dissenting*
The American ideal of political equality, conceived in the earliest days of our colonial existence and fostered by the *104egalitarian language of the Declaration of Independence, could not forever tolerate the limitation of the right to vote to white propertied males. Our Constitution has been amended six times in the movement toward a democracy for more than the few,1 and this Court has interpreted the Fourteenth Amendment to provide that “a citizen has a constitutionally protected right to participate in elections on an equal basis with other citizens in the jurisdiction,” Dunn v. Blumstein, 405 U. S. 330, 336 (1972). The Court’s decision today is in a different spirit. Indeed, a plurality of the Court concludes that, in the absence of proof of intentional discrimination by the State, the right to vote provides the politically powerless with nothing more than the right to cast meaningless ballots.
The District Court in both of these cases found that the challenged multimember districting schemes unconstitutionally diluted the Negro vote. These factual findings were upheld by the Court of Appeals, and the plurality does not question them. Instead, the plurality concludes that districting schemes do not violate the Equal Protection Clause unless it is proved that they were enacted or maintained for the purpose of minimizing or canceling out the voting potential of a racial minority. The plurality would require plaintiffs in vote-dilution cases to meet the stringent burden of establishing discriminatory intent within the meaning of Washington v. Davis, 426 U. S. 229 (1976); Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U. S. 252 (1977); and Personnel Administrator of Mass. v. Feeney, 442 U. S. 256 (1979). In my view, our vote-dilution decisions require only a showing of discriminatory impact to justify the invalidation of a multimember districting scheme, and, because they are premised on the fundamental interest in voting protected by the Fourteenth Amendment, the discriminatory-impact standard adopted by them is unaffected by Washington v. Davis, supra, and its progeny. Furthermore, an intent re*105quirement is inconsistent with the protection against denial or abridgment of the vote on account of race embodied in the Fifteenth Amendment and in § 2 of the Voting Rights Act of 1965, 79 Stat. 437, as amended, 42 U. S. C. § 1973.2 Even if, however, proof of discriminatory intent were necessary to support a vote-dilution claim, I would impose upon the plaintiffs a standard of proof less rigid than that provided by Personnel Administrator of Mass. v. Feeney, supra.
I
The Court does not dispute the proposition that multimem-ber districting can have the effect of submerging electoral minorities and overrepresenting electoral majorities.3 It is *106for this reason that we developed a strong preference for single-member districting in court-ordered reapportionment plans. See ante, at 66, n. 12. Furthermore, and more important for present purposes, we decided a series of vote-dilution cases under the Fourteenth Amendment that were designed to protect electoral minorities from precisely the combination of electoral laws and historical and social factors found in the present cases.4 In my view, the plurality’s treatment of *107these cases is fanciful. Although we have held that multi-member districts are not unconstitutional per se, see ante, at 66, there is simply no basis for the plurality’s conclusion that *108under our prior cases proof of discriminatory intent is a necessary condition for the invalidation of multimember districting.
A
In Fortson v. Dorsey, 379 U. S. 433 (1965), the first vote-dilution case to reach this Court, we stated explicitly that such a claim could rest on either discriminatory purpose or effect:
“It might well be that, designedly or otherwise, a multi-member constituency apportionment scheme, under the circumstances of a particular case, would operate to minimize or cancel out the voting strength of racial or political elements of the voting population.” Id., at 439 (emphasis added).
We reiterated these words in Burns v. Richardson, 384 U. S. 73 (1966), interpreted them as the correct test to apply to vote-dilution claims, and described the standard as one involving “invidious effect,” id., at 88. We then held that the plaintiffs had failed to meet their burden of proof:
“[T]he demonstration that a particular multi-member scheme effects an invidious result must appear from evidence in the record. . . . That demonstration was not made here. In relying on conjecture as to the effects of multi-member districting rather than demonstrated fact, the court acted in a manner more appropriate to the body responsible for drawing up the districting plan. Speculations do not supply evidence that the multi-member dis-tricting was designed to have or had the invidious effect necessary to a judgment of the unconstitutionality of the districting.” Id., at 88-89 (emphasis added) (footnote omitted).
It could not be plainer that the Court in Burns considered *109discriminatory effect a sufficient condition for invalidating a multimember districting plan.
In Whitcomb v. Chavis, 403 U. S. 124 (1971), we again repeated and applied the Fortson standard, 403 U. S., at 143, 144, but determined that the Negro community’s lack of success at the polls was the result of partisan politics, not racial vote dilution. Id., at 150-155. The Court stressed that both the Democratic and Republican Parties had nominated Negroes, and several had been elected. Negro candidates lost only when their entire party slate went down to defeat. Id., at 150, nn. 29-30, 152-153. In addition, the Court was impressed that there was no finding that officials had been unresponsive to Negro concerns. Id., at 152, n. 32, 155.5
More recently, in White v. Regester, 412 U. S. 755 (1973), we invalidated the challenged multimember districting plans because their characteristics, when combined with historical and social factors, had the discriminatory effect of denying *110the plaintiff Negroes and Mexican-Americans equal access to the political process. Id., at 765-770. We stated that
“it is not enough that the racial group allegedly discriminated against has not had legislative seats in proportion to its voting potential. The plaintiffs’ burden is to produce evidence to support findings that the political processes leading to nomination and election were not equally open to participation by the group in question — that its members had less opportunity than did other residents in the district to participate in the political processes and to elect legislators of their choice.” Id., at 765-766.
We held that the three-judge District Court had properly applied this standard in invalidating the multimember dis-tricting schemes in the Texas counties of Dallas and Bexar. The District Court had determined that the characteristics of the challenged electoral systems — multimember districts, a majority-vote requirement for nomination in a primary election, and a rule mandating that a candidate running for a position in a multimember district must run for a specified “place” on the ticket — though “neither in themselves improper nor invidious,” reduced the electoral influence of Negroes and Mexican-Americans. Id., at 766.6 The District Court identified a number of social and historical factors that, when combined with the Texas electoral structure, resulted in vote dilution: (1) a history of official racial discrimination in Texas, including discrimination inhibiting the registration, casting of ballots, and political participation of Negroes; (2) proof that minorities were still suffering the effects of past discrimination; (3) a history of gross underrepresentation of minority interests; (4) proof of official insensitivity to the needs of minority citizens, whose votes were not needed by those in power; (5) the recent use of racial campaign tactics; and (6) a cultural and language barrier inhibiting the participation of *111Mexican-Americans. Id., at 766-770. Based “on the totality of the circumstances,” we affirmed the District Court’s conclusion that the use of multimember districts excluded the plaintiffs “from effective participation in political life.” Id., at 769.7
*112It is apparent that a showing of discriminatory intent in the creation or maintenance of multimember districts is as unnecessary after White as it was under our earlier vote-dilution decisions. Under this line of cases, an electoral dis-tricting plan is invalid if it has the effect of affording an electoral minority “less opportunity than . . . other residents in the district to participate in the political processes and to elect legislators of their choice,” id., at 766. It is also apparent that the Court in White considered equal access to the political process as meaning more than merely allowing the minority the opportunity to vote. White stands for the proposition that an electoral system may not relegate an electoral minority to political impotence by diminishing the importance of its vote. The plurality’s approach requiring proof of discriminatory purpose in the present cases is, then, squarely contrary to White and its predecessors.8
B
The plurality fails to apply the discriminatory-effect standard of White v. Regester because that approach conflicts with what the plurality takes to be an elementary principle of law. “[O]nly if there is purposeful discrimination,” announces the *113plurality, “can there be a violation of the Equal Protection Clause of the Fourteenth Amendment.” Ante, at 66. That proposition is plainly overbroad. It fails to distinguish between two distinct lines of equal protection decisions: those involving suspect classifications, and those involving fundamental rights.
We have long recognized that under the Equal Protection Clause classifications based on race are “constitutionally suspect,” Bolling v. Sharpe, 347 U. S. 497, 499 (1954), and are subject to the “most rigid scrutiny,” Korematsu v. United States, 323 U. S. 214, 216 (1944), regardless of whether they infringe on an independently protected constitutional right. Cf. University of California Regents v. Bakke, 438 U. S. 265 (1978). Under Washington v. Davis, 426 U. S. 229 (1976), a showing of discriminatory purpose is necessary to impose strict scrutiny on facially neutral classifications having a racially discriminatory impact. Perhaps because the plaintiffs in the present cases are Negro, the plurality assumes that their vote-dilution claims are premised on the suspect-classification branch of our equal protection cases, and that under Washington v. Davis, supra, they are required to prove discriminatory intent. That assumption fails to recognize that our vote-dilution decisions are rooted in a different strand of equal protection jurisprudence.
Under the Equal Protection Clause, if a classification “impinges upon a fundamental right explicitly or implicitly protected by the Constitution, . . . strict judicial scrutiny” is required, San Antonio Independent School Dist. v. Rodriguez, 411 U. S. 1, 17 (1973), regardless of whether the infringement was intentional.9 As I will explain, our cases *114recognize a fundamental right to equal electoral participation that encompasses protection against vote dilution. Proof of discriminatory purpose is, therefore, not required to support a claim of vote dilution.10 The plurality’s erroneous conclusion to the contrary is the result of a failure to recognize the central distinction between White v. Regester, 412 U. S. 755 (1973), and Washington v. Davis, supra: the former involved an infringement of a constitutionally protected right, while the latter dealt with a claim of racially discriminatory distribution of an interest to which no citizen has a constitutional entitlement.11
*115Nearly a century ago, the Court recognized the elementary proposition upon which our structure of civil rights is based: “[T]he political franchise of voting is ... a fundamental political right, because preservative of all rights.” Yick Wo v. Hopkins, 118 U. S. 356, 370 (1886). We reiterated that theme in our landmark decision in Reynolds v. Sims, 377 U. S. 533, 561-562 (1964), and stated that, because “the right of suffrage is a fundamental matter in a free and democratic society [,] ... any alleged infringement of the right of citizens to vote must be carefully and meticulously scrutinized.” Ibid. We realized that “the right of suffrage can be denied by a debasement or dilution of the weight of a citizen’s vote just as effectively as by wholly prohibiting the free exercise of the franchise.” Id., at 555. Accordingly, we recognized that the Equal Protection Clause protects “[t]he right of a citizen to equal representation and to have his vote weighted equally with those of all other citizens.” Id., at 576. See also Wes-*116berry v. Sanders, 376 U. S. 1, 17 (1964); Gray v. Sanders, 372 U. S. 368, 379-380 (1963).12
Reynolds v. Sims and its progeny13 focused solely on the discriminatory effects of malapportionment. They recognize that, when population figures for the representational districts of a legislature are not similar, the votes of citizens in larger districts do not carry as much weight in the legislature as do votes cast by citizens in smaller districts. The equal protection problem attacked by the “one person, one vote” principle is, then, one of vote dilution: under Reynolds, each citizen must have an “equally effective voice” in the election of representatives. Reynolds v. Sims, supra, at 565. In the present cases, the alleged vote dilution, though caused by the combined effects of the electoral structure and social and historical factors rather than by unequal population distribution, is analytically the same concept: the unjustified abridgment of a fundamental right.14 It follows, then, that a showing of dis*117criminatory intent is just as unnecessary under the vote-dilution approach adopted in Fortson v. Dorsey, 379 U. S. 433 (1965), and applied in White v. Regester, supra, as it is under our reapportionment cases.15
*118Indeed, our vote-dilution cases have explicitly acknowledged that they are premised on the infringement of a fundamental right, not on the Equal Protection Clause’s prohibition of racial discrimination. Our first vote-dilution decision, Fortson v. Dorsey, supra, involved a 1962 Georgia reapportionment statute that allocated the 54 seats of the Georgia Senate among the State’s 159 counties. Thirty-three of the senatorial districts were made up of from one to eight counties each, and were single-member districts. The remaining 21 districts were allotted among the 7 most populous counties, with each county containing at least 2 districts and electing all of its senators by countywide vote. The plaintiffs, who were registered voters residing in two of the multi-district counties,16 argued that the apportionment plan on its face violated the Equal Protection Clause because countywide voting in the seven multidistrict counties denied their residents a vote equal to that of voters residing in single-member con*119stituencies.17 We were unconvinced that the plan operated to dilute any Georgian's vote, and therefore upheld the facial validity of the scheme. We cautioned, however, that the Equal Protection Clause would not tolerate a multimember districting plan that “designedly or otherwise, . . . operate [d] to minimize or cancel out the voting strength of racial or political 'elements of the voting population.” 379 U. S., at 439 (emphasis added).
The approach to vote dilution adopted in Fortson plainly consisted of a fundamental-rights analysis. If the Court had believed that the equal protection problem with alleged vote dilution was one of racial discrimination and not abridgment of the right to vote, it would not have accorded standing to the plaintiffs, who were simply registered voters of Georgia alleging that the state apportionment plan, as a theoretical matter, diluted their voting strength because of where they lived. To the contrary, we did not question their standing, and held against them solely because we foupd unpersuasive their claim on the merits. The Court did not reach this result by inadvertence; rather, we explicitly recognized that we had adopted a fundamental-rights approach when we stated that the Equal Protection Clause protected the voting strength of political as well as racial groups.
Until today, this Court had never deviated from this principle. We reiterated that our vote-dilution doctrine protects political groups in addition to racial groups in Burns v. Richardson, 384 U. S., at 88, where we allowed a general class of qualified voters to assert such a vote-dilution claim. In Whitcomb v. Chavis, 403 U. S. 124 (1971), we again explicitly recognized that political groups could raise such claims, id., at 143, 144. In White v. Regester, 412 U. S. 755 (1973), *120the plaintiffs were Negroes and Mexican-Americans, and accordingly the Court had no reason to discuss whether non-minority plaintiffs could assert claims of vote dilution.18 In a companion case to White, however, we again recognized that “political elements” were protected against vote dilution. Gaffney v. Cummings, 412 U. S. 735, 751 (1973). Two years later, in Dallas County v. Reese, 421 U. S. 477 (1975) (per curiam), we accorded standing to urban dwellers alleging vote dilution as to the election of the county commission and stated that multimember districting is unconstitutional if it “in fact operates impermissibly to dilute the voting strength of an identifiable element of the voting population.” Id., at 480 (emphasis added). And in United Jewish Organizations v. Carey, 430 U. S. 144 (1977), the plurality opinion of Mr. Justice White stated that districting plans were subject to attack if they diluted the vote of “racial or political groups.” Id., at 167 (emphasis in original).19
Our vote-dilution decisions, then, involve the fundamental-interest branch, rather than the antidiscrimination branch, of our jurisprudence under the Equal Protection Clause. They recognize a substantive constitutional right to participate on an equal basis in the electoral process that cannot be denied or diminished for any reason, racial or otherwise, lacking quite substantial justification. They are premised on a rationale wholly apart from that underlying Washington v. Davis, 426 U. S. 229 (1976). That decision involved application of a different equal protection principle, the prohibition on racial discrimination in the governmental distribution of interests *121to which citizens have no constitutional entitlement.20 Whatever may be the merits of applying motivational analysis, to the allocation of constitutionally gratuitous benefits, that approach is completely misplaced where, as here, it is applied to the distribution of a constitutionally protected interest.21
*122Washington v. Davis, then, in no way alters the discriminatory-impact test developed in Fortson v. Dorsey, 379 U. S. 433 (1965), and applied in White v. Regester, supra, to evaluate claims of dilution of the fundamental right to vote. In my view, that test is now, and always has been, the proper method of safeguarding against inequitable distribution of political influence.
The plurality’s response is that my approach amounts to nothing less than a constitutional requirement of proportional representation for groups. See ante, at 75-80. That assertion amounts to nothing more than a red herring: I explicitly reject the notion that the Constitution contains any such requirement. See n. 7, supra. The constitutional protection against vote dilution found in our prior cases does not extend to those situations in which a group has merely failed to elect representatives in proportion to its share of the population. To prove unconstitutional vote dilution, the group is also required to carry the far more onerous burden of demonstrating that it has been effectively fenced out of the political process. See ibid. Typical of the plurality’s mischaracteri-zation of my position is its assertion that I would provide protection against vote dilution for “every 'political group,’ or at least every such group that is in the minority.” Ante, at 75. The vote-dilution doctrine can logically apply only to groups whose electoral discreteness and insularity allow dominant political factions to ignore them. See nn. 7 and 19, supra. In short, the distinction between a requirement of proportional representation and the discriminatory-effect test I espouse is by no means a difficult one, and it is hard for me to understand why the plurality insists on ignoring it.
The plaintiffs in No. 77-1844 proved that no Negro had ever been elected to the Mobile City Commission, despite the fact that Negroes constitute about one-third of the electorate, and that the persistence of severe racial bloc voting made it highly *123unlikely that any Negro could be elected at large in the foreseeable future. 423 F. Supp. 384, 387-389 (SD Ala. 1976). Contrary to the plurality’s contention, see ante, at 75-76, however, I do not find unconstitutional vote dilution in this case simply because of that showing. The plaintiffs convinced the District Court that Mobile Negroes were unable to use alternative avenues of political influence. They showed that Mobile Negroes still suffered pervasive present effects of massive historical official and private discrimination, and that the City Commission had been quite unresponsive to the needs of the minority community. The City of Mobile has been guilty of such pervasive racial discrimination in hiring employees that extensive intervention by the Federal District Court has been required. 423 F. Supp., at 389, 400. Negroes are grossly underrepresented on city boards and committees. Id., at 389-390. The city’s distribution of public services is racially discriminatory. Id., at 390-391. City officials and police were largely unmoved by Negro complaints about police brutality and a “mock lynching.” Id., at 392. The District Court concluded that “[t]his sluggish and timid response is another manifestation of the low priority given to the needs of the black citizens and of the [commissioners’] political fear of a white backlash vote when black citizens’ needs are at stake.” Ibid. See also the dissenting opinion of my Brother White, ante, p. 94.
A requirement of proportional representation would indeed transform this Court into a “super-legislature,” ante, at 76, and would create the risk that some groups would receive an undeserved windfall of political influence. In contrast, the protection against vote dilution recognized by our prior cases serves as a minimally intrusive guarantee of political survival for a discrete political minority that is effectively locked out of governmental decisionmaking processes.22 So under*124stood, the doctrine hardly “ ‘create [s] substantive constitutional rights in the name of guaranteeing equal protection of the laws/ ” ibid., quoting San Antonio Independent School Dist. v. Rodriguez, 411 U. S., at 33. Rather, the doctrine is a simple reflection of the basic principle that the Equal Protection Clause protects “[t]he right of a citizen to equal representation and to have his vote weighted equally with those of all other citizens.” Reynolds v. Sims, 377 U. S., at 576.23
*125II
Section 1 of the Fifteenth Amendment provides:
“The right of citizens of the United States to vote shall not be denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude.”
Today the plurality gives short shrift to the argument that proof of discriminatory intent is not a necessary condition to relief under this Amendment. See ante, at 61-65:24 I have examined this issue in another context and reached the contrary result. Beer v. United States, 425 U. S. 130, 146-149, and nn. 3-5 (1976) (dissenting opinion). I continue to be*126lieve that “a showing of purpose or of effect is alone sufficient to demonstrate unconstitutionality,” id., at 149, n. 5, and wish to explicate further why I find this standard appropriate for Fifteenth Amendment claims. First, however, it is necessary to address the plurality’s apparent suggestion that the Fifteenth Amendment protects against only denial, and not dilution, of the vote.25
A
The Fifteenth Amendment does not confer an absolute right to vote. See ante, at 62. By providing that the right to vote cannot be discriminatorily “denied or abridged,” however, the Amendment assuredly strikes down the diminution as well as the outright denial of the exercise of the franchise. An interpretation holding that the Amendment reaches only complete abrogation of the vote would render the Amendment essentially useless, since it is no difficult task to imagine schemes in which the Negro’s marking of the ballot is a meaningless exercise.
The Court has long understood that the right to vote encompasses protection against vote dilution. “[T]he right to have one’s vote counted” is of the same importance as “the right to put a ballot in a box.” United States v. Mosley, 238 U. S. 383, 386 (1915). See United States v. Classic, 313 U. S. 299 (1941); Swafford v. Templeton, 185 U. S. 487 (1902); Wiley v. Sinkler, 179 U. S. 58 (1900); Ex parte Yarbrough, 110 U. S. 651 (1884). The right to vote is protected against the diluting effect of ballot-box stuffing. United States v. Saylor, 322 U. S. 385 (1944); Ex parte Siebold, 100 U. S. 371 (1880). Indeed, this Court has explicitly recognized that the Fifteenth Amendment protects against vote dilution. In Terry v. Adams, 345 U. S. 461 (1953), and Smith v. Allwright, 321 U. S. *127649 (1944), the Negro plaintiffs did not question their access to the ballot for general elections. Instead they argued, and the Court recognized, that the value of their votes had been diluted by their exclusion from participation in primary elections and in the slating of candidates by political parties. The Court’s struggles with the concept of “state action” in those decisions were necessarily premised on the understanding that vote dilution was a claim cognizable under the Fifteenth Amendment.
Wright v. Rockefeller, 376 U. S. 62 (1964), recognized that an allegation of vote dilution resulting from the drawing of district lines stated a claim under the Fifteenth Amendment. The plaintiffs in that case argued that congressional district-ing in New York violated the Fifteenth Amendment because district lines had been drawn in a racially discriminatory fashion. Each plaintiff had access to the ballot; their complaint was that because of intentional discrimination they resided in a district with population characteristics that had the effect of diluting the weight of their votes. The Court treated this claim as cognizable under the Fifteenth Amendment. More recently, in United Jewish Organizations v. Carey, 430 U. S. 144 (1977), we again treated an allegation of vote dilution arising from a redistricting scheme as stating a claim under the Fifteenth Amendment. See id., at 155, 161—162, 165-168 (opinion of White, J.). Indeed, in that case Mr. Justice Stewart found no Fifteenth Amendment violation in part because the plaintiffs had failed to prove “that the redistricting scheme was employed ... to minimize or cancel out the voting strength of a minority class or interest; or otherwise to impair or burden the opportunity of affected persons to participate in the political process.” Id., at 179 (Stewart, J., joined by Powell, J., concurring in judgment) (citing, e. g., White v. Regester, 412 U. S. 755 (1973); Fortson v. Dorsey, 379 U. S. 433 (1965); Wright v. Rockefeller, supra). See also Gomillion v. Lightfoot, 364 U. S. 339 (1960).
*128It is plain, then, that the Fifteenth Amendment shares the concept of vote dilution developed in such Fourteenth Amendment decisions as Reynolds v. Sims, 377 U. S. 533 (1964), and Fortson v. Dorsey, supra. In fact, under the Court's unified view of the protections of the right to vote accorded by disparate portions of the Constitution, the concept of vote dilution is a core principle of the Seventeenth and Nineteenth Amendments as well as the Fourteenth and Fifteenth:
“The Fifteenth Amendment prohibits a State from denying or abridging a Negro's right to vote. The Nineteenth Amendment does the same for women. If a State in a statewide election weighted the male vote more heavily than the female vote or the white vote more heavily than the Negro vote, none could successfully contend that that discrimination was allowable. See Terry v. Adams, 345 U. S. 461. . . . Once the geographical unit for which a representative is to be chosen is designated, all who participate in the election are to have an equal vote — whatever their race, whatever their sex, whatever their occupation, whatever their income, and wherever their home may be in that geographical unit. This is required by the Equal Protection Clause of the Fourteenth Amendment.
“The conception of political equality from the Declaration of Independence, to Lincoln's Gettysburg Address, to the Fifteenth, Séventeenth, and Nineteenth Amendments can mean only one thing — one person, one vote.” Gray v. Sanders, 372 U. S., at 379, 381.
The plurality’s suggestion that the Fifteenth Amendment reaches only outright denial of the ballot is wholly inconsistent not only with our prior decisions, but also with the gloss the plurality would place upon the Fourteenth Amendment’s protection against vote dilution. As I explained in Part I, supra, I strongly disagree with the plurality's conclusion that our *129Fourteenth Amendment vote-dilution decisions have been based upon the Equal Protection Clause’s prohibition of racial discrimination. Be that as it may, the plurality at least does not dispute that the Fourteenth Amendment’s language — that “[n]o State shall... deny to any person within its jurisdiction the equal protection of the laws” — protects against dilution, as well as outright denial, of the right to vote on racial grounds, even though the Amendment does not mention any right to vote and speaks only of the denial, and not the diminution, of rights. Yet, when the plurality construes the language of the Fifteenth Amendment — which explicitly acknowledges the right to vote and prohibits its denial or abridgment on account of race — it seemingly would accord protection against only the absolute abrogation of the ballot.
An interpretation of the Fifteenth Amendment limiting its prohibitions to the outright denial of the ballot would convert the words of the Amendment into language illusory in symbol and hollow in substance. Surely today’s decision should not be read as endorsing that interpretation.26
B
The plurality concludes that our prior decisions establish the principle that proof of discriminatory intent is a necessary element of a Fifteenth Amendment claim.27 In contrast, I *130continue to adhere to my conclusion in Beer v. United States, 425 U. S., at 148, n. 4 (dissenting opinion), that “[t]he Court’s decisions relating to the relevance of purpose-and/ or-effect analysis in testing the constitutionality of legislative enactments are somewhat less than a seamless web.” As I there explained, at various times the Court’s decisions have seemed to adopt three inconsistent approaches: (1) that purpose alone is the test for unconstitutionality; (2) that effect alone is the test; and (3) that purpose or effect, either alone or in combination, is sufficient to show unconstitutionality. Ibid. In my view, our Fifteenth Amendment jurisprudence on the necessity of proof of discriminatory purpose is no less unsettled than was our approach to the importance of such proof in Fourteenth Amendment racial discrimination cases prior to Washington v. Davis, 426 U. S. 229 (1976). What is called for in the present cases is a fresh consideration — similar to our inquiry in Washington v. Davis, supra, with regard to Fourteenth Amendment discrimination claims— of whether proof of discriminatory purpose is necessary to establish a claim under the Fifteenth Amendment. I will first justify my conclusion that our Fifteenth Amendment precedents do not control the outcome of this issue, and then turn to an examination of how the question should be resolved.
1
The plurality cites Guinn v. United States, 238 U. S. 347 (1915); Gomillion v. Lightfoot, 364 U. S. 339 (1960); Wright v. Rockefeller, 376 U. S. 52 (1964); Lassiter v. Northampton Election Bd., 360 U. S. 45 (1959); and Lane v. Wilson, 307 U. S. 268 (1939), as holding that proof of discriminatory purpose is necessary to support a Fifteenth Amendment claim. To me, these decisions indicate confusion, not resolution of this issue. As the plurality suggests, ante, at 62, the Court in Guinn v. United States, supra, did examine the purpose of a “grandfather clause” in the course of invalidating it. Yet 24 years later, in Lane v. Wilson, supra, at 277, the Court *131struck down a more sophisticated exclusionary scheme because it “operated unfairly” against Negroes. In accord with the prevailing doctrine of the time, see Arizona v. California, 283 U. S. 423, 455, and n. 7 (1931), the Court in Lane seemingly did not question the motives of public officials.
In upholding the use of a literacy test for voters in Lassiter v. Northampton Election Bd., supra, the Court apparently concluded that the plaintiff had failed to prove either discriminatory purpose or effect. Gomillion v. Lightfoot, supra, can be read as turning on proof of discriminatory motive, but the Court also stressed that the challenged redrawing of municipal boundaries had the “essential inevitable effect” of removing Negro voters from the city, 364 U. S., at 341, and that “the inescapable human effect of this essay in geometry and geography is to despoil colored citizens, and only colored citizens, of their theretofore enjoyed voting rights,” id., at 347. Finally, in Wright v. Rockefeller, supra, the plaintiffs alleged only purposeful discriminatory redistricting, and therefore the Court had no reason to consider whether proof of discriminatory effect would satisfy the Fifteenth Amendment.28
The plurality ignores cases suggesting that discriminatory purpose is not necessary to support a Fifteenth Amendment claim. In Terry v. Adams, 345 U. S. 461 (1953), a case in which no majority opinion was issued, three Justices approvingly discussed two decisions of the United States Court of Appeals for the Fourth Circuit29 holding “that no election machinery could be sustained if its purpose or effect was to deny Negroes on account of their race an effective voice in the governmental affairs of their country, state, or community.” Id., at 466 (opinion of Black, J., joined by Douglas and Burton, JJ.) (emphasis added). More recently, in rejecting a First Amendment challenge to a federal statute provid*132ing criminal penalties for knowing destruction of a Selective Service registration certificate, the Court in United States v. O’Brien, 391 U. S. 367, 383 (1968), stated that “[i]t is a familiar principle of constitutional law that this Court will not strike down an otherwise constitutional statute on the basis of an alleged illicit legislative motive.” The Court in O’Brien, supra, at 385, interpreted Gomillion v. Lightfoot, supra, as turning on the discriminatory effect, and not the alleged discriminatory purpose, of the challenged redrawing of municipal boundaries. Three years later, in Palmer v. Thompson, 403 U. S. 217, 224-225 (1971), the Court relied on O’Brien to support its refusal to inquire whether a city had closed its swimming pools to avoid racial integration. As in O’Brien, the Court in Palmer, supra, at 225, interpreted Gomillion v. Lightfoot as focusing “on the actual effect” of the municipal boundary change, and not upon what motivated the city to redraw its borders. See also Wright v. Council of City of Emporia, 407 U. S. 451, 461-462 (1972).
Ill holding that racial discrimination claims under the Equal Protection Clause must be supported by proof of discriminatory intent, the Court in Washington v. Davis, supra, signaled some movement away from the doctrine that such proof is irrelevant to constitutional adjudication. Although the Court, 426 U. S., at 242-244, and n. 11, attempted mightily to distinguish Palmer v. Thompson, supra, its decision was in fact based upon a judgment that, in light of modern circumstances, the Equal Protection Clause’s ban on racial discrimination in the distribution of constitutional gratuities should be interpreted as prohibiting only intentional official discrimination.30
These vacillations in our approach to the relevance of discriminatory purpose belie the plurality’s determination that our prior decisions require such proof to support Fifteenth Amendment claims. To the contrary, the Court today is in *133the same unsettled position with regard to the Fifteenth Amendment as it was four years ago in Washington v. Davis, supra, regarding the Fourteenth Amendment's prohibition of racial discrimination. The absence of old answers mandates a new inquiry.
2
The Court in Washington v. Davis required a showing of discriminatory purpose to support racial discrimination claims largely because it feared that a standard based solely on disproportionate impact would unduly interfere with the far-ranging governmental distribution of constitutional gratuities.31 Underlying the Court’s decision was a determination that, since the Constitution does not entitle any person to such governmental benefits, courts should accord discretion to those officials who decide how the government shall allocate its scarce resources. If the plaintiff proved only that governmental distribution of constitutional gratuities had a disproportionate effect on a racial minority, the Court was willing to presume that the officials who approved the allocation scheme either had made an honest error or had foreseen that the decision would have a discriminatory impact and had found persuasive, legitimate reasons for imposing it nonetheless. These assumptions about the good faith of officials allowed the Court to conclude that, standing alone, a showing that a governmental policy had a racially discriminatory impact did not indicate that the affected minority had suffered the stigma, frustration, and unjust treatment prohibited *134under the suspect-classification branch of our equal protection jurisprudence.
Such judicial deference to official decisionmaking has no place under the Fifteenth Amendment. Section 1 of that Amendment differs from the Fourteenth Amendment’s prohibition on racial discrimination in two crucial respects: it explicitly recognizes the right to vote free of hindrances related to race, and it sweeps no further. In my view, these distinctions justify the conclusion that proof of racially discriminatory impact should be sufficient to support a claim under the Fifteenth Amendment. The right to vote is of such fundamental importance in the constitutional scheme that the Fifteenth Amendment’s command that it shall not be “abridged” on account of race must be interpreted as providing that the votes of citizens of all races shall be of substantially equal weight. Furthermore, a disproportionate-impact test under the Fifteenth Amendment would not lead to constant judicial intrusion into the process of official decisionmak-ing. Rather, the standard would reach only those decisions having a discriminatory effect upon the minority’s vote. The Fifteenth Amendment cannot tolerate that kind of decision, even if made in good faith, because the Amendment grants racial minorities the full enjoyment of the right to vote, not simply protection against the unfairness of intentional vote dilution along racial lines.32
In addition, it is beyond dispute that a standard based solely upon the motives of official decisionmakers creates significant problems of proof for plaintiffs and forces the inquiring court to undertake an unguided, tortuous look into the minds of officials in the hope of guessing why certain policies were adopted and others rejected. See Palmer v. Thomp*135son, 403 U. S., at 224-225; United States v. O’Brien, 391 U. S., at 382-386; cf. Keyes v. School District No. 1, Denver, Colo., 413 U. S. 189, 224, 227 (1973) (Powell, J., concurring in part and dissenting in part). An approach based on motivation creates the risk that officials will be able to adopt policies that are the products of discriminatory intent so long as they sufficiently mask their motives through the use of subtlety and illusion. Washington v. Davis is premised on the notion that this risk is insufficient to overcome the deference the judiciary must accord to governmental decisions about the distribution of constitutional gratuities. That risk becomes intolerable, however, when the precious right to vote protected by the Fifteenth Amendment is concerned.
I continue to believe, then, that under the Fifteenth Amendment an “[e] valuation of the purpose of a legislative enactment is just too ambiguous a task to be the sole tool of constitutional analysis. . . . [A] demonstration of effect ordinarily should suffice. If, of course, purpose may conclusively be shown, it too should be sufficient to demonstrate a statute’s unconstitutionality.” Beer v. United States, 425 U. S., at 149-150, n. 5 (Marshall, J., dissenting). The plurality’s refusal in this case even to consider this approach bespeaks an indifference to the plight of minorities who, through no fault of their own, have suffered diminution of the right preservative of all other rights.33
*136Ill
If it is assumed that proof of discriminatory intent is necessary to support the vote-dilution claims in these cases, the question becomes what evidence will satisfy this requirement.34
The plurality assumes, without any analysis, that these cases are appropriate for the application of the rigid test developed in Personnel Administrator of Mass. v. Feeney, 442 U. S., at 279, requiring that “the decisionmaker . . . selected or reaffirmed a particular course of action at least in part 'because of,’ not merely 'in spite of,’ its adverse effects upon an identifiable group.” In my view, the Feeney standard creates a burden of proof far too extreme to apply in vote-dilution cases.35
*137This Court has acknowledged that the evidentiary inquiry involving discriminatory intent must necessarily vary depending upon the factual context. See Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U. S., at 264-268; Washington v. Davis, 426 U. S., at 253 (Stevens, J., concurring). One useful evidentiary tool, long recognized by the common law, is the presumption that “[e]very man must be taken to contemplate the probable consequences of the act he does.” Townsend v. Wathen, 9 East. 277, 280, 103 Eng. Rep. 579, 580-581 (K. B. 1808). The Court in Feeney, supra, at 279, n. 25, acknowledged that proof of foreseeability of discriminatory consequences could raise a “strong inference that the adverse effects were desired,” but refused to treat this presumption as conclusive in cases alleging discriminatory distribution of constitutional gratuities.
I would apply the common-law foreseeability presumption to the present cases. The plaintiffs surely proved that maintenance of the challenged multimember districting would have the foreseeable effect of perpetuating the submerged electoral influence of Negroes, and that this discriminatory effect could be corrected by implementation of a single-member districting plan.36 Because the foreseeable disproportionate impact was so severe, the burden of proof should have shifted to the defendants, and they should have been required to show that they refused to modify the districting schemes in spite of, not because of, their severe discriminatory effect. See Feeney, supra, at 284 (Marshall, J., dissenting). Reallocation of the burden of proof is especially appropriate in these cases, where the challenged state action infringes the exercise of a fundamental right. The defendants would carry their burden of proof only if they showed that they considered submergence *138of the Negro vote a detriment, not a benefit, of the multi-member systems, that they accorded minority citizens the same respect given to whites, and that they nevertheless decided to maintain the systems for legitimate reasons. Cf. Mt. Healthy City Board of Ed. v. Doyle, 429 U. S. 274, 287 (1977); Arlington Heights v. Metropolitan Housing Dev. Corp., supra, at 270-271, n. 21.
This approach recognizes that
“[frequently the most probative evidence of intent will be objective evidence of what actually happened rather than evidence describing the subjective state of mind of the actor. For normally the actor is presumed to have intended the natural consequences of his deeds. This is particularly true in the case of governmental action which is frequently the product of compromise, of collective decisionmaking, and of mixed motivation.” Washington v. Davis, supra, at 253 (Stevens, J., concurring).
Furthermore, if proof of discriminatory purpose is to be required in these cases, this standard would comport with my view that the degree to which the government must justify a decision depends upon the importance of the interests infringed by it. See San Antonio Independent School Dist. v. Rodriguez, 411 U. S., at 109-110 (Marshall, J., dissenting).37
*139The plurality also fails to recognize that the maintenance of multimember districts in the face of foreseeable discriminatory consequences strongly suggests that officials are blinded by “racially selective sympathy and indifference.”38 Like outright racial hostility, selective racial indifference reflects a belief that the concerns of the minority are not worthy of the same degree of attention paid to problems perceived by whites. When an interest as fundamental as voting is diminished along racial lines, a requirement that discriminatory purpose must be proved should be satisfied by a showing that official action was produced by this type of pervasive bias. In the present cases, the plaintiffs presented strong evidence of such bias: they showed that Mobile officials historically discriminated against Negroes, that there are pervasive present effects of this past discrimination, and that officials have not been responsive to the needs of the minority community. It takes only the smallest of inferential leaps to conclude that the decisions to maintain multimember districting having obvious discriminatory effects represent, at the very least, selective racial sympathy and indifference resulting in the frustration of minority desires, the stigmatization of the minority as second-class citizens, and the perpetuation of inhumanity.39
*140IV
The American approach to government is premised on the theory that, when citizens have the unfettered right to vote, *141public officials will make decisions by the democratic accommodation of competing beliefs, not by deference to the mandates of the powerful. The American approach to civil rights is premised on the complementary theory that the unfettered right to vote is preservative of all other rights. The theoretical foundations for these approaches are shattered where, as in the present cases, the right to vote is granted in form, but denied in substance.
It is time to realize that manipulating doctrines and drawing improper distinctions under the Fourteenth and Fifteenth Amendments, as well as under Congress’ remedial legislation enforcing those Amendments, make this Court an accessory to the perpetuation of racial discrimination. The plurality’s requirement of proof of intentional discrimination, so inappropriate in today’s cases, may represent an attempt to bury the legitimate concerns of the minority beneath the soil of a doctrine almost as impermeable as it is specious. If so, the superficial tranquility created by such measures can be but short-lived. If this Court refuses to honor our long-recognized principle that the Constitution “nullifies sophisticated as well as simple-minded modes of discrimination,” Lane v. Wilson, 307 U. S., at 275, it cannot expect the victims of discrimination to respect political channels of seeking redress. I dissent.
[This opinion applies also to No. 78-357, Williams et al. v. Brown et al., post, p. 236.]
U. S. Const., Arndts. 15, 17,19, 23, 24, 26.
I agree with the plurality, see ante, at 60-61, that the prohibition on denial or infringement of the right to vote contained in § 2 of the Voting Rights Act, 42 U. S. C. § 1973, contains the same standard as the Fifteenth Amendment. I disagree with the plurality’s construction of that Amendment, however. See Part II, infra.
The Court does not quarrel with the generalization that in many instances an electoral minority will fare worse under multimember districting than under single-member districting. Multimember districting greatly enhances the opportunity of the majority political faction to elect all representatives of the district. In contrast, if the multimember district is divided into several single-member districts, an electoral minority will have a better chance to elect a candidate of its choice, or at least to exert greater political influence. It is obvious that the greater the degree to which the electoral minority is homogeneous and insular and the greater the degree that bloc voting occurs along majority-minority lines, the greater will be the extent to which the minority’s voting power is diluted by multi-member districting. See E. Banfield & J. Wilson, City Politics 91-96, 303-308 (1963); R. Dixon, Jr., Democratic Representation 12, 476-484, 503-527 (1968); Bonapfel, Minority Challenges to At-Large Elections: The Dilution Problem, 10 Ga. L. Rev. 353, 358-360 (1976); Derfner, Racial Discrimination and the Right to Vote, 26 Vand. L. Rev. 523, 553-555 (1973); Comment, Effective Representation and Multimember Districts, 68 Mich. L. Rev. 1577, 1577-1579 (1970). Recent empirical studies have documented the validity of this generalization. See Berry & Dye, The Discriminatory Effects of At-Largé Elections, 7 Fla. St. U. L. Rev. 85, 113-122 (1979); Jones, The Impact of Local Election Systems on Black *106Political Representation, 11 Urb. Aff. Q. 345 (1976); Kamig, Black Resources and City Council Representation, 41 J. Pol. 134 (1979); Kamig, Black Representation on City Councils: Tbe Impact of District Elections and Socioeconomic Factors, 12 Urb. Aff. Q. 223 (1976); Sloan, “Good Government” and the Politics of Race, 17 Soc. Prob. 161 (1969) ; The Impact of Municipal Reformism: A Symposium, 59 Soc. Sci. Q. 117 (1978).
The electoral schemes in these cases involve majority-vote, numbered-post, and staggered-term requirements. See Bolden v. City of Mobile, 423 F. Supp. 384, 386-387 (SD Ala. 1976); Brown v. Moore, 428 F. Supp. 1123, 1126-1127 (SD Ala. 1976). These electoral rales exacerbate the vote-dilutive effects of multimember districting. A requirement that a candidate must win by a majority of the vote forces a minority candidate who wins a plurality of votes in the general election to engage in a runoff election with his nearest competitor. If the competitor is a member of the dominant political faction, the minority candidate stands little chance of winning in the second election. A requirement that each candidate must run for a particular “place” or “post” creates head-to-head contests that minority candidates cannot survive. When a number of positions on a governmental body are to be chosen in the same election, members of a minority will increase the likelihood of election of a favorite candidate by voting only for him. If the remainder of the electorate splits its votes among the other candidates, the minority’s candidate might well be elected by the minority’s “single-shot voting.” If the terms of the officeholders are staggered, the opportunity for single-shot voting is decreased. See City of Rome v. United States, post, p. 156; Zimmer v. McKeithen, 485 F. 2d 1297, 1305 (CA5 1973) (en banc), aff’d on other grounds sub nom. East Carroll Parish School Bd. v. Marshall, 424 U. S. 636 (1976) (per curiam) ; Bonapfel, supra; Derfner, supra.
The plurality notes that at-large elections were instituted in cities as a reform measure to correct corruption and inefficiency in municipal government, and suggests that it “may be a rash assumption” to apply vote-dilu*107tion concepts to a municipal government elected in that fashion. See ante, at 70, and n. 15. To the contrary, local governments are not exempt from the constitutional requirement to adopt representational districting ensuring that the votes of each citizen will have equal weight. Avery v. Midland County, 390 U. S. 474 (1968). Indeed, in Beer v. United States, 425 U. S. 130, 142, n. 14 (1976), and Abate v. Mundt, 403 U. S. 182, 184, n. 2 (1971), we assumed that our vote-dilution doctrine applied to local governments.
Furthermore, though municipalities must be accorded some discretion in arranging their affairs, see Abate v. Mundt, supra, there is all the more reason to scrutinize assertions that municipal, rather than state, multi-member districting dilutes the vote of an electoral minority:
“In statewide elections, it is possible that a large minority group in one multi-member district will be unable to elect any legislators, while in another multi-member district where the same group is a slight majority, they will elect the entire slate of legislators. Thus, the multi-member electoral system may hinder a group in one district but prove an advantage in another. In at-large elections in cities this is not possible. There is no way to balance out the discrimination against a particular minority group because the entire city is one huge election district. The minority’s loss is absolute.” Berry & Dye, supra n. 3, at 87.
That at-large elections were instituted as part of a “reform” movement in no way ameliorates these harsh effects. Moreover, in some instances the efficiency and breadth of perspective supposedly resulting from a reform structure of municipal government are achieved at a high cost. In a white-majority city in which severe racial bloc voting is common, the citywide view allegedly inculcated in city commissioners by at-large elections need not extend beyond the white community, and the efficiency of the commission form of government can be achieved simply by ignoring the concerns of the powerless minority.
It would be a mistake, then, to conclude that municipal at-large elections provide an inherently superior representational scheme. See also n. 3, supra; Chapman v. Meier, 372 F. Supp. 371, 388-392 (ND 1974) (three-judge court) (Bright, J., dissenting), rev’d, 420 U. S. 1 (1975). It goes without saying that a municipality has the freedom to design its own governance system. When that system is subjected to constitutional attack, however, the question is whether it was enacted or maintained with *108a discriminatory purpose or.has a discriminatory effect, not whether it comports with one or another of the competing notions about “good government.”
As the plurality notes, see ante, at 66, we indicated in Whitcomb v. Chavis, 403 U. S., at 149, that multimember districts were unconstitutional if they were “conceived or operated as purposeful devices to further racial or economic discrimination.” The Court in Whitcomb did not, however, suggest that discriminatory purpose was a necessary condition for the invalidation of multimember districting. Our decision in Whitcomb, supra, at 143, acknowledged the continuing validity of the discriminatory-impact test adopted in Fortson v. Dorsey, 379 U. S. 433, 439 (1965), and restated it as requiring plaintiffs to prove that “multi-member districts unconstitutionally operate to dilute or cancel the voting strength of racial or political elements.” Whitcomb, supra, at 144 (emphasis added).
Abate v. Mundt, supra, decided the same day as Whitcomb, provides further evidence that Whitcomb did not alter the discriminatory-effects standard developed in earlier cases. In Abate, supra, at 184, n. 2, we rejected the argument that a multimember districting scheme had a vote-dilutive effect because “[p)etitioners . . . have not shown that these multi-member districts, by themselves, operate to impair the voting strength of particular racial or political elements , , see Burns v. Richardson, 384 U. S. 73, 88 (1966).”
See r. 3, supra.
White v. Regester, makes clear the distinction between the concepts of vote dilution and proportional representation. We have held that, in order to prove an allegation of vote dilution, the plaintiffs must show more than simply that they have been unable to elect candidates of their choice. See 412 U. S., at 765-766; Whitcomb v. Chavis, supra, at 149-150, 153. The Constitution, therefore, does not contain any requirement of proportional representation. Cf. United Jewish Organizations v. Carey, 430 U. S. 144 (1977); Gaffney v. Cummings, 412 U. S. 735 (1973). When all that is proved is mere lack of success at the polls, the Court will not presume that members of a political minority have suffered an impermissible dilution of political power. Rather, it is assumed that these persons have means available to them through which they can have some effect on governmental decisionmaking. For example, many of these persons might belong to a variety of other political, social, and economic groups that have some impact on officials. In the absence of evidence to the contrary, it may be assumed that officials will not be improperly influenced by such factors as the race or place of residence of persons seeking governmental action. Furthermore, political factions out of office often serve as watchdogs on the performance of the government, bind together into coalitions having enhanced influence, and have the respectability necessary to affect public policy.
Unconstitutional vote dilution occurs only when a discrete political minority whose voting strength is diminished by a districting scheme proves that historical and social factors render it largely incapable of effectively utilizing alternative avenues of influencing public policy. See n. 19, injra. In these circumstances, the only means of breaking down the barriers encasing the political arena is to structure the electoral dis-tricting so that the minority has a fair opportunity to elect candidates of its choice.
The test for unconstitutional vote dilution, then, looks only to the discriminatory effects of the combination of an electoral structure and historical and social factors. At the same time, it requires electoral minorities to prove far more than mere lack of success at the polls.
We have also spoken of dilution of voting power in cases arising under the Voting Rights Act of 1965, 42 U. S. C. § 1973 et seq. Under § 5 of *112that Act, 42 U. S. C. § 1973c, a state or local government covered by the Act may not enact new electoral procedures having the purpose or effect of denying or abridging the right to vote on account of race or color. We have interpreted this provision as prohibiting any retrogression in Negro voting power. Beer v. United States, 425 U. S. 130, 141 (1976). In some cases, we have labeled such retrogression a “dilution” of the minority vote. See, e. g., City of Rome v. United States, post, p. 156. Vote dilution under § 5, then, involves a standard different from that applied in cases such as White v. Regester, supra, in which diminution of the vote violating the Fourteenth or Fifteenth Amendment is alleged.
The plurality’s approach is also inconsistent with our statement in Dallas County v. Reese, 421 U. S. 477, 480 (1975) (per curiam), that multimem-ber districting violates the Equal Protection Clause if it “in fact operates impermissibly to dilute the voting strength of an identifiable element of the voting population.” See also Chapman v. Meier, 420 U. S., at 17.
See Shapiro v. Thompson, 394 U. S. 618 (1969) (right to travel); Reynolds v. Sims, 377 U. S. 533 (1964) (right to vote); Douglas v. California, 372 U. S. 353 (1963); and Griffin v. Illinois, 351 U. S. 12 (1956) (right to fair access to criminal process). Under the rubric of the fundamental right of privacy, we have recognized that individuals have freedom from unjustified governmental interference with personal decisions involv*114ing marriage, Zablocki v. Redhail, 434 U. S. 374 (1978); Loving v. Virginia, 388 U. S. 1 (1967); procreation, Skinner v. Okahoma ex rel. Williamson, 316 U. S. 535 (1942); contraception, Carey v. Population Services International, 431 U. S. 678 (1977); Eisenstadt v. Baird, 405 U. S. 438 (1972); Griswold v. Connecticut, 381 U. S. 479 (1965); abortion, Roe v. Wade, 410 U. S. 113 (1973); family relationships, Prince v. Massachusetts, 321 U. S. 158 (1944); and child rearing and education, Pierce v. Society of Sisters, 268 U. S. 510 (1925); Meyer v. Nebraska, 262 U. S. 390 (1923). See also Moore v. East Cleveland, 431 U. S. 494 (1977).
As the present cases illustrate, a requirement of proof of discriminatory intent seriously jeopardizes the free exercise of the fundamental right to vote. Although the right to vote is indistinguishable for present purposes from the other fundamental rights our cases have recognized, see n. 9, supra, surely the plurality would not require proof of discrimina^ tory purpose in those cases. The plurality fails to articulate why the right to vote should receive such singular treatment. Furthermore, the plurality refuses to recognize the disutility of requiring proof of discriminatory purpose in fundamental rights cases. For example, it would make no sense to require such a showing when the question is whether a state statute regulating abortion violates the right of personal choice recognized in Roe v. Wade, supra. The only logical inquiry is whether, regardless of the legislature’s motive, the statute has the effect of infringing that right. See, e. g., Planned Parenthood of Central Missouri v. Danforth, 428 U. S. 52 (1976).
Judge Wisdom of the Court of Appeals below recognized this distinction in a companion case, see Nevett v. Sides, 571 F. 2d 209, 231-234 (CA5 1978) (specially concurring opinion). See also Comment, Proof of *115Racially Discriminatory Purpose Under the Equal Protection Clause: Washington v. Davis, Arlington Heights, Mt. Healthy, and Williamsburgh, 12 Harv. Civ. Rights-Civ. Lib. L. Rev. 725, 758, n. 175 (1977); Note, Racial Vote Dilution in Multimember Districts: The Constitutional Standard After Washington v. Davis, 76 Mich. L. Rev. 694, 722-726 (1978); Comment, Constitutional Challenges to Gerrymanders, 45 U. Chi. L. Rev. 845, 869-877 (1978).
Washington v. Davis, 426 U. S. 229 (1976), involved alleged racial discrimination in public employment. By describing interests such as public employment as constitutional gratuities, I do not, of course, mean to suggest that their deprivation is immune from constitutional scrutiny. Indeed, our decisions have referred to the importance of employment, see Hampton v. Mow Sun Wong, 426 U. S. 88, 116 (1976); Meyer v. Nebraska, supra, at 399; Truax v. Raich, 239 U. S. 33, 41 (1915), and we have explicitly recognized that in some circumstances public employment falls within the categories of liberty and property protected by the Fifth and Fourteenth Amendments, see, e. g., Arnett v. Kennedy, 416 U. S. 134 (1974); Perry v. Sindermann, 408 U. S. 593 (1972). The Court has not held, however, that a citizen has a constitutional right to public employment.
We have not, however, held that the Fourteenth Amendment contains an absolute right to vote. As we explained in Dunn v. Blumstein, 405 U. S. 330 (1972):
“In decision after decision, this Court has made clear that a citizen has a constitutionally protected right to participate in elections on an equal basis with other citizens in the jurisdiction. [Citing cases.] This ‘equal right to vote’ ... is not absolute; the States have the power to impose voter qualifications, and to regulate access to the franchise in other ways. . . . But, as a general matter, ‘before that right [to vote] can be restricted, the purpose of the restriction and the assertedly overriding interests served by it must' meet close constitutional scrutiny.’ ” Id., at 336 (quoting Evans v. Cornman, 398 U. S. 419, 426, 422 (1970)).
Avery v. Midland County, 390 U. S. 474 (1968), applied the equal-representation standard of Reynolds v. Sims to local governments. See also, e. g., Connor v. Finch, 431 U. S. 407 (1977); Lockport v. Citizens for Community Action, 430 U. S. 259 (1977); Hadley v. Junior College Dist., 397 U. S. 50 (1970).
In attempting to limit Reynolds v. Sims to its facts, see ante, at 77-79, the plurality confuses the nature of the constitutional right recognized in that decision with the means by which that right can be violated. Reynolds held that under the Equal Protection Clause each citizen must *117be accorded an essentially equal voice in the election of representatives. The Court determined that unequal population distribution in a multi-district representational scheme was one readily ascertainable means by which this right was abridged. The Court certainly did not suggest, however, that violations of the right to effective political participation mattered only if they were caused by malapportionment. The plurality’s assertion to the contrary in this case apparently would require it to read Reynolds as recognizing fair apportionment as an end in itself, rather than as simply a means to protect against vote dilution.
Proof of discriminatory purpose has been equally unnecessary in our decisions assessing whether various impediments to electoral participation are inconsistent with the fundamental interest in voting. In the seminal case, Harper v. Virginia Bd. of Elections, 383 U. S. 663 (1966), we invalidated a $1.50 poll tax imposed as a precondition to voting. Relying on our decision two years earlier in Reynolds v. Sims, see Harper, supra, at 667-668, 670, we determined that “the right to vote is too precious, too fundamental to be so burdened or conditioned,” 383 U. S., at 670. We analyzed the right to vote under the familiar standard that “where fundamental rights and liberties are asserted under the Equal Protection Clause, classifications which might invade or restrain them must be closely scrutinized and carefully confined.” Ibid. In accord with Harper, we have applied heightened scrutiny in assessing the imposition of filing fees, e. g., Lubin v. Panish, 415 U. S. 709 (1974); limitations on who may participate in elections involving specialized governmental entities, e. g., Kramer v. Union School District, 395 U. S. 621 (1969); dura-tional residency requirements, e. g., Dunn v. Blumstein, supra; enrollment time limitations for voting in party primary elections, e. g., Kusper v. Pontikes, 414 U. S. 51 (1973); and restrictions on candidate access to the ballot, e. g., Illinois Elections Bd. v. Socialist Workers Party, 440 U. S. 173 (1979).
To be sure, we have approved some limitations on the right to vote. Compare, e. g., Salyer Land Co. v. Tulare Water District, 410 U. S. 719 (1973), with Kramer v. Union School District, supra. We have never, however, required a showing of discriminatory purpose to support a claim of infringement of this fundamental interest. To the contrary, the Court *118has accepted at face value the purposes articulated for a qualification of this right, and has invalidated such a limitation under the Equal Protection Clause only if its purpose either lacked sufficient substantiality when compared to the individual interests affected or could have been achieved by less restrictive means. See, e. g., Dunn v. Blumstein, supra, at 335, 337, 343-360.
The approach adopted in this line of cases has been synthesized with the one-person, one-vote doctrine of Reynolds v. Sims in the following fashion: “It has been established in recent years that the Equal Protection Clause confers the substantive right to participate on an equal basis with other qualified voters whenever the State has adopted an electoral process for determining who will represent any segment of the State’s population.” San Antonio Independent School Dist. v. Rodriguez, 411 U. S. 1, 59, n. 2 (1973) (Stewart, J., concurring) (citing Reynolds v. Sims, 377 U. S. 533 (1964); Kramer v. Union School District, supra; Dunn v. Blumstein, supra). It is plain that this standard requires no showing of discriminatory purpose to trigger strict scrutiny of state interference with the right to vote.
See Dorsey v. Fortson, 228 F. Supp. 259, 261 (ND Ga. 1964) (three-judge court), rev’d, 379 U. S. 433 (1965).
Specifically, the plaintiffs contended that countywide voting in the multidistrict counties could, as a matter of mathematics, result in the nullification of the unanimous choice of the voters of one district. Fortson v. Dorsey, 379 U. S., at 436-437.
The same is true of our most recent case discussing vote dilution, Wise v. Lipscomb, 437 U. S. 535 (1978).
In contrast to a racial group, however, a political group will bear a rather substantial burden of showing that it is sufficiently discrete to suffer vote dilution. See Dallas County v. Reese, 421 U. S. 477 (1975) (per curiam) (allowing city dwellers to attack a countywide multimember district). See generally Comment, Effective Representation and Multi-member Districts, 68 Mich. L. Rev. 1577, 1594-1596 (1970).
The dispute in Washington v. Davis concerned alleged racial discrimination in public employment, an interest to which no one has a constitutional right, see n. 11, supra. In that decision, the Court held only that “the invidious quality of a law claimed to be racially discriminatory must ultimately be traced to a racially discriminatory purpose.” 426 U. S., at 240 (emphasis added). The Court’s decisions following Washington v. Davis have also involved alleged discrimination in the allocation of interests falling short of constitutional rights. Personnel Administrator of Mass. v. Feeney, 442 U. S. 256 (1979) (alleged sex discrimination in public employment); Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U. S. 252 (1977) (alleged racial discrimination in zoning). As explained in Feeney, supra, “[w]hen some other independent right is not at stake . . . and when there is no ‘reason to infer antipathy,’ ... it is presumed that ‘even improvident decisions will eventually be rectified by the democratic process.’ ” 442 U. S., at 272 (quoting Vance v. Bradley, 440 U. S. 93, 97 (1979)).
Professor Ely has recognized this distinction:
“The danger I see is . . . that the Court, in its newfound enthusiasm for motivation analysis, will seek to export it to fields where it has no business. It therefore cannot be emphasized too strongly that analysis of motivation is appropriate only to claims of improper discrimination in the distribution of goods that are constitutionally gratuitous (that is, benefits to which people are not entitled as a matter of substantive constitutional right).... However, where what is denied is something to which the complainant has a substantive constitutional right — either because it is granted by the terms of the Constitution, or because it is essential to the effective functioning of a democratic government — the reasons it was denied are irrelevant. It may become important in court what justifications counsel for the state can articulate in support of its denial or nonprovision, but the reasons that actually inspired the denial never can: To have a right to something is to have a claim on it irrespective of why it is denied. It would be a tragedy of the first order were the Court to expand its burgeoning awareness of the relevance of motivation into the thoroughly mistaken notion that a denial of a constitutional right does not count as such unless it was intentional.” Ely, The Centrality and Limits of Motivation Anal*122ysis, 15 San Diego L. Rev. 1155, 1160-1161 (1978) (emphasis in original) (footnotes omitted).
It is at this point that my view most diverges from the position expressed by my Brother Stevens, ante, p. 83. He would strictly scrutinize *124state action having an adverse impact on an individual’s right to vote. In contrast, he would apply a less stringent standard to state action diluting the political influence of a group. See ante, at 83-85. The facts of the present cases, however, demonstrate that severe and persistent racial bloc voting, when coupled with the inability of the minority effectively to participate in the political arena by alternative means, can effectively disable the individual Negro as well as the minority community as a whole. In these circumstances, Mr. Justice Stevens’ distinction between the rights of individuals and the political strength of groups becomes illusory.
The foregoing disposes of any contention that, merely by citing Wright v. Rockefeller, 376 U. S. 52 (1964), the Court in Washington v. Davis, 426 U. S., at 240, and Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U. S., at 264, intended to bring vote-dilution cases within the discriminatory-purpose requirement. Wright v. Rockefeller, supra, was a racial gerrymander case, and the plaintiffs had alleged only that they were the victims of an intentional scheme to draw districting lines discriminatorily. In focusing solely on whether the plaintiffs had proved intentional discrimination, the Court in Wright v. Rockefeller was merely limiting the scope of its inquiry to the issue raised by the plaintiffs. If Wright v. Rockefeller had been brought after this Court had decided our vote-dilution decisions, the plaintiffs perhaps would have recognized that, in addition to a claim of intentional racial gerrymandering, they could allege an equally sufficient cause of action under the Equal Protection Clause — that the districting lines had the effect of diluting their vote.
Wright v. Rockefeller, then, treated proof of discriminatory purpose as a sufficient condition to trigger strict scrutiny of a districting scheme, but had no occasion to consider whether such proof was necessary to invoke that standard. Its citations in Washington v. Davis, supra, and Arlington *125Heights, supra, were useful to show the relevancy, but not the necessity, of evidence of discriminatory intent. These citations are in no way inconsistent with my view that proof of discriminatory purpose is not a necessary condition to the invalidation of multimember districts that dilute the vote of racial or political elements.
In addition, any argument that, merely by citing Wright v. Rockefeller, the Court in Washington v. Davis and Arlington Heights intended to apply the discriminatory-intent requirement to vote-dilution claims is premised on two unpalatable assumptions. First, because the discussion of Wright v. Rockefeller was unnecessary to the resolution of the issues in both of those decisions, the argument assumes that the Court in both cases decided important issues in brief dicta. Second, the argument assumes that the Court twice intended covertly to overrule the discriminatory-effects test applied in White v. Regester, 412 U. S. 755 (1973), without even citing White. Neither assumption is tenable.
It is important to recognize that only the four Members of the plurality are committed to this view. In addition to my Brother Brennan and myself, my Brother Stevens expressly states that proof of discriminatory effect can be a sufficient condition to support the invalidation of dis-tricting, see ante, at 90. My Brother White finds the proof of discriminatory purpose in these cases sufficient to support the decisions of the Courts of Appeals, and accordingly he does not reach the issue whether proof of discriminatory impact, standing alone, would suffice under the Fifteenth Amendment. My Brother Blackmun also expresses no view on this issue, since he too finds the proof of discriminatory intent sufficient to support the findings of violations of the Constitution.
The plurality states that “[h]aving found that Negroes in Mobile 'register and vote without hindrance,’ the District Court and Court of Appeals were in error in believing that the appellants invaded the protection of that Amendment in the present case.” Ante, at 65.
Indeed, five Members of the Court decline the opportunity to ascribe to this view. In addition to my Brother Brennan and myself, my Brother Stevens expressly states that the Fifteenth Amendment protects against diminution as well as denial of the ballot, see ante, at 84, and n. 3. The dissenting opinion of my Brother White and the separate opinion of my Brother Blackmun indicate that they share this view.
The plurality does not attempt to support this proposition by relying on the history surrounding the adoption of the Fifteenth Amendment. I agree that we should resolve the issue of the relevancy of proof of discriminatory purpose and effect by examining our prior decisions and by considering the appropriateness of alternative standards in light of contemporary circumstances. That was, of course, the approach used in Washington v. Davis, 426 U. S. 229 (1976), to evaluate that issue with regard to Fourteenth Amendment racial discrimination claims.
See n. 23, supra.
Rice v. Elmore, 165 F. 2d 387 (1947), cert. denied, 333 U. S. 875 (1948), and Baskin v. Brown, 174 F. 2d 391 (1949).
See rm. 20, 21, supra, and accompanying text.
The Court stated:
“A rule that a statute designed to serve neutral ends is nevertheless invalid, absent compelling justification, if in practice it benefits or burdens one race more than another would be far reaching and would raise serious questions about, and perhaps invalidate, a whole range of tax, welfare, public service, regulatory, and licensing statutes that may be more burdensome to the poor and to the average black than to the more affluent white.” 426 U. S., at 248.
See n. 20, supra.
Even if a municipal policy is shown to dilute the right to vote, however, the policy will not be struck down if the city shows that it serves highly important local interests and is closely tailored to effectuate only those interests. See Dunn v. Blumstein, 405 U. S. 330 (1972). Cf. Abate v. Mundt, 403 U. S. 182 (1971).
In my view, the standard of White v. Regester, 412 U. S. 755 (1973), see n. 7, supra, and accompanying text, is the proper test under both the Fourteenth and Fifteenth Amendments for determining whether a district-ing scheme has the unconstitutional effect of diluting the Negro vote. It is plain that the District Court in both of the cases before us made the “intensely local appraisal” necessary under White, supra, at 769, and correctly decided that the at-large electoral schemes for the Mobile City Commission and County School Board violated the White standard. As I earlier note with respect to No. 77-1844, see supra, at 122-123, the District Court determined: (1) that Mobile Negroes still suffered pervasive present effects of massive historical official and private discrimination; (2) that the City Commission and County School Board had been quite *136unresponsive to the needs of the minority community; (3) that no Negro had ever been elected to either body, despite the fact that Negroes constitute about one-third of the electorate; (4) that the persistence of severe racial bloc voting made it highly unlikely that any Negro could be elected at large to either body in the foreseeable future; and (5) that no state policy favored at-large elections, and the local preference for that scheme was outweighed by the fact that the unconstitutional vote dilution could be corrected only by the imposition of single-member districts. Bolden v. City of Mobile, 423 F. Supp. 384 (SD Ala. 1976); Brown v. Moore, 428 F. Supp. 1123 (SD Ala. 1976). The Court of Appeals affirmed these findings in all respects. Bolden v. City of Mobile, 571 F. 2d 238 (CA5 1978); Brown v. Moore, 575 F. 2d 298 (CA5 1978). See also the dissenting opinion of my Brother White, ante, p. 94.
The statutes providing for at-large election of the members of the two governmental bodies involved in these cases, see n. 33, supra, have been in effect since the days when Mobile Negroes were totally disenfranchised by the Alabama Constitution of 1901. The District Court in both cases found, therefore, that the at-large schemes could not have been adopted for discriminatory purposes. Bolden v. City of Mobile, 423 F. Supp., at 386, 397; Brown v. Moore, 428 F. Supp., at 1126-1127, 1138. The issue is, then, whether officials have maintained these electoral systems for discriminatory purposes. Cf. Arlington Heights v. Metropolitan Housing Dev. Corp., 429 U. S., at 257-258, 267-271, and n. 17.
As the dissenting opinion of my Brother White demonstrates, however, the facts of these cases compel a finding of unconstitutional vote dilution even under the plurality’s standard.
Indeed, the District Court in the present cases concluded that the evidence supported the plaintiffs' position that unconstitutional vote dilution was the natural and foreseeable consequence of the maintenance of the challenged multimember districting. Brown v. Moore, 428 F. Supp., at 1138; Bolden v. City of Mobile, 423 F. Supp., at 397-398.
Mr. Justice Stevens acknowledges that both discriminatory intent and discriminatory effect are present in No. 77-1844. See ante, at 92-94. Nonetheless, he finds no constitutional violation, apparently because he believes that the electoral structure of Mobile conforms to a commonly used scheme, the discriminatory impact is in his view not extraordinary, and the structure is supported by sufficient noninvidious justifications so that it is neither wholly irrational nor entirely motivated by discriminatory animus. To him, racially motivated decisions in this setting are an inherent part of the political process and do not involve invidious discrimination.
The facts of the' present cases, however, indicate that in Mobile considerations of race are far more powerful and pernicious than are considerations of other divisive aspects of the electorate. See supra, at 122-123. In Mobile, as elsewhere, “the experience of Negroes . . . has been different *139in kind, not just in degree, from that of other ethnic groups.” University of California Regents v. Bakke, 438 U. S. 265, 400 (1978) (opinion of Marshall, J.). An approach that accepts intentional discrimination against Negroes as merely an aspect of “politics as usual” strikes at the very hearts of the Fourteenth and Fifteenth Amendments.
Brest, The Supreme Court, 1975 Term — Foreword: In Defense of the Antidiscrimination Principle, 90 Harv. L. Rev. 1, 7 (1976). See also Note, Racial Vote Dilution in Multimember Districts: The Constitutional Standard After Washington v. Davis, 76 Mich. L. Rev. 694, 716-719 (1978).
The plurality, ante, at 74r-75, n. 21, indicates that on remand the lower courts are to examine the evidence in these cases under the discriminatory-intent standard of Personnel Administrator of Mass. v. Feeney, 442 U. S. 256 (1979), and may conclude that this test is met by proof of the refusal of Mobile’s state-legislative delegation to stimulate the passage *140of legislation changing Mobile’s city government into a mayor-council system in which council members are elected from single-member districts. The plurality concludes, then, only that the District Court and the Court of Appeals in each of the present cases evaluated the evidence under an improper legal standard, and not that the evidence fails to support a claim under Feeney, supra. When the lower courts examine these cases under the Feeney standard, they should, of course, recognize the relevancy of the plaintiffs’ evidence that vote dilution was a foreseeable and natural consequence of the maintenance of the challenged multimember districting, and that officials have apparently exhibited selective racial sympathy and indifference. Cf. Dayton Board of Education v. Brinkman, 443 U. S. 526 (1979); Columbus Board of Education v. Penick, 443 U. S. 449 (1979).
Finally, it is important not to confuse the differing views the plurality and I have on the elements of proving unconstitutional vote dilution. The plurality concludes that proof of intentional discrimination, as defined in Feeney, supra, is necessary to support such a claim. The plurality finds this requirement consistent with the statement in White v. Regester, 412 U. S., at 766, that unconstitutional, vote dilution does not occur simply because a minority has not been able to elect representatives in proportion to its voting potential. The extra necessary element, according to the plurality, is a showing of discriminatory intent. In the plurality’s view, the evidence presented in White going beyond mere proof of under-representation of the minority properly supported an inference that the multimember districting scheme in question was tainted with a discriminatory purpose.
The plurality’s approach should be satisfied, then, by proof that an electoral scheme enacted with a discriminatory purpose effected a retrogression in the minority’s voting power. Cf. Beer v. United States, 425 U. S. 130, 141 (1976). The standard should also be satisfied by proof that a scheme maintained for a discriminatory purpose has the effect of submerging minority electoral influence below the level it would have under a reasonable alternative scheme.
The plurality does not address the question whether proof of discriminatory effect is necessary to support a vote-dilution claim. It is clear from the above, however, that if the Court at some point creates such a requirement, it would be satisfied by proof of mere disproportionate impact. Such a requirement would be far less stringent than the burden of proof re*141quired under the rather, rigid discriminatory-effects test I find in White v. Regester, supra. See n. 7, supra, and accompanying text.