Mr. Justice DOUGLAS, with whom Mr. Justice MARSHALL concurs, dissenting.
The Court today denies certiorari to a Black man who stands convicted by an all-White jury which had been selected through a process which petitioner alleges methodically excluded members of minority racial groups. The most pernicious of the practices used to exclude Black and Chicano jurors was what purported to be an intelligence test which, because of its cultural bias and its blatant unreliability, excluded nearly 50% of the otherwise qualified prospective jurors from minority groups. We would all agree that the brand of justice received in our courts should not depend upon the color of one's skin and that the selection of jury panels should not be tainted by the exclusion of racial groups. With all respect I fear precisely that result has obtained in the case.
Petitioner was convicted of the unlawful possession of marihuana. The State's evidence consisted of the testimony of two White police officers that they observed and smelled petitioner smoke a marihuana cigarette and that he dropped the cigarette on the ground before he was arrested. Petitioner steadfastly denied his guilt. He offered witnesses who 'testified that, because of an obstruction, it was not possible to see' the area where [404 U.S. 968 , 969] petitioner was allegedly smoking marihuana, who supported his story 'that no one was smoking marihuana,' and who gave credence to the apparent theory of the defense that a Black man was being framed by White police officers. People v. Donaldson, Crim.No. 17615, at 3-4 (Cal.Ct.App., Dec. 21, 1970.)
The issue for the jury was the relative credibility of two White police officers and the four Black defense witnesses. Cf. Note, 79 Yale L. J. 531 (1970). That this question was close is indicated by the fact that after some 5 1/4 hours of deliberation the jury reported itself deadlocked eight-to-four and an Allen charge was then given. See Allen v. United States, 164 U.S. 492 (1896). It took an additional hour of deliberation before the jury resolved the credibility issue against petitioner and returned a verdict of guilty.
On appeal the California Court of Appeal affirmed the conviction, holding that petitioner, as a Black, had no standing to challenge the exclusion of Chicanos from the jury panel,1 that the exclusion of racial minorities was unintentional and that, in any event, it did not deny equal protection of the laws. The Supreme Court of California denied a petition for a hearing and petitioner now seeks a writ of certiorari.
The culturally-based intelligence test used in this case excludes almost half of the otherwise qualified prospective jurors from minority groups. It is also argued that members of racial minority groups are effectively barred from jury service by (1) the selection of prospec- [404 U.S. 968 , 970] tive jurors solely from voter registration lists despite the significantly lower incidence of registration by Blacks and Chicanos, (2) the absence of any follow-up procedures to correct for the high frequency with which Blacks and Chicanos fail to respond to jury service notices, and (3) the token $5 per diem paid jurors which precludes those in low economic strata from serving as jurors. 2
Petitioner concludes that '[i]n the 1968 South District Jury Draw [ from which his jury was selected], the cumulative effect of the ... selection procedures ... was such that the percentage selected from a white middle-class comparison area was 13 times as great as the percentage selected from a low-income black comparison area, and the exclusion of low- income Mexican-Americans was virtually total.'3 The opinion of the Court of Appeal and the argument of the respondent do not, in my mind, sufficiently rebut the prima facie showing petitioner has made that he has been denied the equal protection of the law by his conviction by a jury selected [404 U.S. 968 , 971] through racially biased procedures. 4 'If there has been discrimination, whether accomplished ingeniously or ingenuously, the conviction cannot stand.' Smith v. Texas, 311 U.S. 128, 132 , 166 ( 1940). It would seem enough, therefore, to reverse this conviction out of hand because of the racial bias built into the jury selection and because of the improper legal standard applied by the court below.
The test in question was drafted in 1935 and it consists of some 33 questions designed to measure vocabulary and reading comprehension. The Jury Commissioner testified that it was administered '[t]o see if a prospective juror has sufficient knowledge of the English language,' but expert scientific testimony indicated that the test was 'not a good measure ... of average verbal ability.' The author of the examination, Psychology Professor Dr. Neil Warren, testified that he validated the test and established the passing score by administering it to some 100 to 200 white middle-class college students. Over the years, the test questions themselves have remained unchanged despite changes in the common vocabulary. 5 At some point, however, the test was divided into two parts and, for some inexplicable reason, the passing grade was changed. Dr. Warren testified that these changes would 'reduce the reliability to the point where [he] would not want to condone the results.' Another expert witness testified that the test was 'completely arbitrary to use in its present form.' [404 U.S. 968 , 972] While the test may have been unreliable as a measure of juror competence, it effectively excluded minority group members from jury service. The rate of failure in minority group areas was almost four times that in a white, middle-class comparison area:
Area Failure Rate Long Beach (White, Middle-Class) 13% (14/106) Wilmington (48% Spanish Surname) and San Pedro (41% Mexican- American and 11% Black) 45% (23/51) Central Long Beach (55% Black) 48% (12/25) District-Wide 18% (349/1937)
It would seem that this device is illsuited to serve any legitimate state interest and does serve to exclude from jury panels significantly disproportionate numbers of minority groups. The Constitution does not of course require racially-balanced juries because, in our pluralistic society, a group of 12 men and women could not possibly represent all of the ethnic, racial and economic groups which comprise our diverse culture. Swain v. Alabama, 380 U.S. 202, 208 , 829 ( 1965); Cassell v. Texas, 339 U.S. 282 , 286-287, 631-632 (1950); Akins v. Texas, 325 U.S. 398 (1945). What the Constitution demands, however, is that no such groups be consciously excluded from the selection process. See Hill v. Texas, 316 U.S. 400, 404 , 1161; Smith v. Texas, 311 U.S. 128, 130 , 165; 5 U.S. Commission on Civil Rights Report, Justice 89-103 (1961).
What the facts might show after oral argument, no one knows. I would hold constitutionally infirm a conviction returned by a jury from which a disproportionately large number of minority group members were excluded through the use of a culturally biased intelligence test where the test is shown to be unreliable as a measure of jury competence. There is a prima facie showing that this test is vulnerable. Accordingly, I would grant the petition for a writ of certiorari.
Mr. Justice BRENNAN would also grant certiorari. [404 U.S. 968 , 973] APPENDIX
A comparison of the number of jurors approved from White, Chicano, and Black residential areas indicates the discriminatory effect of the jury selection method used here:
Long Beach Wilmington Central Long (White Middle-) (48% Spanish Beach Class) Surname) (55% Black) Median Family Income $14,252 $6,584 $ 6,534 Total Population 20,215 36,796 26,400 Population Over 21 11,522 18,002 13,423 Registered Voters (Percent of Those Over 21) 8,991 (78%) 8, 945 (41%) 5,747 (43%) Notices Sent (Percent of Registered Voters) 315(3.50%) 163 ( 1.82%) 110 1.91%) Responses Received (Percent of Notices) 216 (65%) 68 (42%) 46 (42%) Approved by Jury Commission 83 13 12
[ Footnote 1 ] Such reasoning is contrary to Ballard v. United States, 329 U.S. 187, 195 , 265 (1946), where we said, '[R] eversible error does not depend on a showing of prejudice in an individual case. The evil lies in the admitted exclusion of an eligible class or group in the community in disregard of the prescribed standards of jury selection. ... The inquiry is not limited to the defendant-there is injury to the jury system, to the law as an institution, to the community at large, and to the democratic ideal reflected in the process of our courts.'
[ Footnote 2 ] Even if it were assumed that these exclusions were the unavoidable consequences of Los Angeles' method of jury selection, it does not follow that jury panels must be racially unrepresentative of the community. The names of prospective jurors are selected by a computer from voter registration lists. With modern sampling techniques, it would be a simple matter to program the computer so that its initial selection of names would-after the operation of these exclusions-yield a racially representative jury panel. Mills, A Statistical Study of Occupations of Jurors in a United States District Court, 22 Md.L.Rev. 205, 214 (1962).
[ Footnote 3 ] See Appendix to this opinion.
[ Footnote 4 ] The Court of Appeal argues, for example, that '[t]he poor, the Black and Mexican-Americans have the power to register to vote, and voter registration is free and relatively uncomplicated.' Crim. No. 17615, at 14. Be this as it may, this argument is unresponsive to petitioner's contention that jurors are selected from lists which are racially unbalanced.
[ Footnote 5 ] Evidence was adduced at trial which indicated that, even apart from the unsupported changes in the grading of the examination, the test was likely invalid solely because of its vintage and the unrepresentative sample used to validate it in 1935.