skip navigation

ALEXANDER V. SANDOVAL (99-1908) 532 U.S. 275 (2001)
197 F.3d 484, reversed.
[ Scalia ]
[ Stevens ]
HTML version
PDF version
HTML version
PDF version
HTML version
PDF version

Stevens, J., dissenting


No. 99—1908

v. MARTHA SANDOVAL, individually and on
behalf of all others similarly situated


[April 24, 2001]

    Justice Stevens, with whom Justice Souter, Justice Ginsburg, and Justice Breyer join, dissenting.

    In 1964, as part of a groundbreaking and comprehensive civil rights Act, Congress prohibited recipients of federal funds from discriminating on the basis of race, ethnicity, or national origin. Title VI of the Civil Rights Act of 1964, 78 Stat. 252, 42 U.S.C. § 2000d to 2000d—7. Pursuant to powers expressly delegated by that Act, the federal agencies and departments responsible for awarding and administering federal contracts immediately adopted regulations prohibiting federal contractees from adopting policies that have the “effect” of discriminating on those bases. At the time of the promulgation of these regulations, prevailing principles of statutory construction assumed that Congress intended a private right of action whenever such a cause of action was necessary to protect individual rights granted by valid federal law. Relying both on this presumption and on independent analysis of Title VI, this Court has repeatedly and consistently affirmed the right of private individuals to bring civil suits to enforce rights guaranteed by Title VI. A fair reading of those cases, and coherent implementation of the statutory scheme, requires the same result under Title VI’s implementing regulations.

    In separate lawsuits spanning several decades, we have endorsed an action identical in substance to the one brought in this case, see Lau v. Nichols, 414 U.S. 563 (1974); demonstrated that Congress intended a private right of action to protect the rights guaranteed by Title VI, see Cannon v. University of Chicago, 441 U.S. 677 (1979); and concluded that private individuals may seek declaratory and injunctive relief against state officials for violations of regulations promulgated pursuant to Title VI, see Guardians Assn. v. Civil Serv. Comm’n of New York City, 463 U.S. 582 (1983). Giving fair import to our language and our holdings, every Court of Appeals to address the question has concluded that a private right of action exists to enforce the rights guaranteed both by the text of Title VI and by any regulations validly promulgated pursuant to that Title, and Congress has adopted several statutes that appear to ratify the status quo.

    Today, in a decision unfounded in our precedent and hostile to decades of settled expectations, a majority of this Court carves out an important exception to the right of private action long recognized under Title VI. In so doing, the Court makes three distinct, albeit interrelated, errors. First, the Court provides a muddled account of both the reasoning and the breadth of our prior decisions endorsing a private right of action under Title VI, thereby obscuring the conflict between those opinions and today’s decision. Second, the Court offers a flawed and unconvincing analysis of the relationship between §§601 and 602 of the Civil Rights Act of 1964, ignoring more plausible and persuasive explanations detailed in our prior opinions. Finally, the Court badly misconstrues the theoretical linchpin of our decision in Cannon v. University of Chicago, 441 U.S. 677 (1979), mistaking that decision’s careful contextual analysis for judicial fiat.


    The majority is undoubtedly correct that this Court has never said in so many words that a private right of action exists to enforce the disparate-impact regulations promulgated under §602. However, the failure of our cases to state this conclusion explicitly does not absolve the Court of the responsibility to canvass our prior opinions for guidance. Reviewing these opinions with the care they deserve, I reach the same conclusion as the Courts of Appeals: This Court has already considered the question presented today and concluded that a private right of action exists.1

    When this Court faced an identical case 27 years ago, all the Justices believed that private parties could bring lawsuits under Title VI and its implementing regulations to enjoin the provision of governmental services in a manner that discriminated against non-English speakers. See Lau v. Nichols, 414 U.S. 563 (1974). While five Justices saw no need to go beyond the command of §601, Chief Justice Burger, Justice Stewart, and Justice Blackmun relied specifically and exclusively on the regulations to support the private action, see id., at 569 (Stewart, J., concurring in result) (citing Mourning v. Family Publications Service, Inc., 411 U.S. 356, 369 (1973); Thorpe v. Housing Authority of Durham, 393 U.S. 268, 280—281 (1969)). There is nothing in the majority’s opinion in Lau, or in earlier opinions of the Court, that is not fully consistent with the analysis of the concurring Justices or that would have differentiated between private actions to enforce the text of §601 and private actions to enforce the regulations promulgated pursuant to §602. See Guardians, 463 U.S., at 591 (principal opinion of White, J.) (describing this history and noting that, up to that point, no Justice had ever expressed disagreement with Justice Stewart’s analysis in Lau).2

    Five years later, we more explicitly considered whether a private right of action exists to enforce the guarantees of Title VI and its gender-based twin, Title IX. See Cannon v. University of Chicago, 441 U.S. 677 (1979). In that case, we examined the text of the statutes, analyzed the purpose of the laws, and canvassed the relevant legislative history. Our conclusion was unequivocal: “We have no doubt that Congress intended to create Title IX remedies comparable to those available under Title VI and that it understood Title VI as authorizing an implied private cause of action for victims of the prohibited discrimination.” Id., at 703.

    The majority acknowledges that Cannon is binding precedent with regard to both Title VI and Title IX, ante, at 3—4, but seeks to limit the scope of its holding to cases involving allegations of intentional discrimination. The distinction the majority attempts to impose is wholly foreign to Cannon’s text and reasoning. The opinion in Cannon consistently treats the question presented in that case as whether a private right of action exists to enforce “Title IX” (and by extension “Title VI”),3 and does not draw any distinctions between the various types of discrimination outlawed by the operation of those statutes. Though the opinion did not reach out to affirmatively preclude the drawing of every conceivable distinction, it could hardly have been more clear as to the scope of its holding: A private right of action exists for “victims of the prohibited discrimination.” 441 U.S., at 703 (emphasis added). Not some of the prohibited discrimination, but all of it.4

    Moreover, Cannon was itself a disparate-impact case. In that case, the plaintiff brought suit against two private universities challenging medical school admissions policies that set age limits for applicants. Plaintiff, a 39-year-old woman, alleged that these rules had the effect of discriminating against women because the incidence of interrupted higher education is higher among women than among men. In providing a shorthand description of her claim in the text of the opinion, we ambiguously stated that she had alleged that she was denied admission “because she is a woman,” but we appended a lengthy footnote setting forth the details of her disparate&nbhyph;impact claim. Other than the shorthand description of her claim, there is not a word in the text of the opinion even suggesting that she had made the improbable allegation that the University of Chicago and Northwestern University had intentionally discriminated against women. In the context of the entire opinion (including both its analysis and its uncontested description of the facts of the case), that single ambiguous phrase provides no basis for limiting the case’s holding to incidents of intentional discrimination. If anything, the fact that the phrase “because she is a woman” encompasses both intentional and disparate&nbhyph; impact claims should have made it clear that the reasoning in the opinion was equally applicable to both types of claims. In any event, the holding of the case certainly applied to the disparate&nbhyph;impact claim that was described in detail in footnote 1 of the opinion, id., at 680.

    Our fractured decision in Guardians Assn. v. Civil Serv. Comm’n of New York City, 463 U.S. 582 (1983), reinforces the conclusion that this issue is effectively settled. While the various opinions in that case took different views as to the spectrum of relief available to plaintiffs in Title VI cases, a clear majority of the Court expressly stated that private parties may seek injunctive relief against governmental practices that have the effect of discriminating against racial and ethnic minorities. Id., at 594—595, 607 (White, J.); id., at 634 (Marshall, J., dissenting); id., at 638 (Stevens, J., joined by Brennan and Blackmun, JJ., dissenting). As this case involves just such an action, its result ought to follow naturally from Guardians.

    As I read today’s opinion, the majority declines to accord precedential value to Guardians because the five Justices in the majority were arguably divided over the mechanism through which private parties might seek such injunctive relief.5 This argument inspires two responses. First, to the extent that the majority denies relief to the respondents merely because they neglected to mention 42 U.S.C. § 1983 in framing their Title VI claim, this case is something of a sport. Litigants who in the future wish to enforce the Title VI regulations against state actors in all likelihood must only reference §1983 to obtain relief; indeed, the plaintiffs in this case (or other similarly situated individuals) presumably retain the option of re-challenging Alabama’s English-only policy in a complaint that invokes §1983 even after today’s decision.

    More important, the majority’s reading of Guardians is strained even in reference to the broader question whether injunctive relief is available to remedy violations of the Title VI regulations by nongovernmental grantees. As Guardians involved an action against a governmental entity, making §1983 relief available, the Court might have discussed the availability of judicial relief without addressing the scope of the implied private right of action available directly under Title VI. See 463 U. S., at 638 (Stevens, J.) (“Even if it were not settled by now that Title VI authorizes appropriate relief, both prospective and retroactive, to victims of racial discrimination at the hands of recipients of federal funds, the same result would follow in this case because the petitioners have sought relief under 42 U.S.C. § 1983” (emphasis deleted)). However, the analysis in each of the relevant opinions did not do so.6 Rather than focusing on considerations specific to §1983, each of these opinions looked instead to our opinion in Cannon, to the intent of the Congress that adopted Title VI and the contemporaneous executive decisionmakers who crafted the disparate&nbhyph;impact regulations, and to general principles of remediation.7

    In summary, there is clear precedent of this Court for the proposition that the plaintiffs in this case can seek injunctive relief either through an implied right of action or through §1983. Though the holding in Guardians does not compel the conclusion that a private right of action exists to enforce the Title VI regulations against private parties, the rationales of the relevant opinions strongly imply that result. When that fact is coupled with our holding in Cannon and our unanimous decision in Lau, the answer to the question presented in this case is overdetermined.8 Even absent my continued belief that Congress intended a private right of action to enforce both Title VI and its implementing regulations, I would answer the question presented in the affirmative and affirm the decision of the Court of Appeals as a matter of stare


    Underlying the majority’s dismissive treatment of our prior cases is a flawed understanding of the structure of Title VI and, more particularly, of the relationship between §§601 and 602. To some extent, confusion as to the relationship between the provisions is understandable, as Title VI is a deceptively simple statute. Section 601 of the Act lays out its straightforward commitment: “No person in the United States shall, on the ground of race, color, or national origin, be excluded from participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance.” 42 U.S.C. § 2000d. Section 602 “authorize[s] and direct[s]” all federal departments and agencies empowered to extend federal financial assistance to issue “rules, regulations, or orders of general applicability” in order to “effectuate” §601’s antidiscrimination mandate. 42 U.S.C. § 2000d—1.10

    On the surface, the relationship between §§601 and 602 is unproblematic–§601 states a basic principle, §602 authorizes agencies to develop detailed plans for defining the contours of the principle and ensuring its enforcement. In the context of federal civil rights law, however, nothing is ever so simple. As actions to enforce §601’s antidiscrimination principle have worked their way through the courts, we have developed a body of law giving content to §601’s broadly worded commitment. E.g., United States v. Fordice, 505 U.S. 717, 732, n. 7 (1992); Guardians Assn. v. Civil Serv. Comm’n of New York City, 463 U.S. 582 (1983); Regents of Univ. of Cal. v. Bakke, 438 U.S. 265 (1978). As the majority emphasizes today, the Judiciary’s understanding of what conduct may be remedied in actions brought directly under §601 is, in certain ways, more circumscribed than the conduct prohibited by the regulations. See, e.g., ante, at 5.

    Given that seeming peculiarity, it is necessary to examine closely the relationship between §§601 and 602, in order to understand the purpose and import of the regulations at issue in this case. For the most part, however, the majority ignores this task, assuming that the judicial decisions interpreting §601 provide an authoritative interpretation of its true meaning and treating the regulations promulgated by the agencies charged with administering the statute as poor step-cousins–either parroting the
text of §601 (in the case of regulations that prohibit intentional discrimination) or forwarding an agenda untethered
to §601’s mandate (in the case of disparate&nbhyph;impact

    The majority’s statutory analysis does violence to both the text and the structure of Title VI. Section 601 does not stand in isolation, but rather as part of an integrated remedial scheme. Section 602 exists for the sole purpose of forwarding the antidiscrimination ideals laid out in §601.11 The majority’s persistent belief that the two sections somehow forward different agendas finds no support in the statute. Nor does Title VI anywhere suggest, let alone state, that for the purpose of determining their legal effect, the “rules, regulations, [and] orders of general applicability” adopted by the agencies are to be bifurcated by the judiciary into two categories based on how closely the courts believe the regulations track the text of §601.

    What makes the Court’s analysis even more troubling is that our cases have already adopted a simpler and more sensible model for understanding the relationship between the two sections. For three decades, we have treated §602 as granting the responsible agencies the power to issue broad prophylactic rules aimed at realizing the vision laid out in §601, even if the conduct captured by these rules is at times broader than that which would otherwise be prohibited.

    In Lau, our first Title VI case, the only three Justices whose understanding of §601 required them to reach the question explicitly endorsed the power of the agencies to adopt broad prophylactic rules to enforce the aims of the statute. As Justice Stewart explained, regulations promulgated pursuant to §602 may “go beyond . . . §601” as long as they are “reasonably related” to its antidiscrimination mandate. 414 U.S., at 571 (Stewart, J., joined by Burger, C. J., and Blackmun, J., concurring in result). In Guardians, at least three Members of the Court adopted a similar understanding of the statute. See 463 U.S., at 643 (Stevens, J., joined by Brennan and Blackmun, JJ., dissenting). Finally, just 16 years ago, our unanimous opinion in Alexander v. Choate, 469 U.S. 287 (1985), treated this understanding of Title VI’s structure as settled law. Writing for the Court, Justice Marshall aptly explained the interpretation of §602’s grant of regulatory power that necessarily underlies our prior caselaw: “In essence, then, we [have] held that Title VI [has] delegated to the agencies in the first instance the complex determination of what sorts of disparate impacts upon minorities constituted sufficiently significant social problems, and [are] readily enough remediable, to warrant altering the practices of the federal grantees that ha[ve] produced those impacts.” Id., at 293—

    This understanding is firmly rooted in the text of Title VI. As §602 explicitly states, the agencies are authorized to adopt regulations to “effectuate” §601’s antidiscrimination mandate. 42 U.S.C. § 2000d—1. The plain meaning of the text reveals Congress’ intent to provide the relevant agencies with sufficient authority to transform the statute’s broad aspiration into social reality. So too does a lengthy, consistent, and impassioned legislative history.12

    This legislative design reflects a reasonable–indeed inspired–model for attacking the often-intractable problem of racial and ethnic discrimination. On its own terms, the statute supports an action challenging policies of federal grantees that explicitly or unambiguously violate antidiscrimination norms (such as policies that on their face limit benefits or services to certain races). With regard to more subtle forms of discrimination (such as schemes that limit benefits or services on ostensibly race-neutral grounds but have the predictable and perhaps intended consequence of materially benefiting some races at the expense of others), the statute does not establish a static approach but instead empowers the relevant agencies to evaluate social circumstances to determine whether there is a need for stronger measures.13 Such an approach builds into the law flexibility, an ability to make nuanced assessments of complex social realities, and an admirable willingness to credit the possibility of progress.

    The “effects” regulations at issue in this case represent the considered judgment of the relevant agencies that discrimination on the basis of race, ethnicity, and national origin by federal contractees are significant social problems that might be remedied, or at least ameliorated, by the application of a broad prophylactic rule. Given the judgment underlying them, the regulations are inspired by, at the service of, and inseparably intertwined with §601’s antidiscrimination mandate. Contrary to the majority’s suggestion, they “appl[y]” §601’s prohibition on discrimination just as surely as the intentional discrimination regulations the majority concedes are privately enforceable. Ante, at 7.

    To the extent that our prior cases mischaracterize the relationship between §§601 and 602, they err on the side of underestimating, not overestimating, the connection between the two provisions. While our cases have explicitly adopted an understanding of §601’s scope that is somewhat narrower than the reach of the regulations,14 they have done so in an unorthodox and somewhat haphazard fashion.

    Our conclusion that the legislation only encompasses intentional discrimination was never the subject of thorough consideration by a Court focused on that question. In Bakke, five Members of this Court concluded that §601 only prohibits race-based affirmative action programs in situations where the Equal Protection Clause would impose a similar ban. 438 U.S., at 287 (principal opinion of Powell, J.); id., at 325, 328, 352 (Brennan, J., joined by White, Marshall, and Blackmun, JJ., concurring in judgment in part and dissenting in part).15 In Guardians, the majority of the Court held that the analysis of those five Justices in Bakke compelled as a matter of stare decisis the conclusion that §601 does not on its own terms reach disparate impact cases. 463 U.S., at 610—611 (Powell, J., concurring in judgment); id., at 612 (O’Connor, J., concurring in judgment); id., at 642 (Stevens, J., joined by Brennan and Blackmun, JJ.). However, the opinions adopting that conclusion did not engage in any independent analysis of the reach of §601. Indeed, the only writing on this subject came from two of the five Members of the Bakke “majority,” each of whom wrote separately to reject the remaining Justices’ understanding of their opinions in Bakke and to insist that §601 does in fact reach some instances of unintentional discrimination. 463 U.S., at 589—590 (White, J.); id., at 623—624 (Marshall, J., dissenting).16 The Court’s occasional rote invocation of this Guardians majority in later cases ought not obscure the fact that the question whether §601 applies to disparate&nbhyph;impact claims has never been analyzed by this Court on the merits.17

    In addition, these Title VI cases seemingly ignore the well-established principle of administrative law that is now most often described as the “Chevron doctrine.” See Chevron U.S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984). In most other contexts, when the agencies charged with administering a broadly-worded statute offer regulations interpreting that statute or giving concrete guidance as to its implementation, we treat their interpretation of the statute’s breadth as controlling unless it presents an unreasonable construction of the statutory text. See ibid. While there may be some dispute as to the boundaries of Chevron deference, see, e.g., Christensen v. Harris County, 529 U.S. 576 (2000), it is paradigmatically appropriate when Congress has clearly delegated agencies the power to issue regulations with the force of law and established formal procedures for the promulgation of such regulations.18

    If we were writing on a blank slate, we might very well conclude that Chevron and similar cases decided both before and after Guardians provide the proper framework for understanding the structure of Title VI. Under such a reading there would be no incongruity between §§601 and 602. Instead, we would read §602 as granting the federal agencies responsible for distributing federal funds the authority to issue regulations interpreting §601 on the assumption that their construction will–if reasonable–be incorporated into our understanding of §601’s meaning.19        To resolve this case, however, it is unnecessary to answer the question whether our cases interpreting the reach of §601 should be reinterpreted in light of Chevron. If one understands the relationship between §§601 and 602 through the prism of either Chevron or our prior Title VI cases, the question presented all but answers itself. If the regulations promulgated pursuant to §602 are either an authoritative construction of §601’s meaning or prophylactic rules necessary to actualize the goals enunciated in §601, then it makes no sense to differentiate between private actions to enforce §601 and private actions to enforce §602. There is but one private action to enforce Title VI, and we already know that such an action exists.20 See Cannon, 441 U.S., at 703.


    The majority couples its flawed analysis of the structure of Title VI with an uncharitable understanding of the substance of the divide between those on this Court who are reluctant to interpret statutes to allow for private rights of action and those who are willing to do so if the claim of right survives a rigorous application of the criteria set forth in Cort v. Ash, 422 U.S. 66 (1975). As the majority narrates our implied right of action jurisprudence, ante, at 10—11, the Court’s shift to a more skeptical approach represents the rejection of a common-law judicial activism in favor of a principled recognition of the limited role of a contemporary “federal tribunal.” Ante, at 10. According to its analysis, the recognition of an implied right of action when the text and structure of the statute do not absolutely compel such a conclusion is an act of judicial self-indulgence. As much as we would like to help those disadvantaged by discrimination, we must resist the temptation to pour ourselves “one last drink.” Ante, at 11. To do otherwise would be to “ventur[e] beyond Congress’s intent.” Ibid.

    Overwrought imagery aside, it is the majority’s approach that blinds itself to congressional intent. While it remains true that, if Congress intends a private right of action to support statutory rights, “the far better course is for it to specify as much when it creates those rights,” Cannon, 441 U.S., at 717, its failure to do so does not absolve us of the responsibility to endeavor to discern its intent. In a series of cases since Cort v. Ash, we have laid out rules and developed strategies for this task.

    The very existence of these rules and strategies assumes that we will sometimes find manifestations of an implicit intent to create such a right. Our decision in Cannon represents one such occasion. As the Cannon opinion iterated and reiterated, the question whether the plaintiff had a right of action that could be asserted in federal court was a “question of statutory construction,” 441 U. S, at 688, see also id., at 717 (Rehnquist, J., concurring), not a question of policy for the Court to decide. Applying the Cort v. Ash factors, we examined the nature of the rights at issue, the text and structure of the statute, and the relevant legislative history.21 Our conclusion was that Congress unmistakably intended a private right of action to enforce both Title IX and Title VI. Our reasoning–and, as I have demonstrated, our holding–was equally applicable to intentional discrimination and disparate impact claims.22

    Underlying today’s opinion is the conviction that Cannon must be cabined because it exemplifies an “expansive rights-creating approach.” Franklin v. Gwinnett County Public Schools, 503 U.S. 60, 77 (1992) (Scalia, J. concurring in judgment). But, as I have taken pains to explain, it was Congress, not the Court, that created the cause of action, and it was the Congress that later ratified the Cannon holding in 1986 and again in 1988. See 503 U.S., at 72—73.

    In order to impose its own preferences as to the availability of judicial remedies, the Court today adopts a methodology that blinds itself to important evidence of congressional intent. It is one thing for the Court to ignore the import of our holding in Cannon, as the breadth of that precedent is a matter upon which reasonable jurists may differ. It is entirely another thing for the majority to ignore the reasoning of that opinion and the evidence contained therein, as those arguments and that evidence speak directly to the question at issue today. As I stated above, see n. 21, supra, Cannon carefully explained that both Title VI and Title IX were intended to benefit a particular class of individuals, that the purposes of the statutes would be furthered rather than frustrated by the implication of a private right of action, and that the legislative histories of the statutes support the conclusion that Congress intended such a right. See also Part IV, infra. Those conclusions and the evidence supporting them continue to have force today.

    Similarly, if the majority is genuinely committed to deciphering congressional intent, its unwillingness to even consider evidence as to the context in which Congress legislated is perplexing. Congress does not legislate in a vacuum. As the respondent and the Government suggest, and as we have held several times, the objective manifestations of congressional intent to create a private right of action must be measured in light of the enacting Congress’ expectations as to how the judiciary might evaluate the question. See Thompson v. Thompson, 484 U.S. 174 (1988); Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U.S. 353, 378—379 (1982); Cannon, 441 U.S., at 698—699.23

    At the time Congress was considering Title VI, it was normal practice for the courts to infer that Congress intended a private right of action whenever it passed a statute designed to protect a particular class that did not contain enforcement mechanisms which would be thwarted by a private remedy. See Merrill Lynch, 456 U.S., at 374—375 (discussing this history). Indeed, the very year Congress adopted Title VI, this Court specifically stated that “it is the duty of the courts to be alert to provide such remedies as are necessary to make effective the congressional purpose.” J. I. Case Co. v. Borak, 377 U.S. 426, 433 (1964). Assuming, as we must, that Congress was fully informed as to the state of the law, the contemporary context presents important evidence as
to Congress’ intent–evidence the majority declines to consider.

    Ultimately, respect for Congress’ prerogatives is measured in deeds, not words. Today, the Court coins a new rule, holding that a private cause of action to enforce a statute does not encompass a substantive regulation issued to effectuate that statute unless the regulation does nothing more than “authoritatively construe the statute itself.” Ante, at 7.24 This rule might be proper if we were the kind of “common-law court” the majority decries, ante, at 10, inventing private rights of action never intended by Congress. For if we are not construing a statute, we certainly may refuse to create a remedy for violations of federal regulations. But if we are faithful to the commitment to discerning congressional intent that all Members of this Court profess, the distinction is untenable. There is simply no reason to assume that Congress contemplated, desired, or adopted a distinction between regulations that merely parrot statutory text and broader regulations that are authorized by statutory text.25


    Beyond its flawed structural analysis of Title VI and an evident antipathy toward implied rights of action, the majority offers little affirmative support for its conclusion that Congress did not intend to create a private remedy for violations of the Title VI regulations.26 The Court offers essentially two reasons for its position. First, it attaches significance to the fact that the “rights-creating” language in §601 that defines the classes protected by the statute is not repeated in §602. Ante, at 13—14. But, of course, there was no reason to put that language in §602 because it is perfectly obvious that the regulations authorized by §602 must be designed to protect precisely the same people protected by §601. Moreover, it is self-evident that, linguistic niceties notwithstanding, any statutory provision whose stated purpose is to “effectuate” the eradication of racial and ethnic discrimination has as its “focus” those individuals who, absent such legislation, would be subject to discrimination.

    Second, the Court repeats the argument advanced and rejected in Cannon that the express provision of a fund cut-off remedy “suggests that Congress intended to preclude others.” Ante, at 14. In Cannon, 441 U.S., at 704—708, we carefully explained why the presence of an explicit mechanism to achieve one of the statute’s objectives (ensuring that federal funds are not used “to support discriminatory practices”) does not preclude a conclusion that a private right of action was intended to achieve the statute’s other principal objective (“to provide individual citizens effective protection against those practices”). In support of our analysis, we offered policy arguments, cited evidence from the legislative history, and noted the active support of the relevant agencies. Ibid. In today’s decision, the Court does not grapple with–indeed, barely acknowledges–our rejection of this argument in Cannon.

    Like much else in its opinion, the present majority’s unwillingness to explain its refusal to find the reasoning in Cannon persuasive suggests that today’s decision is the unconscious product of the majority’s profound distaste for implied causes of action rather than an attempt to discern the intent of the Congress that enacted Title VI of the Civil Rights Act of 1964. Its colorful disclaimer of any interest in “venturing beyond Congress’s intent,” ante, at 11, has a hollow ring.


    The question the Court answers today was only an open question in the most technical sense. Given the prevailing consensus in the Courts of Appeals, the Court should have declined to take this case. Having granted certiorari, the Court should have answered the question differently by simply according respect to our prior decisions. But most importantly, even if it were to ignore all of our post-1964 writing, the Court should have answered the question differently on the merits.

    I respectfully dissent.


1.  Just about every Court of Appeals has either explicitly or implicitly held that a private right of action exists to enforce all of the regulations issued pursuant to Title VI, including the disparate-impact regulations. For decisions holding so most explicitly, see, e.g. Powell v. Ridge, 189 F.3d 387, 400 (CA3 1999); Chester Residents Concerned for Quality Living v. Seif, 132 F.3d 925, 936—937 (CA3 1997), summarily dism’d, 524 U.S. 974 (1998); David K. v. Lane, 839 F.2d 1265, 1274 (CA7 1988); Sandoval v. Hogan, 197 F.3d 484 (CA11 1999) (case below). See also Latinos Unidos De Chelsea v. Secretary of Housing and Urban Development, 799 F.2d 774, 785, n. 20 (CA1 1986); New York Urban League, Inc. v. New York, 71 F.3d 1031, 1036 (CA2 1995); Ferguson v. Charleston, 186 F.3d 469 (CA4 1999), rev’d on other grounds, 532 U.S. __ (2001); Castaneda v. Pickard, 781 F.2d 456, 465, n. 11 (CA5 1986); Buchanan v. Bolivar, 99 F.3d 1352, 1356, n. 5 (CA6 1996); Larry P.. v. Riles, 793 F.2d 969, 981—982 (CA9 1986); Villanueva v. Carere, 85 F.3d 481, 486 (CA10 1996). No Court of Appeals has ever reached a contrary conclusion. But cf. New York City Environmental Justice Alliance v. Giuliani, 214 F.3d 65, 72 (CA2 2000) (suggesting that the question may be open).

2.  Indeed, it would have been remarkable if the majority had offered any disagreement with the concurring analysis as the concurring Justices grounded their argument in well&nbhyph;established principles for determining the availability of remedies under regulations, principles that all but one Member of the Court had endorsed the previous Term. See Mourning v. Family Publications Service, Inc., 411 U.S. 356, 369 (1973); id., at 378 (Douglas, J., joined by Stewart and Rehnquist, JJ., concurring in part and dissenting in part) (agreeing with the majority’s analysis of the regulation in question); but see id., at 383, n. 1 (Powell, J., dissenting) (reserving analysis of the regulation’s validity). The other decision the concurring Justices cited for this well-established principle was unanimous and only five years old. See Thorpe v. Housing Authority of Durham, 393 U.S. 268 (1969).

3.  See Cannon, 441 U.S., at 687, 699, 702, n. 33, 703, 706, n. 40, 709.

4.  The majority is undoubtedly correct that Cannon was not a case about the substance of Title IX but rather about the remedies available under that statute. Therefore, Cannon can not stand as a precedent for the proposition either that Title IX and its implementing regulations reach intentional discrimination or that they do not do so. What Cannon did hold is that all the discrimination prohibited by the regulatory scheme contained in Title IX may be the subject of a private lawsuit. As the Court today concedes that Cannon’s holding applies to Title VI claims as well as Title IX claims, ante, at 3—4, and assumes that the regulations promulgated pursuant to §602 are validly promulgated antidiscrimination measures, ante, at 5, it is clear that today’s opinion is in substantial tension with Cannon’s reasoning and holding.

5.  None of the relevant opinions was absolutely clear as to whether it envisioned such suits as being brought directly under the statute or under 42 U.S.C. § 1983. However, a close reading of the opinions leaves little doubt that all of the Justices making up the Guardians majority contemplated the availability of private actions brought directly under the statute. Justice White fairly explicitly rested his conclusion on Cannon’s holding that an implied right of action exists to enforce the terms of both Title VI and Title IX. Guardians, 463 U.S., at 594—595. Given that fact and the added consideration that his opinion appears to have equally contemplated suits against private and public parties, it is clear that he envisioned the availability of injunctive relief directly under the statute. Justice Marshall’s opinion never mentions §1983 and refers simply to “Title VI actions.” Id., at 625. In addition, his opinion can only be read as contemplating suits on equal terms against both public and private grantees, thus also suggesting that he assumed such suits could be brought directly under the statute. That leaves my opinion. Like Justice White, I made it quite clear that I believed the right to sue to enforce the disparate&nbhyph;impact regulations followed directly from Cannon and, hence, was built directly into the statute. 463 U.S., at 635—636, and n. 1. However, I did also note that, in the alternative, relief would be available in that particular case under §1983.

6.  The Court today cites one sentence in my final footnote in Guardians that it suggests is to the contrary. Ante, at 7 (citing 463 U.S., at 645, n. 18). However, the Court misreads that sentence. In his opinion in Guardians, Justice Powell had stated that he would affirm the judgment for the reasons stated in his dissent in Cannon, see 463 U.S., at 609—610 (opinion concurring in judgment), and that he would also hold that private actions asserting violations of Title VI could not be brought under §1983, id., at 610, and n. 3. One reason that he advanced in support of these conclusions was his view that the standard of proof in a §1983 action against public officials would differ from the standard in an action against private defendants. Id., at 608, n. 1. In a footnote at the end of my opinion, id., at 645, n. 18, I responded (perhaps inartfully) to Justice Powell. I noted that the fact that §1983 authorizes a lawsuit against the police department based on its violation of the governing administrative regulations did not mean, as Justice Powell had suggested, “that a similar action would be unavailable against a similarly situated private party.” Ibid. I added the sentence that the Court quotes today, ante at 7, not to reserve a question, but rather to explain that the record did not support Justice Powell’s hypothesis regarding the standard of proof. I thought then, as I do now, that a violation of regulations adopted pursuant to Title VI may be established by proof of discriminatory impact in a §1983 action against state actors and also in an implied action against private parties. See n. 5, supra. Contrary to the Court’s partial quotation of my opinion, see ante, at 7, n. 3, what I wrote amply reflected what I thought. See 463 U.S., at 635 (“a private action against recipients of federal funds”), id., at 636 (“implied caus[e] of action”); id., at 638 (“Title VI authorizes appropriate relief”). Justice Powell was quite correct in noting that it would be anomalous to assume that Congress would have intended to make it easier to recover from public officials than from private parties. That anomaly, however, does not seem to trouble the majority today.

7.  See n. 5, supra.

8.  See also Bazemore v. Friday, 478 U.S. 385 (1986) (per curiam) (adjudicating on the merits a claim brought under Title VI regulations).

9.  The settled expectations the Court undercuts today derive not only from judicial decisions, but also from the consistent statements and actions of Congress. Congress’ actions over the last two decades reflect a clear understanding of the existence of a private right action to enforce Title VI and its implementing regulations. In addition to numerous other small-scale amendments, Congress has twice adopted legislation expanding the reach of Title VI. See Civil Rights Restoration Act of 1987, §6, 102 Stat. 31 (codified at 42 U.S.C. § 2000d—4a) (expanding definition of “program”); Rehabilitation Act Amendments of 1986, §1003, 100 Stat. 1845 (codified at 42 U.S.C. § 2000d—7) (explicitly abrogating States’ Eleventh Amendment immunity in suits under Title VI). Both of these bills were adopted after this Court’s decision in Lau, Cannon, and Guardians, and after most of the Courts of Appeals had affirmatively acknowledged an implied private right of action to enforce the disparate impact regulations. Their legislative histories explicitly reflect the fact that both proponents and opponents of the bills assumed that the full breadth of Title VI (including the disparate impact regulations promulgated pursuant to it) would be enforceable in private actions. See, e.g., Civil Rights Act of 1984: Hearings on S. 2658 before the Subcommittee. on the Constitution of the Senate Committee on the Judiciary, 98th Cong., 2d Sess., 530 (1984) (memo from the Office of Management and Budget objecting to the Civil Rights Restoration Act of 1987 because it would bring more entities within the scope of Title VI thereby subjecting them to “private lawsuits” to enforce the disparate impact regulations); id. at 532 (same memo warning of a proliferation of “discriminatory effects” suits by “members of the bar” acting as “private Attorneys General”); 134 Cong. Rec. 4257 (1988) (statement of Sen. Hatch) (arguing that the disparate impact regulations go too far and noting that that is a particular problem because “[o]f course, advocacy groups will be able to bring private lawsuits making the same allegations before federal judges”); see also Brief for United States 24, n. 16 (collecting testimony of academics advising Congress that private lawsuits were available to enforce the disparate impact regulations under existing precedent). Thus, this case goes well beyond the normal situation in which “after a comprehensive reeaxmination and significant amendment” Congress “left intact the statutory provisions under which the federal courts had implied a private cause of action.” Merrill Lynch, Pierce, Fenner & Smith, Inc. v. Curran, 456 U.S. 353, 381—382 (1982). Here, there is no need to rest on presumptions of knowledge and ratification, because the direct evidence of Congress’ understanding is plentiful.

10.  The remainder of Title VI provides for judicial and administrative review of agency actions taken pursuant to the statute, §2000d—2; imposes certain limitations not at issue in this case, §§2000d—3 to 2000d—4; and defines some of the terms found in the other provisions of the statute, §200d—4a.

11.  See 42 U.S.C. § 2000d—1 (§602) (“Each Federal department and agency which is empowered to extend Federal financial assistance is authorized and directed to effectuate the provisions of [§601] by issuing rules, regulations, or orders of general applicability”).

12. See, e.g., 110 Cong. Rec. 6543 (1964) (statement of Sen. Humphrey) (“Simple justice requires that public funds, to which all taxpayers of all races contribute, not be spent in any fashion which encourages, entrenches, subsidizes, or results in racial discrimination”); id., at 1520 (statement of Rep. Celler) (describing §602 as requiring federal agencies to “reexamine” their programs “to make sure that adequate action has been taken to preclude … discrimination”) .

13.  It is important, in this context, to note that regulations prohibiting policies that have a disparate impact are not necessarily aimed only–or even primarily–at unintentional discrimination. Many policies whose very intent is to discriminate are framed in a race-neutral

manner. It is often difficult to obtain direct evidence of this motivating animus. Therefore, an agency decision to adopt disparate&nbhyph;impact regulations may very well reflect a determination by that agency that substantial intentional discrimination pervades the industry it is charged with regulating but that such discrimination is difficult to prove directly. As I have stated before: “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.” Washington v. Davis, 426 U.S. 229, 253 (1976) (concurring opinion). On this reading, Title VI simply accords the agencies the power to decide whether or not to credit such evidence.

14.  See, e.g., Alexander v. Choate, 469 U.S. 287, 293 (1985) (stating, in dicta, “Title VI itself directly reach[es] only instances of intentional discrimination”); Guardians Assn. v. Civil Serv. Comm’n of New York City, 463 U.S. 582 (1983) (in separate opinions, seven Justices indicate that §601 on its face bars only intentional discrimination).

15. Of course, those five Justices divided over the application of the Equal Protection Clause–and by extension Title VI–to affirmative action cases. Therefore, it is somewhat strange to treat the opinions of those five Justices in Bakke as constituting a majority for any particular substantive interpretation of Title VI.

16.  The fact that Justices Marshall and White both felt that the opinion they coauthored in Bakke did not resolve the question whether Title VI on its face reaches disparate-impact claims belies the majority’s assertion that Bakke “had drawn precisely that distinction,” ante, at 6, n. 2, much less its implication that it would have been “absurd” to think otherwise, ibid.

17.  In this context, it is worth noting that in a variety of other settings the Court has interpreted similarly ambiguous civil rights provisions to prohibit some policies based on their disparate impact on a protected group. See, e.g., Griggs v. Duke Power Co., 401 U.S. 424, 432 (1971) (Title VII); City of Rome v. United States, 446 U.S. 156, 172—173 (1980) (§5 of the Voting Rights Act); cf. Alexander v. Choate, 469 U.S., at 292—296 (explaining why the Rehabilitation Act of 1973, which was modeled after §601, might be considered to reach some instances of disparate impact and then assuming that it does for purposes of deciding the case).

18.  In relying on the Chevron doctrine, I do not mean to suggest that our decision in Chevron stated a new rule that requires the wholesale reconsideration of our statutory interpretation precedents. Instead, I continue to adhere to my position in Sullivan v. Everhart, 494 U.S. 83, 103—104, n. 6 (1990) (stating that Chevron merely summarized “well-settled principles”). In suggesting that, with regard to Title VI, we might reconsider whether our prior decisions gave sufficient deference to the agencies’ interpretation of the statute, I do no more than question whether in this particular instance we paid sufficient consideration to those “well-settled principles.”

19.  The legislative history strongly indicates that the Congress that adopted Title VI and the administration that proposed the statute intended that the agencies and departments would utilize the authority granted under §602 to shape the substantive contours of §601. For example, during the hearings that preceded the passage of the statute, Attorney General Kennedy agreed that the administrators of the various agencies would have the power to define “what constitutes discrimination” under Title VI and “what acts or omissions are to be forbidden.” Civil Rights–The Presidents Program, 1963: Hearings before the Senate Committee on the Judiciary, 88th Cong., 1st Sess., 399—400 (1963); see also Civil Rights: Hearings before the House Committee on the Judiciary, 88th Cong., 1st Sess., pt. 4, p. 2740 (1963) (remarks of Attorney General Kennedy) (only after the agencies “establish the rules” will recipients “understand what they can and cannot do”). It was, in fact, concern for this broad delegation that inspired Congress to amend the pending bill to ensure that all regulations issued pursuant to Title VI would have to be approved by the President. See 42 U.S.C. § 2000d—1 (laying out the requirement); 110 Cong. Rec. 2499 (1964) (remarks of Rep. Lindsay introducing the amendment). For further discussion of this legislative history, see Guardians, 463 U.S., at 615—624 (Marshall, J., dissenting); Abernathy, Title VI and the Constitution: A Regulatory Model for Defining “Discrimination,” 70 Geo. L. J. 1 (1981).

20.  The majority twice suggests that I “be[g] the question” whether a private right of action to enforce Title VI necessarily encompasses a right of action to enforce the regulations validly promulgated pursuant to the statute. Ante, at 6, n.  2, 17, n. 8. As the above analysis demonstrates, I do no such thing. On the contrary, I demonstrate that the disparate-impact regulations promulgated pursuant to §602 are–and have always been considered to be–an important part of an integrated remedial scheme intended to promote the statute’s antidiscrimination goals. Given that fact, there is simply no logical or legal justification for differentiating between actions to enforce the regulations and actions to enforce the statutory text. Furthermore, as my integrated approach reflects the longstanding practice of this Court, see n.  2, supra, it is the majority’s largely unexplained assumption that a private right of action to enforce the disparate-impact regulations must be independently established that “begs the question.”

21.  The text of the statute contained “an unmistakable focus on the benefited class,” 441 U.S., at 691; its legislative history “rather plainly indicates that Congress intended to create such a remedy,” id., at 694; the legislators’ repeated references to private enforcement of Title VI reflected “their intent with respect to Title IX,” id., at 696—698; and the absence of legislative action to change the prevailing view with respect to Title VI left us with “no doubt that Congress intended to create Title IX remedies comparable to those available under Title VI and that it understood Title VI as authorizing an implied private cause of action for victims of prohibited discrimination,” id., at 703.

22.  We should not overlook the fact that Cannon was decided after the Bakke majority had concluded that the coverage of Title VI was co-extensive with the coverage of the Equal Protection Clause.

23.  Like any other type of evidence, contextual evidence may be trumped by other more persuasive evidence. Thus, the fact that, when evaluating older statutes, we have at times reached the conclusion that Congress did not imply a private right of action does not have the significance the majority suggests. Ante, at 13—14.

24.  Only one of this Court’s myriad private right of action cases even hints at such a rule. See Central Bank of Denver, N. A. v. First Interstate Bank of Denver, N. A., 511 U.S. 164, 173 (1994). Even that decision, however, does not fully support the majority’s position for two important reasons. First, it is not at all clear that the majority opinion in that case simply held that the regulation in question could not be enforced by private action; the opinion also permits the reading, assumed by the dissent, that the majority was in effect invalidating the regulation in question. Id., at 200 (Stevens, J., dissenting) (“The majority leaves little doubt that the Exchange Act does not even permit the SEC to pursue aiders and abettors in civil enforcement actions under §10(b) and Rule 10b—5”). Second, that case involved a right of action that the Court has forthrightly acknowledged was judicially created in exactly the way the majority now condemns. See, e.g., Blue Chip Stamps v. Manor Drug Stores, 421 U.S. 723, 737 (1975) (describing private actions under Rule 10b-5 as “a judicial oak which has grown from little more than a legislative acorn”). As the action in question was in effect a common-law right, the Court was more within its rights to limit that remedy than it would be in a case, such as this one, where we have held that Congress clearly intended such a right.

25.  See Guardians, 463 U.S., at 636 (Stevens, J., dissenting) (“It is one thing to conclude, as the Court did in Cannon, that the 1964 Congress, legislating when implied causes of action were the rule rather than the exception, reasonably assumed that the intended beneficiaries of Title VI would be able to vindicate their rights in court. It is quite another thing to believe that the 1964 Congress substantially qualified that assumption but thought it unnecessary to tell the Judiciary about the qualification”).

26.  The majority suggests that its failure to offer such support is irrelevant, because the burden is on the party seeking to establish the existence of an implied right of action. Ante, at 17, n. 8. That response confuses apples and oranges. Undoubtedly, anyone seeking to bring a lawsuit has the burden of establishing that private individuals have the right to bring such a suit. However, once the courts have examined the statutory scheme under which the individual seeks to bring a suit and determined that a private right of action does exist, judges who seek to impose heretofore unrecognized limits on that right have a responsibility to offer reasoned arguments drawn from the text, structure, or history of that statute in order to justify such limitations. Moreover, in this case, the petitioners have marshaled substantial affirmative evidence that a private right of action exists to enforce Title VI and the regulations validly promulgated thereunder. See supra, at 21—22. It strikes me that it aids rather than hinders their case that this evidence is already summarized in an opinion of this Court. See Cannon, 441 U.S., at 691—703.