dissenting.
I agree -with Mr. Justice Brennan that this case is not moot, and because of the significance of the issues raised I think it is important to reach the merits.
I
The University of Washington Law School received 1,601 applications for admission to its first-year class .beginning in September 1971. There were spaces available for only about 150 students, but in order to enroll this number the school eventually offered admission to 275 applicants. All applicants were put into two groups, one of which was considered under the minority admissions program. Thirty-seven of those, offered admission had indicated on an optional question on. their application that their' “dominant” ethnic origin was either/black, Chicano, American Indian, or Filipino, the four groups included hi the minurity admissions program. Answers to this optional question were apparently the sole basis *321upon "which eligibility for the program was determined. Eighteen of these 37 actually enrolled in the Law School.
In general, the admissions process proceeded as follows: An index called the Predicted First Year Average (Average) was calculated for each applicant on the basis of a formula combining the applicant’s score on the Law School Admission Test (LSAT) and his grades in his last two years in college.1 "On the basis of its experience with previous years’ applications, the Admissions Committee, consisting of faculty, administration, and students, concluded that the most outstanding applicants were those with averages above 77; the highest average of any applicant was 81. Applicants with averages above 77 were considered as their applications arrived by random distribution of their files to the members of the Committee who would read them and report their recommendations back to the Committee. As a result of the first three Committee meetings in. February, March, and April 1971, 78 applicants from this group were admitted, although virtually no other applicants were offered admission this early.2 By the final conclusion of *322the admissions process in August 1971, 147 applicants with averages above 77 had been admitted, including all applicants with averages above 78, and 93. of 105 applicants with averages, between 77 and 78.
Also beginning early in the admissions process was the culling out of applicants with averages below 74.5. These were reviewed by' the Chairman of the Admissions Committee, who had the authority to reject them summarily without further, consideration by the rest of the Committee. A small number of these applications were saved by the Chairman for Committee consideration on the basis of information in the file indicating greater promise than suggested by the Average. Finally during the early months the Committee accumulated the applications of those with averages between 74.5 and 77 to be considered at a later time when most of the applications had been received and thus could be compared with one another. Since DeFunis’ average was 76.23, he was in this middle group.
Beginning in their May meeting the Committee considered this middle group of applicants, whose folders had been randomly distributed to Committee. members for their recommendations to the Committee. Also considered at this time were remaining applicants with averages below 74.5 who had not. been summarily rejected, and some of those .with averages above 77 who had not been symmarily admitted, but instead held for further consideration. Each Committee member would consider the applications - competitively, following rough guide*323lines as to the proportion who could be offered admission. After the Committee had extended offers of admission to somewhat over 200 applicants, a waiting list was constructed in the same fashion, and was divided into four groups ranked by the Committee’s assessment of their applications. DeFunis was on this waiting list, but was ranked in the lowest quarter. He was ultimately told in August 1971 that there would be no room for - him.
Applicants who had indicated .on their application forms that they were either black, Chicano, American Indian, or Filipino were treated differently in several respects. Whatever their Averages, none-were given to the Committee Chairman for consideration of summary rejection, nor were they distributed randomly among Committee members, for consideration along with the other applications. Instead, all applications of black students were assigned separately to two particular Committee members: a first-year black ’daw student on the Committee, and a professor on the Committee who had worked the previous summer in a special program for disadvantaged college students considering application to the Law School.3 Applications from among the other three minority groups were assigned to an assistant dean who was on the Committee. The minority applications, while considered competitively with one another, were never directly compared to the remaining applications, either by the subcommittee or by the full Committee. As in the admissions process generally, the Committee sought to find “within the minority category, those persons who we thought had the highest probability of *324succeeding in Law School.” 4 In reviewing the minority-applications, the Committee attached less weight to the Average “in making a total judgmental evaluation as to the relative ability of the particular applicant to succeed in law school.” 82 Wash. 2d 11, 21, 507 P. 2d 1169, 1175. In its publicly distributed Guide to Applicants, the Committee explained that “[a]n applicant’s racial or ethnic background was considered as one factor in our general, attempt to convert formal credentials into realistic predictions.” 5
Thirty-seven minority applicants were admitted under this procedure. Of these, 36 had Averages below DeFunis’ 76.23, and 30 had Averages below 74.5, and thus would ordinarily have been summarily rejected by the Chairman. There were also 48 nonminority applicants admitted who had Averages below DeFunis. Twenty-three of these were returning veterans, see n. 2, supra, and 25 were others who presumably were admitted because of other *325factors in their applications that made them attractive candidates despite their relatively low Averages.
It is reasonable to conclude from the above facts that while other factors were considered by the Committee, and were on occasion crucial, the Average was for most applicants a heavily weighted factor, and was at the extremes virtually dispositive.6 A different balance was apparently struck, however, with regard to the minority applicants. Indeed, at oral argument, the respondents’ counsel advised us that were the minority applicants considered under the same procedure as was generally used, none of those who eventually enrolled at the Law School would have been admitted.
The educational policy choices confronting a university admissions committee are not ordinarily a subject for judicial oversight; clearly it is not for us but for the law n hool to decide which tests to employ, how heavily to weigh recommendations from professors or undergraduate grades, and what level of achievement on the chosen criteria are sufficient to demonstrate that the candidate is qualified for admission. What places this case in a special category is the fact that the school did not choose one .set of criteria but two, and then determined which to apply to a given applicant on the basis of his race. The *326Committee adopted this policy in order to achieve “a reasonable representation” of minority groups in the Law School. 82 Wash. 2d, at 20, 507 P. 2d, at 1175. Altho gh it may be speculated that the Committee sought to rectify what it perceived to be cultural or racial biases in tbeLSAT or in the candidates’ undergraduate records, the record in this case is devoid of any evidence of such bias, and the school has not sought to justify its procedures nn this basis.
Although testifying that “ [w] e do not have a quota...” the Law School dean explained that “ [w] e want a reasonable representation. We- will go down to reach it if we can,” without “taking people, who are unqualified in an absolute sense . . . .” Statement of Facts 420/ By “unqualified in an absolute sense” the dean meant candidates who “have no reasonable probable likelihood of having a chance of succeeding in the study of law . . . .” Ibid. But the dean conceded that in “reaching,” the school does take “some minority students who at least, viewed as a group, have a less such likelihood than the majority student group taken as a whole.” Id., at 423.
“Q. Of those who have made application to go to the law school, I am saying you are not taking the best qualified? '
“A. In total?
“Q. In total.
“A. In using that definition, yes.” Id., at 423-424.
It thus appears that by the Committee’s own assess-, ment, it admitted minority students who, by the tests given, seemed .less- qualified than some white students who were not accepted, in order to achieve a “reasonable representation.” In this regard it may be pointed out that for the year 1969-1970: — two years before the class, to which DeFunis was seeking admission — the Law School *327reported an enrollment of eight black students out of a total of 356.7 Defendants’’ Ex. 7. That percentage, approximately 2i2%, compares to a percentage of blacks in the population of Washington of approximately 2.1%.8
II
There was a time when law schools could follow tha advice of Wigmore, who believed that “the way to find out whether a boy has the makings of a competent lawyer is to see what he can do in a first year of law studies.” Wigmore, Juristic Psychopoyemetrology — Or, How tó Find Out Whether a Boy Has the Makings of a Lawyer, 24 Ill. L. Rev. 454, 463-464 (1929). In those days there were enough spaces to admit every , applicant who met minimal credentials, and they all could be given the opportunity to prove themselves at law school. But by the 1920’s many law schools found that they could not admit all minimally qualified applicants, and some selection process began.9 The pressure to use some kind- of admissions test mounted, and a number of schools instituted them. One early precursor to the modern day LSAT was the. Ferson-Stoddard Law Aptitude examination. Wigmore conducted his own study of that test with 50 student volunteers, and concluded that it “had" no substantial practical value.” Id., at 463. But his conclusions were not accepted, and • the harried law *328schools still sought some kind of admissions test which would simplify the process of judging applicants, and,in 1948 the LSAT was born. It has been with us ever since.10
The test purports to predict how successful the applicant will be in his first year of law school, and consists of a few hours’ worth of multiple-choice questions. But the answers the student can give to a multiple-choice question are limited by the creativity and intelligence, of the test-maker; the student with a better or more original understanding of the problem than the test-maker may realize that none of the alternative answers are any good, but there is no way for him to-demonstrate his understanding. <!It is obvious from the nature of the tests that they do not give the candidate a significant opportunity to express himself. If he is subtle in his choice of answers it will go against him; and yet there is no other way for him to show any individuality. If he is strong-minded, nonconformist* unusual, original, or creative — as so many of the truly important people are — he must stifle his impulses and conform as best he can to the norms -that the multiple-choice testers set up in their unimaginative, scientific way. The more profoundly gifted the candidate is, the more his resentment will rise against the mental strait jacket into which the testers would force his mind.” B. Hoffmann, The Tyranny of Testing 91-92 (1962). '
Those who make the" tests and the law schools which use them point, of course, to the high correlations between the test scores and the grades at law school the first year. E. g., Winterbottom, Comments ón “A Study of the Criteria for Legal Education and Admission to the *329Bar,” An Article by Dr. Thomas M. Goolsby, Jr., 21 J. Legal Ed. 75 (1968).. Certainly the tests do seem to do better than chance.' But they do not have the value that their deceptively precise scoring system suggests. The proponents’ own data show that, for example, most of those scoring in the bottom 20% on the test do better than that in law school — indeed six of every 100 of them will be in the top 20% of their law school class. Id., at 79. And no one knows how many of those who were not admitted because of their test scores would in fact have done well were they given the chance. There are many relevant factors, such as motivation, cultural backgrounds of specific minorities that the test cannot measure, and they inevitably must impair its value as a predictor.11 Of course, the law school that admits only those with the highest test scores finds that on the average they do much better, and thus the test is a convenient tool for the admissions committee. The price is. paid by the able student who for unknown reasons did not achieve that high score — perhaps even the minority with a different cultural background. Some testsj at least in the past, have been aimed at eliminating Jews.
The school can safely conclude that the applicant with a score of 750 should be admitted before one with a score of 500. The problem is that in many cases the choice will be between 643' and 602 or 574 and 528. The numbers create an illusion of difference tending to overwhelm other factors. “The wiser testers are well aware of the defects of the multiple-choice format and the danger of placing reliance on any one method of assessment to the exclusion of all others. What is distressing is how little their caveats have impressed the people who succumb to the propaganda of the . test-*330makers and use these tests mechanically as though they were a valid substitute for judgment.” Hoffmann, supra,, at 215.
Of course, the tests are not the only thing considered; here they were combined with the prelaw grades to produce a new number called the Average. The- grades have their own problems; one school’s A is another school’s C. And even to the extent that this formula predicts law school grades, its valúe is limited. The law student with lower grades may in the long pull of a legal career surpass those at the top of the class. “[L]aw school admissions criteria have operated within a hermetically sealed system; it is now beginning to leak. The traditional combination of LSAT and GPA [undergraduate grade point average] may have provided acceptable predictors of likely performance in law school in the past. . . . [But] [t]here is no clear evidence that the LSAT and GPA provide particularly good evaluators of the intrinsic or enriched ability of an individual to perform as a law student or lawyer in a functioning society undergoing change. Nor is there any clear evidence that grades and other evaluators of law school performance, and the bar examination, are particularly good predictors of competence or success as a lawyer.” Rosen, Equalizing Access to Legal Education: Special Programs for Law Students Who Are Not Admissible by Traditional Criteria, 1970 U. Tol. L. Rev. 321, 332-333.
But, by whatever techniques, the law school must make choices. Neither party has challenged the validity of the Average employed here as an admissions tool, and therefore consideration of its possible deficiencies is not presented as an issue.'. The Law School presented no evidence to show that adjustments in the process employed were used in order validly to compare ' applicants ' of • different races; instead, it' chose to avoid making such comparisons, Finally, *331although the Committee did consider other information in the files of all applicants, the Law School has made no effort to show that it was because of these additional factors that it admitted minority applicants who would otherwise have been rejected. To the contrary, the school appears to have conceded that by its own assessment — taking all factors into account — it admitted minority applicants who would have been rejected had they been white. We have no choice but to evaluate the Law School’s case as it has been made.
Ill
The Equal Protection Clause did not enact a requirement that law schools employ as the sole criterion for admissions-a formula based upon the LSAT and undergraduate grades, nor does it prohibit law schools from, evaluating an applicant’s prior achievements in light of the barriers that he had to Overcome. A black applicant who pulled himself out of the ghetto into a junior college may thereby demonstrate a level of motivation, perseverance, and ability that would lead a fairmintled admissions committee to conclude that he shows more promise for law study trian the son of a rich alumnus who achieved better grades at Harvard. That applicant would be offered admission not because he is black, but because as an individual he has shown he has the potential, while the Harvard man may have taken less advantage of the vastly superior opportunities offered him. Because of the weight of. the prior handicaps, that black applicant may not realize his full potential in the first year of law school, or even in the full three years, but in the long pull of a legal career his achievements may far outstrip those of his classmates whose esrlier records appeared superior by conventional criteria. There is currently no test available to the Admissions *332Committee that can predict such possibilities with assurance, but the Committee may nevertheless seek to gauge-it as best it can, and weigh this factor in its decisions. Such a policy would not be limited to blacks, or Chicanos or Filipinos, or American Indians, although undoubtedly groups such as these may in practice be the principal beneficiaries of it. But a poor Appalachian white, or a second generation Chinese in San Francisco, or some other American whose lineage is so diverse as to defy ethnic labels, may demonstrate similar potential and thus be accorded favorable consideration by the Committee.
The difference between such a policy and the one presented by this case is that the Committee would be making decisions on the basis of individual attributes, rather than according a preference solely on the basis of race. To be sure, the racial preference here was -not absolute— the Committee did not admit all applicants from the four favored groups. -But it did accord all such applicants a preference by applying, to an extent not precisely ascertainable from the record, different standards by which to judge their applications, with the result that the Committee admitted minority applicants who, in the school’s own judgment, were less promising than other applicants who were rejected. Furthermore, it is apparent that because the Admissions Committee compared minority applicants only with one another, it was necessary to reserve some proportion of the class for them, even if at the outset a precise number of places were not set aside.12 That proportion, apparently 15% to *33320%, was chosen because the school determined it to be “reasonable,” 13 although no explanation is provided as to how that number rather than some other was found appropriate. Without becoming embroiled in a semantic debate over whether this practice constitutes a “quota,” it is clear that, given the limitation on the total number of applicants who could be accepted, this policy did reduce the total number of places for which DeFunis could compete — solely on account of his race. Thus, as the Washington Supreme Court concluded, whatever label 'one wishes to apply to it, “the minority admissions policy is certainly not benign with respect to nonminority students who are displaced by it.” 82 Wash. 2d, at 32, 507 P. 2d, at 1182. A finding that the state school employed a racial classification in selecting its students subjects it to the strictest scrutiny under the Equal Protection Clause.
The consideration of race as a -measure of an applri cant’s qualification normally introduces a capricious and irrelevant factor working an invidious discrimination, Anderson v. Martin, 375 U. S. 399, 402; Loving v. Virginia, 388 U. S. 1, 10; Harper v. Virginia Board of Elections, 383 U. S. 663, 668. Once race is a starting point educators and courts are immediately embroiled in competing claims of different racial arid ethnic groups that would make difficult, manageable standards consistent *334with the Equal Protection Clause. “The clear and central purpose of the Fourteenth Amendment was to eliminate all official state sources of invidious racial discrimination in the States.” Loving, supra, at 10. The Law School's admissions policy cannot be reconciled with that purpose, unless cultural standards of a diverse rather than a homogeneous society are taken into account. The reason is that professional persons, particularly lawyers, are not selected for life in a computerized society. The Indian who walks to the beat of Chief Seattle of the Muckleshoot Tribe in Washington 14 has a different culture from examiners at law. schools.
The key to the problem is the consideration of each application in a racially neutral way. Since the LSAT reflects questions touching on cultural backgrounds, the Admissions Committee acted properly in my view in setting minority applications apart for separate processing. These minorities have cultural backgrounds that are vastly different' from the dominant Caucasian. Many Eskimos, American Indians, Filipinos, Chícanos, Asian Indians, Burmese, and Africans come from such, disparate backgrounds that a test sensitively tuned for most applicants would be wide of the mark for many minorities.
The melting pot is not designed to homogenize people, making them uniform in consistency. The melting pot as I understand it is a figure, of speech that depicts the wide diversities tolerated by the First Amendment under one flag. See 2 S. Morison & H. Commager, The Growth of the American Republic, c. VIII (4th ed. 1950). Minorities in our midst who are to serve actively in our public affairs should be chosen on talent and character alone, not on cultural orientation or. leanings.
*335I do know, coming as I do from Indian country in Washington, that many of the young Indians know little about Adam Smith or Karl Marx but are deeply imbued with the spirit and philosophy of Chief Robert B. Jim of the Yakimas, Chief Seattle of the Muckleshoots, and Chief Joseph of the Nez Perce which offer competitive attitudes towards life, fellow man, and nature.15
I do not know the extent to which blacks in this country are imbued with ideas of African Socialism.16 Leopold Senghor and Sékou Touré, the most articulate of African leaders, have held that modern African political philosophy is not oriented either to Marxism Or to capitalism.17 How far the reintroduction into educational curricula of ancient Africán art and history has reached the minds of young Afro-Americans I do not. know. But at least as respects Indians, blacks, and Chícanos — as well as those from Asian cultures — I think a separate classification of these applicants is warranted, lest race be a subtle force in eliminating minority members because of cultural differences.
Insofar as LSAT’s reflect the dimensions and orientation of the Organization Man they do a disservice to minorities. . I personally know that admissions tests were, once used to eliminate Jews. How many other minorities they aim at I do not know. My reaction is that the presence of an LSAT is sufficient warrant for a school to put racial minorities into a separate class in order better to probe their capacities and potentials.
. The merits of the present controversy cannot in my view be resolved on this record. A trial would *336involve the disclosure of hidden prejudices, if any, against certain minorities and the manner in which substitute measurements of one’s talents and character were employed in the conventional tests. I could agree with the majority of the Washington Supreme Court only if, on the record, it could be said that the Law School’s selection was racially neutral. The case, in my view", should be remanded for a new trial to consider, inter alia, whether the established LRAT's should be eliminated so far as racial minorities are concerned.
This does not mean that a separate LSAT must be designed for minority racial groups, although that might be a possibility. The reason for the separate treatment of minorities as a class is to make more certain that racial factors do not militate against an applicant or on his behalf.18
There is no constitutional right for any race to be preferred. The years of slavery did more than retard the progress of blacks. Even a,greater wrong was done the whites by creating arrogance instead of humility and by encouraging the growth of the fiction of a superior race. *337There is no superior person by constitutional standards. A DeFunis who is white is entitled to no advantage by reason of that fact; nor is he subject to any disability, no matter what his race or color; Whatever his race, he had a constitutional right to have his application considered on. its individual merits in a racially neutral manner..
The slate is not entirely clean. First, we have held that pro rata representation of the races is not required either on juries, see Cassell v. Texas, 339 U. S. 282, 286-287, or in public schools; Swann v. Charlotte-Mecklenburg Board of Education, 402 U. S. 1, 24. Moreover, in Hughes v. Superior Court, 339 U. S. 460, we reviewed the contempt convictions of pickets who sought by their demonstration to force an employer to prefer Negroes to whites in his hiring of clerks, in order to ensure that 50% of the employees were Negro. In finding that California could constitutionally enjoin the picketing there involved we quoted from the opinion of the California Supreme. Court, which noted that the pickets would “ ‘make the right to work for Lucky dependent not on fitness for the work nor on an equal right of all, regardless of race, to compete in an open market, but, rather, on 'membership in a particular race. If petitioners were upheld in their demand then other races, white, yellow, brown and red. would have equal rights to demand discriminatory hiring on a racial basis.' ” Id., at 463-464. We then noted that
“[t]o deny to California the right to ban picketing in the circumstances of this case would mean that there could be no prohibition of the pressure of picketing to secure proportional employment on ancestral grounds of Hungarians in Cleveland, of Poles in Buffalo, of Germans in Milwaukee, of Portuguese in New Bedford, of Mexicans in San Antonio, of the *338numerous minority groups in New York, and so on through the whole gamut of racial and religious concentrations in various cities.” Id., at 464.
The reservation of a proportion of the law school class for members of selected minority groups is fraught with similar dangers, for one must immediately determine which groups are to receive such favored treatment and which are to be excluded, the proportions of the class that are to be allocated to each, and even the criteria by which to determine whether an individual is a member of a favored group. There is no assurance that a common agreement can be reached, and first the schools, and then the courts, will be buffeted with the competing claims. The University of Washington included Filipinos, but excluded Chinese and Japanese; another school may limit its program to blacks, or to blacks and Chicanos. Once the Court sanctioned racial preferences such as these, it could not then wash its hands of the matter, leaving it entirely in the discretion of the school, for then we would have effectively overruled Sweatt v. Painter, 339 U. S. 629, and allowed imposition of a “zero” allocation.19 But what standard is the Court to apply when a rejected applicant of Japanese ancestry brings suit to require the University of Washington to extend the same privileges to his group? The Committee might conclude that the population of Washington is now 2% Japanese, and that Japanese also constitute 2% of the *339Bar, but that had they not been handicapped by a history of discrimination, Japanese would now constitute 5% of the Bar, or 20%. Or, alternatively, the Court could attempt to assess how grievously each group has suffered from discrimination, and allocate proportions accordingly ; if that were the standard, the current University of Washington policy would almost surely fall, for there is no Western State which can claim that it has always" treated Japanese and Chinese in a fair and evenhanded manner. See, e. g., Yick Wo v. Hopkins, 118 U. S. 356; Terrace v. Thompson, 263 U. S. 197; Oyama v. California, 332 U. S. 633. This Court has not sustained a racial classification since the wartime cases of Korematsu v. United States, 323 U. S. 214, and Hirabayashi v. United States, 320 U. S. 81, involving curfews and relocations imposed upon Japanese-Americans.20
*340Nor obviously will the problem be solved if next year the 'Law School included only Japanese and Chinese, for then Norwegians and Swedes, Poles and Italians, Puerto Ricans and Hungarians, and all other groups which form this diverse Nation would have just complaints.
The key to the problem is consideration of such applications in a racially neutral way. Abolition of the LSAT would be a start. The invention of substitute tests might be made to get a measure of an applicant’s cultural background, perception, ability to analyze, and his or her relation to groups. They are highly subjective, but unlike, the LSAT they are not concealed, but in the open. A law school is not bound by any legal principle to admit students by mechanical criteria which are insensitive to the potential of such an applicant which may be realized in a more hospitable environment. It will be necessary under such an approach to put more effort into assessing each individual than is required when LSAT scores and undergraduate grades dominate the selection process. Interviews with the applicant and others who know him is a time-honored test. Some schools currently run summer programs in which potential students who likely would be bypassed under conventional admissions criteria are given the opportunity to try their hand at law courses,21 and certainly their performance in such programs could- be weighed heavily. There is, moreover, no bar to considering an individual’s prior achievements in *341light of the racial discrimination that barred his way, as a factor in attempting to assess his true potential for a successful legal career. Nor is there any bar to considering on an individual basis, rather than according to racial classifications, the likelihood that a particular candidate will more likely - employ his legal skills to service communities that are not now adequately represented than will competing candidates. Not. every student benefited by such an expanded admissions program would fall into one of the four racial groups involved here, but it is no drawback that other deserving applicants will also get an opportunity they would otherwise have been denied. Certainly such a program would substantially fulfill the Law School’s interest in giving a more diverse group access to the legal profession, Such a program might be less convenient administratively than simply sorting students by race, but we have never held administrative convenience to justify racial discrimination.
The argument is that a “compelling” state interest can easily justify the racial discrimination that is practiced here. To 'many, “compelling” would give members of one race even more than pro rata representation. The public payrolls might then be deluged say with Chícanos because they are as a group the poorest of the poor and need work more, than others, leaving desperately poor individual blacks and whites without employment. By the same token large quotas of blacks or browns could be added to the Bar, waiving examinations required of other groups, so that it would-be better racially balanced.22 *342The State, however, may not proceed by racial classification to force strict population equivalencies for every group in every occupation, overriding individual preferences. The Equal Protection Clause commands the elimination of racial barriers, not their creation in order to satisfy our theory as to how society ought to be organized. The purpose of the University of Washington cannot be to produce black lawyers for blacks, Polish lawyers for Poles, Jewish lawyers for Jews, Irish lawyers for Irish. It should be to produce good lawyers for Americans and not to place First Amendment barriers against anyone.23 That is the point at the heart of all our *343school desegregation cases, from Brown v. Board of Education, 347 U. S. 483, through Swann v. Charlotte-Mecklenburg Board of Education, 402 U. S. 1. A segregated admissions process creates suggestions of stigma and caste no less than a segregated classroom, and in the end it may produce that result despite its contrary intentions. One other assumption must be clearly disapproved: that blacks or browns cannot make it on their individual merit. That is a stamp of inferiority that a State is not permitted to place on any lawyer.
If discrimination based on race is constitutionally permissible when those who hold the reins can come up with “compelling” reasons to justify it, then constitutional guarantees acquire an accordionlike quality. Speech is closely brigaded with action when it triggers a fight, Chaplinsky v. New Hampshire, 315 U. S. 568, as shouting “fire” in a crowded theater triggers a riot. It may well be that racial strains, racial susceptibility to certain diseases, racial sensitiveness to environmental conditions that other races do not experience, may in an extreme . situation justify differences in racial treatment that no fairminded 'person would call “invidious” discrimination. Mental ability is not in that category. All races can compete fairly at all professional levels. So *344far as race is concerned, any state-sponsored preference to one race over another in that competition is in my view “invidious” and violative of the Equal Protection Clause.
The problem tendered by this case is important and crucial to the • operation of our constitutional system; and educators must be given leeway. It may well be that1 a whole congeries of applicants in the marginal group defy known methods of selection. Conceivably, an admissions committee might conclude that a selection by lot of, say, the. last 20 seats is the only fair solution. Courts'aré not educators; their expertise is limited; and Our task ends with the inquiry whether, judged by the main purpose of the Equal Protection Clause — the protection against racial discrimination24 — there has been an “invidious” discrimination.'
We would have a different case if the suit were one to displace the applicant who was chosen in lieu of DeFunis. What the record would show concerning his potentials would have to • be considered and weighed. The educational decision, provided proper guidelines were used, would reflect an expertise that courts should honor. The problem is not tendered here because the physical facilities were apparently adequate to take DeFunis in addition to the others. My view is only that I cannot say by the tests used and applied he was invidiously discriminated .against because of his race.
I cannot conclude that the admissions procedure of the Law School of the University of Washington that excluded DeFunis is violative of the Equal Protection Clause of the Fourteenth Amendment. .The judgment of the Washington Supreme Court should be vacated and the case' remanded for a new trial.
*345APPENDIX TO OPINION OF DOUGLAS, J., DISSENTING* .
The following are excerpts from the Law School’s current admissions policy, as provided to the Court by counsel for the respondents.
ADMISSIONS
A. Policy Statement Regarding Admission to Entering Classes of Juris Doctor Program — Adopted by the Law Faculty December- 4", 1973. -
§ 1. The objectives of the admissions program are to select and admit those applicants who have the best prospect of high quality academic work at the law school and, in the minority admissions program described below, the further objective there stated.
§ 2. In measuring academic potential the law school relies primarily on the undergraduate grade-point average and the performance on the Law School Admission Test (LSAT). The weighting of these two indicators is determined statistically by reference to past experience at this school. For most applicants the resulting applicant ranking is the most nearly accurate of all available measures of relative academic potential. In truly exceptional cases, i. e., those in which the numerical indicators clearly appear to be an inaccurate measure of academic potential, the admission decision indicated by them alone may be altered by a consideration of the factors listed below. The number of these truly exceptional cases in any particular year should fall somewhere from zero to approximately forty. These factors are used, however, only as an aid in assessing the applicant’s academic potential in its totality, without undue emphasis or- reliance upon one or a few and without an attempt to quantify in advance the strength of their *346application, singly or as a whole, in a particular case. They are:
a) the difficulty or ease of the undergraduate curriculum track pursued;
b) the demanding or non-demanding quality of the undergraduate school or department;
c) the attainment of an advanced degree, the nature thereof, and difficulty or ease of its attainment;
d) the applicant’s pursuits subsequent to attainment of the undergraduate degree and the degree of success therein, as bearing on the applicant’s academic potential;
e) the possibility that an applicant many years away from academic work may do less well on the LSAT than his or her counterpart presently or recently in academic work ;
f) substantial change in mental or physical health that indicates prospect for either higher or ..lower quality of academic work;
g) substantial change in economic pressures or other circumstances that indicates prospect for either higher or lower quality of academic work;
h) exceptionally good or bad performance upon the writing test ingredient of the LSAT, if the current year’s weighting of the numerical indicators does not otherwise take the writing score into account;
i) .the quality and strength of recommendations bearing upon the applicant’s academic potential;
j) objective indicators of motivation to succeed at the academic study of law;
k) variations in the level of academic achievement over time; and
l) any other indicators that serve the objective stated above.
§ 6. Because certain ethnic groups in our society *347have historically been limited in their access to the legal profession and because the resulting underrepresentation can affect the quality of legal services available to members of such groups, as well as limit their opportunity for full participation in the governance of our communities, the faculty recognizes a special obligation in its admissions policy to contribute to the solution of the problem.'
Qualified minority applicants are therefore admitted under the minority admissions program in such number that the entering class will have a reasonable proportion of minority persons, in view of the obligation stated above and of the overall objective of the law school to provide legal education for qualified persons generally. For the purpose of determining the number to be specially admitted under the program, and not as a ceiling on minority admissions generally, the‘faculty currently believes that approximately 15 to 20 percent is such a reasonable proportion if there are sufficient qualified applicants, available. Under the minority admissions program, admission is offered to those applicants who have a reasonable prospect of academic success at the law school, determined in each case by considering the numerical indicators along with the listed factors in Section 2, above, but without regard to the restriction upon number contained in that section.
No particular internal percentage or proportion among various minority groups in the entering class is. specified; rather, the law school strives for a reasonable internal balance given the particular makeup of each year’s applicant population.
As to some or all ethnic groups within the scope of the minority admissions program, it may be appropriate to give a preference in some degree to residents of the state; that determination is made each year in view of *348all the particulars of that year’s situation, and the preference is given when necessary to meet some substantial local need for minority representation.
The grades are calculated on a conventional 4.0 scale, and the ,LSAT is scored on a scale ranging from 200 to 800. A Writing Test given on the same day as the LSAT and administered with it is also included in the formula; it is scored on a scale of 20 to 80. The Admissions Committee combines these scores into the Average by calculating the sum of 51.3, 3.4751 X the grade-point average, .0169 X LSAT score, and .0456 X the Writing Test score. App. 24. For a brief discussion of the use of the LSAT in combination with undergraduate grades to predict law school success, see Winter-bottom, Comments on “A Study of the Criteria for Legal Education and Admission to the Bar,” An Article by Dr. Thomas M. Goolsby, Jr.,21J. Legal Ed. 75 (1968).
The only other substantial group admitted at this point were 19 “military” applicants. These were students who had previously been admitted to the'school but who had either been unable to come, or forced to leave during their tenure, because of the draft. They were *322given preferential treatment upon reapplication after completing their military obligation. Since, neither party has raised any issue concerning this group of applicants, the remaining consideration of the admissions procedure will not discuss them. Four minority applicants were also admitted at this time, although none apparently had scores, above 77. App. 31. Their admission was presumably pursuant to the procedure for minority applicants described below.
This was a Council on Legal Education Opportunities program, federally funded by the Office of Economic Opportunity and sponsored by the American Bar Association, the Association of American Law Schools, the National Bar Association, and the Law School Admissions Council.
Testimony of the Chairman of the Admissions -Committee, Statement of Facts 353.
The Guide to Applicants explained:'
“We gauged the potential for outstanding performance in law school not only from the existence of high test scores and grade point averages, but also from -careful- analysis of recommendations, the quality of work in difficult analytical seminars, courses, and writing programs, the academic standards of the school attended by the applicant, the-.applicant’s graduate work (if any), and the nature of. the applicant’s employment (if any), since graduation.
“An applicant’s ability to make significant contributions to law school classes and the community at large was assessed from such factors as his extracurricular and community activities, employment, and general background.
“We gave no preference to, but did not discriminate against, either Washington residents or women in making our determinations. An applicant’s racial or ethnic background was considered as one factor in our general attempt to convert formal credentials into realistic predictions.” 82 Wash. 2d 11, 18-19, 507 P. 2d 1169, 1174.
The respondents provided the following table in response to an interrogatory during the proceedings in the state court:
Predicted - Number of Number
First Year Averages Applications Received Accepted
81 1 1
80 2 2
79 11 11
78 42 42
77 105 93
76 169 53
75 210 22
App. 34.
Although there is apparently no evidence in point in the record, respondents suggest .that at least some of these eight students were also admitted on a preferential basis. Brief for Respondents 40 n. 27.
United States Bureau of the Census, Census of Population: 1970, General Population Characteristics, Washington, Final Report PC (1) — B49, Table 18.
For a history of gradual acceptance among law schools of standardized tests as an admission tool, see Ramsey, Law School Admissions: Science, Art, or Hunch?, 12 J. Legal Ed. 503 (1960).
For a survey of the use of the LSAT by American law schools as of 1965, see .unneborg & Radford, The LSAT: A Survey of Actual Practice, 18 J. Legal Ed. 313 (1966).
Rock, Motivation, Moderators, and Test Bias, 1970 U. Tol. L. Rev. 527, 535.
At the outset the Committee may have chosen only a range, with the precise number to be determined later in the process as the total number of minority applicants, and some tentative assessment of their quality, could be determined. This appears to be the current articulated policy, see App. to this opinion § 6, and we are advised by the respondents that § 6 “represents a more formal statement of the policy which was in effect in 1971 . . . but does not *333represent any change in policy.” Letter to the Court dated March 19, 1974, p. 1. The fact that the Committee did not set a precise number in advanced obviously irrelevant to the legal analysis. Nor does it matter that there is some minimal level of achievement below which the Committee would not reach in order to achieve its stated goal as to the proportion of the class reserved for minority groups, so long as the Committee was willing, in order to achieve that goal, to admit minority applicants who, in the Committee’s own judgment, were less qualified than other rejected applicants and who would not otherwise have been admitted.
See n. 12, supra, and App. to this opinion § 6.
Uncommon Controversy, Report Prepared for American Friends Service Committee 29-30 (1970).
See C. Fee, Chief Joseph, The Biography of a Great Indian (1936).
See. F. Brockway, African Socialism (1963); African Socialism W. Friedland & C. Rosberg ed. 1964).
See L. Senghor, On African Socialism (M. Cook ed. 1964).
We are not faced here with a situation where barriers are overtly or covertly put in the path of members of one racial group which are not. required by others. There was also no showing that the purpose of the school’s policy was to eliminate arbitrary and irrelevant barriers to entry by certain racial groups into the legal profession groups. Griggs v. Duke Power Co., 401 U. S. 424. In Swann v. Charlotte-Mecklenburg Board of Education, 402 U. S. 1, 16, we stated that, as a matter of educational policy school authorities could, within their broad discretion, specify that each school within its district have a prescribed ratio of Negro to white students reflecting the proportion for the district as a whole, in order to disestablish a dual school system. But there is a crucial difference between the. policy suggested in Sivann and that under consideration here: the Swarm policy would impinge on no person’s constitutional rights, because no one would be excluded from a public school and no one has a right to attend a segregated public school.
Sweatt held that a State could not justify denying a. black admission to its regular law school by creating a new law school for blacks'. We held that the .new law school did not meet the requirements of “equality” set forth in Plessy v. Ferguson, 163 U. S. 537.
The student, we said was entitled to “legal education equivalent to that offered by the State to students of. other races. Such education is not available to him in a separate law school as offered by the State.” 339 U. S., at 635.
Those cases involved an exercise of the war power, a great leveler of other rights. Our Navy was sunk at Pearl Harbor and no one knew where the' Japanese fleet was. We were advised on oral argument that if the Japanese landed troops on our west coast nothing could stop them west of the Rockies. The military judgment was that, to aid in the prospective defense of the west coast; the enclaves of Americans of Japanese ancestry should be moved inland, lest the invaders by donning civilian clothes would wreak even more serious havoc on our western ports. The decisions were extreme and went to the verge of wartime power; and they have been severely criticized. It is, however, easy in retrospect to denounce what was done, as there actually was no attempted Japanese invasion of our country. While our Joint Chiefs of Staff were worrying about Japanese soldiers landing on the west coast, they actually were landing in Burma and at Kota Bharu in Malaya. But those making plans for defense of the Nation had no such knowledge and were planning for the worst. Moreover, the day we decided Korematsu we also decided Ex parte Endo, 323 U. S. 283, holding that while evacuation of the Americans of Japanese ancestry was allowable under extreme war conditions, their detention after evacuation was not. We said:
“A citizen who is concededly loyal presents no problem of espió*340nage or sabotage. Loyalty is a matter of the heart and mind, not of race, creed, or color. He who is loyal is by definition not a spy or a saboteur. When the power to detain is derived from the power to protect the war effort against espionage and sabotage, detention which has no relationship +o that objective is unauthorized.” Id., at 302.
See n. 3, supra.
In Johnson v. Committee on Examinations, 407 U. S. 915, we denied certiorari in a case presenting a similar issue. There the petitioner claimed that the bar examiners reconsidered the papers submitted by failing minority applicants whose scores were close to the cutoff point, with the result that some minority appli*342cants were admitted to the Bar although they initially had examination scores lower than those of white applicants who failed.
As the Arizona Supreme Court denied Johnson admission summarily, in an original proceeding, there were no judicial findings either sustaining or rejecting his factual claims of racial bias, putting the case in an awkward posture for review here. Johnson subsequently brought a civil rights action in Federal District Court, seeking both damages and injunctive relief. The District Court dismissed the action and the Court of Appeals affirmed, holding that the lower federal courts did not have jurisdiction to review the decisions of the Arizona Supreme Court on admissions to the state 'Bar. Johnson then sought review here, and we denied his motion for leave to file a petition for mandamus, prohibition and/or certiorari on February 19, 1974. Johnson v. Wilmer, 415 U. S. 911. Thus in the entire history of the case no court had. ever actually sustained Johnson’s factual contentions concerning racial bias in the bar examiners’ procedures. BeFunis thus appears to be the first case here squarely presenting the problem.
Underlying all cultural background tests are potential ideological issues, that have plagued bar associations and the courts. In re Summers, 325 U. S. 561, involved the denial of the practice of law to a man who could not.conscientiously bear arms. The vote against him was five to four. Konigsberg v. State Bar, 353 U. S. 252, followed, after remand, by Konigsberg v. State Bar, 366 U. S. 36, resulted in barring one from admission to a state bar because of his refusal to answer questions concerning Communist Party member*343ship. He, too, was excluded five to four. The petitioner in Schware v. Board of Bar Examiners, 353 U. S. 232, was, however, admitted to practice even though he had about 10 years earlier been a member of the Communist Party. But In re Anastaplo, 366 U. S. 82, a five-to-four decision, barred a man from admission to a state bar not because he invoked the Fifth Amendment when asked abdut membership in the Communist Party, but because he asserted that the First- and Fourteenth Amendments protected him from that inquiry. Baird v. State Bar of Arizona, 401 U. S. 1, held by a divided vote that a person could not be kept out of the state bar for refusing to answer whether he had ever been a member of the Communist Party; and see In re Stolar, 401 U. S. 23.
See Slaughter House Cases, 16 Wall. 36, 81.