dissenting.
The Court reaches the remarkable conclusion that criminal defendants being prosecuted by the State act on behalf of their adversary when they exercise peremptory challenges during jury selection. The Court purports merely to follow *63precedents, but our eases do not compel this perverse result. To the contrary, our decisions specifically establish that criminal defendants and their lawyers are not government actors when they perform traditional trial functions.
I
It is well and properly settled that the Constitution’s equal protection guarantee forbids prosecutors to exercise peremptory challenges in a racially discriminatory fashion. See Batson v. Kentucky, 476 U. S. 79 (1986); Powers v. Ohio, 499 U. S. 400, 409 (1991). The Constitution, however, affords no similar protection against private action. “Embedded in our Fourteenth Amendment jurisprudence is a dichotomy between state action, which is subject to scrutiny under the Amendmen[t] . . . , and private conduct, against which the Amendment affords no shield, no matter how unfair that conduct may be.” National Collegiate Athletic Assn. v. Tarkanian, 488 U. S. 179, 191 (1988) (footnote omitted). This distinction appears on the face of the Fourteenth Amendment, which provides that “No State shall. . . deny to any person within its jurisdiction the equal protection of the laws.” U. S. Const., Arndt. 14, § 1 (emphasis added). The critical but straightforward question this case presents is whether criminal defendants and their lawyers, when exercising peremptory challenges as part of a defense, are state actors.
In Lugar v. Edmondson Oil Co., 457 U. S. 922 (1982), the Court developed a two-step approach to identifying state action in cases such as this. First, the Court will ask “whether the claimed deprivation has resulted from the exercise of a right or privilege having its source in state authority.” Id., at 939. Next, it will decide whether, on the particular facts at issue, the parties who allegedly caused the deprivation of a federal right can “appropriately” and “in all fairness” be characterized as state actors. Ibid.; Edmonson v. Leesville Concrete Co., 500 U. S. 614, 620 (1991). The *64Court’s determination in this case that the peremptory challenge is a creation of state authority, ante, at 51, breáis no new ground. See Edmonson, supra, at 620-621. But disposing of this threshold matter leaves the Court with the task of showing that criminal defendants who exercise pe-remptories should be deemed governmental actors. What our cases require, and what the Court neglects, is a realistic appraisal of the relationship between defendants and the government that has brought them to trial.
We discussed that relationship in Polk County v. Dodson, 454 U. S. 312 (1981), which held that a public defender does not act "under color of state law” for purposes of 42 U. S. C. § 1983 “when performing a lawyer’s traditional functions as counsel to a defendant in a criminal proceeding.” 454 U. S., at 325. We began our analysis by explaining that a public . defender’s obligations toward her client are no different than the obligations of any other defense attorney. Id., at 318. These obligations preclude attributing the acts of defense lawyers to the State: “[T]he duties of a defense lawyer are those of a personal counselor and advocate. It is often said that lawyers are ‘officers of the court.’ But the Courts of Appeals are agreed that a lawyer representing a client is not, by virtue of being an officer of the court, a state actor ....” Ibid.
We went on to stress the inconsistency between our adversarial system of justice and theories that would make defense lawyers state actors. “In our system,” we said, “a defense lawyer characteristically opposes the designated representatives of the State.” Ibid. This adversarial posture rests on the assumption that a defense lawyer best serves the public “not by acting on behalf of the State or in concert with it, but rather by advancing ‘the undivided interests of his client.’ ” Id., at 318-319 (quoting Ferri v. Ackerman, 444 U. S. 193, 204 (1979)). Moreover, we pointed out that the independence of defense attorneys from state control has a constitutional dimension. Gideon v. Wainwright, *65372 U. S. 335 (1963), “established the right of state criminal defendants to the guiding hand of counsel at every step in the proceedings against [them].” 454 U. S., at 322 (internal quotation marks omitted). Implicit in this right “is the assumption that counsel will be free of state control. There can be no fair trial unless the accused receives the services of an effective and independent advocate.” Ibid. Thus, the defense’s freedom from state authority is not just empirically true, but is a constitutionally mandated attribute of our adversarial system.
Because this Court deems the “under color of state law” requirement that was not satisfied in Dodson identical to the Fourteenth Amendment’s state action requirement, see Lugar, supra, at 929, the holding of Dodson simply cannot be squared with today’s decision. In particular, Dodson cannot be explained away as a case concerned exclusively with the employment status of public defenders. See ante, at 54. The Dodson Court reasoned that public defenders performing traditional defense functions are not state actors because they occupy the same position as other defense attorneys in relevant respects. 454 U. S., at 319-325. This reasoning followed on the heels of a critical determination: Defending an accused “is essentially a private function,” not state action. Id., at 319. The Court’s refusal to acknowledge Dodson’s initial holding, on which the entire opinion turned, will not make that holding go away.
The Court also seeks to evade Dodson’s logic by spinning out a theory that defendants and their lawyers transmogrify from government adversaries into state actors when they exercise a peremptory challenge, and then change back to perform other defense functions. See ante, at 54. Dodson, however, established that even though public defenders might act under color of state law when carrying out administrative or investigative functions outside a courtroom, they are not vested with state authority “when performing a lawyer’s traditional functions as counsel to a defendant in a *66criminal proceeding.” 454 U. S., at 325. Since making peremptory challenges plainly qualifies as a “traditional function” of criminal defense lawyers, see Swain v. Alabama, 380 U. S. 202, 212-219 (1965); Lewis v. United States, 146 U. S. 370, 376 (1892), Dodson forecloses the Court’s functional analysis.
Even aside from our prior rejection of it, the Court’s functional theory fails. “[A] State normally can be held responsible for a private decision only when it has exercised coercive power or has provided such significant encouragement ... that the choice must in law be deemed to be that of the State.” Blum v. Yaretsky, 457 U. S. 991, 1004 (1982). Thus, a private party’s exercise of choice allowed by state law does not amount to state action for purposes of the Fourteenth Amendment so long as “the initiative comes from [the private party] and not from the State.” Jackson v. Metropolitan Edison Co., 419 U. S. 345, 357 (1974). See Flagg Bros., Inc. v. Brooks, 436 U. S. 149, 165 (1978) (State not responsible for a decision it “permits but does not compel”). The government in no way influences the defense’s decision to use a peremptory challenge to strike a particular juror. Our adversarial system of criminal justice and the traditions of the peremptory challenge vest the decision to strike a juror entirely with the accused. A defendant "may, if he chooses, peremptorily challenge ‘on his own dislike, without showing any cause;’ he may exercise that right without reason or for no reason, arbitrarily and capriciously.” Pointer v. United States, 151 U. S. 396, 408 (1894) (quoting 1 E. Coke, Institutes 156b (19th ed. 1832)). “The essential nature of the peremptory challenge is that it is one exercised without a reason stated, without inquiry and without being subject to the court’s control.” Swain, supra, at 220. See Dodson, supra, at 321-322; Lewis, supra, at 376, 378.
Certainly, Edmonson v. Leesville Concrete Co. did not render Dodson and its realistic approach to the state action inquiry dead letters. The Edmonson Court distinguished *67Dodson by saying: “In the ordinary context of civil litigation in which the government is not a party, an adversarial relation does not exist between the government and a private litigant. In the jury selection process, the government and private litigants work for the same end.” Edmonson, 500 U. S., at 627. While the nonpartisan administrative interests of the State and the partisan interests of private litigants may not be at odds during civil jury selection, the same cannot be said of the partisan interests of the State and the defendant during jury selection in a criminal trial. A private civil litigant opposes a private counterpart, but a criminal defendant is by design in an adversarial relationship with the government. Simply put, the defendant seeks to strike jurors predisposed to convict, while the State seeks to strike .jurors predisposed to acquit. The Edmonson Court clearly recognized this point when it limited the statement that “an adversarial relation does not exist between the government and a private litigant” to “the ordinary context of civil litigation in which the government is not a ‘party” Ibid, (emphasis added).
From arrest, to trial, to possible sentencing and punishment, the antagonistic relationship between government and the accused is clear for all to see. Rather than squarely facing this fact, the Court, as in Edmonson, rests its finding of governmental action on the points that defendants exercise peremptory challenges in a courtroom and judges alter the composition of the jury in response to defendants’ choices. I found this approach wanting in the context of civil controversies between private litigants, for reasons that need not be repeated here. See id., at 632 (O’Connor, J., dissenting). But even if I thought Edmonson was correctly decided, I could not accept today’s simplistic extension of it. Dodson makes clear that the unique relationship between criminal defendants and the State precludes attributing defendants’ actions to the State, whatever is the case in civil trials. How could it be otherwise when the underlying question is *68whether the accused “c[an] he described in all fairness as a state actor”? 500 U. S., at 620. As Dodson accords with our state action jurisprudence and with common sense, I would honor it.
II
What really seems to bother the Court is the prospect that leaving criminal defendants and their attorneys free to make racially motivated peremptory challenges will undermine the ideal of nondiseriminatory jury selection we espoused in Batson, 476 U. S., at 85-88. The concept that the government alone must honor constitutional dictates, however, is a fundamental tenet of our legal order, not an obstacle to be circumvented. This is particularly so in the context of criminal trials, where we have held the prosecution to uniquely high standards of conduct. See Brady v. Maryland, 373 U. S. 83 (1963) (disclosure of evidence favorable to the accused); Berger v. United States, 295 U. S. 78, 88 (1935) (“The [prosecutor] is the representative not of an ordinary party to a controversy, but of a sovereignty... whose interest... in a criminal prosecution is not that it shall win a ease, but that justice shall be done”).
Considered in purely pragmatic terms, moreover, the Court’s holding may fail to advance nondiseriminatory criminal justice. It is by now clear that conscious and unconscious racism can affect the way white jurors perceive minority defendants and the facts presented at their trials, perhaps determining the verdict of guilt or innocence. See Developments in the Law — Race and the Criminal Process, 101 Harv. L. Rev. 1472,1559-1560 (1988); Colbert, Challenging the Challenge: Thirteenth Amendment as a Prohibition against the Racial Use of Peremptory Challenges, 76 Cornell L. Rev. 1, 110-112 (1990). Using peremptory challenges to secure minority representation on the jury may help to overcome such racial bias, for there is substantial reason to believe that the distorting influence of race is minimized on a racially mixed jury. See id., at 112-115; Developments in *69the Law, supra, at 1559-1560. As amicus NAACP Legal Defense and Educational Fund explained in this case:
“The ability to use peremptory challenges to exclude majority race jurors may be crucial to empaneling a fair jury. In many cases an African American, or other minority defendant, may be faced with a jury array in which his racial group is underrepresented to some degree, but not sufficiently to permit challenge under the Fourteenth Amendment. The only possible chance the defendant may have of having any minority jurors on the jury that actually tries him will be if he uses his peremptories to strike members of the majority race.” Brief for NAACP Legal Defense and Educational Fund, Inc., as Amicus Curiae 9-10 (footnote omitted).
See Brief for National Association of Criminal Defense Lawyers as Amicus Curiae 56-57; Edmonson, supra, at 644 (Scalia, J., dissenting). In a world where the outcome of a minority defendant’s trial may turn on the misconceptions or biases of white jurors, there is cause to question the implications of this Court’s good intentions.
That the Constitution does not give federal judges the reach to wipe all marks of racism from every courtroom in the land is frustrating, to be sure. But such limitations are the necessary and intended consequence of the Fourteenth Amendment’s state action requirement. Because I cannot accept the Court’s conclusion that government is responsible for decisions criminal defendants make while fighting state prosecution, I respectfully dissent.