CBOCS WEST, INC., PETITIONER v. HEDRICK G.
on writ of certiorari to the united states court ofappeals for the seventh circuit
Justice Thomas, with whom Justice Scalia joins, dissenting.
The Court holds that the private right of action it has implied under Rev. Stat. §1977, 42 U. S. C. §1981, encompasses claims of retaliation. Because the Court’s holding has no basis in the text of §1981 and is not justified by principles of stare decisis, I respectfully dissent.
It is unexceptional in our case law that “ ‘[s]tatutory construction must begin with the language employed by Congress and the assumption that the ordinary meaning of that language accurately expresses the legislative purpose.’ ” Engine Mfrs. Assn. v. South Coast Air Quality Management Dist., 541 U. S. 246, 252 (2004) (quoting Park ’N Fly, Inc. v. Dollar Park & Fly, Inc., 469 U. S. 189, 194 (1985) ). Today, that rule is honored in the breach: The Court’s analysis of the statutory text does not appear until Part III of its opinion, and then only as a potential reason to depart from the interpretation the Court has already concluded, on other grounds, must “carry the day.” Ante, at 9. Unlike the Court, I think it best to begin, as we usually do, with the text of the statute. Section 1981(a) provides:
“All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts, to sue, be parties, give evidence, and to the full and equal benefit of all laws and proceedings for the security of persons and property as is enjoyed by white citizens, and shall be subject to like punishment, pains, penalties, taxes, licenses, and exactions of every kind, and to no other.”
Section 1981(a) thus guarantees “[a]ll persons … the same right … to make and enforce contracts … as is enjoyed by white citizens.” It is difficult to see where one finds a cause of action for retaliation in this language. On its face, §1981(a) is a straightforward ban on racial discrimination in the making and enforcement of contracts. Not surprisingly, that is how the Court has always construed it. See, e.g., Domino’s Pizza, Inc. v. McDonald, 546 U. S. 470, 476 (2006) (“Section 1981 offers relief when racial discrimination blocks the creation of a contractual relationship, as well as when racial discrimination impairs an existing contractual relationship”); Patterson v. McLean Credit Union, 491 U. S. 164, 171 (1989) (“[Section] 1981 ‘prohibits racial discrimination in the making and enforcement of private contracts’ ” (quoting Runyon v. McCrary, 427 U. S. 160, 168 (1976) )); Johnson v. Railway Express Agency, Inc., 421 U. S. 454, 459 (1975) (Section 1981 “on its face relates primarily to racial discrimination in the making and enforcement of contracts”).
Respondent nonetheless contends that “[t]he terms of section 1981 are significantly different, and broader, than a simple prohibition against discrimination.” Brief for Respondent 15. It is true that §1981(a), which was enacted shortly after the Civil War, does not use the modern statutory formulation prohibiting “discrimination on the basis of race.” But that is the clear import of its terms. Contrary to respondent’s contention, nothing in §1981 evinces a “concer[n] with protecting individuals ‘based on what they do,’ ” as opposed to “ ‘prevent[ing] injury to individuals based on who they are.’ ” Ibid. (quoting Burlington N. & S. F. R. Co. v. White, 548 U. S. 53, 63 (2006) ). Nor does §1981 “affirmatively guarante[e]” freestanding “rights to engage in particular conduct.” Brief for Respondent 16. Rather, §1981 is an equal-rights provision. See Georgia v. Rachel, 384 U. S. 780, 791 (1966) (“Congress intended to protect a limited category of rights, specifically defined in terms of racial equality”). The statute assumes that “white citizens” enjoy certain rights and requires that those rights be extended equally to “[a]ll persons,” regardless of their race. That is to say, it prohibits discrimination based on race.1
Retaliation is not discrimination based on race. When an individual is subjected to reprisal because he has complained about racial discrimination, the injury he suffers is not on account of his race; rather, it is the result of his conduct. The Court recognized this commonsense distinction just two years ago in Burlington when it explained that Title VII’s antidiscrimination provision “seeks to prevent injury to individuals based on who they are, i.e., their status,” whereas its “antiretaliation provision seeks to prevent harm to individuals based on what they do, i.e., their conduct.” 548 U. S., at 63. This distinction is sound, and it reflects the fact that a claim of retaliation is both logically and factually distinct from a claim of discrimination—logically because retaliation based on conduct and discrimination based on status are mutually exclusive categories, and factually because a claim of retaliation does not depend on proof that any status-based discrimination actually occurred. Consider, for example, an employer who fires any employee who complains of race discrimination, regardless of the employee’s race. Such an employer is undoubtedly guilty of retaliation, but he has not discriminated on the basis of anyone’s race. Because the employer treats all employees—black and white—the same, he does not deny any employee “the same right … to make and enforce contracts … as is enjoyed by white citizens.”2
The Court apparently believes that the status/conduct distinction is not relevant here because this case, unlike Burlington, does not require us to determine whether §1981’s supposed prohibition on retaliation “sweep[s] more broadly” than its antidiscrimination prohibition. Ante, at 13. That is nonsense. Although, as the Court notes, we used the status/conduct distinction in Burlington to explain why Title VII’s antiretaliation provision must sweep more broadly than its antidiscrimination provision in order to achieve its purpose, 548 U. S., at 63–64, it does not follow that the distinction between status and conduct is irrelevant here. To the contrary, Burlington underscores the fact that status-based discrimination and conduct-based retaliation are distinct harms that call for tailored legislative treatment. That is why Congress, in Title VII and a host of other statutes, has enacted separate provisions prohibiting discrimination and retaliation. See Brief for Petitioner 17–18 (citing statutes); see also ante, at 10–11 (same). Construing a general ban on discrimination such as that contained in §1981 to cover retaliation would render these separate antiretaliation provisions superfluous, contrary to the normal rules of statutory interpretation.
Of course, this is not the first time I have made these points. Three Terms ago in Jackson v. Birmingham Bd. of Ed., 544 U. S. 167 (2005) , the Court held that Title IX of the Education Amendments of 1972, 20 U. S. C. §1681 et seq., which prohibits recipients of federal education funding from discriminating “on the basis of sex,” §1681(a), affords an implied cause of action for retaliation against those who complain of sex discrimination. In so doing, the Court disregarded the fundamental distinction between status-based discrimination and conduct-based retaliation, asserting that retaliation against those who complain of sex discrimination “is discrimination ‘on the basis of sex’ because it is an intentional response to the nature of the complaint: an allegation of sex discrimination.” 544 U. S., at 174. But as I explained in my dissenting opinion in Jackson, “the sex-based topic of the complaint cannot overcome the fact that the retaliation is not based on anyone’s sex, much less the complainer’s sex.” Id., at 188.
Likewise here, the race-based topic of the complaint cannot overcome the fact that the retaliation is not based on anyone’s race. To hold otherwise would be to ignore the fact that “protection from retaliation is separate from direct protection of the primary right [against discrimination] and serves as a prophylactic measure to guard the primary right.” Id., at 189; see also Burlington, supra, at 63 (explaining that Title VII’s “antidiscrimination provision seeks a workplace where individuals are not discriminated against because of their racial, ethnic, religious, or gender-based status,” whereas its “antiretaliation provision seeks to secure that primary objective by preventing an employer from interfering (through retaliation) with an employee’s efforts to secure or advance enforcement of the Act’s basic guarantees”).In other words, “[t]o describe retaliation as discrimination on the basis of [race] is to conflate the enforcement mechanism with the right itself, something for which the statute’s text provides no warrant.” Jackson, supra, at 189 (Thomas, J., dissenting).
Notably, the Court does not repeat Jackson’s textual analysis in this case, perhaps because no amount of repetition could make it any more plausible today than it was three years ago. Instead, the Court acknowledges that “the statute’s language does not expressly refer to the claim of an individual (black or white) who suffers retaliation.” Ante, at 9. The Court concludes, however, that the statute’s failure expressly to provide a cause of action for retaliation “is not sufficient to carry the day,” ibid., despite our usual rule that “affirmative evidence of congressional intent must be provided for an implied remedy … for without such intent the essential predicate for implication of a private remedy simply does not exist, ” Alexander v. Sandoval, 532 U. S. 275, 293, n. 8 (2001) (internal quotation marks and emphasis deleted); see also id., at 286–287 (emphasizing that, absent evidence of Congress’ intent to create a cause of action, the “cause of action does not exist and courts may not create one, no matter how desirable that might be as a policy matter, or how compatible with the statute”).
Section 1981’s silence regarding retaliation is not dispositive, the Court says, because “it is too late in the day” to resort to “a linguistic argument” that was supposedly rejected in Sullivan v. Little Hunting Park, Inc., 396 U. S. 229 (1969) . Ante, at 10. As I explain below, the Court’s reliance on Sullivan is entirely misplaced. But it also bears emphasis that the Court does not even purport to identify any basis in the statutory text for the “well-embedded interpretation of §1981,” ante, at 8–9, it adopts for the first time today. Unlike the Court, I find the statute’s text dispositive. Because §1981 by its terms prohibits only discrimination based on race, and because retaliation is not discrimination based on race, §1981 does not provide an implied cause of action for retaliation.
Unable to justify its holding as a matter of statutory interpretation, the Court today retreats behind the figleaf of ersatz stare decisis. The Court’s invocation of stare decisis appears to rest on three considerations: (1) Sullivan’s purported recognition of a cause of action for retaliation under §1982; (2) Jackson’s (re)interpretation of Sullivan; and (3) the Courts of Appeals’ view that §1981 provides a cause of action for retaliation. None of these considerations, separately or together, justifies implying a cause of action that Congress did not include in the statute. And none can conceal the irony in the Court’s novel use of stare decisis to decide a question of first impression.
I turn first to Sullivan, as it bears most of the weight in the Court’s analysis. As I explained in my dissent in Jackson, Sullivan did not “hol[d] that a general prohibition against discrimination permitted a claim of retaliation,” but rather “that a white lessor had standing to assert the right of a black lessee to be free from racial discrimination.” 544 U. S., at 194. Thus, “[t]o make out his third-party claim on behalf of the black lessee, the white lessor would necessarily be required to demonstrate that the defendant had discriminated against the black lessee on the basis of race.” Ibid. Here, by contrast, respondent “need not show that the [race] discrimination forming the basis of his complaints actually occurred.” Ibid. Accordingly, as it did in Jackson, the Court “creates an entirely new cause of action for a secondary rights holder, beyond the claim of the original rights holder, and well beyond Sullivan.” Id., at 194–195.
Having reexamined Sullivan, I remain convinced that it was a third-party standing case. Sullivan did not argue that his expulsion from the corporation—as opposed to the corporation’s refusal to approve the assignment—violated §1982. Instead, he argued that his expulsion was “contrary to public policy” because it was the “direct result of his having dealt with Freeman, as the statute requires, on a non-discriminatory basis.” Brief for Petitioners in Sullivan v. Little Hunting Park, Inc., O. T. 1969, No. 33, p. 32. Sullivan further contended not that his own rights under §1982 had been violated, but that he “ha[d] standing to rely on the rights of the Negro, Freeman,” since he was best situated to vindicate those rights.3 Id., at 33; see also Pet. for Cert. in Sullivan v. Little Hunting Park, Inc., O. T. 1969, No. 33, p. 17, n. 13 (“Although the statute declares the rights of Negroes not to be discriminated against, Sullivan, a Caucasian, has standing to rely on the invasion of the rights of others, since he is the only effective adversary capable of vindicating them in litigation arising from his expulsion” (internal quotation marks omitted)). Similarly, the United States, appearing as amicus curiae in support of Sullivan, argued that because “the private action involved in refusing to honor the assignment was itself illegal,” “relief should be available to all persons injured by it, or as a consequence of their efforts to resist it.” Brief for United States, O. T. 1969, No. 33, p. 34.
Thus, both Sullivan and the United States argued that Sullivan had standing to seek relief for injuries he suffered as a result of the corporation’s violation of Freeman’s rights—not that Sullivan’s own rights under §1982 were violated. And that is the best interpretation of what the Court subsequently held. Tracking the parties’ arguments, the Court first concluded that that the corporation’s “refus[al] to approve the assignment of the membership share … was clearly an interference with Freeman’s right to ‘lease’ ” under §1982. 396 U. S., at 237. Only then did it conclude—based on Barrows v. Jackson, 346 U. S. 249 (1953) , a third-party standing case in which another white litigant was permitted to “rely on the invasion of the rights of others,” id., at 255—that Sullivan “ha[d] standing to maintain this action.” Sullivan, 396 U. S., at 237. The word “retaliation” does not appear in the Court’s opinion. Nor is there any suggestion that Sullivan would have had “standing” absent the violation of Freeman’s rights.
Of course, Sullivan is not a model of clarity, and Justice Harlan, writing in dissent, was correct to criticize the “undiscriminating manner” in which the Court dealt with Sullivan’s claims. Id., at 251. Sullivan had sought relief both for the corporation’s refusal to approve the assignment and for his expulsion. Id., at 253. But in stating that Sullivan had standing to maintain “this action,” id., at 237,the Court did not specify what relief Sullivan was entitled to pursue on remand. Lamenting the Court’s “failure to provide any guidance as to the legal standards that should govern Sullivan’s right to recovery on remand,” id., at 252 (dissenting opinion), Justice Harlan provided an instructive summary of the ambiguities in the Court’s opinion:
“One can imagine a variety of standards, each based on different legal conclusions as to the ‘rights’ and ‘duties’ created by §1982, and each having very different remedial consequences. For example, does §1982 give Sullivan a right to relief only for injuries resulting from Little Hunting Park’s interference with his statutory duty to Freeman under §1982? If so, what is Sullivan’s duty to Freeman under §1982? Unless §1982 is read to impose a duty on Sullivan to protest Freeman’s exclusion, he would be entitled to reinstatement under this standard only if the Board had expelled him for the simple act of assigning his share to Freeman.
“As an alternative, Sullivan might be thought to be entitled to relief from those injuries that flowed from the Board’s violation of its ‘duty’ to Freeman under §1982. Such a standard might suggest that Sullivan is entitled to damages that resulted from Little Hunting Park’s initial refusal to accept the assignment to Freeman but again not to reinstatement. Or does the Court think that §1982 gives Sullivan a right to relief from injuries that result from his ‘legitimate’ protest aimed at convincing the Board to accept Freeman?” Id., at 254–255.
It is noteworthy that of the three possible standards Justice Harlan outlined, the first two clearly depend on a showing that Freeman’s §1982 rights were violated. Only the third—“Or does the Court think that §1982 gives Sullivan a right to relief from injuries that result from his ‘legitimate’ protest”—resembles a traditional retaliation claim and, in context, even it is probably best read to presuppose that Sullivan was protesting an actual violation of Freeman’s rights. Id., at 255. Which, if any, of these standards the Court had in mind is anybody’s guess. It did not say.
I thus adhere to my view that Sullivan is best read as a third-party standing case. That is how the parties argued the case, and that is the most natural reading of the Court’s opinion. But even if Sullivan could fairly be read as having inferred a freestanding cause of action for retaliation—which I doubt it can, at least not without superimposing an anachronistic outlook on a Court that was not as familiar with retaliation claims as we are today—the Court’s one-paragraph discussion of the issue was, at best, both cursory and ambiguous. This is hardly the stuff of which stare decisis is made.
Steadfastly refusing to acknowledge any ambiguity, the Court asserts that it is “not surprising that following Sullivan, federal appeals courts concluded, on the basis of Sullivan or its reasoning, that §1981 encompassed retaliation claims.” Ante,at 5. But given Sullivan’s use of the word “standing” and its reliance on a third-party standing case, what is unsurprising is that each of the cases the Court cites either characterized the issue as one of standing, Winston v. Lear-Siegler, Inc., 558 F. 2d 1266, 1270 (CA6 1977) (characterizing the issue as “whether or not the white plaintiff in this action has standing to sue his former employer under 42 U. S. C. §1981 for discharging him in alleged retaliation for plaintiff’s protesting the alleged discriminatory firing of a black co-worker”), or recognized that it was taking a step beyond Sullivan in inferring a cause of action for retaliation, Choudhury v. Polytechnic Inst. of N. Y., 735 F. 2d 38, 42 (CA2 1984) (stating that the Second Circuit “ha[d] never decided whether §1981 creates a cause of action for retaliation,” even though it had previously held, based on Sullivan, “that a white person who claimed to have suffered reprisals as a result of his efforts to vindicate the rights of non-whites had standing to sue under §1981”); Goff v. Continental Oil Co., 678 F. 2d 593, 598, n. 7 (CA5 1982) (recognizing that Sullivan and a previous Fifth Circuit decision relying on Sullivan were “essentially standing cases holding that white people can assert civil rights claims when they are harmed by someone’s discrimination against blacks,” which is distinct from holding that “a particular type of conduct—retaliation for the filing of a §1981 law suit—is actionable in the first place”).
Moreover, even if Sullivan had squarely and unambiguously held that §1982 provides an implied cause of action for retaliation, it would have been wrong to do so because §1982, like §1981, prohibits only discrimination based on race, and retaliation is not discrimination based on race.4 The question, then, would be whether to extend Sullivan’s erroneous interpretation of §1982 to §1981. The Court treats this as a foregone conclusion because “our precedents have long construed §§1981 and 1982 similarly.” Ante, at 4. But erroneous precedents need not be extended to their logical end, even when dealing with related provisions that normally would be interpreted in lockstep.5 Otherwise, stare decisis, designed to be a principle of stability and repose, would become a vehicle of change whereby an error in one area metastasizes into others, thereby distorting the law. Two wrongs do not make a right, and an aesthetic preference for symmetry should not prevent us from recognizing the true meaning of an Act of Congress.
The Court’s remaining reasons for invoking stare decisis require little discussion. First, the Court relies on the fact that Jackson interpreted Sullivan as having recognized a cause of action for retaliation under §1982. See ante, at 3, 9–10. That is true but irrelevant. It was only through loose language and creative use of brackets that Jackson was able to assert that Sullivan “upheld Sullivan’s cause of action under 42 U. S. C. §1982 for ‘[retaliation] for the advocacy of [the black person’s] cause.’ ” 544 U. S., at 176 (quoting Sullivan, 396 U. S., at 237; brackets in original). Of course, Sullivan did not use the word “retaliation,” did not say anything about a “cause of action,” and did not state that Sullivan had rights under §1982. It most certainly did not “interpre[t] a general prohibition on racial discrimination to cover retaliation against those who advocate the rights of groups protected by that prohibition.” Jackson,544 U. S.,at 176. Jackson’s assertion that Sullivan “plainly held that the white owner could maintain his own private cause of action under §1982,” id., at 176, n. 1, misses the point entirely. While Sullivan held that “the white owner” had standing to maintain his own suit, it said nothing to suggest that he could sue to vindicate his own right to be free from retaliation under §1982. Rather, as I have explained, Sullivan’s “standing” was derivative of the violation of Freeman’s rights. In short, Jackson’s characterization of Sullivan was erroneous, and I am aware of no principle of stare decisis that requires us to give decisive weight to a precedent’s erroneous characterization of another precedent—particularly where, as here, the cases involved different statutes, neither of which was the statute at issue in the case at bar.
Second, the Court appears to give weight to the fact that, since Congress passed the Civil Rights Act of 1991, §101, 105 Stat. 1071, “the lower courts have uniformly interpreted §1981 as encompassing retaliation actions.” Ante, at 8. This rationale fares no better than the others. The Court has never suggested that rejection of a view uniformly held by the courts of appeals violates some principle of stare decisis. To the contrary, we have not hesitated to take a different view if convinced the lower courts were wrong. Indeed, it has become something of a dissenter’s tactic to point out that the Court has decided a question differently than every court of appeals to have considered it. See, e.g., McConnell v. Federal Election Comm’n, 540 U. S. 93 , n. 11 (2003) (Thomas, J., concurring in part, concurring in result in part, concurring in judgment in part, and dissenting in part); Buckhannon Board & Care Home, Inc. v. West Virginia Dept. of Health and Human Resources, 532 U. S. 598, 643 (2001) (Ginsburg, J., dissenting); Sandoval, 532 U. S., at 294 (Stevens, J., dissenting); Jones v. United States, 526 U. S. 227, 254 (1999) (Kennedy, J., dissenting); McNally v. United States, 483 U. S. 350, 365 (1987) (Stevens, J., dissenting). The Court does not explain what makes this particular line of lower court authority any more sacrosanct than those we have rejected in the past.
Of course, lower court decisions may be persuasive, and when the Court rejects the unanimous position of the courts of appeals, it is fair to point out that fact. But the point has traction only to the extent it tends to show that the Court’s reasoning is flawed on the merits, as demonstrated by the number of judges who have reached the opposite conclusion. See, e.g., Buckhannon, supra, at 643–644 (Ginsburg, J., dissenting) (“When this Court rejects the considered judgment prevailing in the Circuits, respect for our colleagues demands a cogent explanation”). Unlike decisions of this Court, decisions of the courts of appeals, even when unanimous, do not carry stare decisis weight, nor do they relieve us of our obligation independently to decide the merits of the question presented. That is why, when we have affirmed a view unanimously held by the courts of appeals, we have done so (at least until today) not because we gave precedential weight to the lower courts’ decisions, but because we agreed with their resolution of the question on the merits. See, e.g., Gonzalez v. Crosby, 545 U. S. 524, 531 (2005) (“Virtually every Court of Appeals to consider the question has held that such a pleading … is in substance a successive habeas petition … . We think those holdings are correct”); Lampf, Pleva, Lipkind, Prupis & Petigrow v. Gilbertson, 501 U. S. 350, 362 (1991) (“Thus, we agree with every Court of Appeals that has been called upon to apply a federal statute of limitations to a §10(b) claim”).
As in Jackson, “[t]he question before us is only whether [§1981] prohibits retaliation, not whether prohibiting it is good policy.” 544 U. S., at 195 (Thomas, J., dissenting). “By crafting its own additional enforcement mechanism, the majority returns this Court to the days in which it created remedies out of whole cloth to effectuate its vision of congressional purpose.” Ibid. That the Court does so under the guise of stare decisis does not make its decision any more justifiable. Because the text of §1981 provides no basis for implying a private right of action for retaliation, and because no decision of this Court holds to the contrary, I would reverse the judgment below.
1 The United States, appearing as amicus curiae in support of respondent, contends that §1981 prohibits not only racial discrimination, but also any other kind of “discrimination” that “impair[s]” the rights guaranteed by §1981(a). Brief for United States 17. In support of this argument, the United States points to §1981(c), which provides that “[t]he rights protected by this section are protected against impairment by nongovernmental discrimination and impairment under color of State law.” Thus, the argument goes, retaliation is prohibited because it is discrimination (differential treatment for those who complain) and it impairs the right granted in §1981(a) to be free from racial discrimination in the making and enforcement of contracts (by penalizing assertion of that right). Although I commend the United States for at least attempting to ground its position in the statutory text, its argument is unconvincing. Section 1981(c) simply codifies the Court’s holding in Runyon v. McCrary, 427 U. S. 160 (1976) , that §1981 applies to private, as well as governmental, discrimination. Nothing in §1981(c) indicates that Congress otherwise intended to expand the scope of §1981. To the contrary, §1981(c) refers to “[t]he rights protected by this section,” i.e., the rights enumerated in §1981(a) to make and enforce contracts on the same terms as white citizens. Moreover, the word “discrimination” in §1981(c) does not refer to “all discrimination,” as the United States would have it. See Brief for United States as Amicus Curiae 16, n. 4. Rather, it refers back to the type of discrimination prohibited by §1981(a), i.e., discrimination based on race. Thus, §1981 is violated only when racial discrimination impairs the right to make and enforce contracts.
2 Of course, if an employer had a different retaliation policy for blacks and whites—firing black employees who complain of race discrimination but not firing similarly situated white employees—a black employee who was fired for complaining of race discrimination would have a promising §1981 claim. But his claim would not sound in retaliation; rather, it would be a straightforward claim of racial discrimination. In his briefs before this Court, respondent attempts to shoehorn his claim into this category, asserting that petitioner “retaliated against [him] because he was a black worker who exercised his right” to lodge a grievance under petitioner’s open-door policy. Brief for Respondent 27; see also id., at 33 (“[S]ection 1981 forbids an employer from having one dismissal policy for blacks who complain about race discrimination, and another for whites who complain about such discrimination”). But respondent cites no record evidence to support his assertion that petitioner treated him differently than it would have treated a similarly situated white complainant. And while the Court of Appeals found that respondent had established a prima facie case of retaliation, 474 F. 3d 387, 406–407 (CA7 2007), it did not identify any evidence that would permit a jury to conclude that the alleged retaliation was race based. Indeed, the Court of Appeals held that respondent had “waived … his discrimination claim by devoting only a skeletal argument [to it] in response to [petitioner’s] motion for summary judgment.” Id., at 407.
3 In contrast to his argument based on §1982, which he consistently tied to the violation of Freeman’s rights, Sullivan also argued that his own First Amendment rights were violated: “Since Sullivan’s expulsion was in retaliation for his having obeyed the dictate of the law the expulsion was against public policy, and he should be reinstated. For the law to sanction punishment of a person such as Sullivan for refusing to discriminate against Negroes would be to render nugatory the rights guaranteed to Negroes by 42 U. S. C. §§1981 … . Furthermore, by giving sanction to Sullivan’s expulsion, the state court deprived Sullivan of his rights, guaranteed by the First Amendment to criticize the conduct of the association’s directors.” Brief for Petitioners, O. T. 1969, No. 33, p. 14 (emphasis added).
4 The majority claims that Sullivan “did not embrace” this “linguistic argument.” Ante, at 10. That is because the argument was not before the Court. The corporation did not argue that §1982’s text could not reasonably be construed to create a cause of action for retaliation; nor did Justice Harlan in dissent. No one made this argument because that was not how the issue was framed, either by Sullivan or by the Court. The majority suggests that the argument was “apparent at the time the Court decided Sullivan.” Ibid. But the only evidence it cites is Justice Harlan’s observation that the Court’s holding in Jones v. Alfred H. Mayer Co., 392 U. S. 409 (1968) , that §1982 prohibits private as well as governmental discrimination was “in no way required by [§1982’s] language.” Sullivan, 396 U. S., at 241. I fail to see how that observation—or Justice Harlan’s further observation that the Court in Sullivan had gone “yet beyond Jones,” ibid.—shows that the Court considered and rejected the entirely different argument that §1982’s text does not provide a cause of action for retaliation.
5 For example, we have refused to extend the holding of J. I. Case Co. v. Borak, 377 U. S. 426 (1964) , which inferred a private right of action for violations of §14(a) of the Securities Exchange Act of 1934, to other sections of the Act. Borak applied the understanding—later abandoned in Cort v. Ash, 422 U. S. 66, 78 (1975) —that “it is the duty of the courts to be alert to provide such remedies as are necessary to make effective the congressional purpose” expressed by a statute. 377 U. S., at 433. As Chief Judge Easterbrook explained in dissent below, the analogy to the present case is obvious: “The argument goes that, because Sullivan ignored the language of §1982 and drafted an ‘improved’ version of the statute, we are free to do the same today for §1981, its neighbor. The Supreme Court requires us to proceed otherwise. Borak dealt with §14(a) of the Securities Exchange Act of 1934, 15 U. S. C. §78n(a). It was as freewheeling in ‘interpreting’ that law as Sullivan was with §1982. Yet the Court has held that the change of interpretive method announced in Cort applies to all other sections of the Securities Exchange Act. See Piper v. Chris-Craft Industries, Inc., 430 U. S. 1 (1977) (§14(e)); Touche Ross & Co. v. Redington, 442 U. S. 560 (1979) (§17(a)). Borak and similar decisions from the 1960s have not been overruled, but we have been told in no uncertain terms that they must not be extended. Indeed, in Virginia Bankshares, Inc. v. Sandberg, 501 U. S. 1083 (1991) , the Court declined to apply Borak to a portion of §14(a) that had not been involved in Borak. So that case has been limited to a single sentence of one subsection. Why, then, may the method of Sullivan be applied to other sections of the Civil Rights Act of 1866 despite intervening precedent?” 474 F. 3d, at 410–411 (citations omitted).