson County had “eliminated the vestiges associated with the former policy of segregation and its pernicious effects,” and thus had achieved “unitary” status. Hampton, 102 F. Supp. 2d, at 360. Jefferson County accordingly does not rely upon an interest in remedying the effects of past intentional discrimination in defending its present use of race in assigning students. See Tr. of Oral Arg. in No. 05–915, at 38.

   Nor could it. We have emphasized that the harm being remedied by mandatory desegregation plans is the harm that is traceable to segregation, and that “the Constitution is not violated by racial imbalance in the schools, without more.” Milliken v. Bradley, 433 U. S. 267, 280, n. 14 (1977). See also Freeman, supra, at 495–496; Dowell, 498 U. S., at 248; Milliken v. Bradley, 418 U. S. 717, 746 (1974). Once Jefferson County achieved unitary status, it had remedied the constitutional wrong that allowed race-based assignments. Any continued use of race must be justified on some other basis.[Footnote 10]

   The second government interest we have recognized as compelling for purposes of strict scrutiny is the interest in diversity in higher education upheld in Grutter, 539 U. S., at 328. The specific interest found compelling in Grutter was student body diversity “in the context of higher education.” Ibid. The diversity interest was not focused on race alone but encompassed “all factors that may contribute to student body diversity.” Id., at 337. We described the various types of diversity that the law school sought:

“[The law school’s] policy makes clear there are many possible bases for diversity admissions, and provides examples of admittees who have lived or traveled widely abroad, are fluent in several languages, have overcome personal adversity and family hardship, have exceptional records of extensive community service, and have had successful careers in other fields.” Id., at 338 (brackets and internal quotation marks omitted).

The Court quoted the articulation of diversity from Justice Powell’s opinion in Regents of the University of California v. Bakke, 438 U. S. 265 (1978), noting that “it is not an interest in simple ethnic diversity, in which a specified percentage of the student body is in effect guaranteed to be members of selected ethnic groups, that can justify the use of race.” Grutter, supra, at 324–325 (citing and quoting Bakke, supra, at 314–315 (opinion of Powell, J.); brackets and internal quotation marks omitted). Instead, what was upheld in Grutter was consideration of “a far broader array of qualifications and characteristics of which racial or ethnic origin is but a single though important element.” 539 U. S., at 325 (quoting Bakke, supra, at 315 (opinion of Powell, J.); internal quotation marks omitted).

   The entire gist of the analysis in Grutter was that the admissions program at issue there focused on each applicant as an individual, and not simply as a member of a particular racial group. The classification of applicants by race upheld in Grutter was only as part of a “highly individualized, holistic review,” 539 U. S., at 337. As the Court explained, “[t]he importance of this individualized consideration in the context of a race-conscious admissions program is paramount.” Ibid. The point of the narrow tailoring analysis in which the Grutter Court engaged was to ensure that the use of racial classifications was indeed part of a broader assessment of diversity, and not simply an effort to achieve racial balance, which the Court explained would be “patently unconstitutional.” Id., at 330.

   In the present cases, by contrast, race is not considered as part of a broader effort to achieve “exposure to widely diverse people, cultures, ideas, and viewpoints,” ibid.; race, for some students, is determinative standing alone. The districts argue that other factors, such as student preferences, affect assignment decisions under their plans, but under each plan when race comes into play, it is decisive by itself. It is not simply one factor weighed with others in reaching a decision, as in Grutter; it is the factor. Like the University of Michigan undergraduate plan struck down in Gratz, 539 U. S., at 275, the plans here “do not provide for a meaningful individualized review of applicants” but instead rely on racial classifications in a “nonindividualized, mechanical” way. Id., at 276, 280 (O’Connor, J., concurring).

   Even when it comes to race, the plans here employ only a limited notion of diversity, viewing race exclusively in white/nonwhite terms in Seattle and black/“other” terms in Jefferson County.[Footnote 11] But see Metro Broadcasting, Inc. v. FCC, 497 U. S. 547, 610 (1990) (“We are a Nation not of black and white alone, but one teeming with divergent communities knitted together with various traditions and carried forth, above all, by individuals”) (O’Connor, J., dissenting). The Seattle “Board Statement Reaffirming Diversity Rationale” speaks of the “inherent educational value” in “[p]roviding students the opportunity to attend schools with diverse student enrollment,” App. in No. 05–908, at 128a, 129a. But under the Seattle plan, a school with 50 percent Asian-American students and 50 percent white students but no African-American, Native-American, or Latino students would qualify as balanced, while a school with 30 percent Asian-American, 25 percent African-American, 25 percent Latino, and 20 percent white students would not. It is hard to understand how a plan that could allow these results can be viewed as being concerned with achieving enrollment that is “ ‘broadly diverse,’ ” Grutter, supra, at 329.

   Prior to Grutter, the courts of appeals rejected as unconstitutional attempts to implement race-based assignment plans—such as the plans at issue here—in primary and secondary schools. See, e.g., Eisenberg v. Montgomery Cty. Public Schools, 197 F. 3d 123, 133 (CA4 1999); Tuttle v. Arlington Cty. School Bd., 195 F. 3d 698, 701 (CA4 1999); Wessman v. Gittens, 160 F. 3d 790, 809 (CA1 1998). See also Ho v. San Francisco Unified School Dist., 147 F. 3d 854, 865 (CA9 1998). After Grutter, however, the two Courts of Appeals in these cases, and one other, found that race-based assignments were permissible at the elementary and secondary level, largely in reliance on that case. See Parents Involved VII, 426 F. 3d, at 1166; McFarland II, 416 F. 3d, at 514; Comfort v. Lynn School Comm., 418 F. 3d 1, 13 (CA1 2005).

   In upholding the admissions plan in Grutter, though, this Court relied upon considerations unique to institutions of higher education, noting that in light of “the expansive freedoms of speech and thought associated with the university environment, universities occupy a special niche in our constitutional tradition.” 539 U. S., at 329. See also Bakke, supra, at 312, 313 (opinion of Powell, J.). The Court explained that “[c]ontext matters” in applying strict scrutiny, and repeatedly noted that it was addressing the use of race “in the context of higher education.” Grutter, supra, at 327, 328, 334. The Court in Grutter expressly articulated key limitations on its holding—defining a specific type of broad-based diversity and noting the unique context of higher education—but these limitations were largely disregarded by the lower courts in extending Grutter to uphold race-based assignments in elementary and secondary schools. The present cases are not governed by Grutter.

B

   Perhaps recognizing that reliance on Grutter cannot sustain their plans, both school districts assert additional interests, distinct from the interest upheld in Grutter, to justify their race-based assignments. In briefing and argument before this Court, Seattle contends that its use of race helps to reduce racial concentration in schools and to ensure that racially concentrated housing patterns do not prevent nonwhite students from having access to the most desirable schools. Brief for Respondents in No. 05–908, at 19. Jefferson County has articulated a similar goal, phrasing its interest in terms of educating its students “in a racially integrated environment.” App. in No. 05–915, at 22.[Footnote 12] Each school district argues that educational and broader socialization benefits flow from a racially diverse learning environment, and each contends that because the diversity they seek is racial diversity—not the broader diversity at issue in Grutter—it makes sense to promote that interest directly by relying on race alone.

   The parties and their amici dispute whether racial diversity in schools in fact has a marked impact on test scores and other objective yardsticks or achieves intangible socialization benefits. The debate is not one we need to resolve, however, because it is clear that the racial classifications employed by the districts are not narrowly tailored to the goal of achieving the educational and social benefits asserted to flow from racial diversity. In design and operation, the plans are directed only to racial balance, pure and simple, an objective this Court has repeatedly condemned as illegitimate.

   The plans are tied to each district’s specific racial demographics, rather than to any pedagogic concept of the level of diversity needed to obtain the asserted educational benefits. In Seattle, the district seeks white enrollment of between 31 and 51 percent (within 10 percent of “the district white average” of 41 percent), and nonwhite enrollment of between 49 and 69 percent (within 10 percent of “the district minority average” of 59 percent). App. in No. 05–908, at 103a. In Jefferson County, by contrast, the district seeks black enrollment of no less than 15 or more than 50 percent, a range designed to be “equally above and below Black student enrollment systemwide,” McFarland I, 330 F. Supp. 2d, at 842, based on the objective of achieving at “all schools … an African-American enrollment equivalent to the average district-wide African-American enrollment” of 34 percent. App. in No. 05–915, at 81. In Seattle, then, the benefits of racial diversity require enrollment of at least 31 percent white students; in Jefferson County, at least 50 percent. There must be at least 15 percent nonwhite students under Jefferson County’s plan; in Seattle, more than three times that figure. This comparison makes clear that the racial demographics in each district—whatever they happen to be—drive the required “diversity” numbers. The plans here are not tailored to achieving a degree of diversity necessary to realize the asserted educational benefits; instead the plans are tailored, in the words of Seattle’s Manager of Enrollment Planning, Technical Support, and Demographics, to “the goal established by the school board of attain-ing a level of diversity within the schools that approximates the district’s overall demographics.” App. in No. 05–908, at 42a.

   The districts offer no evidence that the level of racial diversity necessary to achieve the asserted educational benefits happens to coincide with the racial demographics of the respective school districts—or rather the white/nonwhite or black/“other” balance of the districts, since that is the only diversity addressed by the plans. Indeed, in its brief Seattle simply assumes that the educational benefits track the racial breakdown of the district. See Brief for Respondents in No. 05–908, at 36 (“For Seattle, ‘racial balance’ is clearly not an end in itself but rather a measure of the extent to which the educational goals the plan was designed to foster are likely to be achieved”). When asked for “a range of percentage that would be diverse,” however, Seattle’s expert said it was important to have “sufficient numbers so as to avoid students feeling any kind of specter of exceptionality.” App. in No. 05–908, at 276a. The district did not attempt to defend the proposition that anything outside its range posed the “specter of exceptionality.” Nor did it demonstrate in any way how the educational and social benefits of racial diversity or avoidance of racial isolation are more likely to be achieved at a school that is 50 percent white and 50 percent Asian-American, which would qualify as diverse under Seattle’s plan, than at a school that is 30 percent Asian-American, 25 percent African-American, 25 percent Latino, and 20 percent white, which under Seattle’s definition would be racially concentrated.

   Similarly, Jefferson County’s expert referred to the importance of having “at least 20 percent” minority group representation for the group “to be visible enough to make a difference,” and noted that “small isolated minority groups in a school are not likely to have a strong effect on the overall school.” App. in No. 05–915, at 159, 147. The Jefferson County plan, however, is based on a goal of replicating at each school “an African-American enrollment equivalent to the average district-wide African-American enrollment.” Id., at 81. Joshua McDonald’s requested transfer was denied because his race was listed as “other” rather than black, and allowing the transfer would have had an adverse effect on the racial guideline compliance of Young Elementary, the school he sought to leave. Id., at 21. At the time, however, Young Elementary was 46.8 percent black. Id., at 73. The transfer might have had an adverse effect on the effort to approach district-wide racial proportionality at Young, but it had nothing to do with preventing either the black or “other” group from becoming “small” or “isolated” at Young.

   In fact, in each case the extreme measure of relying on race in assignments is unnecessary to achieve the stated goals, even as defined by the districts. For example, at Franklin High School in Seattle, the racial tiebreaker was applied because nonwhite enrollment exceeded 69 percent, and resulted in an incoming ninth-grade class in 2000–2001 that was 30.3 percent Asian-American, 21.9 percent African-American, 6.8 percent Latino, 0.5 percent Native-American, and 40.5 percent Caucasian. Without the racial tiebreaker, the class would have been 39.6 percent Asian-American, 30.2 percent African-American, 8.3 percent Latino, 1.1 percent Native-American, and 20.8 percent Caucasian. See App. in No. 05–908, at 308a. When the actual racial breakdown is considered, enrolling students without regard to their race yields a substantially diverse student body under any definition of diversity.[Footnote 13]

   In Grutter, the number of minority students the school sought to admit was an undefined “meaningful number” necessary to achieve a genuinely diverse student body. 539 U. S., at 316, 335–336. Although the matter was the subject of disagreement on the Court, see id., at 346–347 (Scalia, J., concurring in part and dissenting in part); id., at 382–383 (Rehnquist, C. J., dissenting); id., at 388–392 (Kennedy, J., dissenting), the majority concluded that the law school did not count back from its applicant pool to arrive at the “meaningful number” it regarded as necessary to diversify its student body. Id., at 335–336. Here the racial balance the districts seek is a defined range set solely by reference to the demographics of the respective school districts.

   This working backward to achieve a particular type of racial balance, rather than working forward from some demonstration of the level of diversity that provides the purported benefits, is a fatal flaw under our existing precedent. We have many times over reaffirmed that “[r]acial balance is not to be achieved for its own sake.” Freeman, 503 U. S., at 494. See also Richmond v. J. A. Croson Co., 488 U. S. 469, 507 (1989); Bakke, 438 U. S., at 307 (opinion of Powell, J.) (“If petitioner’s purpose is to assure within its student body some specified percentage of a particular group merely because of its race or ethnic origin, such a preferential purpose must be rejected … as facially invalid”). Grutter itself reiterated that “outright racial balancing” is “patently unconstitutional.” 539 U. S., at 330.

   Accepting racial balancing as a compelling state interest would justify the imposition of racial proportionality throughout American society, contrary to our repeated recognition that “[a]t the heart of the Constitution’s guarantee of equal protection lies the simple command that the Government must treat citizens as individuals, not as simply components of a racial, religious, sexual or national class.” Miller v. Johnson, 515 U. S. 900, 911 (1995) (quoting Metro Broadcasting, 497 U. S., at 602 (O’Connor, J., dissenting); internal quotation marks omitted).[Footnote 14] Allowing racial balancing as a compelling end in itself would “effectively assur[e] that race will always be relevant in American life, and that the ‘ultimate goal’ of ‘eliminating entirely from governmental decisionmaking such irrelevant factors as a human being’s race’ will never be achieved.” Croson, supra, at 495 (plurality opinion of O’Connor, J.) (quoting Wygant v. Jackson Bd. of Ed., 476 U. S. 267, 320 (1986) (Stevens, J., dissenting), in turn quoting Fullilove, 448 U. S., at 547 (Stevens, J., dissenting); brackets and citation omitted). An interest “linked to nothing other than proportional representation of various races … would support indefinite use of racial classifications, employed first to obtain the appropriate mixture of racial views and then to ensure that the [program] continues to reflect that mixture.” Metro Broadcasting, supra, at 614 (O’Connor, J., dissenting).

   The validity of our concern that racial balancing has “no logical stopping point,” Croson, supra, at 498 (quoting Wygant, supra, at 275 (plurality opinion); internal quotation marks omitted); see also Grutter, supra, at 343, is demonstrated here by the degree to which the districts tie their racial guidelines to their demographics. As the districts’ demographics shift, so too will their definition of racial diversity. See App. in No. 05–908, at 103a (describing application of racial tiebreaker based on “current white percentage” of 41 percent and “current minority percentage” of 59 percent (emphasis added)).

   The Ninth Circuit below stated that it “share[d] in the hope” expressed in Grutter that in 25 years racial preferences would no longer be necessary to further the interest identified in that case. Parents Involved VII, 426 F. 3d, at 1192. But in Seattle the plans are defended as necessary to address the consequences of racially identifiable housing patterns. The sweep of the mandate claimed by the district is contrary to our rulings that remedying past societal discrimination does not justify race-conscious government action. See, e.g., Shaw v. Hunt, 517 U. S. 899, 909–910 (1996) (“[A]n effort to alleviate the effects of societal discrimination is not a compelling interest”); Croson, supra, at 498–499; Wygant, 476 U. S., at 276 (plurality opinion) (“Societal discrimination, without more, is too amorphous a basis for imposing a racially classified remedy”); id., at 288 (O’Connor, J., concurring in part and concurring in judgment) (“[A] governmental agency’s interest in remedying ‘societal’ discrimination, that is, discrimination not traceable to its own actions, cannot be deemed sufficiently compelling to pass constitutional muster”).

   The principle that racial balancing is not permitted is one of substance, not semantics. Racial balancing is not transformed from “patently unconstitutional” to a compelling state interest simply by relabeling it “racial diversity.” While the school districts use various verbal formulations to describe the interest they seek to promote—racial diversity, avoidance of racial isolation, racial integration—they offer no definition of the interest that suggests it differs from racial balance. See, e.g., App. in No. 05–908, at 257a (“Q. What’s your understanding of when a school suffers from racial isolation? A. I don’t have a definition for that”); id., at 228a–229a (“I don’t think we’ve ever sat down and said, ‘Define racially concentrated school exactly on point in quantitative terms.’ I don’t think we’ve ever had that conversation”); Tr. in McFarland I, at 1–90 (Dec. 8, 2003) (“Q. How does the Jefferson County School Board define diversity … ?” “A. Well, we want to have the schools that make up the percentage of students of the population”).

   Jefferson County phrases its interest as “racial integration,” but integration certainly does not require the sort of racial proportionality reflected in its plan. Even in the context of mandatory desegregation, we have stressed that racial proportionality is not required, see Milliken, 433 U. S., at 280, n. 14 (“[A desegregation] order contemplating the substantive constitutional right [to a] particular degree of racial balance or mixing is … infirm as a matter of law” (internal quotation marks omitted)); Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U. S. 1, 24 (1971) (“The constitutional command to desegregate schools does not mean that every school in every community must always reflect the racial composition of the school system as a whole”), and here Jefferson County has already been found to have eliminated the vestiges of its prior segregated school system.

   The en banc Ninth Circuit declared that “when a racially diverse school system is the goal (or racial concentration or isolation is the problem), there is no more effective means than a consideration of race to achieve the solution.” Parents Involved VII, supra, at 1191. For the foregoing reasons, this conclusory argument cannot sustain the plans. However closely related race-based assignments may be to achieving racial balance, that itself cannot be the goal, whether labeled “racial diversity” or anything else. To the extent the objective is sufficient diversity so that students see fellow students as individuals rather than solely as members of a racial group, using means that treat students solely as members of a racial group is fundamentally at cross-purposes with that end.

C

   The districts assert, as they must, that the way in which they have employed individual racial classifications is necessary to achieve their stated ends. The minimal effect these classifications have on student assignments, however, suggests that other means would be effective. Seattle’s racial tiebreaker results, in the end, only in shifting a small number of students between schools. Approximately 307 student assignments were affected by the racial tiebreaker in 2000–2001; the district was able to track the enrollment status of 293 of these students. App. in No. 05–908, at 162a. Of these, 209 were assigned to a school that was one of their choices, 87 of whom were assigned to the same school to which they would have been assigned without the racial tiebreaker. Eighty-four students were assigned to schools that they did not list as a choice, but 29 of those students would have been assigned to their respective school without the racial tiebreaker, and 3 were able to attend one of the oversubscribed schools due to waitlist and capacity adjustments. Id., at 162a–163a. In over one-third of the assignments affected by the racial tiebreaker, then, the use of race in the end made no difference, and the district could identify only 52 students who were ultimately affected adversely by the racial tiebreaker in that it resulted in assignment to a school they had not listed as a preference and to which they would not otherwise have been assigned.

   As the panel majority in Parents Involved VI concluded:

“[T]he tiebreaker’s annual effect is thus merely to shuffle a few handfuls of different minority students between a few schools—about a dozen additional Latinos into Ballard, a dozen black students into Nathan Hale, perhaps two dozen Asians into Roosevelt, and so on. The District has not met its burden of proving these marginal changes … outweigh the cost of subjecting hundreds of students to disparate treatment based solely upon the color of their skin.” 377 F. 3d, at 984–985 (footnote omitted).

   Similarly, Jefferson County’s use of racial classifications has only a minimal effect on the assignment of students. Elementary school students are assigned to their first- or second-choice school 95 percent of the time, and transfers, which account for roughly 5 percent of assignments, are only denied 35 percent of the time—and presumably an even smaller percentage are denied on the basis of the racial guidelines, given that other factors may lead to a denial. McFarland I, 330 F. Supp. 2d, at 844–845, nn. 16, 18. Jefferson County estimates that the racial guidelines account for only 3 percent of assignments. Brief in Opposition in No. 05–915, p. 7, n. 4; Tr. of Oral Arg. in No. 05–915, at 46. As Jefferson County explains, “the racial guidelines have minimal impact in this process, because they ‘mostly influence student assignment in subtle and indirect ways.’ ” Brief for Respondents in No. 05–915, pp. 8–9.

   While we do not suggest that greater use of race would be preferable, the minimal impact of the districts’ racial classifications on school enrollment casts doubt on the necessity of using racial classifications. In Grutter, the consideration of race was viewed as indispensable in more than tripling minority representation at the law school—from 4 to 14.5 percent. See 539 U. S., at 320. Here the most Jefferson County itself claims is that “because the guidelines provide a firm definition of the Board’s goal of racially integrated schools, they ‘provide administrators with the authority to facilitate, negotiate and collaborate with principals and staff to maintain schools within the 15–50% range.’ ” Brief in Opposition in No. 05–915, at 7 (quoting McFarland I, supra, at 842). Classifying and assigning schoolchildren according to a binary conception of race is an extreme approach in light of our precedents and our Nation’s history of using race in public schools, and requires more than such an amorphous end to justify it.

   The districts have also failed to show that they considered methods other than explicit racial classifications to achieve their stated goals. Narrow tailoring requires “serious, good faith consideration of workable race-neutral alternatives,” Grutter, supra, at 339, and yet in Seattle several alternative assignment plans—many of which would not have used express racial classifications—were rejected with little or no consideration. See, e.g., App. in No. 05–908, at 224a–225a, 253a–259a, 307a. Jefferson County has failed to present any evidence that it considered alternatives, even though the district already claims that its goals are achieved primarily through means other than the racial classifications. Brief for Respondents in No. 05–915, at 8–9. Compare Croson, 488 U. S., at 519 (Kennedy, J., concurring in part and concurring in judgment) (racial classifications permitted only “as a last resort”).

IV

   Justice Breyer’s dissent takes a different approach to these cases, one that fails to ground the result it would reach in law. Instead, it selectively relies on inapplicable precedent and even dicta while dismissing contrary holdings, alters and misapplies our well-established legal framework for assessing equal protection challenges to express racial classifications, and greatly exaggerates the consequences of today’s decision.

   To begin with, Justice Breyer seeks to justify the plans at issue under our precedents recognizing the compelling interest in remedying past intentional discrimination. See post, at 18–24. Not even the school districts go this far, and for good reason. The distinction between segregation by state action and racial imbalance caused by other factors has been central to our jurisprudence in this area for generations. See, e.g., Milliken, 433 U. S., at 280, n. 14; Freeman, 503 U. S., at 495–496 (“Where resegregation is a product not of state action but of private choices, it does not have constitutional implications”). The dissent elides this distinction between de jure and de facto segregation, casually intimates that Seattle’s school attendance patterns reflect illegal segregation, post, at 5, 18, 23,[Footnote 15] and fails to credit the judicial determination—under the most rigorous standard—that Jefferson County had eliminated the vestiges of prior segregation. The dissent thus alters in fundamental ways not only the facts presented here but the established law.

   Justice Breyer’s reliance on McDaniel v. Barresi, 402 U. S. 39 (1971), post, at 23–24, 29–30, highlights how far removed the discussion in the dissent is from the question actually presented in these cases. McDaniel concerned a Georgia school system that had been segregated by law. There was no doubt that the county had operated a “dual school system,” McDaniel, supra, at 41, and no one questions that the obligation to disestablish a school system segregated by law can include race-conscious remedies—whether or not a court had issued an order to that effect. See supra, at 12. The present cases are before us, however, because the Seattle school district was never segregated by law, and the Jefferson County district has been found to be unitary, having eliminated the vestiges of its prior dual status. The justification for race-conscious remedies in McDaniel is therefore not applicable here. The dissent’s persistent refusal to accept this distinction—its insistence on viewing the racial classifications here as if they were just like the ones in McDaniel, “devised to overcome a history of segregated public schools,” post, at 47—explains its inability to understand why the remedial justification for racial classifications cannot decide these cases.

   Justice Breyer’s dissent next relies heavily on dicta from Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U. S., at 16—far more heavily than the school districts themselves. Compare post, at 3, 22–28, with Brief for Respondents in No. 05–908, at 19–20; Brief for Respondents in No. 05–915, at 31. The dissent acknowledges that the two-sentence discussion in Swann was pure dicta, post, at 22, but nonetheless asserts that it demonstrates a “basic principle of constitutional law” that provides “authoritative legal guidance.” Post, at 22, 30. Initially, as the Court explained just last Term, “we are not bound to follow our dicta in a prior case in which the point now at issue was not fully debated.” Central Va. Community College v. Katz, 546 U. S. 356, 363 (2006). That is particularly true given that, when Swann was decided, this Court had not yet confirmed that strict scrutiny applies to racial classifications like those before us. See n. 16, infra. There is nothing “technical” or “theoretical,” post, at 30, about our approach to such dicta. See, e.g., Cohens v. Virginia, 6 Wheat. 264, 399–400 (1821) (Marshall, C. J.) (explaining why dicta is not binding).

   Justice Breyer would not only put such extraordinary weight on admitted dicta, but relies on the statement for something it does not remotely say. Swann addresses only a possible state objective; it says nothing of the permissible means—race conscious or otherwise—that a school district might employ to achieve that objective. The reason for this omission is clear enough, since the case did not involve any voluntary means adopted by a school district. The dissent’s characterization of Swann as recognizing that “the Equal Protection Clause permits local school boards to use race-conscious criteria to achieve positive race-related goals” is—at best—a dubious inference. Post, at 22. Even if the dicta from Swann were entitled to the weight the dissent would give it, and no dicta is, it not only did not address the question presented in Swann, it also does not address the question presented in these cases—whether the school districts’ use of racial classifications to achieve their stated goals is permissible.

   Further, for all the lower court cases Justice Breyer cites as evidence of the “prevailing legal assumption” embodied by Swann, very few are pertinent. Most are not. For example, the dissent features Tometz v. Board of Ed., Waukegan City School Dist. No. 61, 39 Ill. 2d 593, 596–598, 237 N. E. 2d 498, 500–502 (1968), an Illinois decision, as evidence that “state and federal courts had considered the matter settled and uncontroversial.” Post, at 25. But Tometz addressed a challenge to a statute requiring race-consciousness in drawing school attendance boundaries—an issue well beyond the scope of the question presented in these cases. Importantly, it considered that issue only under rational-basis review, 39 Ill. 2d, at 600, 237 N. E. 2d, at 502 (“The test of any legislative classification essentially is one of reasonableness”), which even the dissent grudgingly recognizes is an improper standard for evaluating express racial classifications. Other cases cited are similarly inapplicable. See, e.g., Citizens for Better Ed. v. Goose Creek Consol. Independent School Dist., 719 S. W. 2d 350, 352–353 (Tex. App. 1986) (upholding rezoning plan under rational-basis review).[Footnote 16]

   Justice Breyer’s dissent next looks for authority to a footnote in Washington v. Seattle School Dist. No. 1, 458 U. S. 457, 472, n. 15 (1982), post, at 56–57, but there this Court expressly noted that it was not passing on the propriety of race-conscious student assignments in the absence of a finding of de jure segregation. Similarly, the citation of Crawford v. Board of Ed. of Los Angeles, 458 U. S. 527 (1982), post, at 24, in which a state referendum prohibiting a race-based assignment plan was challenged, is inapposite—in Crawford the Court again expressly reserved the question presented by these cases. 458 U. S., at 535, n. 11. Such reservations and preliminary analyses of course did not decide the merits of this question—as evidenced by the disagreement among the lower courts on this issue. Compare Eisenberg, 197 F. 3d, at 133, with Comfort, 418 F. 3d, at 13.

   Justice Breyer’s dissent also asserts that these cases are controlled by Grutter, claiming that the existence of a compelling interest in these cases “follows a fortiori” from Grutter, post, at 41, 64–66, and accusing us of tacitly overruling that case, see post, at 64–66. The dissent overreads Grutter, however, in suggesting that it renders pure racial balancing a constitutionally compelling interest; Grutter itself recognized that using race simply to achieve racial balance would be “patently unconstitutional,” 539 U. S., at 330. The Court was exceedingly careful in describing the interest furthered in Grutter as “not an interest in simple ethnic diversity” but rather a “far broader array of qualifications and characteristics” in which race was but a single element. 539 U. S., at 324–325 (internal quotation marks omitted). We take the Grutter Court at its word. We simply do not understand how Justice Breyer can maintain that classifying every schoolchild as black or white, and using that classification as a determinative factor in assigning children to achieve pure racial balance, can be regarded as “less burdensome, and hence more narrowly tailored” than the consideration of race in Grutter, post, at 47, when the Court in Grutter stated that “[t]he importance of … individualized consideration” in the program was “paramount,” and consideration of race was one factor in a “highly individualized, holistic review.” 539 U. S., at 337. Certainly if the constitutionality of the stark use of race in these cases were as established as the dissent would have it, there would have been no need for the extensive analysis undertaken in Grutter. In light of the foregoing, Justice Breyer’s appeal to stare decisis rings particularly hollow. See post, at 65–66.

   At the same time it relies on inapplicable desegregation cases, misstatements of admitted dicta, and other noncontrolling pronouncements, Justice Breyer’s dissent candidly dismisses the significance of this Court’s repeated holdings that all racial classifications must be reviewed under strict scrutiny, see post, at 31–33, 35–36, arguing that a different standard of review should be applied because the districts use race for beneficent rather than malicious purposes, see post, at 31–36.

   This Court has recently reiterated, however, that “ ‘all racial classifications [imposed by government] … must be analyzed by a reviewing court under strict scrutiny.’ ” Johnson, 543 U. S., at 505 (quoting Adarand, 515 U. S., at 227; emphasis added by Johnson Court). See also Grutter, supra, at 326 (“[G]overnmental action based on race—a group classification long recognized as in most circumstances irrelevant and therefore prohibited—should be subjected to detailed judicial inquiry” (internal quotation marks and emphasis omitted)). Justice Breyer nonetheless relies on the good intentions and motives of the school districts, stating that he has found “no case that … repudiated this constitutional asymmetry between that which seeks to exclude and that which seeks to include members of minority races.” Post, at 29 (emphasis in original). We have found many. Our cases clearly reject the argument that motives affect the strict scrutiny analysis. See Johnson, supra, at 505 (“We have insisted on strict scrutiny in every context, even for so-called ‘benign’ racial classifications”); Adarand, 515 U. S., at 227 (rejecting idea that “ ‘benign’ ” racial classifications may be held to “different standard”); Croson, 488 U. S., at 500 (“Racial classifications are suspect, and that means that simple legislative assurances of good intention cannot suffice”).

   This argument that different rules should govern racial classifications designed to include rather than exclude is not new; it has been repeatedly pressed in the past, see, e.g., Gratz, 539 U. S., at 282 (Breyer, J., concurring in judgment); id., at 301 (Ginsburg, J., dissenting); Adarand, supra, at 243 (Stevens, J., dissenting); Wygant, 476 U. S., at 316–317 (Stevens, J., dissenting), and has been repeatedly rejected. See also Bakke, 438 U. S., at 289–291 (opinion of Powell, J.) (rejecting argument that strict scrutiny should be applied only to classifications that disadvantage minorities, stating “[r]acial and ethnic distinctions of any sort are inherently suspect and thus call for the most exacting judicial examination”).

   The reasons for rejecting a motives test for racial classifications are clear enough. “The Court’s emphasis on ‘benign racial classifications’ suggests confidence in its ability to distinguish good from harmful governmental uses of racial criteria. History should teach greater humility… . ‘[B]enign’ carries with it no independent meaning, but reflects only acceptance of the current generation’s conclusion that a politically acceptable burden, imposed on particular citizens on the basis of race, is reasonable.” Metro Broadcasting, 497 U. S., at 609–610 (O’Connor, J., dissenting). See also Adarand, supra, at 226 (“ ‘[I]t may not always be clear that a so-called preference is in fact benign’ ” (quoting Bakke, supra, at 298 (opinion of Powell, J.))). Accepting Justice Breyer’s approach would “do no more than move us from ‘separate but equal’ to ‘unequal but benign.’ ” Metro Broadcasting, supra, at 638 (Kennedy, J., dissenting).

   Justice Breyer speaks of bringing “the races” together (putting aside the purely black-and-white nature of the plans), as the justification for excluding individuals on the basis of their race. See post, at 28–29. Again, this approach to racial classifications is fundamentally at odds with our precedent, which makes clear that the Equal Protection Clause “protect[s] persons, not groups,” Adarand, 515 U. S., at 227 (emphasis in original). See ibid. (“[A]ll governmental action based on race—a group classification long recognized as ‘in most circumstances irrelevant and therefore prohibited,’ Hirabayashi [v. United States, 320 U. S. 81, 100 (1943)]—should be subjected to detailed judicial inquiry to ensure that the personal right to equal protection of the laws has not been infringed” (first emphasis in original); Metro Broadcasting, supra, at 636 (“[O]ur Constitution protects each citizen as an individual, not as a member of a group” (Kennedy, J., dissenting)); Bakke, supra, at 289 (opinion of Powell, J.) (Fourteenth Amendment creates rights “guaranteed to the individual. The rights established are personal rights”). This fundamental principle goes back, in this context, to Brown itself. See Brown v. Board of Education, 349 U. S. 294, 300 (1955) (Brown II) (“At stake is the personal interest of the plaintiffs in admission to public schools … on a nondiscriminatory basis” (emphasis added)). For the dissent, in contrast, “ ‘individualized scrutiny’ is simply beside the point.” Post, at 55.

   Justice Breyer’s position comes down to a familiar claim: The end justifies the means. He admits that “there is a cost in applying ‘a state-mandated racial label,’ ” post, at 67, but he is confident that the cost is worth paying. Our established strict scrutiny test for racial classifications, however, insists on “detailed examination, both as to ends and as to means.” Adarand, supra, at 236 (emphasis added). Simply because the school districts may seek a worthy goal does not mean they are free to discriminate on the basis of race to achieve it, or that their racial classifications should be subject to less exacting scrutiny.

   Despite his argument that these cases should be evaluated under a “standard of review that is not ‘strict’ in the traditional sense of that word,” post, at 36, Justice Breyer still purports to apply strict scrutiny to these cases. See post, at 37. It is evident, however, that Justice Breyer’s brand of narrow tailoring is quite unlike anything found in our precedents. Without any detailed discussion of the operation of the plans, the students who are affected, or the districts’ failure to consider race-neutral alternatives, the dissent concludes that the districts have shown that these racial classifications are necessary to achieve the districts’ stated goals. This conclusion is divorced from any evaluation of the actual impact of the plans at issue in these cases—other than to note that the plans “often have no effect.” Post, at 46.[Footnote 17] Instead, the dissent suggests that some combination of the development of these plans over time, the difficulty of the endeavor, and the good faith of the districts suffices to demonstrate that these stark and controlling racial classifications are constitutional. The Constitution and our precedents require more.

   In keeping with his view that strict scrutiny should not apply, Justice Breyer repeatedly urges deference to local school boards on these issues. See, e.g., post, at 21, 48–49, 66. Such deference “is fundamentally at odds with our equal protection jurisprudence. We put the burden on state actors to demonstrate that their race-based policies are justified.” Johnson, 543 U. S., at 506, n. 1. See Croson, 488 U. S., at 501 (“The history of racial classifications in this country suggests that blind judicial deference to legislative or executive pronouncements of necessity has no place in equal protection analysis”); West Virginia Bd. of Ed. v. Barnette, 319 U. S. 624, 637 (1943) (“The Fourteenth Amendment … protects the citizen against the State itself and all of its creatures—Boards of Education not excepted”).

   Justice Breyer’s dissent ends on an unjustified note of alarm. It predicts that today’s decision “threaten[s]” the validity of “[h]undreds of state and federal statutes and regulations.” Post, at 61; see also post, at 27–28. But the examples the dissent mentions—for example, a provision of the No Child Left Behind Act that requires States to set measurable objectives to track the achievement of students from major racial and ethnic groups, 20 U. S. C. §6311(b)(2)(C)(v)—have nothing to do with the pertinent issues in these cases.

   Justice Breyer also suggests that other means for achieving greater racial diversity in schools are necessarily unconstitutional if the racial classifications at issue in these cases cannot survive strict scrutiny. Post, at 58–62. These other means—e.g., where to construct new schools, how to allocate resources among schools, and which academic offerings to provide to attract students to certain schools—implicate different considerations than the explicit racial classifications at issue in these cases, and we express no opinion on their validity—not even in dicta. Rather, we employ the familiar and well-established analytic approach of strict scrutiny to evaluate the plans at issue today, an approach that in no way warrants the dissent’s cataclysmic concerns. Under that approach, the school districts have not carried their burden of showing that the ends they seek justify the particular extreme means they have chosen—classifying individual students on the basis of their race and discriminating among them on that basis.

*  *  *

   If the need for the racial classifications embraced by the school districts is unclear, even on the districts’ own terms, the costs are undeniable. “[D]istinctions between citizens solely because of their ancestry are by their very nature odious to a free people whose institutions are founded upon the doctrine of equality.” Adarand, 515 U. S., at 214 (internal quotation marks omitted). Government action dividing us by race is inherently suspect because such classifications promote “notions of racial inferiority and lead to a politics of racial hostility,” Croson, supra, at 493, “reinforce the belief, held by too many for too much of our history, that individuals should be judged by the color of their skin,” Shaw v. Reno, 509 U. S. 630, 657 (1993), and “endorse race-based reasoning and the conception of a Nation divided into racial blocs, thus contributing to an escalation of racial hostility and conflict.” Metro Broadcasting, 497 U. S., at 603 (O’Connor, J., dissenting). As the Court explained in Rice v. Cayetano, 528 U. S. 495, 517 (2000), “[o]ne of the principal reasons race is treated as a forbidden classification is that it demeans the dignity and worth of a person to be judged by ancestry instead of by his or her own merit and essential qualities.”

   All this is true enough in the contexts in which these statements were made—government contracting, voting districts, allocation of broadcast licenses, and electing state officers—but when it comes to using race to assign children to schools, history will be heard. In Brown v. Board of Education, 347 U. S. 483 (1954) (Brown I), we held that segregation deprived black children of equal educational opportunities regardless of whether school facilities and other tangible factors were equal, because government classification and separation on grounds of race themselves denoted inferiority. Id., at 493–494. It was not the inequality of the facilities but the fact of legally separating children on the basis of race on which the Court relied to find a constitutional violation in 1954. See id., at 494 (“ ‘The impact [of segregation] is greater when it has the sanction of the law’ ”). The next Term, we accordingly stated that “full compliance” with Brown I required school districts “to achieve a system of determining admission to the public schools on a nonracial basis.” Brown II, 349 U. S., at 300–301 (emphasis added).

   The parties and their amici debate which side is more faithful to the heritage of Brown, but the position of the plaintiffs in Brown was spelled out in their brief and could not have been clearer: “[T]he Fourteenth Amendment prevents states from according differential treatment to American children on the basis of their color or race.” Brief for Appellants in Nos. 1, 2, and 4 and for Respondents in No. 10 on Reargument in Brown I, O. T. 1953, p. 15 (Summary of Argument). What do the racial classifications at issue here do, if not accord differential treatment on the basis of race? As counsel who appeared before this Court for the plaintiffs in Brown put it: “We have one fundamental contention which we will seek to develop in the course of this argument, and that contention is that no State has any authority under the equal-protection clause of the Fourteenth Amendment to use race as a factor in affording educational opportunities among its citizens.” Tr. of Oral Arg. in Brown I, p. 7 (Robert L. Carter, Dec. 9, 1952). There is no ambiguity in that statement. And it was that position that prevailed in this Court, which emphasized in its remedial opinion that what was “[a]t stake is the personal interest of the plaintiffs in admission to public schools as soon as practicable on a nondiscriminatory basis,” and what was required was “determining admission to the public schools on a nonracial basis.” Brown II, supra, at 300–301 (emphasis added). What do the racial classifications do in these cases, if not determine admission to a public school on a racial basis?   Before Brown, schoolchildren were told where they could and could not go to school based on the color of their skin. The school districts in these cases have not carried the heavy burden of demonstrating that we should allow this once again—even for very different reasons. For schools that never segregated on the basis of race, such as Seattle, or that have removed the vestiges of past segregation, such as Jefferson County, the way “to achieve a system of determining admission to the public schools on a nonracial basis,” Brown II, 349 U. S., at 300–301, is to stop assigning students on a racial basis. The way to stop discrimination on the basis of race is to stop discriminating on the basis of race.

   The judgments of the Courts of Appeals for the Sixth and Ninth Circuits are reversed, and the cases are remanded for further proceedings.

It is so ordered.

Footnote 1  The plan was in effect from 1999–2002, for three school years. This litigation was commenced in July 2000, and the record in the District Court was closed before assignments for the 2001–2002 school year were made. See Brief for Respondents in No. 05–908, p. 9, n. 9. We rely, as did the lower courts, largely on data from the 2000–2001 school year in evaluating the plan. See 426 F. 3d 1162, 1169–1171 (CA9 2005) (en banc) (Parents Involved VII).

Footnote 2  The racial breakdown of this nonwhite group is approximately 23.8 percent Asian-American, 23.1 percent African-American, 10.3 percent Latino, and 2.8 percent Native-American. See 377 F. 3d 949, 1005–1006 (CA9 2004) (Parents Involved VI) (Graber, J., dissenting).

Footnote 3  For the 2001–2002 school year, the deviation permitted from the desired racial composition was increased from 10 to 15 percent. App. in No. 05–908, p. 38a. The bulk of the data in the record was collected using the 10 percent band, see n. 1, supra.

Footnote 4  “No State shall … deny to any person within its jurisdiction the equal protection of the laws.” U. S. Const., Amdt. 14, §1.

Footnote 5  “No person in the United States shall, on the ground of race … be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.” 78 Stat. 252, 42 U. S. C. §2000d.

Footnote 6  “The state shall not discriminate against, or grant preferential treatment to, any individual or group on the basis of race, sex, color, ethnicity, or national origin in the operation of public employment, public education, or public contracting.” Wash. Rev. Code §49.60.400(1) (2006).

Footnote 7  Middle and high school students are designated a single resides school and assigned to that school unless it is at the extremes of the racial guidelines. Students may also apply to a magnet school or program, or, at the high school level, take advantage of an open enrollment plan that allows ninth-grade students to apply for admission to any nonmagnet high school. App. in No. 05–915, pp. 39–41, 82–83.

Footnote 8  It is not clear why the racial guidelines were even applied to Joshua’s transfer application—the guidelines supposedly do not apply at the kindergarten level. Id., at 43. Neither party disputes, however, that Joshua’s transfer application was denied under the racial guidelines, and Meredith’s objection is not that the guidelines were misapplied but rather that race was used at all.

Footnote 9  Meredith joined a pending lawsuit filed by several other plaintiffs. See id., at 7–11. The other plaintiffs all challenged assignments to certain specialized schools, and the District Court found these assignments, which are no longer at issue in this case, unconstitutional. McFarland I, 330 F. Supp. 2d 834, 837, 864 (WD Ky. 2004).

Footnote 10  The districts point to dicta in a prior opinion in which the Court suggested that, while not constitutionally mandated, it would be constitutionally permissible for a school district to seek racially balanced schools as a matter of “educational policy.” See Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U. S. 1, 16 (1971). The districts also quote with approval an in-chambers opinion in which then-Justice Rehnquist made a suggestion to the same effect. See Bustop, Inc. v. Los Angeles Bd. of Ed., 439 U. S. 1380, 1383 (1978). The citations do not carry the significance the districts would ascribe to them. Swann, evaluating a school district engaged in court-ordered desegregation, had no occasion to consider whether a district’s voluntary adoption of race-based assignments in the absence of a finding of prior de jure segregation was constitutionally permissible, an issue that was again expressly reserved in Washington v. Seattle School Dist. No. 1, 458 U. S. 457, 472, n. 15 (1982). Bustop, addressing in the context of an emergency injunction application a busing plan imposed by the Superior Court of Los Angeles County, is similarly unavailing. Then-Justice Rehnquist, in denying emergency relief, stressed that “equitable consideration[s]” counseled against preliminary relief. 439 U. S., at 1383. The propriety of preliminary relief and resolution of the merits are of course “significantly different” issues. University of Texas v. Camenisch, 451 U. S. 390, 393 (1981).

Footnote 11  The way Seattle classifies its students bears this out. Upon enrolling their child with the district, parents are required to identify their child as a member of a particular racial group. If a parent identifies more than one race on the form, “[t]he application will not be accepted and, if necessary, the enrollment service person taking the application will indicate one box.” App. in No. 05–908, at 303a.

Footnote 12     Jefferson County also argues that it would be incongruous to hold that what was constitutionally required of it one day—race-based assignments pursuant to the desegregation decree—can be constitutionally prohibited the next. But what was constitutionally required of the district prior to 2000 was the elimination of the vestiges of prior segregation—not racial proportionality in its own right. See Freeman v. Pitts, 503 U. S. 467, 494–496 (1992). Once those vestiges were eliminated, Jefferson County was on the same footing as any other school district, and its use of race must be justified on other grounds.

Footnote 13 Data for the Seattle schools in the several years since this litigation was commenced further demonstrate the minimal role that the racial tiebreaker in fact played. At Ballard, in 2005–2006—when no class at the school was subject to the racial tiebreaker—the student body was 14.2 percent Asian-American, 9 percent African-American, 11.7 percent Latino, 62.3 percent Caucasian, and 2.8 percent Native-American. Reply Brief for Petitioner in No. 05–908, p. 7. In 2000–2001, when the racial tiebreaker was last used, Ballard’s total enrollment was 17.5 percent Asian-American, 10.8 percent African-American, 10.7 percent Latino, 56.4 percent Caucasian, and 4.6 percent Native-American. App. in No. 05–908, at 283a. Franklin in 2005–2006 was 48.9 percent Asian-American, 33.5 percent African-American, 6.6 percent Latino, 10.2 percent Caucasian, and 0.8 percent Native-American. Reply Brief for Petitioner in No. 05–908, at 7. With the racial tiebreaker in 2000–2001, total enrollment was 36.8 percent Asian-American, 32.2 percent African-American, 5.2 percent Latino, 25.1 percent Caucasian, and 0.7 percent Native-American. App. in No. 05–908, at 284a. Nathan Hale’s 2005–2006 enrollment was 17.3 percent Asian-American, 10.7 percent African-American, 8 percent Latino, 61.5 percent Caucasian, and 2.5 percent Native-American. Reply Brief for Petitioner in No. 05–908, at 7. In 2000–2001, with the racial tiebreaker, it was 17.9 percent Asian-American, 13.3 percent African-American, 7 percent Latino, 58.4 percent Caucasian, and 3.4 percent Native-American. App. in No. 05–908, at 286a.

Footnote 14     In contrast, Seattle’s website formerly described “emphasizing individualism as opposed to a more collective ideology” as a form of “cultural racism,” and currently states that the district has no intention “to hold onto unsuccessful concepts such as [a] … colorblind mentality.” Harrell, School Web Site Removed: Examples of Racism Sparked Controversy, Seattle Post-Intelligencer, June 2, 2006, pp. B1, B5. Compare Plessy v. Ferguson, 163 U. S. 537, 559 (1896) (Harlan, J., dissenting) (“Our Constitution is color-blind, and neither knows nor tolerates classes among citizens. In respect of civil rights, all citizens are equal before the law”).

Footnote 15  Justice Breyer makes much of the fact that in 1978 Seattle “settled” an NAACP complaint alleging illegal segregation with the federal Office for Civil Rights (OCR). See post, at 5, 8–9, 18, 23. The memorandum of agreement between Seattle and OCR, of course, contains no admission by Seattle that such segregation ever existed or was ongoing at the time of the agreement, and simply reflects a “desire to avoid the incovenience [sic] and expense of a formal OCR investigation,” which OCR was obligated under law to initiate upon the filing of such a complaint. Memorandum of Agreement between Seattle School District No. 1 of King County, Washington, and the Office for Civil Rights, United States Department of Health, Education, and Welfare 2 (June 9, 1978); see also 45 CFR §80.7(c) (2006).

Footnote 16  In fact, all the cases Justice Breyer’s dissent cites as evidence of the “prevailing legal assumption,” see post, at 25–27, were decided before this Court definitively determined that “all racial classifications … must be analyzed by a reviewing court under strict scrutiny.” Adarand Constructors, Inc. v. Peńa, 515 U. S. 200, 227 (1995). Many proceeded under the now-rejected view that classifications seeking to benefit a disadvantaged racial group should be held to a lesser standard of review. See, e.g., Springfield School Comm. v. Barksdale, 348 F. 2d 261, 266 (CA1 1965). Even if this purported distinction, which Justice Stevens would adopt, post, at 2, n. 3 (dissenting opinion), had not been already rejected by this Court, the distinction has no relevance to these cases, in which students of all races are excluded from the schools they wish to attend based solely on the racial classifications. See, e.g., App. in No. 05–908, at 202a (noting that 89 nonwhite students were denied assignment to a particular school by operation of Seattle’s racial tiebreaker).

Justice Stevens’s reliance on School Comm. of Boston v. Board of Ed., 352 Mass. 693, 227 N. E. 2d 729 (1967), appeal dism’d, 389 U. S. 572 (1968) (per curiam), post, at 3–5, is inapposite for the same reason that many of the cases cited by Justice Breyer are inapposite; the case involved a Massachusetts law that required school districts to avoid racial imbalance in schools but did not specify how to achieve this goal—and certainly did not require express racial classifications as the means to do so. The law was upheld under rational-basis review, with the state court explicitly rejecting the suggestion—which is now plainly the law—that “racial group classifications bear a far heavier burden of justification.” 352 Mass., at 700, 227 N. E. 2d, at 734 (internal quotation marks and citation omitted). The passage Justice Stevens quotes proves our point; all the quoted language says is that the school committee “shall prepare a plan to eliminate the imbalance.” Id., at 695, 227 N. E. 2d, at 731; see post, at 4, n. 5. Nothing in the opinion approves use of racial classifications as the means to address the imbalance. The suggestion that our decision today is somehow inconsistent with our disposition of that appeal is belied by the fact that neither the lower courts, the respondent school districts, nor any of their 51 amici saw fit even to cite the case. We raise this fact not to argue that the dismissal should be afforded any different stare decisis effect, but rather simply to suggest that perhaps—for the reasons noted above—the dismissal does not mean what Justice Stevens believes it does.

Footnote 17  Justice Breyer also tries to downplay the impact of the racial assignments by stating that in Seattle “students can decide voluntarily to transfer to a preferred district high school (without any consideration of race-conscious criteria).” Post, at 46. This presumably refers to the district’s decision to cease, for 2001–2002 school year assignments, applying the racial tiebreaker to students seeking to transfer to a different school after ninth grade. See App. in No. 05–908, at 137a–139a. There are obvious disincentives for students to transfer to a different school after a full quarter of their high school experience has passed, and the record sheds no light on how transfers to the oversubscribed high schools are handled.

THOMAS, J., CONCURRING
PARENTS INVOLVED IN COMMUNITY SCHOOLS V.SEATTLE SCHOOL DIST. NO. 1
551 U. S. ____ (2007)
SUPREME COURT OF THE UNITED STATES
NOS. 05-908 AND 05-915

PARENTS INVOLVED IN COMMUNITY SCHOOLS, PETITIONER

05–908   v.

SEATTLE SCHOOL DISTRICT NO. 1 et al.

on writ of certiorari to the united states court of appeals for the ninth circuit

CRYSTAL D. MEREDITH, custodial parent and next friend of JOSHUA RYAN McDONALD, PETITIONER

05–915   v.

JEFFERSON COUNTY BOARD OF EDUCATION et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 28, 2007]

   Justice Thomas, concurring.

   Today, the Court holds that state entities may not experiment with race-based means to achieve ends they deem socially desirable. I wholly concur in The Chief Justice’s opinion. I write separately to address several of the contentions in Justice Breyer’s dissent (hereinafter the dissent). Contrary to the dissent’s arguments, resegregation is not occurring in Seattle or Louisville; these school boards have no present interest in remedying past segregation; and these race-based student-assignment programs do not serve any compelling state interest. Accordingly, the plans are unconstitutional. Disfavoring a color-blind interpretation of the Constitution, the dissent would give school boards a free hand to make decisions on the basis of race—an approach reminiscent of that advocated by the segregationists in Brown v. Board of Education, 347 U. S.483 (1954). This approach is just as wrong today as it was a half-century ago. The Constitution and our cases require us to be much more demanding before permitting local school boards to make decisions based on race.

I

   The dissent repeatedly claims that the school districts are threatened with resegregation and that they will succumb to that threat if these plans are declared unconstitutional. It also argues that these plans can be justified as part of the school boards’ attempts to “eradicat[e] earlier school segregation.” See, e.g., post, at 4. Contrary to the dissent’s rhetoric, neither of these school districts is threatened with resegregation, and neither is constitutionally compelled or permitted to undertake race-based remediation. Racial imbalance is not segregation, and the mere incantation of terms like resegregation and remediation cannot make up the difference.

A

   Because this Court has authorized and required race-based remedial measures to address de jure segregation, it is important to define segregation clearly and to distinguish it from racial imbalance. In the context of public schooling, segregation is the deliberate operation of a school system to “carry out a governmental policy to separate pupils in schools solely on the basis of race.” Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U. S. 1, 6 (1971); see also Monroe v. Board of Comm’rs of Jackson, 391 U. S. 450, 452 (1968). In Brown, this Court declared that segregation was unconstitutional under the Equal Protection Clause of the Fourteenth Amendment. Swann, supra, at 6; see also Green v. School Bd. of New Kent Cty., 391 U. S. 430, 435 (1968) (“[T]he State, acting through the local school board and school officials, organized and operated a dual system, part ‘white’ and part ‘Negro.’ It was such dual systems that 14 years ago Brown I[, 347 U. S. 483,] held unconstitutional and a year later Brown II[, 349 U. S. 294 (1955)] held must be abolished”).[Footnote 1]

   Racial imbalance is the failure of a school district’s individual schools to match or approximate the demographic makeup of the student population at large. Cf. Washington v. Seattle School Dist. No. 1, 458 U. S. 457, 460 (1982). Racial imbalance is not segregation.[Footnote 2] Although presently observed racial imbalance might result from past de jure segregation, racial imbalance can also result from any number of innocent private decisions, including voluntary housing choices. See Swann, supra, at 25–26; Missouri v. Jenkins, 515 U. S. 70, 116 (1995) (Thomas, J., concurring). Because racial imbalance is not inevitably linked to unconstitutional segregation, it is not unconstitutional in and of itself. Dayton Bd. of Ed. v. Brinkman, 433 U. S. 406, 413 (1977); Dayton Bd. of Ed. v. Brinkman, 443 U. S. 526, 531, n. 5 (1979) (“Racial imbalance … is not per se a constitutional violation”); Freeman v. Pitts, 503 U. S. 467, 494 (1992); see also Swann, supra, at 31–32; cf. Milliken v. Bradley, 418 U. S. 717, 740–741, and n. 19 (1974).

   Although there is arguably a danger of racial imbalance in schools in Seattle and Louisville, there is no danger of resegregation. No one contends that Seattle has established or that Louisville has reestablished a dual school system that separates students on the basis of race. The statistics cited in Appendix A to the dissent are not to the contrary. See post, at 69–72. At most, those statistics show a national trend toward classroom racial imbalance. However, racial imbalance without intentional state action to separate the races does not amount to segregation. To raise the specter of resegregation to defend these programs is to ignore the meaning of the word and the nature of the cases before us.[Footnote 3]

B

   Just as the school districts lack an interest in preventing resegregation, they also have no present interest in remedying past segregation. The Constitution generally prohibits government race-based decisionmaking, but this Court has authorized the use of race-based measures for remedial purposes in two narrowly defined circumstances. First, in schools that were formerly segregated by law, race-based measures are sometimes constitutionally compelled to remedy prior school segregation. Second, in Croson, the Court appeared willing to authorize a government unit to remedy past discrimination for which it was responsible. Richmond v. J. A. Croson Co., 488 U. S. 469, 504 (1989). Without explicitly resting on either of these strands of doctrine, the dissent repeatedly invokes the school districts’ supposed interests in remedying past segregation. Properly analyzed, though, these plans do not fall within either existing category of permissible race-based remediation.

1

   The Constitution does not permit race-based government decisionmaking simply because a school district claims a remedial purpose and proceeds in good faith with arguably pure motives. Grutter v. Bollinger, 539 U. S. 306, 371 (2003) (Thomas, J., concurring in part and dissenting in part) (citing Adarand Constructors, Inc. v. Peńa, 515 U. S. 200, 239 (1995) (Scalia, J., concurring in part and concurring in judgment)). Rather, race-based government decisionmaking is categorically prohibited unless narrowly tailored to serve a compelling interest. Grutter, supra, at 326; see also Part II–A, infra. This exacting scrutiny “has proven automatically fatal” in most cases. Jenkins, supra, at 121 (Thomas, J., concurring); cf. Hirabayashi v. United States, 320 U. S. 81, 100 (1943) (“[R]acial discriminations are in most circumstances irrelevant and therefore prohibited”). And appropriately so. “The Constitution abhors classifications based on race, not only because those classifications can harm favored races or are based on illegitimate motives, but also because every time the government places citizens on racial registers and makes race relevant to the provision of burdens or benefits, it demeans us all.” Grutter, supra, at 353 (opinion of Thomas, J.). Therefore, as a general rule, all race-based government decisionmaking—regardless of context—is unconstitutional.

2

   This Court has carved out a narrow exception to that general rule for cases in which a school district has a “history of maintaining two sets of schools in a single school system deliberately operated to carry out a governmental policy to separate pupils in schools solely on the basis of race.”[Footnote 4] See Swann, 402 U. S., at 5–6. In such cases, race-based remedial measures are sometimes required.[Footnote 5] Green, 391 U. S., at 437–438; cf. United States v. Fordice, 505 U. S. 717, 745 (1992) (Thomas, J., concurring).[Footnote 6] But without a history of state-enforced racial separation, a school district has no affirmative legal obligation to take race-based remedial measures to eliminate segregation and its vestiges.

   Neither of the programs before us today is compelled as a remedial measure, and no one makes such a claim. Seattle has no history of de jure segregation; therefore, the Constitution did not require Seattle’s plan.[Footnote 7] Although Louisville once operated a segregated school system and was subject to a Federal District Court’s desegregation decree, see ante, at 7; Hampton v. Jefferson Cty. Bd. of Ed., 102 F. Supp. 2d 358, 376–377 (WD Ky. 2000), that decree was dissolved in 2000, id., at 360. Since then, no race-based remedial measures have been required in Louisville. Thus, the race-based student-assignment plan at issue here, which was instituted the year after the dissolution of the desegregation decree, was not even arguably required by the Constitution.

3

   Aside from constitutionally compelled remediation in schools, this Court has permitted government units to remedy prior racial discrimination only in narrow circumstances. See Wygant v. Jackson Bd. of Ed., 476 U. S. 267, 277 (1986) (plurality opinion). Regardless of the constitutional validity of such remediation, see Croson, supra, at 524–525 (Scalia, J., concurring in judgment), it does not apply here. Again, neither school board asserts that its race-based actions were taken to remedy prior discrimination. Seattle provides three forward-looking—as opposed to remedial—justifications for its race-based assignment plan. Brief for Respondents in No. 05–908, pp. 24–34. Louisville asserts several similar forward-looking interests, Brief for Respondents in No. 05–915, pp. 24–29, and at oral argument, counsel for Louisville disavowed any claim that Louisville’s argument “depend[ed] in any way on the prior de jure segregation,” Tr. of Oral Arg. in No. 05–915, p. 38.

   Furthermore, for a government unit to remedy past discrimination for which it was responsible, the Court has required it to demonstrate “a ‘strong basis in evidence for its conclusion that remedial action was necessary.’ ” Croson, 488 U. S., at 500 (quoting Wygant, supra, at 277 (plurality opinion)). Establishing a “strong basis in evidence” requires proper findings regarding the extent of the government unit’s past racial discrimination. Croson, 488 U. S., at 504. The findings should “define the scope of any injury [and] the necessary remedy,” id., at 505, and must be more than “inherently unmeasurable claims of past wrongs,” id., at 506. Assertions of general societal discrimination are plainly insufficient. Id., at 499, 504; Wygant, supra, at 274 (plurality opinion); cf. Regents of Univ. of Cal. v. Bakke, 438 U. S. 265, 310 (1978) (opinion of Powell, J.). Neither school district has made any such specific findings. For Seattle, the dissent attempts to make up for this failing by adverting to allegations made in past complaints filed against the Seattle school district. However, allegations in complaints cannot substitute for specific findings of prior discrimination—even when those allegations lead to settlements with complaining parties. Cf. Croson, supra, at 505; Wygant, supra, at 279, n. 5 (plurality opinion). As for Louisville, its slate was cleared by the District Court’s 2000 dissolution decree, which effectively declared that there were no longer any effects of de jure discrimination in need of remediation.[Footnote 8]

   Despite the dissent’s repeated intimation of a remedial purpose, neither of the programs in question qualifies as a permissible race-based remedial measure. Thus, the programs are subject to the general rule that government race-based decisionmaking is unconstitutional.

C

   As the foregoing demonstrates, racial balancing is sometimes a constitutionally permissible remedy for the discrete legal wrong of de jure segregation, and when directed to that end, racial balancing is an exception to the general rule that government race-based decisionmaking is unconstitutional. Perhaps for this reason, the dissent conflates the concepts of segregation and racial imbalance: If racial imbalance equates to segregation, then it must also be constitutionally acceptable to use racial balancing to remedy racial imbalance.

   For at least two reasons, however, it is wrong to place the remediation of segregation on the same plane as the remediation of racial imbalance. First, as demonstrated above, the two concepts are distinct. Although racial imbalance can result from de jure segregation, it does not necessarily, and the further we get from the era of state-sponsored racial separation, the less likely it is that racial imbalance has a traceable connection to any prior segregation. See Freeman, 503 U. S., at 496; Jenkins, 515 U. S., at 118 (Thomas, J., concurring).

   Second, a school cannot “remedy” racial imbalance in the same way that it can remedy segregation. Remediation of past de jure segregation is a one-time process involving the redress of a discrete legal injury inflicted by an identified entity. At some point, the discrete injury will be remedied, and the school district will be declared unitary. See Swann, 402 U. S., at 31. Unlike de jure segregation, there is no ultimate remedy for racial imbalance. Individual schools will fall in and out of balance in the natural course, and the appropriate balance itself will shift with a school district’s changing demographics. Thus, racial balancing will have to take place on an indefinite basis—a continuous process with no identifiable culpable party and no discernable end point. In part for those reasons, the Court has never permitted outright racial balancing solely for the purpose of achieving a particular racial balance.

II

   Lacking a cognizable interest in remediation, neither of these plans can survive strict scrutiny because neither plan serves a genuinely compelling state interest. The dissent avoids reaching that conclusion by unquestioningly accepting the assertions of selected social scientists while completely ignoring the fact that those assertions are the subject of fervent debate. Ultimately, the dissent’s entire analysis is corrupted by the considerations that lead it initially to question whether strict scrutiny should apply at all. What emerges is a version of “strict scrutiny” that combines hollow assurances of harmlessness with reflexive acceptance of conventional wisdom. When it comes to government race-based decisionmaking, the Constitution demands more.

A

   The dissent claims that “the law requires application here of a standard of review that is not ‘strict’ in the traditional sense of that word.” Post, at 36. This view is informed by dissents in our previous cases and the concurrences of two Court of Appeals judges. Post, at 34–36 (citing 426 F. 3d 1162, 1193–1194 (CA9 2005) (Kozinski, J., concurring); Comfort v. Lynn School Comm., 418 F. 3d 1, 28–29 (CA1 2005) (Boudin, C. J., concurring)). Those lower court judges reasoned that programs like these are not “aimed at oppressing blacks” and do not “seek to give one racial group an edge over another.” Comfort, supra, at 27 (Boudin, C. J., concurring); 426 F. 3d, at 1193 (Kozinski, J., concurring). They were further persuaded that these plans differed from other race-based programs this Court has considered because they are “certainly more benign than laws that favor or disfavor one race, segregate by race, or create quotas for or against a racial group,” Comfort, 418 F. 3d, at 28 (Boudin, C. J., concurring), and they are “far from the original evils at which the Fourteenth Amendment was addressed,” id., at 29; 426 F. 3d, at 1195 (Kozinski, J., concurring). Instead of strict scrutiny, Judge Kozinski would have analyzed the plans under “robust and realistic rational basis review.” Id., at 1194.

   These arguments are inimical to the Constitution and to this Court’s precedents.[Footnote 9] We have made it unusually clear that strict scrutiny applies to every racial classification. Adarand, 515 U. S., at 227; Grutter, 539 U. S., at 326; Johnson v. California, 543 U. S. 499, 505 (2005) (“We have insisted on strict scrutiny in every context, even for so-called ‘benign’ racial classifications”).[Footnote 10] There are good reasons not to apply a lesser standard to these cases. The constitutional problems with government race-based decisionmaking are not diminished in the slightest by the presence or absence of an intent to oppress any race or by the real or asserted well-meaning motives for the race-based decisionmaking. Adarand, 515 U. S., at 228–229. Purportedly benign race-based decisionmaking suffers the same constitutional infirmity as invidious race-based decisionmaking. Id., at 240 (Thomas, J., concurring in part and concurring in judgment) (“As far as the Constitution is concerned, it is irrelevant whether a government’s racial classifications are drawn by those who wish to oppress a race or by those who have a sincere desire to help those thought to be disadvantaged”).

   Even supposing it mattered to the constitutional analysis, the race-based student assignment programs before us are not as benign as the dissent believes. See post, at 34–35. “[R]acial paternalism and its unintended consequences can be as poisonous and pernicious as any other form of discrimination.” Adarand, supra, at 241 (opinion of Thomas, J.). As these programs demonstrate, every time the government uses racial criteria to “bring the races together,” post, at 29, someone gets excluded, and the person excluded suffers an injury solely because of his or her race. The petitioner in the Louisville case received a letter from the school board informing her that her kindergartener would not be allowed to attend the school of petitioner’s choosing because of the child’s race. App. in No. 05–915, p. 97. Doubtless, hundreds of letters like this went out from both school boards every year these race-based assignment plans were in operation. This type of exclusion, solely on the basis of race, is precisely the sort of government action that pits the races against one another, exacerbates racial tension, and “provoke[s] resentment among those who believe that they have been wronged by the government’s use of race.” Adarand, supra, at 241 (opinion of Thomas, J.). Accordingly, these plans are simply one more variation on the government race-based decisionmaking we have consistently held must be subjected to strict scrutiny. Grutter, supra, at 326.

B

   Though the dissent admits to discomfort in applying strict scrutiny to these plans, it claims to have nonetheless applied that exacting standard. But in its search for a compelling interest, the dissent casually accepts even the most tenuous interests asserted on behalf of the plans, grouping them all under the term “ ‘integration.’ ” See post, at 37. “ ‘[I]ntegration,’ ” we are told, has “three essential elements.” Ibid. None of these elements is compelling. And the combination of the three unsubstantiated elements does not produce an interest any more compelling than that represented by each element independently.

1

   According to the dissent, integration involves “an interest in setting right the consequences of prior conditions of segregation.” Post, at 37. For the reasons explained above, the records in these cases do not demonstrate that either school board’s plan is supported by an interest in remedying past discrimination. Part I–B, supra.

   Moreover, the school boards have no interest in remedying the sundry consequences of prior segregation unrelated to schooling, such as “housing patterns, employment practices, economic conditions, and social attitudes.” Post, at 38. General claims that past school segregation affected such varied societal trends are “too amorphous a basis for imposing a racially classified remedy,” Wygant, 476 U. S., at 276 (plurality opinion), because “[i]t is sheer speculation” how decades-past segregation in the school system might have affected these trends, see Croson, 488 U. S., at 499. Consequently, school boards seeking to remedy those societal problems with race-based measures in schools today would have no way to gauge the proper scope of the remedy. Id., at 498. Indeed, remedial measures geared toward such broad and unrelated societal ills have “ ‘no logical stopping point,’ ” ibid., and threaten to become “ageless in their reach into the past, and timeless in their ability to affect the future,” Wygant, supra, at 276 (plurality opinion). See Grutter, 539 U. S., at 342 (stating the “requirement that all governmental use of race must have a logical end point”).

   Because the school boards lack any further interest in remedying segregation, this element offers no support for the purported interest in “integration.”

2

   Next, the dissent argues that the interest in integration has an educational element. The dissent asserts that racially balanced schools improve educational outcomes for black children. In support, the dissent unquestioningly cites certain social science research to support propositions that are hotly disputed among social scientists. In reality, it is far from apparent that coerced racial mixing has any educational benefits, much less that integration is necessary to black achievement.

   Scholars have differing opinions as to whether educational benefits arise from racial balancing. Some have concluded that black students receive genuine educational benefits. See, e.g., Crain & Mahard, Desegregation and Black Achievement: A Review of the Research, 42 L. & Contemp. Probs. 17, 48 (1978). Others have been more circumspect. See, e.g., Henderson, Greenberg, Schneider, Uribe, & Verdugo, High Quality Schooling for African American Students, in Beyond Desegregation 166 (M. Shujaa ed. 1996) (“Perhaps desegregation does not have a single effect, positive or negative, on the academic achievement of African American students, but rather some strategies help, some hurt, and still others make no difference whatsoever. It is clear to us that focusing simply on demographic issues detracts from focusing on improving schools”). And some have concluded that there are no demonstrable educational benefits. See, e.g., Armor & Rossell, Desegregation and Resegregation in the Public Schools, in Beyond the Color Line: New Perspectives on Race and Ethnicity in America 239, 251 (A. Thernstrom & S. Thernstrom eds. 2002).

   The amicus briefs in the cases before us mirror this divergence of opinion. Supporting the school boards, one amicus has assured us that “both early desegregation research and recent statistical and econometric analyses … indicate that there are positive effects on minority student achievement scores arising from diverse school settings.” Brief for American Educational Research Association as Amicus Curiae 10. Another brief claims that “school desegregation has a modest positive impact on the achievement of African-American students.” App. to Brief for 553 Social Scientists as Amici Curiae 13–14 (footnote omitted). Yet neither of those briefs contains specific details like the magnitude of the claimed positive effects or the precise demographic mix at which those positive effects begin to be realized. Indeed, the social scientists’ brief rather cautiously claims the existence of any benefit at all, describing the “positive impact” as “modest,” id., at 13, acknowledging that “there appears to be little or no effect on math scores,” id., at 14, and admitting that the “underlying reasons for these gains in achievement are not entirely clear,” id., at 15.[Footnote 11]

   Other amici dispute these findings. One amicus reports that “[i]n study after study, racial composition of a student body, when isolated, proves to be an insignificant determinant of student achievement.” Brief for Dr. John Murphy et al. as Amici Curiae in No. 05–908, p. 8; see also id., at 9 (“[T]here is no evidence that diversity in the K–12 classroom positively affects student achievement”). Another amicus surveys several social science studies and concludes that “a fair and comprehensive analysis of the research shows that there is no clear and consistent evidence of [educational] benefits.” Brief for David J. Armor et al. as Amici Curiae 29.

    Add to the inconclusive social science the fact of black achievement in “racially isolated” environments. See T. Sowell, Education: Assumptions Versus History 7–38 (1986). Before Brown, the most prominent example of an exemplary black school was Dunbar High School. Id., at 29 (“[I]n the period 1918–1923, Dunbar graduates earned fifteen degrees from Ivy League colleges, and ten degrees from Amherst, Williams, and Wesleyan”). Dunbar is by no means an isolated example. See id., at 10–32 (discussing other successful black schools); Walker, Can Institutions Care? Evidence from the Segregated Schooling of African American Children, in Beyond Desegregation 209–226 (M. Shujaa ed. 1996); see also T. Sowell, Affirmative Action Around the World: An Empirical Study 141–165 (2004). Even after Brown, some schools with predominantly black enrollments have achieved outstanding educational results. See, e.g., S. Carter, No Excuses: Lessons from 21 High-Performing, High-Poverty Schools 49–50, 53–56, 71–73, 81–84, 87–88 (2001); A. Thernstrom & S. Thernstrom, No Excuses: Closing the Racial Gap in Learning 43–64 (2003); see also L. Izumi, They Have Overcome: High-Poverty, High-Performing Schools in California (2002) (chronicling exemplary achievement in predominantly Hispanic schools in California). There is also evidence that black students attending historically black colleges achieve better academic results than those attending predominantly white colleges. Grutter, supra, at 364–365 (Thomas, J., concurring in part and concurring in judgment) (citing sources); see also Fordice, 505 U. S., at 748–749 (Thomas, J., concurring).

   The Seattle school board itself must believe that racial mixing is not necessary to black achievement. Seattle operates a K–8 “African-American Academy,” which has a “nonwhite” enrollment of 99%. See App. in No. 05–908, p. 227a; Reply Brief in No. 05–908, p. 13, n. 13. That school was founded in 1990 as part of the school board’s effort to “increase academic achievement.”[Footnote 12] See African American Academy History, online at http://www. seattleschools.org/schools/aaa/history.htm (all Internet materials as visited June 26, 2007, and available in Clerk of Court’s case file). According to the school’s most recent annual report, “[a]cademic excellence” is its “primary goal.” See African American Academy 2006 Annual Report, p. 2, online at http://www.seattleschools.org/area/ siso/reports/anrep/altern/938.pdf. This racially imbalanced environment has reportedly produced test scores “higher across all grade levels in reading, writing and math.” Ibid. Contrary to what the dissent would have predicted, see post, at 38–39, the children in Seattle’s African American Academy have shown gains when placed in a “highly segregated” environment.

   Given this tenuous relationship between forced racial mixing and improved educational results for black children, the dissent cannot plausibly maintain that an educational element supports the integration interest, let alone makes it compelling.[Footnote 13] See Jenkins, 515 U. S., at 121–122 (Thomas, J., concurring) (“[T]here is no reason to think that black students cannot learn as well when surrounded by members of their own race as when they are in an integrated environment”).

   Perhaps recognizing as much, the dissent argues that the social science evidence is “strong enough to permit a democratically elected school board reasonably to determine that this interest is a compelling one.” Post, at 38. This assertion is inexplicable. It is not up to the school boards—the very government entities whose race-based practices we must strictly scrutinize—to determine what interests qualify as compelling under the Fourteenth Amendment to the United States Constitution. Rather, this Court must assess independently the nature of the interest asserted and the evidence to support it in order to determine whether it qualifies as compelling under our precedents. In making such a determination, we have deferred to state authorities only once, see Grutter, 539 U. S., at 328–330, and that deference was prompted by factors uniquely relevant to higher education. Id., at 328 (“Our holding today is in keeping with our tradition of giving a degree of deference to a university’s academic decisions”). The dissent’s proposed test—whether sufficient social science evidence supports a government unit’s conclusion that the interest it asserts is compelling—calls to mind the rational-basis standard of review the dissent purports not to apply, post, at 36-37. See Williamson v. Lee Optical of Okla., Inc., 348 U. S. 483, 488 (1955) (“It is enough that there is an evil at hand for correction, and that it might be thought that the particular legislative measure was a rational way to correct it”). Furthermore, it would leave our equal-protection jurisprudence at the mercy of elected government officials evaluating the evanescent views of a handful of social scientists. To adopt the dissent’s deferential approach would be to abdicate our constitutional responsibilities.[Footnote 14]

3

   Finally, the dissent asserts a “democratic element” to the integration interest. It defines the “democratic element” as “an interest in producing an educational environment that reflects the ‘pluralistic society’ in which our children will live.” Post, at 39.[Footnote 15] Environmental reflection, though, is just another way to say racial balancing. And “[p]referring members of any one group for no reason other than race or ethnic origin is discrimination for its own sake.” Bakke, 438 U. S., at 307 (opinion of Powell, J.). “This the Constitution forbids.” Ibid.; Grutter, supra, at 329–330; Freeman, 503 U. S., at 494.

   Navigating around that inconvenient authority, the dissent argues that the racial balancing in these plans is not an end in itself but is instead intended to “teac[h] children to engage in the kind of cooperation among Americans of all races that is necessary to make a land of three hundred million people one Nation.” Post, at 39–40. These “generic lessons in socialization and good citizenship” are too sweeping to qualify as compelling interests. Grutter, 539 U. S., at 348 (Scalia, J., concurring in part and dissenting in part). And they are not “uniquely relevant” to schools or “uniquely ‘teachable’ in a formal educational setting.” Id., at 347. Therefore, if governments may constitutionally use racial balancing to achieve these aspirational ends in schools, they may use racial balancing to achieve similar goals at every level—from state-sponsored 4–H clubs, see Bazemore v. Friday, 478 U. S. 385, 388–390 (1986) (Brennan, J., concurring), to the state civil service. See Grutter, 539 U. S. 347–348 (opinion of Scalia, J.).

   Moreover, the democratic interest has no durational limit, contrary to Grutter’s command. See id., at 342; see also Croson, 488 U. S., at 498; Wygant, 476 U. S., at 275 (plurality opinion). In other words, it will always be important for students to learn cooperation among the races. If this interest justifies race-conscious measures today, then logically it will justify race-conscious measures forever. Thus, the democratic interest, limitless in scope and “timeless in [its] ability to affect the future,” id., at 276 (plurality opinion), cannot justify government race-based decisionmaking.[Footnote 16]

   In addition to these defects, the democratic element of the integration interest fails on the dissent’s own terms. The dissent again relies upon social science research to support the proposition that state-compelled racial mixing teaches children to accept cooperation and improves racial attitudes and race relations. Here again, though, the dissent overstates the data that supposedly support the interest.

   The dissent points to data that indicate that “black and white students in desegregated schools are less racially prejudiced than those in segregated schools.” Post, at 40 (internal quotation marks omitted). By the dissent’s account, improvements in racial attitudes depend upon the increased contact between black and white students thought to occur in more racially balanced schools. There is no guarantee, however, that students of different races in the same school will actually spend time with one another. Schools frequently group students by academic ability as an aid to efficient instruction, but such groupings often result in classrooms with high concentrations of one race or another. See, e.g., Yonezawa, Wells, & Serna, Choosing Tracks: “Freedom of Choice” in Detracting Schools, 39 Am. Ed. Research J., No. 1, p. 38 (Spring 2002); Mickelson, Subverting Swann: First- and Second-Generation Segregation in the Charlotte-Mecklenburg Schools, 38 Am. Ed. Research J., No. 2, pp. 233–234 (Summer 2001) (describing this effect in schools in Charlotte, North Carolina). In addition to classroom separation, students of different races within the same school may separate themselves socially. See Hallinan & Williams, Interracial Friendship Choices in Secondary Schools, 54 Am. Sociological Rev., No. 1, pp. 72–76 (Feb. 1989); see also Clotfelter, Interracial Contact in High School Extracurricular Activities, 34 Urban Rev., No. 1, pp. 41–43 (Mar. 2002). Therefore, even supposing interracial contact leads directly to improvements in racial attitudes and race relations, a program that assigns students of different races to the same schools might not capture those benefits. Simply putting students together under the same roof does not necessarily mean that the students will learn together or even interact.

   Furthermore, it is unclear whether increased interracial contact improves racial attitudes and relations.[Footnote 17] One researcher has stated that “the reviews of desegregation and intergroup relations were unable to come to any conclusion about what the probable effects of desegregation were … [;] virtually all of the reviewers determined that few, if any, firm conclusions about the impact of desegregation on intergroup relations could be drawn.” Schofield, School Desegregation and Intergroup Relations: A Review of the Literature, in 17 Review of Research in Education 356 (G. Grant ed. 1991). Some studies have even found that a deterioration in racial attitudes seems to result from racial mixing in schools. See N. St. John, School Desegregation Outcomes for Children 67–68 (1975) (“A glance at [the data] shows that for either race positive findings are less common than negative findings”); Stephan, The Effects of School Desegregation: An Evaluation 30 Years After Brown, in Advances in Applied Social Psychology 183–186 (M. Saks & L. Saxe eds. 1986). Therefore, it is not nearly as apparent as the dissent suggests that increased interracial exposure automatically leads to improved racial attitudes or race relations.

   Given our case law and the paucity of evidence supporting the dissent’s belief that these plans improve race relations, no democratic element can support the integration interest.[Footnote 18]

4

   The dissent attempts to buttress the integration interest by claiming that it follows a fortiori from the interest this Court recognized as compelling in Grutter. Post, at 41. Regardless of the merit of Grutter, the compelling interest recognized in that case cannot support these plans. Grutter recognized a compelling interest in a law school’s attainment of a diverse student body. 539 U. S., at 328. This interest was critically dependent upon features unique to higher education: “the expansive freedoms of speech and thought associated with the university environment,” the “special niche in our constitutional tradition” occupied by universities, and “[t]he freedom of a university to make its own judgments as to education[,] includ[ing] the selection of its student body.” Id., at 329 (internal quotation marks omitted). None of these features is present in elementary and secondary schools. Those schools do not select their own students, and education in the elementary and secondary environment generally does not involve the free interchange of ideas thought to be an integral part of higher education. See 426 F. 3d, at 1208 (Bea, J., dissenting). Extending Grutter to this context would require us to cut that holding loose from its theoretical moorings. Thus, only by ignoring Grutter’s reasoning can the dissent claim that recognizing a compelling interest in these cases is an a fortiori application of Grutter.

C

   Stripped of the baseless and novel interests the dissent asserts on their behalf, the school boards cannot plausibly maintain that their plans further a compelling interest. As I explained in Grutter, only “those measures the State must take to provide a bulwark against anarchy … or to prevent violence” and “a government’s effort to remedy past discrimination for which it is responsible” constitute compelling interests. 539 U. S., at 351–352, 353. Neither of the parties has argued—nor could they—that race-based student assignment is necessary to provide a bulwark against anarchy or to prevent violence. And as I explained above, the school districts have no remedial interest in pursuing these programs. See Part I–B, supra. Accordingly, the school boards cannot satisfy strict scrutiny. These plans are unconstitutional.

III

   Most of the dissent’s criticisms of today’s result can be traced to its rejection of the color-blind Constitution. See post, at 29. The dissent attempts to marginalize the notion of a color-blind Constitution by consigning it to me and Members of today’s plurality.[Footnote 19] See ibid.; see also post, at 61. But I am quite comfortable in the company I keep. My view of the Constitution is Justice Harlan’s view in Plessy: “Our Constitution is color-blind, and neither knows nor tolerates classes among citizens.” Plessy v. Ferguson, 163 U. S. 537, 559 (1896) (dissenting opinion). And my view was the rallying cry for the lawyers who litigated Brown. See, e.g., Brief for Appellants in Brown v. Board of Education, O. T. 1953, Nos. 1, 2, and 4 p. 65 (“That the Constitution is color blind is our dedicated belief”); Brief for Appellants in Brown v. Board of Education, O. T. 1952, No. 1, p. 5 (“The Fourteenth Amendment precludes a state from imposing distinctions or classifications based upon race and color alone”);[Footnote 20] see also In Memoriam: Honorable Thurgood Marshall, Proceedings of the Bar and Officers of the Supreme Court of the United States, X (1993) (remarks of Judge Motley) (“Marshall had a ‘Bible’ to which he turned during his most depressed moments. The ‘Bible’ would be known in the legal community as the first Mr. Justice Harlan’s dissent in Plessy v. Ferguson, 163 U. S. 537, 552 (1896). I do not know of any opinion which buoyed Marshall more in his pre-Brown days …”).

   The dissent appears to pin its interpretation of the Equal Protection Clause to current societal practice and expectations, deference to local officials, likely practical consequences, and reliance on previous statements from this and other courts. Such a view was ascendant in this Court’s jurisprudence for several decades. It first appeared in Plessy, where the Court asked whether a state law providing for segregated railway cars was “a reasonable regulation.” 163 U. S., at 550. The Court deferred to local authorities in making its determination, noting that in inquiring into reasonableness “there must necessarily be a large discretion on the part of the legislature.” Ibid. The Court likewise paid heed to societal practices, local expectations, and practical consequences by looking to “the established usages, customs and traditions of the people, and with a view to the promotion of their comfort, and the preservation of the public peace and good order.” Ibid. Guided by these principles, the Court concluded: “[W]e cannot say that a law which authorizes or even requires the separation of the two races in public conveyances is unreasonable, or more obnoxious to the Fourteenth Amendment than the acts of Congress requiring separate schools for colored children in the District of Columbia.” Id., at 550–551.

   The segregationists in Brown embraced the arguments the Court endorsed in Plessy. Though Brown decisively rejected those arguments, today’s dissent replicates them to a distressing extent. Thus, the dissent argues that “[e]ach plan embodies the results of local experience and community consultation.” Post, at 47. Similarly, the segregationists made repeated appeals to societal practice and expectation. See, e.g., Brief for Appellees on Reargument in Briggs v. Elliott, O. T. 1953, No. 2, p. 76 (“[A] State has power to establish a school system which is capable of efficient administration, taking into account local problems and conditions”).[Footnote 21] The dissent argues that “weight [must be given] to a local school board’s knowledge, expertise, and concerns,” post, at 48, and with equal vigor, the segregationists argued for deference to local authorities. See, e.g., Brief for Kansas on Reargument in Brown v. Board of Education, O. T. 1953, No. 1, p. 14 (“We advocate only a concept of constitutional law that permits determinations of state and local policy to be made on state and local levels. We defend only the validity of the statute that enables the Topeka Board of Education to determine its own course”).[Footnote 22] The dissent argues that today’s decision “threatens to substitute for present calm a disruptive round of race-related litigation,” post, at 2, and claims that today’s decision “risks serious harm to the law and for the Nation,” post, at 65. The segregationists also relied upon the likely practical consequences of ending the state-imposed system of racial separation. See, e.g., Brief for Appellees on Reargument in Davis v. County School Board, O. T. 1953, No. 3, p. 37 (“Yet a holding that school segregation by race violates the Constitution will result in upheaval in all of those places not now subject to Federal judicial scrutiny. This Court has made many decisions of widespread effect; none would affect more people more directly in more fundamental interests and, in fact, cause more chaos in local government than a reversal of the decision in this case”).[Footnote 23] And foreshadowing today’s dissent, the segregationists most heavily relied upon judicial precedent. See, e.g., Brief for Appellees on Reargument in Briggs v. Elliott, O. T. 1953, No. 2, p. 59 (“[I]t would be difficult indeed to find a case so favored by precedent as is the case for South Carolina here”).[Footnote 24]

   The similarities between the dissent’s arguments and the segregationists’ arguments do not stop there. Like the dissent, the segregationists repeatedly cautioned the Court to consider practicalities and not to embrace too theoretical a view of the Fourteenth Amendment.[Footnote 25] And just as the dissent argues that the need for these programs will lessen over time, the segregationists claimed that reliance on segregation was lessening and might eventually end.[Footnote 26]

   What was wrong in 1954 cannot be right today.[Footnote 27] Whatever else the Court’s rejection of the segregationists’ arguments in Brown might have established, it certainly made clear that state and local governments cannot take from the Constitution a right to make decisions on the basis of race by adverse possession. The fact that state and local governments had been discriminating on the basis of race for a long time was irrelevant to the Brown Court. The fact that racial discrimination was preferable to the relevant communities was irrelevant to the Brown Court. And the fact that the state and local governments had relied on statements in this Court’s opinions was irrelevant to the Brown Court. The same principles guide today’s decision. None of the considerations trumpeted by the dissent is relevant to the constitutionality of the school boards’ race-based plans because no contextual detail—or collection of contextual details, post, at 2–22—can “provide refuge from the principle that under our Constitution, the government may not make distinctions on the basis of race.” Adarand, 515 U. S., at 240 (Thomas, J., concurring in part and concurring in judgment).[Footnote 28]

   In place of the color-blind Constitution, the dissent would permit measures to keep the races together and proscribe measures to keep the races apart.[Footnote 29] See post, at 28–34, 64–65. Although no such distinction is apparent in the Fourteenth Amendment, the dissent would constitutionalize today’s faddish social theories that embrace that distinction. The Constitution is not that malleable. Even if current social theories favor classroom racial engineering as necessary to “solve the problems at hand,” post, at 21, the Constitution enshrines principles independent of social theories. See Plessy, 163 U. S., at 559 (Harlan, J., dissenting) (“The white race deems itself to be the dominant race in this country. And so it is, in prestige, in achievements, in education, in wealth and in power. So, I doubt not, it will continue to be for all time … . But in view of the Constitution, in the eye of the law, there is in this country no superior, dominant, ruling class of citizens. … Our Constitution is color-blind, and neither knows nor tolerates classes among citizens”). Indeed, if our history has taught us anything, it has taught us to beware of elites bearing racial theories.[Footnote 30] See, e.g., Dred Scott v. Sandford, 19 How. 393, 407 (1857) (“[T]hey [members of the “negro African race”] had no rights which the white man was bound to respect”). Can we really be sure that the racial theories that motivated Dred Scott and Plessy are a relic of the past or that future theories will be nothing but beneficent and progressive? That is a gamble I am unwilling to take, and it is one the Constitution does not allow.

*  *  *

   The plans before us base school assignment decisions on students’ race. Because “[o]ur Constitution is color-blind, and neither knows nor tolerates classes among citizens,” such race-based decisionmaking is unconstitutional. Plessy, supra, at 559 (Harlan, J., dissenting). I concur in the Chief Justice’s opinion so holding.

Footnote 1  In this Court’s paradigmatic segregation cases, there was a local ordinance, state statute, or state constitutional provision requiring racial separation. See, e.g., Brief for Petitioners in Bolling v. Sharpe, O. T. 1952, No. 4, pp. 28–30 (cataloging state laws requiring separa- tion of the races); id., at App. A (listing “Statutory and Consti- tutional Provisions in the States Where Segregation in Education is Institutionalized”).

Footnote 2  The dissent refers repeatedly and reverently to “ ‘integration.’ ” However, outside of the context of remediation for past de jure segregation, “integration” is simply racial balancing. See post, at 37. Therefore, the school districts’ attempts to further “integrate” are properly thought of as little more than attempts to achieve a particular racial balance.

Footnote 3  The dissent’s assertion that these plans are necessary for the school districts to maintain their “hard-won gains” reveals its conflation of segregation and racial imbalance. Post, at 38. For the dissent’s purposes, the relevant hard-won gains are the present racial compositions in the individual schools in Seattle and Louisville. However, the actual hard-won gain in these cases is the elimination of the vestiges of the system of state-enforced racial separation that once existed in Louisville. To equate the achievement of a certain statistical mix in several schools with the elimination of the system of systematic de jure segregation trivializes the latter accomplishment. Nothing but an interest in classroom aesthetics and a hypersensitivity to elite sensibilities justifies the school districts’ racial balancing programs. See Part II–B, infra. But “the principle of inherent equality that underlies and infuses our Constitution” required the disestablishment of de jure segregation. See Adarand Constructors, Inc. v. Peńa, 515 U. S. 200, 240 (1995) (Thomas, J., concurring in part and concurring in judgment). Assessed in any objective manner, there is no comparison between the two.

Footnote 4  The dissent makes much of the supposed difficulty of determining whether prior segregation was de jure or de facto. See, e.g., post, at 19–20. That determination typically will not be nearly as difficult as the dissent makes it seem. In most cases, there either will or will not have been a state constitutional amendment, state statute, local ordinance, or local administrative policy explicitly requiring separation of the races. See, e.g., n. 1, supra. And even if the determination is difficult, it is one the dissent acknowledges must be made to determine what remedies school districts are required to adopt. Post, at 43.

Footnote 5  This Court’s opinion in McDaniel v. Barresi, 402 U. S. 39 (1971), fits comfortably within this framework. There, a Georgia school board voluntarily adopted a desegregation plan. At the time of Brown, v. Board of Education, 347 U. S. 483 (1954), Georgia’s Constitution required that “[s]eparate schools shall be provided for the white and colored races.” Ga. Const., Art. VII, §1, ch. 2–6401 (1948). Given that state law had previously required the school board to maintain a dual school system, the county was obligated to take measures to remedy its prior de jure segregation. This Court recognized as much in its opinion, which stated that the school board had an “affirmative duty to disestablish the dual school system.” McDaniel, supra, at 41.

Footnote 6  As I have explained elsewhere, the remedies this Court authorized lower courts to compel in early desegregation cases like Green and Swann were exceptional. See Missouri v. Jenkins, 515 U. S. 70, 124–125 (1995), (Thomas, J., concurring). Sustained resistance to Brown prompted the Court to authorize extraordinary race-conscious remedial measures (like compelled racial mixing) to turn the Constitution’s dictate to desegregate into reality. 515 U. S., at 125 (Thomas, J., concurring). Even if these measures were appropriate as remedies in the face of widespread resistance to Brown’s mandate, they are not forever insulated from constitutional scrutiny. Rather, “such powers should have been temporary and used only to overcome the widespread resistance to the dictates of the Constitution.” 515 U. S., at 125 (Thomas, J., concurring).

Footnote 7  Though the dissent cites every manner of complaint, record material, and scholarly article relating to Seattle’s race-based student assignment efforts, post, at 73–75, it cites no law or official policy that required separation of the races in Seattle’s schools. Nevertheless, the dissent tries to cast doubt on the historical fact that the Seattle schools were never segregated by law by citing allegations that the National Association for the Advancement of Colored People and other organizations made in court filings to the effect that Seattle’s schools were once segregated by law. See post, at 7–9, 23. These allegations were never proved and were not even made in this case. Indeed, the record before us suggests the contrary. See App. in No. 05–908, pp. 214a, 225a, 257a. Past allegations in another case provide no basis for resolving these cases.

Footnote 8  Contrary to the dissent’s argument, post, at 44, the Louisville school district’s interest in remedying its past de jure segregation did vanish the day the District Court found that Louisville had eliminated the vestiges of its historic de jure segregation. See Hampton v. Jefferson Cty. Bd. of Ed., 102 F. Supp. 2d 358, at 360 (WD Ky. 2000). If there were further remediation to be done, the District Court could not logically have reached the conclusion that Louisville “ha[d] eliminated the vestiges associated with the former policy of segregation and its pernicious effects.” Ibid. Because Louisville could use race-based measures only as a remedy for past de jure segregation, it is not “incoherent,” post, at 56, to say that race-based decisionmaking was allowed to Louisville one day—while it was still remedying—and forbidden to it the next—when remediation was finished. That seemingly odd turnaround is merely a result of the fact that the remediation of de jure segregation is a jealously guarded exception to the Equal Protection Clause’s general rule against government race-based decisionmaking.

Footnote 9  The dissent’s appeal to stare decisis, post, at 65, is particularly ironic in light of its apparent willingness to depart from these precedents, post, at 36–37.

Footnote 10  The idea that government racial classifications must be subjected to strict scrutiny did not originate in Adarand. As early as Loving v. Virginia, 388 U. S. 1 (1967), this Court made clear that government action that “rest[s] solely upon distinctions drawn according to race” had to be “subjected to the ‘most rigid scrutiny.’ ” Id., at 11 (quoting Korematsu v. United States, 323 U. S. 214, 216 (1944)); see also McLaughlin v. Florida, 379 U. S. 184, 196 (1964) (requiring a statute drawing a racial classification to be “necessary, and not merely rationally related, to accomplishment of a permissible state policy”); id., at 197 (Harlan, J., concurring) (“The necessity test … should be equally applicable in a case involving state racial discrimination”).

Footnote 11  At least one of the academic articles the dissent cites to support this proposition fails to establish a causal connection between the supposed educational gains realized by black students and racial mixing. See Hallinan, Diversity Effects on Student Outcomes: Social Science Evidence, 59 Ohio St. L. J. 733 (1998). In the pages following the ones the dissent cites, the author of that article remarks that “the main reason white and minority students perform better academically in majority white schools is likely that these schools provide greater opportunities to learn. In other words, it is not desegregation per se that improves achievement, but rather the learning advantages some desegregated schools provide.” Id., at 744. Evidence that race is a good proxy for other factors that might be correlated with educational benefits does not support a compelling interest in the use of race to achieve academic results.

Footnote 12  Of course, if the Seattle school board were truly committed to the notion that diversity leads directly to educational benefits, operating a school with such a high “nonwhite” enrollment would be a shocking dereliction of its duty to educate the students enrolled in that school.

Footnote 13  In fact, the available data from the Seattle school district appear to undercut the dissent’s view. A comparison of the test results of the schools in the last year the racial balancing program operated to the results in the 2004-to-2005 school year (in which student assignments were race-neutral) does not indicate the decline in black achieve- ment one would expect to find if black achievement were contin- gent upon a particular racial mix. See Washington State Report Card, online at http://reportcard.ospi.k12.wa.us/summary.aspx?schoolId= 1099&OrgType=4&reportLevel=School; http://reportcard.ospi.k12.wa.us/ summary.aspx?schoolId=1104&reportLevel=School&orgLinkId=1104&yrs=; http: // reportcard . ospi . k12 . wa . us / summary . aspx ? schoolId = 1061 & report Level = School & orgLinkId = 1061 & yrs = ; http: // reportcard . ospi.k12 . wa.us/ summary . aspx ? schoolId = 1043 & reportLevel = School & orgLinkId = 1043 & yrs= (showing that reading scores went up, not down, when Seattle’s race-based assignment program ended at Sealth High School, Ingraham High School, and Franklin High School—some of the schools most affected by the plan).

Footnote 14  The dissent accuses me of “feel[ing] confident that, to end invidious discrimination, one must end all governmental use of race-conscious criteria” and chastises me for not deferring to democratically elected majorities. See post, at 62. Regardless of what Justice Breyer’s goals might be, this Court does not sit to “create a society that includes all Americans” or to solve the problems of “troubled inner city schooling.” Ibid. We are not social engineers. The United States Constitution dictates that local governments cannot make decisions on the basis of race. Consequently, regardless of the perceived negative effects of racial imbalance, I will not defer to legislative majorities where the Constitution forbids it.

   It should escape no one that behind Justice Breyer’s veil of judicial modesty hides an inflated role for the Federal Judiciary. The dissent’s approach confers on judges the power to say what sorts of discrimination are benign and which are invidious. Having made that determination (based on no objective measure that I can detect), a judge following the dissent’s approach will set the level of scrutiny to achieve the desired result. Only then must the judge defer to a democratic majority. In my view, to defer to one’s preferred result is not to defer at all.

Footnote 15  The notion that a “democratic” interest qualifies as a compelling interest (or constitutes a part of a compelling interest) is proposed for the first time in today’s dissent and has little basis in the Constitution or our precedent, which has narrowly restricted the interests that qualify as compelling. See Grutter v. Bollinger, 539 U. S. 306, 351–354 (2003) (Thomas, J., concurring in part and dissenting in part). The Fourteenth Amendment does not enact the dissent’s newly minted understanding of liberty. See Lochner v. New York, 198 U. S. 45, 75 (1905) (Holmes, J., dissenting) (“The Fourteenth Amendment does not enact Mr. Herbert Spencer’s Social Statics”).

Footnote 16  The dissent does not explain how its recognition of an interest in teaching racial understanding and cooperation here is consistent with the Court’s rejection of a similar interest in Wygant. In Wygant, a school district justified its race-based teacher-layoff program in part on the theory that “minority teachers provided ‘role models’ for minority students and that a racially ‘diverse’ faculty would improve the education of all students.” Grutter, supra, at 352 (opinion of Thomas, J.) (citing Brief for Respondents, O. T. 1984, No. 84–1340, pp. 27–28; 476 U. S., at 315 (Stevens, J., dissenting)). The Court rejected the interests asserted to justify the layoff program as insufficiently compelling. Wygant, 476 U. S., at 275–276 (plurality opinion); id., at 295 (White, J., concurring in judgment). If a school district has an interest in teaching racial understanding and cooperation, there is no logical reason why that interest should not extend to the composition of the teaching staff as well as the composition of the student body. The dissent’s reliance on this interest is, therefore, inconsistent with Wygant.

Footnote 17  Outside the school context, this Court’s cases reflect the fact that racial mixing does not always lead to harmony and understanding. In Johnson v. California, 543 U. S. 499 (2005), this Court considered a California prison policy that separated inmates racially. Id., at 525–528 (Thomas, J., dissenting). That policy was necessary because of “numerous incidents of racial violence.” Id., at 502; id., at 532–534 (Thomas, J., dissenting). As a result of this Court’s insistence on strict scrutiny of that policy, but see id., at 538–547, inmates in the California prisons were killed. See Beard v. Banks, 548 U. S. ___, ___ (2006) (Thomas, J., concurring in judgment) (noting that two were killed and hundreds were injured in race rioting subsequent to this Court’s decision in Johnson).

Footnote 18  After discussing the “democratic element,” the dissent repeats its assertion that the social science evidence supporting that interest is “sufficiently strong to permit a school board to determine … that this interest is compelling.” Post, at 40. Again, though, the school boards have no say in deciding whether an interest is compelling. Strict scrutiny of race-based government decisionmaking is more searching than Chevron-style administrative review for reasonableness. See Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837, 845 (1984).

Footnote 19  The dissent half-heartedly attacks the historical underpinnings of the color-blind Constitution. Post, at 28–29. I have no quarrel with the proposition that the Fourteenth Amendment sought to bring former slaves into American society as full members. Post, at 28 (citing Slaughter-House Cases, 16 Wall. 36, 71–72 (1873)). What the dissent fails to understand, however, is that the color-blind Constitution does not bar the government from taking measures to remedy past state-sponsored discrimination—indeed, it requires that such measures be taken in certain circumstances. See, e.g., Part I–B, supra. Race-based government measures during the 1860’s and 1870’s to remedy state-enforced slavery were therefore not inconsistent with the color-blind Constitution.

Footnote 20  See also Juris. Statement in Davis v. County School Board, O. T. 1952, No. 3, p. 8 (“[W]e take the unqualified position that the Fourteenth Amendment has totally stripped the state of power to make race and color the basis for governmental action”); Tr. of Oral Arg. in Brown v. Board of Education, O. T. 1952, No. 1, p. 7 (“We have one fundamental contention which we will seek to develop in the course of this argument, and that contention is that no State has any authority under the equal-protection clause of the Fourteenth Amendment to use race as a factor in affording educational opportunities among its citizens”); Tr. of Oral Arg. in Briggs v. Elliott, O. T. 1953, No. 2, p. 50 (“[T]he state is deprived of any power to make any racial classifications in any governmental field”).

Footnote 21  See also Brief for Appellees in Davis v. County School Board, O. T. 1952, No. 3, p. 1 (“[T]he Court is asked … to outlaw the fixed policies of the several States which are based on local social conditions well known to the respective legislatures”); id., at 9 (“For this purpose, Virginia history and present Virginia conditions are important”); Tr. of Oral Arg. in Davis v. County School Board, O. T. 1952, No. 3, p. 57 (“[T]he historical background that exists, certainly in this Virginia situation, with all the strife and the history that we have shown in this record, shows a basis, a real basis, for the classification that has been made”); id., at 69 (describing the potential abolition of segregation as “contrary to the customs, the traditions and the mores of what we might claim to be a great people, established through generations, who themselves are fiercely and irrevocably dedicated to the preservation of the white and colored races”). Accord, post, at 68 (“Today, almost 50 years later, attitudes toward race in this Nation have changed dramatically. Many parents, white and black alike, want their children to attend schools with children of different races. Indeed, the very school districts that once spurned integration now strive for it. The long history of their efforts reveals the complexities and difficulties they have faced”); post, at 21 (emphasizing the importance of “local circumstances” and encouraging different localities to “try different solutions to common problems and gravitate toward those that prove most successful or seem to them best to suit their individual needs” (citations and internal quotation marks omitted)); post, at 48 (emphasizing the school districts’ “40-year history” during which both school districts have tried numerous approaches “to achieve more integrated schools”); post, at 63 (“[T]he histories of Louisville and Seattle reveal complex circumstances and a long tradition of conscientious efforts by local school boards”).

Footnote 22  See also Brief for Appellees in Brown v. Board of Education, O. T. 1952, No. 1, p. 29 (“ ‘It is universally held, therefore, that each state shall determine for itself, subject to the observance of the fundamental rights and liberties guaranteed by the federal Constitution, how it shall exercise the police power … . And in no field is this right of the several states more clearly recognized than in that of public education’ ” (quoting Briggs v. Elliott, 98 F. Supp. 529, 532 (SC 1951))); Brief for Appellees in Briggs v. Elliott, O. T. 1952, No. 2, p. 7 (“Local self-government in local affairs is essential to the peace and happiness of each locality and to the strength and stability of our whole federal system. Nowhere is this more profoundly true than in the field of education”); Tr. of Oral Arg. in Briggs v. Elliott, O. T. 1952, No. 2, pp. 54–55 (“What is the great national and federal policy on this matter? Is it not a fact that the very strength and fiber of our federal system is local self-government in those matters for which local action is competent? Is it not of all the activities of government the one which most nearly approaches the hearts and minds of people, the question of the education of their young? Is it not the height of wisdom that the manner in which that shall be conducted should be left to those most immediately affected by it, and that the wishes of the parents, both white and colored, should be ascertained before their children are forced into what may be an unwelcome contact?”). Accord, post, at 48 (“[L]ocal school boards better understand their own communities and have a better knowledge of what in practice will best meet the educational needs of their pupils”); post, at 66 (“[W]hat of respect for democratic local decisionmaking by States and school boards?”); ibid. (explaining “that the Constitution grants local school districts a significant degree of leeway”).

Footnote 23  See also Reply Brief for Appellees in Davis v. County School Board, O. T. 1953, No. 3, p. 17 (“The Court is … dealing with thousands of local school districts and schools. Is each to be the subject of litigation in the District Courts?”); Brief for Kansas on Reargument in Brown v. Board of Education, O. T. 1953, No. 1, p. 51 (“The delicate nature of the problem of segregation and the paramount interest of the State of Kansas in preserving the internal peace and tranquility of its people indicates that this is a question which can best be solved on the local level, at least until Congress declares otherwise”). Accord, post, at 61 (“At a minimum, the plurality’s views would threaten a surge of race-based litigation. Hundreds of state and federal statutes and regulations use racial classifications for educational or other purposes… . In many such instances, the contentious force of legal challenges to these classifications, meritorious or not, would displace earlier calm”); post, at 65 (“Indeed, the consequences of the approach the Court takes today are serious. Yesterday, the plans under review were lawful. Today, they are not”); post, at 66 (predicting “further litigation, aggravating race-related conflict”).

Footnote 24  See also Statement of Appellees Opposing Jurisdiction and Motion to Dismiss or Affirm in Davis v. County School Board, O. T. 1952, No. 3, p. 5 (“[I]t would be difficult to find from any field of law a legal principle more repeatedly and conclusively decided than the one sought to be raised by appellants”); Brief for Appellees in Davis v. County School Board, O. T. 1953, No. 3, p. 46–47 (“If this case were to be decided solely on the basis of precedent, this brief could have been much more limited. There is ample precedent in the decisions of this Court to uphold school segregation”); Brief for Petitioners in Gebhart v. Belton, O. T. 1952, No. 5, p. 27 (“Respondents ask this Court to upset a long established and well settled principle recognized by numerous state Legislatures, and Courts, both state and federal, over a long period of years”); Tr. of Oral Arg. in Briggs v. Elliott, O. T. 1953, No. 2, p. 79 (“But be that doctrine what it may, somewhere, sometime to every principle comes a moment of repose when it has been so often announced, so confidently relied upon, so long continued, that it passes the limits of judicial discretion and disturbance… . We relied on the fact that this Court had not once but seven times, I think it is, pronounced in favor of the separate but equal doctrine. We relied on the fact that the courts of last appeal of some sixteen or eighteen States have passed upon the validity of the separate but equal doctrine vis-a-vis the Fourteenth Amendment. We relied on the fact that Congress has continuously since 1862 segregated its schools in the District of Columbia”); Brief for Appellees in Briggs v. Elliott, O. T. 1952, No. 2, App. D (collecting citations of state and federal cases “[w]hich [e]nunciate the [p]rinciple that [s]tate [l]aws [p]roviding for [r]acial [s]egregation in the [p]ublic [s]chools do not [c]onflict with the Fourteenth Amendment”). Accord, post, at 22 (“[T]he Court set forth in Swann a basic principle of constitutional law—a principle of law that has found wide acceptance in the legal culture” (citations and internal quotation marks omitted)); post, at 25 (“Lower state and federal courts had considered the matter settled and uncontroversial even before this Court decided Swann”); post, at 26 (“Numerous state and federal courts explicitly relied upon Swann’s guidance for decades to follow”); post, at 27 (stating “how lower courts understood and followed Swann’s enunciation of the relevant legal principle”); post, at 30 (“The constitutional principle enunciated in Swann, reiterated in subsequent cases, and relied upon over many years, provides, and has widely been thought to provide, authoritative legal guidance”); post, at 61 (“[T]oday’s opinion will require setting aside the laws of several States and many local communities”); post, at 66 (“And what has happened to Swann? To McDaniel? To Crawford? To Harris? To School Committee of Boston? To Seattle School Dist. No. 1? After decades of vibrant life, they would all, under the plurality’s logic, be written out of the law”).

Footnote 25  Compare Brief for Appellees in Davis v. County School Board, O. T. 1952, No. 3, p. 16–17 (“ ‘It is by such practical considerations based on experience rather than by theoretical inconsistencies that the question of equal protection is to be answered’ ” (quoting Railway Express Agency, Inc. v. New York, 336 U. S. 110 (1949))); Brief for Appellees on Reargument in Davis v. County School Board, O. T. 1953, No. 3, p. 76 (“The question is a practical one for them to solve; it is not subject to solution in the theoretical realm of abstract principles”); Tr. of Oral Arg. in Davis v. County School Board, O. T. 1953, No. 4, p. 86 (“[Y]ou cannot talk about this problem just in a vacuum in the manner of a law school discussion”), with post, at 57 (“The Founders meant the Constitution as a practical document”).

Footnote 26  Compare Brief for Kansas on Reargument in Brown v. Board of Education, O. T. 1953, No. 1, p. 57 (“[T]he people of Kansas . . . are abandoning the policy of segregation whenever local conditions and local attitudes make it feasible”), Brief for Appellees on Reargument in Davis v. County School Board, O. T. 1953, No. 3, p. 76 (“As time passes, it may well be that segregation will end”), with post, at 19 (“[T]hey use race-conscious criteria in limited and gradually diminishing ways”); post, at 48 (“[E]ach plan’s use of race-conscious elements is diminished compared to the use of race in preceding integration plans”); post, at 55 (describing the “historically-diminishing use of race” in the school districts).

Footnote 27  It is no answer to say that these cases can be distinguished from Brown because Brown involved invidious racial classifications whereas the racial classifications here are benign. See post, at 62. How does one tell when a racial classification is invidious? The segregationists in Brown argued that their racial classifications were benign, not invidious. See Tr. of Oral Arg. in Briggs v. Elliott, O. T. 1953, No. 2, p. 83 (“It [South Carolina] is confident of its good faith and intention to produce equality for all of its children of whatever race or color. It is convinced that the happiness, the progress and the welfare of these children is best promoted in segregated schools”); Brief for Appellees on Reargument in Davis v. County School Board, O. T. 1953, No. 3, p. 82–83 (“Our many hours of research and investigation have led only to confirmation of our view that segregation by race in Virginia’s public schools at this time not only does not offend the Constitution of the United States but serves to provide a better education for living for the children of both races”); Tr. of Oral Arg. in Davis v. County School Board, O. T. 1952, No. 3, p. 71 (“[T]o make such a transition, would undo what we have been doing, and which we propose to continue to do for the uplift and advancement of the education of both races. It would stop this march of progress, this onward sweep”). It is the height of arrogance for Members of this Court to assert blindly that their motives are better than others.

Footnote 28  See also id., at 8–9 (“It has been urged that [these state laws and policies] derive validity as a consequence of a long duration supported and made possible by a long line of judicial decisions, including expressions in some of the decisions of this Court. At the same time, it is urged that these laws are valid as a matter of constitutionally permissible social experimentation by the States. On the matter of stare decisis, I submit that the duration of the challenged practice, while it is persuasive, is not controlling… . As a matter of social experimentation, the laws in question must satisfy the requirements of the Constitution. While this Court has permitted the States to legislate or otherwise officially act experimentally in the social and economic fields, it has always recognized and held that this power is subject to the limitations of the Constitution, and that the tests of the Constitution must be met”); Reply Brief for Appellants in Briggs v. Elliott, O. T. 1953, No. 2, pp. 18–19 (“The truth of the matter is that this is an attempt to place local mores and customs above the high equalitarian principles of our Government as set forth in our Constitution and particularly the Fourteenth Amendment. This entire contention is tantamount to saying that the vindication and enjoyment of constitutional rights recognized by this Court as present and personal can be postponed whenever such postponement is claimed to be socially desirable”).

Footnote 29  The dissent does not face the complicated questions attending its proposed standard. For example, where does the dissent’s principle stop? Can the government force racial mixing against the will of those being mixed? Can the government force black families to relocate to white neighborhoods in the name if bringing the races together? What about historically black colleges, which have “established traditions and programs that might disproportionately appeal to one race or another”? United States v. Fordice, 505 U. S. 717, 749 (1992) (Thomas, J., concurring). The dissent does not and cannot answer these questions because the contours of the distinction it propounds rest entirely in the eye of the beholder.

Footnote 30  Justice Breyer’s good intentions, which I do not doubt, have the shelf life of Justice Breyer’s tenure. Unlike the dissenters, I am unwilling to delegate my constitutional responsibilities to local school boards and allow them to experiment with race-based decisionmaking on the assumption that their intentions will forever remain as good as Justice Breyer’s. See The Federalist No. 51, p. 349 (J. Cooke ed. 1961) (“If men were angels, no government would be necessary”). Indeed, the racial theories endorsed by the Seattle school board should cause the dissenters to question whether local school boards should be entrusted with the power to make decisions on the basis of race. The Seattle school district’s Website formerly contained the following definition of “cultural racism”: “Those aspects of society that overtly and covertly attribute value and normality to white people and whiteness, and devalue, stereotype, and label people of color as ‘other,’ different, less than, or render them invisible. Examples of these norms include defining white skin tones as nude or flesh colored, having a future time orientation, emphasizing individualism as opposed to a more collective ideology, defining one form of English as standard . . . .” See Harrell, School Web Site Removed: Examples of Racism Sparked Controversy, Seattle Post-Intelligencer, June 2, 2006, p. B1. After the site was removed, the district offered the comforting clarification that the site was not intended “ ‘to hold onto unsuccessful concepts such as melting pot or colorblind mentality.’ ” Ibid.; see also ante, at 22, n. 15 (plurality opinion).

   More recently, the school district sent a delegation of high school students to a “White Privilege Conference.” See Equity and Race Relations White Privilege Conference, https://www.seattleschools. org/area/equityandrace/whiteprivilegeconference.xml. One conference participant described “white privilege” as “an invisible package of unearned assets which I can count on cashing in each day, but about which I was meant to remain oblivious. White Privilege is like an invisible weightless knapsack of special provisions, maps, passports, codebooks, visas, clothes, tools, and blank checks.” See White Privilege Conference, Questions and Answers, http://www.uccs.edu/~wpc/ faqs.htm; see generally Westneat, School District’s Obsessed with Race, Seattle Times, Apr. 1, 2007, p. B1 (describing racial issues in Seattle schools).

OPINION OF KENNEDY, J.
PARENTS INVOLVED IN COMMUNITY SCHOOLS V.SEATTLE SCHOOL DIST. NO. 1
551 U. S. ____ (2007)
SUPREME COURT OF THE UNITED STATES
NOS. 05-908 AND 05-915

PARENTS INVOLVED IN COMMUNITY SCHOOLS, PETITIONER

05–908   v.

SEATTLE SCHOOL DISTRICT NO. 1 et al.

on writ of certiorari to the united states court of appeals for the ninth circuit

CRYSTAL D. MEREDITH, custodial parent and next friend of JOSHUA RYAN McDONALD, PETITIONER

05–915   v.

JEFFERSON COUNTY BOARD OF EDUCATION et al.

on writ of certiorari to the united states court of appeals for the sixth circuit

[June 28, 2007]

   Justice Kennedy, concurring in part and concurring in the judgment.

   The Nation’s schools strive to teach that our strength comes from people of different races, creeds, and cultures uniting in commitment to the freedom of all. In these cases two school districts in different parts of the country seek to teach that principle by having classrooms that reflect the racial makeup of the surrounding community. That the school districts consider these plans to be necessary should remind us our highest aspirations are yet unfulfilled. But the solutions mandated by these school districts must themselves be lawful. To make race matter now so that it might not matter later may entrench the very prejudices we seek to overcome. In my view the state-mandated racial classifications at issue, official labels proclaiming the race of all persons in a broad class of citizens—elementary school students in one case, high school students in another—are unconstitutional as the cases now come to us.

   I agree with The Chief Justice that we have jurisdiction to decide the cases before us and join Parts I and II of the Court’s opinion. I also join Parts III–A and III–C for reasons provided below. My views do not allow me to join the balance of the opinion by The Chief Justice, which seems to me to be inconsistent in both its approach and its implications with the history, meaning, and reach of the Equal Protection Clause. Justice Breyer’s dissenting opinion, on the other hand, rests on what in my respectful submission is a misuse and mistaken interpretation of our precedents. This leads it to advance propositions that, in my view, are both erroneous and in fundamental conflict with basic equal protection principles. As a consequence, this separate opinion is necessary to set forth my conclusions in the two cases before the Court.

I

   The opinion of the Court and Justice Breyer’s dissenting opinion (hereinafter dissent) describe in detail the history of integration efforts in Louisville and Seattle. These plans classify individuals by race and allocate benefits and burdens on that basis; and as a result, they are to be subjected to strict scrutiny. See Johnson v. California, 543 U. S. 499, 505–506 (2005); ante, at 11. The dissent finds that the school districts have identified a compelling interest in increasing diversity, including for the purpose of avoiding racial isolation. See post, at 37–45. The plurality, by contrast, does not acknowledge that the school districts have identified a compelling interest here. See ante, at 17–25. For this reason, among others, I do not join Parts III–B and IV. Diversity, depending on its meaning and definition, is a compelling educational goal a school district may pursue.

   It is well established that when a governmental policy is subjected to strict scrutiny, “the government has the burden of proving that racial classifications ‘are narrowly tailored measures that further compelling governmental interests.’ ” Johnson, supra, at 505 (quoting Adarand Constructors, Inc. v. Peńa, 515 U. S. 200, 227 (1995)). “Absent searching judicial inquiry into the justification for such race-based measures, there is simply no way of determining what classifications are ‘benign’ or ‘remedial’ and what classifications are in fact motivated by illegitimate notions of racial inferiority or simple racial politics.” Richmond v. J. A. Croson Co., 488 U. S. 469, 493 (1989) (plurality opinion). And the inquiry into less restrictive alternatives demanded by the narrow tailoring analysis requires in many cases a thorough understanding of how a plan works. The government bears the burden of justifying its use of individual racial classifications. As part of that burden it must establish, in detail, how decisions based on an individual student’s race are made in a challenged governmental program. The Jefferson County Board of Education fails to meet this threshold mandate.

   Petitioner Crystal Meredith challenges the district’s decision to deny her son Joshua McDonald a requested transfer for his kindergarten enrollment. The district concedes it denied his request “under the guidelines,” which is to say, on the basis of Joshua’s race. Brief for Respondents in No. 05–915, p. 10; see also App. in No. 05–915, p. 97. Yet the district also maintains that the guidelines do not apply to “kindergartens,” Brief for Respondents in No. 05–915, at 4, and it fails to explain the discrepancy. Resort to the record, including the parties’ Stipulation of Facts, further confuses the matter. See App. in No. 05–915, at 43 (“Transfer applications can be denied because of lack of available space or, for students in grades other than Primary 1 (kindergarten), the racial guidelines in the District’s current student assignment plan”); id., at 29 (“The student assignment plan does not apply to . . . students in Primary 1”); see also Stipulation of Facts in No. 3:02–CV–00620–JGH; Doc. 32, Exh. 44, p. 6 (2003–04 Jefferson County Public Schools Elementary Student Assignment Application, Section B) (“Assignment is made to a school for Primary 1 (Kindergarten) through Grade Five as long as racial guidelines are maintained. If the Primary 1 (Kindergarten) placement does not enhance racial balance, a new application must be completed for Primary 2 (Grade One)”).

   The discrepancy identified is not some simple and straightforward error that touches only upon the peripheries of the district’s use of individual racial classifications. To the contrary, Jefferson County in its briefing has explained how and when it employs these classifications only in terms so broad and imprecise that they cannot withstand strict scrutiny. See, e.g., Brief for Respondents in No. 05–915, at 4–10. While it acknowledges that racial classifications are used to make certain assignment decisions, it fails to make clear, for example, who makes the decisions; what if any oversight is employed; the precise circumstances in which an assignment decision will or will not be made on the basis of race; or how it is determined which of two similarly situated children will be subjected to a given race-based decision. See ibid.; see also App. in No. 05–915, at 38, 42 (indicating that decisions are “based on … the racial guidelines” without further explanation); id., at 81 (setting forth the blanket mandate that “[s]chools shall work cooperatively with each other and with central office to ensure that enrollment at all schools [in question] is within the racial guidelines annually and to encourage that the enrollment at all schools progresses toward the midpoint of the guidelines”); id., at 43, 76–77, 81–83; McFarland v. Jefferson Cty. Public Schools, 330 F. Supp. 2d 834, 837–845, 855–862 (WD Ky. 2004).

   When litigation, as here, involves a “complex, comprehensive plan that contains multiple strategies for achieving racially integrated schools,” Brief for Respondents in No. 05–915, at 4, these ambiguities become all the more problematic in light of the contradictions and confusions that result. Compare, e.g., App. in No. 05–915, at 37 (“Each [Jefferson County] school … has a designated geographic attendance area, which is called the ‘resides area’ of the school[, and each] such school is the ‘resides school’ for those students whose parent’s or guardian’s residence address is within the school’s geographic attendance area”); id., at 82 (“All elementary students … shall be assigned to the school which serves the area in which they reside”); and Brief for Respondents in No. 05–915, at 5 (“There are no selection criteria for admission to [an elementary school student’s] resides school, except attainment of the appropriate age and completion of the previous grade”), with App. in No. 05–915, at 38 (“Decisions to assign students to schools within each cluster are based on available space within the [elementary] schools and the racial guidelines in the District’s current student assignment plan”); id., at 82 (acknowledging that a student may not be assigned to his or her resides school if it “has reached … the extremes of the racial guidelines”).

   One can attempt to identify a construction of Jefferson County’s student assignment plan that, at least as a logical matter, complies with these competing propositions; but this does not remedy the underlying problem. Jefferson County fails to make clear to this Court—even in the limited respects implicated by Joshua’s initial assignment and transfer denial—whether in fact it relies on racial classifications in a manner narrowly tailored to the interest in question, rather than in the far-reaching, inconsistent, and ad hoc manner that a less forgiving reading of the record would suggest. When a court subjects governmental action to strict scrutiny, it cannot construe ambiguities in favor of the State.

   As for the Seattle case, the school district has gone further in describing the methods and criteria used to determine assignment decisions on the basis of individual racial classifications. See, e.g., Brief for Respondents in No. 05–908, p. 5–11. The district, nevertheless, has failed to make an adequate showing in at least one respect. It has failed to explain why, in a district composed of a diversity of races, with fewer than half of the students classified as “white,” it has employed the crude racial categories of “white” and “non-white” as the basis for its assignment decisions. See, e.g., id., at 1–11.

   The district has identified its purposes as follows: “(1) to promote the educational benefits of diverse school enrollments; (2) to reduce the potentially harmful effects of racial isolation by allowing students the opportunity to opt out of racially isolated schools; and (3) to make sure that racially segregated housing patterns did not prevent non-white students from having equitable access to the most popular over-subscribed schools.” Id., at 19. Yet the school district does not explain how, in the context of its diverse student population, a blunt distinction between “white” and “non-white” furthers these goals. As the Court explains, “a school with 50 percent Asian-American students and 50 percent white students but no African-American, Native-American, or Latino students would qualify as balanced, while a school with 30 percent Asian-American, 25 percent African-American, 25 percent Latino, and 20 percent white students would not.” Ante, at 15–16; see also Brief for United States as Amicus Curiae in No. 05–908, pp. 13–14. Far from being narrowly tailored to its purposes, this system threatens to defeat its own ends, and the school district has provided no convincing explanation for its design. Other problems are evident in Seattle’s system, but there is no need to address them now. As the district fails to account for the classification system it has chosen, despite what appears to be its ill fit, Seattle has not shown its plan to be narrowly tailored to achieve its own ends; and thus it fails to pass strict scrutiny.

II

   Our Nation from the inception has sought to preserve and expand the promise of liberty and equality on which it was founded. Today we enjoy a society that is remarkable in its openness and opportunity. Yet our tradition is to go beyond present achievements, however significant, and to recognize and confront the flaws and injustices that remain. This is especially true when we seek assurance that opportunity is not denied on account of race. The enduring hope is that race should not matter; the reality is that too often it does.

   This is by way of preface to my respectful submission that parts of the opinion by The Chief Justice imply an all-too-unyielding insistence that race cannot be a factor in instances when, in my view, it may be taken into account. The plurality opinion is too dismissive of the legitimate interest government has in ensuring all people have equal opportunity regardless of their race. The plurality’s postulate that “[t]he way to stop discrimination on the basis of race is to stop discriminating on the basis of race,” ante, at 40–41, is not sufficient to decide these cases. Fifty years of experience since Brown v. Board of Education, 347 U. S. 483 (1954), should teach us that the problem before us defies so easy a solution. School districts can seek to reach Brown’s objective of equal educational opportunity. The plurality opinion is at least open to the interpretation that the Constitution requires school districts to ignore the problem of de facto resegregation in schooling. I cannot endorse that conclusion. To the extent the plurality opinion suggests the Constitution mandates that state and local school authorities must accept the status quo of racial isolation in schools, it is, in my view, profoundly mistaken.

   The statement by Justice Harlan that “[o]ur Constitution is color-blind” was most certainly justified in the context of his dissent in Plessy v. Ferguson, 163 U. S. 537, 559 (1896). The Court’s decision in that case was a grievous error it took far too long to overrule. Plessy, of course, concerned official classification by race applicable to all persons who sought to use railway carriages. And, as an aspiration, Justice Harlan’s axiom must command our assent. In the real world, it is regrettable to say, it cannot be a universal constitutional principle.

   In the administration of public schools by the state and local authorities it is permissible to consider the racial makeup of schools and to adopt general policies to encourage a diverse student body, one aspect of which is its racial composition. Cf. Grutter v. Bollinger, 539 U. S. 306 (2003); id., at 387–388 (Kennedy, J., dissenting). If school authorities are concerned that the student-body compositions of certain schools interfere with the objective of offering an equal educational opportunity to all of their students, they are free to devise race-conscious measures to address the problem in a general way and without treating each student in different fashion solely on the basis of a systematic, individual typing by race.

   School boards may pursue the goal of bringing together students of diverse backgrounds and races through other means, including strategic site selection of new schools; drawing attendance zones with general recognition of the demographics of neighborhoods; allocating resources for special programs; recruiting students and faculty in a targeted fashion; and tracking enrollments, performance, and other statistics by race. These mechanisms are race conscious but do not lead to different treatment based on a classification that tells each student he or she is to be defined by race, so it is unlikely any of them would demand strict scrutiny to be found permissible. See Bush v. Vera, 517 U. S. 952, 958 (1996) (plurality opinion) (“Strict scrutiny does not apply merely because redistricting is performed with consciousness of race… . Electoral district lines are ‘facially race neutral’ so a more searching inquiry is necessary before strict scrutiny can be found applicable in redistricting cases than in cases of ‘classifications based explicitly on race’ ” (quoting Adarand, 515 U. S., at 213)). Executive and legislative branches, which for generations now have considered these types of policies and procedures, should be permitted to employ them with candor and with confidence that a constitutional violation does not occur whenever a decisionmaker considers the impact a given approach might have on students of different races. Assigning to each student a personal designation according to a crude system of individual racial classifications is quite a different matter; and the legal analysis changes accordingly.

   Each respondent has asserted that its assignment of individual students by race is permissible because there is no other way to avoid racial isolation in the school districts. Yet, as explained, each has failed to provide the support necessary for that proposition. Cf. Croson, 488 U. S., at 501 (“The history of racial classifications in this country suggests that blind judicial deference to legislative or executive pronouncements of necessity has no place in equal protection analysis”). And individual racial classifications employed in this manner may be considered legitimate only if they are a last resort to achieve a compelling interest. See id., at 519 (Kennedy, J., concurring in part and concurring in judgment).

   In the cases before us it is noteworthy that the number of students whose assignment depends on express racial classifications is limited. I join Part III–C of the Court’s opinion because I agree that in the context of these plans, the small number of assignments affected suggests that the schools could have achieved their stated ends through different means. These include the facially race-neutral means set forth above or, if necessary, a more nuanced, individual evaluation of school needs and student characteristics that might include race as a component. The latter approach would be informed by Grutter, though of course the criteria relevant to student placement would differ based on the age of the students, the needs of the parents, and the role of the schools.

III

   The dissent rests on the assumptions that these sweeping race-based classifications of persons are permitted by existing precedents; that its confident endorsement of race categories for each child in a large segment of the community presents no danger to individual freedom in other, prospective realms of governmental regulation; and that the racial classifications used here cause no hurt or anger of the type the Constitution prevents. Each of these premises is, in my respectful view, incorrect.

A

   The dissent’s reliance on this Court’s precedents to justify the explicit, sweeping, classwide racial classifications at issue here is a misreading of our authorities that, it appears to me, tends to undermine well-accepted principles needed to guard our freedom. And in his critique of that analysis, I am in many respects in agreement with The Chief Justice. The conclusions he has set forth in Part III–A of the Court’s opinion are correct, in my view, because the compelling interests implicated in the cases before us are distinct from the interests the Court has recognized in remedying the effects of past intentional discrimination and in increasing diversity in higher education. See ante, at 12–13. As the Court notes, we recognized the compelling nature of the interest in remedying past intentional discrimination in Freeman v. Pitts, 503 U. S. 467, 494 (1992), and of the interest in diversity in higher education in Grutter. At the same time, these compelling interests, in my view, do help inform the present inquiry. And to the extent the plurality opinion can be interpreted to foreclose consideration of these interests, I disagree with that reasoning.

   As to the dissent, the general conclusions upon which it relies have no principled limit and would result in the broad acceptance of governmental racial classifications in areas far afield from schooling. The dissent’s permissive strict scrutiny (which bears more than a passing resemblance to rational-basis review) could invite widespread governmental deployment of racial classifications. There is every reason to think that, if the dissent’s rationale were accepted, Congress, assuming an otherwise proper exercise of its spending authority or commerce power, could mandate either the Seattle or the Jefferson County plans nationwide. There seems to be no principled rule, moreover, to limit the dissent’s rationale to the context of public schools. The dissent emphasizes local control, see post, at 48–49, the unique history of school desegregation, see post, at 2, and the fact that these plans make less use of race than prior plans, see post, at 57, but these factors seem more rhetorical than integral to the analytical structure of the opinion.

   This brings us to the dissent’s reliance on the Court’s opinions in Gratz v. Bollinger, 539 U. S. 244 (2003), and Grutter, 539 U. S. 306. If today’s dissent said it was adhering to the views expressed in the separate opinions in Gratz and Grutter, see Gratz, 539 U. S., at 281 (Breyer, J., concurring in judgment); id., at 282 (Stevens, J., dissenting); id., at 291 (Souter, J., dissenting); id., at 298 (Ginsburg, J., dissenting); Grutter, supra, at 344 (Ginsburg, J., concurring), that would be understandable, and likely within the tradition—to be invoked, in my view, in rare instances—that permits us to maintain our own positions in the face of stare decisis when fundamental points of doctrine are at stake. See, e.g., Federal Maritime Comm’n v. South Carolina Ports Authority, 535 U. S. 743, 770 (2002) (Stevens, J., dissenting). To say, however, that we must ratify the racial classifications here at issue based on the majority opinions in Gratz and Grutter is, with all respect, simply baffling.

   Gratz involved a system where race was not the entire classification. The procedures in Gratz placed much less reliance on race than do the plans at issue here. The issue in Gratz arose, moreover, in the context of college admissions where students had other choices and precedent supported the proposition that First Amendment interests give universities particular latitude in defining diversity. See Regents of Univ. of Cal. v. Bakke, 438 U. S. 265, 312–314 (1978) (opinion of Powell, J.). Even so the race factor was found to be invalid. Gratz, supra, at 251. If Gratz is to be the measure, the racial classification systems here are a fortiori invalid. If the dissent were to say that college cases are simply not applicable to public school systems in kindergarten through high school, this would seem to me wrong, but at least an arguable distinction. Under no fair reading, though, can the majority opinion in Gratz be cited as authority to sustain the racial classifications under consideration here.

   The same must be said for the controlling opinion in Grutter. There the Court sustained a system that, it found, was flexible enough to take into account “all pertinent elements of diversity,” 539 U. S., at 341 (internal quotation marks omitted), and considered race as only one factor among many, id., at 340. Seattle’s plan, by contrast, relies upon a mechanical formula that has denied hundreds of students their preferred schools on the basis of three rigid criteria: placement of siblings, distance from schools, and race. If those students were considered for a whole range of their talents and school needs with race as just one consideration, Grutter would have some application. That, though, is not the case. The only support today’s dissent can draw from Grutter must be found in its various separate opinions, not in the opinion filed for the Court.

B

   To uphold these programs the Court is asked to brush aside two concepts of central importance for determining the validity of laws and decrees designed to alleviate the hurt and adverse consequences resulting from race discrimination. The first is the difference between de jure and de facto segregation; the second, the presumptive invalidity of a State’s use of racial classifications to differentiate its treatment of individuals.

   In the immediate aftermath of Brown the Court addressed other instances where laws and practices enforced de jure segregation. See, e.g., Loving v. Virginia, 388 U. S. 1 (1967) (marriage); New Orleans City Park Improvement Assn. v. Detiege, 358 U. S. 54 (1958) (per curiam) (public parks); Gayle v. Browder, 352 U. S. 903 (1956) (per curiam) (buses); Holmes v. Atlanta, 350 U. S. 879 (1955) (per curiam) (golf courses); Mayor of Baltimore v. Dawson, 350 U. S. 877 (1955) (per curiam) (beaches). But with reference to schools, the effect of the legal wrong proved most difficult to correct. To remedy the wrong, school districts that had been segregated by law had no choice, whether under court supervision or pursuant to voluntary desegregation efforts, but to resort to extraordinary measures including individual student and teacher assignment to schools based on race. See, e.g., Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U. S. 1, 8–10 (1971); see also Croson, 488 U. S., at 519 (Kennedy, J., concurring in part and concurring in judgment) (noting that racial classifications “may be the only adequate remedy after a judicial determination that a State or its instrumentality has violated the Equal Protection Clause”). So it was, as the dissent observes, see post, at 13–14, that Louisville classified children by race in its school assignment and busing plan in the 1970’s.

   Our cases recognized a fundamental difference between those school districts that had engaged in de jure segregation and those whose segregation was the result of other factors. School districts that had engaged in de jure segregation had an affirmative constitutional duty to desegregate; those that were de facto segregated did not. Compare Green v. School Bd. of New Kent Cty., 391 U. S. 430, 437–438 (1968), with Milliken v. Bradley, 418 U. S. 717, 745 (1974). The distinctions between de jure and de facto segregation extended to the remedies available to governmental units in addition to the courts. For example, in Wygant v. Jackson Bd. of Ed., 476 U. S. 267, 274 (1986), the plurality noted: “This Court never has held that societal discrimination alone is sufficient to justify a racial classification. Rather, the Court has insisted upon some showing of prior discrimination by the governmental unit involved before allowing limited use of racial classifications in order to remedy such discrimination.” The Court’s decision in Croson, supra, reinforced the difference between the remedies available to redress de facto and de jure discrimination:

“To accept [a] claim that past societal discrimination alone can serve as the basis for rigid racial preferences would be to open the door to competing claims for ‘remedial relief’ for every disadvantaged group. The dream of a Nation of equal citizens in a society where race is irrelevant to personal opportunity and achievement would be lost in a mosaic of shifting preferences based on inherently unmeasurable claims of past wrongs.” Id., at 505–506.

   From the standpoint of the victim, it is true, an injury stemming from racial prejudice can hurt as much when the demeaning treatment based on race identity stems from bias masked deep within the social order as when it is imposed by law. The distinction between government and private action, furthermore, can be amorphous both as a historical matter and as a matter of present-day finding of fact. Laws arise from a culture and vice versa. Neither can assign to the other all responsibility for persisting injustices.

   Yet, like so many other legal categories that can overlap in some instances, the constitutional distinction between de jure and de facto segregation has been thought to be an important one. It must be conceded its primary function in school cases was to delimit the powers of the Judiciary in the fashioning of remedies. See, e.g., Milliken, supra, at 746. The distinction ought not to be altogether disregarded, however, when we come to that most sensitive of all racial issues, an attempt by the government to treat whole classes of persons differently based on the government’s systematic classification of each individual by race. There, too, the distinction serves as a limit on the exercise of a power that reaches to the very verge of constitutional authority. Reduction of an individual to an assigned racial identity for differential treatment is among the most pernicious actions our government can undertake. The allocation of governmental burdens and benefits, contentious under any circumstances, is even more divisive when allocations are made on the basis of individual racial classifications. See, e.g., Regents of Univ. of Cal. v. Bakke, 438 U. S. 265 (1978); Adarand, 515 U. S. 200.

   Notwithstanding these concerns, allocation of benefits and burdens through individual racial classifications was found sometimes permissible in the context of remedies for de jure wrong. Where there has been de jure segregation, there is a cognizable legal wrong, and the courts and legislatures have broad power to remedy it. The remedy, though, was limited in time and limited to the wrong. The Court has allowed school districts to remedy their prior de jure segregation by classifying individual students based on their race. See North Carolina Bd. of Ed. v. Swann, 402 U. S. 43, 45–46 (1971). The limitation of this power to instances where there has been de jure segregation serves to confine the nature, extent, and duration of governmental reliance on individual racial classifications.

   The cases here were argued upon the assumption, and come to us on the premise, that the discrimination in question did not result from de jure actions. And when de facto discrimination is at issue our tradition has been that the remedial rules are different. The State must seek alternatives to the classification and differential treatment of individuals by race, at least absent some extraordinary showing not present here.

C

   The dissent refers to an opinion filed by Judge Kozinski in one of the cases now before us, and that opinion relied upon an opinion filed by Chief Judge Boudin in a case presenting an issue similar to the one here. See post, at 35 (citing 426 F. 3d 1162, 1193–1196 (CA9 2005) (concurring opinion) (citing Comfort v. Lynn School Comm., 418 F. 3d 1, 27–29 (CA1 2005) (Boudin, C. J., concurring))). Though this may oversimplify the matter a bit, one of the main concerns underlying those opinions was this: If it is legitimate for school authorities to work to avoid racial isolation in their schools, must they do so only by indirection and general policies? Does the Constitution mandate this inefficient result? Why may the authorities not recognize the problem in candid fashion and solve it altogether through resort to direct assignments based on student racial classifications? So, the argument proceeds, if race is the problem, then perhaps race is the solution.

   The argument ignores the dangers presented by individual classifications, dangers that are not as pressing when the same ends are achieved by more indirect means. When the government classifies an individual by race, it must first define what it means to be of a race. Who exactly is white and who is nonwhite? To be forced to live under a state-mandated racial label is inconsistent with the dignity of individuals in our society. And it is a label that an individual is powerless to change. Governmental classifications that command people to march in different directions based on racial typologies can cause a new divisiveness. The practice can lead to corrosive discourse, where race serves not as an element of our diverse heritage but instead as a bargaining chip in the political process. On the other hand race-conscious measures that do not rely on differential treatment based on individual classifications present these problems to a lesser degree.

   The idea that if race is the problem, race is the instrument with which to solve it cannot be accepted as an analytical leap forward. And if this is a frustrating duality of the Equal Protection Clause it simply reflects the duality of our history and our attempts to promote freedom in a world that sometimes seems set against it. Under our Constitution the individual, child or adult, can find his own identity, can define her own persona, without state intervention that classifies on the basis of his race or the color of her skin.

*  *  *

   This Nation has a moral and ethical obligation to fulfill its historic commitment to creating an integrated society that ensures equal opportunity for all of its children. A compelling interest exists in avoiding racial isolation, an interest that a school district, in its discretion and expertise, may choose to pursue. Likewise, a district may consider it a compelling interest to achieve a diverse student population. Race may be one component of that diversity, but other demographic factors, plus special talents and needs, should also be considered. What the government is not permitted to do, absent a showing of necessity not made here, is to classify every student on the basis of race and to assign each of them to schools based on that classification. Crude measures of this sort threaten to reduce children to racial chits valued and traded according to one school’s supply and another’s demand.

   That statement, to be sure, invites this response: A sense of stigma may already become the fate of those separated out by circumstances beyond their immediate control. But to this the replication must be: Even so, measures other than differential treatment based on racial typing of individuals first must be exhausted.

   The decision today should not prevent school districts from continuing the important work of bringing together students of different racial, ethnic, and economic backgrounds. Due to a variety of factors—some influenced by government, some not—neighborhoods in our communities do not reflect the diversity of our Nation as a whole. Those entrusted with directing our public schools can bring to bear the creativity of experts, parents, administrators, and other concerned citizens to find a way to achieve the compelling interests they face without resorting to widespread governmental allocation of benefits and burdens on the basis of racial classifications.

   With this explanation I concur in the judgment of the Court.